Radio Monitoring: Problems, Methods and Equipment

Radio Monitoring A volume in the Nanostructure Science and Technology series. Further titles in the series can be foun...

Author: Anatoly Rembovsky | Alexander Ashikhmin | Vladimir Kozmin | Sergey M. Smolskiy

451 downloads 1599 Views 12MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

Radio Monitoring

A volume in the Nanostructure Science and Technology series. Further titles in the series can be found at: http://www.springer.com/series/7818

Lecture Notes in Electrical Engineering Volume 43 Radio Monitoring: Problems, Methods, and Equipment Anatoly Rembovsky, Alexander Ashikhmin, Vladimir Kozmin, and Sergey Smolskiy 978-0-387-98099-7 Incorporating Knowledge Sources into Statistical Speech Recognition Sakti, Sakriani, Markov, Konstantin, Nakamura, Satoshi, and Minker, Wolfgang 978-0-387-85829-6 Intelligent Technical Systems Martínez Madrid, Natividad; Seepold, Ralf E.D. (Eds.) 978-1-4020-9822-2 Languages for Embedded Systems and their Applications Radetzki, Martin (Ed.) 978-1-4020-9713-3 Multisensor Fusion and Integration for Intelligent Systems Lee, Sukhan; Ko, Hanseok; Hahn, Hernsoo (Eds.) 978-3-540-89858-0 Designing Reliable and Efficient Networks on Chips Murali, Srinivasan 978-1-4020-9756-0 Trends in Communication Technologies and Engineering Science Ao, Sio-Iong; Huang, Xu; Wai, Ping-kong Alexander (Eds.) 978-1-4020-9492-7 Functional Design Errors in Digital Circuits: Diagnosis Correction and Repair Chang, Kai-hui, Markov, Igor, Bertacco, Valeria 978-1-4020-9364-7 Traffic and QoS Management in Wireless Multimedia Networks: COST 290 Final Report Koucheryavy, Y., Giambene, G., Staehle, D., Barcelo-Arroyo, F., Braun, T., Siris,V. (Eds.) 978-0-387-85572-1 Proceedings of the 3rd European Conference on Computer Network Defense Siris, V.; Ioannidis, S.; Anagnostakis, K.; Trimintzios, P. (Eds.) 978-0-387-85554-7 Data Mining and Applications in Genomics Ao, Sio-Iong 978-1-4020-8974-9, Vol. 25 Informatics in Control, Automation and Robotics: Selected Papers from the International Conference on Informatics in Control, Automation and Robotics 2007 Filipe, J.B.; Ferrier, Jean-Louis; Andrade-Cetto, Juan (Eds.) 978-3-540-85639-9, Vol. 24 Continued after index

Anatoly Rembovsky · Alexander Ashikhmin · Vladimir Kozmin · Sergey Smolskiy

Radio Monitoring Problems, Methods, and Equipment

123

Anatoly Rembovsky JSC IRCOS Staroalexeevskaya str. 14 Moskva Bldg. 2, Apt. 133 Russia 129626 [email protected]

Alexander Ashikhmin JSC IRCOS Staroalexeevskaya str. 14 Moskva Bldg. 2, Apt. 133 Russia 129626 [email protected]

Vladimir Kozmin JSC IRCOS Staroalexeevskaya str. 14 Moskva Bldg. 2, Apt. 133 Russia 129626 [email protected]

Sergey Smolskiy Department of Radio Receivers Technical University Moscow Power Engineering Institute Lefortovskly Val ul, 7, Apt. 66 Moskva E-116 Russia 111116 [email protected]

ISBN 978-0-387-98099-7 e-ISBN 978-0-387-98100-0 DOI 10.1007/978-0-387-98100-0 Springer Dordrecht Heidelberg London New York Library of Congress Control Number: 2008943693 © Springer Science+Business Media, LLC 2009 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)

Preface

Automated radio monitoring (ARM) technology obtained wide distribution as a tool for problem-solving in various areas, beginning from radio frequency spectrum usage control to the use of radio environment checks to search for illegal radio transmitters. Radio monitoring equipment serves as the basis of technical measures for counteracting unapproved information pick-up, including the all-important investigation of compromising emanations. The list of problems solved with the help of ARM equipment includes: – Revelation and analysis of radio emissions, for the identification of signal and interference sources, – Measurement of radio emission parameters, and the estimation of their danger or value for the user, – Electromagnetic field strength, or the power flow density measurement, – Radio signals and interference direction-finding in the terrain. In particular, ARM equipment allows radio engineering facilities and computer hardware to be checked for the presence and level of incidental emanations. As such, the main functions of ARM equipment are the permanent or periodic observations of airwaves in the wide frequency range, the effective detection, analysis and localization of potential or specially-organized channels of information drain. Based on the authors’ development experience, fundamental information concerning the described ARM systems, reference data, and recommendations on the best methods and approaches for obtaining solutions to the above-mentioned problems are included in the book, together with the classification and detailed description of modern high-efficient hardware-software ARM equipment, including equipment for detection, radio direction-finding, parameters measurement and their analysis, and the identification and localization of electromagnetic field sources. Examples of ARM equipment structure and application, within the complicated interference environments found in industrial centers, inside of buildings, and in the open terrain, are included, together with the software required for such applications.

v

vi

Preface

The book is prepared on the basis of Russian and foreign publications and as a result of various research and implementation activities of IRCOS1 company experts, under the supervision and direct participation of the authors. The book contains 12 chapters. In Chapter 2, the list of problems solved by ARM systems is discussed in detail. An analysis of the nomenclature, structure, functions and parameters of ARM equipment is performed, and the system hierarchy of the facilities is developed. The composition, the functions, and the main technical characteristics for each class of equipment are determined. Chapter 3 is devoted to the basic parameters of up-to-date radio receivers affecting ARM problem fulfillment. The peculiarities of the digital receiver structure for the 9 kHz – 18 GHz frequency range are shown. Design examples and the characteristics of single-channel and double-channel digital receivers are discussed. Chapter 4 is dedicated to the mathematical aspects of narrow-band signal detection, as well as the signals with dynamic frequency-time distribution (with frequency hopping) for single- and double-channel radio equipment. ARM problem-solving via multi-channel panoramic digital receivers is analyzed in Chapter 5, together with the hardware and software structure peculiarities of these receivers and their main technical data. Chapters 6 and 7 are devoted to the radio signals used in communication, broadcasting, TV and data transmission systems, and to the technical analysis and parameter measurement of modulated and non-modulated signals. Examples of radio signal parameter measurement are discussed and recommendations for software applications are given. A review of and the theoretical bases for direction-finding methods are presented in Chapter 8, and the main parameters of radio direction finders are explained. Examples of multifunctional radio monitoring and direction-finding equipment in VHF, UHF, and microwave ranges are described. The affect of used digital receivers on direction-finding effectiveness is shown. Chapter 9 is devoted to the development of geographically-distributed radio monitoring systems and to direction-finding systems for radio emission sources. The application of stationary, mobile, portable and hand-held ARM equipment is considered. Moreover, the problems related to ARM station system equipment, organization of data transmission through the communication, navigation and power supply channels, are considered in this chapter as well. The possible uses of software for signal detection, their parameter measurement, and direction-finding of radio emission sources – with positions indicated on an electronic map – are discussed. Chapter 10 includes information on determining the position of radio emission sources by mobile radio monitoring stations, and estimation of field strength distribution, taking into account terrain relief and area reclamation, to obtain covering zones of broadcasting and communication. Solutions to the problems of

1 IRCOS

means: Investigations on Radio Control and System design

Preface

vii

electromagnetic compatibility and the parameters testing of radio electronic equipment are discussed also. Chapter 11 describes the structural peculiarities of radio monitoring equipment inside the premises and the revelation of technical channels of information leakage and unapproved radio emission sources. Revelation methods are discussed, together with these source localization methods on checked objects. Implementation examples for hardware-software facilities for technical channel leakage revelation, used both inside the premises and on the boundary of the checked zone, are presented. In Chapter 12, the problems of radio system structure in performing compromising emanations investigation are considered. The theoretical aspects and the practical approaches for the revelation of the informative components are discussed, with calculation of the checked area and object immunity radii. The equipment and the software examples for these investigations are given. The authors are confident that the materials offered in the book will be useful to experts in the area of radio monitoring, to operators and leaders of civil and military radio-checking services, and to security service employees of both state and commercial structures. The book can be recommended to the students of technical universities and colleges, studying in the appropriate fields.

“This page left intentionally blank.”

Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1

2 Problems, Classification and Structure of ARM Equipment Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Classification of Radio Monitoring Equipment . . . . . . . . . Operation Zone Sizes . . . . . . . . . . . . . . . . . . . . . Application . . . . . . . . . . . . . . . . . . . . . . . . . . Equipment Performance . . . . . . . . . . . . . . . . . . . . Design Constraints . . . . . . . . . . . . . . . . . . . . . . Radio Monitoring Equipment Design Philosophy . . . . . . . . Requirements for RM Equipment Technical Parameters . . . . Quality Criterion Selection . . . . . . . . . . . . . . . . . . Main Technical Parameters of RM Equipment . . . . . . . . Characteristics of RM Equipment Families . . . . . . . . . . . Radio Monitoring and RES Location Detection Systems . . . Stationary and Mobile RM Stations . . . . . . . . . . . . . . Portable RM Equipment . . . . . . . . . . . . . . . . . . . Manpack ARM Equipment . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . .

5 5 6 7 7 8 8 9 12 12 13 16 16 16 18 19 21 22

3 Radio Receiver Applications for Radio Monitoring System Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . Tuned Radio Receiver . . . . . . . . . . . . . . . . . . . . . Main Radio Receiver Parameters . . . . . . . . . . . . . . . Operating Frequency Range . . . . . . . . . . . . . . . . Amplitude-Frequency Response of the Linear Receive Path Voltage Standing Wave Ratio . . . . . . . . . . . . . . . . Main Channel and Spurious Channels . . . . . . . . . . . RR Selectivity . . . . . . . . . . . . . . . . . . . . . . . . Inherent Noise and Receiver Sensitivity . . . . . . . . . . Sensitivity Increase with the Help of Pre-amplifiers . . . . Pre-amplifier Gain Factor Selection . . . . . . . . . . . . . Receiver Multi-Signal Selectivity . . . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

23 23 23 27 28 28 29 30 34 36 39 42 43

. . . . . . . . . . . . .

ix

x

Contents

Intermodulation Noise . . . . . . . . . . . . . . . . . . . . Intercept Points on IP2 and IP3 Intermodulation . . . . . . . Intermodulation-Free Dynamic Range Determination . . . . Attenuator Influence on the Intermodulation Value . . . . . . Determining the Intercept Points . . . . . . . . . . . . . . . Blockage Effect . . . . . . . . . . . . . . . . . . . . . . . . Crosstalk Distortions . . . . . . . . . . . . . . . . . . . . . Phase Noise and Retuning Rate of the Panoramic RR . . . . Digital Radio Receivers . . . . . . . . . . . . . . . . . . . . . General Principles of Digital Radio Receiver Implementation Types of ARM Receivers . . . . . . . . . . . . . . . . . . . Development of Russian Arm Systems . . . . . . . . . . . . . First- and Second-Generation Systems . . . . . . . . . . . . Radio Receivers of the Third and Fourth Generation . . . . . Fifth-Generation Radio Receivers . . . . . . . . . . . . . . ARK-CT1 Digital Radio Receiver . . . . . . . . . . . . . . . . ARK-D1TP Digital Panoramic Measuring Receiver . . . . . . ARK-CT3 Digital Receiver . . . . . . . . . . . . . . . . . . . ARK-KNV4 External Remote-Controlled Converter . . . . . . ARK-PR5 “Argamak” Digital Radio Receiver . . . . . . . . . ARGAMAK-I Panoramic Measuring Receiver . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . .

43 47 50 51 52 52 53 54 56 56 58 60 60 62 66 67 73 75 79 81 93 93 94

. . . . . . . . . . . .

95 95 97

4 Single-Channel and Multi-Channel Radio Signal Detection Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Single-Channel Signal Detection . . . . . . . . . . . . . . . . Characteristics of Single-Channel Detection of Narrow-Band Signals . . . . . . . . . . . . . . . . . . . . . Single-Channel Detection of Radio Signals With POFT . . . . Probabilistic Features of the Frequency Observation Time . . Probability of Separate Frequency Registration . . . . . . . Estimate of the Total Number of Registered Frequencies . . Optimization of ARM System Parameters . . . . . . . . . . Detection Characteristics . . . . . . . . . . . . . . . . . . . Double-Channel Detection of Narrow-Band Signals . . . . . . Comparison of Single-Channel and Double-Channel Processing Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

. . . . . . . . . . .

105 108 109 111 112 113 115 117 119 121 121

5 Multi-Channel Digital Receivers . . . . . Introduction . . . . . . . . . . . . . . . . . Panoramic Multi-Channel Receivers . . . . ARK-D11 Double-Channel Complex . . . . ARK-RD8M Multi-Channel Complex . . . SMO-MCRM Customized Software Package

. . . . . .

. . . . . .

. . . . . .

. . . . . .

123 123 123 125 126 130

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

. . . . . .

Contents

Software Purpose and Performance Capabilities Software Operation Modes . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . .

xi

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

. . . .

130 131 133 133

6 Modulation and Signal Types in Modern Radioelectronic Means Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Administrative Division of the Frequency Spectrum . . . . . . . . . Modulation in Communication and Broadcast Systems . . . . . . . . General Information . . . . . . . . . . . . . . . . . . . . . . . . . Types of Analog Modulation . . . . . . . . . . . . . . . . . . . . Types of Discrete (Digital) Modulation . . . . . . . . . . . . . . . Signals of Modern Radio Electronic Means . . . . . . . . . . . . . . SW Range Signals (Less Than 30 MHz) . . . . . . . . . . . . . . VHF Range Signals (More Than 30 MHz) . . . . . . . . . . . . . International System for Signal Designation . . . . . . . . . . . . International Frequency Range Distribution . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . .

135 135 135 138 138 140 147 158 158 165 182 188 191 192

7 Measurement of Radio Signal Parameters . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Frequency Measurement . . . . . . . . . . . . . . . . . . . . . . . . Instantaneous Frequency Measurement Method . . . . . . . . . . FFT Method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Measurement of Spectrum Width . . . . . . . . . . . . . . . . . . Determination of Modulation Type and Its Parameter Measurement . Determination of Modulation Type . . . . . . . . . . . . . . . . . Modulation and the Determination of Shift-Keying Characteristics SMO-STA Software for the Analysis of Automated Radio Signals . . STA Software Possibilities and Its Functional Diagram . . . . . . Examples of Radio Signal Modulation Type and Parameters’ Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . Automated Technical Analysis of Radio Signals . . . . . . . . . . . Unit for Automated Radio Signal Analysis . . . . . . . . . . . . . Peculiarities of SMO-PA Application . . . . . . . . . . . . . . . . Application of Automatic Signal Analysis in SMO-RD2 . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . .

195 195 195 196 197 199 201 201 204 206 206

. . . . . . .

208 219 219 221 233 235 236

8 Direction Finding of Radio Emission Sources . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . History of Radio Direction-Finding Technique . . . . . . . . . . . Structural Diagram and Characteristics of Radio Direction Finders Main Technical Parameters of Radio Direction Finders . . . . . . . Accuracy of Direction Finding . . . . . . . . . . . . . . . . . .

. . . . . .

237 237 238 241 242 243

. . . . . .

xii

Contents

Sensitivity . . . . . . . . . . . . . . . . . . . . . . . . . Noise Immunity . . . . . . . . . . . . . . . . . . . . . . Operating Rate . . . . . . . . . . . . . . . . . . . . . . Resolution . . . . . . . . . . . . . . . . . . . . . . . . . Operating Frequency Range . . . . . . . . . . . . . . . Types of Being-Found Signals . . . . . . . . . . . . . . Deployment Time . . . . . . . . . . . . . . . . . . . . . Weight and Size . . . . . . . . . . . . . . . . . . . . . . Complexity in Manufacture and Operation . . . . . . . . Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . Classification of Direction-Finding Methods . . . . . . . . Systems Based on a Rotating Directional Antenna . . . . . ARK-RP3 Handheld Radio Direction Finder . . . . . . . . ARK-RP4 Handheld Radio Direction Finder . . . . . . . . Automatic Radio Compass . . . . . . . . . . . . . . . . . Automatic Radio Direction Finder with Low Antenna Base Doppler and Quasi-Doppler Direction Finders . . . . . . . Phase and Correlation Interferometers . . . . . . . . . . . . Peculiarities of Correlation Interferometer . . . . . . . . . Algorithm of Correlation Interferometer Measuring System Single-Channel Measuring System on the Basis of a Correlation Interferometer . . . . . . . . . . . . . . . ARTIKUL-M4 Foldable Correlation Interferometer . . . . ARTIKUL-M1 Mobile Direction Finder . . . . . . . . . . ARTIKUL-P Portable Foldable Direction Finder . . . . . . ARTIKUL-P11 Portable Foldable Direction Finder . . . . . Direction Finding Error Correction in Mobile Systems . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . .

246 247 248 248 249 249 249 249 249 250 250 252 256 262 263 266 270 273 278 281

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

289 295 299 301 306 307 315 316

. . . . . . . . . .

317 317

. . . . .

318

. . . . . . . . .

320 321 322 322 323 326 326 328 328

9 Radio Monitoring Systems and Determination of Radio Emission Sources Location . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . Requirements for Radio Monitoring and Location Determination Systems . . . . . . . . . . . . . . . . . . . . Structure of the Radio Monitoring System and Determination of RES Location . . . . . . . . . . . . . . . . . . . . . . . . ARK-POM1 System . . . . . . . . . . . . . . . . . . . . ARK-POM2 System . . . . . . . . . . . . . . . . . . . . ARK-POM3 Geographically-Distributed System . . . . . Combined ARK-POM System . . . . . . . . . . . . . . . Control Arrangement in the System . . . . . . . . . . . . . . Data Exchange between Stationary Posts . . . . . . . . . . Data Exchange with the Mobile and Deployed Posts . . . . Peculiarity of the Low-Speed Radio Channel Application .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

Contents

xiii

Usage of Radio Modems of the Cellular Communication Systems Data Exchange Implementation in Combined ARK-POM System . “Archa” Stationary Station . . . . . . . . . . . . . . . . . . . . . . . “Argument” Mobile Station . . . . . . . . . . . . . . . . . . . . . . System-Wide Car Equipment . . . . . . . . . . . . . . . . . . . . “Arena” Portable Station . . . . . . . . . . . . . . . . . . . . . . . . “Arena” Station Structure . . . . . . . . . . . . . . . . . . . . . . Mast Devices for Radio Monitoring Stations . . . . . . . . . . . . . Navigation Systems for Radio Monitoring Stations . . . . . . . . . . Features of Modern Navigation Systems . . . . . . . . . . . . . . Navigation Systems for Mobile Stations . . . . . . . . . . . . . . Electric Power Supply Systems . . . . . . . . . . . . . . . . . . . . Requirements for Electric Power Sources . . . . . . . . . . . . . Electric Power Sources for Radio Equipment . . . . . . . . . . . . Secondary Electric Supply Sources . . . . . . . . . . . . . . . . . Example of Pulse Power Supply of Low Power . . . . . . . . . . Multi-Channel Pulse Power Source . . . . . . . . . . . . . . . . . ARK-UPS12 Universal Power Supply Unit . . . . . . . . . . . . Autonomous Electric Station Usage . . . . . . . . . . . . . . . . Special Software Support and Operation Modes of Stations . . . . . Software Support Structure . . . . . . . . . . . . . . . . . . . . . “Spectrum” Mode . . . . . . . . . . . . . . . . . . . . . . . . . . “Search” Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . “Bearing” Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . “Measurement” and “Technical Analysis” Modes . . . . . . . . . “Review” Mode . . . . . . . . . . . . . . . . . . . . . . . . . . . “Multi-Channel Direction Finding” Mode . . . . . . . . . . . . . Peculiarities of the Direction Finding of POFT Stations . . . . . . “Electronic Map” Mode . . . . . . . . . . . . . . . . . . . . . . . Post-processing Mode . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Radio Emission Source Localization Using Mobile Stations and Field Strength Measurement . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods of RES Localization Using the Mobile Station . . . Drive Method . . . . . . . . . . . . . . . . . . . . . . . . . Quasi-Stationary Method . . . . . . . . . . . . . . . . . . . Method of Automatic Calculation of RES Coordinates During Movement . . . . . . . . . . . . . . . . . . . . . . . Peculiarities of Multi-Channel Direction Finding . . . . . . Simultaneous Direction Finding . . . . . . . . . . . . . . . Electromagnetic Field Strength Measurement . . . . . . . . . . Main Mathematical Relations . . . . . . . . . . . . . . . . .

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

331 332 334 338 343 346 346 347 349 350 355 356 356 357 360 362 364 365 369 372 372 373 374 377 378 381 382 383 385 388 390 391

. . . . .

. . . . .

. . . . .

. . . . .

393 393 395 395 396

. . . . .

. . . . .

. . . . .

. . . . .

398 401 402 403 404

xiv

Contents

Peculiarities of the Field Strength Distribution Estimation . Field Strength Measurement . . . . . . . . . . . . . . . . On-Site Calculation of Field Strength Distribution . . . . . District Topography . . . . . . . . . . . . . . . . . . . . . Urban Build-Up . . . . . . . . . . . . . . . . . . . . . . . Vegetation Influence . . . . . . . . . . . . . . . . . . . . . Calculation of Field Strength in the SMO-KN Application Processing of Field Strength Measurements . . . . . . . . Determination of RES Location . . . . . . . . . . . . . . . Checking Transmitters for Announced Parameters . . . . . Calculation of Electromagnetic Compatibility . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

. . . . . . . . . . . . .

Detection and Localization of Technical Channels of Information Leakage . . . . . . . . . . . . . . . . . . . . . . . . . Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Main Search Stages for Electromagnetic Channels of Information Leakage . . . . . . . . . . . . . . . . . . . . . . . . Detection of Radio Signals Emitted in Monitored Premise . . . . . . Radio Signal Intensity in Near-Field and Far-Field Regions . . . . Generalized Structure of Equipment for TCIL Detection . . . . . . Comparison Technique for Signal Intensities . . . . . . . . . . . . Detection Algorithm for Radio Signal Sources in Monitored Area Detection Effectiveness Dependence on the Equipment and the Ways of “Standard” Panorama Obtaining . . . . . . . . . Identification and Localization of Radio Microphones . . . . . . . . Distant Radio Monitoring Systems of Remote Premises . . . . . . . Construction Principles of Remote Radio Monitoring Systems . . Examples of Remote Radio Monitoring Systems . . . . . . . . . . Peculiarities of ARK-D3T Remote Radio Monitoring System . . . Peculiarities of the ARK-D9 Remote Radio Monitoring System . . Peculiarities of the ARK-D13 Remote Radio Monitoring System . Software for Remote Radio Monitoring Systems . . . . . . . . . . . Purpose and Possibilities of SMO-DX Application . . . . . . . . . Peculiarities of Radio Microphone Detection . . . . . . . . . . . . Joint Usage of the Various Detection Algorithms . . . . . . . . . . Radio microphone Localization Inside of Monitored Premises . . Equipment Operation in the Remote Radio Monitoring System . . Detection of TCIL Sources by the Mobile Station . . . . . . . . . . Antenna System Selection . . . . . . . . . . . . . . . . . . . . . Methods of RES Detection . . . . . . . . . . . . . . . . . . . . . Equipment Structure of ARTIKUL-M6 Mobile Direction Finder . Software Structure and Search Procedure Implementation . . . . . Aggregation of the Initial Data Frames . . . . . . . . . . . . . . .

. . . . . . . . . . . . .

406 407 408 409 411 412 414 415 419 421 423 424 424

. .

427 427

. . . . . .

428 430 430 432 433 435

. . . . . . . . . . . . . . . . . . . .

436 437 438 438 441 443 448 450 451 452 453 454 455 455 457 458 459 461 462 464

Contents

xv

Frame Processing and Generation of “Suspicious” Frequency List Checking the Frequencies from the List and More Precise RES Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

465

. . .

467 468 469

. . . . . . . . . . . .

471 471 471 473 475 475 476 478 479 482 484 485

. . . . . . . .

486 488

. . . .

. . . .

490 492 496 496

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

497

Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

499

12

Methods and Equipment for Protection Against Information Leakage Via CEE Channels . . . . . . . . . . . General Information . . . . . . . . . . . . . . . . . . . . . . . Special Investigation Types and Information Security Index . . Calculation of Information Security Index . . . . . . . . . . . Estimation of the Testing Mode Parameters for a LCD Monitor Estimation of the Testing Mode Parameters for a CRT Monitor Methods of Detection of CEE Informative Components . . . . Probabilistic Features of Periodogram Samples . . . . . . . . . TDM Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . Application of ARK-D1TI Measuring Complex . . . . . . . . Search of CEE Informative Components . . . . . . . . . . . Measurement of CEE Informative Component Intensity . . . Calculation of the Monitored Zone Radius by SMO-PRIZ Application . . . . . . . . . . . . . . . . . . . . . . . . . . Information Security Monitoring . . . . . . . . . . . . . . . . SMO-PRIZ Application Operation for Information Security Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . Purposes and Functions of SMO-THESIS Application . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . .

. . . . . . . . . . . .

. . . .

“This page left intentionally blank.”

Acronyms

AA ACS ADC AE AFH AFR AGC ALE AM AnM APM APSK ARI ARM ARME ARQ ARU AS ASK BP BPT BRPSK BRPT BS BWLL CB CDCS CDMA CE CEE CEEP CEPT CF CIM

- antenna array - amplifying-converting section - analog-digital converter - antenna element - automatic frequency hopping - amplitude-frequency response - automated gain control - automatic link establishment - amplitude modulation - angle modulation - amplitude-pulse modulation - amplitude-phase shift-keying - Autofahrer Rundfunk Information - automated radio monitoring - automated radio monitoring equipment - Auto ReQuest - antenna-receiver unit - antenna system - amplitude shift-keying - bearing pair - binary phase telegraphy - binary relative phase shift-keying - binary relative phase telegraphy - basing station - Broadband Wireless Local Loop - Citizen s Band - Continuous Dynamic Channel Selection - Code Division Multiple Access - consumer equipment - compromising electromagnetic emanation - compromising electromagnetic emanation and pick-up - Conference European for post and telecommunication - computing facility - Correlation interferometric meter (correlative interferometer) xvii

xviii

COFDM CP CPFSK CRC CRT CTF CW DAC DAM DARC DB DC DDM DECT DF DFT DPRS DPSK DRA DRM DRMS DRR DSBAM DSBSC DSP DSSS DVBT EBU EDGE EHF EMA EMC EMF EMW ETSI FCU FDMA FEC FFSK FFT FH FM FP FS FTD

Acronyms

- Coded Orthgonal Frequency Division Multiplexing - central post - continuous phase frequency shift-keying - Cyclic Redundance Check - cathode-ray tube - complex transfer factor - continuous wave - digital-analog converter - DAM modulation - Data Radio Channel - database - distant control panel - difference-distance measuring - Digital Enhanced Cordless Telecommunications - direction finding, direction finder - discrete Fourier transform - DECT Packet Radio Services - differential phase shift-keying - distributed random antenna - Digital Radio Mondiale - distant radio monitoring system - digital radio receiver - double sideband amplitude modulation - double sideband suppressed carrier - digital signal processing - Direct Sequence Spread Spectrum - Digital Video Broadcasting - European Broadcasting Union - Enhanced Data rates for Global Evolution - extremely high frequency - electromagnetic availability - electromagnetic compatibility - electromagnetic field - electromagnetic wave - European Telecommunication Standards Institute - frequency conversion unit - Frequency Division Multiple Access - Forward Error Correction - fast frequency shift-keying - fast Fourier transform - frequency hopping - frequency modulation - frequency position - frequency synthesizer - frequency-time diagram

Acronyms

FV GEG GFSK GIS GPRS GPS GSM GTC HFF HiperLAN ICAO IEEE IF IFM IMC INS IP ISSB ITA2 ITU LCD LF LMSK LNA LO LRA LSB LW MASK MF (UHF) MFSK MMDS MP MPC MSK MUSIC MW NB NFM NICAM OBW OFDM OOK OQPSK PBF

xix

- flying vehicle - gasoline electric generator - Gaussian frequency shift-keying - geo-information system - General Packet Radio Service - Global Positioning System - Global System for Mobile communications - gain-transfer characteristic - high-frequency filter - High Performance Local Area Network - International Civil Aviation Organization - Institute of Electrical and Electronic Engineers - intermediate frequency - instantaneous frequency measurement - intermodulation component - inertial navigation system - interception point - Independent Single Sideband - International Teleprinter Alphabet - International Telecommunication Union - liquid-crystal display - low frequency - frequency shift-keying with minimal shift and with level regulation - low-noise amplifier - local oscillator - lumped random antenna - lower sideband - long waves - multiple amplitude shift-keying - microwave frequency - Multiple frequency shift-keying - Multichannel Multipoint Distribution System - monitored premise - microprocessor control - Minimum Shift Keying - Multiple Signal Classification - medium waves - Normal Burst - Narrow Frequency Modulation - Near Instantaneous Companded Audio Multiplex - occupied bandwidth - Orthogonal Frequency Division Multiplexing - On/Off Keying - offset quadrature phase shift-keying - pass-band filter

xx

PC PM POFT PPM PS PSA PSF PSK PTA PWM QAM QASK QM QPSK RDS REE REM RES RF RFA RFS RMD RMS RO RPSK RPU RR RRMS RSS RTTY SA SAN SBD SCA SFH SG SGU SHF SMPS SMS SNR SPS SQPSK SR SRNS

Acronyms

- personal computer - phase modulation - programmable operating frequency tuning - phase-pulse modulation - phase-shifter - panoramic spectral analysis - power source filter - phase shift-keying - panoramic-technical analysis - pulse-width modulation - quadrature amplitude modulation - quadrature amplitude shift-keying - quadrature modulation - quadrature phase shift-keying - Radio Data System - radio electronic environment - radio electronic means - radio emission source - radio frequency - radio frequency amplifier - radio frequency spectrum - reference-methodical documentation - root-mean-square value (deviation) - reference oscillator - relative phase shift-keying - reception and processing unit - radio receiver - Remote Radio Monitoring System - reference spatial signal - Radio Tele Type - spectrum analyzer, space apparatus - system of active noisiness - spectral and bearing data - Sub-carrier Communication Allocation - Slow Frequency Hopping - signal generator - signal generation unit - super high frequency - switch-mode power supply - special mathematical software - signal/noise ratio - secondary power source - staggered quadrature phase shift-keying - Selective Repeat - satellite radio navigation system

Acronyms

SS SSBh SSBl SSBSC SV SW SWRV TCIL TCP/IP T-DAB TDM TDMA TDS TOI TTF UE UHF UMTS UPS URES US USB VHF VLF VSB VSWR WARC WCDMA WFM WLAN WMAN WPAN WTSC

xxi

- software support - single-side band (higher) - single-side band (lower) - Single Sideband Suppressed Carrier - space vehicle - short waves - standing-wave factor on voltage - technical channel of information leakage - Transport Control Protocol/Internet Protocol - Terrestrial Digital Audio Broadcasting - Testing and Detection Mutual - Time Division Multiple Access - Testing and Detection Separate - Third Order Intercept - tactical-technical features - user equipment - Ultra High Frequency - Universal Mobile - uninterrupted power supply - unwanted radio emission source - user station - Upper Sideband - very high frequency - very low frequency - Vestigal Side Band - Voltage Standing Wave Ratio - World Administration Radio Conference - Wideband-Code Division Multiple Access - Wide Frequency Modulation - Wireless Local Area Network - Wireless Metropolitan Area Network - Wireless Personal Area Network - World Telecommunication Standards Conference

Chapter 1

Introduction

Equipment for automated radio monitoring (ARM) can be considered an information extraction system. ARM equipment is widely used in various areas and was developed at the same rate as information transmission systems through radio channels. There are many problems which ARM can address: planned checking of regular equipment parameters, unpremeditated interference level measurement, detection and determination of non-licensed transmitter locations, measurement of energy covering zones during the estimation of radio communication quality, and determination of radio frequency resource usage intensity. ARM equipment also solves the problem of informational security. The increase in ARM equipment at present is caused by several reasons, the first of which relates to the continued technical progress of radio communication equipment, but the second relates to changes of an economical and political nature, which have occurred in the world. It should be noted that, in Russia, prior to 1992, radio frequency band loading, new frequency allocation, and radio frequency usage regulations were effectively and strictly controlled by the appropriate state agencies, including the security services of the various levels, and that strong restrictions on new radio communications equipment import and usage were simultaneously enacted. Under these conditions, ARM problems were effectively solved by existing and newly-developed native equipment. Standard modernization and replacement of equipment were fulfilled in planned order. An evident and rapid increase of ARM problems, in solving the tasks of radio monitoring and technical informational security, became apparent in Russia after 1992, due to the political and economical changes that occurred there. A strong reduction in the large number of radio electronic enterprises, which had earlier occupied leading positions in the development and manufacture of ARM equipment, caused leading-expert outflow from this field and, hence, a large reduction of modern ARM equipment delivery by these companies. This circumstance caused the slowing down of high-quality native ARM equipment delivery, both in product assortment and in the parameters and performance of the equipment. At the same time, in highly-developed, foreign countries, the radio monitoring equipment evolution advanced, as earlier, by an increasing rate, since the high effectiveness of radio electronic means (REM) for various types of information transmission – at A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_1,

1

2

1

Introduction

constant cost price reduction – stimulated REM distribution greatly to all corners of the world. The fundamental complications of the radio electronic environment (REE) observed now in Russia (and typical, apparently, for the other countries), can be related to the following factors: – Increase in the number of regular TV and radio transmitters, introduction and further modernization of cellular communication systems and an increase in its usage intensity, a process which is far from complete; – Overloading of several regions of the radio range (e.g., sub-ranges 40, 100, 400, 800, and 2450 MHz), caused by a series of objective circumstances, such as the best conditions for radio wave propagation, no need to grant a license, etc.; – Permanent increase of REM operating range upper limit (at present to 18– 60 GHz), corresponding to the rapid development of modern technologies and instruments; – Application of various types of new waveforms: narrow-band with fixed frequency distribution, or with the dynamic frequency-time distribution of emission, and wide-band with code user division; – General tendency of REM transmitter power increase, dictated by the attempt to extend their action range, which is equivalent to the REM number increase acting in the receiver site point of the ARM equipment and leading to the unpremeditated interference level rise both at the main frequency and the harmonics; – Successful research on various receiver sensitivity giving rise to the necessity of an appropriate increase in ARM receiver sensitivity, required for reliable REE revelation and analysis. Additionally, the number of non-licensed radio emission sources (RES) with various power levels, and the large (over level and spectrum) number of spurious emissions not corresponding to the permissible norm and the international standards, increases permanently in the cities and industrial centers of many countries, which requires the responsible agencies to keep a closer watch for its number, parameters, and territorial allocation. One of the results of the last 10–15 years in Russia has been the definite liberalization of the radio frequency spectrum usage, which has become apparent, in particular, in the distribution of a huge number of uncontrolled devices capable of intercepting private information and the non-licensed equipment for its transmission . . .. Nevertheless, after some time passed, the appropriate legislative documents were issued and confirmed in Russia, but these efforts, unfortunately, have had limited effectiveness. As a result, not only has legalized equipment for information interception been placed in the market now – the manufacture and delivery of which can be controlled by responsible authorities – but uncontrolled equipment with very “exotic” modulation types – and very dangerous from the standpoint of economic security – is also available. A consequence of this period of time is the drastic increase in the volume of used office equipment and electronic equipment for household and industrial

1

Introduction

3

applications. These devices have the compromising electromagnetic emanations that are, in many cases, information leakage channels, for example, due to the microphone effect of the HF i UHF oscillators, the correlation between the monitor emission parameters and the computers with processed data. Moreover, it is necessary to note the following several factors related to REE complications on protected (controlled) objects: – The first relates to the large number of used REM, located in limited and often in enough low space, which can lead to the great complexity of unwanted RES (URES) revelation; – The second is the essential increase of the information transmission rate and the redundancy application to increase the secrecy and noise immunity of several REM, to which the equipment used in measuring and information radio systems of state and commercial enterprises, the wide-band systems with dynamic frequencytime structure, etc. can be, primarily, concerned; – Non-uniform (in time) REM usage leads to additional REE complications, at instances of radio system maximal intensity operation. The increasing problem of ARM effectiveness is redoubled by the fact that, with the growth in the number of international contacts and due to the liberalization of the REM market in Russia, the threats from foreign states – which collect data concerning the industrial and economic secrets of Russian enterprises and watch the scientific and technological developments in the field of perspective technologies – also grows as well. Technical facilities, in particular, the radio electronic ones, are the most important for this activity, because they are very suitable for secretive information transmission. A similar technique for veiled information theft is oriented toward obtaining and transmitting through radio channels any and all messages: from acoustic signals and speech, phone and fax signals, to emissions from computers and monitors, and other information signals, modulating the radio waveforms by many various methods. It is quite evident that the information security services of private and state organizations, and state agencies as a whole, cannot ignore the problem of possible veiled information theft and should take reciprocal measures to use radio systems as an effective counteraction to these threats. On a new level, the appearance of information security problems on controlled objects clearly shows the definite scientific, and particularly the technical, lag in ARM techniques capable of adequately resisting this threat during REE checking, of revealing and localizing potentially dangerous RES, and of detecting the electromagnetic emission and cross-talks capable of leading to important information leakage. As a matter of fact, the technical and procedural level of ARM equipment must be equal to any future achievements within the field of “information transmission,” otherwise the information struggle will be lost. All these factors stimulate both the development of ARM technologies and equipment, and the creation, by the experts, of ARM technical systems, integrated

4

1

Introduction

by the generality of the problem, by the unity approach to its structures, and by the universality and multifunctionality of its solutions. The main goal of the present book consists in the description of the structure and function of digital radio receivers and radio systems intended for radio monitoring and technical information security tasks, beginning from the characteristics and structural diagrams of radio electronic sets and systems, including ARM systems to the description and explanation of the functioning of complicated systems. To this end, we consider the methods, algorithms and peculiarities of the appropriate software. The theoretical discussions are explained via specific equipment examples of REM, created at present in Russia.

Chapter 2

Problems, Classification and Structure of ARM Equipment

Introduction The fundamental purposes of radio monitoring (RM) equipment are: • • • • •

Permanent or intermittent monitoring of airwaves in the wide frequency range Detection and analysis of new emissions Determination of the emission sources location Evaluation its danger or value Detection of unintentional or specially-organized radio channels, for information leakage.

Each of these tasks is a complex, multistage one. Each can be solved under the conditions of the complex electromagnetic environment, and each requires the application of a wide range of radio electronic means (REM), which execute definite functions [1, 2]. These functions can be divided into the following main groups: 1. Universal functions, which, as a rule, are executed by modern, automated RM (ARM) systems 2. Additional functions for specific RM task solutions in the field 3. Additional functions for RM task solutions at one, separately-controlled location, or at a group of the most important premises of the controlled object 4. Additional functions for detecting compromising electromagnetic emanations (CEE). Regarding the first RM functions group (universal functions) one can consider the following: • Real-time panoramic spectral analysis with the maximum high rate, resolving capacity, and adaptation to the complex electromagnetic environment • Fast search for new emissions, including wide-band, and emissions with the dynamic time-frequency structure, its parameters measurement, and comparison to the database, to determine its danger (value) for the user A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_2,

5

6

2

Problems, Classification and Structure of ARM Equipment

• Creation of a signals database, its replenishment, and the registered data comparison with the references stored in the database • Control of the radio emission sources (RES) with emission parameters estimation • Radio signals recording, including digital signals, simultaneously with the service parameters (frequency, time, signal level, spectrum data, etc.) and its further play-back • Real-time and post-processing technical analysis of radio signals. The following functions can be classified to the second functional group: • Field strength measurement • RES direction finding with arbitrary types of modulation on azimuth and elevation angles • Stationary and mobile RES location determination in the field and on extensive objects, and its representation in cartographic diagrams (digital object image). The third group of tasks includes: • Search and detection of the technical channels of information leakage, at the separate or combined premises • RES identification as a radio microphone • RES site location. The compromising electromagnetic emanation (CEE) detection (the fourth group function) provides the following: • Technical means emission parameters and electromagnetic field strength measurement in the receiving antenna near-field zone • Confidential information immunity examination during the course of its processing and storing by the intended technical facilities • Survey of confidential information immunity against leakage, due to the pick-up from the auxiliary technical facilities, systems, and its communication lines • Allocated premises immunity analysis against speech information leakage through the acoustic and electric transformation channels • Measures effectiveness control concerning information security against the CEE leakage.

Classification of Radio Monitoring Equipment It is expedient to classify RM equipment based on specific signs, with further determination of the RM facilities’ efficient structure within each group. These signs are: • Size of the RM operation zone (territory) • RM means application

Classification of Radio Monitoring Equipment

7

• Executed functions • Performance of RM means • Design constraints. Let us consider, in detail, the RM means categories, according to these signs.

Operation Zone Sizes Based on operation zone size, all radio monitoring means can be grouped as follows [3]: • Means for RM task solutions in the field and RES direction finding • Means for information protection measures on the external boundaries of the controlled objects • Means for RM task solutions within separate or several controlled premises of the object; these facilities will be referred to as eavesdropping detection means • Facilities for CEE special investigations. The first and second group means should be able to cover substantial territories with the possibility of RES detection at the exits, and at the external boundaries of the controlled objects. The third group means should provide RM task solutions with maximum operating rate, to detect the RES location and to identify it as a radio microphone. These tasks should be solved at both the separate premise and the premises group, under control from one post. Control facilities are located inside the premises. Special investigations of the technical means for CEE presence can be executed, as a rule, in specially-allocated premises, but investigations are possible directly at the place of the means location, as well.

Application Based on the applications, RM means can be classified into three groups: • For open operation at stationary or temporary posts, as well as when moving on different transport carriers • For concealed operation with RM means carried in an attaché case, handbag, or on the operator’s body. In this instance, appropriate measures for camouflaging the antenna system should be provided, as well as measures to conceal the technical means design, and, in several cases, in combination with fully-autonomous functioning during operator movement • For combined (open or concealed) RM means application, with the possibility of carrier control, and the necessary measures for camouflaging the antenna systems and the appropriate, RM means design.

8

2

Problems, Classification and Structure of ARM Equipment

Equipment Performance RM means performance can be characterized by the signal panoramic spectral analysis operating rate at the given resolution and the dynamic range. Typically, the following classifications are used: • • • •

Low performance (10–100 MHz/s) Medium performance (100–1,000 MHz/s) High performance (1,000–10,000 MHz/s) Ultra-high performance (more than 10 GHz/s).

Design Constraints Modern RM equipment has been created based on a system approach: hardwaresoftware means united by overall design. This approach provides the ability to link each separate means with its weight and size parameters, its electromagnetic compatibility, the decoupling of its power supply, and the development of its design implementation, all of which correspond to the used-carrier parameters. Such problems can be solved by effective classification of the means into groups. Each of the groups shall fulfill each – or a number of – stated conditions. It is often suggested to divide all means into families: stationary, mobile, portable and handheld means. When developing the means of each family, those technical solutions are preferable, which first of all comply with the set of main parameters, secondly, which comply with the minimal weight and size parameters, and, lastly, which cost the least. For stationary RM means, weight and size constraints are practically absent, and therefore the best technical parameters can be achieved via RM means. Thus, to ensure a large operating area for stationary posts, an antenna system located on remote masts can be applied, which can then be mounted on high buildings or in elevated areas. For mobile RM means, which are located on a vehicle or air transport carrier that is able to execute the main function while moving, it is important to take into consideration any constraints on weight, dimensions, and power consumption. This relates to the dimensions and carrying capacity of the carriers themselves, as well as the power capacity and the power of the sources located on the transport carriers. Since the above-mentioned constraints are not very strict, in the mobile RM means family, similar to the stationary means family, one can use multi-channel digital panoramic receivers to obtain high values on the dynamic range, the rate of panoramic analysis, and on received-information processing. Portable RM means are intended for transportation by one or a number of operators and are destined for further operation at stationary or temporary posts equipped or not equipped with power sources, and in the field. There is no requirement for these means to function during transport. Thus, serious constraints are formulated as to the weight, power consumption, and dimensions of the detection and

Radio Monitoring Equipment Design Philosophy

9

direction-finding antenna system. Moreover, for portable RM means, it is necessary to have an autonomous power source to provide for its function (e.g., accumulator charge, unfolding solar batteries, fuel supply for gasoline-electric generators, etc.). Hand-held RM means are intended, first of all, for operation during operator movement when placed on the operator’s body (or in his arms). Additionally, these means can be used to solve RM tasks at temporary or stationary posts. From the point of view their application, these means are universal and their usage is appropriate to detect RES locations in out-of-the-way places or where concealed operation is needed. Due to the serious constraints of energy consumption and weight and size parameters, such means parameters should be selected taking into account the unit’s operating life with a single power source set. Measuring radio-receiving devices and antenna systems are required for the measurement of regular radio electronic means (REM) parameters at the emission control of the officially-registered communications equipment, and also to estimate the effectiveness of information-leakage prevention measures at the boundaries of the controlled objects, and for SEE investigations. Usage of RM means for measurements can occur in the stationary, mobile, portable or hand-held versions. In Russia, the possibility of such usage for measurements must be approved by the respective certificates of Gosstandard and the Federal Service for Technical and Export Supervision of the Russian Federation. Therefore, we formalize the classification of all RM means into the following groups: • Stationary RM means family • Mobile RM means family, mounted on vehicles, and on air and sea transport carriers • Portable RM means family, operation of which is provided only after its deployment at the temporal location posts • Hand-held RM means family, for concealed and open operation, intended for operation while the operator is moving (without operator participation in the mean control, or with partial or complete participation) • Measuring means, to ensure effective control of the attempts made at information-leakage prevention, and also to measure the emission parameters for regular radio facilities. In order to decrease the number of means, it is expedient to combine the first and second families (stationary and mobile) into one family, provided that the execution by mobile means equals all the functions of the stationary means, taking into account the constraints to the antenna systems and the electric power systems of mobile means.

Radio Monitoring Equipment Design Philosophy The main purpose of RM equipment development is the creation of universal hardware-software systems, using the limited range of devices to carry out the

10

2

Problems, Classification and Structure of ARM Equipment

maximum possible RM task scope [4, 5]. The main requirements for RM equipment, aimed at the minimization and unification of the equipment and software, are the following: • Universality and multifunctionality of basic RM equipment for each family • Universality and multifunctionality of the additional means • Provision for the combined operation of the family’s basic equipment with the additional equipment, common for all families of RM equipment • Unification of the different families’ equipment • Unification of software, using similar modules, data formats, and interface formats for the different families • Unification of power supply • Effective distribution of processing tasks between the hardware signal processors and the controlling computer • Creation of code libraries for the basic set of each family’s equipment • End-to-end solutions for electromagnetic compatibility problems. A partial decrease the amount of necessary RM equipment can be achieved at the development stage of each family’s equipment, based on the functionally-modular principle of combining each family’s basic equipment with the additional means common for all RM families’ equipment. Investigations of the various types of digital receiver structures with wide operating frequency ranges show that minimizing the number of means can be achieved by restricting the operating range of the family’s basic equipment combined with the additional means common for all RM families. Implementation of this principle allows the selection of a fixed, basic equipment structure for each family. Another argument in favor of this principle is the consideration that, at present, technologically, the implementation of all or most of the functions mentioned in section “Introduction” into one constructively-completed mean would lead to an unreasonable increase of weight, dimensions, power consumption, and cost. Realization of this multi-functionality principle assumes that it is possible to reduce the structure of RM equipment, based on hardware digital-unit usage, with the possibility of quick reprogramming to execute the various signal processing algorithms, to combine the functions of separately-manufactured devices, and to effectively distribute problems between two software layers, namely, those used in the hardware digital unit and in the controlling computer. An end-to-end solution to the power supply problem assumes the unification of voltages supplied, including, in the equipment structure, any units that provide power supply from the AC network, from the onboard network of the mobile vehicle (car, helicopter, etc.), as well as from the battery with the charging device for autonomous operation, and for fail prevention at any supply interruption. The following principles defining the RM equipment structure are: • Unification of different families’ equipment; possibility of combining the equipment of various families, for example, combining the radio signals analog-digital

Radio Monitoring Equipment Design Philosophy

11

converter (ADC) of the mobile unit with the double-channel or single-channel unit of the analog-digital processor of the portable family • Unification of software packages, application of a similar data structure and format to achieve the possibility of using the same software package (with various drivers) within the different families • End-to-end solution of the electromagnetic compatibility problem, accounting for the carrier’s electric equipment. Minimization of the total expenses spent on RM equipment development relates directly to the possibility of its modernization during duplication. The open command library for each equipment type allows the possibility for the user himself to program and solve individual specific tasks, using the available RM equipment hardware. RM equipment development and usage experience shows that the equipment structure should include: • Single-channel or multi-channel (with coherently-related local oscillator) radio signal converter • Single or multi-channel analog-digital processing unit • Equipment for digital radio signals recording, at the intermediate frequency (IF), to magnetic or other storage devices • Equipment for real-time signal technical analysis and post-processing • Digital demodulation unit • Equipment for recording the demodulated signal simultaneously with the service signals (current time in the record moment, current frequency, etc.) • Power supply unit with reduced interference level • Universal control equipment allowing for the possibility of fast replacement and changing of modes, based on special mathematical software (SMS) program selection • Uniform SMS packages. Measuring radio-monitoring devices should be certified by the authorized, statestandard agencies. The additional equipment includes: • Wide-range unidirectional antennas of various applications • Antenna system sets for automated direction finding when moving, at stops, and for the stationary posts • Antenna modules sets with directional properties for hand-held direction finders of open and concealed application • Radio signals tuners, to widen the operating frequency ranges • Digital signal recorders • Equipment for positioning the RM means at the geographical coordinates.

12

2

Problems, Classification and Structure of ARM Equipment

Additionally, the following argument should be taken into account in favor of separate implementation of the basic equipment and the additional means. At the stationary and mobile RM posts, one can try, if possible, to move the receiving antenna to a very high place (on the roof, on the mast, etc.) in order to widen the post’s area of operation. In that case, the RF cable from the antenna to the basic equipment may be rather long. The losses and noise generated in the cable increase with the frequency. If all units, for example, in up to the 18 GHz range are concentrated in one place (say, on the operator’s desk), then this situation – even in the case of very good cable application – would lead to unjustifiable significant signal damping in the RF cable with receiving signal frequency increase and, hence, to a reduction in system sensitivity. Moreover, in this situation, the so-called antenna effect will reveal radio signal crosstalk to the RF cable and will distort the directional pattern of the antenna. Using an additional frequency converter, for example, in the 1–18 GHz range, provided that it is located near the antenna, essentially decreases the requirements for the upper boundary frequency of the RF cable, increases the sensitivity, and reduces the antenna effect.

Requirements for RM Equipment Technical Parameters Quality Criterion Selection It is nearly impossible to execute all the parameters necessary for complete optimization of all RM tasks, due to large number of parameters. Nevertheless, it is evident that, for most executed tasks, there is a common approach. This approach consists of estimating the necessary RM equipment using an “effectiveness-cost” criterion. Under this method, the area of possible decisions is restricted as follows: • Minimal number of important parameters is defined for each task or group of tasks • Permissible (or acceptable, in the absence of clear recommendations) limit for each parameter is fixed. In a number of cases, the probability P(t ≤ Ts ) of appropriate RM task execution during the operating time interval, not exceeding the given value of signal time Ts , can be successfully used as the main index of ARM equipment effectiveness. In this instance, the important technical parameters of the equipment should not be worse than required. RM equipment can be considered optimal when it provides the greatest probability of task execution during the same time at the same cost. At the same time, at the selection of the specific equipment by the user, other indexes can be the most important, for example, the accuracy of the current frequency measurement or the accuracy of the direction finding, as well as the equipment cost.

Requirements for RM Equipment Technical Parameters

13

Let us use the probability criterion to estimate the performance of a variety of equipment, for the task of signal detection. We consider detection probability functions at the panoramic spectral analysis, under the assumption that the radio signal has the duration of, say, 3 s. This time interval is the typical average value at the radio interchange. The probability of single-frequency signal detection with Ts duration, under assumption that the signal/noise ratio (SNR) is high, is defined in Chapter 4. Let us assume that the scan range is equal to 1,800 MHz. The calculation results for the case of single-frequency REM detection are shown in Fig. 2.1. In Chapter 4, the suggested probability criterion is used for more complex cases of signal detection, for example, signals with programmable operating frequency tuning (POFT). Pdet 0.8

0.6

0.4

0.2

0

400

800

1200

1600 g, MHz/s

Fig. 2.1 Detection probability of single-frequency signal vs. a function of system performance during time interval T=3 s. Search range is 1,800 MHz at the analysis bandwidth 2 MHz

Fig. 2.1 shows that, at search range 1,800 MHz, the continuous wave (CW) signal with duration 3 s can be detected with the probability P = 0.5 at the system performance 300 MHz/s, and with unit probability beginning with the panoramic analysis speed 600 MHz/s. Plots of the “new” signal detection for the panoramic analysis rate of 1,500 MHz for several values of the search range are shown in Fig. 2.2. At this given rate, the CW signal for the maximum search range of 3,000 MHz is positively detected in only 2 s.

Main Technical Parameters of RM Equipment The basis of any RM equipment is the panoramic radio receiver, executing the functions of panoramic analysis and signal detection during its search.

14 Fig. 2.2 Detection probability of single-frequency signal vs. time at performance of 1,500 MHz/s, bandwidth 2 MHz at the various search range (1 – search range is 300 MHz; 2 – search range is 900 MHz; 3 – search range is 2000 MHz; 4 – search range is 3000 MHz)

2

Problems, Classification and Structure of ARM Equipment

Pdet

2

0.9

1

3 4

0.7

0.5

0.3

0.1 0

0,5

1,5

2,5

t, s

As mentioned above, the probability P(t ≤ Ts ) of the appropriate RM task during the definite signal time interval Ts can be used as the RM equipment performance index. At that time, the main parameter values of this equipment are fixed and recorded. At detection-problem solution, this probability depends mainly on panoramic spectral analysis speed, which is ensured by the radio receiver (RR). However, the analysis speed cannot be examined separately from the other receiver parameters: the dynamic range on the intermodulation of 2nd and 3rd order, the frequency resolution, the sensitivity, the operating frequency range, the simultaneous bandwidth, the frequency stability, and the spurious rejection. Thus, the needless increase of the resolution can essentially reduce the detection possibilities of a fast radio signal with the dynamic time-frequency distribution. At present, the panoramic digital RR’s (DRR) have the widest application in the area of RM tasks. DRR is the combination of radio signal frequency converters with the fixed IF and the analog-digital processing unit, which provide the parallel signal processing within the simultaneous bandwidth with the necessary frequency resolution [6]. This implementation provides the maximum operating rate, however, it is necessary to take into account that the simultaneous bandwidth growth at high range load leads to ADC overloading, and the application of the attenuator leads to weak signal suppression, i.e., to reducing its electromagnetic accessibility zone. The solution for this situation is the usage of frequency selection sections (so-called “comb” sections) adjoined to each other. For example, to obtain the simultaneous bandwidth equal to 80 MHz one can use 8 sections of 10 MHz; however it essentially complicates signal processing and increases the cost. The methods for creating RM multi-channel panoramic DRR are discussed in Chapter 5. The lowest frequency range boundary in RM applications is usually equal to 9 kHz for both Russian and foreign RM equipment. The upper frequency range boundary for the RR equipment base is equal to 3 GHz; it can be extended to 6, 8 or 18 GHz, with additional devices, and, at that point, the trend of upper range boundary growth is steady, as mentioned above. In any case, the implementation of the principle formulated above ensures minimum expenses on the existing equipment modernization. According to this principle, the basis DRR and the additional devices are provided in order to increase the upper boundary of the operating range.

Requirements for RM Equipment Technical Parameters

15

Nowadays, the dynamic range of 70–80 dB and resolution of 6–25 kHz are considered sufficient for RM equipment. This corresponds to 3–12 kHz spectrum discretization. The spurious rejection should be not less than 70 dB, the relative frequency stability of the reference oscillator should be not worse than 10–6 –10–7 . When necessary to obtain the better frequency stability, for example, for measuring equipment, it is possible to use the highly stable external or internal reference frequency oscillator in DRR. Table 2.1 presents the typical tactical technical characteristics of stationary and mobile RM equipment, manufactured by one Russian company.1

Table 2.1 Typical tactical and technical characteristics of stationary and mobile RM equipment

Characteristic

Stationary RM station Mobile station with antenna system with mast-mounted Vehicle-mounted Mast-mounted antenna system Panoramic spectral analysis (PSA)

Operating range, MHz Basic equipment With optional equipment PSA rate in the operating range, MHz/s For medium performance For high and ultra-high performance, MHz/s Frequency sampling, kHz For medium performance For high performance, MHz/s Dynamic range, dB Sensitivity, μV

25–3,000

25–3,000 0.009–18,000

100–1,000 More than 1,000

3 6–12 75 Not worse than 3 Direction finding

Rate in the range, MHz/s For low radio range load For high load Signal bandwidth, MHz Sensitivity, μV/m Instrumental accuracy (RMS), degrees Technical analysis Analysis bandwidth, kHz/resolution, Hz

1–12 1

50–100 More than 300 Arbitrary 2–15 1.5

2,000 (5,000)/15; 250/500, 120/240; 50/100; 9/20; 6/12 Multi-channel radio monitoring

No. of monitored channels: For low range load For high load

1 IRCOS

(www.ircos.ru)

3–15 1.5

2–4 6–8

16

2

Problems, Classification and Structure of ARM Equipment

Characteristics of RM Equipment Families Radio Monitoring and RES Location Detection Systems The required terrain coverage and RES location detection can be achieved using a system of distant RM and direction-finding (DF) stations (the central and several peripheral stations), which provide detection and signal-receiving by the central post and also the simultaneous (synchronous) direction finding by the central station command, as well as RES location calculation with representation on the map [7, 8]. The number of stations for the stationary system is defined by the relief, the possibility of using high-rise buildings for antenna mast-mounting, by the controlled RES power and the detection and DF equipment’s sensitivity For locations where it is difficult to receive signals via stationary stations, mobile RM stations shall be used as additional support for stationary RES detection and direction-finding systems. These mobile stations are intended for more accurate localization of the detected RES. Handheld direction-finding and manpack RM equipment shall be provided for RES localization inside of buildings and in places that are beyond the reach of mobile equipment. The same regularities are true for the mobile system of detection-finding and RES location detection, with the only peculiarity being that the antenna system is mounted on the remote mast with less length, in the flat country, which leads to a reduction of the REM-monitoring operating zone. The antenna system dimensions (diameter) for the mobile equipment, evidently, will be less than the appropriate dimensions of the stationary system, due to application conditions, which leads to less direction-finding accuracy at the low section of the operating range (less than 100 MHz). The portable system can be characterized by the more strict limitations on weight, dimensions, and power consumption, which inevitably adversely affect the performance and the functions. The necessity of operation under field conditions requires autonomous power-supply means in the system structure. Similar system creation tasks are discussed in detail in Chapter 9.

Stationary and Mobile RM Stations Possible organization of the stationary and mobile RM stations includes the following posts, along with the handheld direction finder or the manpack RM unit. • Post No.1. The direction finder with the stationary (Fig. 2.3a) antenna system or with the mast-mounted and vehicle-mounted antenna system (Fig. 2.3b). • Post No.2. Panoramic radio receiver. • Post No.3. Multi-channel panoramic radio receiver. • Post No.4. The cartographic and RES location-calculation equipment.

Characteristics of RM Equipment Families

a)

17

b)

Fig. 2.3 Stationary (a) and mobile (b) RM stations

a)

b)

c)

d)

Fig. 2.4 Single-channel (a), double-channel (b) and multi-channel (c, d) DRR

The single-channel DRR structure used in RM stations (Fig. 2.4a,b) is: • • • • • •

Wide-band RM antenna Tuner Analog-digital processing unit Control system with customized software package Power supply from the vehicle’s on-board power or from the AC net Additionally, the measuring antennas set and the frequency converter for the frequency range extension. The multi-channel DRR structure (Fig. 2.4c,d) is:

• • • • •

Wide-band RM antenna Multi-channel panoramic DRR with remote control Control system with customized software package Power supply from the vehicle’s on-board power or from the AC net Additionally, the frequency converter for the frequency range extension.

The parameters, the structural diagrams, and DRR examples are considered in Chapter 3, and the examples of multi-channel DRR are given in Chapter 5.

18

2

Problems, Classification and Structure of ARM Equipment

Portable RM Equipment As previously mentioned, portable equipment is mainly purposed for radio monitoring at temporary and stationary posts, as well as in open terrain and in out-ofthe-way places, where mobile and stationary equipment usage is impossible. The portable equipment functions for executing RM tasks in the field should correspond, if possible, to the functions of the stationary or mobile RM stations. It is clear that the direction-finding antenna system of the portable equipment will never be completely equivalent to the stationary or mobile antenna system. Taking into account the restrictions on power consumption and the strict weight requirements, the portable, automated, direction finders are developed, which are described in Chapter 8. RM system structure for RES location detection, consisting of portable stations, (Fig. 2.5) is: • Three or more portable stations (central and several peripheral) • Handheld direction-finding equipment of open or concealed application • Manpack RM unit. In order to achieve multi-functionality, additional tasks related to CEE, may be entrusted to portable RM equipment. Possible versions of such units are described in Chapter 11. The family of CEE detection equipment (Fig. 2.6) should include the following: control equipment for one or several premises and mobile RM, direction-finding and emission parameters measurement stations for the controlled zone boundaries. The problems of CEE detection are discussed in Chapter 11.

Fig. 2.5 System comprising the portable stations

Characteristics of RM Equipment Families

a)

b)

19

c)

Fig. 2.6 Portable ARM units, with: single-channel panoramic DRR (a); double-channel DRR (b); and multi-channel panoramic DRR (c)

Certified RM equipment with measuring antennas and additional facilities can be used together with the appropriate customized mathematical software for special CEE investigations. Chapter 12 is devoted to these applications.

Manpack ARM Equipment Implementation options for open and concealed RM applications relate mainly to manpack equipment. On the basis of our experience with communications surveillance services in Russia, we can affirm that the following manpack equipmentimplementation options are desirable [9]: in the document-case, in the handbag (rucksack) or in the multi-pocket vest (Fig. 2.7). If possible, manpack equipment should have the functions of stationary or mobile RM equipment. Taking into consideration the need to minimize equipment weight and dimensions, as well as the variety of application conditions and the absence of strict requirements for direction-finding accuracy, it is reasonable to use the amplitude

a)

b)

c)

d)

e)

Fig. 2.7 The implementation options of RM manpack equipment: RM equipment in the document-case (a); in the multi-pocket vest (b); RM handheld equipment for open applications (c); concealed direction finder (d); automatic direction finder (e)

20

2

Problems, Classification and Structure of ARM Equipment

method of direction finding, based on the directional antennas in the manpack equipment. The manpack direction finders may have both open and concealed application options. The RES search process using manpack automatic direction finders differs from the similar process using mobile facilities, by convenience mostly, since it is possible to use manpack automatic direction finders in places that are inaccessible for portable and mobile facilities. In such instances, the fundamental method for RES position determination – the “homing” method – is based on operator motion, with the manpack direction finder in the RES position area, along the bearing direction. When the distance to the RES decreases, the direction-finding signal amplitude increases, an additional sign that the direction finder is moving in the right direction. The structure of the manpack automatic direction finder is close to the portable facility structure. With the help of handheld direction finders, the RES search process is provided by means of some basic stages, which are as follows [10]: • Fast panoramic spectral analysis in the given (operating) range and the detection of “new” signals • Qualitative or quantitative estimation of the detected emission parameters • Obtained parameters comparison with the database and the determination of RES value (danger) • RES location detection is an iterative process of operator-executed RES direction-finding stages, where the estimate of each stage’s level, its comparison with the previous iteration level, and the operator’s choice of movement direction, with the equipment, for the next iteration, are fulfilled. The handheld direction finder consists of: • Exchangeable directional antennas for open application and the indicator • Exchangeable directional antennas for concealed application and the control panel • Panoramic digital radio receiver • Additional facilities providing the signal level indication, level variation, signal demodulation and audition, and (when necessary) the operating frequency range extension • Power source from the autonomous accumulators, car power net and AC net, as well as the accumulator re-charge. The problem of field strength measurement using manpack equipment can be solved by using the measuring antennas only, and including required masts or tripods for their mounting, as part of the equipment structure, if necessary. To reduce the equipment range, and to provide the unification of high-quality radio monitoring, the main technical requirements for all types of equipment, including the portable and manpack ones, should not differ greatly from the appropriate requirements for the stationary and mobile equipment. The requirements for weight, dimensions and

Conclusion

21

power consumption may be the exception to this rule. Therefore, it is expedient to compile the single-channel and multi-channel DRR from the unified modules, to put them in cases, in keeping with the main requirements for the tactical technical characteristics – and for multi-functionality – and to provide a power supply from various power sources (AC net, car on-board supply, and accumulators). Moreover, it is desirable to provide full-scale radio monitoring, with the manpack equipment in operation at the temporary or stationary posts. Such an approach to RM equipment development allows for a lightening of the workload, with regard to the interaction between the technical facilities for the various families, and provides for unified database formation, as well as for electrical and informational compatibility. Sensitivity and direction-finding accuracy are also the main parameters of the manpack equipment. Moreover, its weight and operation duration from the single power supply set are also important. For the manpack equipment, expert estimates concerning the needed accuracy of RES direction finding show that, for practical purposes, it is enough to have the angle error 100 –150 . Equipment sensitivity (across the field) defines the action zone size, which can sometimes influence the ability to safely execute the operation. With this aim in mind, it can be recognized as necessary to have the option of concealed equipment application. The sensitivity (across the field) of modern handheld direction finders, for open application, is equal to 5–25 μV/m in the frequency range of 25–3,000 MHz. In the opinion of the professional experts, this solves most of the problems. The weight of the equipment set should not exceed 5–10 kg, and the operating duration from the single power source set should be not less than 3–5 h.

Conclusion In the present chapter, ARM technical means are classified by territorial coverage zone, by application, by the character of the function, by the equipment performance, and by the construction constraints. It is expedient to divide the equipment range into the following families: • • • •

Stationary equipment Mobile equipment Portable equipment Handheld equipment for open and concealed application.

ARM equipment can be used for CEE measurements and investigations, in the presence of the authorized organization certificates and the additional facilities. This chapter proves the rationality of the approach that, in each equipment family, there is basic ARM equipment, the possibilities of which can be improved by the additional facilities that are mutual for all families. To reduce the ARM equipment set, it is necessary to use the programmable units of digital-signals processing, to share efficiently the problems that exists among the hardware and software means. It is desirable to have unified customized mathematical software packages and the

22

2

Problems, Classification and Structure of ARM Equipment

usage of the similar data structure and format to be able to use the same package (with various drivers) in all ARM equipment families. A unified set of hardware-software facilities is offered. It includes: • Single-channel or multi-channel frequency converter • Single-channel and double-channel unit of analog-digital processing • Multi-channel equipment for the radio signal digital record in the bandwidth of simultaneous analysis • Equipment for the real-time and post-processing of signal technical analysis. Digital demodulators unit • Equipment for recording the demodulated signal simultaneously with the service signals • Means for RM equipment localization, as per geographical coordinates • Power supply • Customized mathematical software for the solution of RM problems, suitable for all families. As the general estimation index for ARM equipment, the criterion “effectivenesscost” can be chosen as the most convenient criterion. At this time, equipment effectiveness is the best of all criteria at characterizing the probability of executing the appropriate RM problem solution during the fixed-time interval, under the condition of the presence of the essential additional parameters for the given equipment.

References 1. Poisel, R.A., Target Acquisition in Communication Electronic Warfare Systems. Artech House, 2004, 370 pp. ISBN:1580539130 2. Poisel, R.A., Modern Communications Jamming Principles and Techniques. Artech House, 2003, 502 pp. ISBN:158053743X 3. Rembovsky, A.M., Automated Radio Emission Monitoring – Problems and Facilities (in Russian). Special technologies. 2002. Special Edition, pp. 2–6. 4. Rembovsky, A.M., Combined Solutions of the Automated Radio Monitoring Problems by the Restricted Set of Facilities (in Russian). INFORMOST – Communication facilities No. 5 (29), Sept. 2003. p. 23–29. 5. Rembovsky, A.M., Search Facility Effectiveness Increase for Automated Radio Monitoring (in Russian). Special technologies. No. 4, 2003, pp. 40–47. 6. Ashikhmin, A.V., Sergeev, V.B., and Sergienko, A.R., Radio Receiver Front-Ends for Automated Radio Monitoring Complexes: Peculiarities, Solutions and Prospects (in Russian). Special technologies. 2002. Special Edition, pp. 57–64. 7. Rembovsky, A.M., Automated Radio Monitoring and Emission Bearing – Problems and Facilities (in Russian). Uspekhi Sovremennoi Radioelektroniki, No. 6, 2003, pp. 3–21. 8. Ashikhmin, A.V., Kozmin, V.A., and Rembovsky, A.M., Ground-based Mobile Complexes of Radio Monitoring and Direction Finding (in Russian). Special technologies. 2003. Special Edition, pp. 30–41. 9. Rembovsky, A.M., Problems and Facility Structure of Automated Radio Monitoring Facilities (in Russian). Special technologies. 2003. Special Edition, pp. 2–7. 10. Ashikhmin, A.V., and Rembovsky, A.M., Carried Direction-Finders for Emission Sources (in Russian). Special technologies. 2003. Special Edition, pp. 34–40.

Chapter 3

Radio Receiver Applications for Radio Monitoring System

Introduction Radio receivers are a system of interconnected units used for the extraction of energy from electromagnetic fields, as well as for the selection, amplification and conversion necessary to recover information from radio signals. The structural diagram of the radio receiver (RR) is shown in Fig. 3.1. Fig. 3.1 Structural diagram of the radio receiver

Radio receiver

Terminal unit

The receiving antenna executes the first main RR function: it extracts the electromagnetic field energy and converts it into the electric signal. The radio receiver fulfills the second main function: selection and conversion of the electric signal generated by the antenna, as well as its amplification. This conversion is executed in such a way as to ensure the normal operation of the terminal unit, which fulfills the third main RR function: the extraction of useful information from the received signal. Radio receivers, in turn, are sub-systems of the more complicated systems of communication, radio broadcasting, TV, radio navigation, radar, radio direction finding, radio monitoring, radio control, etc. At present, the direct conversion receiver and superheterodyne receiver (receiver with frequency conversion) are the most well known RR types. Structural diagrams of these receivers differ according to the structure of the radio frequency (RF) section.

Tuned Radio Receiver In tuned radio receivers (Fig. 3.2), the RF path contains the input circuit and the RF amplifier (RFA). In this case, all resonant circuits are tuned to the received signal frequency fs , on which the main pre-detection amplification is fulfilled. Signal and noise from the receiving antenna enter the input circuit. The input circuit is designed to match the antenna output and RFA, which ensures the main frequency A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_3,

23

24

3

Input circuit

Radio Receiver Applications for Radio Monitoring System

RF amplifier

Detector (demodulator)

Base-band amplifier

Terminal unit

Fig. 3.2 Structural diagram of the tuned radio receiver

selection and pre-detection signal amplification. RFA resonant circuits are tuned within operating frequency range. Since both high selectivity and amplification are usually required (RFA gain may have the order of 106 –107 ), several amplification cascades and resonant circuits may be required. Synchronous frequency tuning of all these units is not a simple task. In microwave range, it is difficult to match the RR bandwidth with the useful signal spectrum width for filtering a noise, which does not coincide with the signal frequency. The number of resonant circuits is rarely more than three or four, due to the design complexity of the tuning. Therefore, at the operating frequency fs , the amplifier can be non-stable, and its selectivity will be insufficient because the selective circuit bandwidth B with the quality-factor Q is related to its resonant frequency f0 = fs by the formula B = f0 /Q.

(3.1)

As the tuning varies, the selectivity and gain coefficient vary: at fs growth, the bandwidth B widens and thus the selectivity decreases. The detector (or the demodulator) extracts the message from the RF signal. A low frequency amplifier intensifies the message signal to the value necessary for normal operation of the terminal unit. The advantage of the tuned radio receiver is its simplicity and relatively low level of inherent noise. The crystal receiver is the simplest direct conversion RR, having the minimum number of functional units required for signal receiving: the antenna, the selective circuit, detector, and the terminal unit. Since there are no amplifying elements in this receiver, the noise immunity and receiving quality are small. These receivers are used restrictedly in microwave, millimeter and optical ranges. The reflex receiver is a type of tuned radio receiver in which the same amplifier is used simultaneously for both pre-detection and post-detection amplification. The main principle of the reflex receiver is as follows: firstly, the active element amplifies the RF signal, which is detected, and then the audio frequency signal enters the input of the same amplifier. To reduce the amplifying cascade number and to simplify the design, in the past, regenerative and super-regenerative amplifiers were widely used in tuned radio receivers. In regenerative receivers, the negative resistance, which partially compensates for the losses, is introduced into the resonant circuit, due to the positive feedback, and it increases the equivalent Q−factor and the gain coefficient. However, these receivers have low stability because they operate in a mode close to self-excitation. Thus, it is possible to expect generated oscillations penetration

Tuned Radio Receiver

25

into the antenna, and its emission leads to the interference amplification for another receiver, which is extremely undesirable from the point of view of electromagnetic compatibility (EMC). The super-regenerative receiver is a tuned radio receiver that contains the amplifying cascade with smoothly-controlled self-excitation. In the super-regenerative receiver, the positive feedback from the RFA varies periodically with some auxiliary frequency, which greatly exceeds the signal modulation frequency. At that, during part of the cycle, the introduced resistance becomes negative and the oscillations are excited in the resonant circuit. During the next part of the cycle, the oscillation break occurs. These oscillation amplitudes exceed the received signal amplitude for 104 times, and more. The intensity is proportional to the receiving signals acting on the resonant circuit, i.e., the generated signals are, in essence, the amplified signals. The super-regenerative receivers have somewhat better stability than the regenerative receivers. Their advantage is high sensitivity in the presence of a simple electric circuit. Their shortcomings are the signal distortions and intensive spurious emissions that do not meet EMC requirements. In the superheterodyne receiver, radio signal frequency conversion occurs, which is the linear spectrum transition into the range suitable for the received-signal processing. Such receivers found the most circulation. The feature that most distinguishes the superheterodyne receiver from the direct conversion receiver is the presence of a special cascade for the frequency conversion. The structural diagram of the superheterodyne receiver is shown in Fig. 3.3. The linear receive path contains the relatively wide-band tuning pre-selector operating at the signal frequency fs , and the intermediate frequency (IF) path,

Linear receive section Signal frequency section Preselector Input circuit

RF÷àñòîò amplifier û

Tuning

Terminal unit

Base-band amplifier

Detector (demodulator)

Fig. 3.3 Structural diagram of the superheterodyne receiver

Intermediate frequency section Frequency converter Mixer

Local oscillator

Intermediate frequency amplifier

26

3

Radio Receiver Applications for Radio Monitoring System

which operates at the fixed frequency fIF with the bandwidth corresponding the signal spectrum. The pre-selector, which consists of the input circuit and RFA, provides the preliminary amplification, necessary for the signal extraction, and the receiver selectivity over the spurious channels, mainly, over the image channel. The frequency converter is the unit executing the frequency conversion (down or up) and which contains the local oscillator (so-called, heterodyne) and a mixer. In this cascade, the high frequency oscillations are converted into oscillations of another frequency, so-called, intermediate frequency (high enough), which is constant for any frequency of the received signal. As a result of the conversion, the received signal spectrum undergoes the linear transition from one frequency range to another, usually to the lower frequency range. Compared to the tuned radio receiver, the superheterodyne receiver has the following advantages: high selectivity and sensitivity, selectivity and sensitivity persistence over the frequency range, and increased stability. The high selectivity of the superheterodyne receiver is provided by the filtering on the reduced intermediate frequency. As is well known, the selectivity depends on the relative detuning f /f0 , which, at non-variable absolute detuning f , increases with the frequency fall. Therefore, the selective properties of the oscillating systems are improved. The high sensitivity of the superheterodyne receiver is also the result of processing frequency decrease, because the IFA can have a rather large stable gain. The sensitivity and selectivity persistence over the frequency range can be explained by the IF nonvariability, which allows the amplifying and selective properties to be maintained almost without variations, for any frequency of the received signal. The increased stability of the superheterodyne receiver is provided as a result of amplification distribution between the various frequency paths: radio frequency and intermediate frequency. The reduction of the number of cascades, operating at the same frequency, decreases the amplifier self-excitation danger due to the feedback. The tuner operation considerably defines superheterodyne RR performance. First of all, the tuner affects such RR parameters as the frequency range. The given range coverage depends on the local oscillator (LO) operation, which should provide stable oscillation throughout the whole receiving frequency range. LO voltage amplitude within the range should stay relatively permanent because it defines the tuning parameters and, hence, the tuner transfer factor constancy over the frequency range. The IF value strongly affects the superheterodyne RR operation and its sensitivity, selectivity and bandwidth. High selectivity over the adjacent channel is provided at low IF. Good selectivity over the image channel is ensured at high IF. It is difficult to meet simultaneously the high selectivity requirements over both channels; therefore, IF selection represents the engineering compromise. Thus, several frequency tunings are applied in professional receivers. The first IF has the high value, to provide the selectivity over the image channel. The last IF usually has the low value, to provide the selectivity over the adjacent channel. When selecting IF values, one must take into account that the IF should

Main Radio Receiver Parameters

27

not fall in the receiving signal bandwidth. This means that the first IF value must vary, depending on the received signal frequency. Moreover, it is important that the LO harmonics do not fall into the IF bandwidth. In a number of cases, it is sufficient to have double frequency conversion (Fig. 3.4). In such superheterodyne RR, there are two tuners and two IFA. To ensure high selectivity over the image channel, the first IF is selected large enough; the high selectivity over the adjacent channel, as well as the narrow bandwidth, is ensured at the low enough second IF. Tuning

Input circuit

RF amplifier

Mixer 1

Local oscillator 1

IFamplifier 2

Mixer 2

IFamplifier 1

Local oscillator 2

Fig. 3.4 Structural diagram of superheterodyne linear path with double conversion

The serious shortcoming of the superheterodyne receiver is the appearance of spurious channels: some signals can enter the receiver through the antenna circuit and may cause the output signal to appear, even at useful signal absence for the tuned frequency. The spurious channel is the frequency band outside the main receiving channel, in which radio interference creates the response appearance, caused by its passage to the demodulator or detector input. The channels including intermediate frequencies, combination frequencies, and the frequencies which are less in integer times with respect to the RR frequency tuning, IF and image frequencies, can be attributed to the spurious channels.

Main Radio Receiver Parameters The most important parameters, defining RR usage effectiveness in RM systems, are the following: • • • • • • • •

Operating frequency range Amplitude-frequency response (AFR) Transfer function irregularity (ripples) Voltage standing-wave ratio (VSWR) at RR input Selectivity over the spurious channels Selectivity over the adjacent receiving channels Noise factor and RR limit sensitivity RR sensitivity at the demodulator output

28

• • • • • •

3

Radio Receiver Applications for Radio Monitoring System

Dynamic range and intercept points for 2nd and 3rd order intermodulation Threshold of the blockage effect appearance Crosstalk value Phase noise, stability and tuning speed of the synthesizer Weight and dimensions Manufacturing and exploitation complexity, cost.

Operating Frequency Range The operating frequency range of a radio receiver is the range of possible tuning frequencies within which the main RR characteristics are ensured. At smooth tuning, the range can be defined by the limit frequencies f0 min − f0 max . The relative range is characterized by the coverage coefficient kc = f0 max /f0 min . The measuring receivers, the spectrum analyzers, the selective micro-voltmeters are remarkable for large coverage coefficients.

Amplitude-Frequency Response of the Linear Receive Path Amplitude-frequency response (AFR) of the linear receive path is the frequency dependence of the through transfer function at fixed RR tuning frequency. An example of AFR is shown in Fig. 3.5. AFR can be evaluated quantitatively with the following parameters: the selectivity kf at the given detuning f (see Fig. 3.5); the bandwidth B1 at the given irregularity SB1 ; the squareness coefficient B2 /B1 at the given attenuation levels SB2 , SB1 . Fig. 3.5 AFR irregularity estimation for linear receive path

k0 SB1

B1

SB2

B2 kΔf 0

Δf f0

f

In practice, RR path transfer function varies with the frequency. An example of this function is shown in Fig. 3.6. Usually within the limits of one sub-range, the transfer function irregularity varies smoothly; but, at the interfaces between the subranges, the jump in the through AFR occurs, which corresponds to the commutation of RR paths at tuning, e.g., from the last frequency in the previous sub-range to the first frequency of the next sub-range.

Main Radio Receiver Parameters

29

k k max k0 k min

0

f1

f2

fK-2

fK-1

fK

RR tuning frequency

Fig. 3.6 AFR of RR path: f1 is lower frequency of the frequency range; fk is higher frequency of the frequency range

The transfer function irregularity Sk is evaluated by maximal relative deviation of the transfer function kmax from its average value k0 : Sk = 20 lg [kmax /k0 ]

(3.2)

Where Kmax =max(Kmax –K0 ;K0 –Kmin ). When RR is used for the signal level measurement, AFR irregularity defines a more strict (metrological) characteristic: the limit of the permissible relative error of level measurement, which is the maximal relative deviation of the measured value from its true value.

Voltage Standing Wave Ratio If the receiver input impedance (complex resistance) differs from the cable resistance of the antenna system, then not all power transferred through the cable will arrive at the receiver. Part of this power will be reflected backward. The reflected signal will be added to the incident signal when their phases coincide, and will be subtracted when the reflected signal acts in reverse phase. As a result, in the inlet cable, the set of voltage maximums and minimums enters the intervals equal to the half wavelength. Voltage standing wave ratio (VSWR) is the ratio of the voltage maximum to its minimum: S = Umax /Umin

(3.3)

Since Umax = Uinc + Urefl , and Umax = Uinc − Urefl , where Uinc is the incident signal voltage, and Urefl is the reflected signal voltage, Equation (3.3) can be rewritten in the form: 1 + Urefl /Uinc Umax 1+r S= = = (3.4) Umin 1 − Urefl /Uinc 1−r where r = Urefl /Uinc is the voltage ratio of the reflected and incident signals. Since the power ratio of the reflected Prefl and incident Pinc signals is r = 2

2 Urefl 2 Uinc

=

Prefl , Pinc

(3.5)

30

3

Radio Receiver Applications for Radio Monitoring System

The VSWR equation can be rewritten as Prefl /Pinc . S= 1 − Prefl /Pinc 1+

(3.6)

Figure 3.7 shows VSWR vs. the power ratio of the reflected and incident waves. If a receiver’s input resistance is purely active and equal to the wave impedance of the input cable, S = 1 and reflected power is absent. If the input impedance is not equal to the input cable wave impedance, VSWR is more than 1. If the reflected power is equal to 10% of the incident power, S ≈ 2; if the reflected power is equal to 25% of the incident power, S = 3. In practice, the value S < 3 is usually considered as acceptable for RR input cascades. Fig. 3.7 VSWR vs. the power ratio of the reflected and incident waves

S 5

3

1 0 0.05

0.15

0.25

0.35

0.45 Prefl Pinc

Main Channel and Spurious Channels The frequency bandwidth in which the received signal spectrum falls forms the main receiving channel. The frequency bands, which join to the main channel and which may be occupied by the outside (spurious) signal spectra, form the adjacent receiving channels. Spurious channel formation can be explained by the frequency conversion in the superheterodyne receivers. The frequency conversion process is the high frequency f0 voltage transformation (signal at the tuning frequency) into the voltage at another (intermediate) frequency fIF , without changing the modulation type and character. The frequency conversion is executed in the frequency converter and can be fulfilled both with down-conversion (fIF < f0 ), or with up-conversion (fIF > f0 ). Usually, the tuner includes the local oscillator, which is the low-power oscillator generating the oscillations with fLO frequency, and the mixer, in which the oscillations of signal and local oscillators are mixed and one of the combination frequencies is extracted, for example, by decreasing the frequency (Fig. 3.8): fIF = fLO − f0 .

(3.7)

Main Radio Receiver Parameters

31

Pre-selector AFR

0

IF section AFR

f0

fIF

fLO

f

f

0

fIF

Fig. 3.8 Received signal frequency shift

This combination frequency is indeed the intermediate frequency. Thus, in the tuner, the variation of the signal carrying frequency occurs without the distortion of the information carried with the signal. The frequency mixer creates the combination frequency spectrum, in response to the arrival of two or more signals of different frequencies. In the mixer, the signal spectrum transition occurs into intermediate frequency range without the amplitude and phase ratios of the spectrum components disturbance. However, in a similar way, the radio emission can be received for the frequencies falling higher LO frequency fIF = f0 − fLO .

(3.8)

Thus, the spurious channel with the receiving at image frequency fim fim = fs + 2fIF .

(3.9)

corresponds to the useful channel with signal receiving at fs frequency. All inherent parameters of the tuner for the image channel and the channel on the signal frequency are absolutely the same. Therefore, the image channel is the one of the most dangerous spurious channels (Fig. 3.9). When using the total (sum) frequency fIF = fs + fLO , the image frequency is

fLO fIF

0

f 0

fIF

Fig. 3.9 Image channel formation

fim

f

32

3

0

fsp

0

fIF

Radio Receiver Applications for Radio Monitoring System

f

f

Fig. 3.10 Direct leakage channel formation

fim = 2fIF − fs .

(3.10)

Another spurious channel, which is referred to as the direct leakage channel, is the channel whose frequency is equal to IF (Fig. 3.10). If the signal with IF frequency acts at the tuner input, signal direct leakage without frequency conversion occurs, but with the amplification in the tuner and in the cascades of IF path. It should be noted that the frequency of the direct leakage channel is constant and equal to IF, while the image channel frequency varies at each RR retuning. The direct leakage channel non-related to the frequency conversion is as dangerous as the image channel. However, when evaluating the danger of these spurious channels, we must take into consideration that the direct leakage channel with the chosen fIF is fixed, but the image channel moves after the useful signal channel. Therefore, the probability of interference passing through the image channel is greater than through the direct leakage channel. Measures against the spurious channels are possible in the circuits before the tuner only, namely, in the input circuits, in the pre-selector, or in RFA. To eliminate the interference from RES, the frequency of which is equal or close to IF, the special rejecters are often used at RR input (in signal frequency path). The channels whose frequencies differ from LO harmonics 2fLO , 3fLO ,. . ., kfLO by fIF value fsp = kfLO ± fIF ,

(3.11)

where k is any integer, are the spurious channels as well. The spurious channel at combination frequencies fcom is formed as a result of the interaction of the tuner spectrum components with LO frequency or LO harmonics (Fig. 3.11): mfcom = nfLO ± fIF ,

(3.12)

where m, n are any positive and negative integers. The frequency of the spurious combination channel is

Main Radio Receiver Parameters

33

fcom 2fLO fIF

0

f

f 0

fIF

Fig. 3.11 Combination channel formation

fcom = (1/m)fIF ± (n/m)fLO .

(3.13)

The interference squeal (at demodulator output) or the received signal “twins” appearance (on the spectral diagram) is the type of distortion related to non-linear processes occurring in the tuner. In addition to the intermediate frequency in the tuner output circuit, the combination frequencies close to IF can appear. Thus, if the receiving signal is at a frequency which is k times lower than the frequency of any spurious channel fs = fsp /k

(3.14)

interference from signal k th harmonic forms, due to distortions in the mixer. For the main receiving channel, the input signal frequency shift, by some testing step f , corresponds to the same frequency shift of the output signal to the value f . At this, the shifted sign depends on the specific LO frequency position, with respect to the received signal frequency position for the given tuning frequency. For the image receiving channel, the input signal frequency shift to some testing step f corresponds to the output signal inverse shift in frequency to the value minus f (see Figs. 3.8 and 3.9). This property can be used to distinguish the signal for the main and image receiving channels. For the spurious combination channels – the spurious channels at the frequencies, which are integer times less than RR tuning frequencies – for IF, image frequencies, the input signal frequency shift by some testing step f corresponds to the output signal frequency displacement by the value, multiple to f , proportionally to the values and signs of the coefficients m and n (respectively, to combination order). Transfer functions within the bandwidth of the main channel, the spurious channels at image and intermediate frequencies do not depend on the input signal level (provided that the testing signals are rather small and the overload mode does not occur). At that, the input signal level variation by some testing value U corresponds to RR adequate output signal level variation by the same value U. Transfer functions within the bandwidth of the spurious combination channels, of the spurious channels at the frequencies (integer times less than RR tuning frequency), of IF, of image frequencies, depend on the input signal level. Therefore,

34

3

Radio Receiver Applications for Radio Monitoring System

the input signal level variation, by some testing value U, corresponds to the inadequate RR output signal level variation kU, where k is the multiple depending on the input signal amplitude and its harmonics level, type of tuner non-linearity, value and sign of the testing impact U. Thus, there are many signals at the various frequencies, which are converted into the oscillations at the same IF, and, in the general case, the number of these signals is infinite. From this set, only one signal is useful, and others correspond to the spurious receiving channels; therefore, suppression of the spurious channels and direct leakage channel should be strictly regulated in RR technical requirements.

RR Selectivity Radio receiver selectivity is its ability to extract the useful signal, at which the receiver is tuned, from the spurious signals arriving from the antenna system. In most cases, the interference level in the receiving antenna exceeds the useful signal level, which emphasizes the particular importance of this RR characteristic. The receiver’s ability to extract the useful signal from the spurious ones is based on the usage of distinguishing features between useful and spurious signals, namely, the emission arrival direction and time of activity, the amplitude, frequency and phase. The first feature is used at the spatial selectivity, which is realized with the help of antennas with the sharp pattern. The second feature allows the fulfillment of time selectivity, which is the receiver opening only for the period of the useful signal action. The amplitude, frequency and phase distinction of the useful and spurious signals is based on the amplitude, frequency and phase selectivity, respectively. The frequency selectivity is of primary importance. It can be explained by the fact that, in radio communication systems, the signals differ in frequency and signal separation can be fulfilled with the help of the resonant circuits and filters. One can distinguish two types of selectivity: one-signal and real. The one-signal selectivity is defined by filters AFR of the RR radio frequency path without the non-linear phenomena account at the single input signal (or useful, or spurious). The one-signal selectivity is quantitatively estimated by the ratio of the testing signal level at the interference frequency to its value at the useful signal frequency, for invariable tuning and the same output voltage. This selectivity can be estimated as well by the ratio, showing how much larger the gain of the radio path or the gain of the receiver’s separate cascade is compared with the gain for the spurious signal. Measurement of the one-signal selectivity is used for performance determination at the small enough levels of the input radio signal, which allows for the avoidance of the influence of non-linear processes (e.g., caused by overload) on the measurement results. The circuits of automatic control (of frequency, gain, etc.) are disconnected during these measurements. If there are no defined frequencies, at which the selectivity should be measured, in the RR technical requirements, then it should be measured at the end frequencies

Main Radio Receiver Parameters

Signal generator

35

Attenuator

Selective microvoltmeter

RR

Fig. 3.12 Structural diagram for determination of one-signal selectivity

and in the middle of each sub-range. To determine the selectivity curve by the onesignal approach, one can use the structural diagram shown in Fig. 3.12. The general approach for determining the selectivity curve consists in the following. The receiver and signal generator are tuned to the required frequency. The output signal level and the additional attenuator attenuation value are set in such a manner that the non-linear phenomena are absent during the testing signal receiving at the tuning frequency. The receiver output signal level Un (as the normal level of the receiver output signal) and the receiver input signal level U1 are fixed. After that, the generator frequency is increased by some value f . Then, the level of the generator signal is increased till the value, at which the receiver output signal level will again be normal (Un ). We measure the receiver output signal level U2 at the second point of the measuring curve. The measurements are repeated, increasing the generator detuning f to the necessary value. After that, the researcher changes the generator detuning, by the same steps, into the range of frequencies less than the receiver tuning frequency. On the basis of data obtained, one can plot the selectivity curve (Fig. 3.13). This curve is used for the determination of the bandwidth, the selectivity curve squareness coefficient, and the attenuation in the adjacent channel. The selectivity for the image channel as well as for the direct leakage and the combination channels are determined for the large detuning. The signal attenuation in the spurious channel is the ratio of the input radio signal level, required for the given output signal level, to the useful radio signal level, necessary to obtain the same output signal. Thus, the frequency selectivity characteristic in the spurious channels defines the susceptibility level function for the spurious receiving channels, with respect to the testing signal frequency. The bandwidth B is the band limited by two frequencies, at which the signal level attenuation does not exceed the given limits.

Uosc

Noise level

fosc Fig. 3.13 Radio receiver selectivity curve

fsp

f0

fim

fcom

36

3

Radio Receiver Applications for Radio Monitoring System

The slope of receiver AFR decays depending on the selection filter complexity inside the linear receiving path, and shows the reduction speed of the transfer function outside the bandwidth. The decay slope can be measured in decibels/Hz, in decibels/octave, or in decibels/decade (the octave means the twice frequency variation, the decade means ten-time variation). Sufficient information about the selectivity at one-signal approach measurement can be obtained on the basis of an analysis of the frequency differences corresponding to the signal attenuation by 20, 40, 60, 80, and 100 dB, beginning from the end frequencies of the bandwidth. If the attenuation values obtained in such a manner are close to the lower and higher end of the bandwidth, one can indicate average values only.

Inherent Noise and Receiver Sensitivity The inherent resistance of antenna, its thermal noise and RR input circuits noise are the factors that affect RR sensitivity, i.e. RR ability to ensure that weak radio signals are received. The main influence is exerted by the noise appearing in RR input cascades because this noise is amplified in the same manner as the useful signal. It is well known that each conductor having the inherent resistance creates electrical fluctuations, i.e., a noise in the whole frequency spectrum. This noise is conditioned on the thermal motion of the electrical charge carriers. The random thermal motion of the charge carriers in the conductor causes the random electric potential between its ends. This electric potential oscillates around the average value equal to zero, and its average square is proportional to the absolute temperature. This noise is referred to as the thermal noise. The noise magnitude depends on the conductor ohm resistance, its temperature, and the bandwidth of the transmitted signal. Root-meansquare (RMS) voltage of the thermal noise Un (expressed in Volts) is defined as: Un =

√

4kTBR

(3.15)

where k = 1,38 ·10−23 J/K is Boltzmann constant; T is the temperature (in K); B is the bandwidth (in Hz); R is the resistance (in Ohm). The receiver is a system consisting of the active and passive elements set, and possessing an active resistance. The linear part of the radio receiver, from the input to the detector, can be characterized by the non-dimensional noise coefficient F, which shows how many times larger the input signal and noise power ratio Ps /Pn is than the output signal and noise power ratio Ps,out /Pn,out [1]: F=

Ps /Pn . Ps out /Pn out

(3.16)

At present, it is acceptable to use the noise coefficient, expressed in decibels. The noise coefficient in the ideal noiseless receiver is F = 1(0 dB), because the

Main Radio Receiver Parameters

37

signal and the noise are amplified in the same manner (with the similar gain). In real receivers, the noise coefficient is increased due to the inherent noise and, as a result, the output noise power grows and the output signal-noise-ratio (SNR) decreases. The output power Pn,out can be presented as a sum of two items: Pn G, caused by the amplification of the input (source) noise, and Pinh , caused by the inherent own noise, where G is the receiver power gain coefficient. In this case, Equation (3.16) can be converted in the form: F=

Pinh Ps (Pn G + Pinh ) =1+ . GPs Pn GPn

(3.17)

In order to be able to compare various receivers by their noise properties, the standard value of the resistor R thermal noise power at T = 293 K is used as the input noise power Pn = 4kTBR.

(3.18)

Sometimes, another temperature value (299 or 300K) is used and, at that, the numerical value Pn changes insignificantly. RR sensitivity evaluated by the signal power value Ps only, at which signal receiving is ensured, takes into consideration the RR amplification properties only. It can appear that, by means of the amplification increase, one can ensure the receiving of any arbitrarily weak signals. As a matter of fact, however, the receiver with the greatest amplification inevitably amplifies its own inherent noise, and that restricts its sensitivity. The ratio of signal power to noise power at the RR linear path output characterizes the SNR, which is often referred to as the discrimination coefficient q = Ps,out /Pn,out .

(3.19)

RR ultimate sensitivity is equal to the minimal input signal power Ps = Ps min for the discrimination coefficient q = 1. At that Ps min = FPn

(3.20)

Thus, RR ultimate sensitivity is proportional to the noise coefficient. Consistent reception of the useful signal is ensured at the considerable useful signal power Ps,out excess over the noise Pn,out , i.e., at the discrimination coefficient q > 1. The real RR sensitivity is estimated by the minimal input signal power Ps , at which the required value q > 1 of the discrimination coefficient is achieved: Ps = FPn q.

(3.21)

Let us obtain the calculation formula for determination of the real RR sensitivity. Assuming that the signal source has the inherent resistance Rss , as shown in Fig. 3.14, the noise power at RR input can be written as: Pn =

In2 Rin

=

Un Rss + Rin

2 Rin .

(3.22)

38

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.14 Equivalent circuit for RR sensitivity determination

Rss

Un

In

Rin

In order to ensure the maximal power in the load, we should satisfy the condition of the inherent source resistance Rss and the load resistance Rin equality: Rin = Rss , then Pn = Un2 /(4Rss ).

(3.23)

Assuming that the noise has a thermal origin and is defined by Equation (3.15), we can determine the real sensitivity of the receiver as: Ps = qFPn = qF

4kTRss B = qFkTB. 4Rss

(3.24)

Power sensitivity can be transformed into voltage sensitivity. For the matched load Ps = Us2 /4Rss and (3.25) Us = 2 qFkTBRss . It should be remembered that, at Rss = Rin , the voltage at RR input is twice as less than the voltage acting at the source output in the no-load condition. To calculate the radio equipment parameters, it is convenient to use the logarithmic unity. The noise coefficient expressed in decibels (noise-factor), is NF = 10 lg F.

(3.26)

The sensitivity, expressed in decibels with respect to mW (dBm), can be presented as: qFkTB = 10 lg q + 10 lg (1.38 · 10−20 T) + 10 lg B + NF. (3.27) Ps = 10 lg 10−3 Let us check how much RR sensitivity changes at varying ambient temperature. It is evident that, in the last equation, only the second item a = 10 lg (1.38 · 10−20 T) ◦ depends on the temperature. At T = 223 K (i.e., –50 C) we get a = −175.1 dBm; ◦ ◦ at T = 353 K (+60 C) a = −173.4 dBm. Thus, for temperature variation by 110 C, ◦ the sensitivity changes less than 2 dB. For room temperature T = 293 K (20 C), Equation (3.24) can be rewritten in the simplified form: Ps = Q − 174 + 10 lg B + NF, dBm

(3.28)

where Q = 10 lg q is the required SNR at RR output (discrimination coefficient) in dB.

Main Radio Receiver Parameters

39

For the ideal receiver without the inherent noise F = 1, and in 1 Hz-bandwidth, the threshold sensitivity, i.e., the sensitivity at output SNR Qout = 0 dB, is equal to –174 dBm. Using Equation (3.25), at the temperature T = 293 K and the input resistance Rss = 50 Ohm, we can calculate the voltage sensitivity, expressed in decibels with respect to μV (dB μV), as: Us = 20 lg (2·106 qFkTBRss ) = Q − 61 + 10 lg (B + NF).

(3.29)

For example, RR sensitivity for the bandwidth B = 10 kHz and the noise-factor NF = 12dB at the output SNR Q = 10 dB will be Us = 10 − 61 + 10 lg 1,000 + 12 = 1 dB μV

(3.30)

or in micro-volts Us = 1.08 μV.

Sensitivity Increase with the Help of Pre-amplifiers Any amplifier assimilates the noise signal as the input signal. At the cascade connection of the electronic units (cascades), each cascade amplifies both the signals and a noise, passed through the previous cascades, adding, at that point, its own inherent noise. Let us determine the noise factor of three cascades connected consecutively as shown in Fig. 3.15. In accordance with Equations (3.16) and (3.17), the total noise factor is Ps

Fig. 3.15 Consecutive cascade connection in RR

Pn

F=

F1, G1

Ps1 Pn1

F2, G2

Ps2 Pn2

Ps /Pn Ps (Pn2 G3 + Pinh3 ) = Ps3 /Pn3 Pn G1 G2 G3 Ps

F3, G 3

Ps3 Pn3

(3.31)

where G1 ,G2 ,G3 are power gain factors of the first, second and third cascades; Pn2 is the output noise of the second cascade; Pinh3 is the inherent (own) noise of the third cascade. Representing the second cascade output noise in the form of a sum of the inherent noise and the amplified input noise, and then, in the same manner, the output noise of the first cascade, we find F=

[(Pn G1 + Pinh1 )G2 + Pinh2 ]G3 + Pinh3 Pn G1 G2 G3

Pn G1 G2 G3 + Pinh1 G2 G3 + Pinh2 G3 + Pinh3 = . Pn G1 G2 G3

(3.32)

40

3

Radio Receiver Applications for Radio Monitoring System

From (3.17) we find Pinh = (F − 1)GPn .

(3.33)

Substituting Pinh into (3.32) we get F=

Pn [G1 G2 G3 + (F1 − 1)G1 G2 G3 + (F2 − 1)G2 G3 + (F3 − 1)G3 ] . Pn G1 G2 G3

(3.34)

Executing the cancellations, we fulfill the final formula for the noise factor for three consecutively connected cascades: F = F1 +

F2 − 1 F3 − 1 + . G1 G1 G2

(3.35)

On the analogy of (3.35), we can form the equation for the noise factor for an arbitrary number of cascades: F = F1 +

F2 − 1 F3 − 1 FM − 1 + + ... + M−1 G1 G1 G2 Gm

(3.36)

m=1

where M is the cascade number. This formula is called Früs’ noise equation [1]. From (3.36), we can see that the whole system noise is defined first of all by the first cascade parameters. The contribution of the other cascades can be practically neglected, if the first cascade gain is large. It should be noted that, in the general case, the noise factor and transfer function of each cascade will depend on frequency, i.e., they will have different values in various frequency ranges. This means that the specific calculations can be executed in definite frequency intervals only. As a rule, a RR is connected to an antenna system by the connecting cable. As any electric device with losses, the coaxial cable has its own noise level [1]. At room temperature, the noise factor of a coaxial transmission line is equal to the losses in it. When frequency grows, the losses in the coaxial cable increase. Figure 3.16 shows the plots of linear attenuation (over 1 m length) vs. frequency, for several types of coaxial cables manufactured in Russia. As we can see from the figure, the signal attenuation value in the cable and, hence, its noise factor increases with transferred signal frequency growth. Attenuation values for the flexible coaxial cables at 1,000 MHz frequency are within the limits 0.1–0.6 dB/m, but, at 2,000 MHz, frequency values are within the limits 0.2–1 dB/m. At sufficient cable length, the cable’s noise factor will be rather significant and that will decrease RR sensitivity. For example, if cable losses at a 2,000 MHz frequency are 0.5 dB/m, a cable 30 m in length will have the noise factor NF = 15dB. The first possible option to decrease the noise factor of the cable line is the usage of the cable with minimal losses. Unfortunately, such cable with minimal losses has a very high cost.

Main Radio Receiver Parameters Fig. 3.16 Coaxial cable attenuation vs. frequency

41

G, dB RK-50-2-22

0.8 0.6 0.4

RK-50-4.8-31(32)

RK-50-7-11

0.2 RK-50-7-34

0

RK-50-7-35

RK-50-9-11

600

1000

1400

1800 f, MHz

The second way is by cable length minimization or, in the ideal case, in RR mounting just near the receiving antenna. If the receiver has small dimensions, this problem can be essentially simplified, e.g., the direction finder’s receiver can be mounted directly on the antenna array basis (see Chapter 8). Finally, the third option to decrease the influence of the cable noise factor is the application of a low-noise amplifier (LNA) based just near the receiving antenna (Fig. 3.17). This LNA should have a noise factor not exceeding several decibels and also the required gain. Low-noise pre-amplifier

Cable line

Radio receiver

Fig. 3.17 Example of low-noise pre-amplifier application for decreasing the influence of the cable line noise factor

Let us examine the example. We assume that LNA is used with the noise factor NF1 = 4dB and the gain g1 = 30dB. The connecting cable has the noise factor NF2 = 10dB and attenuates the signal by g2 = −10dB. The receiver has the noise factor NF3 = 12dB. Let us transform these values to the absolute values. For LNA F1 = 10NF1 /10 = 104/10 = 2.51; G1 = 10g1 /10 = 1030/10 = 1000. Similarly, for the cable and the receiver, we get: F2 = 10; G2 = 0.1; F3 = 15.85. Let us obtain the total noise factor F = F1 +

F2 − 1 F3 − 1 10 − 1 15.85 + = 2.512 + + = 2.67 G1 G1 G2 1000 1000 · 0.1

(3.37)

or in decibels NF = 4.3 dB. If we have no pre-amplifier, the total noise factor would be: F3 − 1 15.85 − 1 = 10 + = 158.69 (3.38) F ∗ = F2 + G2 0.1

42

3

Radio Receiver Applications for Radio Monitoring System

or in decibels NF ∗ = 22 dB. Thus, a pre-amplifier with the inherent noise factor NF1 = 4dB and with gain g1 = 30 dB increases the system sensitivity by = NF ∗ − NF= 22 – 4.3 = 17.7 dB. The question is: how to choose the LNA gain factor correctly, for the given noise factor? When increasing the LNA gain factor g1 , the total noise factor will asymptotically tend to its own noise factor value. Figure 3.18 shows the system noise factor plot versus the LNA gain factor for three types of the connecting cable with the noise factor 5, 10, and 15 dB. The other system parameters were kept unaltered. From these curves, we see that, when using the cable with the noise factor NF2 = 5 dB, the required LNA gain should be near 20 dB; for the cable with the noise factor 10 dB the required gain is 25 dB; and, finally, for the cable with the noise factor 15 dB, the required gain is 30 dB. Thus, it is evident that further LNA gain growth does not improve practically the system noise factor. NF, dB

Fig. 3.18 System noise factor vs. LNA gain factor

25

15 dB

20 10 dB 15 5 dB

10 5 0

5

15

25

35

g1, dB

If the wide-band signals have a large level and borrow the wide frequency band, the pre-amplifier can be overloaded. Hence, the main attention should be placed on its linearity, especially, if there are no pre-selection filters in its input. Moreover, in the measuring systems, LNA with the calibrated gain should be used to minimize the measurement errors.

Pre-amplifier Gain Factor Selection The dynamic range D of the receiver or its separate cascades is understood to be the ratio of maximally possible and minimally possible input signal levels. Usually the dynamic range is expressed in decibels as D = 20 lg

Uin max Uin min

= 10 lg

Pin max Pin min

= 10 lg (Pin max ) − 10 lg (Pin min ). (3.39)

The minimal level values are usually equal to the threshold RR sensitivity. The maximal values are defined by the acceptable non-linear distortion level at the output.

Main Radio Receiver Parameters

43

Let us return to the typical circuit of an antenna system connected to a RR by a connecting cable line. As shown in section “Sensitivity increase with the help of preamplifiers”, in order to reduce the harmful influence of the cable’s inherent noise, it is necessary to apply a pre-amplifier with a low noise factor, just after the antenna system. The gain factor growth decreases asymptotically the total noise factor of the system. For the hypothetical case, when the gain factor is equal to infinity, the noise factor of the whole system is equal to the noise factor of the amplifier. In the above-mentioned example, the LNA with its own noise figure of NF1 = 4 dB and a gain factor of g1 = 30 dB increased the system sensitivity by = 17.7 dB, i.e., it actually extended the system’s dynamic range into the small values range, for this value. On the contrary, with gain factor growth, the system’s dynamic range decreases into the large values range, by the difference between the amplifier gain factor and the value by which the dynamic range was extended in the small values range. For example, in the above-mentioned example, the dynamic range decreases by g1 − = 30 − 17.7 − 12.3 dB. As we can see from Fig. 3.18, beginning from the definite value, the gain factor growth does not practically decrease the noise factor. Hence, in order to avoid excessive dynamic range reduction, the LNA gain factor should not exceed some necessary value, enough for fulfillment of the required noise factor and system sensitivity. One can see from Fig. 3.18, that to establish the total system noise factor NF ≤ 5 dB, and the LNA gain factor g1 ≈ 20 dB, then, for the cable with NF2 = 5 dB, we have g1 ≈ 25 dB, and, for the cable with NF2 = 15 dB, we have g1 ≈ 30 dB.

Receiver Multi-Signal Selectivity Multi-signal selectivity describes the receiver’s ability to extract the weak useful signal in the presence of powerful disturbing signals falling outside the receiving bandwidth. The interference from these signals appears in the mixer. If the mixer fulfilled absolutely exactly the voltage multiplication operation for the signals from the receiver input and the local oscillator, no interference due to out-of-band signals would appear at all. Each input signal would create its own different frequency at mixer output, and the receiver’s multi-signal selectivity would coincide with the one-signal selectivity. Real mixers do not have this feature. Firstly, they mix up the different input signals so that the first one serves as the local oscillator signal for another, thus causing intermodulation noise. Secondly, they detect signals, which lead to crosstalk noise, namely, to modulation transition from the disturbing signal to the useful one. Thirdly, they detect the powerful input signal, which leads to the blockage, i.e., transfer function variation of the linear cascades.

Intermodulation Noise Intermodulation in the receiver is the interference occurrence at RR output when two or more disturbing signals act at its input, the frequencies of which fall outside the main and spurious receiving channels. These interferences are referred to as

44

3

Radio Receiver Applications for Radio Monitoring System

intermodulation. The reason intermodulation occurs is due to the amplitude nonlinearity of the transfer function of RF path elements. Gain-transfer characteristic (GTC) of the receiver or its separate cascades is referred to as the function of the output voltage amplitude (or the active value) versus the input sinusoidal voltage of fixed frequency. Figure 3.19 shows the GTC of the ideal path, by the dotted line, and the real path, by the continuous line.

Fig. 3.19 Gain-transfer characteristic of the path

Uout

Uout comp Uout max

III

II Uout min Un

I Uin 0

Uin min

Uin max Uin comp

The real path GTC can be divided into the following parts: Part I– the part of the signal and noise (interference) superposition (between Un and Uin min points); Part II– the linear part (between Uin min and Uin max points); Part III–the overload part (between Uin max and Uout comp ). GTC of the ideal and real paths coincide at the linear part II from Uin min to Uin max . In this part, GTC is the straight line, the slope angle of which defines the voltage gain factor of the path. In part II, at Uin < Uin min , the real path GTC does not pass through the coordinates origin. Even at Uin = 0 some voltage Un acts at the path output, which is caused by the action of the fluctuations and interference in the path. In part III, at Uin > Uin max , the real path GTC becomes detached from the ideal path GTC, which is related to the overload of the real path at large levels of the input signal. The condition Uin min < Uin < Uin max should be satisfied for normal path operation. Let us analyze the influence of the analog path transfer function non-linearity on the useful signal amplitude variation. Approximation of the path transfer function is quite complex, but the main nonlinear transformation regularities can be understood if we use a simple model, in the form of a non-linear two-port network, where volt-ampere (amplitude) characteristic, i.e., the function of output signal voltage versus that of input signal voltage, has a polynomial form: iout =

∞ k=0

bk ukin ≈ (b0 + b1 uin + b2 u2in + b3 u3in + b4 u4in + ...).

(3.40)

Main Radio Receiver Parameters

45

We shall be limited by the cubic polynomial for the analysis, the combination components occurring as a result of the non-linear transformation: iout =

3

bk ukin = b0 + b1 uin + b2 u2in + b3 u3in .

(3.41)

k=0

The two signal sum can be accepted as the input signal uin instantaneous value uin = u1 + u2 = U1 cos ω1 t + U2 cos ω2 t.

(3.42)

Substituting Equation (3.42) into (3.41) we get, after exponentiation: iout (t) = b0 + b1 U1 cos ω1 t + b1 U2 cos ω2 t + b2 U12 cos2 ω1 t+ +2b2 U1 U2 cos ω1 t cos ω2 t + b2 U22 cos2 ω2 t + b3 U13 cos3 ω1 t+ (3.43) +3b3 U12 U2 cos2 ω1 t cos ω2 t + 3b3 U22 U1 cos2 ω2 t cos ω1 t+ +b3 U23 cos3 ω2 t Using the known trigonometric relations cos2 α =

1 (1 + cos 2α); 2

1 [cos (α − β) + cos (α + β)]; 2 cos (2α−β) cos2 α cos β = 12 cos β + cos (2α+β) + ; 2 2 cos α cos β =

cos3 α =

3 4

cos α +

1 4

(3.44) cos 3α,

formula (3.43) can be presented in the form: iout (t) = b0 + b1 U1 cos ω1 t + b1 U2 cos ω2 t +

b2 U12 2

+

b2 U12 cos 2ω1 t + 2

+b2 U1 U2 cos (ω1 t + ω2 t) + b2 U1 U2 cos (ω1 t − ω2 t) + +

b2 U22 cos 2ω2 t 2

+

3b3 U13 cos ω1 t 4

+

3b3 U12 U2 cos ω2 t 2

+

3b3 U12 U2 cos (2ω1 t+ω2 t) 4

+

+

b3 U13 cos 3ω1 t + 4

3b3 U12 U2 cos (2ω1 t−ω2 t) + 4

+

3b3 U22 U1 cos (2ω2 t−ω1 t) + 4

b2 U22 2 +

46

3

+

Radio Receiver Applications for Radio Monitoring System

3b3 U22 U1 cos (2ω2 t+ω1 t) 4

3b U 3 cos ω t + 3 24 2

+

+

3b3 U22 U1 cos (2ω2 t−ω1 t) + 4

(3.45)

b3 U23 cos 3ω2 4 4

It should be noted that the cubic polynomial use for receiver path transfer function approximation allows the illustration of harmonics and new frequency components occurrence, but it does not ensure the correct calculation of these frequency components, corresponding to practical GTC. Nevertheless, the considered example shows that in the spectrum of the current, passing through the non-linear element – the characteristic of which is given by the third order polynomial – besides the components with frequencies ω1 and ω2 , the additional spectrum components occur, the frequencies of which are presented in Table 3.1. Table 3.1 Combination components Combination frequency order N

Frequencies

1 2 3

ω1 , ω2 2ω1 , 2ω2 , ω1 + ω2 , ω1 − ω2 3ω1 , 3ω2 ,2ω1 + ω2 , 2ω1 − ω2 , 2ω2 + ω1 , 2ω2 − ω1

Spectral component frequencies at non-linear element output are referred to as combination frequencies. The combination frequencies are described by the equation: ω = [n1 ω1 + n2 ω2 + ... + nm ωm + ...]

(3.46)

where ni are any positive or negative integers, including zero. The combination frequencies are usually grouped combining together all frequencies for which N = |n1 | + |n2 | + ... + |nm | .

(3.47)

The number N is referred to as the combination frequency order. There exists the following regularity [2]: the item with exponent N in the path non-linear transfer function causes the combination component occurrence with the maximal order equal to N. If N is an even number, the combination components of the even order occur: N, N − 2, N − 4 till DC component N = 0. If N is an odd number, the combination components of the odd order occur: N, N − 2, N − 4 till N = 1. Intermodulation characteristics are extremely important properties for quality determination because, in the most cases, the receiver must operate in a complex electromagnetic environment in the presence of powerful disturbing signals at different frequencies.

Main Radio Receiver Parameters

47

Output signal

Frequency 0 f2 – f 1

2f1 2f2 f1 f2 2f1 – f2 2f2 – f1 f2 + f1 Input signals Second order products Third order products

3f1 3f2 2f1 + f2 2f2 + f1

Fig. 3.20 Second and third order intermodulation products

Figure 3.20 shows the possible location of intermodulation components of the second and third order, occurring when two sinusoidal signals of a similar level act on the path input. We can see that the even order products are formed farther from the input signals on the frequency axis compared with the odd order products at the frequencies 2 f1 − f2 and 2 f2 − f1 . The parameters describing quantitatively the ratio of useful signal and intermodulation components are more important for the receiver.

Intercept Points on IP2 and IP3 Intermodulation At present, there are three approaches for determining the linearity of radio receivers. Approach 1. Measurement of intermodulation components (IC) of the third or second order, expressed in decibels, with respect to microvolt (dB μV), or in decibels, with respect to milliwatt (dBm). This is the level of disturbing signals acting on the receiver input, which causes the intermodulation components at its output, the level of which is equal to output signal, obtained at introducing the input signal corresponding to the receiver sensitivity level. Approach 2. Intermodulation factor (or the dynamic range on intermodulation) expressed in decibels. For example, if the intermodulation factor is not worse than 70 dB, it means that the disturbing signals should be at the least 70 dB higher than the useful signal in order to create output products of the same level. Approach 3. Intercept points on the intermodulation of third IP3 or the second IP2 order. Sometimes the TOI (third order intercept) abbreviation is used for the designation of the intercept point. Prior to the 1980s, the concepts of intermodulation factor or the dynamic range on the intermodulation were used as a rule for the description of intermodulation properties. However, later on, the almost ubiquitous transition to the intercept point concept was put into practice. This concept turned out to be more convenient due to its universality, since it unambiguously characterizes both the linearity

48

3

Radio Receiver Applications for Radio Monitoring System

and the dynamic range of the receiver (amplifier or any other non-linear circuit). The intercept point value is the rather permanent meaning for the specific receiver, in contrast to the IC level, which depends on the signal level at the input [3, 4]. Third order intercept point IP3 can be calculated for the assumption that the function of IC power of the third order versus the input signal power in the receiver follows exactly the cube law, i.e., at 1 dB increase of the input-disturbing signal level, the third-order products of intermodulation distortions increase by 3 dB. Actually, in accordance with Equation (3.45), the intermodulation third-order product has a level of 3b3 U12 U1 /4 . If two signals with same amplitudes U1 = U2 act at RR input, the output value of the intermodulation product will be proportional to the amplitude cube. In log-scale, this means that the output signal will grow three times faster than the input signal, i.e., the curve has a linear form with the slope 3:1. Second-order intercept point IP2 is defined under the assumption that the function of second-order products power versus the receiver input signals power follows exactly the quadratic law, i.e., at 1 dB increase of the input signal level the intermodulation distortion products of second order will increase by 2 dB. Actually, in accordance with Equation (3.45), the second-order intermodulation product has the level of b2 U1 U2 . If two components with the same amplitudes U1 = U2 enter at the receiver input, the output value of the intermodulation product is proportional to the amplitude squared. In log-scale, this means that the output signal power will grow twice as fast as the input signal power, i.e., the curve will have a linear form with the slope 2:1. At the same time, the useful signal growth at the receiver output should follow the linear law, i.e., at 1 dB increase of the input signal power the output signal power should increase by 1 dB. In log-scale, this means that the output signal power increases at the same rate as the input signal power, i.e., the function has a linear law with the slope 1:1. Figure 3.21 shows the output signal power versus the input signal power for the useful signal (P1 ), the intermodulation product of the second (P2 ) and the third (P3 ) order in log-scale. The third-order intermodulation product power P3 increases three times faster compared to the input signal power P1 . This means that these curves must have an intersection point. It should be noted that in reality the growth rate of these curves reduces as the input signal power decreases. The values do not tend to infinity since the real GTC has the form of the curve with saturation, as shown in Fig. 3.19. However, at small input signals, the curves have practically a linear character. If the curves are extrapolated by the straight lines, they will cross at IP3 point. Similarly, the second-order intermodulation product power P2 grows twice as fast as the useful signal power P1 . Accordingly, the hypothetical intersection point of lines P1 and P2 is referred to as the intercept point on the second-order intermodulation IP2 . In receiver specifications, the second and third-order intercept points are usually defined with respect to the input, i.e., by the input signal power, expressed in decibels with respect to milliwatt (dBm).

Main Radio Receiver Parameters

49

Pout,dBm

Extrapolated

IP2out

IP2 curves IP3

IP3out

Actual curves

P1dB

1dB P1dB

P1

1 dB-compression point

P3 P2 Pin,dB I P3in

I P2in

Fig. 3.21 Second and third-order intercept points

Intercept points IP2 and IP3 usage is the convenient engineering approach, allowing the quantitative estimation of the receiving path linearity, to determine the difference (in decibels) between the useful signal and the intermodulation component level. For example, let the value IP3 be 12 dBm. How much will the useful signal level exceed the intermodulation components level at the input signal power Pin = −10 dBm? The input signal is less than IP3 on input, by IP3 − Pin = 12 − ( −10) = 22 dBm. Hence, the useful signal power P1 is less than IP on the output, by 22 dBm, and the intermodulation components power P3 is less than IP3 on the output, by 66 dBm. The useful signal will be larger than the intermodulation components by 3 = 44dBm. In the general case, the useful signal excess over the third-order intermodulation components can be determined as: 3 = P1 − P3 = (IP3 − Pin ) − 3(IP3 − Pin ) = −2(IP3 − Pin ).

(3.48)

For the useful signal and the second-order intermodulation components, we get 2 = P1 − P2 = (IP2 − Pin ) − 2(IP2 − Pin ) = −(IP2 − Pin ).

(3.49)

For the useful signal and the n-order intermodulation components, we have n = P1 − Pn = (IPn − Pin ) − n(IPn − Pin ) = −(n − 1)(IPn − Pin ).

(3.50)

Moreover, knowing the intercept points, one can estimate the receiver dynamic range, free of intermodulation.

50

3

Radio Receiver Applications for Radio Monitoring System

Intermodulation-Free Dynamic Range Determination Usually, the full receiver dynamic range is defined by Equation (3.39), as the ratio of the largest to the smallest signals received by the receiver. The level of limited sensitivity Ps lim , determined by Equation (3.20), is accepted as the lower limit, and the higher limit is related to the receiver’s characteristics, describing its nonlinearity, such as, for example, the blockage level of 1 dB-compression point (P1dB in Fig. 3.21). The compression point is the absolute limit of the GTC linear part, and it is usually less than IP by 10–20 dB [5]. One additional receiver feature has the greatest practical importance, namely, the dynamic range on the intermodulation. This shows in which input-values range the receiver is able to operate without the intermodulation distortions. Let us look at the plot in Fig. 3.22. The third-order intermodulation dynamic range D3 can be obtained by determining the length of the vertical segment BC. The segment starting point – point B – is the intersection of the straight line P3 with the limited sensitivity level or with inherent noise Ps lim . Point C is situated on the straight line P1 . Since the straight line P1 has the slope 1:1, the triangle ABC is isosceles one. Therefore, the dynamic range D3 can be obtained as the length of the AB horizontal segment from the intersection point of the straight line P1 with the straight line Ps lim till the intersection point of the line P3 with the line Ps lim . As mentioned above, the straight line P3 has the slope 3:1, therefore, the segment BD is three times less than the segment AD. At the same time, the segment AD is equal to the segment DIP . Thus, the third-order intermodulation dynamic range is 2 D3 = (IP3 − Ps lim ). (3.51) 3 Similarly, we get the expression for the second-order intermodulation dynamic range: D2 =

1 (IP2 − Ps lim ). 2

(3.52)

P3

Pout,dB 20

P1

IP3

–20

C

–60 D3 –100

Fig. 3.22 Dynamic range of the third-order intermodulation

D3

A –100

–60 2/3

B –20 1/3

D Ps lim Pin,dB

Main Radio Receiver Parameters

51

In the general case, the dynamic range Dn on the n−order intermodulation is Dn =

n−1 (IPn − Ps lim ). n

(3.53)

where IPn is the intercept point on the n−order intermodulation. Recall that the limited receiver sensitivity can be obtained by Equation (3.20).

Attenuator Influence on the Intermodulation Value A serious question arises: Does the attenuator at the receiver input influence the intermodulation components value? Reducing the receiver input voltage by the attenuator, for example, by 1 dB, decreases the useful signal P1 by 1 dB also, but then the second-order intermodulation products P2 decreases by 2 dB, and the thirdorder products P3 decreases by 3 dB, as shown in Figs. 3.23 and 3.24. This property can be used for estimating the intermodulation component order. Fig. 3.23 1 dB attenuator influence on the second-order intermodulation

1dB

2dB

(f2-f1)

Fig. 3.24 The level variation of the third-order intermodulation components with 1 dB attenuator

f1 f2

(f2-f1)

f1 f2

1dB

3dB (2f1-f2) f1 f2 (2f2-f1)

(2f1-f2) f1 f2 (2f2-f1)

In the general case, by including the attenuator with the transfer function g1 = −A dB, n−order intermodulation components level will be decreased by nA dB. At that, intercept point value on input will be increased by A value, since all straight lines in Fig. 3.21will displace parallel to the right by A value. Since the intercept point displaces to the right by A when including the attenuator, it can appear that receiver linearity will increase. As a matter of fact, this is not so, because, at that point, the receiver’s limited sensitivity becomes worse by exactly the same A value.

52

3

Radio Receiver Applications for Radio Monitoring System

Determining the Intercept Points There are many approaches to determine the intercept points [3]. In most cases, these approaches are based on the measurement of the dynamic range on intermodulation Dn . In one such approach, used for determining third-order intermodulation distortions, two signals enter the input, as shown in Fig. 3.25. One signal displaces from the channel central frequency by 20 kHz, and another by 40 kHz. Third-order intercept point IP is calculated assuming that the third-order IC level depends on the receiver input signal levels following exactly the cube law, i.e., at 1 dB-increase of the disturbing signal level, the third-order IC increases by 3 dB, and the difference by 2 dB. Signal generator 1

Attenuator 1 Adder

Signal generator 2

Radio receiver

Spectrum analyzer

Attenuator 2

Fig. 3.25 Structural diagram of measurement to determine the intercept point on intermodulation

In the general case, to calculate the arbitrary order intercept points on input, one can use the following equations, obtained from (3.48) to (3.50): 3 + Pin 2

(3.54)

IP2 = 2 + Pin

(3.55)

IP3 =

IPn =

n + Pin n−1

(3.56)

where 3 , 2 , n are differences, expressed in decibels, between the useful signal power and the IC power of the third, second and n-th order, respectively; Pin is the input signal power.

Blockage Effect The useful signal blockage (compression) becomes apparent when reducing the gain factor in the input path of the receiver, or, in decreasing SNR, when disturbing signal action, the frequency of which is outside the main channel range. Blockage occurs in the RF path active elements (RFA and tuners), due to the non-linear law of the useful signal transfer function variation, simultaneously with the disturbing signal (Fig. 3.26). Useful signal blockage occurs in this or that receiver cascade if the amplitude signal transfer function has the saturation character, in which case it is as though the

Main Radio Receiver Parameters

53

Fig. 3.26 Explanation of the useful signal blockage effect in the receiver

ftun

fbl

ftun

fbl

output signal increment lags with respect to the input signal increment, in the wide variation interval. The blockage does not occur if the amplitude signal-transfer function has the linear character in the wide interval of the input signal variation. The channel, in which the blocking/disturbing signal acts, is out-of-band; the nominal frequency of such a signal may accept the different values within some frequency band, depending on the disturbing signal level and the RF path circuit’s selectivity before the mixer. At rather large detunings f of the disturbing signal, relative to the receiver tuning frequency, this signal is attenuated by the RF path resonant circuits. To describe the receiver property to detect the useful signal in the presence of the powerful disturbing signal, one may use the concept of the “dynamic range on the blockage” (in decibels): Db = 20

Umax b Us min

(3.57)

where Umax b is the maximal permissible voltage of the disturbing signal, corresponding to the blockage threshold; Us min is the minimal voltage of the useful signal, corresponding to the receiver sensitivity.

Crosstalk Distortions Crosstalk distortions are useful signal spectrum structure variations of the modulated disturbing signal, occurring at simultaneous impact to the receiver, the frequencies of which do not coincide with the main and the spurious receiving channels. Such a distortion-occurring process is defined by the non-linear variation of the amplitude signal transfer function in the RF path active elements. The “crosstalk distortions” concept refers to the useful signal with amplitude modulation (AM), when the modulation components of the disturbing AM signal occur in its structure (Fig. 3.27). In this case, the RF path non-linear element – just as for the blockage in RFA – can be presented by way of the power polynomial model, the only difference being that the input signal is imitated by the sum of two AM signals. In order to simplify the analysis, one can limit oneself to the third order polynomial, as at the blockage. After some transformations we can obtain the crosstalk

54

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.27 Explanation of the crosstalk distortions process

ftun

ftun

fctd

distortion factor. This factor represents the ratio of the spectral component level in the useful signal structure, occurring as a result of the crosstalk distortions, to the useful signal level, for the given parameters of the disturbing and useful signals. To describe the receiver property necessary to detect the useful signal in the presence of the powerful disturbing signal – up to the crosstalk distortion threshold – we can use the concept of dynamic range on the crosstalk distortions. The channel, in which the disturbing signal acts, and which creates the crosstalk distortions, is out-of-band; the nominal frequency value of such a signal may differ within some frequency band, depending on the disturbing signal level and on the RF path circuit’s selectivity before the tuner. As mentioned above, the “crosstalk distortions” concept is often attributed to the disturbing signal affect on the useful AM signal. Nevertheless, it can be more widely understood. The crosstalk distortions may become apparent in the form of the useful signal phase variations (distortions), i.e., in the form of interference in the phase (generally angle) modulation systems. In this case, the term “phase crosstalk distortions” is expedient. Phase crosstalk distortions are not essential for AM systems. They become apparent in the form of interference, in systems where the useful information consists in the phase structure of the received signal.

Phase Noise and Retuning Rate of the Panoramic RR Frequency synthesizers (FS) are used in receivers to vary the tuning frequency. In the general case, FS form the discrete frequency set by means of coherent frequency conversion of the single reference oscillator, usually crystal. The long-term relative stability of any frequency at the output of such a coherent synthesizer is equal to the long-term frequency stability of the reference oscillator. Summarizing the various FS versions, we shall list the most important FS characteristics. Then, we shall take into consideration the widespread case, when the oscillation from one set of equidistant frequencies occurs in the FS output, at each time moment. The main FS characteristics are: the output signal frequency stability, the operating frequency range, the discreteness of the frequency or phase retuning of the output signal, the output signal type, the level of spurious discrete components, the phase noise level, and the switching time.

Main Radio Receiver Parameters

55

The FS operating frequency range is defined by the operating frequency range of the receiver. If a FS is used as the first local oscillator, its frequency should differ from the input frequency of the receiver by the intermediate frequency value. Instead of the frequency range, one may apply the coverage factor of the operating frequency range, which is equal to the ratio of the maximal frequency of the operating frequency range to the minimal frequency of the same range. The frequency retuning discreteness (the step of frequency grid) is defined by the synthesizer’s purpose. At fixed-frequency communication and at the frequency jump tracking, the step of synthesizer frequency grid, used as a local oscillator, is defined by the step of the transmitter frequency grid (from hundreds of hertz to tens of megahertz). At the programmed Doppler shift compensation and with the use of FS as a tracking system (on the phase or in time), the required frequency grid step may be very low (till some hundredth and some thousandth part of hertz). In this, and some other, cases, the phase (not frequency) retuning discreteness of the output oscillation has a higher profile. As a result, the phase jumps are inadmissible during the transition from one frequency to another, and these jumps cannot exceed the definite value (usually some tenth part or parts of a degree). The output signal view is defined to a great extent by the signal processing character in the device in which a FS is used. For the synthesizer to be used as a local oscillator of the analog receiver, the sinusoidal form of the output signal is usually required. For a FS in digital-analog tracking systems, the pulse form is preferable. Finally, in the path with completely digital signal processing, a FS must generate the number sequence (codes) corresponding to the sinusoidal signal samples in the fixed (equidistant) time moments. In technical specifications, the phase noise spectral density is usually given in decibels, by hertz, with respect to carrier level at the given offset from the carrier frequency (or dBc, for example, –120 dBc/Hz at 10 kHz offset). The phase noise level of the FS output oscillation, with respect to the generated signal level (the carrier level), usually falls within –60 to –120 dB/Hz at 10 kHz offset from the carrier frequency. At frequency multiplication with FS, the spectral density increases proportionally to the multiplication factor, applying hard restrictions on the reference oscillator noises. The frequency switching time (permissible) varies widely depending on the synthesizer’s purpose. In particular, when a FS is used in a communication receiver at different frequencies, the switching time may be equal to one second. In that case, during the frequency switching, not only are phase jumps permitted, but the complete short-term miss of signal is permitted as well. On the other hand, at FS application in the phase-locked loop, and in some other cases, it is desirable to ensure the complete absence of the transients. For panoramic receivers, the frequency switching time defines the receiverretuning rate over the operating frequency range. For modern panoramic receivers, the switching time is equal to units of milliseconds. We should mention that the less the frequency switching time is, the more difficult it is to ensure the low phase noise level.

56

3

Radio Receiver Applications for Radio Monitoring System

The frequency accuracy of the receiver frec includes both the initial error of the given tuning frequency setting and the receiver tuning instability. The setting error depends on the setting method and the tuning frequency indication method, and the tuning instability depends on the tuning frequency offset due to the system warming-up, the climatic and mechanical impacts, the supply voltage variations, etc. The high frequency accuracy of the receiver is necessary for the preset communication entry, the communication maintenance without adjustment. For measuring receivers, frequency accuracy is the governing factor for the accurate measurements of the radio signal frequency. The technical specification may fix the receiver’s frequency accuracy, or, separately, the setting error and the tuning instability. In some cases, tuning instability is defined in parts, due to the necessity of separate calculation and testing: the selfwarming-up, the temperature variations, the shocks and vibrations, the variations of the supply power voltage. The absolute frequency instability of the reference oscillator (RO) is the oscillation frequency deviation f at its output during the definite time interval, caused by the external destabilizing factor influences, against the specified nominal frequency f0 : f = f − f0 . The strictest requirements for frequency accuracy are placed on the receivers, intended for radio signal receiving with single-sideband (SSB) modulation (frec = 5 − 10Hz) and the signals with differential phase modulation (frec = 0.5 − 1Hz). This follows from the fact that the relative frequency accuracy of the receivers should have the value 10−7 − 10−8 . When the receiver uses the frequency stabilization system with one RO, it complies with the requirements for relative frequency accuracy defined by this oscillator. The relative RO frequency instability is the ratio of the absolute frequency instability to the specified frequency: δ = f /f0 . To reduce from the total frequency error the error portion caused by the initial frequency setting inaccuracy, the possibility of RO frequency correction on the basis of the external frequency standard or the operation from the external RO, which is more accurate and stable, should be provided. Tuning instability of the receiver can also occur via the permissible variation of the receiver’s local oscillator frequencies in the time interval after the self-warmingup. The temperature coefficient of the local oscillator is the important parameter, i.e., the relative frequency offset at air temperature 1◦ C variation around the receiver. The long-term RO frequency instability is the total frequency deviation caused by its slow variation due to element aging, the external destabilizing factor influence. It is defined during the long-term period, namely, an hour, round the clock, a month, a year.

Digital Radio Receivers General Principles of Digital Radio Receiver Implementation The digital RR is the RR that performs the signal processing in analog and digital forms [6–9].

Digital Radio Receivers

57

The complete or partial digital signal processing (DSP) is executed in the digital RR (DRR), intended for receiving the analog signal. In accordance with DSP applications, DRR can be divided into two groups: • Receivers, where the received signal conversion into the digital form is not executed, and the separate units are implemented on the digital element base, e.g., the control devices, the monitoring devices, the information image devices, the systems for establishment of communication, the automatic control systems, the digital frequency synthesizers, etc. • Receivers, where the signal is converted into the digital form and part of the main receiving path is fulfilled on the digital element base, including the digital filter for the main signal selection, the digital demodulators, the digital devices for signal recognition and parameter measurement, and the auxiliary units. The first group of RRs is the most numerous, at present. The majority of modern, professional, RRs have separate units implemented on the digital base of their structure. However, receivers in the second group are also considered receivers, where the preliminary signal filtering, its amplification, and the frequency conversion to the intermediate frequency are executed in the analog domain; and, after that, the IF signal is subjected to the analog-digital conversion and all further signal processing is executed in the digital domain. Practically all, modern, RRs intended for radio-monitoring problems can be classified as the second group. The generalized structural diagram of DRR consists of five functional units, as shown in Fig. 3.28. The amplifying-converting path (ACP) accepts the signal from the antenna, provides filtering against the interference, displaces the input signal spectrum to the IF, at which time the analog-digital conversion is executed. The auxiliary units: the automatic gain control (AGC) system, the attenuators, the limiters, etc., which affect From reference oscillator Frequency synthesizer From antenna

Amplifyingconverting section

DSP section

Secondary power supply

Fig. 3.28 Generalized structural diagram of DRR

Control and display device

To customer

58

3

Radio Receiver Applications for Radio Monitoring System

the amplitude characteristics of the amplifying section, but do not introduce distortions into the received information, may be included in ACP structure. The main signal processing is executed in the DSP section. It includes the filter, defining to a considerable degree the noise immunity of the receiver, the demodulator, and the circuits of post-detector signal processing. The frequency synthesizer converts the frequency of the external or inherent reference oscillator and, from this signal, generates the frequency grid necessary for operation of the ACP tuners. The synthesizer permits the receiver to adjust on another input frequency. The separate synthesizers may belong to the tracking system structure. Moreover, the synthesizer can generate the frequency grid required for the DSP unit’s operation. In the autonomous mode, the control and display unit executes the given algorithm of the receiver operation (switching on, switching out, search and choice of the signal, adapting to the changing operating conditions, etc.) and allows the operator to control the receiver manually or automatically. The secondary power supply is intended for the energy conversion of the primary power source, e.g., stationary grid (220 V) or on-board grid, into the form suitable for application directly in RR.

Types of ARM Receivers In spite of the generality of these operation principles, it is possible to single out several characteristic types of RRs, as listed in Table 3.2. These types are used, at present, for ARM problems solving. High real sensitivity and selectivity, as well as the application of methods providing noise immunity and reliability under the conditions of strong pulse, fluctuation, and concentrated over the spectrum interference, are typical for scanning receivers. As a rule, the scanning receiver can be assigned to the first DRR group. The selective micro-voltmeter is a voltmeter equipped with an adjusted narrowband filter, and, thanks to this filter, the voltmeter can measure frequency band voltage, up to the separate spectrum components. The most advanced selective microvoltmeters differ from spectrum analyzers by the manual adjustment only, as well as by the absence of panoramic representation. Selective micro-voltmeters allow us to measure the signal level at the antenna path output in the given bandwidth for the wide operating frequency range. The spectrum analyzer (SA) is a universal measuring device designed for the investigation of the signal spectral structure and the measurement of its parameters. The spectrum analyzer structure coincides with the superheterodyne receiver structure. Depending on the input signal sensor type (with the appropriate matching circuits), the SA can be used in various areas of science and technology. In particular, when using the antenna (as the radio signal sensor), the SA plays the role of the panoramic receiver. Upon additional application of the antenna to the SA – the circuits of preliminary selection – such a SA is then capable of solving the problems of the panoramic measuring receiver.

Digital Radio Receivers

59

Table 3.2 Types of radio receivers used for radio monitoring Spectrum imaging Calibration Demodulator

Signal parameter

Defined by signal bandwidth for which receiverreception is designed (from hundreds of Hz to hundreds of kHz) Adjustable (from hundreds Hz to hundreds kHz)

Usually no

No

Yes

No

Usually no

Yes

Desirable

Yes

Usually adjustable

Yes

Usually yes

Desirable

Yes

Wide. Usually from hundreds kHz to tens MHz Adjustable (from tens Hz to tens MHz)

Yes

No

Usually yes

Usually yes

Yes

Yes

Usually yes

Yes

Preselector Bandwidth Scanning radio receiver

Yes

Selective Yes microvolt meter Spectrum Usually anano lyzer Panoramic Yes receiver

Panoramic Yes measuring radio receiver

In spite of all the mentioned SA advantages, its application for solving the ARM tasks is not always expedient, since the basic technology of the analyzer implementation is, as a rule, intended for realization of universal functional possibilities. In the basic configuration, a SA, as a rule, has no preliminary signal selection units, therefore, its application for ARM problems is rather difficult. The panoramic receiver is a RR with a wide bandwidth (from hundreds kHz to some tens MHz), capable of representing the signal spectral structure, and with the high speed of spectral analysis (from hundreds MHz to tens GHz per second). As a rule, the panoramic receiver has the demodulators of AM, FM, PM, and SSB signals and the interface for connection to a PC. The panoramic measurement receiver is a panoramic receiver with high metrological characteristics for the measurement of the level and other radio signal parameters. The measuring receiver is the “heart” of the advanced radio-monitoring unit. At present, the hardware of ARM equipment is limited by the antenna complex and the measuring receiver. As a rule, the set of peak, quasi-peak and RMS detectors designed for signal level measurements is included in the measuring receiver structure. The measuring receiver must provide the frequency resolution, from several hertz to tens of kilohertz, at the signal spectral analysis, and must operate under PC control.

60

3

Radio Receiver Applications for Radio Monitoring System

The main type of measuring receiver is the voltage and power measuring system for RF signals and noise, which operates for the given receiving antenna parameters as the field-strength meter. It is a superheterodyne receiver with a measuring device at the output, which has high sensitivity (up to 10–15 W) and selectivity (50–60 dB). As a rule, the measuring receiver is a multi-range receiver with bandwidth adjustment in the IF section. The linear receiving path operates to the linear and quadratic amplitude detector with the given time constants of the load circuit, which allows for the possibility to measure the averaged rectified, effective, and peak values of sinusoidal and noise signals.

Development of Russian Arm Systems First- and Second-Generation Systems Prior to the 1990s, automated radio monitoring systems used in Russia were based on foreign scanning receivers. In the first-generation systems, the foreign receivers were used without serious updating, but, in the second-generation systems, serious modernization in receivers was made, which allowed for the improved technical performance of ARM systems, in which they were applied. In the first-generation systems, the receiver update was reduced to the additional buffer arrangement, so that the analog IF outputs could pass the signals to the external digital processing unit. Moreover, a switch was added as an AGC circuit break, to ensure the possibility of operating in the multiple-pass panoramic coverage mode with the permanent path transfer factor. More serious updates were provided in the second-generation systems, including the arrangement of additional units of “fast,” properly developed and manufactured, synthesizers, permitting the time of receiver adjustment to the given frequency to be reduced substantially and the panoramic coverage rate to be increased.

Fig. 3.29 Central unit of ARK-PK3KU automated radio monitoring system (AOR-3000 radio receiver)

Development of Russian Arm Systems

61

At that time, the following ARM equipment manufacturing was launched, namely, the handheld direction finders, the multi-channel RM systems, the various ARM systems, the distributed systems of remote RM to distant premises and the reveal of the information leakage channel, multi-functional RM and DF systems, and so on. Figures 3.29, 3.30, 3.31, 3.32 and 3.33 show some of these systems.

Fig. 3.30 Central unit of ARK-PK3KU automated radio monitoring system (AOR-5000 radio receiver)

Fig. 3.31 Control unit and receiver of the handheld ARK-RP1 direction finder (AOR-3000 radio receiver)

62

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.32 ARK-KPC six-channel radio monitoring system (two IC-8500 radio receivers and four IC-PSR1000 radio receivers)

Fig. 3.33 ARK-RD4 four-channel radio monitoring system (four IC-PCR1000 radio receivers)

Radio Receivers of the Third and Fourth Generation Experience gained from application of the hardware-software ARM systems of the first and second generation showed the total accuracy of the selected hardwaresoftware solutions, but, at the same time, the following restrictions were discovered:

Development of Russian Arm Systems

63

• Large irregularity of the gain factor and the inherent noise within the operating frequency range, which could not allow the system to be applied for measuring purposes • Insufficient dynamic range in the wide-band path • Impossibility of optimal receiver control in the system structure, leading to operation rate reduction • Low tuning rate, which could not allow the 50 MHz/s limit to be overcome • Impossible operation in the RM structure required the wide-band IF output • Large number of staggered frequencies falling into IF bandwidth • Absence of double-channeled coherent receiving possibility from two receivers, without serious updating • Extent of the electromagnetic compatibility problems for the receivers and for other system units • Structural features of receivers, prohibiting application within the systems, due to the complicated requirements for the mechanical and climatic operation conditions. Taking these shortcomings into consideration, Russia decided to launch the development of a new DRR generation, which would ensure the following features in the interests of ARM systems: • Possibility for use in multi-channel coherent receiving systems • Dynamic range within wide-band path on the second- and third-order intermodulation not less than 70 dB • Retuning duration from one frequency to another not more than 15 ms for frequency setting accuracy 500 Hz • Digital processing bandwidth not less than 2 MHz • Spectral analysis rate not less than 100 MHz/s, for the spectral sample discreteness 3 kHz • Increased firmness to the mechanical and climatic impacts permitting DRR application in the equipment of various purposes based on the transport carrier • Minimization of the consumed power. As a result of the intensive development of new circuitry and design solutions, in 1999 a new Russian DRR of the third generation came into the world. One of its implementations is shown in Fig. 3.34. The developed ARK-CT1 receiver had the acceptable technical specifications, including dynamic range on the third-order intermodulation not worse than 70 dB, synthesizer switching time of about 10 ms, which, at 2 MHz bandwidth, enabled a digital spectrum analysis rate of 140–150 MHz/s at the spectral sample discreteness 3.125 kHz. The irregularity within the receiver bandwidth did not exceed 1.5 dB. The receiver’s consumed power was not more than 30 W. On the basis of that RR, the third-generation ARM equipment was manufactured including RM, DF, and CEE equipment, described in subsequent chapters. High specifications of the receiver shown in Fig. 3.34, and its further modernization

64

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.34 DRR ARK-CT1 of the third generation

Fig. 3.35 ARK-D1TP panoramic measuring receiver

permitted the creation of the panoramic measuring receiver (Fig. 3.35), certified in 2002 by the authorized Russian committee as measuring equipment. The constantly growing requirements for RM, CEE and DF complexes, in combination with the successfully-solved problem of full-scale production of the singlechannel and double-channel DRR of the third generation, made for the development of the fourth DRR generation in Russia, with an extended set of functional possibilities and higher specifications. The main requirements for such DRR are listed below: • Operating frequency range 0.01–3 GHz • Synthesizer retuning time not more than 5 ms • Presence of built-in digital unit for radio signal demodulation and technical analysis • Increased one-signal and multi-signal selectivity especially in the upper end of the operating frequency range

Development of Russian Arm Systems

65

• Extended possibilities regarding digital signal processing, due to the application of multi-processor units and the rate increase of information exchange with the control computer • Receiving section bandwidth not less than 4 MHz • Spectrum analysis rate not less than 500 MHz/s • Implementation of the modular construction principle, in particular, rackmounted realization on the basis of the Compact PCI standard constructions • Possibility for the coherent combination of receivers, along with the opportunity to equalize the amplitude-phase characteristics of the receiving paths. At the design step, special attention was devoted to the development of functionally and constructively completed modules, which could be easily combined in the various system types by means of different unifying plates and under-frames. As a result of this development, the new ARK-CT3 digital receiver was delivered into the Russian market in 2003. The frequency range of the new receiver is 9 kHz–3 GHz. The synthesizer switching time is about 5 ms, and a bandwidth is equal to 5 MHz, which allows for the possibility to calculate spectral panorama with a rate of more than 700 MHz/s. The dynamic range on the third-order intermodulation is not worse than 75 dB. This new receiver was used for the creation of ARM equipment of the fourth generation, including the ARK-D7K double-channel RM and CEE system, in ARK-MK1, ARK-MK2 mobile and stationary direction-finding systems, and in a new ARM station called ARGUMENT. The external view of the ARK-MK1 system central unit, on the basis of two coherently connected receivers, is shown in Fig. 3.36.

Fig. 3.36 Central module of ARK-MK1 system with the double-channel DRR of ARK-CT3 Type

The external remote-controlled radio signal converter, providing the higher limit enhancement of RM operating range till 18 GHz, was developed and designed in parallel with this receiver development. This frequency converter assures suppression of the spurious receiving channel up to 40 dB, has a noise factor of not more than 14 dB, can operate both with the internal directional antenna system, and with the external antenna. Its external view is shown in Fig. 3.37. In 2004, this external converter was certified as measuring equipment.

66

3

Radio Receiver Applications for Radio Monitoring System

Fig.3.37 ARK-KNV4 external frequency converter

Fifth-Generation Radio Receivers Unfortunately, equipment based on the DRR ARK-CT1 and ARK-CT3 required rather large weight and dimensions, in order to effectively apply it in the carried systems. Taking into consideration the cumulative experience on development and manufacture of the direction finders and the equipment for handheld DFs, as well as the experience of equipment-mounting on various ground and air carriers, in 2003,

ARK-CT1 Digital Radio Receiver

67

the following main requirements for the portable DRR of the next (fifth) generation were formulated in Russia: • Receiver design should include two modules: radio frequency (RF) – very high frequency (VHF) – ultra high frequency (UHF) radio signal frequency converter and the digital signal processing module • Basic size of the module plate should be accepted as 100 × 160 mm, which is the world standard for industrial equipment • Module implementation must ensure its integration into the user’s equipment • DRR should be used in multi-channel coherent receiving systems • DRR should have sufficient stable metrological characteristic for application as measuring equipment. The requirements for class B ARM equipment were selected as the technical requirements for new DRR. At that time, some additional restrictions were imposed concerning the consumed power, namely, the analog radio signal frequency converter should consume not more than 8 W, and the DSP module not more than 10 W. In the middle of 2004, after 12 months of intensive development, the first versions of the fifth generation DRR arrived and were named ARGAMAK. In 2004–2005, on the basis of the ARGAMAK family, the fifth generation of RM, DF and CEE technical equipment was developed, including the stationary, mobile, portable and carried equipment.

ARK-CT1 Digital Radio Receiver Let us consider the structure of Russian DRR of the third generation of ARK-CT1 type with the bandwidth of 2 MHz. Constructively, this receiver consists of two units: the analog radio receiving path of CT1 type and the digital signal processing unit of AC01 type, as shown in Fig. 3.38. Technical characteristics are presented in Table 3.3. Fig. 3.38 Structural diagram of ARK-CT1R

ARK-CT1 IF 10,7 MHz Input

CT1 unit

DSP unit

LPT COM

PC

The structural diagram of the RR section is shown in Fig. 3.39. The main section parts are: the pre-selectors unit; the mixers unit; the frequency synthesizers unit with the reference oscillator; and the control unit. To achieve high selectivity on the combination channels, it is necessary to ensure good signal selection in the pre-selector. The usage of the band-pass filters with fixed bandwidth is acceptable only in the case of rather narrow operating ranges of the receiver. Since the main area of this receiver application is radio monitoring in

68

3

Radio Receiver Applications for Radio Monitoring System

Table 3.3 Technical specifications of ARK-CT1 DRR Parameter name

Value

General parameters Operating frequency range of the basic set, MHz Input attenuator, dB Maximal permissible input signal, dBm Noise factor, dB: In range 25–1,000 MHz above 1,000 MHz Relative frequency instability of the reference oscillator ◦ ◦ Relative error of frequency setting (–20 C to +50 C) Tuning time of synthesizer, ms, not more than LO phase noise at offset 10 kHz, dB/Hz Selectivity and non-linear distortions Interference suppression at IF, dB, not less than Selectivity on the image channel, dB, not less than Dynamic range on the 3rd and 2nd order intermodulation, dB, not less than Gain factor irregularity within operating frequency range, dB, not more than

20–2020 10, 20, 30 23 not > 14 not > 16 5 · 10−6 10–6 10 –95 70 70 70 ±3

IF signal Analog IF signal frequency, MHz Bandwidth before IF output, MHz Demodulator frequency setting discreteness, Hz

10.7 2 1

Operating temperature, weight, dimensions, power consumption ◦ Operating temperature interval, C Supply voltage, V Consumed power, VA, not more than Dimensions (width × height × depth), mm Weight, kg

–10 to +60 24–30 18 300 × 65 × 255 not > 4

the wide frequency range, a set of ten tracking filters is used in pre-selector, which are switched by the analog switchers (AS), for the frequency sub-ranges: 20–35, 35–60, 60–100, 100–170, 170–240, 240–333, 333–465, 465–700, 700–1012, 1,012– 2,020 MHz. The medium frequency coverage factor is about 1.58. Within the frequency range of 20–1012 MHz the band-pass filters (BPF) are used, and in the range of 1,012–2,020 MHz the high-pass filter (HPF) with the cutoff frequency defined by the receiver tuning frequency. The filters bandwidth in the range of 20–1,012 MHz on –3dB-level is about 10%, and the bandwidth on –70 dB-level is 50–150% from the receiver tuning frequency. Figure 3.40 shows the experimental curves of the pre-selector transfer functions obtained at the testing of the receiver commercial specimen at the tuning frequencies 20 and 701 MHz. The control over switchers and pre-selector filters is fulfilled by the special controller. During factory adjustment for each individual receiver, a record of the controlling voltages, made into the controller, is executed in sixteen points of each

ARK-CT1 Digital Radio Receiver

69

Pre-selector unit

S w i t c h

Antenna input

Mixers unit

HPF (1012–2020 MHz)

Amp

HPF

Amp

BHPF

BPF (700–1012 MHz)

Amp

HPF

Amp

BPF

BPF (465–700 MHz)

Amp

HPF

Amp

BPF

BPF (333–465 MHz)

Amp

HPF

Amp

BPF

BPF (240–333 MHz)

Amp

HPF

Amp

BPF

BPF (170–240 MHz)

Amp

BPF

BPF (100–170 MHz)

Amp

BPF

BPF (60–100 MHz)

Amp

BPF

BPF (35–60 MHz)

Amp

BPF

Amp

BPF

BPF (20–35 MHz) Attenuator

S w i t c h

S w i t c h

Mixer 2 BPF

Amp

BPF

BPF

Amp

BPF

Switch Mixer 2

BPF

Amp

BPF

Mixer 3

BPF

Amp

Analogous IF output 10,7 MHz

Pre-selector unit controller

From ACO1 unit

M i x e r 1

Amp

Amp

Amp

Amp

Osc

Osc

Osc

Osc

Synthesizers unit

Synthesizers unit controller

Control unit

Fig. 3.39 Structural diagram of CT1 unit K/K0,dB –10

K/K0,dB –10

–30

–30

–50

–50

–70

–70

a)

10

20

30 f, MHz

b)

700

900 f, MHz

Fig. 3.40 Pre-selector frequency responses at the tuning frequencies of 20 MHz (a) and 701 MHz (b)

frequency sub-range. Obtaining the array in such a manner ensures filter fine-tuning at each tuning frequency of the receiver. In receiver operation mode, the voltages applied to the filter variable capacitance diodes from the outputs of the controller DAC unit are calculated by means of linear interpolation of the stored controlling values. Indexes, by which the controlling value selection is executed, are calculated on the basis of the receiver tuning frequency. A similar approach is used to maintain the regularity of the pre-selector transfer function. The transfer function irregularity does not exceed 3 dB at each frequency within the receiver operating range. To achieve the pre-selector tuning automation, the specific process software applications are used together with the specific hardware. The external view of pre-selector of ARK-CT1 DRR with removed shielding is shown in Fig. 3.41. As shown, the attenuator at the receiver input regulates the input signal attenuation by 0, 10, 20, or 30 dB. The next important element in the receiving path determining the resolution capacity and the receiver’s tuning rate is the frequency synthesizer unit. The single reference frequency 12.8 MHz from the reference oscillator with the crystal frequency stabilization, and with the temperature stabilization, is applied to the synthesizer unit. DRR retuning with 10 kHz-step is executed by the first LO

70

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.41 ARK-CT1 RR pre-selector

retuning, and the 500 Hz-step tuning is executed by the third LO retuning. The tuning with 1 Hz-step is fulfilled in the DSP unit. We can attribute the suppression method of the spurious harmonics penetrating into the IF section to the characteristic properties of the synthesizer units of the ARK-CT1 and ARK-CT2 receivers. This can be achieved by means of the simultaneous frequency offset of the 1st and 2nd LO at the given receiver tuning to the selected in advance offset. As a result, within the whole receiver operating range, from 20 to 2,020 MHz, the spurious harmonics are absent. Another synthesizer peculiarity is the frequency correction possibility of the built-in reference oscillator. The relative frequency instability of a RO in the temper◦ ature range from 0 to +50 C achieves 3 × 10-6 . RO program frequency correction, stipulated for the application as being in the widened temperature range, provides a reduction of temperature instability not less than three times. To increase the receiver frequency stability, the external reference oscillator, on 10 MHz or 12.8 MHz, can be connected to the receiver. The switching between internal and external RO is executed by the program approach. The synthesizer unit of the receiver provides the frequency setting, with the accuracy not worse than 250 Hz for less than 15 ms. One of the important factors defining receiver resolution capacity is the phase noise of the frequency synthesizer. In ARK-CT1 DRR, the phase noise level is mainly caused by the first LO noise, which increases with its frequency growth. Due to the fact that range coverage is provided by the mutual operation of all LOs, the noise distribution over the tuning range depends on the tuning frequency, but does not exceed 95 dB/Hz. Figure 3.42 shows a typical picture of spectral density of the phase noise power with respect to generated signal frequency offset. The receiving path is designed on the basis of the superheterodyne receiver with three frequency conversions. The fourth frequency conversion is fulfilled in the analog-digital signal-processing unit. The first IF value depends on the frequency of the received signal and is equal to 712 MHz or 302 MHz. The second IF value is equal to 45 MHz, and, lastly, the third IF value is equal to 10.7 MHz. DRR has the buffered third IF output, which can be used for the additional connections, for example, for radio signals decoders. Receiver selectivity on the adjacent channel is defined by the characteristics of the IF filters used. Therefore, when designing the DRR analog path, much attention

ARK-CT1 Digital Radio Receiver Fig. 3.42 Phase noise spectrogram at IF3 output for 460 MHz tuning frequency

71 N, dBc/Hz –20 –40 –60 –80 –100 –15 –10

–5

0

5

10

15 f, MHz

was paid to the development of the IF filters. The resulting filter parameters are listed in Table 3.4. Such filter applications ensure the suppression of the image and spurious channels, occurring as a result of the spectra transition, by not less than 70 dB. The typical value of the combination interference suppression level is equal to 90 dB. Table 3.4 IF filters selectivity characteristics Parameter name AFR irregularity in bandwidth, dB, not more than –1 dB-bandwidth, MHz –70 dB-bandwidth, MHz

IF1 302 MHz

IF1 712 MHz

IF2 45 MHz

IF3 IF4 10.7 MHz 1.6 MHz

±0.5

±0.5

±0.5

±0.5

±0.5

17 60

24 90

6 20

3.2 6.4

2 3.2

The structural diagram of the ACO1 unit is shown in Fig. 3.43. The unit consists of two main assemblies of the 3rd IF 10.7 MHz frequency converter into the 4th IF 1.6 MHz; ADC; the digital signal processor; the controller unit. ACO unit 10.7 MHz input from CT1 unit

CSP

ADC

AO unit 10.7 MHz channel

Converter

1.6 MHz channel

DSP

+12V Controller USB

To controller of CT1 unit

To PC

Fig. 3.43 Structural diagram of ACO1 unit

COM

Buffer Power supply +27V

+27V

DAC

From power supply To CT1 unit

LPT

To PC CSP testing output

72

3

Radio Receiver Applications for Radio Monitoring System

The signal at 1.6 MHz frequency is exposed to analog-digital conversion in 12-bit ADC with the sampling frequency 6.4 MHz. The fourth IF value 1.6 MHz is related to ADC sampling frequency. The selected relation between the input frequency 1.6 MHz and the sampling frequency, equal to 1:4, allows the algorithms to be used to reduce the computing operation number when obtaining the digital signal complex envelope. The operations related to the spectrum transition from the 3rd IF to the 4th IF, are shown in Fig. 3.44. AFRBPF

Spectrum transition

AFRRR

AFRIF

0

1.6 2

AFRRR

0

1.6 2

3.2

6.4

10.7 RR IF

Image channels of the sampled signal (without filtering)

F,MHz

Operating bandwidth

3.2

6.4

F,MHz

Image channels of the sampled signal (with Transition LF band filtering) Transition HF band

Fig. 3.44 Transition to IF 1.6 MHz

Before AD conversion, the signal in the 1.6 MHz channel is filtered additionally in the band-pass filter, which has a 2 MHz bandwidth. With respect to the bandwidth central frequency, at 2.6 MHz offset, a suppression of not less than 80 dB is ensured. The filter transfer function irregularity within the bandwidth does not exceed 1 dB. From the band-pass filter, the signal passes at ADC input. After ADC, the digital signal enters the digital signal processor, manufactured by Analog Devices Ltd. The processor’s main purpose consists of signal spectral analysis on the basis of fast Fourier transform (FFT). The length of a typical spectrum is equal to 1,024 complex samples. The spectral sample discreteness is 3.125 kHz. To reduce the Gibbs effect, a Kaiser-Bessel window is used to weight the input time sample. At selected sampling frequency, the equivalent noise bandwidth of this window is equal to 6 kHz. The calculation time for the 1,024 sample complex spectrum does not exceed 5.5 ms. If necessary, the time sample length used in calculation

ARK-D1TP Digital Panoramic Measuring Receiver

73

may achieve 16,000 samples. In a DSP unit, additional kinds of processing can be executed, such as signal detection, phase difference determination, peak, quasi-peak and rms detection, and digital demodulation. Information exchange with the PC is fulfilled through the printer parallel port. At that time, the EPP, Bi-Di or SPP protocols can be used. The ACO1 unit has an additional feature: DSP program loading is executed from the PC. This presents the possibility for fast DSP program changes directly during the unit’s operation. The ARK-CT1 digital radio receiver has within its structure a secondary impulse power supply, ensuring that the receiver can operate from power sources with the voltage range 9–33 V and not corrupt the sensitivity of the receiver analog paths. The inherent temperature regime in RRs is provided by the hardware-software controller, by adjusting the internal warming elements and the external forced aircooling used in combination with the heat radiators mounted on the receiver’s case. ARK-CT1 DRR can successfully operate under conditions of considerable mechanical impacts. It is stable to vibration loading in the frequency range from 10 to 55 Hz, shift amplitude 0.15 mm and to shock load with 10 g peak acceleration at shock impulse duration 16 ms. On the basis of the ARK-CT1 receiver, it was possible to develop the ARKCT2 double-channel coherent receiver, which consists of two identical ARK-CT1 receiver paths and the double-channel analog-digital processing unit. The above-mentioned properties of ARK-CT1 DRR made it possible to use it as a hardware base for the ARK-MK1–ARK-MK6 mobile and deployed RM and DF systems, and in ARK-D1T single-channel CEE complexes and ARK-D7 doublechannel CEE complexes (see below).

ARK-D1TP Digital Panoramic Measuring Receiver The high stability of ARK-CT1 DRR characteristics gave rise to the possibility to create the ARK-D1TP panoramic measuring receiver on its basis [10]. The ARK-D1TP panoramic measuring receiver is a Russian receiver certified as measuring equipment by the authorized state institution. As the ARK-CT1 DRR, this receiver is designed for operation in the frequency range from 20 to 2,020 MHz in the automated RM system structures. To widen the operating frequency range till 18 GHz, one can connect to this receiver the ARK-KNV4 external converter, which is also certified as measuring equipment. The receiver is manufactured in the form of portable, desk-size equipment and consists of two units: the CT1 unit and the ACO1 unit. The main technical specifications of ARK-D1TP are listed in Table 3.5. The possibility of using this DRR as measuring equipment is achieved in the first place due to the automation of the AFR adjustment process in DRR during factory adjustments and the possibility of equipment calibration during normal operation. In the D1TP panoramic measuring receiver the possibility of forming and using the calibration system files “as a whole” is realized. In this case, the calibration peculiarity is the calibration files usage, which contains the correction factors array for the frequency and temperature lists.

74

3

Radio Receiver Applications for Radio Monitoring System

Table 3.5 Technical specification of ARK-D1TP panoramic measuring receiver Parameter name

Value

General parameters Operating frequency range of the basic set, MHz Operating frequency range in maximal configuration (with ARK-KNV4 converter), MHz Limit of permissible error of level measurement, dB Frequency stability External reference oscillator frequency, MHz VSWR on input Maximal measured signal at receiver input, dBm Panoramic analysis speed in operating range, MHz/s Selectivity and non-linear distortions Dynamic range of measured radio signals, with built-in attenuators, dB Dynamic range on 3rd and 2nd order intermodulation, dB Image and spurious channels attenuation, dB Sensitivity restricted by noise (S+N)/N till 6 dB, dBm At frequencies 20–1,012 MHz

20–2,020 0.009–18,000 ±3 2 × 10–6 12.8 Not more than 3 –7 (100 dBμV ) 150 100 70 70

At frequencies 1,012–2,020 MHz

–110 (–3 dBμV , 0,71 μV ) –107 (0 dBμV , 1.0 μV )

Weight, dimensions, power consumption Supply voltage, V Consumed current, A Weight, kg Dimensions (width × height × depth), mm

27 ±3 1.2 7.5 340 × 130 × 260

The complex calibration is fulfilled with the help of the standard signal sources, attenuators and connectors, the errors of which are 3–5 times less than the main permissible error of the complex. The calibration is executed at several reference points of the frequency range by means of the transfer function variation till the nominal reading achievement. The amendment calculation between two reference points is conducted on the basis of the linear interpolation method, which essentially simplifies the correction algorithm. Having increased the reference point number, one may decrease the bias error up to the level defined by the calibration process error. The calibration allows for the execution of a serviceability check for the hardwaresoftware system and the reduction of its bias error. The interpolation allows us to obtain, with the given margin of error, the correction values falling in the interval between two reference points. The interpolation accuracy depends on the argument variation interval (t2 − t1 ) and the function increment F(t2 ) − F(t1 ). At linear interpolation, the value F(t12 ) in the arbitrary point t12 between the points t1 and t2 is determined as:

ARK-CT3 Digital Receiver

75

F(t12 ) = F(t1 ) +

(t12 − t1 )[F(t2 ) − F(t1 )] . t2 − t1

(3.58)

Calibration by the linear interpolation method between the reference points is executed in the ARK-D1TP receiver at the stage of the pre-selector unit adjustment. To satisfy the standard requirements, the peak, quasi-peak and rms detectors are software realized. They operate over the signal time sampling on IF. Before the signal is detected, it is exposed to filtering in the band-pass digital filter. The digital filter bandwidth can be defined from the list. In this list, there are bandwidths of 9 and 120 kHz, which are recommended by state standards. The possibility is stipulated to change the time constants of charge or discharge for the quasi-peak detector. For this, the inertial part of the quasi-peak detector is implemented in the form of the 1st-order filter digital model with the transfer function K(p) =

1 , 1 + pτ

(3.59)

where p is a complex variable, and τ is the filter time constant. When using the integration by trapezium method, the transfer function of the discrete filter will have the form:

K(z−1 ) =

1 2 1−z−1 Td 1+z−1 τ

+1

=

Td −1 Td +2τ (1 + z ) Td −2τ −1 +1 Td +2τ z

(3.60)

where z is a complex variable, Td is the IF signal sampling period, and τ is the time constant. The difference equation of this digital filter can be written as: y[k] =

Td − 2τ Td (x[k] + x[k − 1]) − y[k − 1] Td + 2τ Td + 2τ

(3.61)

where x[k] is the signal sample at filter input, y[k] is the signal sample at detector output, k is a number of time sample. The time samples period is equal to Td . If the input signal x[k] increases, filter coefficients are used that are calculated on the basis of the recharge time constant; if the x[k] signal decreases, filter coefficients are used that are calculated by the discharge time constant.

ARK-CT3 Digital Receiver Compared to the ARK-CT1 DRR, the ARK-CT3 receiver has the more advanced characteristics of receiving path linearity, selectivity, the operating frequency range, the bandwidth and the frequency-retuning rate. The dynamic range on the 3rd order intermodulation is 75 dB, the operating frequency range is from 9 kHz to 3 GHz, the receiving path bandwidth is 5 MHz, the synthesizer retuning time does not exceed 5 ms. This receiver uses a module construction based on the standard Compact PCI, with module printed plate sized 100 × 160 mm. The basic receiver type has two

76

3

Radio Receiver Applications for Radio Monitoring System

Table 3.6 Technical specifications of ARK-CT3 DRR Parameter name

Value

General parameters Operating frequency range of the basic set, MHz Input attenuator, dB Maximal permissible input signal, dBm Noise factor, dB: In the range 25–1,000 MHz Above 1,000 MHz Frequency instability of reference oscillator ◦ ◦ Relative error of frequency setting (–20 C to +50 C) Tuning time of the synthesizer, ms, not more than LO phase noise at 10 kHz offset, dB/Hz Selectivity and non-linear distortions Suppression of IF interference, dB, not less than Selectivity on image channel, dB Dynamic range on the 3rd and 2nd order intermodulation, dB Transfer function irregularity in operating frequency range, dB, not more than IF signal Analog IF signal frequency, MHz Bandwidth before IF output, MHz Demodulator frequency setting discreteness, Hz Operating temperature, weight, dimensions, power consumption ◦ Operating temperature interval, C Supply voltage, V Consumed power, VA, not more than Dimensions (width × height × depth), mm Weight, kg

0.009–3,000 10, 20, 30 23 not > 14 not > 16 1 · 10−6 1 · 10−6 5 –95

70 70 75 ±3

41.6 4 or 5 1

–20–+50 24–30 50 450 × 140 × 300 not > 6.5

coherent receiving channels. This receiver’s external view in rack-mount implementation is shown in Fig. 3.36. ARK-CT3 DRR technical specifications for two coherent channels are listed in Table 3.6. ARK-CT3 DRR consists of the range switch and two identical channels of signal receiving (channels A and B), which are connected to a double-channel DSP unit. The structural diagram of ARK-CT3 DRR is shown in Fig. 3.45. The signals from the antennas, or from the antenna switch, pass to the doublechannel range switch, which connects them to the appropriate channels A or B depending on tuning frequency. In the 25 MHz–1 GHz range the input signals are connected to input 1 of the channels, in 1–3 GHz, they are connected to input 2. The receiver control is executed through the RS-485 serial interface. The structural diagram of one receiving path is shown in Fig. 3.46. The following units are included in the channel structure:

ARK-CT3 Digital Receiver

77 Output A IF 41.6 MHz

Input 1

Input B 25–3000 MHz

Range switch

Input A 25–3000 MHz

Channel A

Input 2

Supply 9–16 V

RO LO LO output1 output2 output LO LO RO input1 input 2 input

Output B IF 41.6 MHz

Input 1 Input 2

Channel B

Control RS-485

Fig. 3.45 Structural diagram of ARK-CT3

BPF 3 (285–465 MHz) BPF 4 (160–285 MHz) BPF 5 (85–160 MHz)

Switch 7/1

Switch 1/7

Switch 1/2

Switch 2/1

Attenuator 0..–30dB

Input 1 25–1000 MHz

Buf

BPF 2 (465–700 MHz)

IFA 1 299.2 ÌÃö

Mix

IFA 2 708.8 ÌÃö

Switch 2/1

IFU RFU-1

BPF 1 (700–1000 MHz)

Buf

Mix

Buf

BPF 38.6-44.6 IF output

Buf

Mix

Buf

RO

41.6 MHz VCO 1 257.6

PLL system

Switch 2/1

Buf

Buf

Ðâ LO output 2

Switch 2/1

BPF 6 (45–85 MHz) BPF 7 (25–45 MHz)

RO

Switch 2/1

VCO 2 667.2

PLL system

Buf

Ðâ LO input 2 LO output1

RFU-2

Buf

BPF 3 (2250–2500 MHz)

Buf Buf

BPF 4 (2000–2250 MHz) BPF 5 (1750–2000 MHz)

Switch 8/1

1000–3000 MHz

Switch 1/8

Input 2

Attenuator 0..–30dB

BPF 1 (2750–3000 MHz) BPF 2 (2500–2750 MHz)

Mix

Buf

VCO 3 1131–1400

BPF (1826–2262)

Buf

VCO 2 913–1131

BPF (1550–1826)

BPF 6 (1500–1750 MHz)

VCO 1 733-913

BPF 7 (1250–1500 MHz) BPF 8 (1000–1250 MHz)

BPF (2262–2800) BPF (913–1131)

Switch 3/1 BPF (73311–1913)

Buf

BPF (1131–1400)

Switch 3/1

Synthesiz ers RO unit

Switch 2/1

Switch 2/1

Power supply

RO

PLL system

RO

PLL system

–5V +5V +3.3V+12V+30V

Switch 2/1

PLL system

Control buses and lines Control unit

Switch 1/3

RO Reference oscillator

BUF

CPSU

LO input 1

RO output RO input Control RS-485 Power supply

9 - 16 V

Fig. 3.46 Structural diagram of ARK-CT3 (one receiving channel)

• RFU1 and RFU2 radio frequency units, fulfilling the preliminary frequency selection, signal amplification, and the frequency conversion to the first IF, which is equal to 299.2 MHz or 708.8 MHz (the frequency plan) depending on the frequency tuning • Synthesizer unit (SU) intended for first LO signal formation for RFU1 and RFU2 mixers • IF unit (IFU) fulfilling the main frequency selection, signal amplification and frequency conversion to the second IF, which is equal to 41.6 MHz • Control and power supply unit (CPSU) intended for power supply voltage generation necessary for all units, for receiving and processing the commands from

78

3

Radio Receiver Applications for Radio Monitoring System

the DSP unit, for the unified reference frequency 12.8 MHz generation necessary for synthesizer operation, including in SU and IFU units. The input signal, at frequencies of 25–1,000 MHz, enters the attenuator of RFU1 and then on to the pre-selector assembly. The set of seven sub-range frequency channels is arranged in the pre-selector, which can be selected by seven-channel input and output switches. Each channel has two adjustable band-pass filters controlled by a micro-controller, and the buffer cascade. The signal from the buffer cascade passes to the mixer input. To its second input, the LO signal enters the frequency from 733.8 to 1,299.2 MHz. The input signal at the frequency of 1–3 GHz enters the input attenuator of RFU2 and then on to the pre-selector assembly. The RFU2 pre-selector has eight frequency channels, which can be selected by eight-channel input and output switches. Each channel includes two non-adjustable band-pass filters and the buffer cascade. The signal from the buffer cascade enters the mixer input. The signal from the adjustable LO of the synthesizer unit, at a frequency from 1,550 to 2,800 MHz is applied to the second input of the mixer. The synthesizer unit is intended for reference frequency grid generation for RFU1 and RFU2 mixers. SU consists of three adjustable LO with the adjustable frequencies 73–913 MHz, 913–1,131 MHz, 1,131–1,400 MHz. All local oscillators are implemented on the basis of the frequency synthesizer. The unified reference frequency signal enters the SU from the RO. The LO signal is applied to the RFU1 mixer input, and the signal to the RFU2 mixer is applied after frequency doubling. To arrange the channel’s coherent operation in the direction-finding structure, the possibility is ensured for both to pass the internal LO signals to the synchronization output connectors and to obtain the external oscillator signals from the input connectors. The intermediate frequency unit (IFU) consists of two channels from the first IF processing: the synthesizers and the switches of input and output data. The input switches fulfills the IF signals switching from RFU1 and RFU2 outputs to one of the first IF processing channels (IFA1 or IFA2). The first IF and first LO frequency values are shown in Table 3.7, for various tuning frequencies. Table 3.7 First IF and first LO frequencies of ARK-CT3 Tuning frequency, MHz

LO frequency, MHz

IF frequency, MHz

from 25 to 464.8 from 464.8 to 1,001.2 from 1,001.2 to 1,250.8 from 1,250.8 to 2,550.4 from 2,550.4 to 2,750 from 2,750 to 3,000

from 733.8 to 1,173.6 from 767.4 to 1,300.4 from 1,710 to 1,959.6 from 1,550 to 2,799.6 from 2,201.2 to 2,450.8 from 2,041.2 to 2,291.2

708.8 299.2 708.8 299.2 299.2 708.8

IFA1 and IFA2 have similar structures and consist of the first IF filter and buffer amplifier, and the second mixer. The signals from the buffer amplifiers

ARK-KNV4 External Remote-Controlled Converter

79

are applied to the mixer inputs. Another input of the mixers absorbs the signals from the appropriate non-adjustable LOs of IFU synthesizers module (257.6 and 667.2 MHz). Similarly to SU in IFU, to arrange the coherent channel operation in the direction-finding equipment structure, it is possible both to pass the internal LO synchronization signals to the output connectors and to obtain the external source signals at the input connectors, for further application as the local oscillator signals. The control and power supply unit (CPSU) consists of the impulse voltage converter generating the voltage set necessary for other units operation; the receiving and processing micro-controller for the commands from the DSP unit, and the reference oscillator generating the unified reference frequency 12.8 MHz for the frequency synthesizers included in the SU and IFU structures. To ensure that both channels operate from the unified reference frequency there is the possibility in CPSU both to pass the reference oscillator synchronization signal to the output connector and to obtain the reference oscillator signal from the external source (another channel) from the input connectors.

ARK-KNV4 External Remote-Controlled Converter The ARK-KNV4 radio signals converter operates in the structure of the wide-band automated RM and CEE systems. It is intended for radio signal transition from the range of 1–18 GHz to the intermediate frequencies 299.2 and 708.8 MHz. This converter can be used as portable equipment, or as a stationary device, mounted on a mast or a tripod. The converter connects to ARK-CT1, ARK-CT2 or ARK-PR5 DRR. The main technical specifications of the converter are listed in Table 3.8. The operating temperature range of the ARK-KNV4 converter is from –10◦ C to +50◦ C;. Relative humidity of the environment should not exceed 80%. The external view of the ARK-KNV4 converter is shown in Fig. 3.37, and Figs. 3.47 and 3.48 show the converter mounted on a tripod and on a dielectric mast, together with the horn measuring antenna. The structural diagram of the ARK-KNV4 converter is shown in Fig. 3.49. This converter consists of the active antenna unit, pre-selector and first IF unit, synthesizer 1, mixer 1, synthesizer 2, the second IF unit, the reference oscillator and the control unit. The built-in antenna system of the ARK-KNV4 contains eight wide-band active antennas implemented as the active phased array with linear signal polarization. The signal from the antenna system, which corresponds to the switch tuning frequency, and which, with the help of switches “9” to “1” passes through the group path to the mixer. The ninth switch input is used for the transition to operation mode from the external antenna. The signal transfer factor from the external antenna input to IF output is 0 dB. Thus, the ARK-KNV4 converter, together with the measuring antenna, can be used for the field strength measurement.

80

3

Radio Receiver Applications for Radio Monitoring System

Table 3.8 Main parameters of ARK-KNV4 radio signals converter Parameter name Received frequencies range, GHz: At operation from the internal antenna At operation from the external antenna input Output signal frequency, MHz Relative LO frequency instability, not worse than –3 dB-level bandwidth, MHz, not less than Converter sensitivity across the field, at operation from the internal antenna in the bandwidth 10 kHz, μV/m, not worse than Transfer factor at operation from external antenna, dB Transfer factor error at operation from external antenna, dB, not > Relative level of 3rd-order intermodulation noise at the 10 MHz offset for the receiver 120 kHz bandwidth, dB, not more than Spurious receiving channels suppression at operation from internal antenna, dB, not less than Noise factor (without accounting for the transfer factor of antenna circuit), dB, not worse than Medium level of inherent noise within the bandwidth 120 kHz, W, not more than Dynamic range, dB Through AFR irregularity, dB, not more than Phase noise at 10 kHz-offset, dB/Hz, not more than Intercept point on the 3rd-order intermodulation, dBm SWR on external antenna input at input impedance 50 , not worse than SWR on output, not worse than Supply voltage, V Consumed power, VA, not more than Dimensions (width × height × depth), mm, not more than Converter weight, kg, not more than

Value

3–18 1–18 299 or 708.6 2.10–7 10 50 0 6 64 45 14 4.10–11 70 ±6 –80 –5 2 3 12±3.0 or 27±3.0 19 250 × 220 × 90 2.0

From the mixer output the signal passes through the first IF unit. Here, the signal is amplified and restricted in bandwidth. After that, the first IF signal enters the second IF unit, which carries out the additional frequency selection, the amplification, and the signal transfer to the output frequency of the second IF. Synthesizers 1 and 2 generate the signals required for the mixers. After filtering, amplification, and frequency conversion, the signals from ARKKNV4 output at IF 299.2 MHz or 708.8 MHz enter the receiver input, for example, ARK-D1TP or ARGAMAK. The intermediate frequency value is defined by the ARK-KNV4 tuning frequency, in accordance with the frequency plan accepted at the development stage. The control unit accepts the commands from the external devices through the serial bus RS-485, transfers the data to the active antenna unit, to the synthesizer units and to the second IF unit. The control unit of ARK-KNV4 has a nonvolatile storage device, in which the adjustable parameters are stored, namely, the frequency correction coefficients for the reference oscillator, attenuator adjustment data, etc.

ARK-PR5 “Argamak” Digital Radio Receiver

81

Fig. 3.47 ARK-KNV4 converter on the tripod

These parameters are automatically loaded after switching-on the converter or after the hardware reset signal is received. The ARK-KNV4 external remote-controlled radio signal converter is included into the measuring equipment register, by State Standard of the Russian Federation.

ARK-PR5 “Argamak” Digital Radio Receiver Compared to ARK-CT1 and ARK-CT3, the ARK-PR5 ARGAMAK digital radio receiver has several times as less weight and dimensions. Therefore, its accuracy and sensitivity parameters are better than those of the ARK-CT3, and the synthesizer tuning time on frequency is shorter than 2 ms. This receiver consists of two modules: the ARK-PS5 RF-VHF-UHF radio signal converter and the ARK-CO DSP module, as shown in Fig. 3.50. Each module is located on the multi-layer printed board with dimensions of 100 × 160 mm. The implementation of each module allows for the possibility of

82

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.48 ARK-KNV4 converter on the dielectric mast

ARK-KNB4 IF1 unit Active antenna unit First IF unit

Active antenna 1

Second IF unit

IF output

Active antenna 4 Active antenna 5

Switch 9 to 1

Active antenna 3

Group section

Active antenna 2 Mixer

Synthesizer 2

Reference oscillator

Synthesizer 1

Active antenna 6 Control unit

Control RS-485

Active antenna 7 Active antenna 8

Fig. 3.49 Structural diagram of ARK-KNV4 converter

their joint or separate application in the equipment. The important DRR feature is the possibility of its application – without any revisions – in the structure of multichannel coherent systems. Moreover, DRR has stable characteristics, including an interface to connect the external reference oscillator, which allows it to be used

ARK-PR5 “Argamak” Digital Radio Receiver

83

IF Input

ARK-PS5

ARK-CO

USB 2.0

PC

(RS485) Control

Fig. 3.50 Enlarged structural diagram of ARGAMAK SRR

as measuring equipment. One more helpful feature is the possibility to select the receiver bandwidth from 2, 5, or 10 MHz. ARK-CO2, ARK-CO5, and ARK-CO10 DSP modules with bandwidths 2, 5, and 10 MHz, respectively, can be used together with this receiver, depending on the bandwidth. Depending on the implementation version, the ARGAMAK DRR can be accommodated in a separate case, as shown in Fig. 3.51, inside the protective case shown in Fig. 3.52, or can be built directly into the equipment, e.g., into the antenna array of the ARK-MK11 DF system, or into the case of the ARK-RP4 handheld direction finder, etc.

a)

b) Fig. 3.51 ARK-PR5 DRR: (a) Front view; (b) Rear view

84

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.52 ARK-PR5 DRR in protective case with emergency battery

The ARGAMAK DRR has analog IF outputs with bandwidths of 2, 5, or 10 MHz, for connection to the various decoding or demodulating units. ARK-PR5 DRR specifications are listed in Table 3.9. The ARK-CO DSP module has an audio line and adjustable outputs from the digital signal demodulator. The receiving of commands and the data transfer to the PC are executed via a USB 2.0 serial interface. The calculated spectral panoramas or the time signal samples are transferred to the PC. The rate of spectral panoramas obtained depends on the receiver, and DSP unit bandwidths and may exceed 3,000 MHz/s, at a bandwidth of 10 MHz (see Table 3.9). The continuous time sample of arbitrary duration can be transferred to the PC for the received signal with a bandwidth up to 2 MHz. For 5 and 10 MHz signal bandwidths, time samples of limited duration are transferred. Achievement of the DRR’s low weight and dimensions became possible due to the application of advanced passive electronic components with 0402, 0603, and 0805 standard size, and the integrated circuits in MLP, QFN leadless cases, which allow a surface mounting density of more than 10 elements per square cm to be achieved. The ARK-PR5 receiver plate, with dimensions 100 × 160 mm, has about

ARK-PR5 “Argamak” Digital Radio Receiver

85

Table 3.9 ARK-PR5 DRR technical specifications Parameter name General parameters Operating frequency range of basic set, MHz Input attenuator, dB Maximal permissible input signal, dBm Noise factor, dB: In range 25–1,000 MHz Above 1,000 MHz Frequency instability of reference oscillator ◦ ◦ Relative error of frequency setting (–20 C–+50 C) Tuning time of synthesizer, ms, not more than LO phase noise at 10-kHz-offset, dB/Hz In range 25–1,000 MHz Above 1,000 MHz Selectivity and non-linear distortions Intercept point on the 3rd-order intermodulation IP3, dBm IF interference suppression, dB, not less than Image attenuation, dB Dynamic range on the 3rd and 2nd order intermodulation, dB Transfer factor irregularity within operating frequency range, dB, not more than IF signal IF analog signal frequency, MHz Bandwidth before IF output, MHz DSP unit Demodulator frequency setting discreteness, Hz Built-in demodulator Detectors for signal level measurement Spectrum calculation speed at usage of, MHz/s: ARK-CO2 DSP module ARK-CO5 DSP module ARK-C010 DSP module Outputs, control and data transfer interface Low frequency audio output Audio output for earphones Control and data transfer interface Operating temperature, weight, dimensions, power consumption Operating temperature interval, ◦ C Power supply voltage, V Consumed power, VA Dimensions (width × height × depth), mm Weight, kg, max

Value 0.009–3,000 0–30, step 2 23 not more than 12 not more than 14 5 × 10–7 5 × 10–7 2 –95 –85 0 70 70 75 ±3 10.7 or 41.6 2, 5, or 10 1 AM, FM, PM, SSB, CW Peak, Quasi-peak, rms 600 1,500 3,000

Is present Is present USB 2.0 –20–+50 9–16 15 108 × 42 ×200 1.5

86

3

Radio Receiver Applications for Radio Monitoring System

1,800 mounted elements. In printed plate topology, special reference marks are provided for automated assembling; and, checking points are used for the automated adjustment. To manufacture the multi-layer printed circuit board, the FR-4 and RO4350 materials were used, which allows for the ensuring of parameter recurrence over specimens, to take these parameters into account in the mathematical models. At each of the design stages, mathematical modeling was widely used, permitting the receiver unit characteristics to be balanced, and the best location of the printed-circuit wiring components to be selected, without the manufacture of a preproduction model. Figure 3.53 depicts the external view of the ARK-PS5 HF-VHF-UHF radio signal converter module for the ARGAMAK DRR, without the shields, and Fig. 3.54 shows its structural diagram.

Fig. 3.53 External View of ARK-PS5 Radio Signal Converter Module

The ARK-PS5 module represents a device with dimensions 100 × 160 × 20 mm, based on the single, printed circuit board in “Euroboard” standard. The purpose of this module is to receive radio signals in the frequency range of 9 kHz–3 GHz. The output signal is an IF signal at frequency 10.7 MHz or 41.6 MHz. IF value selection for each of the two outputs is executed by the software. For reception in the UHF range of 25–3,000 MHz at the output IF 41.6 MHz, two frequency conversions are used, at the output IF 10.7 MHz, three frequency conversions are used. A single frequency conversion is used for reception in the range of 0.009–30 MHz at output IF 41.6 MHz. But, this frequency conversion should be doubled for output IF 10.7 MHz. The signal at output IF arrives at two independent output connectors and for each connector an individual IF value can be assigned. To arrange the synchronous operation of the LOs in the structure of several ARKPR6, four connectors are mounted at its printed circuit board. Using them, one can ensure the following synchronization modes: • Completely autonomous operation from the internal reference oscillator at 12.8 MHz frequency with 5 · 10−7 accuracy • Autonomous operation of frequency synthesizers from the external reference oscillator, the signal of which enters one of the four synchronization connec-

IF1 In

BPF 4.4 (225–465 MHz) BPF 4.5 (110–225 MHz)

VCO 1.2 1700-2400

BPF 4.6 (53–110 MHz)

Channel Â 706.8 MHZ

Frequency conversion unit

BA

VCO 2.1 258.4

Mixer 1

BPF 4.7 (25–53 MHz) Frequency conversion unit VCO 3 44.1–71.6

Switch 2/1

BA

Filter unit IF1 Channel Â 300 MHZ

BA LO2 out/in

RO PLL system

BA

VCO 2.2 665.2

Channel 0.009-30 MHz

BA

Switch 2/1

BPF 4.3 (465–850 MHz)

Switch7/1

Switch 1/8

Att 0…–30 (0.009-30 MHz) dB

25-3000 MHz

PLL system

Switch 2/1

RO

BPF 4.2 (0.85–1.6 GHz)

BA

Switch 3/1

BPF 4.1 (1.6–3.0 GHz)

Switch 1/2

VCO1.1 730–1300

Channel 25–3000 MHz

Switch 2/1

Frequency conversion unit

Switch 2/1

Preselector unit

BA

RO PLL system

Switch 4/1

BPF 4.11 (9–1500 kHZ)

LO3 out/in

–3V 9–16 V

+15V +5V +3.3V +25V

Power supply unit Control buses and lines

Control RS–485

BPF 41.6 (5 MHZ

Filter unit IF1

Frequency conversion unit

BA

Mixer 4

RO +5 V RO

BPF 10.7 (2 MHz)

PLL system BA

Control unit Reference oscillator

IF outputs switch

Switch 2/1

BPF 4.10 (1.5–4.5 MHz)

Mixer 2 Mixer 3

Switch 2/1

BPF 4.9 (4.5–12 MHz)

Switch 2/1

Switch 1/4

0.009-30 MHz Àtt 0…–30 dB

Switch 2/1

BPF 4.8 (12–30 MHz)

ARK-PR5 “Argamak” Digital Radio Receiver

LO1 out/in

IF 41.6/10.7MHz

LO4 out/in. RO Switch 2/1

VCO 4 52.3

87

Fig. 3.54 Structural diagram of ARK-PS5 of ARGAMAK DRR HF-VHF-UHF radio signal converter

88

3

Radio Receiver Applications for Radio Monitoring System

tors to increase the frequency setting accuracy, and for the complex frequency synchronization of several ARK-PS5 modules • Autonomous operation from the internal reference oscillator with reference frequency signal delivery to one of the four synchronization connectors for the complex frequency synchronization of several ARK-PS5 modules • Synthesizer operation from the internal reference oscillator with LOs signal delivery to four external connectors, to ensure the frequency synchronization of several ARK-PS5 modules, up to accuracy within the phase (e.g., operation in the direction finder’s complex structure) • The mode of completely-driven LOs with disconnected internal reference oscillators and frequency synthesizers, to ensure the synchronization of several ARK-PS5 modules, up to accuracy within the phase (e.g., operation in the direction finder system’s structure). The ARK-PS5 module represents a device with dimensions 100 × 160 × 20 mm, based on the single, printed circuit board in “Euroboard” standard. The purpose of this module is to receive radio signals in the frequency range of 9 kHz–3 GHz. The output signal is an IF signal at frequency 10.7 MHz or 41.6 MHz. IF value choice for each of the two outputs is executed by the software. For reception in the UHF range of 25–3,000 MHz at the output IF 41.6 MHz, two frequency conversions are used, and, at the output IF 10.7 MHz, three frequency conversions are used. A single frequency conversion is used for reception in the range of 0.009– 30 MHz at output IF 41.6 MHz. But, this frequency should be doubled for output IF 10.7 MHz. The signal at output IF enters two independent output connectors and, for each connector, an individual IF value can be assigned. To arrange the synchronous operation of the LOs in the structure of several ARKPR6, four connectors are mounted at its printed circuit board. Using them, one can ensure the following synchronization modes: • Completely autonomous operation from the internal reference oscillator at 12.8 MHz frequency with 5 · 10−7 accuracy • Autonomous operation of frequency synthesizers from the external reference oscillator, the signal of which enters one of the four synchronization connectors to increase the frequency setting accuracy, and for the complex frequency synchronization of several ARK-PS5 modules • Autonomous operation from the internal reference oscillator with reference frequency signal delivery to one of the four synchronization connectors for the complex frequency synchronization of several ARK-PS5 modules • Synthesizer operation from the internal reference oscillator with LOs signal delivery to four external connectors, to ensure the frequency synchronization of several ARK-PS5 modules, up to accuracy within the phase (e.g., operation in the direction finder’s complex structure) • The mode of completely-driven LOs with disconnected internal reference oscillators and frequency synthesizers, to ensure the synchronization of several ARK-

ARK-PR5 “Argamak” Digital Radio Receiver

89

PS5 modules, up to accuracy within the phase (e.g., operation in the direction finder’s complex structure). In all of the above-mentioned modes, the pre-selector and other unit tuning is independently executed through the control bus for each ARK-PS5 module, in accordance with the commands from the controlling PC or another device. Let us consider the structural diagram of ARK-PS5 ARGAMAK DRR module in Fig. 3.54. Upon input of the HF, VHF, and UHF pre-selector sections, the attenuator units are applied; they are intended for the input signal attenuation within the limits 0–30 dB with 2 dB-step. The pre-selector section units of 25–3,000 MHz are used for the preliminary selection, to suppress the signals in the spurious receiving channels and to improve the signal-noise ratio (SNR) at the first mixer input. The input signal arrives through the input switch to the inputs of the adjustable filters, which select the input signal in the given frequency ranges. There are seven adjustable wide-range filters (BPF1–BPF7) switching by the input and output switches, for the following frequency sub-ranges: 25–53; 53–110; 110–225; 225–465; 465–850; 850–1,600; 1,600–3,000 MHz. Within these frequency sub-ranges, each filter is adjusted by the variable capacitance diodes, which are controlled by DAC. Pre-selector units of 0.009–30 MHz consist of four band-pass filters for the frequency sub-ranges 9–1,500 kHz, 1,500–4,500 kHz, 4.5–12 MHz, 12–30 MHz. At that point, the lowest sub-range filter is non-adjustable, but three other filters are adjustable with the bandwidths of 1 and 2 MHz. Pre-selector DAC and switches are controlled by the micro-processor control unit. The individual pre-selectors adjustments for each module are stored in a nonvolatile storage device, which allows for the achievement of the small transfer factor irregularity – both within the bandwidth and within the full operating range. The frequency converter unit is used for the radio signal frequencies transfer to the intermediate frequencies, for its amplification and image channels suppression, and consists of three frequency conversion units and IF filters. The signal from the VHF-UHF range pre-selector unit enters the frequency conversion input 1, which converts the input signal to the intermediate frequencies 300 and 706.8 MHz, in accordance with the frequency plan. The signal enters the first mixer (M1) input, and the signal from the adjustable LO is applied to another mixer input. The LO signal is applied through the switch, from one of three sources: from external LO or from one of two internal oscillators: VCO1.1 or VCO1.2, selected in accordance with the frequency plan. As a result of the signal conversion, the signal of the first IF1 with the frequencies 300 MHz or 706.8 MHz is generated at mixer output, which arrives further to the IF1 filter unit. The values of the first intermediate frequency IF1 and the adjustable LO frequency of the frequency conversion unit 1 are listed in Table 3.10, for various receiver frequencies. From the filter unit IF1 output, the signal passes to frequency conversion unit 2. This unit converts the signal from the first IF to the IF2 signal at 41.6 MHz. The

90

3

Radio Receiver Applications for Radio Monitoring System

Table 3.10 First IF and LO Frequency Values Tuning frequency, MHz

LO frequency, MHz

IF frequency, MHz

from 25 to 464.8 from 464.9 to 1,000 from 1,000.1 to 1,599.1 from 1,599.2 to 2,600.6 from 2,600.7 to 3,000

from 731.8 to 1,171.8 from 764.9 to 1,300 from 1,706.9 to 2,305.9 from 1,899.2 to 2,300.6 from 1,893.9 to 2,293.2

706.8 300 706.8 300 706.8

signal from filter unit IF1 enters the second mixer input, and the signal from one of the LOs, selected in accordance with the frequency plan, is sent to the second input of this mixer. The IF2 signal at 41.6 MHz, from the mixer output, enters the IF2 filter unit and arrives further to the frequency converter unit 4 and to the IF outputs switch. The signal from the HF range pre-selector passes to the frequency converter unit 3 where it converts into IF 41.6 MHz. The frequency converter unit 4 converts the second IF signal into the third IF 10.7 MHz. The output IF signal enters two independent output connectors, and, for each, an IF value of 41.6 MHz or 10.7 MHz can be assigned by software. All LOs of the frequency converter unit are made on the basis of the frequency synthesizers. The unified reference signal frequency of 12.8 MHz, applied to the synthesizers, is generated by the temperature-controlled internal reference oscillator with relative instability 5 · 10−7 or by the external reference oscillator at frequency 12.8 MHz. The control unit provides the receiving command and controls the voltage and commands the transfer into the receiver units. The control unit consists of the microprocessor-controlled (MPC) unit, along with two DACs. The commands from an external source, e.g., from the ARK-CO module, arrive at the control unit via the serial bus of RS-485 protocol. DAC modules generate the control voltages for pre-selectors and the crystal reference oscillator. The power supply unit for the receiver converts the DC input voltage 9–16 V into a set of DC output voltages (–3 V; +5 V; +3.3 V; +15 V, +25 V), required for the unit’s power supply. The structural diagram of the power supply unit is shown in Fig. 3.55. The power supply unit consists of the secondary power supply (SPS) units, the input and output power supply filters, which reduce the output voltage rippling, and the voltage regulator, which provides the power supply on duty for the microprocessor. Let us consider the second module included in ARGAMAK DRR, namely, the DSP module. The structural diagram of the ARK-CO DSP module is shown in Fig. 3.56, and a photo of the printed circuit board with the mounted components is shown in Fig. 3.57. In this photo, we can see the ARK-CO module without the shields. After adjustment, its modules are exposed to the careful shielding. In the CO unit, the analog filters are provided, allowing unambiguously representation of the signal in digital form. Signal demodulation for an acoustical

ARK-PR5 “Argamak” Digital Radio Receiver

91 Power supply unit

External power supply

Power supply filter

9–16 V

Secondary power supply module 1

Power supply filter

–3 V

Power supply filter

+15 V

Power supply filter

+25 V

Power supply filter

+5 V

In/Out PS +5 V

Secondary power supply module 2

Voltage regulator Power supply filter

+3.3 V

Fig. 3.55 Structural diagram of power supply unit

check-up is executed by soft-hardware means that allow the number of permissible modulation types to increase without hardware modernization, and allow the weight and dimension parameters to change. The possibility of radio signal recording in vector form exists, for further technical analysis. This module provides simultaneous operation in the panoramic analysis and signal demodulation modes. The ARK-CO module has two channels that allow for its coherent signal processing. The module provides high performance when direction finding, radio signal demodulation, and when recording in vector form. It serves as the connecting link Analogous output Input filters

1 Channel1In 1 Channel 2In

2 Channel1In 2 Channel 2In

S w i t c h

S w i t c h

Band-pass filter 10.7±1 41.6±2.5 41.6±1 41.6±5 MHz Band-pass filter 10.7±1 41.6±2.5 41.6±1 41.6±5 MHz

Frequency converter 12.3–10.7 41.6–38.4 41.6–40.0 41.6–35.2 MHz Frequency converter 12.3–10.7 41.6–38.4 41.6–40.0 41.6–35.2 MHz

Analogous-digital unit Band-pass filter 1.6±1 3.2±2.5 1.6±1 6.4±5 MHz Band-pass filter 1.6±1 3.2±2.5 1.6±1 6.4±5 MHz

Signal processor

Interface

PLD

Signal processor

PLD

Digital receiver

Signal processor

Codec

A D C

RS485 Nº1 RS485 Nº2

A D C

fdig

fïp

USB

Digital receiver

I2C

Synthesizer unit f ref1 f ref2 12.8 MHz

9–16 V

Switch

Switch

Synthesizer

Reference oscillator

Synthesizer

Reference oscillator

Buffer

Power supply unit

+5 V –5 V +3.3 V +2.5 V

Fig. 3.56 Structural diagram of ARK-CO5 DSP unit

I2C Control processor with ROM

Audio unit

Control

Amplifier

Control

Buffer

Line output Head-phones

92

3

Radio Receiver Applications for Radio Monitoring System

Fig. 3.57 External view of ARK-CO5 DSP module

in RM complex control systems, ensuring the information exchange between PC or other controlling device, DSP processors, the radio signal converters and the additional equipment. Similar to the ARK-PS5 receiver, the ARK-CO unit is implemented as a module with dimensions 100 × 60 × 20 mm. An IF signal at frequency of 41.6 MHz or 10.7 MHz enters it. At IF 41.6 MHz, the unit provides analog signal processing with a bandwidth of 5 MHz or 10 MHz, but, at IF 10.7 MHz, it processes the signal with a 2 MHz bandwidth. This unit executes input signal filtering, its conversion into digital form, further signal processing, e.g., demodulation or spectrum calculation, and the preliminary amplification of the output audio signal. Moreover, the unit provides external device control via RS-485 serial interface, e.g., the ARK-PS5 receiving section. The module has two processing channels; each channel has two switched inputs. There are also two analog input-outputs and the digital differential input-output at 12.8 MHz frequency, to synchronize the operation with other devices. The module can operate in autonomous mode or can be controlled by the external PC through a USB2.0 bus. Moreover, the module has two serial ports of RS-485 standard for communication or control by the other devices. As we can see from the structural diagram, the ARK-CO module consists of three digital signal processors of Analog Devices-type AD2185 for the digital processing of the received signals and two controlling processors with Intel 8051 core to control the audio signals and the data exchange between the PC, the digital processors, and the ARK-PS5 module or other devices. The selection of the mentioned digital signal processors can be explained by the requirement to reduce the consumed power and the interference level created by these electronic devices. The presence of the built-in reference clock-frequency synthesizer can be considered a module feature that provides the possibility to connect the external high-stable reference oscillator, as well as the possibility to improve the signal AD conversion stability and hence, to increase the signal parameter measurement accuracy, e.g., its carrier frequency.

Conclusion

93

The DRR AD6620 – one for each channel – fulfils the demodulation, or filtering operation, and the signal decimation, to execute the vector analysis. The ARK-CO module can operate in three main modes: in autonomous mode from the remote control panel, under PC control through USB interface, and through RS-485 interface. The last mode can be used to create the net of complexes controlling through radio-modems.

ARGAMAK-I Panoramic Measuring Receiver Based upon the ARGAMAK DRR, the ARGAMAK-I panoramic measuring receiver was developed, and was certified as measuring equipment. The features of this receiver are the specifications required for measuring equipment. These specifications are ensured by the minimization of sensitiveness functions to parameter variations during the design phase, and also by using the digital technologies of path corrections. So, the first adjustable IF introduced allows us to virtually exclude the number of staggering frequencies at the panoramic analysis. The stability of the absolute section transfer function is achieved by means of its digital correction, taking into account the operating temperature. The application of the temperaturecontrolled reference oscillator provides reference frequency stability upon temperature change; and, the ability to adjust the frequency – with the help of specific software – eliminates the aging effect of the crystal resonator.

Conclusion In the present chapter, our main attention is placed on structural diagrams, the construction features, and the parameters of superheterodyne receivers with single or several frequency conversions. This receiver type is the most convenient for radio monitoring problems in the wide frequency range from units to thousands of megahertz and for digital signal processing at intermediate frequencies. The effectiveness of radio receiver applications in radio monitoring systems is defined by its main parameters: operating frequency range, amplitude-frequency response, standing-wave ratio at receiver input, selectivity on adjacent and spurious receiving channels, receiver sensitivity in the given bandwidth, linearity, phase noise of the frequency synthesizer, adjustment (re-tuning) speed, bandwidth, time of spectrum calculation, weight and dimension, and the exploitation complexity. In this chapter, we show that noise factor is, at present, the most universal parameter for RR sensitivity determination in the given frequency band, for estimation of the 2nd and 3rd order intermodulation intercept points, which are the most suitable for determining receiver linearity. Thus, knowing the noise factor and the intermodulation intercept points, one can easily find out the dynamic range of the receiver. The various RRs in use now for radio monitoring applications are discussed. These include the scanning receiver, the selective micro-voltmeter, the spectrum analyzer, the panoramic receiver, and the panoramic measuring receiver. We prove

94

3

Radio Receiver Applications for Radio Monitoring System

that the digital panoramic measuring receiver is the most suitable for radio monitoring applications. Finally, a short developmental history of hardware-software RM systems in the Russian Federation during last 20 years is considered in this chapter. The structural diagrams and the construction features of the Russian ARK-CT1, ARK-CT3, ARKPR5 panoramic digital receivers, and ARK-KNV4 converter, which are widely used in radio monitoring, measuring, and direction–finding systems are described.

References 1. Davies, J., and Carr, J.J., Newnes Radio and RF Engineer’s Pocket Book. 2nd Edition, Butterworth-Heinemann, 2000, 594 pp. 2. Baskakov, S.I., Radio Engineering Circuits and Systems (in Russian). Moscow, Vyschaya Skola, 2005. 462 p. 3. Kurganov, A.N., and Pavlyuk, A.P., Linearity Features of Measuring Radio Receivers (in Russian). Trudy NIIR, 2003. 4. Raucher, Ch., Janssen, V., and Minihold, R., Fundamentals of Spectrum Analysis. Munchen, Rohde & Schwarz, 2001. 5. Red, T.E., Arbeitsbuch fur den HF-Techniker. München, Franzis-Verlag GmbH, 1986, ISBN 3-7723-8151-0. 6. Poberezhskiy, E.S., Digital Radio Receivers (in Russian). Moscow, Radio i sviaz, 1987, 184 pp. 7. Rabiner, L.R., and Gold, B., Theory and Applications of Digital Signal Processing. Prentice Hall, New Jersey, 1975. 8. Digital Radio Receiving Systems: Reference Book. Zhodzizhskiy, M.I., Mazepa, R.B., Ovsiannikov, E.P. etc. (in Russian). Under edition of Zhodzizhskiy, M.I. Moscow, Radio i sviaz, 1990, 208 pp. 9. Rembovsky, A.M., Automated Radio Emission Monitoring – Problems and Facilities (in Russian). Special technologies. 2002. Special Edition. pp. 2–6. 10. Sergeev, V.B., Sergienko, A.R., and Pereversev. S.B., ARK-D1TP Panoramic Measuring Receiver. Special technologies. No. 3, 2004, pp. 50–57.

Chapter 4

Single-Channel and Multi-Channel Radio Signal Detection

Introduction Complexity of radio monitoring-problem solutions, as applied to the various types of signals, can differ considerably. Thus, for example, the detection of wide-band signals in the absence of a priori information is a serious problem. At the same time, narrow-band signal processing, which, as a rule, is described by a large SNR, can be solved successfully on the basis of a rather simple model of examined processes requiring a minimum of a priori information. Below, we consider the digitalprocessing algorithms for the narrow-band radio signal group, observed together on the wide-band additive noise background [1]. Let us assume the following conditions for radio environment investigation: • Analyzed random process has an unknown number of narrow-band components, generated by RES • A priori information on the carrier frequency, and on the type and parameters of the examined signal modulation, are absent. We know only that the signal spectrum width does not exceed some limit set in advance • Noise intensity data are absent, however, within the processed frequency bandwidth, restricted by the possible sample frequency values, the noise power does not change practically. Further, in this chapter, we shall interpret the signal detection task as the determination of the number of simultaneously-observed narrow-band signals and their location along the frequency axis. Let us formulate this task as follows. An unknown number M of narrow-band radio signals um (t), whose spectrum width dfm is limited and cannot be more than some definite value dfrch (radio channel width), acts within the frequency band F on the additive normal white noise ξ (t) background with unknown intensity σξ2 · The observed process can be written in the form: uin (t) = ξ (t) +

M

um (t, fm , dfm )

(4.1)

m=1

A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_4,

95

96

4

Sin (f)

df1

Single-Channel and Multi-Channel Radio Signal Detection dfM

df2

dfrch

f1

f2

F

fì

f

Fig. 4.1 Spectral power density of observed process

where f m is the central frequency of the um (t) signal spectrum. The signals do not overlap in requency, and are equally likely to be located in any part of the analyzed range. Their spectrum form is unknown (see, e.g., Fig. 4.1). It is required to determine the number M of the narrow-band signals included in the uin (t) structure, to estimate their spectrum width and central frequencies on the basis of observation results of one or several process (4.1) samples. As the theoretical basis of the stated-problem solution, we use the statistical synthesis theory of joint optimal algorithms [2]. In terms of this theory, the above-stated problem may be formulated as the problem of compound vector → → T λ = σξ2 , M, λ M estimation, where M is the signal number, falling into the ana M is the vector of the unknown parameters collection of the lyzed frequency band, λ signals being resolved. Vector x samples of the observed co-ordinates of the input random process, calculated by the registered realization uin (t), are the basis for vec determination. The probabilistic features of x samples depend on the true value tor λ M ). λ, and redefined by the conditional probability density W(x|λ) = WM (x|σ 2 ,M,λ ξ of sigThe search object is such a detection rule at which the received estimation λ nal number and parameters is the best of all those possible, on the basis of some criterion selected in advance. Independent from the selected optimization criterion, the solution of the stated problem is related to the research of the behavior of the above-mentioned condi called the likelihood functions. At the selection tional probability densities W(x|λ), of the time sample collection of uin (t) process, each value included in x depends in a complex way on parameters of all resolved signals, and determination of the estimation is rather time-consuming. On the contrary, in the freoptimal vector λ quency domain, the dependence of the spectral sample probability characteristics on the resolved signal parameters is local, and the samples themselves can be considered as independent [3], which leads to the likelihood function factorization, and simplifies both the optimal algorithm search process and the obtained processing

Single-Channel Signal Detection

97

procedure itself. Due to this, we shall use the spectral samples of the uin (t) process as a vector of observed co-ordinates x.

Single-Channel Signal Detection Discrete Fourier transform (DFT) N−1 1 c˙ (n) = uin (kT)e−i2πnk/N N

(4.2)

k=0

is the generally-accepted spectral characteristic of signals represented by the fixed volume samples. However, the complex sample c˙ (n) collection is not the optimal vector of the M , included in the W(x|λ) observed co-ordinates x because the parameter vector λ structure, contains the unknown amplitudes and phases of the spectral components of all signals. At likelihood function maximization, with respect to these parameters (in the absence of a priori information about the phase spectrum of the detected signals), the ratio of the real and imagined parts of c˙ (n) will affect on only the noninformative vector of the phase spectral components for the solved problem. The signal set detection results depend on the amplitude vector of the analyzed process components only, therefore, one can take the sample collection 1 |˙c(n)|2 R R

XR (n) =

(4.3)

r=1

of the process uin (t) energy spectrum as the vector x. If the spectrum (4.3) is obtained by R samples of observed process uin (kT), each N samples in volume, the signal um (t) can be presented in this spectrum dnm = N · T · dfm

(4.4)

by the spectral samples, beginning from the sample which is more right than nm = int[(fm − 0.5dfm )NT], where int[.] is the function of the integer part of the number-taking. The spectral sample sub-set, which satisfies the inequality nm ≤ n ≤ nm + dnm and characterizes the m−th signal, we designate, in the future, as θm . In fact, these samples represent the independent random variables following the non-central χ 2 - distribution: Wncχ 2 (x; a, λ, J, δ) = x−a J/2−1 −[δ+(x−a)/λ] J x−a 1 e = λJ/2 (J/2) 0 F1 2 , δ λ λ where 0 F1 (α,z) is the generalized hyper-geometric function. These distribution parameters are:

(4.5)

98

4

Single-Channel and Multi-Channel Radio Signal Detection

a = 0, λ =

σξ2 RN

, J = 2R, δ =

2 RNSmq

4σξ2

,

(4.6)

where σξ2 is the additive noise power; q = n − nm is the sample sequence number within the useful signal spectrum; and, Smq is the amplitude of the signal um (t) component at the appropriate frequency. The noise samples of the averaged energy spectrum (4.3) represent the random variables, following the central χ 2 −distribution 1 Wχ 2 (x;a,λ,J) = λ(J/2)

x−a λ

J/2−1

e−(x−a)/λ

(4.7)

with J = 2R degrees of freedom and parameters a = 0, λ =

σξ2 RN

.

(4.8)

The corresponding samples sub-set we shall designate as θn . Taking into account the selected observation vector x, we can formulate the detection problem as follows. There exists an averaged energy spectrum (4.3) of the random process, which consists of noise, and may be of several useful narrow-band signals um (t). Taking into consideration the mutual independence of the XR (n) samples, this spectrum can be described by the likelihood function = W(x|λ)

M

m=1

⎡ ⎣

dnm

q=1

⎤ Wncχ 2 (xnm +q |Smq )⎦

Wχ 2 (xn )

(4.9)

n∈θn

where Wχ 2 (xn ) is the central χ 2 − distribution with parameters (4.8), and Wncχ 2 (xnm +q |Smq ) is the non-central χ 2 − distribution with parameters (4.6). It is required to determine the number M of the narrow-band signals included into the uin (t) structure, and also to estimate the values nm and dnm collection defining the frequency limits of the detected signals. At the stated problem solution, it is necessary to take into account the following: • This problem is distinguished by its essential a priori uncertainty, i.e., not only the resolved signal parameter values, but also their distribution laws are a priori unknown. To overcome this a priori uncertainty, the statistical theory of optimal algorithms synthesis recommends the use of the adaptive approach [4], in accordance with which we should use its maximal believable estimates instead of the unknown variables. • If we have a priori information concerning the probabilities of PM arriving at the specific number M of narrow-band components in the process (4.1) structure, the resolution and estimation of signals can be executed on the basis of the

Single-Channel Signal Detection

99

maximal a posteriori probability criterion. The appropriate processing algorithm a posteriori probabilities of arriving at the should maximize over the vector λ averaged energy spectrum x for the set of hypotheses H0 , H1 ,. . .,HM max , concerning the presence of a specific number of um (t) signals in the uin (t) structure. Unfortunately, this maximization can be arranged only in the form of an iteration procedure, and that is why this optimal algorithm will be extremely complicated for the calculations. • The optimal procedure corresponding to the case of probabilities PM absence, based, for example, on the maximal likelihood criterion, is merely slightly simpler for calculation because the necessity of the mutual maximization of the multi-dimensional likelihood function over the parameters set is kept, in this case. • Together with the optimal approaches to the signal detection problem, which have a great complexity, one can offer a number of quasi-optimal methods for this problem solution, based on the concept of quasi-full resolution [5]. This method allows the resolution problem to be considered as a problem of the complex detection of some signal set at which the signal resolution problem can be described by the normalized probability of false detection Pfd1 and by the probability of missing the separate, arbitrarily-selected, signal Pms1 . At that point, the algorithm ensuring, at the fixed probability of false detection Pfd1 , the minimal value of the probability of the missing signal Pms1 , is the optimal one (as per the Neiman-Pirson criterion). In practical implementation of RM systems, the operating-rate factor is extremely important. The threshold detection procedures discussed below permit us to simplify essentially the calculating complexity of the processing, at the expense of some decreasing of obtained estimation accuracy. In this connection, the exact same quasioptimal algorithms are the most claimed in practice. The simplification offered by the threshold quasi-optimal procedure, consists in the fact that the global maximization of the likelihood function (4.9) is changed by the preliminary division of all sample sets to “noise” θn and “signal” θm (m ≥ 1) sub-sets, on the basis of the differences between the distribution laws of noise and signal spectrum samples. Let H1 be a hypothesis asserting the affiliation of a XR (n) sample, having the value xn , with the spectrum of some signal um (t), the amplitude of which is Smq at the frequency n/NT; and let H0 be a hypothesis of the sample with a value xn affiliation with the frequency axis segment, where there are no useful signals. In this case, in accordance with the above-mentioned statistical characteristics of the noise and signal spectra samples, the hypotheses H1 and H0 likelihood ratio has the form: L(xn ) =

Wncξ 2 (xn |Smq ) Wξ 2 (xn )

= λ1−R e−δ 0 F1 (R, δ

xn ). λ

(4.10)

At any δ>0 with xn growth, the likelihood ratio is monotonically increased. This means that the optimal rule of spectral samples division into sub-sets θn and θm consists in the comparison of xn with the some threshold xthr :

100

4 Single-Channel and Multi-Channel Radio Signal Detection

H1 > Xn < X1¨ 1ˆ ð H0

(4.11)

The threshold level unambiguously defines the error probabilities contained in the assignment of some number of maximal noise spectrum samples to the sub-set θm , and some missing signal components, in this case, falling into sub-set θn . Since it is assumed that the decision about the number of observed useful signals is made on the basis of the assigned sub-sets θm of the signal spectrum components, the selection of xthr exclusively affects on the parameters of the resolution algorithm as a whole. The threshold algorithm of the narrow-band signal detection includes the following steps: 1. The set of signal spectrum components is assigned on the basis of the rule (4.11). 2. Signal spectrum samples, located in order of their increasing number n, are divided into several sub-ranges Qj , 1≤ j ≤ J. At that point, the signal sample with the least number XR (n1 ) always belongs to the sub-range Q1 , and the affiliation of the subsequent samples is determined by the iterative grouping rule XR (nk+1 ) ∈

Qj , if XR (nk ) ∈ Qj and nk+1 − nk ≤ dnrch Qj+1 , if XR (nk ) ∈ Qj and nk+1 − nk > dnrch

(4.12)

where dnrch = N · T · dfrch is the maximal possible sample number representing the signal spectrum in one radio channel. 3. For each sub-range, the minimal and maximal number of samples falling in this sub-range is determined, and the number of useful signals in each sub-range is determined by the rule: Mj = int[1 + (nmax j − nmin j )/dnrch ]

(4.13)

where int[.] is the symbol of the integer part of the number-taking. 4. For Mj signals formed in the sub-range Qj , the most probable variant of the sample set nmin j ...nmax j is division into the sub-sets θm , and, for each of these sub-sets, the central frequency and the signal-spectrum width estimation is executed by the correlation approach. Let the frequency band of the simultaneous analysis correspond to the averaged spectrum-sample set with the numbers nmin ≤ n ≤ nmax . From the above-mentioned description of the threshold algorithm, we find that, for false signal resolution to occur, it is enough that at least one signal from the noise spectral sample exceeds the level xthr1 . If the number of such samples is equal to Nn , the probability of false signal detection is:

Single-Channel Signal Detection

101

Pfd1 = 1 − P{XR (n) < xthr1 , n ∈ θn } = 1 − FχN2n (xthr1 )

(4.14)

where Fχ 2 (x) is the function of noise sample distribution. To ensure acceptable detection quality, it is necessary to provide a low false-detection probability, at least Pfd1 < 0.01, and hence Fχ 2 (xthr1 ) ≈ 1, and thus Pfd1 ≈ Nn [1 − Fχ 2 (xthr1 )]. Evidently, before the analysis is finished, the exact number of noise samples Nn is unknown; however, to assure fulfillment of the requirements for the permissible false-detection probability, it is quite reasonable to accept Nn as equal to the analysis range width expressed in the samples Noper = nmax − nmin . In this case, the permissible probability of threshold excess, by the separate noise sample of the averaged spectrum, will be defined by the value ε = 1 − Fχ 2 (xthr1 ) ≈ Pfd1 /Noper .

(4.15)

For central χ 2 −distribution, the argument χε2 for which P{χ 2 ≥ χε2 } = ε, is approximately defined by the equation

2 χε2 ≈ λJ 1 − + xε 9J

2 9J

3 (4.16)

where J is the degree of distribution freedom, and xε is the percentage point of Gaussian distribution, which can be calculated using the approximate expression: xε = t −

c0 + c1 t + c2 t2 1 + d1 t + d2 t2 + d3 t3

(4.17)

where t = ln (1/ε2 ); c0 = 2.515517, c1 = 0.802853, c3 = 0.010328, d1 = 1.432788, d2 = 0.189269, d3 = 0.001308. Using the mentioned approximations and Equation (4.8), we get the following rule for the calculation of the threshold xthr1 , which divides the spectrum samples into noise and signal sub-sets at the single-channel processing: xthr1 (σξ2 )

=

σξ2 N

1 1− + xε 9R

1 9R

3 (4.18)

where xε is defined by (4.17), and the probability of ε is defined by Equation (4.15). The algorithm offered above, for the sample division into noise and signal sub-sets, assumes the noise intensity σξ2 as a known quantity, on the background of which the narrow-band components are observed, in particular, see (4.18). However, in accordance with the initial statement of the detection problem, this intensity is a random parameter, and not only the value σξ2 itself, but its distribution law, too, is unknown. The attempt to use the optimal maximal-likelihood estimate of noise levels leads to an algorithm that is complicated, in terms of the calculations. In this vein, let

102

4 Single-Channel and Multi-Channel Radio Signal Detection

us analyze the following approach to the noise level estimate, one that is a bit less labor-intensive. For the central χ 2 − distribution with J degrees of freedom and parameters (4.8), the expected value is defined as: MXn =

σξ2 Jλ = . 2 N

(4.19)

Hence, if the narrow-band signal’s um (t) location on the frequency axis is known, and the number Nn of noise spectrum samples is large enough, it is expedient to estimate the noise level σξ2 by means of the noise spectrum components only, rejecting the signal components. Of course, in this case, the quality of the σξ2 estimation decreases; but, if the signal component number is a small portion of all the spectral samples, the accuracy reduction is negligible and the need to determine the unknown amplitudes of the signal components Smq is eliminated. In practice, however, the sub-set of the noise samples θn , as a rule, is not known in advance and can be determined only approximately. Let us analyze two different methods that allow the division of the noise samples sub-set, at little computing expense. 1. If we ignore the presence of signal components and define the initial noiseintensity estimate by the equation 2 = NMXn = σest1,0

N Noper

nmin +Noper −1

xn

(4.20)

n=nmin

where nmin is the sample corresponding to the left frequency-band border of the simultaneous analysis, and Noper is the range width in the samples, then the obtained value (at M > 0) will obviously be too high. Nevertheless, the following equation will be true for signals with considerable intensity, even for a similar inaccurate estimation, 2 ) ≈1, P XR (n) n∈θm > xthr1 (σest1,k

(4.21)

that allows us to attribute these samples to the signal ones. Ascertainment of the fact that the energy spectrum-samples portion is affiliated with the signal sub-set allows the substitution of the rough estimate (4.20) by the more accurate 2 σest1, k =

N xn Noper − Ns

(4.22)

n∈θ / sig

where Ns is the number of signal samples found to exceed the threshold, and k is the iteration number. As a rule, after two to three similar iterations, the estimate (4.22) becomes already rather close to the true value of the noise power.

Single-Channel Signal Detection

103

2. Another approach to estimate noise level is based on the fact that, in any signalfree frequency segment, the expected value of the energy spectrum samples coincides with the variable σξ2 /N, but, at the “signal” segments, it increases. If we smooth the spectrum – whose width B W does not exceed the maximal (on frequency) interval Bmax between signal spectra (Fig. 4.2) – over frequencies, by the window, such smoothing will cause noticeable spectrum distortion only near the central frequencies of the signals um (t), but, at noise intervals, it will decrease the spectral sample variance by BW -times, without changing the expected value.

XR(n) B0

dn1

B1 = Bmax

dn2

B2

dnM

BM

Approximated result of smoothing

dnrch

nmin

n1

n2

nM

n max

Fig. 4.2 Window width selection for smoothing in frequency

At BW > 20, the smoothing effect becomes seriously apparent, and both minimal and maximal samples (among “purely noise”) differ slightly from σξ2 /N. The minimal samples belonging to the θn sub-set are the least simultaneous among all samples of the smoothed spectrum. As a result, it is not necessary to determine the useful signal location in the frequency range, in order to estimate noise intensity. The main complexity occurring at the practical application of the offered approach is the fact that, before finishing the signal resolution procedure, information on the size of the widest interval Bmax between the useful signal spectra is absent. If the maximal possible signal number M0 which can be present in the analyzed frequency spectrum, is known, we may ensure that, among the signal spectra, at least one interval can be found, which exceeds BW min =

nmax − nmin − M0 dnrch . M0 + 1

(4.23)

However, the estimate (4.23) assumes the presence of all M0 signals and their location in the frequency through strictly similar intervals, which is unlikely. In this connection, it is expedient to accept the maximal frequency interval between the spectra as being equal to the BW estimate satisfying the equation:

104

4 Single-Channel and Multi-Channel Radio Signal Detection

FB max (BW |M0 ) = δ

(4.24)

where δ ≈ 0 is the probability of the absence of an inter-spectral interval in the analyzed frequency range, which exceeds BW . Since Equation (4.24) is oriented on the maximal possible number M0 of narrowband signals, in reality, the probability of situation Bmax < BW occurring will be much less than δ, i.e., practically zero. The final curves permitting us to find the Bmax value are shown in Fig. 4.3. From the presented data, we can see that a δ increase to 10–15%, compared with the case δ = 0, which corresponds to the rule (4.23), allows the smoothing window width BW to grow by 2–2.5 times.

Bw Nn 0.35

δ = 0.15

0.25

δ = 0.10

0.15

δ = 0.05

0.05

δ=0 0

5

15

25

M

Fig. 4.3 Recommended width of smoothing window BW

Thus, the second method to estimate noise intensity assumes the calculation of the smoothed energy spectrum X˜ R (n) =

1 BW

n+(B W /2)

xi i=n−[(BW −1)/2]

(4.25)

where the smoothed window width BW is defined on the basis of (4.24), and usage of the least of the obtained values for determination of noise intensity: 2 ˜ R (n). = N · min X σest2 nNoper

(4.26)

Analysis shows that both of the above-considered, quasi-believable, estimations of noise intensity, found on the basis of maximal component elimination (1st type estimations), as well as determined by minimal smoothed energy spectrum (2nd type estimations), have the bias (systematic) error, the magnitude of which grows with the increasing of the simultaneously-observed signal number. Nevertheless, the character of this error is different: the 1st type estimations, as a rule, are too high,

Characteristics of Single-Channel Detectionof Narrow-Band Signals

105

while the 2nd type errors are too low. Practice shows that the best results can be achieved by combining the two previous procedures into a unified procedure. To perform it, noise intensity is preliminarily estimated by rule (4.26) and, after that, is made more exact on the basis of (4.22). At large volume of the presented spectral data, and for not very large loading of the analyzed frequency range by the narrowband signals (less than 30%), the accuracy of such type of algorithm does not yield to the strictly-optimal estimation procedure. Now, assuming that the error at noise-level determination is negligibly small, we can estimate signal detection quality by the single-channel algorithm formulated above as follows. The signal components of the averaged energy spectrum follow the non-central χ 2 −distribution: −δ

Fncχ 2 (xn ) = e

∞ k δ γ (0.5n + k,xn ) k! (0.5n + k)

(4.27)

k=0

where (x) is gamma-function, γ (a, x) is non-full gamma-function, and other designations and parameters are as introduced above [see (4.6)]. Since signal presence is registered after the threshold is exceeded by only one spectral sample, applying to the signal described by dnm a set of spectral components with Smq amplitudes, the probability of missing the signal is defined as Psm1 =

dnm

Fncχ 2 (xthr1 |Smq ) ·

(4.28)

q=1

Evidently, the increasing of the spectral sample number representing the signal causes the decreasing of its signal-missing probability. In this connection, it is expedient to analyze the dependence of the signal-missing probability on radio environment parameters for the most complicated case, when the signal is represented by only one spectral sample. The curves corresponding to this case are shown in Fig. 4.4. Using samples of the observed, random process uin (t) of large size (N>>1), we can ensure each signal spectrum representation, by a large enough number of spectral samples and in accordance with (4.28), to provide a rather high quality of signal detection.

Characteristics of Single-Channel Detection of Narrow-Band Signals Let us apply the above-mentioned results to the detection of a specific signal of small duration, whose position on the frequency axis is unknown. To do this, let us estimate the probability P{tdet ≤ Ts } of this signal-detection during the definite time interval, and its dependence on the signal’s properties and the ARM system’s features.

106

4 Single-Channel and Multi-Channel Radio Signal Detection Psm1

Pfd1 = 0.001

{ { R=1

0.1

Pfd1 = 0.005

R=2

Pfd1 = 0.01

Pfd1 = 0.001

0.03

P fd1 = 0.05

Pfd1 = 0.005 0.01 P fd1= 0.01 0.003 Pfd1= 0.05 0.001

a)

6

14

h2

22

Psm1

{

R =16 0.1

{

R =8

0.03

P Pfd1 = 0.001 0.001 fd1= Pfd1= 0.005

0.01

Pfd1 = 0.01 Pfd1= 0.05

0.003 0.001 0 1.5

2.5

3.5

4.5

h2

b) Fig. 4.4 Qualitative characteristics of single-channel algorithm at small (a) and large (b) number of spectrum averaging actions R

At panoramic spectral analysis, an ARM system scans cyclically L frequency bands with width F, one of which contains the signal under detection. The analysis of the separate frequency band by the ARM system assumes the taking and processing successively of R samples, after which the ARM receiver is re-tuned on a new frequency. As a result, the processing of each frequency band requires τproc1 = τret + Rτsamp

(4.29)

where τret is the time of ARM receiver re-tuning on a new frequency; and, τsamp is the time of taking (and processing) one sample. As a consequence, an ARM system’s single cycle of analysis has the length

Characteristics of Single-Channel Detectionof Narrow-Band Signals

107

τcyc = Lτ proc1

(4.30)

and ARM system performance in this mode is defined by the variable: g=

LF F = · Lτproc1 τret + Rτsamp

(4.31)

For signal detection, it is necessary, firstly, that, before time interval Ts ends, the ARM system, at least once, has time to re-tune on the frequency band containing this signal, and, secondly, that the child spectral peak after this emission was higher than the detection threshold – in at least one of the observation cycles (if there were several peaks). The small duration of the detected signal assumes that the maximal permissible duration of detection procedure Ts should not exceed several seconds. Nevertheless, time of taking and processing of the single sample is defined by portions of milliseconds, and the coincidence of ARM equipment re-tuning time with the signal beginning or finishing is unlikely; therefore, at ARM system tuning on the appropriate frequency band corresponding to the signal, we can consider that the signal will be observed in this analysis cycle during all R samples. Let us introduce the designation k = int[Ts /τcyc ]

(4.32)

to represent the minimal possible number of analysis cycles, for which the ARM system will have the time to fulfill, during time Ts . Thus, the actually observed cycle number Kcyc is equal to k or k + 1, and probabilities of the appropriate events are determined as P{Kcyc = k} = 1 − v

(4.33)

P{Kcyc = k + 1} = v

(4.34)

where v=

Ts Ts Ts − int −k = τcyc τcyc τcyc

(4.35)

is the fractional part of the ratio of observation interval and the analysis cycle duration of the ARM system. If, during the Ts time interval, the detected signal falls Kcyc times into the ARM system scanning band, using the averaged, over R, spectral estimate samples, the probability of its missing the signal at all of these cycles is Psm (R, Kcyc ) =

nfin

n=nst

Kcyc Fncχ 2 (xthr |J = 2R, δ =

h2n )

(4.36)

108

4 Single-Channel and Multi-Channel Radio Signal Detection

where xthr is the detection threshold determined by (4.18); nst ,...,nfin is the range of the spectral sample numbers corresponding to the detected signal; and, h2n is the SNR on power, at the frequencies of these samples. If the sample averaging is not used, and the signal detection is fulfilled on the basis of each sample individually, the probability of signal-missing in all Kcyc cycles is Psm (R, Kcyc ) =

nfin

RKcyc Fncχ 2 (xthr |J = 2, δ =

h2n )

.

(4.37)

n=nst

Thus, the full detection probability can be calculated as P{tdet < Ts } =

v[1 − Psm (R,1)] at Ts ≤ τcyc (1 − v)[1 − Psm (R, k)] + v[1 − Psm (R, k + 1) at Ts > τ (4.38)

where R is a samples number accumulated by the ARM system, without the retuning to the new scanning band; k and v are defined from (4.32) and (4.35), and, for calculation of Psm (..), we must apply (4.36) when using the averaged spectrum, or (4.37) for the non-averaged spectrum.

Single-Channel Detection of Radio Signals With POFT Let us apply the above-obtained results to the detection problem for signals with programmable operating frequency tuning (POFT). Such signals possess high energy security and are used for the transfer of a large number of messages, alternatively re-adjusting among them by the random rule. For POFT signals, the averaged power (over the large time interval) falling on each frequency is comparable with the natural additive noise, which is the serious obstacle for this power detection. At the same time, the instantaneous power accompanying the data transmitting at the given frequency is rather large; therefore, at coincidence of the radio environment analysis moment with the time interval of using the frequency, the radio emission event can be registered. When using the modern, high-performance, ARM systems in the panoramic spectral analysis mode, the intensity peaks corresponding to typical radio signals with fixed operating frequency are observed during many analysis cycles, as a rule. The peaks caused by POFT signals are almost always singular because the probability of recurring usage of some frequency – strictly after the time interval equal to one ARM system analysis cycle length – is very small. Thus, the presence of the activity peaks periodically observed in some frequency sub-set, which can be described by the large enough amplitude and by the approximately constant spectrum width, is the characteristic sign of POFT signals used within the frequency range controlled by an ARM system. Of course, the single spectral peaks can be caused by the noise also, but, in this case, the peak amplitude,

Single-Channel Detection of Radio Signals With POFT

109

as a rule, is not large, and the spectrum width fluctuates randomly. As a result, if the observation at the given frequency of at least two single peaks with the spectrum width corresponding to a POFT signal is selected as a criterion of some frequency affiliation to the POFT frequency set, then the probability of the mistaken classification is small.

Probabilistic Features of the Frequency Observation Time The time diagram of the mutual operation of a POFT signal transmitter and an ARM system is shown in Fig.4.5. This diagram assumes that the ARM system scans cyclically L frequency bands in which M frequencies are located randomly, and that these positions are used at the formation of the POFT signal. As in section “Characteristics of Single-Channel Detection of Narrow-Band Signals”, we assume that, at the analysis of the single scanning band by the ARM system, before the re-tuning to the new frequency, R samples are accumulated and mutually processed. However, now, the interval τem1 of the continuous usage of the separate frequency by the POFT signal is very small (it coincides with the time of the sample set-taking, up to the order of magnitude). Therefore, the observation time ξ of the separate frequency by the ARM system is random variable and the maximal possible ξ value is τmax = min τem1 ;Rτsamp .

(4.39)

Let us select any arbitrary m− th position from the total M number of frequencies used with a POFT signal, and examine the one arbitrary activity case for this frequency, having analyzed the random observation time ξ by the ARM system of

Frequency position of POFT signal

F, MHz

L scanning bands of RM system

POFT signal operating frequency variation

fM

… l-th scanning band

fm

f3 f2 f1

τ ret t0 t0 + tobs1

Rτ samp t0 + τcyc

Fig. 4.5 Time diagram of POFT signal detection

t,s

110

4 Single-Channel and Multi-Channel Radio Signal Detection Partial or maximal registration region

Fig. 4.6 Calculation of distribution law of active position observation time ξ

–τem 1

0

Region of missing the emission of the frequency position R•τsamp

t, s τcyc–τem1

Moment of analysis start for l-th frequency band

the given radio signal emission. Since both the moments of radio signal emission and the transition to the necessary frequency band analysis are random variables, let us attach the time axis to the moment of the monitoring start for exactly the same (l−th) frequency band, which contains m− th frequency (see Fig. 4.6). The emission start we consider equiprobable within the ARM system analysis cycle (4.30). At the activation moment falling into the interval from −τem1 to Rτsamp for m− th frequency, the time interval of this frequency usage at least partially coincides with the analysis time of the l− th spectrum band and hence, the registration of this event is potentially possible. If the active position falls into another part of the cycle, the frequency observation is principally impossible, due to the emission and monitoring interval non-overlap. As a result, the probability density ξ can be written as

τem1 + Rτsamp δ(z)+ Wξ (z) = 1 − τcyc 2 |τ −Rτ | , 0 ≤ z ≤ τmax + em1 τcyc samp δ(z − τmax ) + τcyc

(4.40)

If the sample number η, during which one succeeds in observing the selected position on the POFT signal, is of interest, we can follow the relations

0 < r < rmax 2τsamp /τcyc , P{η = r|τem1 ≥ Rτsamp } = τem1 − (R − 1)τsamp /τcyc , r = rmax

(4.41)

0 < r < rmax 2τsamp /τcyc , P{η = r|τem1 < Rτsamp } = τem1 + (R − 2rmax + 1)τsamp /τcyc , r = rmax (4.42) where rmax =

R, τem1 ≥ Rτsamp int[0.5 + τem1 /τsamp ,τem1 < Rτsamp

is the maximal possible number of samples within the overlap interval.

(4.43)

Single-Channel Detection of Radio Signals With POFT

111

Probability of Separate Frequency Registration Let each accumulated sample be processed independently from the others and a decision concerning the signals presence at any frequencies is made if the detection threshold is exceeded in at least one of these samples. In this case, for each specific number of samples η = r, during which, at the next emission, the ARM system will observe some m− th frequency, the conditional probability of successful spectral peak registration can be represented in the form: Pr = 1 −

nst +n

Fncχ 2 (xthr |δ = h2n )

nfin

r Fncχ 2 (xthr |δ = h2n

(4.44)

n=nfin −n

n=nst

where xthr is the detection threshold of signal samples defined by (4.18), nst , nfin are the start and the end samples included in the monitored spectrum peak, n is the permissible measurement error of the signal spectrum width (in samples), h2n is the SNR on power at the frequency of the tested sample. The freedom degree number of the non-central χ 2 − distribution J = 2, parameter λ of this distribution is equal to σn2 /N. Combining, on the basis of total probability relation, the cases, differing by the number of the observation samples η, we get the following equation for unconditional probability of m− th frequency registration at its next emission: Preg1 =

rmax

P{η = r}Pr

(4.45)

r=1

where the probabilities P{η = r} depending on τem1 are calculated on the basis of (4.41) or in accordance with (4.42). The case, when the samples accumulated during the time of l− th frequency band observation are used for the averaged spectrum calculation, is the alternative one to that considered above. Due to the mistiming of the emission and monitoring moments, m− th frequency, including to POFT signal, is observed during the random time ξ ≤ Rτsamp instead of the time intervalRτsamp . In this connection, the spectral peak caused by the activity of this position will have the lesser intensity, namely, the specific value ξ = z will correspond to the following probability of the true registration of the appropriate spectral peak: P(z) = 1 − ×

nst +n n=nst n fin

Fncχ 2 xthr |δ =

n=nfin −n

z 2 Rτsamp hn

Fncχ 2 xthr |δ =

z

×

2 Rτsamp hn

(4.46)

where, contrary to (4.44), the parameters of the non-central χ 2 − distribution are defined by the relations J = 2R, λ = σn2 /(RN).

112

4 Single-Channel and Multi-Channel Radio Signal Detection

Taking into consideration the continuous character of the random observation time ξ , the total probability of m− th frequency registration at its next emission should now be described by the equation: τmax

Preg1 =

P(z)Wξ (z)dz

(4.47)

0

Estimate of the Total Number of Registered Frequencies If the transferred message duration is Ts millisecond, the total number of frequencies, which will chaotically change each other, and appear and disappear during the emission, can be obtained as A=

Ts (1 + γ )τem1

(4.48)

where γ is the reserve factor on the POFT signal transmitter re-construction (γ ≈ 0.1). Let us examine the regular a− th message segment with one frequency duration. The probability that the specific m− th frequency will be used, and, moreover, registered, on this segment is equal to pa = Preg1 /M

(4.49)

where Preg1 is the unconditional probability of frequency registration during its next emission, calculated in accordance with (4.45) or (4.47). It can be shown that the registration cases on the specific m− th frequency will represent Poisson event flow. In accordance with this, the probability that the specific m− th frequency will be registered at least twice during all data transmitting time, will be defined as p2+ = 1 − (1 + Apa )e−Apa .

(4.50)

This probability is extended equally to all M frequencies used by the POFT signal, and their registrations happen relatively independently from each other. As a result, the total number of frequencies Mreg , which can be registered during the current data transmitting time, will represent the binomial random variable described by the distribution series k k p2+ (1 − p2+ )M−k . P{Mreg = k} = CM

(4.51)

Single-Channel Detection of Radio Signals With POFT

113

Optimization of ARM System Parameters Practical application of the calculation formulas obtained above shows that, for modern ARM systems and POFT signals with message length in some seconds, the most probable number of revealed frequencies is not large. At the same time, there is a set of parameters, which are selected at the design stage and/or ARM system application stage, which essentially affect on the success of the given problem’s solution. Improvement of the results, in particular, can be achieved using multi-channel ARM equipment or a receiver with lesser re-tuning time to the new frequency, because, usually, time spent on completing the transients is comparable or even exceeds the data accumulation (processing) time. At the same time, there is also one important parameter, the variation of which does not require any material expenses. It is the sample number R, which is processed before the re-tuning on the new frequency. The typical curve of the most probable number of the revealed frequencies as a function of this parameter is shown in Fig. 4.7. Mreg

16

1 12

8

2

4

0

3

9

15

21

R

Fig. 4.7 The most probable number of frequencies revealed for POFT signal with the length 40 s at τret = 4 ms, τsamp = 0.32 ms, L = 60, M = 32, τem1 = 3.125 ms, h2n = 3; 1 – for the averaged energy spectrum; 2 – without averaging

We shall take into consideration the following factors. 1. At small number R of the averaging samples, a significant portion of time is spent on the re-tuning from one frequency to another, because, many times, the receiver re-tuning interval τret exceeds the time of the observation and processing of the spectral sample τsamp 2. At the averaged sample number growth till R ≈ τem1 /τsamp , the portion of time that is wasted unproductively on receiver re-tuning decreases, and the probability

114

4 Single-Channel and Multi-Channel Radio Signal Detection

that registration of the spectral peak increases allows ensuring a Mreg value close to the maximum. 3. Further growth of the R parameter is accompanied by improvement of the ratio of data processing time to the receiver frequency re-tuning time, on the one hand; but, on the other hand, the energy of the short-term observed POFT signal fragment lengthens now to the whole length of the averaged spectra set, which causes a decreasing of the effective SNR. As a result, the long segment of R values is formed, within which the Mreg value virtually does not change. 4. Finally, at redundantly large R, the influence of SNR reduction, determined by (4.46), “outweighs” the growth of the sample taking time, and system operation effectiveness starts to decrease essentially. The loss in the number of revealed positions observed at each individual data processing session, is caused by the small probability of weak signal detection (h2n = 3), when using the non-averaged spectrum. At large SNR, the situation partially changes, and rejection of the spectrum average for large R may allow the detection of more frequencies, compared with the average application. The reason for such deflection is the fact that, at large SNR, the probability of successful spectral peak registration – even on the basis of the single sample – seems to be very large, and independent sample analysis eliminates the typical averaged spectrum negative effect of a “decreasing of effective SNR value”. Nevertheless, in similar cases, the initial segment of the curve Mreg (R) keeps the type shown in Fig. 4.7, i.e., at small sample number, the averaged spectrum application is evidently more effective. From a practical point of view, the overstating of averaging samples is disadvantageous. It causes a reduction in the information renewal rate, in the monitored frequency band, and, therefore, the search of such number of samples R, accumulated before the re-tuning to the new frequency, is of considerable interest. Then, simultaneously with the high effectiveness of the POFT signal-frequency reveal, we can keep the high operation rate. The sample number Rˆ is found optimal, which ensures a probability value (4.51) close to maximum, in conformity with the large range of initial data. The fulfilled analysis shows that, for the most part, the sample number Rˆ is defined by SNR. The value Rˆ is more than 1, only for fulfillment of

R 10 6 Fig. 4.8 Optimal number of averaging samples vs. SNR, in cases differing from τsamp << τem1 << τret

2 0

1

3

5

h2n

Single-Channel Detection of Radio Signals With POFT

115

the following conditions: τsamp << τem1 << τret . If even one of these inequalities is broken, the dependence Rˆ on h2n will take the form shown in Fig. 4.8. Thus, the values R = 3 − 6 are optimal for the detection of POFT signal frequencies.

Detection Characteristics For the problem of POFT signal detection, the definition of criterion, on the basis of which a decision is made that a POFT signal has been detected, is very important. Since we have information only concerning the detected signal’s features – that it changes the operating frequency many times per second – and we have no a priori data as of yet, we shall agree to use the registration of several (not less than three) frequencies (FP) in the monitored range, as this criterion. Thus, we shall treat as open the question of whether these positions belong to a single RES or to several RES, operating perhaps independently each from the other; however, the fact itself– that a programmable operating frequency tuning (POFT) is being used for information transfer – can be considered as prescribed, and the POFT signal presence as detected. To understand the curves mentioned below, it is useful to take into account the following factors. 1. Relatively small specific width of the scanned frequency range, as the ratio L of this range width to the simultaneously-observed width of the spectrum analyzer, affects the detection characteristics of POFT signals by the panoramic spectrum analyzer, under other equal conditions. The exception is the situation when, due to the expansion of the simultaneous scanning width, the total spectral sample number representing POFT signals decreases so much that it leads to a drastic reduction of the probability of true spectral peak detection. If the spectrum resolution on frequency does not change, the L number reduction, which should be subject to analysis of the frequency bands, unambiguously corresponds to POFT signal detection probability growth. 2. At the POFT signal detection criterion accepted above, number M of the used frequencies affects on the detection characteristics rather weakly. Really, M growth means that each position is used more rarely and is more difficult to reveal. Thus, the time interval required, to reveal all POFT signal positions, increases as a geometric series with M growth. At the same time, it is not necessary for the detection application to reveal all POFT positions; it is enough to find at least three used frequencies, and it is easier and easier to discover any three positions from the increasing M number. Variation of M in a wide range weakly signifies the POFT signal detection characteristics. 3. The total length of the data transfer Ts by POFT signal and the duration τem1 of the usage of a separate frequency by POFT signal (this parameter does not include the 10% guard time interval for re-tuning) are undoubtedly a reflection of the total probability of POFT signal detection, but the influence of these parameters is similar for any used equipment and for any detection methods. Therefore,

116

4 Single-Channel and Multi-Channel Radio Signal Detection

while conducting the analysis, we can use the arbitrarily-selected values for Ts and τem1 variables. 4. It was shown above (section “Optimization of ARM System Parameters”) that the value R = 3 − 6 usage is optimal for detection, therefore, at investigation of the detection characteristic, we use R = 4. 5. All other parameters, such as a time for taking the separate sample τsamp , the receiver re-tuning time from one frequency to another τret , and the number of parallel used detection channels, are strictly related to system performance [see (4.31)]. In accordance with the above-mentioned facts, the effectiveness of the POFT signal detection problem solution is defined, in the first place, by the intensity of the detected signal and by the performance g of the ARM system used, as well as by the criterion of detected signal conformity to the POFT signal frequency. Two sets of detection characteristics of a POFT signal in the range 2 GHz-width are listed below. The first set (Fig. 4.9) is oriented to the POFT signal-reveal method, assuming the reveal is not less than two short-term spectral peaks at one and the same frequency during Ts emission time of POFT signal. Figure 4.10 shows detection characteristics, which are distinguished by the fact that even once-detected short-term spectral peaks with the expected spectrum width are treated as POFT signal frequencies. In these figures, curve 1 corresponds to Ts = 20 s, h2 = 3; curve 2 to Ts = 10 s, h2 = 6; curve 3 to Ts = 5 s, h2 = 10. Analysis of the obtained curves shows that, while searching in the wide frequency range, one can rely on short-term POFT signal detection only, when using ARM systems with high and ultra-high performance. For ARM systems with lesser performance, the probability of short-term POFT signal detection does not exceed 30%, even for the optimal choice of detection parameters. We can recommend the double- and more-times arrival of spectral peaks at some frequency as a criterion of fixing frequencies, only when using an ARM system with ultra-high performance. It is more reliable to fix the frequencies on the basis P{tdet < Ts}

P{tdet< Ts} 0.9

0.9

0.7

1

0.7

2

2

0.5

0.5 3

0.3

0.3 0.1 0

a)

0.1

1 800 1000

4000

3

8000 g, MHz/s

0 800 1000

4000

8000

g, MHz/s

b)

Fig. 4.9 Detection characteristics of POFT signal at frequency detection, on the basis of not less than two signal peaks and using the independent samples (a) and the averaged spectrum (b)

Double-Channel Detection of Narrow-Band Signals

117

P{tdet < Ts}

P{tdet < Ts}

0.9

0.9

0.7 0.5

1

2

0.7 2 3

0.5 3

1 0.3

0.3

0.1

0.1

0 800 1000

4000

0 800 1000

8000 g, MHz/s

a)

4000

8000 g, MHz/s

b)

Fig. 4.10 POFT signal detection characteristics at determining the frequencies on the basis of the first spectrum peak, and using the independent samples (a) and the averaged spectrum (b)

of the first-arrived spectral activity peak, with further recheck of the obtained frequency set. Powerful signals are most effectively detected on the basis of independentlyprocessed spectral samples, but signals with lesser intensity are better detected on the basis of the averaged spectrum, at the averaged samples number 3–6. Taking into consideration that gain in the probability of powerful signal detection, achieved at the expense of spectrum accumulation refusal, is insignificant, and that averaging usage is principally important for weak signals, we can recommend detecting signals on the basis of the averaged spectrum, accumulated during 3–6 independent samples.

Double-Channel Detection of Narrow-Band Signals As with the similar considerations for the single-channel processing case, it follows that, at double-channel detection, it is advantageous to use the spectral characteristic set of the observed random process, as an observation vector. At double-channel processing, the magnitude values of the averaged sample values of the energy crossspectrum X˙ csR (n) =

1 c˙ sr (n)c∗rr (n) R R

(4.52)

r=1

will serve as these characteristics, where c˙ sr (n) is the spectrum of the r− th sample of uin (t) process in the “signal” channel, and c∗rr (n) is the spectrum of a simultaneously-obtained sample in the “reference” channel. The spectral samples c˙ sr (n) and c∗rr (n), with n numbers that correspond the noise segments, are small in magnitude, and chaotically change by phase from one sample to another, and therefore, the spectral samples (4.52) for these frequencies will also be small, by modulus. On the contrary, for “signal” segments of the frequency axis, the spectral samples c˙ sr (n) and c∗rr (n) have larger intensity and change

118

4 Single-Channel and Multi-Channel Radio Signal Detection

correspondingly (keeping the phase difference), therefore the value X˙ csR (n) is much more significant. The mentioned distinctions, by analogy with the singlechannel case, allow the division of the total sample set (4.52) into the “noise” θn and “signal” θm (m ≥ 1) sub-sets, which serve as a basis of the double-channel quasi-optimal procedure construction for signal detection. It can be shown that, if this cross-spectrum is obtained over the large sample number R of the uin (t) process, its signal spectral components follow the generalized Rayleigh distribution law Wgen Ray (x|Smq ) =

x σXmq2

2 ) −(x2 +c2 )/(2σXmq

e

I0

xc 2 σXmq

,x ≥ 0

(4.53)

.

(4.54)

with parameter

σXmq

! 2 " 2 " σξ2 σξ Smq # = + , 2N 4 RN

c=

2 Smq

4

The noise spectrum samples are distributed on Rayleigh law WRay (x) with σ parameter defined as σX0

σξ2 = √ . N 2R

(4.55)

Taking into account these distributions, the likelihood function for the crossspectrum samples will take the form: = W(x|λ)

M

m=1

⎛ ⎝

dnm

⎞ Wgen Ray (xnm +q |Smq )⎠

q=1

WRay (xn ).

(4.56)

n∈θn

To obtain (by analogy with the single-channel case) the new quasi-optimal detection procedure, it remains only to define more exactly the calculation rule for the threshold dividing the spectral samples into signal and noise sub-sets, corresponding to the distribution of signal and noise samples X˙ csR (n) . For the probability of false signal detection, we can write the equation: Pfd1 = 1 − P X˙ csR (n) < xthr2 ,

( Nn n ∈ θn = 1 − FRay (xthr2 ).

(4.57)

It should be noted, however, that the mentioned, in (4.56), Rayleigh’s distribution is correct, within the assumption that the number of spectrum averages R is large. At small R, the threshold must be increased. As a result, in order to calculate the threshold dividing the samples into noise and signal sub-sets, we can recommend the following equation:

Comparison of Single-Channel and Double-Channel Processing

xthr2

σξ2

=

119

2

σξ 4 1+ R + 0.75 N

−

ln ε R

(4.58)

where the probability ε is defined as before by (4.15), and the first factor is a correction describing the real distribution difference from the Rayleigh ones. Since the signal components of the energy cross-spectrum follow the generalized law of Rayleigh distribution with parameters (4.54), and the signal presence is registered by the threshold exceeding even a single sample, in conformity with the um (t) signal, characterized by a set of dnm spectral components, the probability of missing a signal is determined as follows: Psm1 =

dnm

Fgen Ray (xthr2 |Smq ).

(4.59)

q=1

Having considered the most complicated case, where the detected signal is represented by only one spectral sample, let us plot the appropriate detection curves in Fig. 4.11. Psm1

Psm1

{{

Pfd1=0.001 Pfd1=0.005

R=1

0.1

R=2

0.1

0.01

0.01

Pfd1 = 0.01 Pfd1 = 0.05

{ { R=8

R =16

Pfd1 = 0.001 Pfd1 = 0.005

Pfd1= 0.01 Pfd1= 0.05

0.001

0.001

0

14

6

h2

a)

0

1.0

2.0

3.0

h2

b)

Fig. 4.11 Qualitative characteristics of double-channel algorithm for small (a) and large (b) number R of spectrum averaging

Comparison of Single-Channel and Double-Channel Processing It is well known that χ 2 − distribution has the following numerical characteristics: M1 {χ } = a + 2

J + δ λ, 2

D{χ } = 2

J + 2δ λ2 2

(4.60)

where λ is a parameter describing the random variable value spread, J is a freedom degree number, and δ is the distribution non-centrality parameter.

120

4 Single-Channel and Multi-Channel Radio Signal Detection

Taking this into consideration, the noise samples of the energy spectrum can be described by the mean value m1 {XRn (n)} =

σξ2 N

for n ∈ θn

(4.61)

and the signal samples by the mean value m1 {XRs (n)} =

2 Smq

4

+

σξ2 N

for n ∈ θm .

(4.62)

As applied to generalized Rayleigh distribution, the mean value is defined as M1 {Xgen Ray } = σ

π 2

1+

c2 2σ 2

I0

c2 4σ 2

+

c2 I1 2σ 2

c2 4σ 2

e−c

2 /(4σ 2 )

(4.63) where I0 (x) and I1 (x) are the modified Bessel functions of the zero and first order. As a result, the noise sample mean value of the averaged energy cross-spectrum is * ) m1 X˙ csR (n) = σX0 π/2 at n ∈ θn

(4.64)

where σX0 is determined by Equation (4.55). For the signal sample, a similar mean value can be obtained by means of substituting into (4.63) σXmq and c parameters determined by (4.54). On the basis of the comparison of signal component excess over noise components, the gain of the

ν

aver ,dB

R = 14

R = 10

5 R=6

3

R=2 1

0

3

9

15

h2

Fig. 4.12 Gain in SNR provided by the double-channel processing procedure compared with the single-channel one

References

121

double-channel processing procedure over the single-channel one can be presented in the form: * ) ) * m1 X˙ csRs (n) m1 XRn (n) * ) ) * , dB. (4.65) vaver = 20 lg m1 X˙ cs Rn (n) m1 XRs (n) The calculation results of the vaver exponent are shown in Fig. 4.12. From the presented data, we can see that, as we increase the signal intensity and the averaging number, the achieved gain grows. The gain in the mean value of SNR provided by the double-channel processing procedure is 1–5 dB, depending on the averaging number. For R > 30, it can achieve 10–12 dB.

Conclusion In this chapter, qualitative relations are obtained for threshold selection, for the situation where, during the detection (single-channel and double-channel), detection of the narrow-band signal group is executed simultaneously, i.e., determination of their number and the main parameters. It is shown that the SNR gain in mean value at double-channel processing, compared with single-channel processing, is 1–5 dB, depending on the averaging number R. The possibility of single-frequency signals, and signals with dynamic frequencytime structure, to be detected by panoramic digital RR during the finite time interval is confirmed. For searches in the wide frequency range, we can expect to detect short-term POFT signals only in the case when ARM systems with high and ultra-high performance are used. We can recommend the use of double, and more, spectral peak arrival at some frequency as a detection criterion, only when an ARM system with ultra-high performance is used. It is more reliable to fix the frequencies on the basis of the first peak of spectral activity, with further re-check of the obtained frequency set. Powerful signals can be most effectively detected on the basis of independentlyprocessed spectral samples, and weaker signals on the basis of the averaged spectrum, with the number of averaged samples ranging from 3 to 6.

References 1. Rembovsky, A.M., and Tokarev, A.B., Automated Radio Monitoring on the Basis of SingleChannel and Double-Channel Data Processing (in Russian). Moscow, Vestnik MGTU (Bauman Moscow State Technical University), No 3(56), 2004, pp. 42–62. 2. Levin, B.R., and Shinakov, Yu.S., Mutually Optimal Algorithms for Signal Detection and Its Parameter Evaluation (review) (in Russian). Radiotekhnika i elektronika, No. 11, 1977, pp. 2239–2256.

122

4 Single-Channel and Multi-Channel Radio Signal Detection

3. Marple Jr, S.L., Digital Spectral Analysis with Applications, Prentice-Hall, Inc., 1987. 4. Repin, V.G., and Tartakovskiy, G.P., Statistical Synthesis Under a priori Ambiguity and Adaptation of Information Systems (in Russian). Moscow, Sovetskoe Radio, 1977, 432 pp. 5. Shirman, Ya.D., Signal Resolution and Compression (in Russian). Moscow, Sovetskoe Radio, 1974, 360 pp.

Chapter 5

Multi-Channel Digital Receivers

Introduction The hierarchical structure of ARM equipment, its composition, functions, and the main technical specifications for solving ARM problems in industrial centers and in open terrain, to reveal CEE in the monitored regions and its boundaries, and to check the measure’s effectiveness to prevent information leakage, is well described in Chapter 2. The solution to all these problems in corpore is related to certain financial difficulties. The way to solve this situation is to use multifunctional equipment to execute the greatest number of ARM tasks with the highest effectiveness and at the least expense. In the present chapter, we consider how to execute ARM tasks on the basis of one piece of equipment only, namely, the multi-channel (in the minimal configuration of double-channel) digital receiver. The detection of radio signals with the dynamic frequency-time structure is one of the additional possibilities of this equipment. The high rate of panoramic analysis ensures the possibility to reliably reveal these signals. Figure 5.1 shows the panoramic spectrum analysis results with the extreme value accumulation in 1, 20, and 50 s after detection begins, when observing the real POFT communication line. In this case, the spectral analysis mode with the rate of 140 MHz/s is used. The presented result analysis shows that, 30–50 s after POFT signal detection, the equipment allows the determination of most frequency positions. One of the most important parameters of the technical means for the search and detection of new RES, i.e., defining the radius of the RM post’s action zone, is the real sensitivity of the panoramic analysis equipment.

Panoramic Multi-Channel Receivers The most important technical specifications, as well as the ARM complex’s functionality, depend directly on the fact that receivers are used at RM posts. For a long period of time, Russian manufacturers were constrained to use foreign communications receivers, which were widely represented in the Russian market, in combination with digital processing units. The undoubted advantage of this approach A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_5,

123

124

5

Multi-Channel Digital Receivers

(a)

(b)

(c)

Fig. 5.1 Signal spectral diagrams at emission detection of the real POFT communication line 1 s (a), 20 s (b), and 50 s (c) after the operation’s start

ARK-D11 Double-Channel Complex

125

consists in its low cost. Nevertheless, the communications receiver approach has several serious shortcomings, among which we can emphasize the limitations on functionality, performance, and on several other technical characteristics. In the end, communications receivers do not ensure high indices when used for the purposes of spectral analysis. The construction and circuitry features of digital receivers, considered in Chapter 3, allow for the creation of multi-channel radio monitoring systems, on their basis. Prior to 2005, the double-channel DRR of the third generation (ARK-CT2) and the fourth generation (ARK-CT3) were produced. On their basis, the RM, DF, and CEE systems were developed and manufactured. Nevertheless, it was very difficult to create the portable DRR with more than two channels, on the basis of 3rd and 4th receiver generations, due to their large dimensions and the absence of builtin software and hardware functions to provide the simplicity of coherent receipt channel formation. The features of the 5th generation equipment of the ARGAMAK family, namely, the ARK-PC5 radio signal frequency converter modules and the ARKCO digital processing modules, which were discussed in section “ARGAMAK-I Panoramic Measuring Receiver”, allowed for the design of portable and carried systems of various purposes, on their basis, including the multi-channel RM systems. Let us consider the features of the ARK-D11 double-channel complex and the ARK-RD8 eight-channel complex, which are intended for multifunctional RM and CEE complexes, for multi-channel radio monitoring, and for mono-pulse radio direction-finding systems.

ARK-D11 Double-Channel Complex The ARK-D11 complex (Fig. 5.2) is intended for ARM and CEE task-solving. It looks like the ARK-D7K complex [1, 2], but it has better technical specifications, namely, 1.5 times higher performance at lesser weight and power consumption. The ARK-D11 complex provides the following: • Double-channel or single-channel search and CEE reveal, accumulation and management of database on sources and results processing • Correlative receipt of noise-like signals • Radio signal recording in vector form on the computer hard disk • Technical analysis and measurement of the radio signal parameters • Wire network monitoring. When combined with an additional direction-finding antenna system, the ARK-D11 equipment can also be used for DF problems, and can be combined with antenna switchers and external modules, for distant monitoring of remote premises (see Chapter 11). The main specifications of the ARK-D11 complex are determined

126

5

Multi-Channel Digital Receivers

Fig. 5.2 ARK-D11 central unit

by DRR ARGAMAK characteristics, which are given in section “ARGAMAK-I Panoramic Measuring Receiver”. Other specifications of the complex are listed in Table 5.1.

ARK-RD8M Multi-Channel Complex To improve the performance of ARM equipment in the wide frequency range, we can use two main approaches. The first one consists in the bandwidth widening of simultaneously-processed frequencies up to 30–100 MHz values at the appropriate growth of ADC digit capacity and signal processor power. Such an approach is undoubtedly warranted at low radio-range loading or at the processing of wide-band signals from a single source. When RM technical means are used in urban conditions, where a large number of radio emission sources operates, the presence of at least one powerful REE in the simultaneous analysis bandwidth leads to DRR overload. Therefore, under these conditions, another approach is warranted, which, in essence, consists of using several receiving sections, each of which has comparatively narrow frequency bandwidth from 2 to 10 MHz. The tuning frequency of each receiving section is displaced, with respect to the adjacent sections, by the pass-band.

ARK-RD8M Multi-Channel Complex

127

Table 5.1 Additional technical specifications of ARK-D11 Parameter

Value

Sensitivity of receiver in AM and FM modes, μ V not worse than, Panoramic analysis and fast signal search Panoramic spectral analysis rate using, MHz/s: ARK-CO2 module with the 2 MHz bandwidth at 3 kHz FFT discreteness ARK-CO5 module with the 5 MHz bandwidth at 6 kHz FFT discreteness ARK-C010 module with the 10 MHz bandwidth at 12 kHz FFT discreteness Technical channel reveal of the information leakage (25–3,000 MHz): Coupling loss of antenna switchers between the channels, dB, not less than Total system sensitivity (transmitter power in the 8 m × 8 m size premise, detected with the probability 0.99), μ W The capability to detect radio microphones with AM, narrow-band and wide-band FM, static technical shielding inside the premises (without changing in time the shielding parameters) Wire network monitoring (in the 0.05 kHz–30 MHz range) with the detected signal level, μ V: In range 0.05–10 kHz, less than In range 10–1,000 kHz, less than In range 1–30 MHz, less than Input resistance of the wire network external sensor, k not less than Wire network voltage, V Radio signal recording, technical analysis and parameter measurement: Processing frequency bandwidth/resolution capacity

0.5

Double-channel radio monitoring, demodulated transmission recording: Demodulator frequency bandwidth

Discreteness of signal tuning Modulation types

Operating temperature, weight, dimensions, power consumption: Interval of operating temperatures, ◦ C Power supply voltage, V: From AC network From on-board car electric net From autonomous battery Consumed power, VA, not more than Dimensions (length × width × height), mm Basic set weight, kg

1,200 3,000 6,000

40 100 Yes

1,000 100 10 1,000 Up to 400 5 MHz/15 kHz, 250 kHz/500 Hz, 120 kHz/240 Hz, 50 kHz/100 Hz, 25 kHz/50 Hz, 9 kHz/20 Hz, 6 kHz/12 Hz 250 kHz, 120 kHz, 50 kHz, 25 kHz, 9 kHz, 6 kHz, 3 kHz 1 Hz AM, FM, SSB, telegraphic messages –20 to +50 90–250 10.6–13.6 9–16 20 486 × 398 × 194 11

128

5

Multi-Channel Digital Receivers

Let us examine the structure of the ARK-RD8M multi-channel panoramic radio receiver, which can have up to eight ARK-PC5 independent radio signal frequency converters controlled by a single PC, and, correspondingly, up to four ARK-CO double-channel, digital-processing modules with the simultaneous analysis bandwidth 2, 5, and 10 MHz in each channel. Moreover, to ensure the maximal panoramic analysis rate, the ARK-C5 high-performance double-channel special calculators, which reduce the spectrum calculation time up to 100 μs, are included in the ARK-RD8M structure. The functional diagram of the ARK-RD8M equipment, with four physical channels of frequency selection, each from 25 to 3,000 MHz, is shown in Fig. 5.3. The ARK-RD8 central unit consists of the antenna splitter, the control panel, four boards of ARK-PC5 signal-frequency converters, two boards of ARK-CO2 double-channel analogous-digital processing, two boards of ARK-C5 double-channel special calculators, the power supply unit, and a cooler. The active audio system and the power supply unit of the complex from AC network 90–240 V are mounted in separate cases. The control and audio board can simultaneously execute audio recording in eight channels and control the receiver. The ARK-C5 special calculator is designed for digital spectrum analysis of the signals, and its application provides almost double growth of the system’s operation rate. The technical specification of the receiving channel is defined by DRR ARGAMAK (see section “Introduction”). The additional characteristics of the multichannel receiver are listed in Table 5.2. The control of the receiver is executed with the help of an external PC. For receiver operation, usage of the following customized software packages (SP) is provided:

25-3000 MHz

ARK-PC5

41.6 MHz 6.4 MHz

RS-485

ARK-AB1-4

41.6 MHz 25-3000 MHz

25-3000 MHz

ARK-PC5

ARK-C5

ARK-CO10

ARK-PC5

6.4 MHz 41.6 MHz

HUB USB 2.0 6.4 MHz RS-485

ARK-CO10

25-3000 MHz

ARK-PC5

41.6 MHz

Fig. 5.3 Functional diagram of ARK-RD8M

ARK-C5 6.4 MHz

ARK-RD8M Multi-Channel Complex

129

Table 5.2 Additional technical specifications of the ARK-RD8M multi-channel receiver Parameter Panoramic analysis, fast signal search: Simultaneous spectral analysis bandwidth in each channel, MHz For ARK-CO2 For ARK-CO5 For ARK-CO10 Total bandwidth of the simultaneous spectral analysis for 4–8 channels, MHz For ARK-CO2 For ARK-CO5 For ARK-CO10 Speed in the operating range for 4–8 channels, GHz/s: For ARK-CO2 (discreteness 3 kHz) For ARK-CO5 (discreteness 6 kHz) For ARK-CO10 (discreteness 12 kHz) Power supply voltage, V From AC net From car on-board net From battery Operative radio monitoring, demodulated message recording: Number of monitored channels Frequency number in the task on scanning Number of ranges in the task on scanning Radio signal tuning discreteness, Hz Demodulation types

Radio signals recording, technical analysis: Processing frequency bandwidth/Resolution capacity For ARK-CO2

For ARK-CO5

For ARK-CO10

Value

2 5 10

8–16 20–40 40–80 4.5–6 16–32 32–64 90–250 10.6–13.6 12 4–8 255 255 1 AM, FM, SSB, telegraphic messages

2 MHz/15 kHz, 250 kHz/500 Hz, 120 kHz/240 Hz, 50 kHz/100 Hz, 25 kHz/50 Hz, 9 kHz/20 Hz, 6 kHz/12 Hz 5 MHz/15 kHz, 250 kHz/500 Hz, 120 kHz/240 Hz, 50 kHz/100 Hz, 25 kHz/50 Hz, 9 kHz/20 Hz, 6 kHz/12 Hz 10 MHz/30 kHz, 250 kHz/500 Hz, 20 kHz/240 Hz, 50 kHz/100 Hz, 25 kHz/50 Hz, 9 kHz/20 Hz, 6 kHz/12 Hz

130

5

Multi-Channel Digital Receivers

• SMO-PA software for panoramic analysis • SMO-MCRM software for multi-channel radio monitoring • SMO-TA software for technical analysis. The receiver provides the possibility to fulfill the following functions: • Panoramic spectral analysis of radio signals at the joint operation of all channels, from a single antenna • Panoramic spectral analysis of radio signals in each channel, with a separate independent task for each channel • Accumulation of spectra panorama in the given frequency range, saving the panorama of the range loading for further analysis • Statistical analysis of the panoramic analysis results • Coherent in-couples multi-channel signal processing, for use in various applications • Search of active radio channels in the frequency range or by frequency list, automatic registration of detected emission sources • Radio IF signals recording in vector form on computer hard disk • Technical analysis, determination of modulation type, and radio signal parameter measurement • Recording of the demodulated messages on computer hard disk • Playback of the demodulated signals recorded on computer hard disk • Deferred demodulation of the given frequency channel on the basis of the recorded IF signal on computer hard disk • Listening of demodulated signals in real-time scale • Forming the reports with radio monitoring and signal analysis results. It should be noted that, at operation in the panoramic spectral analysis mode with the connection of all channels to a single antenna, it is possible to achieve the total performance of 64 GHz/s at discreteness 12.5 kHz.

SMO-MCRM Customized Software Package Software Purpose and Performance Capabilities SMO-MCRM customized software, when in operation with the multi-channel complex, provides the following functions: • Simultaneous operation with eight receivers, with the possibility to listen to one receiver on operator choice • Receiver tuning correction during listening or recording • Manually-controlled frequency selection for listening and recording • Automated adjustment to RES on the basis of range list, by specifying the receiver tuning step within the range

SMO-MCRM Customized Software Package

131

• Automated storage in database of data obtained during RES search, recording arrangement • Fast task correction during scanning • Representation in panorama of all monitored radio signals during the current operation session, keeping it on the hard disk and loading from the file • Representation in real-time scale of all signals and the active task conditions • Registration of demodulated transmissions on computer hard disk in WAV-format standard, their export to the external carriers • Audio records playback on the micro-telephones via computer audio card, autonomous operation without the equipment during playback • Audio records stenographing support with saving of the text in database • Forming the reports with the radio monitoring results and their export to the external carriers • RES accumulation in database, with the possibility to arrange its classification on types • Operation in the structure of distributed complex, with the possibility to get commands via the network. The SMO-MCRM software window (Fig. 5.4) contains (top-down) the following: • The main menu • Toolbar comprising control buttons and the dropdown menus for selection of the mode, search tasks and scanning tasks • Status bar and receiver control panel (the status bar is at the left, the control panel is at the right) • Multi-page notebook with the panorama bookmarks, the search task table, the scanning task table, the audio records table with the playback • Total status bar.

Software Operation Modes The software has four main independent operation modes: • • • •

Manual mode for RES switching to listening/recording Automatic operation mode, in accordance with the formed task Playback mode for the accumulated audio records Panoramic spectral analysis mode.

The manual mode is intended for the manual RES switching to listening or recording. The frequency can be introduced manually or chosen from any software table, as well as from the panorama. In the manual mode, any receiver can

132

5

Multi-Channel Digital Receivers

Fig. 5.4 Main window of SMO-MCRM software

be switched on, irrespective of the fact of whether it operates in automatic mode or not. The features of automated operation modes are illustrated in Table 5.3. In automatic mode, starting the task is distributed among all active receivers. We can exclude some of the receivers from the automatic mode, having reserved them for the manual operations.

Table 5.3 Automatic modes of software operation Purpose

Function

Search on list Search on frequency list: when source is detected, it starts recording Search on ranges Search on ranges with the defined step: when source is detected, it starts recording Detection Search on ranges with the defined step: when source is detected, its frequency is noted down to the scanning task table Combined One part of the receiver operates in “Detection” mode, i.e., searches on ranges and saves the obtained frequencies in the scanning task table. The other receiver part operates in “Search on list” mode, in accordance with the task dynamically formed by the first part of the receiver.

References

133

Playback mode is intended for playback of audio information recorded on the computer hard disk via the computer audio card. This mode does not depend on the operation modes of the receivers or on the equipment connection at all. This mode provides such performance capabilities as: • Widened positioning functions inside the audio record • Representation of estimation plot of voice or noise presence in the audio signal, allowing operator to find the informative segments in long audio records • Definite audio record fragments cycling • Audio records filtering and sorting on the different criteria • Consecutive playback of all audio records included in the table • Audio records stenographing with saving of the text in the database • Forming the reports with the radio monitoring results and its export to the external carriers. Panoramic spectral analysis mode. In the ARK-RD8 equipment, the first receiver is used to obtain the spectrum for panoramic analysis. In this connection, it is not used in the modes related to the step-by-step frequency adjustment within the range. It can be used only in the “Manual” and “Search on list” modes, as well as in the “Spectrum” and “Panorama” special modes.

Conclusion It is shown in this chapter that, under the conditions of radio frequency range high loading, when a large number of RES operate, application of the multi-channel panoramic DRR consisting of several devices of frequency selection is justified. These DRR may include up to eight ARK-PS5 signal converter modules and up to four ARK-CO2, ARK-CO5, and ARK-CO10 digital signal processing modules with the simultaneous analysis of bandwidths 2, 5, and 10 MHz in each channel. Additionally, to achieve the maximal panoramic analysis speed of 64 GHz/s at the spectrum discreteness 12.5 kHz, the ARK-S5 high-performance double-channel special calculators can be included in the structure of this equipment. Control of the multi-channel DRR can be fulfilled by the SMO-PA software, for panoramic spectral analysis, by the SMO-MCRM software, for multi-channel radio monitoring, and by the SMO-TA software, for signal technical analysis. Effective multi-channel RM problem solutions, under the control of the SMOMCRM software, are illustrated on examples of the multi-channel ARM complexes.

References 1. Rembovsky, A.M., Search Facility Effectiveness Increase for Automated Radio Monitoring (in Russian). Special technologies. No. 4, 2003, pp. 40–47.

134

5

Multi-Channel Digital Receivers

2. Rembovsky, A.M., Combined Solution of Radio Monitoring Problems on the Basis of Multifunctional Facilities with Double Reception-Analysis Section (in Russian). Special technologies. No. 5, 2003, pp. 2–7.

Chapter 6

Modulation and Signal Types in Modern Radioelectronic Means

Introduction Modern radio space is occupied by various emissions: from the hand telegraph to complicated time-varying radio signals with digital modulation and coding. In recent years, rapid growth in the number of radio channels for broadcast service has been observed, as well as for systems of official, personal, and amateur communication. Data stream organization is changing dramatically, with progress in the field of microelectronics promoting the further transition to digital methods. The mastering of new high frequency ranges is moving forward at a high rate. At present, communication technologies ensure the global exchange of any data. This chapter focuses on the main methods of modulation at data transfer via radio channel. Preliminary data on modulation theory are discussed, which are necessary for the main measurements of radio signal parameters and their technical analysis. This chapter contains references to the appropriate technical literature, to allow the reader further information on communication systems.

Administrative Division of the Frequency Spectrum The intensity of radio frequency spectrum (RFS) usage grows constantly and is accompanied by the critical need to eliminate mutual interference. Therefore, on the international scale, procedure development is constantly being performed on operation co-ordination of various communication systems, which differ by construction principles and by technical solutions. The international standards for the parameters of various radio equipment applications are stated. At that, it is very important to provide recommendations regarding unified radio monitoring methods for the acting radio system operation, and to perfect frequency allocation methods for the radio communication and broadcast networks, and the spectrum distribution at the used frequency spectrum extension. To this end, the joint intellectual and technical resources of various states established a number of international technical organizations, which deal with the

A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_6,

135

136

6

Modulation and Signal Types in Modern Radioelectronic Means

standardization problems of the various radio system applications. Besides the International Telecommunication Union (ITU), the European Telecommunication Standard Institute (ETSI) plays an important role in development of the standards on radio communication and broadcasting systems. ETSI developed the set of standards on the digital systems of mobile and stationary communication and broadcasting, which are used for the manufacture of equipment distributed widely all over the world. Important solutions are offered at conferences held under the Conference European of Post and Telecommunication (CEPT), the World Administrative Radio Communication Conference (WARS), and the World Technical Standard Communication Conference (WTSC). Moreover, there are corporate standards on several communication systems, for example, the standards of the Intelsat organization, for satellite communication. Radio communication regulations of the Russian Federation [1] developed in consideration of the international standards; they contain the main features of the frequency band distribution between the radio services and the main juridical aspect of radio spectrum usage on Russian territory by the civil and military systems. It includes: • The frequency band distribution table between the Russian Federation radio services in the frequency range from 3 kHz to 400 GHz • Frequency layout for the main radio services • The main juridical acts regulating the order of frequency allocation and assignment, the monitoring on their usage in the Russian Federation, the order of manufacture, purchasing and import into the Russian Federation and the radio electronic means (REM) usage on Russian territory, the procedure for action type licensing in the area of radio communication and radio- and TV broadcasting in Russia, the procedure for REM certification, and the list of the main rules and standards on REM technical characteristics defining electromagnetic compatibility (EMC). Radio regulations of the Russian Federation can serve as the guidelines for radio communication and radio broadcast system operators, concerning the rules and procedures of radio communication, TV and broadcast means usage in the Russian Federation. Manufacture and purchasing of radio equipment used in the Russian Federation territory should be performed with consideration of the national peculiarities of radio frequency spectrum usage represented in the frequency band distribution table between radio services. The allocation of radio frequency spectrum, in accordance with International regulations of radio communication and the radio wave propagation peculiarities in specific ranges, are shown in Table 6.1.

Administrative Division of the Frequency Spectrum

137

Table 6.1 Administrative division of radio frequency spectrum Range

Name

Propagation peculiarities

3–30 kHz 100–10 km

Very low frequency (VLF, super-long or myriameter waves).

Long-haul radio in daytime and at night, practically without evident fading influence. Reception under water is possible (at depths of several meters).

30–300 kHz 10–1 km

Low frequency (LW, long or kilometer waves).

Intra-continent communication in daytime and at night. Sometimes global communication is possible. Limitation of communication distance often occurs in the high-frequency part of the range, especially at daytime.

300– 3,000 kHz 1,000–100 m

Medium frequency (MW, medium or hectometer waves).

Communication in daytime at distance <1,500 km, at night at distance <4,000 km. Under particularly favorable conditions at night, global communication is possible. In daytime, the communication distance decreases at frequency growth. In the upper part of the range, the influence of solar activity can be evidently observed.

3–30 MHz 100–10 m

Short frequency (SW, short or decameter waves), HF (high frequency).

The application of reflections from ionosphere layers is typical. Depends on solar activity.

3–6 MHz 100–50 m

Communication in daytime at distance < 600 km, at night at distance < 3,000 km. Under particularly favorable conditions, global communication is possible. Not-strongly dependent on solar activity.

6–10 MHz 50–30 m

Communication in daytime at distance < 5,000 km, at night, intra-continent communication is possible, even global communication. Evidently depends on solar activity.

10–20 MHz 30–15 m

Intra-continent communication in daytime and at night, global communication is often possible. Strongly depends on solar activity.

20–30 MHz 15–10 m

Intra-continent communication. Additionally, global communication is possible at daytime. High frequency part of the range is used successfully for space communication. Extremely strongly depends on solar activity.

138

6

Modulation and Signal Types in Modern Radioelectronic Means Table 6.1 (continued)

Range

Name

Propagation peculiarities

30–300 MHz 10–1 m

Very high frequency (VHF, meter waves).

Communication, as a rule, in the limits of straight visibility (quasi-optical communication). However, intra-continent communication is possible. Under particularly favorable conditions, global communication is possible. Successfully used for space communication. At the low frequency part of the range, sometimes evidently depends on solar activity.

300– 3,000 MHz 1-0,1 m

Ultra high frequency (UHF, decimeter waves).

In general, the communication is quasi-optical or optical. At the low frequency part of the range, the communication is intra-continent. Under particularly favorable conditions, global communication is possible. Space communication is used for very large distance measurement.

3–30 GHz 10–1 cm

Super high frequency (SHF, centimeter waves)

In general, the communication is optical. At the low frequency part of the range, the communication distance evidently increases. Under particularly favorable conditions, global communication is possible. Space communication is used for very large distance measurement.

30–300 GHz 10–1 mm

Extremely high frequency (EHF, millimeter waves)

In general, the communication is optical, however, due to the great absorption in the atmosphere, the communication distance is rather small. Until now, this range has not been used for distant communication. Communication can occur between space objects (the atmosphere influence is absent); at that, communication over a large distance is possible.

Modulation in Communication and Broadcast Systems General Information As a rule, the data signal spectrum, which is transferred via the communication channel, in particular speech, concentrates in the restricted low frequency band. On the other hand, radio communication at the given distance is provided in the high enough frequency ranges. Modulation usage allows transmission of the low frequency data via the high frequency communication channel [2, 3]. Modulation is the process whereby the physical carrier parameter changes in accordance with

Modulation in Communication and Broadcast Systems

139

some law. High frequency oscillation (modulated) is used in radio communication as a physical carrier, and, usually, it is sinusoidal. The law of varying is determined by the transmitted message, called the modulating signal. The resulting oscillation, with the time-varying parameters, is referred to as a modulated signal. To extract the modulating signal from the modulated signal at the receiver, the reverse process is performed. This process is called demodulation. The modulating signal, in the case of data transmission from several signal sources via the single radio communication channel, can be a complicated one. It may be multi-channeled at the frequency channel combining, digital aggregated at the time channel combining, as well as with the initial signals on sub-carrier frequency, etc. The analog signal with three varying parameters (the amplitude, the frequency, and the phase) can be used as a modulating signal, as well as any periodic sequence of rectangular pulses with three parameters also (the pulse amplitude, the pulse duration, and the pulse repetition rate). Depending on the type of modulated and modulating signals, and the modulated parameters, one can offer the following classification of modulation types (see Table 6.2). Table 6.2 Modulation type classification

Modulating signal Carrier Modulation type

Analog

Pulse

Digital

Analog Analog AM (amplitude modulation)

Analog or discrete Discrete APM (amplitude-pulse modulation) PWM (pulse-width modulation) PPM (pulse-phase modulation)

Discrete Analog ASK (amplitude-shift keyed carrier) FSK (frequency-shift keyed carrier) PSK (phase-shift keyed carrier) Combined

FM (frequency modulation) PM (phase modulation)

At discrete signal transmission, instead of the term “modulation”, the terms “keying” or “shift keying” are often used. Combined shift keying assumes the simultaneous variation of several carrier parameters (most frequently, the amplitude and the phase). The radio communication transmission channel structure can be presented in simplified form (Fig. 6.1). The initial signal (for example, the speech), acting at the system input in analog form, is analog-digital converted in the source encoder, and then to the coding, which eliminates the initial data redundancy. In analog transmission lines, the frequency filtering is the equivalent of this coding, which extracts the data spectrum section only, in particular, for speech messages, together with the dynamic range compression.

140

6

Message source

Modulation and Signal Types in Modern Radioelectronic Means

Coder source

Coder channel

Modulator

Channel propagation path signal

Demodulator

Decoder channel

Decoder source

Interference

End user

Fig. 6.1 Typical structure of radio data transmission line

In the channel encoder, anti-noise coding is performed. Additional symbols are introduced into the transmitted message at the receiver’s end, which allow for the correction of any error that may have occurred on the physical transmission channel, due to interference. There are several coding schemes, and the application of each one depends on the tasks entrusted to the communication system and on the channel characteristics. In analog systems, the pre-distortions inserted into the modulating signal can be considered as an equivalent of channel coding. Modulation is the next step, on the transmitting side. The problem of having good noise immunity, the power and spectral efficiency of the communication line, depends on the selection of the modulation type and parameters. As a rule, in modern systems, the modulation procedure is combined with the procedure of channel coding, which additionally increases the radio line quality. The radio signal propagation channel, as was mentioned earlier, is characterized by the presence of natural interference, depending on the frequency range, the nautical day and time, the season, the geographical location and the electromagnetic environment. One can distinguish between additive and multiplicative interference. In the first case, i.e., the interference being the noise, the narrow-band or pulse emission is added to the signal. In the second case, the signal and interference multiplication occur as, for example, during the fading in the SW frequency range. The processing of received signals is performed in reverse order. The signal is demodulated, decoded (both operations are often organized as one inseparable process) and recovered, with maximal possible approximation, into the initial message. This recovered message can be passed to the end user.

Types of Analog Modulation Amplitude Modulation Amplitude modulation (AM) can be defined as the variation of carrier amplitude proportionally to the level of the modulating signal (frequency and phase modulation in true AM devices are considered as spurious ones). In the case of the sinusoidal modulating signal, the analytic representation for an AM signal U(t) is

Modulation in Communication and Broadcast Systems

141

U(t) = A0 [1 + k cos (t + ϕ)] cos (ω0 t + ϕ0 )

(6.1)

where A0 , ω0 = 2πf0 , and ϕ0 are the amplitude, the angular frequency and the phase, respectively; k = Am /A0 is the proportionality factor between the modulating signal and the variations of the AM oscillation amplitude or the modulation factor; Am , = 2π F, and φ are the amplitude, the angular frequency and the phase of the modulating oscillation; t is the time. Figure 6.2 shows the AM oscillation versus time. In this figure, we can see that the envelope has the form of the sinusoidal modulating signal. Fig. 6.2 AM oscillation graph

UM(t) t U0(t) t U(t) t

Equation (6.1) can be transformed to the form (for simplicity, the initial phases are omitted): k k U(t) = A0 cos ω0 t + cos (ω0 + )t + cos (ω0 − )t · 2 2

(6.2)

This form shows that, in the spectrum of the modulated oscillation, in addition to the carrier, there are two side components with amplitudes which are proportional to the modulation factor at frequencies that are higher and lower related to the carrier by the modulation frequency = 2πF (Fig. 6.3). A

Fig. 6.3 Spectrum of AM oscillation

f0 − F

f0

f0 + F

f

The spectrum width of such an AM signal is f = 2F ·

(6.3)

142

6

Modulation and Signal Types in Modern Radioelectronic Means

If the low frequency modulating oscillation is the complex one, the modulated oscillation spectrum will, in addition to the carrier, consist of two side bands: upper and lower. They represent the modulating signal transferred to the area of carrier frequency, without any variations and with inversion, respectively. In this case, to determine the full spectrum width of an AM signal, we can substitute the maximal frequency of the modulated oscillation spectrum to (6.3). The vector diagram of the modulated signal is very representative (Fig. 6.4). The sinusoidal carrier oscillation is represented by the vector A0 e jω0 t rotated counter-clockwise with the constant rate of ω0 radians per second. In turn, the side components are represented by the vectors A0 ke jt /2 and A0 ke−jt /2, which are symmetrical relative to the first vector, and are fixed at its end. They rotate counterclockwise and clockwise with the angular modulation rate , moving together with the carrier vector. The resulting vector of the modulated signal changes its length depending on two symmetrical vector positions, and its rotation frequency remains constant. Fig. 6.4 Vector representation of AM oscillation

(ω 0 + Ω)t Ωt −Ω t

Ω

(ω 0 − Ω)t Ω

ω0 ω 0t

AM oscillation power depends on the modulation depth. The carrier frequency power is constant and proportional to A20 /2 The power of each side component is proportional to its amplitude square, i.e., to the value of A20 k2 /8 At the deepest modulation (k = 1), the AM signal power (equal to the sum of all three components) exceeds the non-modulated signal power by 1.5 times only. In practice, the average value of the AM factor does not exceed 0.5, to decrease the over-modulation probability at peaks of the modulation function. In order to increase the transmitter usage effectiveness and to use sparingly the frequency band occupied by the modulated signal, we can transmit not the full spectrum, but one side band of the AM signal only. At this, the carrier and the other side band are suppressed. This modulation type is referred to as AM with the single side band (SSB). It should be noted that, strictly speaking, this oscillation is now the signal with the complicated amplitude-phase modulation. We can distinguish the following variety of amplitude modulation:

Modulation in Communication and Broadcast Systems

143

• • • •

Double-sideband AM (DSB) Double-sideband AM with suppressed carrier (DSBSC) Single sideband (SSB) Single sideband with suppressed carrier (SSBSC), in the variants of lower and upper sideband (LSB and USB) • AM with partially-suppressed single sideband (vestigial sideband – VSB) • AM with double independent sidebands (Independent Single Sideband – ISSB). Another way to increase the effectiveness of AM is by the application of so-called dynamic AM (DAM), where the carrier power is regulated in accordance with the modulating signal amplitude. AM and its varieties have found application, in general, in radio and TV broadcasting. In LW and MW ranges, the double sideband AM is used, but, in SW (HF) and UHF ranges, the single sideband AM is used. In the UHF range in TV systems, for transmission of video signals (luminance component), AM with partiallysuppressed single sideband is used; but, for color-difference signals in PAL and NTSC standards, a specific type of balanced modulation – the so-called quadrature AM – is used. The SSB AM principle is used for formation of channel groups in multi-channel communication systems with frequency multiplexing. Furthermore, this type of modulation is used in mobile communication systems and for communication with airplanes (118–136 MHz). Frequency Modulation Frequency modulation (FM) is the specific case of angular modulation. At FM, the carrier frequency is the varying parameter, i.e., frequency deviation from the nominal meaning is proportional to the modulating signal level in every moment. In the case of the sinusoidal modulating oscillation, the instantaneous frequency is ω(t) = ω0 + ω cos (t + φ)

(6.4)

where ω = 2π f is the carrier frequency deviation amplitude or the frequency deviation. The total instantaneous phase is related to the instantaneous frequency, through the following integral (t) =

ω(t)dt = ω0 t +

ω sin (t + φ) ·

(6.5)

The variable m = ω/

(6.6)

is called the frequency modulation index. For the complex modulating signal, we should substitute the maximum spectrum frequency in (6.6).

144

6

Modulation and Signal Types in Modern Radioelectronic Means

Analytical expression for the FM signal U(t) can be written in the following form: U(t) = A0 cos{[ω0 + m sin (t + φ)]t + ϕ0 } ·

(6.7)

The time graph of the FM signal is shown in Fig. 6.5. Fig. 6.5 FM signal graph

UM(t) t U0(t) t U(t) t

The spectrum of the FM signal at single-tone modulation can be obtained by representation of Equation (6.7) in the form of an infinite trigonometric series: U(t) = J0 (m) cos ω0 t +

∞

Jn (m) cos (ω0 + n)t+

n=1

+

∞

( − 1)n Jn (m) cos (ω0 − n)t

n=1

where Jn (x) is a special Bessel function of n order of the argument x . For the fixed-argument value, Bessel function values decrease by module, and, at m > n, have the small value. Therefore, in practice, one can be limited by consideration of the finite number of the spectrum components. FM signal spectrum view at modulation of the sinusoidal oscillation is shown in Fig. 6.6. We can distinguish the wide-band m >> 1 (i.e., ω >> ) and the narrowband m ≤ 1 (i.e., ω ≤ ) frequency modulation. In the first case, as a rule, one considers the components with n ≤ m + 1 numbers. This corresponds to the FM signal spectrum width at sinusoidal modulation within 99% of signal energy concentrates f = 2(f + F) ≈ 2f ·

(6.9)

At small FM index values (from 1 to 2.5), we shall use the formula f = 2F(1 + m +

√ m) ·

(6.10)

Modulation in Communication and Broadcast Systems

145 Δf

A F

f0

m=5

f

Δf

Fig. 6.6 FM signal spectrum

Beyond this band, the component amplitude is 100 times less that the nonmodulated carrier amplitude. At m << 1, the FM signal (6.7) can be approximated in the form U(t) ≈ A0 {cos ω0 t +

m m cos (ω0 + )t − cos (ω0 − )t}, 2 2

(6.11)

i.e., we can consider that, in this FM signal spectrum, only the carrier and two side components deviated on modulation frequency are present. However, in contrast to amplitude modulation, the second side band component has the phase shift of π radian. The vector diagram in this case has the view presented in Fig. 6.7. In contrast to the AM signal, the sum of the side band oscillations is perpendicular to carrier oscillation vector, which leads to rotation acceleration or deceleration of the resulting vector. The length of this vector representing the modulated oscillation amplitude varies slightly, related to the assumed approximations. In the general case, a greater number of vectors will be totaled and the resulting vector end, during its swinging, will move along the circular arc, i.e., the length of resulting vector does not change.

Ωt Ω

Ω

α Fig. 6.7 Vector diagram of FM oscillation

ω 0t

−Ωt

146

6

Modulation and Signal Types in Modern Radioelectronic Means

Since the FM signal spectrum is wider than that of the AM, the noise immunity of FM modulation is higher. Frequency modulation is used due to its wide-band character, and basically in meter and shorter waves. The narrow-band FM (NFM – Narrow Frequency Modulation) is used in mobile communication systems, the wide-band FM (WFM – Wide Frequency Modulation) in radio and TV broadcasting. In stereo broadcasting, the sub-carrier with polar or frequency modulation is present in the modulating signal, depending on the stereo broadcasting system. Furthermore, FM with m ≈ 2 was widely used in microwave relay and satellite communication, and carrier modulation was performed by wide-band aggregated signals; but, at present, these signals are nearly forced out by digital signals. In radar technology, frequency modulation is used as the intra-pulse one, in the variants of linear FM (chirp-signal), symmetrical FM, zigzag FM, etc. Phase Modulation Phase modulation (PM) is also the specific case of angular modulation. The FM oscillation examined above is, at the same time, the PM oscillation. However, at phase modulation, the phase variation (not a frequency variation) should follow the law of the modulating signal variation. In the case of sinusoidal modulating oscillation, the analytic presentation of the PM signal will take the form: U(t) = A0 cos [ω0 t + ϕ sin (t + φ) + ϕ0 ]

(6.12)

where ϕ is the phase deflection amplitude (or phase deviation). When angular modulation is performed by the sinusoidal signal, one can distinguish frequency modulation from phase modulation by comparing the instantaneous phase variation of the modulated oscillation with the variation law of the modulating voltage only. A comparison of Equations (6.7) and (6.12) indicates that the frequency modulation index is equal to the phase deflection amplitude, measured in radians. However, at frequency modulation, the modulation index is in inverse proportion to the modulating frequency; but, at phase modulation, the phase deviation is fixed and does not depend on the modulating frequency. The spectrum of the phase-modulated signal by sinusoidal oscillation will be the same as for the frequency-modulated signal, if their modulation indexes are equal. At m << 1, the PM signal spectrum will contain the carrier and two side band components deflected from the carrier on the modulation frequency. The difference from the AM signal spectrum consists only in the fact that the side band components ◦ are shifted on 90 in phase. At large modulation index, to calculate PM signal spectrum width, it is necessary to use the formulas for FM signals. The spectrum width in both cases is determined by the frequency deviation. But, it is necessary to note that, at modulation frequency growth, the FM signal spectrum width will remain the previous, at a smaller number of spectral components, while the PM signal spectrum width will increase, at the constant number of spectral components.

Modulation in Communication and Broadcast Systems

147

The vector diagram of the PM signal does not differ from the vector diagram of the FM signal. It is necessary to take into account that FM is defined by the resulting vector angular deviation from the position of carrier frequency vector while FM is defined by the speed of this deviation, i.e., by the phase time derivative. Phase modulation is used in radio navigation systems.

Types of Discrete (Digital) Modulation Amplitude-Shift Keyed Signal In the case of amplitude-shift keyed (ASK) signals described by the switching on and off of the carrier (OOK – On/Off Keying), the output oscillation can be presented by the radio pulse sequence. The rectangular envelope can take two values, repeating the binary modulating signal (Fig. 6.8). Fig. 6.8 Graph of ASK signal

UM(t)

0

1

1

1

0

U0(t)

1

0 1

0 1

0

t t

U(t) t

If the amplitude of the ASK signal can accept values different from zero, we use the term ASK (Amplitude Shift Keying), for double-level keying, and MASK (Multiple ASK), for multi-level keying. The amplitude shift keying signal for elementary transmission duration T can be described in the general case by the ensemble si (t) = Ai cos (ω0 t + ϕ0 ), 0 < t < T

(6.13)

where Ai amplitude can accept M discrete values, i.e., i = 1,2,...M. ASK signal spectrum is the spectrum of binary signal sequence transferred to carrier frequency (Fig. 6.9). Its width depends on the keying speed. For digital modulation types, it is convenient to examine the vector scheme by way of the data signal aggregate, without the account of non-modulated carrier vector rotation (or by constellation diagram). This scheme of the ASK signal for different values of M is shown in Fig. 6.10. As a rule, the adjacent amplitude values correspond to the data modulating combinations, differing in one bit. This representation is referred to as Grey code. At demodulation, when errors with deflections by one gradation – with respect to true

148

6

Modulation and Signal Types in Modern Radioelectronic Means

Fig. 6.9 Spectrum envelope of ASK signal

1 T

A

1

τ

τ= T

2

f0

f

Fig. 6.10 Vector representation of ASK signal 0

1

A1

A2

00 A1

01 A2

M=2

11

10

A3 M=4

A4

amplitude value – are the most probable, the error can occur in the one bit only, in received data combination. Frequency-Shift Keyed Signal In the simplest case, the frequency shift keyed (FSK) signal is described by two frequency presences (Mark and Space), corresponding to the levels of the initial binary modulating signals. At an increase in the number of used frequencies, there is a transfer to multiple FSK (MFSK), the application of which allows the growth of the symbol transmission rate (at the expense of the bit number increasing, which is fitted to one tone message). The general analytic expression for the FSK signal has the following form: si (t) = A0 cos (ωi t + ϕi0 ), 0 < t < T

(6.14)

where ϕi0 is the initial phase of the i − th tone; i = 1,2...M is the frequency number of the FSK signal, which is emitted during the elementary package; and A0 is the signal amplitude. In practice, M is usually a non-zero power of two (2, 4, 8, 16, ...). The time diagram of the FSK signal is shown in Fig. 6.11. In order to decrease the spectrum width of the FSK signal (Fig. 6.12), the initial data signal can be frequency filtered, which leads to smoothing of the input modulating pulses. Using the filter with Gaussian amplitude-frequency response, we can form a so-called Gaussian FSK (GFSK) signal. It is used in the DECT wireless phone standard.

Modulation in Communication and Broadcast Systems UM(t)

Fig. 6.11 Time diagram of FSK signal

0

149

1

1

1

0

1

0 1

0 1

0

t

U0(t)

t U(t) t

Δf

A

F

m=5

τ= 1 2F

fH

f0

fB

f

Fig. 6.12 Spectrum view of FSK signal

We can distinguish the orthogonal and non-orthogonal FSK schemes. In order to be orthogonal, the signal ensemble components should be non-correlated during the symbol transmission time T. The minimum frequency deviation at which this condition can be hold is equal to fi − fi+1 = 1/T

(6.15)

for non-coherent FSK signal detection, when the initial phase of elementary packages can be arbitrary. In a coherent system, when the initial phases of all elementary packages are known, it is equal to fi − fi+1 = 1/(2T) ·

(6.16)

The width of the occupied frequency band of the FSK signal, in the case of keying by symmetrical rectangular pulses, can be estimated by formula [4] f =

2.6f + 1.4B with accuracy up to 2% for 2 ≤ m ≤ 8; 2.2f + 3.1B with accuracy up to 2% for 8 ≤ m ≤ 20

(6.17)

150

6

Modulation and Signal Types in Modern Radioelectronic Means

where f is the frequency deviation (a half of frequency keying range), B = 2F = 1/T is the keying rate (measured in baud); F = 1/(2T) is the keying frequency; and m = f /F is the modulation index. For the orthogonal non-coherent MFSK scheme, the spectrum width, taking into account the additional filtering of the modulating signal, can be considered as f = M/T

(6.18)

f = M/(2T) ·

(6.19)

while for the coherent scheme

The vector diagram of the data signals for FSK oscillation, in the case of an orthogonal scheme, can be represented, in the general case, by M orthogonal vectors in the M−dimension Cartesian co-ordinate system. Figure 6.13 shows the diagram for M = 3. Fig. 6.13 Vector diagram for FSK signal

01 f2

f1

00

f3 M=3 10

The signals without phase jump at the package boundary (Continuous Phase FSK – CPFSK) are the specific case of FSK signals. These are used at high transmission rate, due to the low width of the occupied frequencies, the high decrease of the out-of-band emission level, and the envelope level constancy. So, for example, in the NMT-450 cellular communication standard, fast FSK (FFSK) is used at the sub-carrier with 1,200 baud keying rate. The keying subcarrier is the modulating signal for the usual frequency modulation of a highfrequency carrier. During the elementary data pulse with 0.833 μs duration, either the one oscillation period with frequency 1,200 Hz is transmitted, which corresponds to the unit level of the modulating sequence (Mark), or 1.5 periods of oscillation with the frequency 1,800 Hz, corresponding to zero (Space) (Fig. 6.14). At the same time, during the package transmission in the first case, the oscillation phase varies by 2π radians, while for the second case – by 3π/2. Therefore, to satisfy the phase continuity requirement, the package following the unitary should have the same initial phase as the previous unitary one, while the package following after zero should differ from the previous zero package by π radians, in its initial phase.

Modulation in Communication and Broadcast Systems Fig. 6.14 View of the tone package sequence in FFSK standard

151

UM(t)

1 Un(t)

0

1

1

0

1200 Hz 1800 Hz 1200 Hz 1200 Hz 1800 Hz

1 1200 Hz

t

t

In radio modems, the other parameters of fast FSK on sub-carrier are used. For example, the frequencies of Mark and Space of the fast FSK may be 1,200 and 2,400 Hz at transmission rate of 2,400 baud; or 2,400 and 4,800 Hz at the rate of 4,800 baud, respectively. But at the same time, since the oscillation phase varies by π radians at logical unit transmission, and by 2π radians at logical zero transmission, the dependence of the package initial phase alternation on the transmitted data sequences will be reversed. Since the initial oscillation phase of the specific package depends on the initial oscillation phase of the previous package, this modulation is referred to as modulation with memory. In communication systems, another signal of CPFSK class is widely used, in which the carrier shift keying occurs with minimal frequency shift (Minimum Shift Keying – MSK signal). Analytically, this signal can be presented in the following form: sk (t) = cos [2π (f0 +

dk )t + ϕk ], kT < t < (k + 1)T 4T

(6.20)

where f0 is the carrier frequency, dk = ±1 represents the binary sequence with package duration T, and ϕk is the phase constant for the k − th binary package equaled to 0 or π radians. At dk = 1, the frequency f0 + 1/(4T) is transmitted; at dk = −1 the frequency f0 − 1/(4T) is transmitted, and the frequency deflection 2f = 1/(2T), as at the orthogonal coherent FSK scheme, i.e., the deflection is minimally possible. The frequency modulation index for this situation is equal to m = f /F = (1/4T)/(1/2T) = 0.5. Since the phase incursion difference during the package time for the upper and lower frequency is π radians, in order to fulfill the phase continuity condition at package boundaries, the phase constant should satisfy the following equation ϕk = [ϕk−1 +

πk (dk−1 − dk )] mod 2π · 2

Power spectral density at MSK (Fig. 6.15) is described by the equation

(6.21)

152

6

Modulation and Signal Types in Modern Radioelectronic Means

10 logGN(f) 0

–30

–60 –2

–1

0

1

2

( f − f0) TM

Fig. 6.15 Spectrum of MSK signal

G( f ) =

16PT π2

cos 2πf T 1 − 16f 2 T 2

2

where P is the average power of the modulated signal. As in the case of usual FSK, the modulating binary sequence can be passed through the narrow-band Gaussian filter that additionally decreases the spectrum width of the modulated signal. In this case, the modulation is referred to as Gaussian shift keying with minimal shift (Gaussian MSK – GMSK), and it is used in GSM standard, one of the most popular standards of cellular communication. Phase Shift-Keying Signal Phase shift-keying (PSK) signal is the main type of shift keying providing the high rate of symbol transmission. The modulated signal represents the sections of the elementary signals differing in phase ϕi value only: si (t) = A0 cos (ω0 t + ϕi ), 0 < t < T

(6.23)

where i = 1,2,...M; M is the phase gradation number, and A0 is the signal amplitude. The most noise immunity is provided at the uniform distribution of phase gradations ϕi = 2πi/M. The simplest case is the double-phase PSK (BPSK – Binary PSK). So-called antipodal signals are compared to the data symbols 1 and 0. The first antipodal signal coincides in phase with the carrier oscillation, while the second is out-of-phase. The time diagram of the BPSK signal is shown in Fig. 6.16. M phase PSK (MPSK – Multi PSK), as well as MASK and MFSK, no longer uses the binary alphabet with the transmission of one data bit during the channel symbol transmission period, but the alphabet consisting of M symbols which allows the transmission k = log2 M bits during each symbolic interval. In this case, the data transmission rate (bits per second) will be k times higher than the shift keying rate (in baud).

Modulation in Communication and Broadcast Systems UM(t)

Fig. 6.16 Time diagram of BPSK signal

0

153

1

1

1

0

1

0

1

0

1

0

t

U0(t)

t U(t) t

At 4-phase PSK (or at double phase telegraphy – DPT), four elementary signals are taking part in the transmission, and each of them is characterized by its phase. These signals can be used to transmit the code of 4 pairs (dibits) of the binary symbols. In this case, the signal element transmission rate (modulation rate) decreases twice compared to the bit movement rate of the initial stream that twice reduces the occupied frequency band. For more PSK multiplicity, more effective spectrum usage is provided. At 4-phase PSK, in practice, two sets of phases found ◦ ◦ ◦ ◦ ◦ ◦ ◦ ◦ the application (variants A and B): A) 0 , 90 , 180 , 270 ; B) 45 , 135 , 225 , 315 (Fig. 6.17). For the data combination, as a rule, Grey code is used. 01

ϕ 2 = 90°

ϕ 2 = 135°

00 ϕ 1 = 0°

11 ϕ 3 = 180°

M=4 10 ϕ 4 = 270°

01

ϕ 1 = 45°

00 M=4

11 ϕ 3 = 225°

10 ϕ 4 = 315°

Fig. 6.17 Vector scheme of PSK signal

To demodulate the PSK signals, it is necessary to have reference oscillation, which is synchronous and in-phase, with respect to carrier. This oscillation is usually recovered from the base of the PSK signal. But, in every form of carrier recovering from the multi-phase modulated signal, resides the ambiguity divisible to the maximal shift. To overcome it, one needs to transmit the special marker packages indicating which phase of the elementary signal we should consider as zero. Otherwise, we can cause the phenomenon called “the inverse operation”. In random time moments, all packages “1” at detector output are transformed to packages “0”, and packages “0” are transformed to packages “1”. At the next random moment, the normal receiving is recovered till the next beginning of “the inverse operation,” and so on. To avoid this situation, differential or relative coding is used. The coding rule is reduced to transformation of the modulating data sequence {ai } into sequence {bi } so that bi = ai + bi−1 , and composition is provided by a module equal to the base of the modulation code. This is equivalent to when the initial data are contained in

154

6

Modulation and Signal Types in Modern Radioelectronic Means

the phase differences of the signal packages (as a rule, adjacent), and shift keying, in this case, is referred to as relative or differential (Differential PSK – DPSK). Naturally, at the beginning of the communication session, it is necessary to transmit one redundant package, which we can use for the first data package phase counting. This type of phase shift-keying can also be related to modulation with memory. At relative 4-phase PSK (double relative phase telegraphy – DRPT), the dibits are associated with the phase differences of two adjacent elements of transmitted signal, and that is why the angles for DRPT in Fig. 6.17b should be interpreted as the phase variations counted from the previous package phase, at the moment of its termination. Since DPSK signals differ by relative coding presence only, they can also be demodulated as PSK signals, by the coherent method; but, additionally, it is necessary to recode the data to recover the initial sequence in accordance with the rule ai = bi − bi−1 (modulo equaled to the code base). Most frequently, the quadrature method is used for PSK signal formation. It is based on the fact that any sinusoidal oscillation with arbitrary phase can be presented by the linear combination of in-phase (I) and quadrature (Q) components, such that Q component is shifted relatively to the I component by π /2 radians. If we select cos ω0 t as the reference (in-phase) component, then, in the case of the 4-phase PSK signal, the elementary package is formed as follows: A0 A0 sk (t) = √ dIk cos (ω0 t + φ) − √ dQk sin (ω0 t + φ), kT < t < (k + 1)T (6.24) 2 2 where, in the in-phase channel, the variable dIk = 1 if the dibit high-order bit of the data sequence is equal to 1, and dIk = −1 if the dibit high-order bit is equal to 0. In a similar manner, in the quadrature channel, the variable dQk , depending on the value of the dibit low-order bit 1 or 0, is equal to 1 and –1, relatively. A0 is the signal amplitude, T is the package duration (i.e., dibit formed from two elements of the initial binary sequence), and φ is the initial phase of the carrier oscillation. For the cases shown in Fig. 6.17, the initial phase of the modulated oscillation should have the values of 3π/4 and π radians. In accordance with this method of PSK4 signal formation, the term “quadrature PSK – QPSK” is used. Being not specifically attached to the method, the designation Q can be also interpreted as Quaternary or Quadriphase, i.e., 4-phase. At simultaneous symbol changing in both channels of the modulator (when dibit 10 changes to 01, 00 to 11, and vice versa), there are phase jumps by π radians in PSK4 signals. When using the transmitter with small dynamic range, for example, in a satellite retransmitter-transponder, the envelope variation related to this phenomenon leads to the spurious side band occurrence. These bands take off part of the power and create interference in the adjacent channels. To avoid deep envelope modulation, the 4-phase PSK signal with the offset is used (OQPSK – Offset QPSK), which is sometimes referred to as PSK4 with staggering (Staggered QPSK – SQPSK). The signal formation in the quadrature scheme is provided on the basis of the similar algorithm as in usual PSK4 signals, except

Modulation in Communication and Broadcast Systems

155

that the shift-keying sequences dIk and dQk are shifted in time by T/2. As a result, the phase jumps in the channels occur in turn (twice as often as in usual 4-phase ◦ ◦ ◦ PSK) and, therefore, only the phase change to 0 , +90 , –90 remains (Fig. 6.18). ϕ 2 = 90°

Fig. 6.18 Possible phase jumps at OQPSK

ϕ 1 = 0°

ϕ 3 = 180°

ϕ 4 = 270°

The modulated signal spectrum does not widen. This is related to the fact that the spectrum width at OQPSK is defined by the spectrum width of the in-phase and quadrature components, which are the sequences of the independent signals of T duration, as for PSK4. Power spectral density of QPSK and OQPSK signals has the following form (Fig. 6.19): G(f ) = 2PT

sin 2πfT 2πfT

2 (6.25)

where P is the average power of the modulated signal. Fig. 6.19 Spectrum of QPSK signal

10 logGN(f) 0

–30

–60

–2

–1

0

1

2

( f − f0)TM

The signal π/4 DQPSK is widely used in practice, in particular, in cellular communication standards D-AMPS, in trunking communications TETRA and APCO 25. This is the relative quadrature PSK with the additional offset by π/4 radians, which is sometimes referred to as symmetric. The value π/4 is extra, added to the signal phase at the transmission of each, neat, dibit, therefore, the phase jumps occur by the angles ±π /4 and ±3π/4, which decreases the envelope fluctuations. In contrast to the DQPSK signal, we use the combination of two variants A and B – eight phase states at the given modulation, but the specific transition from the given state is possible for four of them. In other words, if the current phase condition belongs

156

6

Modulation and Signal Types in Modern Radioelectronic Means

to the variant A ensemble, the next phase state will be from the variant B ensemble only, and vice versa (Fig. 6.20).

ϕA2

Fig. 6.20 Phase transition diagram at shift keying π/4 DQPSK

ϕB1

ϕB2

ϕA1

ϕA3 ϕB4

ϕB3 ϕA4

Amplitude-Phase Shift Keying Amplitude-phase shift-keying (APSK) is a combination of amplitude and phase methods. During the duration of the elementary package, it can be described as si (t) = Aj cos (ω0 t + ϕz ), 0 < t < T

(6.26)

where j = 1,2,...N, N is the amplitude gradation number, z = 1,2,...L, and L is the phase gradation number. The total position number of the data signal vector is equal to M = NL, but, in each specific scheme, not all of them can be used (Fig. 6.21); asymmetrical variants are also possible. The number of used positions is also defined by the transmitter message alphabet. So, for case “a” in Fig. 6.21, M = 16, and the quadribits (i.e., the four bits at symbol interval) can be transmitted. In case “b”, M = 8, and each package will contain three data bits. The time diagram for APSK signal is shown in Fig. 6.22. 011 90°

0111 90° 0011

0101 0110

0010 010

0100 1100

0000

0001

001 000

110

0°

0°

1110 1111

1000 1010

M = 16

111

100

1001

1011

Fig. 6.21 Vector diagrams of APSK signal (N = 2, L = 8)

101

M=8

Modulation in Communication and Broadcast Systems Fig. 6.22 Time diagram of APSK signal (N = 2, L = 8, M = 8)

157

U(t)

t

001

011 100

010

101

000

111

110

It is possible to form APSK signals, with the help of quadrature methods. Since, then, the usual amplitude shift-keying is performed in each channel, this type of modulation is referred to as quadrature amplitude shift-keying (QASK) or quadrature amplitude modulation (QAM). The ensemble number from M points on the plane is infinite, for the specific M value. At M > 4, we can consider as optimal (i.e., having the minimal average power at given error probability) the ensembles from the signals of different power distributed uniformly within the circle, the radius of which is defined by maximal permissible signal energy. The symmetrical configuration, with regular location of the signal points in the knots of the square grid, has found the most applications in practice (Fig. 6.23). The number of signal points indicating the vector ends on the spatial diagram is used in the designation of specific scheme QASK: 16-QAM, 64-QAM, etc. Fig. 6.23 Vector scheme of QASK signal

1000 1100

90°

1001 1101

0100 0000 0101

180° 1011 1111

0001 0°

0111 0011 M = 16 0110 0010

1010 1110 270°

QASK signal spectrum is similar to PSK signal spectrum, at equal number of used positions of the data vector. However, QASK systems have the better error characteristics, especially with the large position number, because at the same maximal power the distance between points is higher for QASK signals. The combined shift-keying method is not limited by the amplitude-phase only. We know the signal ensembles with good characteristics at simultaneous frequency and phase shift-keying (FSK-N/PSK-M), as well as at the use of all schemes FSKN/APSK-M.

158

6

Modulation and Signal Types in Modern Radioelectronic Means

Signals of Modern Radio Electronic Means SW Range Signals (Less Than 30 MHz) Peculiarities of radio waves propagation, related to ionosphere reflections, allow data transmission to large distances without the arranging of additional infrastructure, at relatively small transmitter power and with relatively simple antennas. The instability of the communication channel and its channel capacity restrictions are the shortcomings of SW range systems. Both international and intra-state broadcasting are organized at decameter waves, as well as the communication of diplomatic, military and police departments of many countries. Communication means of international organizations such as the U.N., Red Cross, etc. also operate in this range; the data of news agencies and meteo data are transmitted in the SW range, as well. Amateur radio-operators work intensively in this exact frequency range. SW range radio communication systems remain in operation and are used as the reserve system for communication with airplanes and navy, presenting an alternative to satellite systems. There is also a SW range allocated for personal communication: the Citizen’s Band (CB).

Radio Broadcasting At present, the majority of widely-broadcast radio stations operating in SW range are using analog amplitude modulation. At that, the lower modulating frequency is 150 Hz (for frequencies less than 150 Hz, damping is introduced with 6 dB per octave), and the upper modulating frequency should not be higher than 4.5 kHz. The broadcasting is organized in range bands allocated in accordance with international agreements (Table 6.3). In LW and MW ranges, the unified grid is approved, with a 9 kHz carrier frequency-diversity (in the U.S., this diversity in MW is equal to 10 kHz). In long waves range there are 15 channels, the first channel frequency is 155 kHz. In MW range there are 120 such channels, the first channel frequency is equal to 531 kHz. The carrier frequency-diversity in the HF (SW) range is approved at 10 kHz, within the limits of one geographic zone; but, if transmitters operate in different zones, usage of a 5 kHz frequency-diversity is allowed. There are three geographic zones only related to radio frequency distribution, in accordance with intergovernmental agreements. Zone 1 includes all territory of Europe and the Asian part of Russia, Asian countries of the former Soviet Union (SU), Mongolia and the territory of Africa. Zone 2 covers the territories of Northern and Southern America and Greenland. Zone 3 includes the territories of Asia (except the former SU and Mongolia) and Australia. In LW range, the main emission energy is transferred by surface (earth) wave, which is not influenced by the condition of the ionosphere, but is strongly absorbed by the soil. The latter factor forces the application of transmitters with 500– 1,000 kW power to support the large territories. The receiving conditions in this range are stable, while the propagation distance, due to the spatial (ionospheric)

Signals of Modern Radio Electronic Means

159

Table 6.3 Digital communication types in SW range

Frequency spacing, Hz

Name

Modulation

Baudot (RTTY)

FSK

ARQ-E

FSK

85; 125; 170; 325; 400; 425; 450; 800; 850 400

FEC-A

FSK

400

PSK31 SITOR-A

BPSK (QPSK) FSK

PACTOR

FSK

PACTOR-II

DPSK

MFSK-16

MFSK

MIL STD 188–141 A (ALE) CLOVER-II

Mode and method of transmission

Alphabet

Transmission rate, Baud

Asynchron., data stream

ITA2

45; 45; 50; 75; 100; 150; 200

Synchronous, data stream

ITA2-P

Synchronous, data stream – Synchronous, data stream 170; 300; 400; Synchronous, 850 data packages 200 Synchronous, data packages 200 (2 tones) Synchronous, data packages

ITA2-P ASCII

48; 64; 72; 86; 96; 144; 192; 288 96; 144; 192; 288; 384 31; 25

CCIR476

100

ITA-5

100

ASCII

200 (DBPSK); 400 (DQPSK); 600 (D8-PSK); 800 (D16-PSK) ASCII ITA-2 15.625

Asynchron., data stream

MFSK

15.625 (16 tones) 250 (8 tones)

PSK, APSK

125 (4 tones)

Data packages ASCII

CLOVER2000

PSK, APSK

250 (8 tones)

Data packages ASCII

FEC-101

FSK

80; 170 (3 channels with diversity 650 or 680 Hz)

Synchronous, data stream

ASCII

ITA2-P

375

125; 250; 375; 500; 750 (31,25 symbol/s) 500; 1,000; 1,500; 2,000; 3,000 (62,5 symbol/s) 96; 144; 192

160

6

Modulation and Signal Types in Modern Radioelectronic Means Table 6.3 (continued)

Frequency spacing, Hz

Name

Modulation

MIL 188–110 A

BPSK QPSK 8-PSK

–

MIL 188–110 B (App. B)

DQPSK

MIL 188–110 B (App. S)/ STANAG 4539 HDR

QPSK 8PSK 16-, 32- and 64-QAM

56.25 (OFDM 39 tones: 675– 2,812.5 Hz) + Doppler tone 393.75 Hz –

STANAG 4197

QPSK

4/16/39 OFDM

Mode and method of transmission

Alphabet

Transmission rate, Baud 75; 150; 300; 600; 1,200; 2,400; 4,800 (without coding) 75; 150; 300; 600; 1,200; 2,400

Coded speech

3,200, 4,800, 6,400, 8,000, 9,600, 12,800 bit/s (without coding) 1,200 (16 × 75); 1,733 (39 × 44,44)

wave, is larger at night than in daytime, and in winter is larger than in summer. Atmospheric and industrial interference strongly influences on the receiving in this range. MW range serves mainly the regional radio broadcast. Propagation conditions in the MW range are characterized by increased absorption in the soil, at frequency growth, and by serious influence of the spatial wave at night, when the propagation distance essentially increases. As a result, interference from the distant station occurs – operating at the same and adjacent frequencies – and the consistent receiving zone is reduced. Depending on the purpose, transmitters of 5–1,000 kW power are used in this range. In SW (HF) range, the earth wave receiving distance reduces to several kilometers. Here, ionosphere propagation plays the main role, at which energy absorption is rather small. The majority of broadcasts are broadcasts to foreign countries. The propagation to the large distance happens by the jumps, with the alternate reflections from the ionosphere and the Earth’s surface. The maximal length of one jump is 3,000–4,000 km. In between the jumps, a so-called “dead zone” occurs, from 50 to 180 km in radius, where the receiving is impossible. To increase the receiving reliability, the broadcast may be transmitted simultaneously in different wave

Signals of Modern Radio Electronic Means

161

sub-ranges. In such a way, for example, broadcasting on four frequencies within the limits of a single jump at the protective ratio 27 dB (exceeding over interference at high frequency) and at minimal permissible field intensity 50 dBμV/m provides 95% reliability of high-quality receiving. For two jumps, it reduces to 85%, for three – to 70% [5]. Moreover, operating wave-lengths can be changed round the clock and annually, in accordance with the radio wave propagation for the specific direction. HF range stations have the nominal power of 50, 100, 150, 250, and 500 kW. The main characteristic property of receiving in HF range is the interference of several spatial waves reflected from the different points of the varying ionosphere, which leads to the random frequency-selected (fast and slow) signal fading up to ten times. This fading, in turn, causes, at detection, the simultaneous appearance of the unavoidable randomly-dependent linear and nonlinear distortions. Most radio broadcast quality deterioration is linked with the carrier frequency fading. The distortions essentially reduce at single-side band AM. That is why, in 1987, at the international level, the decision was approved for the gradual transition to the single-side radio broadcasting system. But, at present, several such SW radio stations are operating regularly. There is another alternative: in recent years, the new radio broadcasting system DRM (Digital Radio Mondiale) was introduced. This system of digital radio broadcasting ensures the receiving of stereo and mono programs at quality close to UHF FM radio broadcasting, at signal levels less than for usual AM systems. At that, the transmission of speech signal and the additional text and graphic data to all or selected users is possible [6, 7]. The bandwidth of the DRM signal coincides with the frequency grid of AM ranges, six frequency variants are stipulated for it: 9 and 10 kHz (main regimes) and also the half (4.5 and 5 kHz) and double (18 and 20 kHz) values. If necessary, the DRM system provides the joint transmission of digital radio broadcasting and analog broadcasting signals with amplitude or single side band (there are 12 variants of spectrum combining) in one signal channel. The organization of DRM broadcasting by the transmitter net is possible, such that it operates in synchronous mode at the same frequency. Technical solutions applied in the DRM system ensure the high stability of signal receiving under unfavorable affected factors in transmission channels (interference, fading, Doppler effect, etc.). The DRM signal represents the ensemble of the orthogonal frequencydivided carriers (Fig. 6.24), so-called, Orthogonal Frequency Division Multiplexing (OFDM) signals, and, at that, the parallel transmission of the digital data is executed. In 10-kHz bandwidth, the carrier number can span from 88 to 226 depending on the mode, and the duration of the transmitted signal (data with the protected interval, which reduces the sensitivity to interference caused by multipath radio wave propagation) can vary from 16.6 to 26.6 ms. The method of each carrier modulation 4-QAM, 16-QAM or 64-QAM with redundant coding (from this the additional letter C in designation COFDM) is used,

162

6

Modulation and Signal Types in Modern Radioelectronic Means

1 T A

f Fig. 6.24 Group of the orthogonal frequency-divided carrier

depending on the given noise immunity degree. The process of concurrent modulation of the large carrier number can be executed mathematically with the help of direct and inverse Fourier transformation. The input audio signal is coded under MPEG-4 standard to reduce the redundancy, and the rate of the digital stream acting to channel encoder is equal to 4–24 kbit/sec.

Communication in SW Range The overwhelming majority of analog radio communication stations of SW range operate in single side-band amplitude modulation mode. The lower side band (LSB) or upper (as a rule, more than 10 MHz) side band (USB) is used. The analog narrowband frequency modulation (NFM) and usual AM in the bandwidth 3 or 6 kHz (AM3 or AM6) are used rarely. Single side-band modulation is used by radio amateurs, in the definite ranges of the bands allocated to them, and also by the land mobile and fixed services, and the sea mobile service. FM is used by railway organizations and in civil range at 27 MHz. Simple AM can also be found in the civil range. Image transmission systems (facsimile) have a certain connection to analog SW systems: images of weather maps, hieroglyphic text, etc. The signal from the image-scanning modulates sub-carrier on frequency, and this sub-carrier, in turn, is the modulating oscillation for the amplitude modulation of the main carrier. Single side band is emitted, and the carrier is fully or partially suppressed. The bandwidth is 3 or 2 kHz, in this case. Moreover, there are facsimile systems with direct FM of the main carrier, as a rule, with a 400 Hz deviation (transmission of black and white levels only, by FSK signal, can already be considered as digital). The main direction of SW-communication development is the further transition to the transmission systems for digital data, which are constantly perfected.

Signals of Modern Radio Electronic Means

163

The transmission rate of 9,600 baud in a 3 kHz channel is already not the limit. The automatic adaptation methods to radio wave-propagation conditions and to interference levels in the channel are widely implemented. Data transmission reliability, due to the various coding methods, increases. At present, in the SW range, the large variety of digital signals and the number of digital communication protocols continue to expand [8–11]. Their names partially reflect the used data transmission technologies. Protocol classification is possible by used signal modulation: FSK, PSK, or QAM; by the mode and data transmission method: synchronous or asynchronous, continuous or by packages; by the message alphabet; by channel number; by coding method: with interleaving, by packages, by cascades; by adaptation method: hardware – the channel choice with the better characteristics or in the given order, or structural – rate and time-modulation transmission parameters varying. Some types of digital SW communication systems are described in Table 6.3. Radio teletype (RTTY – Radio Tele Type) is the simplest and the earliest digital protocol, except for the Morse alphabet. In contrast to the non-uniform Morse alphabet, the Baudot alphabet consists of a five-place combination set. The total number of such combinations is 25 = 32, which is enough to transmit Latin letters. To transmit figures and other symbols, an additional digital register is used consecutively (the same 32 combinations but with other semantic meanings), the transition to which is executed with the help of a special symbol. The inverse transition is possible after the transmission of another similar symbol. There are adaptations of the ITA2 (International Teleprinter Alphabet) alphabet, for national alphabets: Arabian, Chinese, and Cyrillic. There are also the variants of the mutual operation with the Latin alphabet, with the switching into the so-called third register for the national alphabet at transmission of the symbol, which is usually not used. Asynchronous mode of RTTY protocol operation is historically related to the fact that, at the start, the data, which had been entered from the keyboard by letters, was transmitted, and the data stream had no long, regular, structure. For correct operation of the mechanical teletype apparatus, one must send the start package before symbol writing. Its duration is equaled to the single data package duration that, in particular, is 22 ms for a 45.45 baud transmission rate (radio amateur standard). After transmission of the symbol data packages, the stopping package follows with polarity inverse to the starting one, and, in accordance with the protocol, its duration can be the same as the data one or exceed it by 1.5 or 2 times. In SW range, the duration 1.5 is used, which provides maximal noise immunity. RTTY protocol uses frequency shift-keying, the various interpretations of which should be considered as the Mark frequency (transmission of “1”) and as the Space frequency (transmission of “0”), and by which level is used for transmission of start and stop components. Therefore, the equipment for RTTY receiving, as a rule, makes provision for several variants and can even have an adjustable mask for inverting the individual bits of signal combination. RTTY protocol does not provide error protection, and thus the transmitted data, as a rule, is repeated more than once. ARQ (Auto Request) technology, which is the

164

6

Modulation and Signal Types in Modern Radioelectronic Means

automatic message over-interrogation at error detection or transmission acknowledgement at true receiving, is one of the methods of error protection. The error is detected, for example, by violation of 3:4 zeros and units’ parity in ITA3 alphabet (Moore code), by check on evenness in the ITA2 alphabet, or with the help of calculation of the data package check sum. At continuous data transmission, ARQ technology is provided due to duplex, at package transmission, simplex or semiduplex are possible. The relative error value when using ARQ is equal 10–5 , while the usual system has this parameter at a level of 10–2 . Another protection method is the application of various redundant coding algorithms with the help of codes correcting the errors, the so-called Forward Error Correction (FEC). It requires a one-way communication line only, since, in this case, the check parity bit serves for the detection, as well as for the error correction. In time, the symbol and bit interleaving improves the error correction process, as it transforms the group errors in the fading interval into the individual uncoordinated ones. One element of the FEC approach is the twice transmission of each symbol with the shift on the definite number of symbol positions. Joint application of FEC and ARQ technologies are also possible. The adaptive selection of the best operating frequency ALE (Automatic Link Establishment) also improves the reliability of the data transmission. In accordance with the MIL-STD-188-141A standard, this technology provides a survey of the programmed-in-advance operating frequencies, at the rate of 2–5 channels per second after call-in, the station (stations) with triple exchange of secondary messages (code words of 49 bits) in the form of tone packages (8 tones of 8 ms duration). After channel quality estimation on the basis of error probability, signal/noise ratio, and multipath effect, the best frequency selection is executed and equipment switching to the high-rate data transmission of voice communication is carried out. The holding of range data regarding the propagation at different frequencies, depending on the nautical day/time, reduces the period of communication establishment. Another frequency adaptation approach is automatic frequency hopping (FH). Frequency jumps can occur at channel quality deterioration in the allocated band or in the band set, in accordance with the preliminary-defined algorithm, for example, to the adjacent channel or band. Such jumps occur slowly enough, several times per second. This approach can be used to increase transmission hiding; in this case, the number of jumps per second will be more, essentially. There are a series of protocols (for instance, the PACTOR and CLOVER families) that provide automatic adaptation to the channel condition, at expense of varying the data transmission rate. This can be done by simply doubling the growth of bit package duration (PACTOR), by modulation type and code varying (PACTOR-II, CLOVER-II, CLOVER-2000), and by sub-carrier number varying from 2 to 18, at a rate changing from 200 to 3,600 bits/s in the 2.4 kHz channel (PACTOR-III). Modern SW modems, with the high transmission rate and using the phase and quadrature amplitude modulation methods of the single carrier and OFDM technology, have the adaptability property as well.

Signals of Modern Radio Electronic Means

165

VHF Range Signals (More Than 30 MHz) In VHF range, the most reliable radio receiving can be provided within the confines of a transmitter and receiver straight visibility. At that point, in the near zone, direct wave interference with waves reflected from the Earth’s surface takes place, while, in the far zone, interference with re-radiation, for instance, from various types of construction, occurs. The field strength minima and maxima, in the open area, in the town, and inside of buildings, are the interference consequence. The refraction (the deflection of radio wave propagation in the atmosphere, with respect to the straight line resulting from the refraction index varying according to height) and the diffraction (rounding the obstacles by radio waves), which is expressed evidently in meter wave range, become apparent at propagation. Due to these factors, the communication distance increases by approximately 15% compared to the horizon line distance, and, under normal conditions, can be estimated by the approximate formula [12]: r = 4.12

h1 +

h2

(6.27)

where r is the communication distance in km; and h1 ,h2 are the heights of the transmitter and the receiver antenna mounting in m. So, for example, if radio communication is being established between two portable radio stations with enough transmitter power and enough receiver sensitivity, when the transmitter and the receiver antennas are mounted at approximately 1.5 m, then the available distance will be about 10 km in the open area. If the antenna of one of the radio stations is mounted on a high building, the stable communication distance may achieve 60–70 km. The field strength of meter and decimeter waves in the town is less due to building shading, by 3–5 times, while, for centimeter waves, the attenuation value is still more. Inside the buildings, the field attenuation can be 2–30 times more, with respect to the level over the roof. More accurate estimates of field strength in urban and suburban regions can be obtained, for instance, with the help of the Hut model [13]. Sometimes, under definite atmospheric conditions, super-refraction is observed, and VHF radio waves can propagate on the large distance (more that 1,000 km), including so-called atmospheric duct (centimeter and decimeter range). Moreover, diffusive dispersion (re-reflection) of VHF range radio waves occurs on the vertical and foliated troposphere heterogeneity (toward the initial wave movement). The troposphere field component strength (the result of interference of the fields re-radiated by the large number of heterogeneity) is small and depends strongly on time. Usually the area of this component is situated at 100–1,000 km distance from the transmitter. Under some conditions, meter radio waves may reflect from the regular F2 and sporadic Ec ionosphere layers, and may disperse on the turbulent ionosphere irregularities (similar to troposphere dispersion). The receiving zone for the ionosphere field component is at 800–3,000 km from the transmitter, and the field strength is not large.

166

6

Modulation and Signal Types in Modern Radioelectronic Means

There are communication systems using reflections of VHF waves (in particular, in the 30–60 MHz range) from short-term ionized meteor traces. At centimeter radio waves propagation, absorption by the water vapor and dispersion by the atmospheric precipitation exist. Radio Broadcasting In meter wave range, high-quality audio mono- and stereo broadcasting is organized, with the help of frequency modulation. At present, two broadcasting standards are used in Russia: OIRT (East-European) and CCIR (West-European). At standard designation, the abbreviations of organizations issuing the recommendations are indicated: International Organization of Radio and TV Broadcasting, included in 1993 into the European Broadcast Union (EBU), and International Consultative Committee on Radio Broadcast, which is Radio Communication Sector (ITU-R), beginning from 1992. The broadcasting frequency bandwidth for the first standard is equal to 65.9– 74 MHz. The nominal frequency deviation is 50 kHz, and the transmitted audio frequency range is 31.5–15,000 Hz. In stereo variant (compatible with the mono mode), polar amplitude modulation of the sub-carrier with 31.25 kHz in frequency occurs (with the help of commutation: its negative and positive half-waves are modulated in turn by the right and left channels of the audio signal and, as a result, modulation by the channel-difference signal occurs). To increase the noise immunity, the predistortions are introduced i.e., the high-frequency spectrum components of the audio signal are increased (by 15 dB at 15 kHz). The sub-carrier frequency is 5 times suppressed (by 14 dB), and signal deviation does not exceed 10 kHz at one sub-carrier presence. The frequency grid is dividable to 30 kHz (65.9 + 0.03n MHz), but the frequency diversity between the stations should exclude the mutual interference. The frequency band at this FM broadcast signal transmission, in accordance with (6.9), in mono mode is 130 kHz, and in stereo mode is not less than 192.5 kHz. In West-European standard, the frequency band of 87.5–108 MHz is used; frequency deviation is 75 kHz. The transmission of the audio frequency signal is arranged at 40–15,000 Hz range. In stereo mode, the difference signal of the left and right channels is transmitted with the help of the 38 kHz, sub-carrier, amplitude modulation. The sub-carrier itself is suppressed almost fully (not less than 40 dB) and, for its recovering during the receiving, a so-called 19±0.002 kHz pilottone is introduced into the signal. The deviation related to the pilot-tone does not exceed 7.5 kHz. The total signal frequency band for CCIR standard in mono mode is 180 kHz, while in stereo mode it is 256 kHz. The frequency grid is dividable to 100 kHz. At present, broadcasting with pilot-tone is used also in the OIRT range, which corresponds to the imported equipment application. Minimal field strength for the high-quality receiving of an FM signal is equal to 48 dBμ V/m for mono broadcast and 54 dBμV/m for stereo mode, beyond the urban limits; inside the town, taking into account the additional interference, these figures are 70 dBμV/m and 74 dBμV/m, relatively.

Signals of Modern Radio Electronic Means

167

In recent years, systems for the additional transmission of data on additional subcarriers have been used. For the lower VHF range, Russian branch standard OST 45.125–99 regulates the application of the broadcasting channel multiplexing system UVK2, with two sub-carriers 46.875 kHz and 78.125 kHz (both as well as any of the frequencies can be used), and also the “Radiotext” system on the sub-carrier 46.875 kHz. In the first case, the frequency modulation of the sub-carriers with 10 kHz deviation is used; the additional data signal bandwidth is 30–10,000 kHz; at data transmission on 78.125 kHz, the transmission rate is equal to 19 kbit/s. “Radiotext” uses the relative binary-phase shift-keying (OBPSK) of a 46.875 kHz sub-carrier with the clock rate 732 Hz. The sub-carrier frequency deviation caused by the total signal can be increased to 60 kHz. For the second standard systems, the following is allowed: the broadcast channel multiplexing UVK2 (it is possible to use one sub-carrier 78.125 kHz only); data transmission system RDS (Radio Data System) on a 57 kHz sub-carrier (third harmonic of the pilot-tone); the road data transmission system ARI (Autofahrer Rundfunk Data), also on a 57 kHz sub-carrier, possible together with RDS; the additional data transmission system SCA (Sub-carrier Communication Allocation) on a 67 kHz sub-carrier (in other countries on 92 kHz). In the SCA system, the frequency modulation with 6 kHz deviation is used, and the additional signal frequency band is 30–6,000 Hz. In the ARI system, the program recognition signals are transmitted with the help of separate tones in the 23–54 Hz range, and they modulate the sub-carrier amplitude with 60% depth. In the RDS system, the amplitude modulation with the suppressed (not less than 50 dB) carrier is used. In this case, the modulating signal is the bipolar pulse signal passed through the filter with cosine amplitude-frequency response. At transmission of unit level, the pulse signal polarity changes from positive to negative; at zero transmission we have the opposite situation. The initial data is relative coded. The data transmission rate in the RDS system is 1,187.5 bit/s, and, at that, the spectrum width on the –60 dB level is equal to 4.8 kHz. The 57 kHz sub-carrier phase should either coincide with the pilot-tone third harmonic phase with 10◦ accuracy, or be in quadrature with it. At joint application of the RDS and ARI systems, both sub-carriers should be in quadrature with the same accuracy [14]. The RDS standard is the multi-functional one, and it allows the transmission of various data types. These include the recognition of radio stations and programs, data on the alternative frequencies, program schedule, road data, radio text, time data, danger alarm signals, various reference data, the data transparent channel for message transmission, and radio paging. In 1997, in addition to RDS, the system of high-rate (16 kbit/s) data transmission DARC (Data Radio Channel) on 76 kHz sub-carrier (fourth pilot-tone harmonic) was approved as the All-European standard ETS 300 751. The protocol of the similarly-named Japanese system is the basis for it. In the DARC system, the FSK with minimal shift and level regulating (LMSK) are used. The sub-carrier level varies linearly between two extreme values, which are defined by the difference stereo signal level. If the radio signal deviation caused by the audio channel

168

6

Modulation and Signal Types in Modern Radioelectronic Means

difference signal is less than 1.875 kHz, the deviation caused by the system subcarrier is equal to 3 kHz. If the radio signal deviation related to the difference signal is equal to 3.75 kHz, the deviation defined by the sub-carrier should be not more than 7.5 kHz. The frequency band occupied by the DARC system is equal to 35 kHz on –20 dB level [15]. The possibilities of the DARC system are broad compared to the RDS system, with regard to possessing services as well as to the creation of new services: onsite position determination on the basis of global positioning system (GPS) with the differential amendments; risk decreasing on the roads, electronic mail, and voice paging (MobiDARC). Digital audio broadcasting is the alternative to usual analog broadcasting in VHF range. The T-DAB (Terrestrial Digital Audio Broadcasting) system is approved as the All-European standard, by the European Telecommunication Union. This format, like the digital SW radio DRM, uses the CODFM method for transmission. At that, the carrier number is defined by the mode concerned with the used broadcasting frequency and the transmitted data stream capacity, and can be varied from 192 to 1.536. To modulate carrier frequencies, the relative π/4 phase shiftkeying (DQPSK) is used, and it is executed by the calculation approach. The signal bandwidth is 1.54 MHz, and the protective frequency spaces between DAB signals are about 200 kHz. The duration of one OFDM symbol (data plus the protective space) can possess the value from 0.156 to 1.25 ms. The frequency ranges of 174– 230 MHz and 1,452–1,492 MHz are recommended for new broadcasting technology. The required minimum field strengths beyond the town are 58 dBμV/m and 66 dBμV/m, respectively, for these ranges [16, 17]. The audio signal quality provided by the DAB system is comparable with the usual FM broadcasting, even for mobile receiving. At that, the concurrent transmission from 4 to 9 stereo programs (at different quality) and the additional data, of various natures, on the single frequency in the unified digital stream with the rate of 1.168 Mbit/s (2.4 Mbit/s taking into account redundant coding) is possible. The initial audio signals are coded under MPEG-1 Audio Layer 2 (MUSICAM) standard, and the rate of the audio data digital stream can be varied from 8 to 384 kbit/s per channel. At present, in Russia, broadcasting in DAB format is restricted to experimental transmissions only, and its wide application is limited by economical reasons only.

TV Broadcasting Ground-based TV broadcasting is provided in meter (48.5–230 MHz) and decimeter (470–790 MHz) wave range with dividing into the sub-ranges (the total frequency grid of TV channels is shown in Table 6.4). The lower boundary of the frequency range is conditioned by the technical effectiveness of video signal transmission and receiving with the maximal spectrum frequency of 6 MHz, and it is necessary that the radio signal carrier some times exceeds this frequency.

Signals of Modern Radio Electronic Means

169

Table 6.4 Frequency grid of TV channels Sub-range

Frequencies, MHz

TV channels

I II III IV V

48.5–66 76–100 174–230 470–582 582–790

1 and 2 3–5 6–12 21–34 35–60

In large towns, there are also the TV program distribution systems MMDS (Multichannel Multipoint Distribution Systems), in which the signals of channel groups (for example, 24) are transmitted in the 2,500–2,700 MHz frequency-range, for further receiving and usage at cable networks. The image carrier is modulated on amplitude by the total brightness signal. The lower side band is partially suppressed and the audio accompaniment carrier is frequency modulated by the audio signal (30–15,000 Hz) with a 50 kHz deviation, and, at that, it is possible to use the sub-carriers. Figure 6.25 shows the radio signal spectrum envelope for image and audio signals. The lowest radio channel frequency bandwidth is 7.625 MHz for the image (attenuation of the 1.25 and 6.375 MHz components is 20 dB, with respect to the carrier), while it is 0.25 MHz for the audio accompaniment signal. In Russia, the D/K standard is used when the audio and image frequency carrier diversity is equal to 6.5 MHz (the image carrier frequency is less than the audio carrier frequency); the nominal frequency bandwidth of a TV broadcasting radio channel is 8 MHz; power ratio of the image and audio carriers is 5:1–10:1. The power of the signal spectrum components decreases quickly with the frequency growth, and the span of the video signal high-frequency components is

6.5

A

fimage Fig. 6.25 TV signal spectrum

8.0

faudio

f, MHz

170

6

Modulation and Signal Types in Modern Radioelectronic Means

usually not great. Thus, the color frequency sub-carriers (frequency-modulated for SECAM 4.250 and 4.40625 MHz systems) are arranged exactly in this video spectrum range. The video signal spectrum is discrete; it contains the harmonics multiple to the line repetition rate. Narrow-enough bands of side band signals group around this harmonics, which are conditioned by the frame deflection and the image detail movement. The line frequency harmonics, with their side bands, form the discrete zones carrying data on the transmitted image. The stereo audio signal application for TV broadcasting in Russia was limited not long ago by the experiments only. Stereo mode was implemented by the method closed to VHF broadcasting. But, this approach makes serious additional demands on the TV transmitter, with regard to the intermodulation. In 1986, engineers of the British TV-Radio Company (BBC) developed the digital version of stereo programs for audio accompaniment NICAM 728 (Near Instantaneous Companded Audio Multiplex). This system has the following features: discretization frequency is 32 kHz, level quantization at transmission is 10 bit, i.e., 1024 samples, the audio frequency band is 30–15,000 Hz. Together with the service signals, the digital stream rate is 728 kbit/s. This stream modulates a 5.85 MHz sub-carrier by the method of relative phase shift-keying DQPSK, and at that the spectrum width is 700 kHz on –30 dB level. Concurrently, for compatibility with mono TV sets, the usual frequency-modulated audio sub-carrier is transmitted. The maximal value power ratio for carriers of image and digital modulated audio signal is 500:1 for D/K standard [18]. On the horizon are digital technologies for changing the analog image transmission, also. The air-telecasting transition to the European standard of digital ground TV DVB-T (Digital Video Broadcasting) is intended to begin in Russia in 2016, in decimeter wave range (21–69th channel). By now, the experimental digital broadcasting is already issued in Moscow and St. Petersburg. The DVB-T standard is based on the mentioned COFDM technique. The orthogonal carriers number may be equal to 1,705 (2 K mode) and 6.817 (8 K mode). The data symbol duration, together with the protective interval, can change, in the first case, from 231 to 28 ms, while, in the second case, from 924 to 1,120 ms. The carriers are modulated mathematically by QPSK, 16-QAM and 64-QAM method, with uniform and non-uniform data vector locations depending on the digital stream rate and on the required redundancy degree. The data transmission rate (without the redundancy) varies from 4.98 to 31.67 Mbit/s, which allows simultaneous transmission of four programs using the video signal coding in accordance with MPEG-2 standards. The bandwidth of the transmitted signal is about 7.61 MHz [2, 19]. Communication in VHF Range At present, VHF range is occupied by various communication systems, with a large number of services and possibilities. In VHF range, communications based on usual radio stations with narrow-band FM (sometimes with scrambling or spectrum inversion) functions, automatic communications networks with analog and digital transmission of speech data acts, and wireless systems for data transmission,

Signals of Modern Radio Electronic Means

171

are deployed. Many various types of signals are used and allocated for the radio amateurs’ zones, namely, narrow-band FM, single band AM, telegraph, and digital types. The first communication type is the simplest one, and can be used for process purposes, over the territory of not-so-lengthy objects (up to 4 km), by a small group of people. As a rule, one-and-the-same frequency is used; the emission exit is executed alternatively (simplex), and the system of individual or group call, tonal, for example, may possibly be present. If necessary, to operate under conditions of large industrial complexes or in a large town (distance up to 30–50 km), an additional dispatching station (so-called, radial communication system) or a repeater (repeater network) with a highly-mounted antenna can be used. Since, at repeating, the receiving of a distant station’s signal is provided, along with its transfer with amplification to the other frequency, it is necessary to use two frequencies spaced to 3–10 MHz (duplex and semi-duplex for the users). Similar systems are located in the following sections of the frequency ranges: 30– 56 MHz, 136–174 MHz and 300–308, 336–344, 400–512 MHz. So, for instance, the radio frequency band from 154.025 to 154.775 MHz is intended for processing the communication of the local railways of large industrial enterprises; from 168.100 to 168.225 MHz is used for radio communication in car transport control systems. The section in the band from 433.075 to 434.775 MHz is approved for civil communication without frequency assignment, but the transmitter power in this band is limited by 10 mW and that restricts the communication distance by 1 km. It is necessary to discuss separately the air range from 117.975 to 137 MHz (grid step is 19.025 kHz), where radio-phone communication with AM is used for “air to ground” communication. Trunking (hitcher) systems are one version of the second type of systems. They appeared as an answer to the need for more effective usage of the limited frequency resource and united the users in one group, providing priority or mutually equal, automatic, access to the system (the trunk) of radio communication channels, on the basis of the usual telephone network. Trunking systems can be used for professional mobile communication since they provide the service, mainly, for agency-level or corporate users. This, in turn, moves to first place the requirements for communication efficiency and reliability, on the communication configuration flexibility between the user groups, on providing of circular communication, and on transmitted data confidentiality. There are a large number of trunking communication standards, which allow deployment of non-expensive and effective systems, as well as complicated multizone systems with a large number of users. The following parameters are used to classify the trunking systems: • Speech transmission method: analog with the help of frequency modulation at channel diversity 12.5/25 kHz, or digital with the speech data transmission rate not more than 4.8 kbit/s • Organization of access to the system: without the control channel with the scanning usage for finding the unoccupied line; with the distributed control channel

172

• •

• • • • •

6

Modulation and Signal Types in Modern Radioelectronic Means

in sub-tone range 0–300 Hz or with the dedicated control channel in which the data are transmitted with the rate up to 9.6 kbit/s Method of channel keeping: with the channel keeping during all communication sessions or with keeping during the period of one transmission Radio network configuration: single-zone (distance up to 70 km) or multi-zone, in the latter case we can distinguish the approach of the base station commutation, namely, with central switchboard (“star” type connection), with distributed switching or direct connection Method of channel organization: simplex, semi-duplex, or duplex Application area: agency-level, regional (commercial) or combined User number: small systems (up to 300 users), medium (up to 3,000) and large (more than 3,000) Openness of the data transmission protocol Type of multi-station access for the digital systems: with frequency separation (FDMA), time separation (TDMA,) or combined with frequency and time separation (FDMA+TDMA).

The main characteristics of some analog and digital trunking systems are shown in Table 6.5 and 6.6 [20, 21]. Other systems of the second group, namely, cellular communication networks are the most popular. First of all, they are usually considered as individual communication networks. The name of this type of radio communication comes from the main principle of its organization, in accordance with which the service zone is divided into cells or honeycombs, which have the form, depending on the real territory conditions. Each cell has its own base station (not necessarily in the center) designed for a definite number of users. When a user moves from one cell to another, the relay-race service transition occurs from one base station to another. The decision on station-changing can be made on the basis of a cellular phone signal-level comparison of both stations. All base stations are linked with each other through the switching center, which, in turn, is linked with other communication networks. In cellular communication, the principle of second frequency usage is widely used, i.e., the same frequency can be used by different base stations, for operation with different users without interference. This ensures the essential capacity of such systems [22]. Usually, one speaks about three generations of cellular communication: analog systems, which are already losing their position and are gradually being removed from operation; digital systems, which are widely used at present; and the prospective universal systems of the near future, which will allow the transmission of video and multi-media data at the rate of 2–10 Mbit/s. Development of the fourth generation of cellular systems has recently been announced, at which it is implied that mobile communication systems will provide a very high rate of data transfer (up to 100 Mbit/s) on the basis of batch switching. NMT-450 and AMPS are the well-known standards of analog cellular networks, while GSM, D-AMPS and CDMA are used for digital cellular systems. The main characteristics of such systems are shown in Tables 6.7 and 6.8.

Signals of Modern Radio Electronic Means

173

Table 6.5 Main characteristics of analog trunking systems Standard Application area Frequency bandwidth, MHz

Control channel

Altay (Russia)

LTR

SmartNet (Motorola)

MPT1327

Combined

Commerc.

Combined

Combined

136–174/ 400–512/ 800/900

136–174/ 300–350 403–520

300–308 400–512 (301.1375– 305.8125) users/336– 344 (337.1375– 341.8125) base stations No Distribut., 300 baud

Time of connection establishment in dispatcher mode, ms Channel 180 numbers in single-zone systems Maximal user number

EDACS Smart Trunk (Ericsson) II Agency- Combined level 300–344/ 30–56/ 800 136–174/ 806–825 400–512/ users/ 800/900 851–870 base stations)

500

Assigned, Assigned, Assigned 3,600 bit/s FFSK, 1,200 bit/s 500 300 500

500

20

28

24

28

16

5,000

16,383

1,036,800

16,383

4,096

No

Table 6.6 Main characteristics of digital trunking systems AEGIS (EDACS Ericsson)

iDEN (Motorola)

APCO 25

TETRA

TetraPol

Application area Frequency band, MHz

Combined

Commercial

Combined

Combined

Combined

800

800; 900; 1,500

380–403; 136–174; 410–490; 800 400–512

Type of access

FDMA

TDMA 25

136–174; 400–512; 800 FDMA (+TDMA) 12.5 (6.25)

VSELP 16-QAM

Standard

Channel 25 diversity, kHz Speech coding AME type Modulation GFSK type

TDMA

FDMA

25

12.5/10

IMBE

CELP

RPCELP

C4FSK (CQPSK)

Pi/4QPSK

GMSK (BT=0.25)

174

6

Modulation and Signal Types in Modern Radioelectronic Means Table 6.6 (continued)

Standard

AEGIS (EDACS Ericsson)

Data 9.6 – total transmission rate in the bandwidth of 25 kHz, kbit/s

Time of 1,000 connection establishment in dispatcher mode, ms Channel 28 number in single-zone systems Maximal user 16,383 number

iDEN (Motorola)

APCO 25

TETRA

TetraPol

64 – total in 19.2 – 9.6 per 28,8 – 4 voice 8 × 2/20 package channel channels by mode, 6 7.2 s with speech coding (26 s channels with totally). 7.2 with coding 500 500 300 500–800

144

1

32 (200)

24

1,440,000

48,000

–

28,672

Table 6.7 Main standard characteristics of cellular communication of the first generation (analog) Characteristic

NMT-450

Frequency range, MHz

463–467.5 (BS); 453–457.5 869.01–893.97; 824.01–848.97 25 30 2–45 2–20 180 666 30 96 3.5 8

Frequency bandwidth, kHz Cell radius, km Channel number of user station Channel number of base station Maximal deviation in control channel, kHz Frequencies of used service tone signals, kHz Signal type on the control channel sub-carrier/sub-carrier frequency, kHz/transmission rate, kbit/s Maximal deviation in speech channel, kHz Base station transmitter power, W User station transmitter power, W Minimal signal/noise ratio, dB Switching time at cell boundary, ms

AMPS

3.955/3.985/4.015/4.045

5.97/6/6.03 (SAT); 8 (ST)

FFSK/1.5/1.2

OQPSK/7.5/10

5

12

50 (0.15) 1.5–15 15 1,250

45 1–12 10 250

Signals of Modern Radio Electronic Means

175

Table 6.8 Main standard characteristics of cellular communication of the second generation (digital) Characteristic

GSM-900/GSM-1800 D-AMPS

CDMA (IS-95)

Frequency range, MHz

935.2–959.8; 890.2– 869.01–893.97; 914.8/1.805–1.880; 824.01–848.97 1/710–1/785 TDMA TDMA 8/16 3

873–876; 828–831

Method of access Speech channel number per carrier

Channel diversity, kHz Modulation type Cell radius, km Total rate of data transmission, kbit/s Speech conversion rate, kbit/s Speech conversion algorithm Data transmission rate for the user, kbit/s Minimal signal/noise ration, dB Base station transmitter power, W (category) User station transmitter power, W (category)

200 GMSK (BT = 0.3) 0.5–35/ up to 10 270.833

30 π/4DQPSK 0.5–20 48

CDMA Up to 62 (practically 20 for mobile commun.) 1,250 QPSK/OQPSK 0.5–25 1,288

13 RPE-LTR 9.6

7.95 VSELP 9.6

9.6 CELP 14.4

9 2.5 (8); 20 (5); 320 (1) To 0.8 (5); to 20 (1)

16 25–50

6 2–20

0.6

To 0.2

As we can see from Table 6.8, the most popular standard of cellular communication, GSM (Global system for Mobile communications), uses the multiple access technology with time division (Time Division Multiple Access – TDMA). The speech message transmission process is provided by the following approach. The voice codec (vocoder with the long duration linear forecast) provides the quantized audio signal processing by 20 ms segments (160 8-bit samples). At codec output, the segment is represented by a 260-bit sequence, in which 182 bits are channel coded (to block and convolutional coding). As a result, 456 bits of data are divided into eight blocks with interleaving by the block-diagonal algorithm. After interleaving, two blocks of 57 bits are included into the so-called normal time interval (Normal Burst – NB) with duration about 576.9 μs. This interval (slot) has the following time structure: the first three bits are the protective sheet, used as a start flag; 57 bits are the first voice data block; one control bit defining the type of transmitted data (voice/control); 26-bit reference sequence; one control bit defining the type of transmitted data (voice/control); the second block from 57 bits; the protective sheet (3 bits) as a stopping flag, the last 8.25 bits as a protective interval. Only one slot from the eight included in the TDMA frame with 4.625 ms duration is dropped to each user (subscriber). This allows the time division for transmission and receiving processes, by means of the shift of channel interval assigned to the mobile transmitters and to the base station. The base station always transmits three

176

6

Modulation and Signal Types in Modern Radioelectronic Means

time slots earlier than the subscriber terminal. Thus, four TDMA frames transmit 456 bits, corresponding to 20 ms of voice signal. Another GSM standard peculiarity is the message receiving (transmission) in the interval assigned to the subscriber of each following frame on other fixed frequency. At that, the time for frequency changing is about 1 ms. Slow frequency jumps (SFH – Slow Frequency Hopping) such as these improve the system functioning, under the conditions of multi-path radio waves propagation. Each base station can receive up to 16 frequencies, and frequency number and transmission power is defined depending on location and loading. Besides the normal time intervals, four types of the same duration and bit length (156.25) slots are used for data transmission in the control channel: frequency adjustment FB, time synchronization SB, setting interval DB, and access interval AB. The frames contained the voice messages are organized in multi-frames of 120 ms duration (26 TDMA frames, at that, the frames 13 and 26 are concerned with the control channel). The control multi-frames have duration of 235.385 ms and include 51 control channel frames. In turn, the multi-frames are united into super-frames with duration of 6.12 s (51 multi-frames of traffic channel or 26 multiframes of control channel), and 2048 super-frames represent a hyper-frame with duration of 12,533.76 s. The number of simple TDMA frames, within the limits of hyper-frame, is used for the transmitted data encoding. In addition to voice, GSM standard allows the transmission of other data in asynchronous mode, with a rate up to 9.6 kbit/s. The standard extension is possible for batch data transmission inside the radio network with the higher rate (115 kbit/s), due to batch organization of the additional time slots at not-used channel locking on the basis of the “capacity on request” principle (for instance, all eight slots of one TDMA frame) GPRS (General Packet Radio Service). At that, outgoing and incoming traffic organization is possible independently from each other, which is very important at the asymmetrical data exchange. The physical channels are activated during the direct transmission only, which allows their usage together with the other subscribers. The GPRS system supports all widespread protocols of data transmission, ensuring the operation of most various applications. In EDGE (Enhanced Data rates for Global Evolution) technology, which is usually considered as the evolution of GSM, it is possible, also, to join the time slots (their structure remains the previous, as for GSM). But, because of PSK8 phase shift-keying usage, the EDGE standard has the increased physical transmission rate in the channel up to 384 kbit/s (EGPRS). Code division multiple access (CDMA) technology of the CDMA IS-95 standard is based on the principle of spectrum extension by direct sequence method (DSSS – Direct Sequence Spread Spectrum). Each data bit is changed by the definite length binary combination fixed for each channel and units, and zeros differ by the full bit inversion of this long sequence. The necessary condition of channel division is the orthogonality of the used coding sequences, such as Walsh sequences or M-sequence. A voice message in the CDMA system is transmitted as follows. Vocoder transfers the data stream of digitized voice, decreasing its rate from 64 kbits to 8.55

Signals of Modern Radio Electronic Means

177

kbit/s. After that, the correcting symbol addition occurs and the rate is increased to 9.6 kbit/s. Data are divided into 20 ms segments and convolution coded with the factor of 0.5 on the base station and 0.33 on the subscriber station, and then the data are interleaved. The interleaving is provided by the matrix line filling and matrix column reading, and plays the role of protection against group errors. The resulting digital sequence is transformed at the base station by two modules, each having a 64th symbol of the maximal length coded sequence, which is formed as a result of logical operations with a pseudo-random binary sequence with the length of 242 –1 and with the individual 42-bit mask of the subscriber. After that, the stream extension of one Walsh 64-bit sequence is provided, which is assigned to each channel of the base station. At that time, the digital stream rate is increased from 19.2 kbit/s to 1.2286 Mbit/s. Further, the stream is divided between in-phase and quadrature channels, and is additionally extended in each channel with the help of the mutual-for-all-network shorter pseudo-random sequence (with the period of 215 –1 and the rate of 1.3288 Mbit/s), using the “exclusive OR” logical operation. After that, the streams are filtered and pass to adders, when group signals of the I and Q channels are formed, and, further, they pass to the quadrature phase modulator. The output signal contains the pilot-signal, in which the short code is transmitted only (the data signal is absent, zero Walsh sequence consists of units only). The pilot-signal, being the reference one, allows, at the receiving side, for data extraction to be provided from the received signal mixture, using the correlation approach. After interleaving, the data stream, acting at the rate of 28.8 kbit/s at the subscriber side, is divided into 6-bit sections. One of the 64 Walsh sequences (26 = 64) is brought in correspondence with each similar fragment. After this procedure, the stream of 307.2 kbit/s is transformed with the help of a long coding sequence, similar to the sequence at the base station. Each element of the Walsh sequence is represented by four elements of pseudo-random sequence. After that, the stream is divided between the channels, is additionally transformed by the short coding sequence, is delayed by half-symbol in the quadrature channel – as is required by PSK with the shift, is filtered and passes to the modulator. The strict requirements for synchronization and frequency stability are a peculiarity of the CDMA cellular communication system, therefore, the correction at the base stations is provided on the basis of GPS signals. The need to equalize subscriber station power at the receiving place (on the base station) is another peculiarity, since this factor affects on the system’s real capacity and its operating distance. For such equalization, subscriber transmitter-power distance control is provided in the direct, as well as in inverse, channels. At that, the regulation range is 84 dB with 1 dB gradation. The CDMA IS-95 standard (or cdmaOne) has the IS-95b modification, which allows the joining of up to eight channels of the base station, with the maximal permissible transmission rate of up to 115 kbit/s, and makes provision for power control accuracy, up to 0.25 dB. Further evolution of the standard is heading in the direction of the cdma2000 project. The modification of cdma2000 1x has an additional channel group with the orthogonal carrier shift (it was planned earlier for IS-95c), which twice increases the system frequency efficiency while maintaining inverse compatibility with the

178

6

Modulation and Signal Types in Modern Radioelectronic Means

previous standard versions. The data transmission rate is 144 kbit/s. In the variant cdma2000 1xEV-DO (Single carrier EVolution-Data Only), the PSK8 modulation application (earlier planned for IS-95c-HDR) increases the data transmission rate in the direct channel, up to 2.4 Mbit/s. In the next version of cdma2000 1xEV-DV (Single carrier EVolution-Data and Voice), there is a possibility of concurrent transmission of voice and data, due to the dynamic channel extraction and more high-order modulation (16-QAM). The maximal data transmission rate at that is 3.1 Mbit/s, which satisfies the requirements for the third generation (3G) of cellular communication. The next development stage of cdma2000 3x (6x, 9x) anticipates spectral bandwidth growth either at using several carriers, or due to the increasing of the digital transmitted stream rate (3.6864 Mbit/s for 5 MHz bandwidth). In Russia, the standard cdma2000 is implemented as CDMA-450 under the SkyLink system, to substitute for the analog standard NMT-450 in the frequency range of 453.0–457.4/463.0–467.4 MHz. The European variant of 3G cellular networks, UMTS (Universal Mobile Telecommunications System), also anticipates the use of code division technology, in the standard WCDMA (Wideband-CDMA). The main features of WCDMA, compared with cdma2000, are the digital stream rate (3.84 Mbit/s), the direct channel structure, and the asynchronous operation of the base stations, due to Gold coding sequences application. The operating range of WCDMA systems is near 2 GHz (1,920–1,980/2,110–2,170 MHz), and the frequency bandwidth is equal to 5 MHz. The third generation of cordless phones of the DECT (Digital Enhanced Cordless Telecommunications) standard is one more element of the cellular network of the third generation concept, which allows the creation of so-called micro-cellular communication systems [23, 24]. The main characteristic of the DECT standard and the previous generations of cordless phones are shown in Table 6.9. The average power of a DECT subscriber phone is 10 mW, which allows communication realization inside the premises at distances up to 50 m, while in the open area up to 300 m. Minimal signal/noise ratio is 12 dB. TDMA frame duration in the DECT system is 10 ms, 16 frames organize the multi-frame. The frame itself is divided into 24 time segments of 48-bits each (approximately 416.7 μs). The first 12 intervals are purposed for transmission from the base station, but the second 12 are for transmission from the portable terminal. One duplex channel is formed by the intervals spaced by 12 slots. At voice transmission, 320 bits are used. The other bits are the synchro-code (32 bits), signalization code (48 bits), error protection code (16 bits), checking symbols (4 bits) and the protective interval (60 bits). The base DECT station transmits the signal constantly during the multi-frame, at least, via one channel in broadcasting mode, playing the beacon role for the subscribers. The full set of system data is included in the transmitted signal. The principle of continuous dynamic channel selection (Continuous Dynamic Channel Selection – CDCS) is implemented in the system, at which each of the subscriber terminals has access to any of 120 system channels. During the connection, the receiver selects the channel with the best communication quality and, in the future, its

Signals of Modern Radio Electronic Means

179

Table 6.9 Main characteristics of cordless phones Standard

CT1

CT2 (CT2+)

CT3

DECT

Frequency range, MHz Frequency channel number Channel bandwidth, MHz Access method

900

862–866

1,880–1,900

40

864–868.2 (944–952) Up to 6

4

10

0.05

0.1

1

1.728

FDMA

FDMA (TDD)

MC/TDMA (TDD)

MS/TDMA

1

8

(TDD) 12

–

ADPCM

ADPCM

ADPCM

–

32

32

32

FM

GFSK (VT = 0,5) 18

GFSK (VT = 0,5) 160

GFSK (VT = 0,5) 288

–

72

640

1152

Channel number per carrier Voice coding method Voice coding rate, kbit/s Modulation Carrying frequency deviation, kHz Transmission rate, kbit/s

changing may occur during the same session. This can be realized due to the background scanning and signal level estimation among all frequency-time positions. For high-rate data transmission, the specification DPRS (DECT Packet Radio Services) has been developed, which provides asynchronous transmission without connection establishing, at a rate of up to 552 kbit/s. In spite of the progress in cellular communication, systems of personal call or paging communication (so called, “mobile telegraph”) continue to exist and to develop [25]. The parameters of the main paging protocols are shown in Table 6.10. RDS and MobiDARC systems are used for paging communication, which uses the compression in FM broadcasting. We discussed these systems above in the appropriate section. In a number of cases, under the absence of necessary communication infrastructure, communication is organized through radio modems, when the requirement of reliability and independence on the available communication systems (or due to technological and economical reasons) is very important. For most problems of telemetry, control, location data transmission, and of other service data for fixed or mobile application, radio modems, which operate in narrow frequency bandwidth with a relatively low data-transmission rate, are very suitable. At that, one can organize “point to point” radio lines and multiple access systems, for example, in accordance with AX.25 protocol. To illustrate this, the parameters of such specific radio modems of the Dataradio Canadian Company are shown in Table 6.11. Broadband radio modems, possessing a data transmission rate of 2 Mbit/s and higher, have found their application in multi-channel systems, in fixed broadband

180

6

Modulation and Signal Types in Modern Radioelectronic Means

Table 6.10 Characteristics of paging communication protocols

Protocol name POCSAG

ERMES

Used frequencies, MHz

Transmission Modulation rate, baud

Any paging 512, 1,200, range 2,400 (135–175, 278–284, 322–328, 406–423, 435–480, 495–512) 169.425– 6,250 169.800 (16 channels)

Message Channel numbering and band, kHz roaming

2FSK with spacing 25 4.5 kHz; FFSK with deviation 4 kHz

Yes

4FSK with the 25 spacing: +4.6875 kHz 10 +1.5625 kHz 11 –1.5625 kHz 01 –4.6875 kHz 00 2FSK with spacing 25 4.8 kHz; 4FSK with spacing: +4.8 kHz 10 +1.6 kHz 11 –1.6 kHz 01 –4.8 kHz 00 25 or 50

Yes

FLEX

Any paging ranges

1,600, 3,200, 6,400

ReFLEX25 – transmission to pager, receiving from pager ReFLEX50 – transmission to pager, receiving from pager InFLEXion – transmission to pagers, receiving from pagers

929–931, 940–941; 901–902

1,600, 3,200, 6,400; 800, 9,600

930–931, 940–941; 901–902

Up to 25,600

50

Yes

930–931, 940–941; 901–902

112 kbit/s, digital voice compression

50

Yes

Yes

Yes

access systems (BWLL – Broadband Wireless Local Loop), in local radio networks of data transmission (WLAN – Wireless Local Area Network), and in metropolitan radio networks of data transmission (WMAN – Wireless Metropolitan Area Network). The operating frequencies of such types of modems are usually located in the limits of 900–5,800 MHz (mainly 2.4, 3.5, and 5.8 MHz), although there are solutions for higher frequency zones. The radiated power is from 1 to 800 mW. Radio interface can be developed by the company or corresponded to the standard

Signals of Modern Radio Electronic Means

181

Table 6.11 Main characteristics of dataradio narrow-band modems Name

Integra-TR

MobilPac II

GeminiG3 700

Frequency range, MHz

132–174; 380–512; 928–960

138–174; 400–520; 800–870; 896–940

Frequency grid step, kHz Output power, W Maximal time of continuous transmission, s Sensitivity

6.25; 12.5; 25

792–803 (transmission) (766–773); 762–764 (receiving) 50

1–5 30

5–25

10–25

0,35 μV

–110 dBm (19.2 kbit/s) 9.6; 14.4 (12.5 kHz) 9.6; 19.2 (25 kHz)

–95 dBm (128 kbit/s)

MSK

MSK Semi-duplex

Mobile

Mobile

Data transmission rate, kbit/s Modulation type Operation mode Type

2,400, 4,800, 9,600 or 19,200 (25 kHz) DRCMSK Simplex or semi-duplex Stationary

64, 96, 128 (50 kHz)

specifications offered by the IEEE (Institute of Electrical and Electronic Engineers). The topology and architecture of implemented systems depends on the character of the current task. The features of several broadband, fixed, wireless access systems are shown in Table 6.12 [26]. At present, there are four IEEE standard groups for wireless communication. The standards of the IEEE 802.11 group define the components and characteristics Table 6.12 Systems of broadband wireless access Name

WaveNet WaveNet BreezeACCESS Access 2458 Access 3500 3,5 WaveGain

Manufacturer Wireless

Wireless

BreezeCOM

Range, GHz

2.4/5.8

3.4–3.6

3.4–3.6

Radio interface Band, MHz Modulation

FH

FH

FH-CDMA

17–78 GMSK

1.75 4-QAM

2 GFSK

1.625

Up to 3

Up to 8 devices

Up to 54 Mbit/s Up to 96 Mbit/s

Transmission 0.6 rate, Mbit/s Max capacity Up to 6 base station devices

InnoWave ECI Wireless Systems 3.4–3.6

PMP

OnDemand

P-COM

Lucent Technologies

10, 24, 26, 28, 31, 38 DS-CDMA FDMA, TDMA 5, 10, 20 – 8PSK QPSK, 16-QAM, 64-QAM Up to 2 Up to 40

10, 26, 38 ATM 7, 12.5, 14 4-QAM, 16-QAM 8, 13, 16 26

Up to 4,800 – Mbit/s

182

6

Modulation and Signal Types in Modern Radioelectronic Means

of wireless LAN (local area networks) on the physical level and on the environment access level, as well as the interaction possibilities with the existing network. The standards of the IEEE 802.15 group define the interaction characteristics in personal (home) wireless networks (WPAN – Wireless Personal Area Network) and with autonomous devices. The standards group 802.16 contains the recommendations on “last mile” system and the wireless metropolitan network organization. The standard 802.20, which is at the design stage, determines the requirements for mobile wireless broadband networks. There is also the family of European standards (ETSI) of highly productive wireless networks HiperLAN (High Performance Local Area Network). Brief information on the above-mentioned standards of wireless communication is shown in Table 6.13 [27, 28].

International System for Signal Designation International designation of radio emission [1] is approved in accordance with its necessary frequency bandwidth (four symbols – one letter and three numerals) and its classification (five symbols), for example, 100HA1AAN means AM telegraphy with the rate of 25 words/minute (Morse code), with oral receiving; 2K70J3EJN means single-side telephony with commercial quality, suppressed carrier, and frequency band of a standard phone channel, 300KF9EHF means stereo FM broadcasting with deviation of 75 kHz, and the frequency multiplexing of the RDS system and phone channel with bandwidth of 16 kHz with FM on a sub-carrier of 67 kHz. The value of necessary bandwidth is defined by calculation, or by direct measurement [4], and is expressed in hertz (H) at bandwidths from 0.001 to 999 Hz, in kilohertz (K) at bandwidths from 1 to 999 kHz, and, in a similar manner, in megahertz (M) and gigahertz (G) with application of the rounding rule. The letter is placed instead of decimal comma, for instance, H030 – 0.03 Hz; 240H – 240 Hz; 35K5 – 35.5 kHz; 550 K – 550 kHz; 5M00 – 5 MHz; 22G0 – 22 GHz. The first three symbols of radio emission designation are obligatory, since they characterize its main parameters. The first symbol (Table 6.14) indicates the modulation type (short-term modulation, used, for instance, to transmit recognition signals, may not be considered because the necessary bandwidth is not increased). The second symbol (Table 6.15) designates the character of the signal (signals) modulating the main carrier. The third symbol designates the type of transmitted data (Table 6.16). The fourth (Table 6.17) and fifth classification symbols (Table 6.18) are additional, and provide detailed information about the signal and about the used multiplexing method, relatively. In the case where it is impossible to use any of the definite codes for the additional symbols, it is permitted to locate the symbol “-” at this place. It is stated in [29] that five-place designation is quite enough for the purposes of frequency regulation, but, for monitoring purposes, the international designation

Standard (year) 802.11 (1997)

802.11a (1999) 802.11b (1999)

Name

Range, GHz

Transmission rate, Mbit/s

Brief characteristic

WLAN

2.4 (2.4– 2.4835)

1 and 2

5 2.4

6, 12 and 24 (9, 18, 36, 48 and 54)/ (turbo to 108) 5.5 and 11 (22 for b+)

2.4

Up to 54

Radio interface: DSSS (Barker sequence), FHSS. 22 MHz channel band. DBPSK and DQPSK modulation. Access to CSMA/CA environment (Carrier Sense Multiple Access with Collision Avoidance). Cellular architecture. DWEP (Wired Equivalent Privacy)/WPA (Wi-Fi Protected Access) data protection. OFDM radio interface (48 sub-carriers). BPSK, QPSK, 16-QAM, 64-QAM modulation. Communication distance up to 50 m. DSSS (Barker sequence; CCK – Complementary Code Keying) radio interface. DBPSK, DQPSK modulation. Communication distance up to 100 m. OFDM (54 Mbit/s), DSSS (Barker sequence; CCK; PBCC - Packet Binary Convolutional Coding 33 Mbit/s) radio interface. BPSK, QPSK, 16-QAM, 64-QAM; DBPSK, DQPSK modulation. Communication distance up to 100 m. Development of 802.11b. Extends the possibilities of 802.11 standard using MIMO and the channels with extended bandwidth up to 40 MHz. Inverse compatibility with 802.11a/b/g. 23-79 MHz bandwidth (depending on the region). 1 MHz channel bandwidth. FHSS/TDD radio interface (1,600 jumps per second). GFSK modulation (BT = 0.35). Communication distance up to 10 m (class 1), up to 100 m (class 3). OQPSK modulation method. Distance up to 100 m. Up to 245 users. Low energy consumption (home electronics, portable multi-media devices). Data coding in accordance with AES128.

Wi-Fi

802.11g (2002)

802.11n

Design stage

2.4 and 5

From 100

802.15.1 (1999)

Bluetooth WPAN

2.45

0.4339 symmetr., 0.7233/0.0576 asymmetr.

802.15.3

WPAN

2.45

Up to 55 (11, 22, 33 and 44)

Signals of Modern Radio Electronic Means

Table 6.13 Standards of the data transmission wireless networks

183

Standard (year) 802.15.4a

802.15.4

Transmission rate, Mbit/s

Brief characteristic

UWB design stage ZigBee

3.1–10.6

110–480

2.4; 0.915; 0.868

Up to 0.25 at 2.4 MHz

Ultra-broadband high-rate data transmission by short pulses of low power and low-rate location tracking for wireless devices and objects. For autonomous equipment and home devices with integrated wireless sensors of low energy consumption. Coding in accordance with AES128. Uses the existing networks 802.11 and 802.15 as the channels for its traffic. Operation radius up to 10 m. Creation of high-rate scaled multi-user home and business networks supporting any equipment. Straight visibility. 20, 25, 28 MHz channel bandwidth. QPSK, 64-QAM modulation. Cell radius is 2-5 km (typical). OFDM (256 sub-carriers), OFDMA (2048 sub-carriers) radio interface. QPSK, 64-QAM modulation. Cell radius is 6–9 km (50 km max.). Operation beyond straight visibility, using the reflections. Final variant of standards 802.16 and 802.16a. Cell radius up to 50 km. Operation on reflections. Channel bandwidth 1.5–20 MHz.

802.16 (2002)

BWA

10–66

32–134.4

802.16a (2003)

WiMAX

2–11

Up to 70

802.16d

WiMAX design stage Design stage

2–11

Up to 75

2–6

15 (up to 5 km)

MBWA design stage

Up to 3.5

Less than 1 “down”, less than 0.3 “up” (for 1.25 MHz channel) Up to 25

802.16e

802.20

HiperLAN2

5.1–5.3

Channel bandwidth is 5 MHz. Cell radius is 2–5 km. Mobility is up to 150 km/h. Communication on the basis of license. Operation on reflections. OFDM radio interface. Mobility is up to 250 km/h. Channel bandwidth is 2 × 1.25/2 × 5/2 × 10 MHz. Action radius is up to 15 km. Creation of universal networks on the base of IP protocol. OFDM radio interface. Action radius is up to 100 m. Cellular architecture. Mobility is up to 36 km/h.

Modulation and Signal Types in Modern Radioelectronic Means

Range, GHz

6

Name

184

Table 6.13 (continued)

Signals of Modern Radio Electronic Means

185

Table 6.14 Coded designations of modulation type of the main carrier Symbol

Meaning

A B C D F G H J K L M N P Q R V W

AM: two side bands AM: with independent side bands AM: partially suppressed single side band Combination or sequence of AM and angular modulation Angular modulation: FM Angular modulation: PM AM: single side band with full carrier AM: single side band with suppressed carrier Pulse emission: amplitude-pulse modulation Pulse emission: pulse-width modulation Pulse emission: phase-pulse modulation Modulation is absent Pulse emission: non-modulated pulse sequence Pulse emission: angular modulation during the pulse duration AM: single side band with partially suppressed carrier or with carrier variable level Pulse emission: combination or other pulse modulation None from the mentioned, other cases, when the emsission consists of the main carrier which is modulated either simultaneously, or by the settled in advance sequence, with the combination of two or more of the following modulation methods: amplitude, angular, pulse. Other cases

X

Table 6.15 Coded Designation of modulating signal type Symbol

Meaning

0 1 2 3 7 8 9 X

The modulating signal is absent 1 channel/quantized or digital data/ no modulation sub-carrier 1 channel/quantized or digital data/ modulation sub-carrier 1 channel/analog data 2 or more channels/digital data 2 or more channels/analog data Combined system with minimum 2 channels (with digital and analog data) Other cases

should be supplemented by the widespread name of the used system, should show its type (for instance, start/stop, batch, with overinterrogation, with error correction, multi-tone, navigational/radar, etc.), used alphabet, bit number per symbol, rate in baud, repetition cycle, etc. That is why one may use the sixth and seventh classification symbols to designate the system group, and the system in the group (Table 6.19). For multi-tone systems, after the emission designation, the following data can be transmitted: tone duration, ms/intertone shift, Hz/tone number, and, for multichannel systems: shift in channel Hz/channel diversity, Hz/channel number.

186

6

Modulation and Signal Types in Modern Radioelectronic Means

Table 6.16 Coded designation of transmitted data type Symbol

Meaning

A B C D E F N W X

Telegraphy, acoustic receiving Telegraphy, automatic receiving Photo-telegraphy Data transmission, telemetry or telecontrol Telephony (including audio radio broadcasting) Television Data is absent Combination of above-mentioned types None from above-mentioned types

Table 6.17 Coded designation of additional detailed information about the signal Symbol Meaning A B C D E F G H J K L M N W X

Two-position code with different number/duration of code elements Two-position code with equal number of elements equaled on duration, the error correction is absent Two-position code with equal number of elements equaled on duration, the error correction is present Four-position code, each state corresponds to one signal element Multi-position code, each state or their combination corresponds to one symbol Multi-position code, each state corresponds to one signal element Audio for radio broadcasting channel, mono Audio for radio broadcasting channel, stereo or quadraphonic Audio of commercial quality Audio of commercial quality with frequency inverse or the bandwidth splitting Audio of commercial quality with separate frequency modulated signals for signal level regulation at demodulation Monochrome signal Color signal Combination of above-mentioned signals None of the above-mentioned signals

Table 6.18 Coded designation of multiplexing method Symbol

Meaning

C

Multiplexing with coded division (including the spectrum extension methods) Multiplexing with frequency division Multiplexing is absent Multiplexing with time division Combined multiplexing with frequency-time division Other

F N T W X

Signals of Modern Radio Electronic Means

187

Table 6.19 Coded designation of system groups and the system in the specific group Symbol Meaning A C

E

F

H

Symbol Meaning

Morse Asynchronous B C D K

Baudot telex Russian telex Arabian telex ASCII telex

A B C D E F K L M

ARQ-1000 duplex ARQ E-3 324 TOR 1 channel 324 TOR 2 channels 324 TOR 4 channels 242TOR 2 channels ARQ-N FOL-ARQ TORG 10–11

A C D E F G H I K L N X

Simplex/ SITOR ARQ-1000 simplex SWED-ARQ ARQ6-70 ARQ6-90 ARQ6-98 UN-ARQ HC-ARQ RS-ARQ ARTRAC PACKET P 162

A B C D E F G H K L M

SITOR F7B SITOR F7B-1 SITOR F7B-2 SITOR F7B-3 SITOR F7B-4 SITOR F7B-5 SITOR F7B-6 c ASCII C Baudot F7 Baudot/Morse F7 other code / Morse

A

ARTRAC

A B C D E F

FEC-100 SITOR-B FEC1000 simplex Autospex ROU-FEC HNG-FEC

ARQ with continuous data transmission

ARQ with package data transmission

TWINPLEX (doubled system)

J

Unknown

K

System with FEC (direct error correction)

188

6

Modulation and Signal Types in Modern Radioelectronic Means Table 6.19 (continued)

Symbol Meaning M

Symbol

Meaning

A B C F

Piccolo MK6 ITA2 Piccolo MK6 ITA5 Piccolo MK1/330 Piccolo 1◦ 025/040/034 Piccolo 2◦ 025/010/034 Piccolo 3◦ 100/040/034 Piccolo 4◦ 100/010/034 Coquelet MK1 Coquelet MK1 TT2300b

Multi-tone

G H I L M P N

Radio navigation and radar technology

International Frequency Range Distribution The frequency range distribution, in accordance with the international agreements, is shown in Tables 6.20–6.27.

Table 6.20 Frequency ranges for radio broadcasting below 30 MHz.

Frequency range, MHz

Note

0.15–0.285 0.525–1.605 2.300–2.498 3.200–3.400 3.950–4.000 4.750–4.995 5.006–5.06 5.900–6.200 7.100–7.350 9.400–9.900 11.600–12.100 13.570–13.870 15.100–15.800 17.480–17.900 18.900–19.020 21.450–21.850 25.670–26.100

200–735.3 m (LW) 575–187 m (MW) 120-m range (in the tropics) 90-m range (in the tropics) 75-m range (in the tropics) 60-m range (in the tropics) 49-m range 41-m range 31-m range 25-m range 19-m range 16-m range 13-m range 11-m range

International Frequency Range Distribution Table 6.21 Frequency ranges for amateur communications below 30 MHz

189

Range

Frequency, MHz

Wavelength, m

160 m 80 m 40 m 30 m 22 m 20 m 17 m 15 m 12 m 10 m

0.1357–0.1378 1.810–1.850 3.5–3.8 7.0–7.3∗ 10.10–10.15 13.57–13.60 14.00–14.35∗ 18.068–18.168 21.00–21.45∗ 24.89–24.99 28.10–29.7∗

2210.76–2177.07 166.7–150.0 85.7–75.0 42.9–41.1 29.7–29.6 22.1–22.06 21.4–20.9 16.6–16.5 14.3–14.0 12.1–12.0 10.7–10.1

∗

7.0–7.3 MHz (step 100 Hz); 14.00–14.35 MHz (step 250 Hz); 21.00–21.45 MHz (step 450 Hz); 28.00–29.7 MHz (step 1.7 kHz). Table 6.22 SW Frequency ranges for fixed (stationary) transmitters Frequency, kHz 1,606.5–1,625 1,635–1,800 1,850–2,160 2,194–2,498 2,502–2,850 3,155–3,400 3,500–3,900

3,950–4,063 4,438–4,650 4,750–4,995 5,005–5,480 5,730–5,900 6,765–7,000 7,350–8,195

9,040–9,400 9,900–9,995 10,100–11,175 11,400–11,600 12,100–12,230 13,360–13,570 13,870–14,000

14,350–14,990 15,800–16,360 17,410–17,480 18,030–18,068 18,168–18,780 19,020–19,680 19,800–19,990

20,010–21,000 21,850–21,924 22,855–24,890 25,010–25,070 25,210–25,550 26,175–28,000 29,700–30,005

Table 6.23 SW Frequency ranges for the communication between marine objects Frequency ranges, kHz

Notes

1,606.5–1,625 1,635–1,800 2,045–2,160 2,170–2,173.5 2,190.5–2,194 2,625–2,650 4,000–4,063 4,063–4,438 6,200–6,525

Deviations from the indicated values are possible. 2,625–2,650 kHz – marine mobile and radio navigational communication.

8,100–8,195 8,195–8,815 12,230–13,200 16,360–17,410 18,780–18,900 19,680–19,800 22,000–22,855 25,070–25,210 26,100–26,175

The deviations from the indicated values are possible. The frequencies of 26,945 and 26,960 kHz may be used on a secondary basis by the intrusion protection systems with the radiated power up to 2 W.

190

6

Modulation and Signal Types in Modern Radioelectronic Means

Table 6.24 SW frequency ranges for aviation communication Frequency ranges, kHz

Notes

2,850–3,155 3,400–3,500 3,800–3,950 4,650–4,850 5,450–5,730 6,525–6,765 8,815–9,040

Deviations from the indicated values are possible. Frequencies of 3,023 kHz, 5,680 kHz, 8,364 kHz (carrier) can be used by the mobile service stations participating in mutual search and rescue operations, and also for search and rescue of piloted spaceships. .

10,005–10,100 11,175–11,400 13,200–13,360 15,010–15,100 17,900–18,030 21,924–22,000 23,200–23,350

Table 6.25 SW frequency ranges for land mobile Transmitters Frequency ranges, kHz

Notes

1,606.5–1,625 1,635–1,800 1,850–2,160 2,194–2,498 2,502–2,625 2,650–2,850 3,155–3,400 3,500–3,900 4,438–4,650

4,750–4,995 5,060–5,480 5,730–5,900 6,765–7,000 7,350–8,100 10,150–11,175 13,410–13,570 13,870–14,000 14,350–14,990

18,168–18,780 20,010–21,000 23,000–23,200 23,350–24,890 25,010–25,070 25,210–25,550 26,175–27,500 27,500–28,000 29,700–30,005

Deviations from the indicated values are possible. Frequencies of 2,130 and 2,150 kHz are used by the radio stations of railway radio communication systems, in phone mode. Frequencies of 2,444 and 2,464 kHz are used for subway train radio communication means.

Table 6.26 TV channel frequencies

Channel number Meter waves 1 2 3 4 5 6 7 8 9 10 11 12 Decimeter waves 21 22

Frequency boundaries of the channel, MHz

Carrying frequency of image, MHz

Carrying frequency of accompanied sound, MHz

48.5–56.5 58–66 76–84 84–92 92–100 174–182 182–190 190–198 198–206 206–214 214–222 222–230

49.75 59.25 77.25 85.25 93.25 175.25 183.25 191.25 199.25 207.25 215.25 223.25

56.25 65.75 83.75 91.75 99.75 181.75 189.75 197.75 205.75 213.75 221.75 229.75

470–478 478–486

471.25 479.25

477.75 485.75

Conclusion

191 Table 6.26 (continued)

Channel number

Frequency boundaries of the channel, MHz

Carrying frequency of image, MHz

Carrying frequency of accompanied sound, MHz

23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

486–494 494–502 502–510 510–518 518–526 526–534 534–542 542–550 550–558 558–566 566–574 574–582 582–590 590–598 598–606 606–614 614–622 622–630 630–638 638–646 646–654 654–662 662–670 670–678 678–686 686–694 694–702 702–710 710–718 718–726 726–734 734–742 742–750 750–758 758–766 766–774 774–782 782–790

487.25 495.25 503.25 511.25 519.25 527.25 535.25 543.25 551.25 559.25 567.25 575.25 583.25 591.25 599.25 607.25 615.25 623.25 631.25 639.25 647.25 655.25 663.25 671.25 679.25 687.25 695.25 703.25 711.25 719.25 727.25 735.25 743.25 751.25 759.25 767.25 775.25 783.25

493.75 501.75 509.75 517.75 525.75 533.75 541.75 549.75 557.75 565.75 573.75 581.75 589.75 597.75 605.75 613.75 621.75 629.75 637.75 645.75 653.75 661.75 669.75 677.75 685.75 693.75 701.75 709.75 717.75 725.75 733.75 741.75 749.75 757.75 765.75 773.75 781.75 789.75

Conclusion The present chapter details briefly information about the generally-accepted division of radio frequency range and the properties of each sub-range. The main schemes of analog and digital modulation used in the broadcasting, communication and data transmission systems are analyzed. The mathematical relations are discussed, with

192

6

Modulation and Signal Types in Modern Radioelectronic Means

Table 6.27 Frequency ranges above 30 MHz for amateur radio communication (including satellite) Range

Frequency

Wavelength

6m 2m 1.5 m 70 cm 33 cm 23 cm 13 cm 9 cm 5 cm 3 cm 1.25 cm

50–54 MHz 144–148 MHz 220–225 MHz 420–450 MHz 902–928 MHz 1,215–1,300 MHz 2,300–2,450 MHz 3,300–3,500 MHz 5,650–5,925 MHz 10.00–10.50 GHz 24.00–24.25 GHz

6.00–5.56 m 2.08–2.03 m 1.36–1.33 m 71.4–66.7 cm 33.3–32.3 cm 24.7–23.1 cm 13.0–12.2 cm 9.09–8.57 cm 5.31–5.06 cm 3.00–2.86 cm 1.25–1.24 cm

the necessary explanations; and, time and spectral diagrams are used to illustrate the given information. Examples of modern, radio equipment signals are considered. A review of AM and FM radio broadcasting and television systems is given, and the distinctions between the DRM, T-DAB and DVB-T broadcasting formats are discussed. The chapter includes information about trunking, cellular, and paging communication, and about the narrow-band and high-rate digital systems of data transmission, including the IEEE standards for wireless communication. The international designation system for radio signals is given, together with the frequency range distribution, in accordance with the international agreements.

References 1. Radio Communication Regulations of Russian Federation (in Russian). Moscow, 1999. 2. Sklar, B., Digital Communications: Fundamentals and Applications, 2nd Edition, PrenticeHall, 2001. 3. Proakis, J.G., Digital Communications, 4th Edition, Mc Graw-Hill, 2001. 4. Recommendation ITU-R SM.328-10. Spectra and bandwidth of emissions 5. Radio Broadcasting and Electro-Acoustics (in Russian). Under edition of Kovalgin, Yu.A. Moscow, Radio i sviaz, 1999. 6. Rikhter, S.G., Digital Radio Broadcasting (in Russian). Moscow, Goriachaya Linia – Telecom, 2004. 7. ETSI ES 201 980 v 2.2.1 (2005–10). Digital Radio Mondiale (DRM). System Specification. 8. Radio Data Code Manual. Klingenfuss Publication, 2003. 9. Bill Henry K9GWT and Raymond C. Petit W7GHM, “HF Radio Data Communication”, Communication Quarterly, Vol. 2, No. 2 Spring 1992, p. 11. 10. Steve, F., ARRL’s HF Digital Handbook, 2007. 11. Radio Communications in the Digital Age, Volume 1: HF Technology, 2nd Edition, Harris Corporation, 2005, 96 pp. 12. Davies, J., and Carr, J.J., Newnes Radio and RF Engineer’s Pocket Book, 2nd Edition, Butterworth-Heinemann, 2000, 594 pp. 13. Smith, C., Practical Cellular and PCS Design. New York, McGraw-Hill, 1998.

References

193

14. CENELEC EN 50067 (April 1998). Specification of the Radio Data System (RDS) for VHF/FM Sound Broadcasting in the Frequency Range from 87.5 to 108.0 MHz. 15. Andersson, R., The Possibilities of Using DARC/SWIFT for Datacasting. Montreux International Radio Symposium, 1998. 16. Walter, F., Digital Video and Audio Broadcasting Technology: A Practical Engineering Guide (Signals and Communication Technology). Springer, 2008, 586 pp. 17. ETSI EN 300 401 v1.3.3 (2001–05). Radio Broadcasting Systems: Digital Audio Broadcasting (DAB) to Mobile, Portable and Fixed Receivers. 18. ETSI EN 300 163 v1.2.1 (1998–03). Television Systems; NICAM 728: Transmission of TwoChannel Digital Sound with Terrestrial Television Systems B, G, H, I, K1 and L. 19. ETSI EN 300 744 v1.4.1 (2001-01). Digital Video Broadcasting (DVB): Framing Structure, Channel Coding and Modulation for Digital Terrestrial Television. 20. ETSI EN 300 392-2 V2.3.2 (2001–03). Terrestrial Trunked Radio (TETRA); Voice Plus Data (V+D); Part 2: Air Interface (AI). 21. Sokolov, A.V., and Andrianov, V.I., Alternative to Cellular Comminication: Trunking Systems (in Russian). Sankt-Petersburg, BHV-Petersburg Publisher. Arlit, 2002. 22. Andrianov, V.I., and Sokolov, A.V., Mobile Communication Facilities (in Russian). SanktPetersburg, BHV-Peterburg Publisher, 2001. 23. Phillips, J.A., Namee Gerard Mac, Personal Wireless Communication with DECT and PWT, Artech House Publishers, 1998, 360 pp. 24. Dinges, S.I., Mobile Communication: DECT Technology (in Russian). Moscow, SOLONPress Publisher, 2003. 25. The Paging Technology Handbook. John Wiley & Sons, 1998, 327 pp. 26. Igumenov, S.A., Broadband Fixed Wireless Access Systems (broadband WLL) (in Russian). TeleMultiMedia, No. 2, 2000. 27. Engels M., and Petrac F., Broadband Fixed Wireless Access/A System Perspective. Springer, 2006, 211 pp. 28. Schiller, J.H., Mobile Communications, 2nd Edition, Addison-Wesley, 2003. 29. Recommendation ITU-R SM.1270. Additional Information for Monitoring Purposes Related to Classification and Designation of Emission.

“This page left intentionally blank.”

Chapter 7

Measurement of Radio Signal Parameters

Introduction Determination of radio signal-modulation parameters and type is an important task, which is solved at radio monitoring stations. This task necessarily contains the frequency and bandwidth measurement for the received signal. Recognition of modulation type and its parameter measurement becomes possible when using customized software for technical analysis of the radio signals. One can distinguish two varieties of such software: the software for manual analysis, in which the operator prescribes arbitrary, in sequence, radio signal transformations, for the purpose of its analysis, and the automatic software, which carries out the necessary signal transformations and its parameter measurements automatically. In this chapter, radio signal frequency and bandwidth measurement methods are considered and, also, application examples of radio signal, technical analysis software are given.

Frequency Measurement There are some different methods for frequency measurement of radio signals received from radio communication channels. All of them use the comparison with the frequency standard. In the beat method, the coincidence between the measuring and reference frequencies is achieved by varying the reference frequency (obtained from the standard source, with the help of the frequency synthesizer), while the coincidence is determined by the difference beating presence at zero frequency. Measurement using the frequency shift is distinguished by the fact that no zero beating occurs, but there are two frequency differences determined in advance. With the Lissaugout approach, frequency closeness is estimated on the basis of ellipse rotation on an oscilloscope screen, and the compared oscillations act at the vertical and horizontal sweep inputs. The frequency meter measures the signal period number during a definite time interval, which is formed also from the reference oscillation. A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_7,

195

196

7

Measurement of Radio Signal Parameters

Frequency discriminators (frequency detectors) determine the signal frequency deflection from some central frequency, which, in turn, is calibrated in accordance with the standard. Frequency measurement can be executed as well with the help of the analog spectrum analyzer, and, in this case, extraction of the spectral component occurs during the standard-sweep oscillator retuning. The instantaneous frequency measurement (IFM) method and the fast Fourier transformation (FFT) method are realized on the basis of digital signal processing. But, in this case, there is a lock-on to the reference standard oscillation as well. In accordance with [1], three named methods are universal. The methods that use digital signal processing have improved accuracy and increased measurement rate at stable results, allowing simultaneous averaging and other statistical operations. These exact methods are the most suitable for the automated hardware-software complexes for radio monitoring.

Instantaneous Frequency Measurement Method The signal’s instantaneous frequency is the changing rate (i.e., the derivative) of its total phase. At small observation intervals, the following expression for the instantaneous frequency is true: f (t) =

(t) − (t − t) 2π t

(7.1)

where (t) is total signal phase, and t is the time interval, usually between the adjacent signal samples. In the sinusoidal signal case, determination of the instantaneous frequency can be reduced to consecutive measurement of the signal level at near discrete time moments (consecutive determination on three samples), and to the solution of the trigonometric equation system: ⎧ ⎨ s(t1 ) = A cos (2π ft1 + ϕ0 ); s(t2 ) = A cos [2π f (t1 + t) + ϕ0 ]; (7.2) ⎩ s(t3 ) = A cos [2π f (t1 + 2t) + ϕ0 ] where t1 ,t2 ,t3 are the moments of sampling; ϕ0 is the initial phase; A is the oscillation amplitude. The instantaneous frequency can now be written as: 1 s(t1 ) + s(t3 ) arccos · f (t2 ) = 2π t 2s(t2 )

(7.3)

This result is strict, for the pure sinusoidal signal only. In the general case, at noise and modulation presence, one should determine the instantaneous frequency of the complex signal s∗ (t) = s(t) + jˆs(t)

(7.4)

Frequency Measurement

197

where s(t) is the initial signal; sˆ(t) =

1 π

.∞ −∞

s(t) t−x dx

is the initial signal transformed

by Gilbert transformation (which is equivalent to the signal passing through the ideal phase-shifter with π/2 shift), and j is the imaginary unit. The instantaneous frequency, in this case, can be written in the form: f (t2 ) =

sˆ(t2 )s(t1 ) − s(t2 )ˆs(t1 ) · 2π t[s2 (t2 ) + sˆ2 (t2 )]

(7.5)

Discrete Gilbert transformation can be fulfilled on the basis of the recursive and non-recursive digital filters, as well as on the basis of signal expansion into Kotelnikov series. In a number of cases, measurement result averaging is necessary for the stable frequency estimation. The total size of the signal sample, by which the frequency determination will be carried out, depends also on the signal type, signal/noise ratio and on the current tasks during measurement (ensuring of maximal accuracy, shortterm transmitter radiation instability measurement, modulation parameter measurement, simultaneous measurements of several characteristics). So, at estimation of the signal’s instantaneous frequency, for the GSM standard of cellular communication, the sample is limited by the transmission duration of 577 μs of the elementary data burst (at synchronization presence). For other communication systems with time access division, the sample duration lies in the limits of 5–10 ms. For ordinary frequency measurements for the usual signals with increased and normal rate, one can recommend the sample duration 200 ms and 1 s, relatively [1]. To increase the frequency measurement accuracy, preliminary signal filtering can be fulfilled in the bandwidth matched to its spectrum width, or even in the narrow band near the carrier or sub-carrier. The IFM method cannot be applied to signals with single side-band modulation (except the specific case of modulation by the sinusoidal signal) and to COFDM signals. In this case, the central frequency estimate can be obtained on FFT basis.

FFT Method Fast Fourier transformation (FFT) is the method for signal spectrum determination on the basis of the discrete Fourier transformation (DFT) calculation, with the help of effective algorithms, in particular, of time or frequency decimation algorithms [2]. The usual discrete Fourier transformation requiring N 2 arithmetic operations is replaced by a recursive procedure with N log2 N operation number, at the expense of repetition exclusion. For FFT algorithm operation, it is required that the length of the initial data series (the count number N of the signal sample) is the multiple of power of 2, although there are other approaches. At the same time, the signal

198

7

Measurement of Radio Signal Parameters

sample length should be large enough to obtain the required spectral resolution df and is equal to N = Fdis /δf

(7.6)

where Fdis is the sample discretization frequency, which, according to the Kotelnikov theorem, should exceed a minimum of two times the maximal frequency of the transformed signal (the value of analyzing bandwidth at the transformation to zero frequency). The detailed spectrum investigation assumes the presence of spectral zoom (ZOOM FFT), on the basis of sample length growth, at keeping the analysis bandwidth and the corresponding Fdis decreasing at the same sample length. The expression for the spectrum obtained on a DFT basis has the following form:

S(mδf ) =

N T s(nδt) exp ( − j2π mn/N) N

(7.7)

n=0

where m is the spectrum component number, n is the element number of the signal sample, δt = 1/Fdis ; and T is time of analysis. To eliminate the edge effects of spectrum distortion due to sample finite length, the multiplication of its samples by the window functions (weighting functions) is used: Hann, Hamming, Natoll, Barlett, Blackman, Blackman-Harris window function, etc. Signal frequency estimation, in this measurement method, is provided on the basis of the maximal spectral component. In the case when FM carrier frequency or FSK mark and space frequencies are determined, and at small signal/noise ratio, as a rule, spectrum averaging is required, i.e., the calculation of the amplitude average value of the spectral components for the several current spectra. For wide-band FM signals, in order to reduce the modulation index, the frequency of the intermediate frequency signal can be divided (up to 200 times), with the following correction on the division ratio. For signals without the carrier (COFDM, LSB/USB), the central frequency value can be calculated as the average of the bandwidth extreme values, which are obtained by the spectrum FFT method. In accordance with Russian standard R 50657-94, the relative frequency deflection of most transmitters is not worse than the units of 10–6 . This standard includes the instruction to fulfill measurement with the error not worse than 0.1 from the permissible frequency deflection. This agrees with the recommendations of [1], in accordance with which the frequency standard for measuring complexes, and all stages of frequency conversion, should have the error of 10–7 . When it is necessary to execute more accurate measurement, under increased demands to transmitters, the possibility should exist to connect the external frequency standard. This could be the high-quality, temperature-controlled, crystal oscillator, the signal from the reference frequency station, the rubidium oscillator, or the calibration from GPS signals is possible, as well. The accessible error could be in the range 10–8 –10–11 .

Frequency Measurement

199

Measurement of Spectrum Width There are two main methods for the estimation of the radio signal spectrum width: power ratio or “β/2”, and the measurement of “X dB” level. In the first case, the spectrum width of the frequency band is defined as the width of the occupied frequency band of radio emission (occupied bandwidth – OBW), beyond which the given part (β%) of the total averaged transmitter power is radiated [1, 3]. Most often, the radiated power beyond the lower and the higher limits are assumed equal (each is a half of the given averaged power β/2%). The value of β/2 is usually selected equal to 0.5%, if there are no additional instructions for the appropriate emission class. In practice, the frequency band in which the direct power determination is provided often contains background interference, which reduces measurement accuracy. It is possible to improve the measurement results, not taking into consideration these spectrum sections where the spectral power density of the signal does not exceed the noise level by Y dB (as a rule, Y = 6 dB), its value is assumed as zero. To estimate the width of the occupied frequency band, one can use digital signalprocessing methods, including FFT. In this case, at first, the total signal power is calculated on the basis of the spectral components, by summation of its amplitude squares, with the following dividing by 2 or by integration of the spectral power density. The lower and higher spectrum frequencies are determined by noise level. After that, in the same manner, the power in the band is calculated, in which the lower spectrum frequency is fixed as a lower limit, and the higher limit is gradually increased until the values obtained will be closed to β/2% from the total power. Further interpolation serves for more accurate estimation of the higher band frequency, in which the given part of the averaged power is concentrated. The same operation should be repeated for the higher spectrum part, and the width of the occupied frequency band will be the difference between the lower boundary of the higher frequency band and higher boundary of the lower frequency band. This method works well, with enough accuracy even for a relatively small signal/noise ratio (with a 15–20 dB excess of peak values over noise), and has small sensitivity to spectral resolution under the presence of more than 100 spectral lines in the signal bandwidth. The method allows for the obtaining of good results, at analysis of signals with any digital modulation. In the last case, it is recommended to use the main lobe (before the first intersection with the noise level) to determine the width of the occupied frequency band, and to select the analysis time as a time of 1,000 symbol transmission [1]. To examine the non-stationary signals (AM, SSB, FM with modulation by the speech signal), one should execute spectral data accumulation (averaging) over a rather long period, to obtain a valid estimation of the spectrum width. To estimate the maximal signal bandwidth for FM broadcasting, we recommend selecting programs with a wide spectrum of the modulating signal, for instance, symphonic music. If the bandwidth of the analyzed signals exceeds the maximal band of simultaneous processing of the hardware-software complex, to estimate the signal spectrum,

200

7

Measurement of Radio Signal Parameters

it is possible to apply “the joining” (lacing) with the further averaging of spectral data, which are consecutively obtained in the adjacent bands, at fast receiver retuning. The emission is optimal, from the point of view of spectrum saving, when the bandwidth of the occupied band is equal to the necessary bandwidth for this emission class, i.e., to such frequency bandwidth sufficient for this emission class, to ensure the message transmission with the necessary rate and quality, under definite conditions (Fig. 7.1). Fig. 7.1 Determination of bandwidth of occupied frequency band

G

Occupied bandwidth

β % 2

β % 2

f1

f2

f

In the case of interference, and at known signal type, the method of bandwidth measurement on “X dB” level is more preferable (excluding several signals with digital modulation). In this case, the area beyond which any discrete spectrum component or the continuous spectral-power density of the measured signal are, at least, by X dB, less than given in advance reference level 0 dB, is considered as the emission frequency bandwidth [1]. This method can be used to measure the width of the selected emission bandwidth, which is understood to be the frequency zone beyond which any spectrum component that is 30 dB less than the level is equal to 0 dB. When using the testing signals defined for this class of emission, for setting zero level in accordance with the regulatory document recommendations, the value of the selected emission bandwidth should not exceed by more than 20% the normalized one at the same level frequency bandwidth. To compare the measurement results of the different radio monitoring stations, the Bureau on Radio Communication of the ITU recommends, as the temporal norm, the measurement at –26 dB level and the correction coefficients γ application for the necessary frequency band estimation, while the specific “X dB” levels can be used for the specific emission types (Fig. 7.2). For some signals, the levels, which provide bandwidth estimates close to the width of the occupied frequency bandwidth at β = 0.5%, are determined empirically. Zero level setting, with respect to which the count is executed, depends on the specific class of the measuring emission. The recommendations on this setting are included in the regulatory documents. The setting can be determined by the level

Determination of Modulation Type and Its Parameter Measurement Fig. 7.2 Frequency bandwidth determination by X dB method

201

G0, dB 0

–X

B

f0

f

of: non-modulated (non-shift-keying) carrier (for example, for FM broadcasting signals, FSK signals, AM telegraph emission); non-modulated sub-carrier (AM photo-telegraph); maximal spectrum component (pulse non-modulated emission); maximal spectrum envelope level in the limits of the side frequency band (for instance, for AM broadcasting signals, single side-band telephony). This method of bandwidth measurement can be implemented by the direct approach (filtering), as well as on the basis of the digital approach (FFT). The main difficulty in examining real radio signals consists in obtaining zero level value. In particular, to find the non-modulated carrier level, we can use the substitution methods (with the help of the additional oscillator), the peak detector method, and the calculation on the basis of the analytical relations. By using FFT, this level can also be obtained for the long spectrum accumulation (maximal current spectral values). So, in the case of FM broadcast signals, when hitting between the separate programs during pauses observations, the level of non-modulated carrier will be factually registered. On the other hand, the long accumulation can fix the moments with audio signal modulation, with maximally wide spectrum and, respectively, the maximal radio signal width.

Determination of Modulation Type and Its Parameter Measurement Determination of Modulation Type Determination of the type and modulation parameters of received radio signals promotes the correct recognition of known radio stations and permits the distant execution of regular maintenance checks of the announced radio emission characteristics. It factually causes all further steps related to the processing of unknown signals.

202

7

Measurement of Radio Signal Parameters

The simplest method for solving this problem is the consecutive searching of the demodulators disposed in the receiving equipment (fixed or software varying). So, the speech for reproducing the analog signals without distortions at the output of AM, SSB, or FM demodulators (scrambling requires additional transformation) will indicate this or that type of modulation, and will allow simultaneously the approximate estimation of modulation parameters. Using the decompaction equipment, it is possible to make the same conclusion about the signals that use frequency or time channel division. The terminal equipment for TV receiving allows the detecting of the TV signal presence. When receiving the digital signals, the determination accuracy of the given class of emission by the unambiguously corresponding combination demodulator/decoder selection is confirmed by the presence of sense information, call signs, or the auxiliary symbol combinations at the device output. In some cases, the experienced operator can estimate the modulation type and the transmission protocol by the typical sounding at the direct signal listening, from the analog signal demodulator outputs that can reduce the number of checked variants. The conclusions on signal type can be made also on the basis of the type of radio signal and demodulated signal spectrum (including the usage of the spectral mask) and its evolution in time, of the signal form and its variations in the time domain, as well as of the measuring parameter values typical for various types of modulation. For example, for analog FM signals, the variations of spectrum width and spectral components amplitude in time and, simultaneously, the amplitude stability in the time domain, are typical. For AM signals, the deviation value measurement will give zero values and the typical amplitude notches, depending on the modulation depth, will be observed on the oscilloscope pattern, and the spectrum component correlated with the carrier will be observed on a spectrum pattern. For digital FSK signals with relatively small transmission rate, the presence of several spectrum maxima is typical. For various digital signals with high keying rate, conversely, the rather wide and stable spectrum is typical. The signals of the base stations of GSM standard cellular communication have a short-term spectral hit on the space frequency related to the synchronization (at time interval transmission for frequency tuning). The measured frequency diversity, the transmission rate, the duration of various time fragments, and the number of binary symbols in a transmitted block may correspond to the standard values, caused by the specific protocols. The character of spectrum modification is the additional indication of modulation recognition at signal transformation. For instance, in the case of phase shift-keying, raising the signal to the power is equivalent to the current phase multiplication by the exponent s2i (t) = A20 cos2 (ω0 t + ϕi ) =

A20 [1 + cos (2ω0 t + ϕi )],0 < t < T. 2

(7.8)

Raising the PSK2 signal (ϕi = 0,π) to the second power will lead to keying elimination in the resulting oscillation, and, at this operation, the PSK4 signal will be transformed to the PSK2 signal, i.e., one keying level will be removed. Relatively, the spectrum of the signal raised to the second power in the first case will

Determination of Modulation Type and Its Parameter Measurement

203

represent one spectral component on the second carrier harmonic, while, in the second case, the spectrum of the signal with the binary phase shift-keying will be located at the same frequency. In order to eliminate shift-keying completely in the PSK4 signal, it is possible to raise it to the fourth power. At that, the pattern in the frequency domain will be the following: at the second carrier harmonic, the PSK2 signal spectrum will be observed; the single spectrum component will be present at the fourth harmonic. When raising the PSK2 signal to the fourth power, two single spectral components at the doubled carrier frequency and at the fourth harmonic will be observed in the spectrum pattern. In a similar manner, the raising of the FSK signal with minimal shift to the second power will yield the usual FSK signal with doubled central frequency and diversity, with respect to the initial values, as a result. To recognize the signal, we can rely upon the value distribution analysis of its instantaneous frequency, phase, and amplitude during some interval, particularly, with the help of histograms. So, for example, for PSK2 signals, the instantaneous phase histograms will have two maxima, while for quadrature phase shift-keying – four maxima, etc. The time distribution of given amplitude (power) level exceeding is the additional statistical feature, depending on the signal types. Signal structure can be determined by the auto-correlation and correlation methods. In the first case, multiplication of the analyzed signal by its time-shifted replicas is executed, while, in the second case, signal multiplication by the reference oscillation set – with this or that modulation type at various modulation parameters – is executed. Auto-correlation is used for signal parameter determination, such as the package duration, the duration of data block. Correlation permits the specific signal to be identified from the available set, in particular, allows the determination of synchro- and pilot sequence presence. But, probably, the clearest reveal of modulation type is based on vector signal representation [4]. The instantaneous amplitude and phase of the signal (7.4) reflection on the phase (complex) plane is understood as vector representation. For this representation, in the polar coordinates, the frequency modulation is presented as the circle while the amplitude modulation – as the ring. The additional shift of the signal central frequency to zero frequency is required for the digital signal analysis. As a result, we obtain the vector diagram of the appropriate modulation type, containing the total history of the information signal vector position, i.e., including the informational signal ensemble (signal constellation) and the transition lines of the information vector from one position to another. The carrier will be represented by a point on this vector diagram, AM signal – by the radius segment, FM signal – by the circle arc. Based on the vector diagrams, for digital modulation, it is possible not only to determine the specific modulation type, but to reveal these or those distortions, interference and malfunctions in the communication channels, for example, non-linearity of transmission section, insufficient bandwidth, the large phase noise, deregulation of the quadrature channels, etc. All of the above-mentioned concerns relate to automatic signal analysis, when the operator is in dialog mode, with the help of a definite set of instruments, and provides the radio emission classification based on his own experience. At that, the

204

7

Measurement of Radio Signal Parameters

main point is the correct visualization of necessary information and the equipment control convenience. Modern computer technology development allows the solution of the completely automatic determination problem for modulation type, not only for the postprocessing, but in real time as well, which permits the receiver to dynamically adapt to the current signal environment. This mode essentially increases the operation performance of radio monitoring services. The mentioned function can be implemented using the mathematical tool of the statistical theory of pattern recognition. Recognition indicates the presence of the object’s definite signs, and the statistical estimation of its parameters under noise presence. The decision to attribute an object to a specific class is executed on the basis of the obtained estimate comparison, for a signal sample with the reference characteristics of the full set of distinguished classes. The selection of informational signal ensemble as modulation signs seems to be the most logical [5]. We can distinguish the situations with completely definite classes, when the conditional integral or differential probability distributions for the samples are known in advance, together with the a priori probabilities of given class presence, and the peculiarities, when these distribution functions are defined by the learning sample. In the last case, the neuron network method application is possible [6, 7]. But, in any approach, the critical rule is reduced to the comparison of this or that functional value with the threshold.

Modulation and the Determination of Shift-Keying Characteristics Modulation depth k (6.1) is the main characteristic of the typical amplitude modulation. It can be calculated by measuring the maximal Um max and minimal Um min envelope value for the modulated signal, and the calculation is as follows (in %): k=

Um max − Um min · 100% · Um max + Um min

(7.9)

On the AM signal-time diagram, the presence of segments, where the envelope has zero value, corresponds to signal overmodulation. Under testing measurements, when sinusoidal modulating oscillation is used, we can estimate the overmodulation factor in the following way. We take the envelope maximal value Um max and envelope value U0 at the time moment corresponding to the modulating signal transition via zero, i.e., at the moments shifted by one quarter of the low-frequency sinusoid period from the points of maxima. After that, the calculation is executed as follows: k+ =

Um max − U0 · 100% · U0

(7.10)

The comparison of the calculated value of the modulation depth factor for the positive half-wave of the modulating oscillation k+ with the same factor for the negative half-wave

Determination of Modulation Type and Its Parameter Measurement

k− =

U0 − Um min · 100% · U0

205

(7.11)

allows the estimation of the modulation asymmetry caused by the modulator nonlinearity. We can also say about the non-linearity on the basis of the distortions of low frequency sine signal after demodulation, in particular, of the cut curve signal peaks or of the harmonic presence at spectral analysis of the detected signal. When in operation with the usual AM signals, the current (for quality check at peak values of the modulating signal) as well as the averaged values during definite time intervals of the amplitude modulation depth value (to estimate the effectiveness of transmitter usage) are used. For analog FM signals, the frequency deviation is the main feature. Its determination in hardware-software complexes can be executed on the basis of the instantaneous frequency measurement of the radio signal. As a rule, it is necessary to know the peak deviation values unambiguously related to the maximal FM radio signal width and regulated by the regulatory documents. The frequency modulation symmetry can be simultaneously checked by the comparison of maximal and minimal values of measured instantaneous frequency with the central frequency of the signal. To decrease the spurious AM influence due to multipath propagation on the measurement process, the FM signal should be restricted. For FM broadcasting signals, it is recommended to select the minimal time for a maximal deviation single measurement equal to 50 ms, which corresponds to two periods of the smallest modulation frequency 40 Hz, according to the CCIR standard. At that, the total observation time should be 15 minutes. The measurement results can be represented as the distribution histogram of the values obtained, or the time distribution corresponding to any exceeding over the given deviation level [1]. We can measure the deviation (frequency diversity) of FSK signals with the help of instantaneous frequency measurement, although sometimes it can be done more simply directly on the spectrum pattern. Frequency deviation (usually the peak values) is one more analyzed parameter of analog angular modulation. It is defined as the difference of the instantaneous phase and the phase of the non-modulated carrier ϕ = ϕ(t) − 2π f0 t = arctg[

sˆ(t) ] − 2π f0 t s(t)

(7.12)

where s(t) is the receiving signal, and sˆ(t) is the signal transformed according to Hilbert. To eliminate the ambiguities, it is necessary to track the phase variation continuity. Exact estimations of the phase deviation can be fulfilled with the help of the phase-locked-loop methods, for instance, using Costas circuit [8]. For the digital modulation types, one of the main characteristics is the keying rate. One can measure the single package duration, which is the variable inverse to

206

7

Measurement of Radio Signal Parameters

the rate looked for, directly at the output of the demodulator for the specific signal type. Due to the noise presence in the communication channel, the single measure can serve as a gross estimation only, therefore the problem should be solved as a statistical one. On the basis of disposed time intervals measurements, we can construct the histogram and, on its basis, make conclusions about the transmission rate. Another approach to determine the keying rate is based on recovering the clock frequency, with the help of the bit synchronizer of the open or closed type. In the first case, the clock sequence is formed directly from the receiving signal, while, in the second case, the receiving signal is used for local oscillator synchronization [9].

SMO-STA Software for the Analysis of Automated Radio Signals STA Software Possibilities and Its Functional Diagram System for technical analysis (STA) software is a component of the software package of customized mathematical software (SMO) for automated radio monitoring systems, and it is intended for the technical analysis of radio signals: determination of modulation type and its parameter measurement [10]. The software provides signal analysis of the radio frequency, detected signals, and signals transmitted on the sub-carrier. The analysis can be executed in real time, as well as in the postprocessing mode, on the basis of the recorded data. The following functions are implemented in the software:

• • • • • • • • • • • • • •

Hardware system control Signal displaying with time and amplitude scaling Signal spectrum displaying with different resolution and frequency scaling Signal displaying in the phase plane Signal band-pass filtering Frequency shift of signals (for more accurate tuning and for signal demodulation on the sub-carrier) Amplitude, frequency, and phase detection of radio signals Amplitude, frequency, and phase detection of radio signals transmitted on the sub-carrier Determination of frequency and time parameters of radio signals Determination of radio signal bandwidth Signal raising to the second and fourth power, to recognize the modulation type Time and amplitude histogram displaying of radio signals, to recognize the digital type of modulation Maximal and minimal deviation diagram displaying, to determine the signal structure Signal demodulation and analysis parameter saving

SMO-STA Software for the Analysis of Automated Radio Signals

207

• Automatic adjustment of demodulation parameters from the saved configuration files (for fast demodulator selection in accordance with the standard protocols of data transmission) • Signal recording on the hard disk. The functional diagram of the technical analysis software is shown in Fig. 7.3. Input signal

Frequency detector

Frequency shift

Amplitude detector

Band-pass filter (BPF)

Phase detector

Raising to power

Relative phase detector

Frequency shift

Band-pass filter (BPF)

Raising to power

Band-pass filter (BPF)

Frequency detector

Amplitude detector

Band-pass filter (BPF)

Displayed signal on sub-carrier

Phase detector Relative phase detector Displayed detected LF signal after frequency shift

Displayed RF signal

Displayed detected LF signal

Fig. 7.3 Functional diagram of the software for system technical analysis

The complex samples of the input high-frequency signals pass to the input of the frequency shift module. To detect the signal, the shift is fulfilled to zero frequency. Then the signal passes to the band-pass filter, the application of which allows the increase of the signal/noise ratio. The filter is implemented as the digital nonrecursive filter of the 127th order with the finite impulse response. The filter has an amplitude-frequency response close to the rectangular one and the linear phasefrequency response. After the band-pass filter, the signal can be raised to the first, second, or fourth power, to recognize the phase shift-keying signals. Then the signal and its spectrum are reflected in the diagrams in RF radio signal analysis mode. To analyze the modulation type, one of the following detectors may be selected: frequency, amplitude, phase or relative phase detectors. After detection, the signal may be additionally processed with the help of the second band-pass filter. The oscilloscope pattern and the spectrum of the detected and filtered signal are reflected in the detected signal analysis mode D1. To analyze the signal on the sub-carrier, the detected signal is transformed into the complex type, shifts to zero frequency, and passes through the band-pass filter. The signal may be additionally raised to the power for phase shift-keying recognition. After such processing, the signal is displayed on the diagrams (in D1 mode at pressed “Frequency shift” button).

208

7

Measurement of Radio Signal Parameters

For demodulation of the signal on the sub-carrier, the frequency, amplitude, phase, or relative phase detectors from the second set, can be selected. The detected signal on the sub-carrier may also be filtered with the third band-pass filter, after which it is presented on the diagrams in sub-carrier analysis mode D2.

Examples of Radio Signal Modulation Type and Parameters’ Determination After loading the control program, the SMO-STA Dispatcher can be launched via the program icon, by clicking on the Dispatcher panel. Upon operation, the first action should be to tune on the frequency of the signal to be analyzed, with the equipment set in real-time mode. This can be done by specific value-setting in the frequency-setting window on the equipment control panel, and by additional adjustment, with the help of the mentioned window arrows. If signal analysis is executed after its detection in other package programs, the tuning is executed automatically, at the transition to STA. Switching on the attenuators and amplifiers can be the next step, depending on the signal level and the antennas used. After that, one should select the suitable analysis bandwidth, depending on the frequency bandwidth occupied by the signal. To analyze the signal in the post-processing mode, it is necessary to record the file with signal-time samples. The recording process begins when the “Record” button, in the signal-recording window, is pressed. The window, in turn, is started by the red button on the equipment control panel, or by the redundant command “Record” in the “Mode” menu. The same window allows the setting of the recording time. Post-processing signal analysis will be executed in the same bandwidth and with the same settings, which were used during the recording. The frequency, which was used for the recording, is indicated in the file name, with a .taf extension. To load the file with signal data, the software has the standard “Open file” operation. File analysis is possible without connection to the equipment. In this case, the program is initiated exactly from its folder. AM Signal Analysis To recognize the typical AM signal, one may use three specific spectral components: the carrier frequency and two reflection-symmetric side bands, with respect to the carrier. To distinguish the case of angular modulation with low index, it is necessary to take the vector diagram into consideration additionally. The mentioned spectral components are distinctly evident in the window of the STA program (Fig. 7.4), upon file analysis of the recorded AM signal. The modulating oscillation is most likely a harmonic one, due to the small side-band width. Moving the marker to the side-band maximum, one can measure the modulation oscillation frequency. The measurement result can be read at the left side of the spectrum output window in the MX line (marker position on the horizontal axis).

SMO-STA Software for the Analysis of Automated Radio Signals

209

Fig. 7.4 STA application window for AM signal analysis at radio frequency

The signal vector diagram is reflected in the additional “Phase plane” window. In the figure, we can see that, at signal recording, there was no exact tuning to its central frequency because the signal vector rotates with some speed, modulo varying (“lobe” vector diagram). To adjust the frequency, one may use the “Frequency shift” button group (the exact adjustment takes place by the additional pressing of the “Ctrl” button). The adjustment result is shown in Fig. 7.5, where the vector diagram already has a form close to the theoretical one for an AM signal. The spectrum and the time diagram of the demodulated signal, with the help of the AM detector, are shown in the same figure. As one would expect, the modulating signal is practically sinusoidal. In the spectrum-pattern window, one can define more exactly the signal’s frequency, having moved the marker to the maximum of the spectrum pattern of the detected signal. The pattern scale is expanded with the help of the “Spectral frequency scale” button group, for convenience. Figure 7.6 shows the STA-program window displaying the signal amplitude histogram. For a more detailed examination, this window is expanded to the full screen, with the help of the “Expand/recover the window” command in the “Utilities” menu. At allocation of the histogram-displaying area, the modulation depth value is calculated automatically, which is read in the upper line (“Alloc. %”) of the informational table, to the left of the window.

210

7

Measurement of Radio Signal Parameters

Fig. 7.5 STA application window at detected signal analysis after additional adjustment

Fig. 7.6 “Amplitude histogram” window for measurement of the amplitude modulation depth

SMO-STA Software for the Analysis of Automated Radio Signals

211

FM Signal Analysis An FM signal can be recognized by the vector diagram and by the frequency deviation presence. Figure 7.7 shows the application window for FM signal file analysis, with the sinusoidal modulating signal. In the spectrum, we can see the typical spectral “comb”: in this case, which components are spaced each from the other by a distance equal to the modulation frequency. This diversity can be measured with the help of a marker. The vector diagram represents the circle that corresponds to FM.

Fig. 7.7 STA application window for FM signal analysis at radio frequency

Figure 7.8 illustrates the measurement of FM signal-frequency deviation using an amplitude histogram at the FM detector output. This procedure can be executed on the “Amplitude histogram” tab, in the mode of detected signal analysis, by means of area allocation from the edge to the edge of the “well”. Deviation value is read in the second line (“Alloc./2, kHz”) of the informational panel, on the left. An additional check of measurement accuracy can be made by moving the marker to the maxima of the detected-signal, in the time diagram. Deviation value is read again to the left in this window in the line MY (marker position on the vertical axis). FFSK Signal Analysis The FFSK signal is the frequency-modulated signal, and the modulating signal is the frequency shift-keying sub-carrier, where the shift-keying is executed without

212

7

Measurement of Radio Signal Parameters

Fig. 7.8 STA application window for frequency deviation measurement by the amplitude histogram of the demodulated FM signal

the phase jump. Figure 7.9 shows the application window for this signal analysis at radio frequency. We can determine the frequency modulation on the basis of the vector diagram, in the form of circle. Figure 7.10 shows the behavior of the deviation peak value of the analyzed signal. This parameter estimation is executed by the superposition of the horizontal marker with the lines on the diagram. The results are read on the informational panel to the left of the oscilloscope-pattern window for the detected signal. The smooth transitions between the mark and space frequencies are well seen in the detailed-pattern window in this figure. In this case, we can determine the sub-carrier frequency by the maximum of the demodulated signal spectrum (Fig. 7.11) and by the mark and space frequencies at the raising of the demodulated signal to the second power (Fig. 7.12). At this operation, the phase ratios are destroyed and, at doubled sub-carrier frequency, the simple FSK signal occurs with a diversity that is twice as large as the initial one. It is possible to find the mark and space frequencies (with somewhat less accuracy), using the frequency-deviation estimation on the sub-carrier (Fig. 7.13). The keying rate on the sub-carrier can be determined with the help of a time histogram of the detected signal on the sub-carrier (Fig. 7.14). The marker is placed to the extreme left of the histogram maximum, and the value is read at the left, in the “1/MX, bps” line.

SMO-STA Software for the Analysis of Automated Radio Signals

213

Fig. 7.9 STA applicaton window for FFSK signal analysis at radio frequency

Fig. 7.10 Determination of FFSK radio signal frequency deviation on the basis of the deviation diagram

214

7

Measurement of Radio Signal Parameters

Fig. 7.11 Determination of the sub-carrier frequency of FFSK radio signal by spectrum maximum of the detected signal

Fig. 7.12 Determination of mark and space frequencies of FFSK signal at raising the detected signal to the second power

SMO-STA Software for the Analysis of Automated Radio Signals

Fig. 7.13 Determination of FFSK signal frequency deviation on the sub-carrier

Fig. 7.14 Determination of FFSK signal-keying rate on the sub-carrier

215

216

7

Measurement of Radio Signal Parameters

PSK Signal Analysis Figure 7.15 shows the window of the technical analysis software used to analyze a PSK signal with the shift. We can see the typical points on the vector diagrams, which are the square angles corresponding to four discrete states of an OQPSK signal and the lines of movement from one state to another. The absence of cross movements testifies that the state varies during the single transition, by π/2 only. Figures 7.16 and 7.17 show the consecutive raising of the analyzing signal to the second and the fourth power, to confirm the presence of phase shift-keying. The variation of the number and location of loops on the vector diagrams unambiguously indicates the nature of the analyzed signal.

Fig. 7.15 STA application window for OQPSK signal analysis at radio frequency

Figure 7.18 provides additional proof of the phase discreteness of the signal carrier, by showing the histogram maximum presence of the phase value distribution for the keyed carrier approximately corresponding to 45, 135, 225, and 315◦ . As for all shift-keyed signals, the keying rate can be determined on the basis of a time diagram of the detected signal (on this occasion, with the help of a phase detector) (Fig. 7.19). In this case, the transmission rate will be twice as much as the value obtained, because two bits of information are transmitted during one interval.

SMO-STA Software for the Analysis of Automated Radio Signals

Fig. 7.16 Elimination of one shift-keying level at OQPSK signal-raising to the second power

Fig. 7.17 Elimination of shift-keying at OQPSK signal-raising to the fourth power

217

218

7

Measurement of Radio Signal Parameters

Fig. 7.18 Histogram of OQPSK signal phase distribution

Fig. 7.19 Determination of shift-keying rate of PSK signal

Automated Technical Analysis of Radio Signals

219

Automated Technical Analysis of Radio Signals Unit for Automated Radio Signal Analysis At present, all modern hardware-software systems for radio monitoring have the possibility to provide automated determination of modulation type and its parameter measurement. This function is implemented in the panoramic analysis (PA) software, which is part of the SMO software, on the basis of expert system technology of the interpretive type. In the most general type, this expert system can be represented as follows (Fig. 7.20). Fig. 7.20 Simplified scheme of decision-making about the signal type

Frame system

Object (radio signal)

Logical deduction device

Operator

Knowledge base

The frame system is a well-ordered structure of modules – descriptions of the recognizing signals with separate slots in which the specific values of these signals’ essential characteristics are saved. The recognition reduces to the comparison of frame-slot values with the characteristics of the object (radio signal) observed and following logical analysis. To accelerate the procedure, frames are ordered in the form of a tree. At that, the recognition goes from the general signs of signal class to the specific ones. The process is considered to be complete either at the boundary achievement (i.e., successful comparison of the signal characteristics with the records of the last frame-slots at the specific tree branch), or at the absence of such coincidence at the lower hierarchical level. One can essentially increase the operating speed and signal-type determination accuracy at the expense of narrowing the set and the key parameter values range, which can be saved in the knowledge-base of the specific radio sources. At present, the expert system allows the recognition of a wide nomenclature of analog as well as digital signals. On the carrier frequency, the system recognized the following modulation types: • • • • •

Simple carrier, amplitude modulation (AM) TV signal (brightness channel) Amplitude shift-keying (ASK), double side-band modulation (DSB) Angular modulation, frequency modulation (FM) Frequency shift-keying (FSK), binary phase shift-keying (BPSK)

220

7

Measurement of Radio Signal Parameters

• Quadrature modulation (QM), quadrature phase shift-keying (QPSK), differential quadrature phase shift-keying with the shift divisible by π/4 (π/4DQPSK), minimal shift-keying (MSK). On the sub-carrier, the system recognizes the following: • • • •

Tone OIRT and CCIR stereo radio broadcasting standards Fast frequency shift-keying (FFSK) Quadrature phase shift-keying with the shift (OQPSK).

In addition to recognition, some signal parameter measurement is executed: carrier frequency, sub-carrier frequency, bandwidth, or modulation parameters (amplitude modulation factor, frequency deviation, angular modulation index, and data transmission rate for the digital signals). The implemented expert system is open (and this is especially important) because it allows the addition of new frames rather simply (new signal descriptions) without changing its structure and maintaining the succession when extending the possibilities. The system functioning we shall review is based on the example of binary phase shift-keying. The analyzing signal passes from receiver IF output to the analog-digital converter (Fig. 7.21). The real digital data obtained (Fig. 7.22a) are transformed into complex form, with simultaneous filtering in the band defined by the operator, and with the following conversion into zero frequency (Fig. 7.22b). The carrier frequency and the initial signal-phase estimation are carried out by means of the spectral line analysis method [9]. The values obtained serve for the phase and frequency synchronization (Fig. 7.22c). After that, clock synchronization is executed for the following: gating, determination of the discrete transmission presence in the signal, and measurement of the information transmission rate. In order to locate the gate-pulse in the middle of the separate informational symbol, the clock frequency

Determination of period and initial delay

A Output data from analog-digital converter

Conversion and filtering

B

Discrete time

C

Complex data

Gating

X exp − ( j ω t + ϕ )

Measurement of frequency and initial phase. Raising to the second power

Fig. 7.21 Structural scheme of BPSK digital signal analysis

D Signal diagram analysis

N

Automated Technical Analysis of Radio Signals

A

221

Re t

f 0

Fdiskr /4

Fdiskr /2

B

Im f

0

C

Im

t

D Re

Fdiskr /4

Fig. 7.22 Frequency and time diagrams in the different scheme points for BPSK signal recognition

period and the initial delay of the pulse sequence are measured. After gating, the signal vector diagram is examined (Fig. 7.22d). In the case of a BPSK signal (binary phase shift-keying), the frame has the form shown in Table 7.1. The first field contains the name shown to the user. Discreteness and data vectorposition number on the signal diagram are the essential properties. During the comparison process, they play a main part. The keying rate is the data parameter. The advantages of such a system are that it has no strictly-definite processing method and, for each modulation type, the operation sequence can be changed. Table 7.1 Example of signal frame at automatic recognition

Frame

BPSK

Time discreteness Vector position number on the diagram Keying rate

Yes 2 2,400

Peculiarities of SMO-PA Application Automatic technical analysis of radio signals is available in the universal software for panoramic analysis (PA), in two modes: “Measurement” and “Review”. In both cases, the single data sample of the digitized-on intermediate frequency, which is obtained from the ADC, is used for the analysis. But, at that, in “Measurement” mode, the operation is executed in the real-time mode: data are permanently renewed, while, in “Review” mode, the digital data, saved in the scanning results base, are analyzed. Operation in “Measurement” Mode To fulfill the analysis in the “Measurement” mode, one should execute the following actions:

222

7

Measurement of Radio Signal Parameters

• Tune on the analyzed signal • Select the spectral resolution (lens) corresponding to the bandwidth of the analyzed emission signal • Set the analysis bandwidth to match the signal bandwidth • Press the button initiating the technical analysis start. The adjustment is accomplished by setting the central frequency value in the window for editing the tuning frequency, located at the tool panel of the main application window, or, visually, on the basis of the signal spectrum. The frequency can be set from the computer keyboard or by the mouse, with the help of the virtual adjustment arrows on the right of the tuning frequency window. The procedure is finished by pressing the “Enter” button. At the visual setting, the main marker is moving directly to the maximum, or to the center, of the signal spectrum, by means of the keyboard or the mouse, and, after this, the “Enter” button is pressed. At the visual setting, it is recommended to switch-on the spectrum pattern, aggregated as maximal during the observation time. At operation, this will help to avoid the current pattern asymmetry influence, which occurs with some signals, in particular, with wide-band FM. The spectral resolution (lens) is selected by the “+” and “–” buttons on the toolbar of the “Measurement” window. The spectrum resolution parameters may differ, for different types of equipment. The best analysis results are obtained when the signal spectrum fills the program spectral window as much as possible – not exceeding its limits. Figures 7.23–7.25 show some examples of automatic determination of modulation type and its parameter measurement, including for the radio-broadcasting FM signal, the signal with frequency-shift-keying, and the TV image signal. Setting the exact bandwidth is performed via a second group of markers. They are activated by pressing the appropriate button at the tool bar of the “Measurement” window (the button is doubled by the pop-up menu command) or the Space button. At selection of this additional pair, the mouse pointer takes the view of a vertical line with the central thickness, and its movement dislocates the marker intersection point. To position the marker on the frequency via the keyboard, we may use the Left and Right buttons (for acceleration together with the Shift button) on the main keyboard, and, at that, the mode window should be active. When moving the additional vertical marker, the measurement zone is allocated, which is symmetric with respect to the main vertical marker. This zone should be slightly more than the spectrum bandwidth of the analyzed signal. The marker positions are indicated on the information panel. The automatic radio signal-analysis process is initiated by pressing the “Measurement” button. The analysis results appear on the information panel. The following data is indicated: the modulation type, the central frequency value, the signal bandwidth, the frequency deviation and modulation index (for angular modulation), the modulation factor (for amplitude modulation), modulation rate in baud (for digital modulation types). At sub-carrier presence, the following is indicated: the modulation type on the sub-carrier, sub-carrier frequency, and sub-carrier modulation parameters.

Automated Technical Analysis of Radio Signals

223

Fig. 7.23 “Measurement” mode window at the 2 MHz bandwidth analysis of FM broadcasting signal in stereo mode on the basis of the CCIR European Standard (with Pilot-Tone)

Fig. 7.24 “Measurement” mode window at the 25 kHz bandwidth analysis of FSK signal with 850 Hz deviation and 100 baud transmission rate

224

7

Measurement of Radio Signal Parameters

Fig. 7.25 Pop-Up Menu of the information panel for fixing the modulation type at TV 10th channel signal analysis

Once the rather short digital sample is analyzed, the recognition results have the probabilistic character, especially under the evident noise presence. Therefore, the additional lines “Modulation type” and “Sub-carrier, modulation type” on the information panel have the arrow on the right to call the pop-up menu, which has the list of all modulation types determined during this recognition, with the statistical percent distribution. With the help of this menu, it is possible to fix the specific modulation type from the selected types, or, instead of current, to select the pattern of the most-probable modulation type of the carrier and sub-carrier. Information about each modulation type is accumulated separately, and, in further analysis, at fixing the specific modulation type, the measurement data will pass to output for this modulation type only. The additional lines with the signal parameter data have similar arrows to call a menu, in which it is possible to select the average or maximal value, instead of the current value, on the basis of the single measurement set for each characteristic separately or for all simultaneously (Fig. 7.26). The number of single measurements, which are used for the statistical processing, may be changed. This can be done on the “Measurement” page of the “Setting” window of the SMO-PA program, where there is a special window in which this value is set in the range from 1 to 10,000 (Fig. 7.27). Examples of the automatic recognition results, achieved with the help of the SMO-PA program, are presented for the following signal series (Figs. 7.28–7.37):

Automated Technical Analysis of Radio Signals

225

Fig. 7.26 Pop-up menu for selecting the pattern of average or maximal parameter value on the basis of the measurement set during signal analysis of the control channel for the AMPS cellular communication standard

Fig. 7.27 “Measurement” page of the setting window

226

7

Measurement of Radio Signal Parameters

Fig. 7.28 Results of signal analysis without modulation

Fig. 7.29 Results of automatic analysis of AM radio signal with 50% modulation depth and 15 kHz modulation frequency

Automated Technical Analysis of Radio Signals

227

Fig. 7.30 Analysis of FM signal with 10 kHz deviation and 1 kHz modulation frequency

Fig. 7.31 Analysis of FSK signal of the FLEX-3200 paging standard (transmission rate is 3200 baud, frequency deviation is 4.8 kHz)

228

7

Measurement of Radio Signal Parameters

Fig. 7.32 Recognition of the POCSAG paging standard with FFSK on the sub-carrier 1,500 Hz, frequency deviation is 300 Hz at sub-carrier, rate is 1.2 kbaud, deviation is 4 kHz

Fig. 7.33 Recognition of GFSK signal standard of CT3 cordless phone with parameters: modulation rate of 640 kbaud, frequency deviation is 160 kHz

Automated Technical Analysis of Radio Signals

229

Fig. 7.34 Results of GMSK signal analysis of GSM cellular communication standard with the transmission rate of 270.833 kbaud

Fig. 7.35 Results of GMSK signal analysis with modulation rate of 1 Mbaud

230

7

Measurement of Radio Signal Parameters

Fig. 7.36 Results of QPSK radio signal recognition with modulation rate of 24 kbaud

Fig. 7.37 Results of π/4DQPSK radio signal recognition for the TETRA trunking communication standard (transmission rate is 36 kbit/s)

Automated Technical Analysis of Radio Signals

231

carrier signal, AM, FM, FSK, and PSK. Radio signals were sent from the RS SME 03 generator at 407 MHz frequency. Operation in “Survey” Mode To perform analysis in “Survey” mode, it is necessary to provide the following: • • • • •

Record the radio signal scanning frequency into the frequency table Select the technical analysis bandwidth (spectral resolution) Set the measurement bandwidth corresponding to the signal spectrum width Select the record of short signal sample in controller responses Select the response saving option in the database

• • • •

Initiate the scanning process during several cycles Stop the scanning process Launch the automatic analysis Select the necessary record in the result table.

Radio signal frequencies are recorded into the scanning frequency table, manually or automatically, on the basis of the active radio-channel detection results in the “Search” sub-mode of the “Spectrum” mode. In the first case, one needs (having selected the frequency table) to press the <>+<> button combination or to initiate the “New frequency” command from the context menu, called by pressing the right mouse button. After that, it is necessary to set the frequency in MHz and press “ENTER”. When several frequencies are present, the operation should be repeated. In the second case, the “Import of founded frequencies” button is pressed and, after that, the founded-sources frequencies for the “Search” table are transmitted into the scanning frequency table. At that, the signal spectrum width is registered automatically. Spectral resolution is set either with the help of the window located near the frequency table of the “Survey” mode, or in the editing window, called by the “Edit allocated” context-menu command. In the editing window, it is possible to select the spectral resolution for the frequency group (Fig. 7.38). In a similar manner, the necessary measurement bandwidth can be set or corrected. To select the controller response, one needs to make a check mark in the “Controller response” field, near the pictogram of the necessary response. The response “Short signal sample” pictogram is a boxed sinusoid of green color. The responses are saved to the database by pressing the “Save response” button on the tool bar of the “Survey” window. By default, this button is always pressed (Fig. 7.39). Scanning is initiated by the “Start” button, or by the similar menu command. At operation, the scanning results table begins to fill. In the last column of this table, the saved controller responses are indicated by its pictograms. “Survey” mode stops by the “Stop” command.

232

7

Measurement of Radio Signal Parameters

Fig. 7.38 “Survey” mode window with the window of grouped parameter editing

Fig. 7.39 “Survey” mode window with activated automatic analysis of the recorded-in-database stereo signal sample of FM broadcasting signal on QIRT standard

Application of Automatic Signal Analysis in SMO-RD2

233

To launch the automatic technical analysis on the recorded data, it is necessary to press the button with the boxed sinusoid pictogram on the tool bar. The necessary record selection in the scanning results table is executed by pressing the right mouse button at the appropriate line, by browsing the lines via the Up and «Down» buttons of the keyboard. At necessary record selection, an information panel appears in the program window. This panel is similar to the above-mentioned information panel, but it contains additional lines of data regarding the automatic recognition of the radio signal. Since one line contains a single response only, this panel has no statistical options. It follows from the above, that it is profitable to use automatic technical analysis in “Measurement” mode for the detailed selective investigation of separate and new signals, when studying a specific radio environment. Automatic analysis in “Survey” mode is more convenient for group radio equipment monitoring, for instance, for the permanent monitoring of broadcasting stations.

Application of Automatic Signal Analysis in SMO-RD2 The SMO-RD2 program is part of the ARK-RD2 hardware-software system, which is intended for automatic 24-hours radio monitoring of SW radio lines, including the usage of RTTY protocol for data transmission. An overall view of the program interface is shown in Fig. 7.40.

Fig. 7.40 SMO-RD2 program interface

234

7

Measurement of Radio Signal Parameters

One of the system’s operation modes is the selective monitoring of the transmitted information at allocated frequencies, with recording of the program. This can be realized by scanning, in the task mode. Since the operation is executed uninterruptedly, the quantity of registered data is a serious problem. The module for automatic recognition of the transmission type is used for activating the continuousrecording command only after the detection of a signal with the definite modulation type – in particular, frequency shift-keying. At the same time, the signal parameters (mark and space frequencies, frequency diversity) are measured and recorded to the database. Additionally, these parameters’ measured values are used for further automatic decoding of the messages contained in the recorded files. At that, the program additionally determines the transmission rate and RTTY protocol parameters: duration of stop message, frequency inversion presence. Only one operator action may set the initial coding table (for Cyrillic and Latin alphabet, letter or digit register). Figures 7.41 and 7.42 show the decoding window of the SMO-RD2 program with the decoding results for the information transmitted by RTTY protocol. In the first case, the type of signal spectrum pattern is selected, with decoding of the call sign and station frequencies. In the second case, the detected signal is displayed without the additional restrictions, and the meteorological data are decoded. On the time

Fig. 7.41 Decoding window of SMO-RD2 program with signal spectrum pattern

Conclusion

235

Fig. 7.42 Decoding window with detected signal pattern without clipping

diagram, the section of symbol transmission is allocated and displayed, which is indicated by a cursor in the text window.

Conclusion The concepts of measurement methods for the main characteristics of radio signals are given: frequency and the bandwidth of occupied frequencies. The extant approaches to the recognition of modulation type and to the determination of its parameters are discussed. A description of the functions and structure of the unified program for technical analysis included in the SMOS-STA customized mathematical software package is presented. The peculiarities of the practical application of this program for specific signals: AM, FM, FSK on the sub-carrier (FFSK) and PSK, are shown. The implementation of an automatic signal-recognition module in the SMO-PA program for panoramic spectral analysis is considered. Recommendations on its application in different modes are discussed. Examples of various modulation-type determinations are given. In the last section, the application of automatic radio-signals recognition in the SMO-RD2 program is shown, which is intended for parameters’ measurements and for signals decoding at monitoring stations, for SW radio lines with RTTY protocol.

236

7

Measurement of Radio Signal Parameters

References 1. Spectrum Monitoring Handbook, ITU-R, Geneva, 2002. 2. Rabiner, L.R., and Gold, B., Theory and Applications of Digital Signal Processing. New Jersey, Prentice Hall, 1975. 3. Recommendation ITU-R SM.328-10. Spectra and Bandwidth of Emissions. 4. Agilent Vector Signal Analysis Basics. Application Note 150-15. 5989-1121EN. 5. Mobasseri, B.G., Digital Modulation Classification Using Constellation Shape, Signal Process., vol. 80, 2000, pp. 251–277. 6. Kuznetsov, A.V., Neural-Network Algorithm of Telecommunication Signal Class Division (in Russian). Informatsionnye Tekhnologii, No. 7, 1999. 7. Wong, M.L.D., and Nandi, A.K., Automatic Digital Modulation Recognition Using Artificial Neural Network and Genetic Algorithm, Signal Process., Vol. 84, No. 2, February 2004, pp. 351–365. 8. Proakis, J.G., Digital Communications, 4th Edition, Mc Graw-Hill, 2001. 9. Sklar, B., Digital Communications: Fundamentals and Applications, 2nd Edition, PrenticeHall, 2001, ISBN 0-13-084788-7. 10. Beliakov, A.L., Bykovnikov, V.V., and Trembachov, A.V., Hardware and Software for Automated Technical Analysis of Radio Signals (in Russian). Special technologies. 2003. Special Edition, pp. 20–26.

Chapter 8

Direction Finding of Radio Emission Sources

Introduction A radio direction finder measures a radio wave’s angle-of-arrival and allows the determination of its source direction. A bearing is the angle between the direction to the radio emission source (point 0 in Fig. 8.1) from the direction-finding point (point A in Fig. 8.1) and some initial direction. A bearing counted out clockwise from the North direction of the geographical meridian is referred to as true. If the longitudinal axis of the transport object is considered as the initial direction, the bearing will be relative (on-board). Traditionally, radio direction finders are classified as goniometric radio navigation systems. The main task of a navigation system is the position-finding of a mobile object, for instance, a ship or airplane in the Earth’s coordinate system. If the coordinates of two or more radio transmitters are known, it is possible to determine the own-object’s coordinates, having found its direction. With the help of a single direction finder, it is possible to determine bearing only and, hence, the azimuth on the radio emission source. To determine the source position, one needs, at least, two radio direction finders, which are distant from one another by a large space. The position of the unknown radio emission source is defined by the bearing interception point. The bearings can be discovered simultaneously or consecutively. Since the Earth’s surface is not flat, the bearing lines can be represented as straight lines for relatively small distances only. Under middle latitude conditions, these distances correspond approximately to the straight visibility. At large distances, the bearing lines are mapped in the form of geodesic lines (great circle), joining the given points by the shortest way along the Earth’s surface. The geodesic line view depends on the used cartographic projection. At present, direction finding for radio navigation purposes is loosing its significance, due to satellite navigation system distribution. At the same time, the need for determining radio emission sources is as relevant as before, in various important areas, including:

A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_8,

237

238

8 Direction Finding of Radio Emission Sources

Fig. 8.1 True and relative radio bearings Radio emission source

A North

True bearing

Relative bearing

Transport carrier axis

0

• Radio monitoring problems, with the purpose of revealing radio signals and interference source positions • Problems related to terrorism • Military applications • Modern radio communication systems with multiple access • Scientific investigations, for instance, in radio astronomy.

History of Radio Direction-Finding Technique One may consider that the radio direction-finding technique began in 1888 when Heinrich Hertz discovered the properties of a directional antenna, executing his experiments in the decimeter radio waves range [1]. Rotating electrical or magnetic dipoles, which were oriented in accordance with the direction of the electrical or magnetic field components, were the first direction finders. The direction finder with the rotating frame is the most well known direction finder of such type. The usage of direction finders with rotating frame antennas began widely during the First World War. These finders made a good showing in the low-frequency range from 0.2 to 1.5 MHz. At a large signal/noise ratio, they allow measurement within one degree of accuracy. But in the hectometer wave range (3–30 MHz), direction finders with frame antennas provide large errors, especially at night. These errors relate to the peculiarities of short wave propagation, which reflect from the ionosphere and change the polarization angle. The error presence in framed direction finders, first of all, can be explained by the electromagnetic field receiving on the horizontal parts of the frame. Therefore, at the next stage of radio direction-finding development, users began to apply spaced, spike antennas. In 1917, Frank Adcock, a researcher from the U.K., noticed that, with the help of vertical spike antennas spaced not more than a tenth of a wavelength apart, one can ensure a directional diagram (pattern) similar to the frame antennas, but with no need to find out any additional, horizontally-polarized, field components. From approximately 1931, the wide usage of Adcock’s antennas

History of Radio Direction-Finding Technique

239

began, in practice. These antennas can be considered as those having a small base, because the geometric distance between the spike antennas is significantly less than a wavelength. Various modifications to direction finders with the Adcock antenna system were proposed. These modifications were aimed at decreasing signal-receiving on horizontal antenna feeders and reducing the antenna effect. In 1925–1926, Robert Watson-Watt suggested the use of two fixed-frames in combination with a non-directional spike, instead of the direction finders with mechanical antenna rotation, and the use of the electron-beam tube for bearing display. Beginning from 1943, Britain naval ships were equipped with three-channel Watson-Watt direction finders of SW range, which were successfully used for detection of German submarines. Beginning from 1931, concealed direction finders were used in cars, and handheld direction finders also appear. Systems realizing the Watson-Watt and Adcock methods are often considered amplitude-phase systems. Synchronous measurement of the being-found signal amplitude, with the help of three antennas, is used in these systems. Two antennas have the direction pattern in the form of a “figure-eight,” and they are located on orthogonal bases; the third antenna is non-directional. In the classical WatsonWatt method, a three-channel receiver is used with a combined local oscillator, and the angle of the arriving signal is displayed in the form of sinus and cosine of the third non-directional channel. In the Watson-Watt method, frame antennas are used as the antennas, with the direction pattern in the form of a “figure-eight.” In Adcock systems, a pair of phased, spike antennas is used for this purpose. The advantages of Watson-Watt and Adcock direction finders are the small response time, high accuracy and sensitivity. Among the shortcomings are: relatively narrow frequency range, poor stability to interference, due to multi-path receiving, the restricted possibility of several non-coherent signals distinguishing with different azimuth in a single frequency channel. During the Second World War, direction finders with spaced frame antennas were manufactured in the U.K. and the U.S., and were used for direction finding in hectometer (short) waves. The need to develop such radio direction finders was related to the fact that direction finders with the Adcock antenna system provided large errors at source distances of 50–350 km, due to the sharp decrease of signal level upon increasing the incident angle of radio waves. For the direction finding of sharply-incident waves, it is important that the frame has no directional properties in the vertical plane. Their operation in the presence of horizontal polarization components is possible, due to the frame connections being opposite one another. In 1943, the wide-aperture Wullenweber direction finders appeared. Developed by Telefunken, a German company, this antenna is known as the antenna with a large base (compared to wavelength), and it represents a complicated engineering structure consisting of 40 vertically-located wide-band antenna systems positioned on a circle of 120 m radius. A metal grid was mounted behind the antennas on the internal circle, to ensure the receiving of radio waves arriving from outside. From each antenna, feeders with 75 wave impedance were run under the ground to the

240

8 Direction Finding of Radio Emission Sources

operator room, located in the center. This radio direction finder was intended for operation in the frequency range from 6 to 20 MHz. Amplitude systems using the Wullenweber method are based on the electronic scanning of antenna array patterns. In this method, signals from several antenna elements pass through phase-shifters and are summed to obtain the optimal pattern in the definite sector. Therefore, to obtain the optimal pattern in the limits of the whole circle, a large quantity of antenna elements and phase-shifters is required, together with the complicated switching system. Besides its complexity, the shortcoming of these systems is the relatively narrow frequency range (covering factor, as a rule, is not more than 10). The advantages are high sensitivity and accuracy, good stability to interference caused by multi-path propagation, and the possibility to distinguish several non-coherent signals with different azimuth in a single frequency channel. The first SW direction finder based on the Doppler effect was developed in 1941. Fast radar perfection increased the need for direction finding in higher frequency ranges: Doppler direction finders with the frequency of 3 GHz appeared in 1943. Beginning in the 1950s, airports all over the world were equipped with Doppler direction-finding systems, to monitor the air environment. Doppler and pseudo-Doppler systems are phase systems. Typical implementation of the Doppler direction-finding method involves a single non-directional antenna, which rotates with constant angular speed around the axis perpendicular to the radio wave propagation plane. Antenna rotation modulates the received signal by frequency. At that, the maximal Doppler shift is detected at that point in the circle where the direction of radio wave arrival coincides tangentially with the antenna rotation circle. The signal from the antenna passes to the radio receiver input intended for FM signal receiving. The frequency detector output passes to the phase detector, whose reference input is connected to the sinusoidal source used for the antenna system drive. Under appropriate phasing of the reference signal, the direction to the emission source corresponds to the voltage maximum, at phase detector output. Pseudo-Doppler systems imitate the rotation of the single antenna of the Doppler direction finder, by the fast-switching of the antennas located as a circular antenna array. The advantages of Doppler (pseudo-Doppler) radio direction finders are the wide frequency range (covering factor is more than 10), high sensitivity and accuracy, and immunity to multi-path propagation interference. The main shortcomings are: impossibility of signal distinguishing in one frequency channel; proper modulation influence of finding signal on the bearing accuracy, upon the direction finding of signals with angular modulation; restriction of bandwidth of the being-found signal by the operation zone of the frequency or phase discriminator. During the 1960s and 1970s, the concept of direction finding obtained further development in the papers of I.S. Kukes, V.V. Shirkov, E.Ya. Schegolev, O.V. Belavin, V.A. Ventsel, V.S. Ulianov, S.E. Falkovich, V.K. Mezin, L.S. Gutkin, D.R. Rods, and other Russian and foreign researchers. In the early 1970s, digital methods and devices were implemented into radio direction-finding systems. In the 1980s and 1990s, digital signal processing (DSP) began to be used for direction finding more often.

Structural Diagram and Characteristics of Radio Direction Finders

241

The necessity for the detection and direction finding of signals with expanded frequency spectrum, for example, signals with adaptive and programmable frequency tuning, prompted the appearance of a wide-band direction finder, in which the processes of searching and direction finding could be combined on a DSP basis. DSP application allows the realization of methods, well known from spectral analysis theory, and makes it possible to determine the parameters of incident wave, on the basis of processing the signals from the similar antennas (field sensors). DSR allows the realization of the methods based on beam pattern formation, for instance, correlation interferometric meters (CIM), the usage of adaptive antenna arrays and algorithms with increased resolving capacity for direction finding, for example, the MUSIC (Multiple Signal Classification) algorithm.

Structural Diagram and Characteristics of Radio Direction Finders A typical structural diagram of a modern radio direction finder is shown in Fig. 8.2. A radio direction finder consists of the following main parts: antenna system, radio receiver, device for digital processing, and device for displaying direction-finding results. Depending on the present requirements, additional units can be added to the structural diagram, for instance, the navigation system for determining the proper location and for direction finder orientation, the remote control modules through cable lines or radio channel, the efficiency testing modules, devices for radio receiver calibration, etc. An antenna system consists of antenna elements (AE), located in space according to some law, for instance, over the circle, and contains the necessary quantity of AE. The frame antennas, the conical and bi-conical vibrators, the spike antennas, the discoid-conical antennas, directional antennas of wave channel or log-periodic type can be used as the antenna elements. At present, it is considered that, for high direction-finding accuracy (not worse than 1◦ ) and for a wide frequency range (for example, from 1 to 30 MHz or from 20 to 1,300 MHz), it is enough to have a number of antenna elements N = 9. In principle, the antenna element quantity may be larger. So, phase radio direction finders with the quantity of antenna elements N = 17

1

2

N

Radio receiver unit 1 2

Antenna system

Digital signal processing unit

M

Fig. 8.2 Typical structural diagram of direction finder

Indication unit

242

8 Direction Finding of Radio Emission Sources

are manufactured, while the Wullenweber direction finder consisted of 40 antenna elements. The radio receiving section is intended for the selection, amplification and frequency conversion of input signals. In so-called “mono-pulse” direction finders, the number M of receiving sections is equal to the number of antenna elements: M = N. In this case, the maximal direction-finding rate is provided. However, the technical implementation of such a solution is rather complicated. Usually, signal conversion in the receiving sections should be provided with similar phases and amplitudes, thus, it is necessary to use the common frequency synthesizer. In mono-pulse direction finders, before the bearing-finding operation, the receiving sections should be calibrated with the help of a specific testing generator. As a rule, in radio direction-finding technique, the fact that the radio receiver has a restricted number of receiving channels, usually from one to three channels: M = 1−3, can often be limiting. If the number of antenna elements is more than the number of receiving channels, the antenna elements are connected to the receiving sections consecutively, with the help of high-frequency switchers. From the output of the radio receiver unit, the analog signals on the intermediate frequency pass to the digital processing unit, where they are exposed to analogdigital conversion; and, after that, in accordance with the direction-finding method used, the radio emission source azimuth value is determined on the basis of the obtained samples. In addition to the bearing determination, the DSP system often provides the spectral analysis of signals, digital demodulation, or decoding. The indication unit is intended for display of the direction finder’s operation results, in a form suitable for the operator. Sometimes, this unit represents the fullvalue personal computer (PC). Besides displaying the direction-finding results and spectral and technical analysis, the PC is used to control the equipment operation, to save the databases with the direction-finding results and with the signals for the technical analysis, etc., and to generate the reports and operation protocols. The digital signal-processing application allows the avoidance of some shortcomings peculiar to analog direction finders. Digital processing provides the synchronization of receiving channels, correction of the phase and amplitude values for antennas, cables, etc. In the digital section, the temperature drift is absent; the bearing is available in digital form, which, in particular, simplifies the further calculations and data transmission.

Main Technical Parameters of Radio Direction Finders The most important quality indices of radio direction finders are the following: • • • •

Accuracy of direction finding Sensitivity Noise immunity Operation rate

Main Technical Parameters of Radio Direction Finders

• • • • • • •

243

Resolution Operating frequency range Type of being-found signal Deployment time Weight and dimensions Complexity in manufacturing and operation Cost.

Accuracy of Direction Finding Radio direction finder accuracy is defined by the angular error of bearing. Usually, radio direction finder accuracy can be characterized by the root-mean-square (RMS) error of bearing, determined as: ! " N−1 "1 (θi − θˆi )2 δ=# N

(8.1)

i=0

where θi is the true azimuth, θˆi is the azimuth measured by the direction finder, and N is the number of measurements. At determination of direction finder accuracy, a large number N of measurements are made, changing RES location azimuth and the emission frequency. As an example, Fig. 8.3 shows the measured bearings, depending on the frequency and azimuth of the testing generator location, for the ARTIKUL-P direction finder. The direction finder is mounted on a mast with 10 m height, in the center of an open, flat, area with sizes 500 × 800 m. The area was preliminarily marked by “marker pegs” over the circle – with the help of the azimuth circle – through 10◦ . The test generator was consecutively transferred from one marker peg to another. In each position, the generator frequency was changed from 40 to 1,000 MHz. As we see, in the lower part of the measurement range, approximately before 100 MHz, some degradation of direction-finding accuracy is observed. We can assume that accuracy degradation at low frequencies is caused by Earth surface influence, since the mast height is commensurable with the wavelength. Moreover, the accuracy of the correlation interferometer in the lower part of the operating range is decreased, due to the fact that the correlation curve demonstrating the relation between the theoretical and practical field distribution on the antenna array elements has a less sharp maximum (see Fig. 8.31). Figure 8.4 shows the root-mean-square error of direction finding versus a frequency, calculated by the formula (8.1) for the values of measured azimuth shown in Fig. 8.3. From Fig. 8.4, we can see that, in the low part of the measurement range (approximately before 100 MHz) the RMS direction-finding error decreases from 2.5 to 1◦ . For frequencies over 100 MHz, the error, in fact, does not exceed 1◦ . The direction-finding process, like all measurements under interference conditions, is accompanied by random and constant errors. Random direction-finding

244

8 Direction Finding of Radio Emission Sources

θ, degree

0

360

10 20 30

330

40 50

300

60 70 80

270

90 100

240

110 120 130

210

140 150 160

180

170 180

150

190 200 210

120

220 230

90

240 250 260

60

270 280 290

30

300 310

0

320

0

100

200

300

400

500

600

700

800

900

1000

330 340

f, MHz

350

Fig. 8.3 Measured azimuth versus frequency for ARTIKUL-P radio direction finder. The location of test generator was changed with 10◦ step (a scale is to the right)

Fig. 8.4 RMS error versus frequency for ARTIKUL-P radio direction finder

σ,degree 2.5

1.5

0.5 0

100

300

500

700

900 f, MHz

errors occur under the influence of a large number of random factors, the actions of which may change from one measurement to another. Therefore, random errors cannot be taken into account by any corrections in the direction-finding results. Constant errors are caused by factors, which act constantly under given conditions, thus they can be taken into account during measurements.

Main Technical Parameters of Radio Direction Finders

245

The operation accuracy of the direction finder is the accuracy, which can be ensured under normal direction-finding conditions. It can be estimated by the large bearing number obtained under different conditions for the local emission sources, the location of which is known. These objects, where possible, should be placed at non-equal distances and should radiate radio waves of various wavelengths. In a number of cases, the electromagnetic field strength of the sources being found can be close to the direction finder’s sensitivity. Comparing the measured bearings with true azimuth, obtained by map tracing or by means of calculations, we can determine the separate measurement errors and, after that, find the root-mean-square operational error. Instrumental errors of the direction finder are caused by imperfection of its manufacture and alignment. First, any errors caused by non-accuracy of its antenna and feeder system, as well as the difference in characteristics of its antenna elements and feeder sections, may be related to instrumental errors. Instrumental accuracy is usually measured experimentally under the conditions of large signal/noise ratio at a field strength considerably exceeding the direction finder’s sensitivity. Unfortunately, we cannot measure instrumental accuracy for all direction finder systems. It is practically impossible to measure instrumental error for the direction finders with the bulky antenna system with the tens meter radius. In this case, we would have to limit analysis for the instrumental error, by separate sources. The simplest way to estimate the instrumental error of direction finders, which can be mounted on the rotary test bench, is as follows. Having tuned at the signal source, for instance, the test generator, we can determine its bearing. After that, the test generator is re-tuned to the next frequency and it determines the bearing again. Such measurements are provided for full operating-frequency range. Then the antenna system rotates for 10–15◦ and the measurement process is repeated. The measurement cycle is conducted for a whole range of operating angles of the direction finder. In order to take into account the local object influence, the test generator location is changed and the measurement cycle is again repeated. Sometimes, it is possible to define whether the error is instrumental or is caused by the local object, by varying the distance to the test generator at constant azimuth of its location. The instrumental error will not depend on distance, in contrast to an error caused by local objects. Thus, measurement of instrument accuracy is an extremely labor-consuming process. It is possible to essentially reduce some of its labor-intensiveness, if one has a programmable test generator and a program set that provides automatic synchronous re-tuning of the test generator and the direction finder. For example, when using the ARK-TG1 test generator or the ARK-TG3 generator, the time for measurement execution can be ten times reduced. Direction finder testing based on source elevation and polarization errors is particular complicated. To create such conditions, the test generator should be mounted at a large height (for example, on a mast or on a balloon) and should have a radiated antenna with the possibility to rotate on the necessary polarization angle. The instrumental errors of direction finding are component parts of the operational errors. However, determination of instrumental errors only cannot completely

246

8 Direction Finding of Radio Emission Sources

characterize the operational accuracy of the direction finder. On the other hand, determination of operational accuracy only is also insufficient, because it remains unknown which error part may be reduced by perfecting the direction finder structure, and which part may be reduced by its replacement.

Sensitivity The sensitivity of a direction finder defines its capability to take the bearing of remote and low-power sources. Usually, sensitivity is considered as such value of the electromagnetic field strength at which direction finding is executed under given characteristics, for instance, at given RMS error. Direction finder sensitivity is defined by the sensitivity of its receiving sections and the structure of its antenna system. At any decrease of the being-found signal field strength due to the direction finder’s internal noise action as well as any external noise influence, the RMS error of direction finding increases. In other words, the accuracy and the sensitivity of the direction finding are interconnected parameters and therefore, the indication of the direction finder sensitivity value – expressed in field strength units or in power flow density units, should be accompanied by the appropriate RMS direction-finding error value. Sometimes, equipment manufacturers indicate the direction finder sensitivity value at double the value of the RMS direction-finding error. Moreover, the RMS direction finder error and, relatively, the direction-finding sensitivity depend on the direction-finding time (on measurement number). Some manufacturers indicate the sensitivity data for the detection-finding equipment at bearing averaging during 1 s. To determine the direction finder’s sensitivity, a measuring receiver with a set of calibrated measuring antennas and a test generator with a controlled attenuator are necessary. The technique for sensitivity measurement may be, for instance, the following. The test generator is placed at a distance corresponding to the far emission zone. Next, one should check that, at a disconnected antenna, the spurious generator and feeder emission are not detected by the direction finder. The direction finder and the receiver with the measuring antennas are placed in such way so that the distance between them and the test generator will be equal. Since the measurement of weak signal field strength is inevitably accompanied by large errors caused by the external interference and the internal noise of the measuring receiver, the measurements of the direction finder’s sensitivity are carried out in two stages. At the first stage, one sets a small attenuation of the test generator equal to η1 dB; at that, the receiving signal level should be rather large, for instance, by 20–30 dB higher than the noise components level in the measurement bandwidth. The receiving signal electromagnetic field strength level E1 is estimated, for example, in dBμV/m, with the help of the measuring receiver. After that, the level of radiated signal gradually decreases by the attenuator of the test generator. Then, the direction-finding RMS error is calculated, which grows with the reducing of the radiated signal level.

Main Technical Parameters of Radio Direction Finders

247

The signal attenuation η2 dB, at the moment when the RMS direction-finding error begins to accept the overload capacity, is determined. The numerical direction finder sensitivity value is then determined in dBμV/m: E = E1 − |η2 − η1 | .

(8.2)

Figure 8.5 shows the ARTIKUL-P direction finder sensitivity versus frequency and the bearing number. The direction finder sensitivity is measured under the doubled RMS error values. As we can see, the direction finder sensitivity depends on frequency. In this case, this property is caused by the active antenna elements of the direction finder. At averaged bearing number growth, the sensitivity is improved. Thus, for example, at 1,000 MHz frequency, when averaging by three bearings, we have 6μV/m sensitivity (curve 1), while, when averaging by 30 bearings, we have 2.5μV/m sensitivity (curve 2) and, when averaging by 100 bearings, the sensitivity is equal to less than 2μV/m (curve 3). The technical parameters and structural peculiarities of the ARTIKUL-P direction finder will be discussed in section “ARTIKULP Portable Foldable Direction Finder”. E,μV/m

18 14 10 6

1

2

3

2 0

100

300

500

700

900

1100 F,MHz

Fig. 8.5 ARTIKUL-P direction finder sensitivity versus frequency and bearing number

The direction finder sensitivity depends on the bandwidth of the radio receiver sections. If the noise in the receiving channel has the uniform spectrum, the direction finder sensitivity value will be in inverse proportion to the square root of the bandwidth. However, the essential bandwidth narrowing leads to the impossibility of direction finding the pulse signals with small duration, which have the wide-band spectrum.

Noise Immunity Noise immunity is the direction finder parameter that shows its ability to operate under conditions of interference influence. Direction finder noise immunity is defined by the noise immunity of its receiving sections and antenna system, by the spatial discrimination, depending on antenna construction, by the display

248

8 Direction Finding of Radio Emission Sources

device, by processing approach, and by the degree of adaptation to the interference environment. The noise immunity for the direction finder’s receiving section, as for any other receivers, is characterized by the dynamic range, intermodulation selectivity (intercept points of second and third order), and selectivity on the spurious receiving channel. These features were discussed in detail in Chapter 3. If the direction finder uses active antenna systems, they may impair the intermodulation selectivity of the radio direction finder and its capability to operate under high-power signal influence. Other important factors defining radio direction finder selectivity are its immunity to field distortions caused by multi-path radio wave propagation, its stability to polarization errors, and its operational stability under the presence of non-coherent interference in the finding frequency channel.

Operating Rate The radio direction finder operating rate is defined by the minimal time interval during which the tuning process at the required frequency and the taking of the bearing occur. Recently, systems with POFT have been distributed, where the tuning rate at the required frequency is from several tens to several hundred jumps per second. Therefore, the direction finder operation rate becomes the determinative index for its application, especially in the military field. Maximal bearing-taking rate can be achieved in mono-pulse direction finders, where the bearing is measured during the duration of a single signal pulse. At that, for unambiguous bearing determination in the 360◦ angle range, at least three antennas are required. In the case of three parallel radio sections (front-ends), the operating rate of such a system will mainly be defined by the radio receiver tuning time, on the required frequency. In the general case, to increase the direction finder operating rate, we need to use the radio receiving sections with low transient period in the frequency synthesizer, to reduce the signal-taking time from the direction finder antenna elements, to decrease the antenna element number, and to reduce the calculation time at the expense of the resolve capacity degradation.

Resolution Resolution is the direction finder parameter that defines its capability to separate the bearing-taking of radio emission sources with close parameters. We can distinguish between the resolution on frequency and the resolution on angle. The resolution on frequency can vary for direction finders of different operation ranges. For example, in the hectometer wave range, radio transmitters with narrowband modulation types are used, and require a direction finder with higher resolution compared with the decimeter wave range. The resolution on frequency of modern

Main Technical Parameters of Radio Direction Finders

249

radio direction finders is mainly defined by the synthesizer phase noise in the receiving sections and by the digital signal-processing complexity. The problem of angle discrimination is also important for modern direction finders, since several radio emission sources may operate in a single radio frequency channel, for instance, the sources included in a radio network or the several base stations of a cellular radio communication system.

Operating Frequency Range The direction finder’s operating frequency range values the frequency area in which the radio direction finding occurs with the given values of accuracy and sensitivity. At present, radio emission sources use the radio frequency range from units of kilohertz to tens of gigahertz. Therefore, the wider the operating frequency range of the direction finder is, the more preferable it is for application in the field of radio monitoring.

Types of Being-Found Signals This parameter defines radio signal types, the sources of which can be detected by the direction finder. The type of being-found signal directly relates to the bandwidth of the direction finder receiving sections and with its operating rate. The wider the bandwidth is, the more wide-band and short-term signals can be detected and measured. Moreover, the ability to take the bearing of short-term periodic signals will depend on the mathematical processing, which can be fulfilled by the digital signalprocessing unit of the direction finder.

Deployment Time The deployment time of the direction finder is an important parameter for the mobile direction finder, indicating how fast the direction finder can be transformed from the transportation package to the operating mode.

Weight and Size Weight and size are important parameters, especially for application of direction finders in mobile systems. The less the weight and size of a radio direction finder are, the simpler it can be used in ground, air and navy carriers.

Complexity in Manufacture and Operation The complexity in manufacture and in operation parameter defines the possibility of bulk production of a direction finder, its operational convenience, and also its cost.

250

8 Direction Finding of Radio Emission Sources

Cost Cost is a serious direction finder parameter, defining the possibility of its purchase and application. The higher the direction finder quality is, the higher (as a rule) its cost is.

Classification of Direction-Finding Methods We can use different approaches for classification of direction-finding methods. If we wish to determine which signal parameter at output of the receiving antenna system plays the main role during measurement, we can distinguish the amplitude, phase, and amplitude-phase methods for bearing measurement [2]. If the method in which the data is obtained is used as a criterion concerning the direction to the radio emission source, then radio direction finders can be divided into singlechannel (consecutive) and multi-channel (mono-pulse) types. If we wish to take into consideration two known electromagnetic field properties in the far zone: the orthogonality of magnetic and electric vector components to the propagation direction and the orthogonality of the phase front plane to the propagation direction, then the known radio direction-finding methods can be attributed to two large groups [1]: The sensitive to polarization methods based on the determination of the direction of electric and (or) magnetic vectors of field strength can be attributed to the first group. The phase-sensitive methods based on the determination of the equal phase surface orientation are attributed to the second group. In polarization direction finders, the dipoles or the frame antennas are used. The classical rotated frame (minimum of received signal corresponds to the normal wave incidence on the frame plane) is attributed to this category. Currently, polarization direction finders are used under the condition of restricted space, when it is possible to apply compact antennas only, for example, in cars and in ships, for direction finding in the SW range. The direction estimation is executed mainly based on the Watson-Watt principle. Phase direction finders obtain information about the direction of radio wave arrival, from the spatial location of lines or surfaces with equal phase. We describe two main methods below. The first method relates to the directional properties of antennas. The directional pattern of an antenna is formed at superposition of elementary signals, which are received by the antenna. Thus, for instance, if the source signals received in two spaced points of the antenna system (say, by field sensors) are summed, the maximum resulting voltage corresponds to the case when the antenna system is turned out by the angle – with respect to the source, at which the phase difference of the received signals is minimal. The system turning to this or that side will lead to the summed signal decrease. On the contrary, when one signal is substituted from another one, we observe the evident pattern minimum in the direction of the incident wave. The second method relies on field measurement at different points in the limits of the geometrical sizes of the direction finder antenna (so-called, aperture sampling).

Classification of Direction-Finding Methods

251

The measurements can be fulfilled consecutively at displacement of the antenna field sensor, as well as simultaneously by the sensor set. In accordance with the procedure for bearing detection, we can additionally distinguish the approaches with the direct bearing determination and with the digital processing of signals from the antenna array (so-called, sensor array processing). Typical examples of the first technique’s implementation are the interferometers and Doppler direction finders. Another group of incident wave parameter estimation includes two different, in essence, approaches: the beam forming method, on the basis of which, in particular, correlation direction finders operate, and the subspace method used in MUSIC and ESPRIT algorithms of high resolution. At present, the following types of direction finders are the most widespread in radio monitoring systems [3]: • • • • •

Systems on the basis of rotating directional antenna Double-channel automatic direction finders (of Watson-Watt and Adcock) Quasi-Doppler systems Phase interferometer Correlation interferometric meter or correlative interferometer (CIM).

Since, at present, the wide-band systems are the most claimed (with the range covering factor of 100, or even more), the interferometric or phase radio direction finders are the most convenient for such system implementation. Several types of direction finders: amplitude direction finders, systems of Watson-Watt/Adcock, and multi-channel interferometers satisfy the second criterion (modulation type and bandwidth of the analyzed signal). Several types of amplitude direction finders on the basis of directional antennas (including CIM) have the possibility of azimuth discrimination of several signals to a valuable degree. Interferometer shortcomings are: relatively high cost and comparatively large response time (especially for the double-channel correlation interferometers). Each used direction-finding method has its own advantages and shortcomings, but, for the multi-function monitoring systems, CIM application is preferable. CIM allows the possibility of direction finding of practically all radio signal types, including the wide-band with complicated modulation types. CIM has the capability of simultaneous processing and several-signal discrimination in a single frequency, both coherent (at multi-path receiving of the single emission source) and non-coherent (at receiving of radio signals from several sources with overlapping spectrum). For CIM, effective methods for reducing the instrumental errors caused by the local conditions and by the interaction of antenna elements are developed. Moreover, CIM implementation can be simplified on the basis of the unified units: single-type nondirectional antenna elements, double-channel radio receivers with the common local oscillator, antenna switchers, and analog-digital converters. In correlation interferometers, spatial resolution and measurement of the wave arrival direction can be effectively combined with the measurement of field strength of each detected source.

252

8 Direction Finding of Radio Emission Sources

Systems Based on a Rotating Directional Antenna Systems based on a rotating directional antenna use the amplitude methods. We may categorize these systems according to their direction finding: by maximum, minimum and equisignal. When finding on the maximum method, the pattern maximum of the rotating antenna system is the bearing indicator. The resulting voltage U at antenna output is defined by its pattern F(θ,β), where θ is the azimuth, β is the elevation, and depends on the field strength E at the receiving point: U = he EF (θ , β)

(8.3)

where he is the effective antenna height. The RES angle coordinate is counted out at that moment, when signal amplitude at radio receiver output achieves the maximal value (Fig. 8.6). At that, the perpendicular to the phase front of the incident wave is perpendicular to the antenna aperture plane, and the direction of the pattern maximum coincides with the direction to the radio emission source. Fig. 8.6 Direction finding on the basis of the maximum method

North

A

U

θ

Θ0 Ωt

O

θ

Ω

The advantages of the maximum method are: relatively small noise influence on direction-finding accuracy because, in the direction of the pattern maximum, the maximal possible power of the analyzed signal is received; the possibility to resolve several emission sources with different azimuth in a single frequency channel; cheap implementation, since only the single-channel receiver is required. The shortcoming of this method is low direction-finding accuracy because of the low pattern curvature near the maximum. The angle determination accuracy is low and has the order of one fifth from the antenna pattern width 0 : θ ≈ 0.20 .

(8.4)

Systems Based on a Rotating Directional Antenna

253

To determine the bearing, the antenna is rotated observing the output voltage. We may turn the antenna till the output voltage maximum or rotate it continuously. In radio direction finders using the second approach, the voltage is modulated with the antenna rotation frequency. If is the angular frequency of the antenna rotation, the voltage at antenna output will be modulated in accordance with the law U = Umax F(t − θ ).

(8.5)

The phase θ will correspond to the bearing. Therefore, the direction finder with continuous antenna rotation will be an automatic one, i.e., it will determine the bearing without additional observer operations. At that, the phase meter method of direction finding will be implemented. At direction finding on minimum, the pattern minimum is the bearing indicator, and it should be only one (Fig. 8.7). This diagram can be obtained, for instance, using a pair of antennas with narrow pattern. Due to the fact that the minimum of the resulting pattern is rather sharp, one can increase the direction-finding accuracy, compared with the maximum method, up to one tenth from the pattern width: θ ≈ 0.10 .

Fig. 8.7 Direction finding on the basis of the minimum method

(8.6)

North

A

U

θ

O

θ

Ωt

Ω

The shortcoming of systems using the minimum method is a decrease of the direction-finding accuracy at low signal/noise ratio, due to additional signal reducing by the pattern minimum: at the moment of determining the direction to the object, the output voltage level is the same as at signal absence. This may lead to the erroneous determination of object direction. The equisignal method is the compromise between the two above-mentioned methods. It is realized with the help of two directional antennas, which are turned in azimuth plane (Fig. 8.8). The true direction to the emission source is considered

254

8 Direction Finding of Radio Emission Sources

North

ε θ

A

U

ε U1

U3

Ωt

O

θ Ω U2

Fig. 8.8 Direction finding on the basis of equisignal method

those directions, which are located between the pattern maximums of two antennas, assuming that the signal levels at the antenna outputs are equal. If the patterns F(θ ) of the pair of antennas are the same and its maximums are turned by angle ε, then it is possible to describe it by the functions F(θ+ε) and F(θ−ε). The direction θ corresponding to the pattern intersection is referred to as equisignal. The signals from the antennas outputs are filtered, amplified, detected and then compared (subtracting one from another). In result, the dependence of the resulting voltage amplitude on the angular direction to the object is u = U1 (θ ) − U2 (θ ) = k[F(θ + ε) − F(θ − ε)].

(8.7)

This function of the direction finder output signal versus the angle of wave arrival is the bearing characteristic, and, at symmetrical patterns, it will be the odd function. Contrary to the minimum method, the equisignal method provides exact information about signal presence, having analyzed the signal at one antenna output. At small deflection of the equisignal direction from the direction to the radio emission source, the polarity of this voltage will indicate the deflection sign, and the voltage level will show the deflection level. This is due to the oddness of the bearing characteristic and the linearity of its central part. The advantages of this method are higher instrumental accuracy compared to the direction finder on the maximum method, and higher sensitivity compared to

Systems Based on a Rotating Directional Antenna

255

the direction finders on the minimum method. The equisignal method ensures the direction finding accuracy as follows: θ ≈ 0.050 .

(8.8)

The equisignal method is implemented in the automatic radio compass, described in section “Automatic Radio Compass”. Common disadvantages of the direction finders with rotating antennas are: the narrow operating frequency range (the covering factor, as a rule, is not more than 10); long response time (defined by the time of antenna rotation); the complicated mechanical drive of the antenna system; low survey rate. This method is ineffective for short-term signals, the duration of which is small compared to the antenna rotation period. In spite of these disadvantages, the amplitude method, which uses rotating directional antennas, has been applied up to now, as the other method applications often demand large expenses and lead to large weight and size of the direction finder. The

Fig. 8.9 External ARK-KNB4 distantly-controlled converter mounted on rotating mast

256

8 Direction Finding of Radio Emission Sources

amplitude method can be especially applied to the UHF range: here, this method is a good compromise between the quality indexes of direction finder operation and its cost. Sections “Systems Based on a Rotating Directional Antenna” and “ARK-RP3 Handheld Radio Direction Finder” show examples of handheld direction finders, while in section “Automatic Radio Direction Finder with Low Antenna Base” a description of the ARK-KNV4 UHF radio signals converter is given. This device has the built-in directional antenna for 3–18 GHz and can be used as a direction finder. Figure 8.9 shows the ARK-KNV4 converter mounted on a rotating mast. The ARK-DVP indicator of UHF emissions, shown in Fig. 8.10, also has the directional antenna phased arrays and can be used in the range of 6–12 GHz, for handheld direction finding of radio emission sources. Fig. 8.10 ARK-DVP indicator of UHF emission

ARK-RP3 Handheld Radio Direction Finder The amplitude direction-finding method is used in handheld direction finders, which are handheld radio receivers equipped with the directional antenna. In the simplest case, the maximum or minimum methods are applied. The directional antenna is mounted on the special handle. The direction finder operator, holding the handle in his arm and turning it, determines the direction of radio signal arrival. In spite of the fact that the maximum method has less direction-finding accuracy, nevertheless, exactly this approach is, as a rule, used, under complicated operation conditions,

ARK-RP3 Handheld Radio Direction Finder

257

in difficult-to-access places (for instance, on the roof), since it allows provision for (simultaneously with the direction finding) the acoustic monitoring of the radio station. As an example, let us consider the construction and the functions of the ARKRP3 handheld radio direction finder. The direction finder consists of the receiving and processing unit (RPU), the ARK-PP panoramic add-on device combined with the control panel, the handle – for open direction finding, and the set of exchangeable directional antennas. Figure 8.11 shows the main parts of this radio direction finder. Fig. 8.11 Main parts of ARK-RP3 handheld radio direction finder: 1 – ARK-PP wrist control panel; 2 – Handle for directional antennas; 3 – Receiving and processing unit; 4 – Directional antennas in the bag

1 2

3

The direction finder has a reserve rechargeable power source: the charger, the ARK-BP-NK3 power supply unit for power supply from the electric mains and from the vehicle’s onboard network, the shoulder bag and the rucksack for transportation, headphones. To use the ARGAMAK panoramic receiver together with a PC at the stationary or temporal post, software packages are included in the complex structure, for automatic radio monitoring (SMO-PA), for technical analysis (SMO-STA), and for post-processing (SMO-ASPD). The RPU of the direction finder (Fig. 8.11) looks like a small case in which the ARGAMAK panoramic radio receiver is mounted. It consists of the ARK-PS5 controlled radio signal converter and the ARK-CO2 ADC module. There is a compartment with internal rechargeable batteries inside the case. The direction finder can operate continuously from the internal batteries for a period of 6–8 h. The ARK-PP panoramic add-on device is intended for the fast panoramic spectral analysis of the radio emission sources (RES) and for control of the radio direction finder’s operational modes. The add-on device has a built-in LCD display and push-button keyboard. The display shows the spectral diagrams of the radio signals

258 Fig. 8.12 External view of the handle for directional antennas mounting

8 Direction Finding of Radio Emission Sources

ARK-RP3 Handheld Radio Direction Finder

259

in the band 2 MHz or 0.5 MHz, the signal level at the tuned frequency, as well as the setting parameters of the direction finder. With the help of the push-button keyboard, the operational modes are set, and the selection of demodulator type is executed. The handle for open direction finding is shown in Figs. 8.12 and 8.13. There is a connector for directional antennas, which ensures the antenna turn around the longitudinal axis at a 90◦ angle – to select a signal polarization. The digital indicator shows the tuning frequency of the direction finder and the received signal level. The handle has a film keyboard, which can be used to define the tuning frequency; in the upper part of the handle, there is a magnetic compass. The radio direction finder includes antennas systems for open and concealed direction finding. The antenna system for open direction finding includes five portable directional antennas covering the range from 300 kHz to 3 GHz. The antenna system for concealed direction finding includes the control module and several directional antennas for concealed direction finding. A list of antennas is shown in Table 8.1. Figures 8.14 and 8.15 show several antennas for open direction finding. The ARK-RP3 radio direction finder can be used in open and concealed modes, as well as in the mode of automatic radio monitoring, under control from the external PC. In the open mode of direction finding, the handle with one of the directional antennas and the panoramic add-on device are connected to the RPU. The operator takes the handle with the connected antenna module in his arm and tunes to the analyzed signal frequency, using the handle keyboard. After that, he turns around his axis, observing the indicating signal level. He determines the direction of wave

Fig. 8.13 The handle with the ARK-A3-1 directional antenna

260

8 Direction Finding of Radio Emission Sources Table 8.1 List of directional antennas included in direction finder set Name

Range of operating frequencies, MHz

For open direction finding ARK-A3-KV Antenna module ARK-A3-0 Antenna module ARK-A3-1 Antenna module ARK-A3-2 Antenna module ARK-A3-3 Antenna module

0.3–36 25–200 90–450 370–1,000 1,000–3,000

For concealed direction finding ARK-A3-KVS Antenna module ARK-A3-0S Antenna module ARK-A3-1S Antenna module ARK-A3-2S Antenna module ARK-A3-3S Antenna module

0.3–30 20–200 100–400 300–1500 1,500–3,000

Fig. 8.14 ARK-A3-0 antenna module

arrival by the maximal level. At that, he hears the demodulated received signal in one of the headphones, while in another headphone – the tone signal, the frequency of which changes from 100 Hz to 10 kHz depending on the input signal level. The tone signal simplifies the determination of the direction at which the maximal level of received signal occurs. The antenna has a pattern that looks like a cardioid, and the difference between the maximal and minimal signal levels from the various directions may achieve 20–30 dB. As an example, Fig. 8.16 shows the pattern of the ARK-A3-1 antenna for the frequencies 280, 330, 380, and 430 MHz. To find the direction of the weak signals, the expanded scale mode for the signal level is used. The expanded scale corresponds to the level range from –15 to +10 dBμV, while the main scale is from –20 to +80 dB μV.

ARK-RP3 Handheld Radio Direction Finder

261

Fig. 8.15 ARK-A3-1 antenna module

The concealed mode is used when it is undesirable to attract attention to operator work. The ARK-PP panoramic add-on device allows operation in three main modes: “Panorama”, “Search”, “Review”. “Panorama” mode is intended for spectral analysis near the selected frequency, and also for listening to the demodulated signal on the selected frequency. “Search” mode is used for detection of active channels in the selected frequency range. The central frequencies of detected channels are automatically recorded into the table. In “Review” mode, an active channel search is provided on the basis of the frequency list from the formed-in-advance table, which is formed for scanning during the operation stage, in “Search” mode or manually. 0

0 340

350 1.00

10

340

20

330

0.80

310

310

50

0.60 0.40

270

60

0.40

0.20

80

280

0.20

0.00

90 270

0.00

70 80 90

100 260

250

110

240

130 220

140 200

190

160 180

170

110

240

430

120

230

130 220

140 150

210

150

210

100

250

120 230

280

50

0.60

290

70

260

330

40

300

60

280

30

0.80

320

40

300 290

20

330

30

320

10

350 1.00

380

160

200 190

180

170

Fig. 8.16 Patterns of ARK-A3-1 antenna for the frequencies 280, 330, 380, and 430 MHz

262

8 Direction Finding of Radio Emission Sources

Operation mode selection and selection of its parameters are provided with the help of the ARK-PP panoramic add-on device’s menu.

ARK-RP4 Handheld Radio Direction Finder At present, radio systems operating in the frequency range higher than 1 GHz have the widest application. The peculiarities of such radio systems are high transmission rate and, as a rule, a wide band of occupied frequencies (from units of MHz to several tens of MHz). Another feature of the devices operating in the frequency range higher than 1 MHz is relatively low transmission power, as a rule, in the limits of 1 mW to 1 W. The third property of a system with such type devices is the wide application of narrow-beam antennas having a pattern width up to several degrees. The mentioned features of these systems cause location determination problems during the radio monitoring, since radio direction finders with high operation rate, large bandwidth, and with the possibility of spatial displacement (including in vertical plane) are required. The last property of the direction finder is very important, since the stationary means of radio monitoring, located in fixed places on the Earth’s surface, cannot practically detect the radio signals emitted by highdirectional antennas. Under conditions of strong urban development, application of mobile, grounded, radio monitoring stations also does not always lead to positive results because data transmission can be conducted via narrow spatial beam between the sharply directed antennas mounted on high building roofs. To detect the location of radio equipment operating in a wide frequency band, and using the modulation types with time division, together with directional antennas, we can suggest the ARK-RP4 handheld direction finder. External view of this direction finder is shown in Fig. 8.17. This equipment ensures autonomous operation, under control from the panoramic add-on device. At that point, manual direction finding and panoramic analysis in the band of displayed frequencies from 8 MHz or 128 MHz are provided. The main technical parameters of this direction finder are listed in Table 8.2. The following parts are included into the basic direction finder set:

Fig. 8.17 ARK-RP4 handheld direction finder with panoramic add-on device

Automatic Radio Compass

263

Table 8.2 Basic specifications of ARK-RP4 direction finder General parameters Operating frequency range, MHz Dynamic range, dB Noise factor, dB Operating temperature range, ◦ C Weight of basic set, kg Supply voltage from rechargeable batteries, V Supply voltage from vehicle onboard network, V Supply voltage from AC mains, V

1–8,000 60 14 –20 to +55 5 12 10.6–13.6 90–250

Direction finding Direction-finding method Sensitivity, μV/m

Amplitude 10–30

Information display at operation from the panoramic add-on device Displayed frequency band, MHz Spectrum display modes Display of receiver tuning frequency Indication of signal spectrum width Indication of signal level

128. 8 Instantaneous, averaged, aggregated Numerical Numerical Graphical and numerical

Panoramic analysis Simultaneous survey band, MHz Rate in the band 128 MHz (discreteness 800 kHz), GHz/s Rate in the band 8 MHz (discreteness 50 kHz), GHz/s

8 4 8

• Radio signal converter on the basis of the ARK-PS5 module of the ARGAMAK family (see section “ARGAMAK-I Panoramic Measuring Receiver”) • Wrist control panel with the ARK-PP4 panoramic add-on device • Directional antenna • Rechargeable battery, charger and power supply unit from AC mains and from vehicle’s onboard network • Cables set • Shoulder bag for the equipment and for operation while in motion • Rucksack for the equipment • Operation documentation • Software packages for SMO-PA automatic radio monitoring and for the SMOASPD post-processing • PC.

Automatic Radio Compass For implementation of the equisignal method, a single antenna is often used, the pattern of which consequently in time may accept two positions. Such a method

264

8 Direction Finding of Radio Emission Sources

Rotating antenna loop

RF amplifier

Balance modulator

Loop motor

Spike antenna

Phone channel

Receiver

Phase detector

Amplifier

LF generator

N Selsynsensor

Selsynreceiver

W Indicator limb

E S

Fig. 8.18 Structural diagram of automatic radio compass

is, for instance, used in an automatic radio compass. The structural diagram of the radio compass is shown in Fig. 8.18 [2]. The antenna system of a radio compass consists of the rotating frame antenna and a spike antenna located along the rotation axis of the frame one. If, at the receiving point, the analyzing radio station creates the field strength e(t) = E cos ω0 t

(8.9)

then the electromotive force induces in the frame el (t) = El ( sin θ ) sin ω0 t.

(8.10)

The angle of radio wave arrival θ is counted out of the perpendicular to the frame plane, as shown in Fig. 8.19.

y

z

O

O

θ x

Fig. 8.19 Pattern of frame antenna

y

x

A

θ

Automatic Radio Compass

265

Application of the single frame only leads to determination ambiguity regarding the side of wave arrival. Moreover, the used maximum or minimum method has the above-discussed shortcomings. Therefore, in the automatic radio compass, the frame signal after amplification and a 90◦ turn is summed with the following signal from the nondirectional spike antenna: eant = Eant cos ω0 t.

(8.11)

The resulting signal, under the condition El = Eant , takes the form: er (t) = el (t) + eant (t) = Eant (1 + sin θ ) cos ω0 t.

(8.12)

The pattern of such an antenna system in polar coordinates is represented by the cardioid defined by the expression (1+sin θ ). The cardioid plot is shown in Fig. 8.20a. Equisignal direction

– +

+ ep

a)

O

ea

ec

b)

Fig. 8.20 Formation of equisignal direction in the automatic compass

At signal phase varying by 180◦ on the frame antenna output, the direction of the cardioid maximum will also change by 180◦ . In the automatic radio compass, the signal phase from the frame output switches by 180◦ periodically, with the help of the balance modulator. At that, the cardioid will be transferred from the right halfplane to the left one, as is shown in Fig. 8.20b, i.e., two symmetric patterns will be formed in turn. Due to the phase-locked-frame (PLL) system in the given equipment, a single, stable, equisignal direction will be realized only, and it will be directed upwards vertically from 0 point. The motor controlled by the signal from the phase detector output will turn the frame antenna until the equisignal direction will coincide with the direction to the signal source. The rotating frame antenna can be changed by a system comprised of two, fixed, mutually-perpendicular frames.

266

8 Direction Finding of Radio Emission Sources

Automatic Radio Direction Finder with Low Antenna Base The antenna system of such a direction finder represents two, fixed, mutuallyperpendicular frames. Moreover, there is one more antenna with a circular pattern (Fig. 8.21). Fig. 8.21 Antenna system in the form of two perpendicular frames

North θ

+ e p2

eep1 p1

+ –

a)

b)

Signals from the radio emission sources induced in the frames are defined as el1 (t) = El cos θ sin ω0 t,

(8.13)

el2 (t) = El sin θ sin ω0 t.

(8.14)

Signals from the output of each frame pass to the identical receiving channels, where they are filtered, amplified and converted to the intermediate frequency ωIF : u1 (t) = U0 cos θ sin ωIF t,

(8.15)

u2 (t) = U0 sin θ sin ωIF t

(8.16)

where U0 is the maximal output voltage at the intermediate frequency. Then, if the electron-beam tube (EBT) is used as an indicator unit, the signal u1 (t) passes to the vertical deflector, and the signal u2 (t) to the vertical deflector. Depending on the value and the signs of u1 (t) and u2 (t), the electron beam draws on the screen the diameter line deflected by the angle defined by the following relation tan ϕ =

u1 = tan θ . u2

(8.17)

Automatic Radio Direction Finder with Low Antenna Base

267

Two angular values ϕ=θ and ϕ=π+θ correspond to this relation, i.e., there is ambiguity in the bearing count. In order to eliminate the bearing ambiguity, an additional nondirectional antenna is used, the signal from which is phased in such way that it is in-phase (or antiphase) with the signal from the first frame, when the radio emission source is placed exactly on North. The signal from the nondirectional antenna – after filtering, amplification, and conversion to the intermediate frequency – passes to the control electrode of the EBT and, by its own negative half-period, reduces the half-sweep of the beam corresponding to the false direction. Figure 8.22 shows the output voltages of the main channels u1 (t) and u2 (t), and also the signal of the additional third channel u3 (t) for the angle of radio wave arrival θ = 30◦ and θ = 250◦ . θ = 250°

θ = 30°

U1

U1 t

U2

t

U2 t

U3

t

U3 t

t

0

0 270

90

270

180

90 180

Fig. 8.22 Voltage drawings in radio direction finder channels

At radio wave angle-of-arrival θ = 30◦ , the voltage u3 (t) reduces the image corresponding to the negative parts of voltages u1 (t) and u2 (t). If the angle is equal to 250◦ , then the image is reduced corresponding to the positive parts of u1 (t) and u2 (t). Thus, the unambiguous bearing on the indicator takes place upon adding to this direction-finding method the disabling signal obtained with the help of the nondirectional antenna. This method was offered in 1926 by Watson-Watt. The structural

268

8 Direction Finding of Radio Emission Sources

diagram of a direction finder implementing the mentioned principle is shown in Fig. 8.23.

uU11

Radio receiver unit

U2

u2

1 2 Antenna system

u3a U

3 Indication unit

Fig. 8.23 Automatic radio direction finder with frame antennas

In the hectometer range, the direction finder with intersected frame antennas, especially at night, will yield large errors, caused by the horizontal antenna receiving parts of the reflected-from-the-ionosphere steep-incident radio waves. If two pairs of identical, antiphase-connected, vertical, vibrators with mutuallyperpendicular bases, which are much less than the wavelength, are used as the antenna system, the pattern of this system will be identical to the antenna with the mutually-perpendicular frames. But, the antenna system with the spikes ensures fewer direction-finding errors at the presence of the radio waves reflected from the ionosphere, compared with the frame antennas. Let us assume that, at two plane points A1 and A2 at distance 2b, one from the other, the similar antennas are located (Fig. 8.24). The vertically polarized wave, the direction of which is given by the angle θ in the horizontal plane, acts on these antennas. For simplicity, we assume that the angle-of-arrival β in the ver-

Ph

g Si

e as fro nt

y θ

A1

Δ

A2

O

b Fig. 8.24 The bearing pair

x b

na

l

Automatic Radio Direction Finder with Low Antenna Base

269

tical plane is equal to zero. In this case, the voltages at antennas outputs can be written as: uA1 (t) = U0 cos (ω0 t − ϕA )

(8.18)

uA2 (t) = U0 cos (ω0 t + ϕA )

(8.19)

where ω0 is the wave radian frequency; ϕ A is the wave phase shift in A1 point with respect to 0 point, or the phase shift in 0 point with respect to A2 point. Let us connect the antennas in an antiphase manner, then the resulting voltage ur at output will be: ur = uA2 − uA1 = −2U0 sin ϕA sin ω0 t.

(8.20)

The phase shift ϕ A is defined by the distance , as is shown in Fig. 8.24. The wave will pass this distance during time τdel = /c

(8.21)

where c is the light speed. Using the expression c=

ωλ , 2π

(8.22)

2π . λω0

(8.23)

where λ is the wavelength, we get: τdel = Then, ϕA = ω0 τdel =

2π . λ

(8.24)

In its turn, the distance = b sin θ .

(8.25)

Substituting (8.25) into (8.24), we get: ϕA =

2πb sin θ . λ

(8.26)

Using the formula (8.26), we can rewrite the equation for the resulting voltage at antenna pair output:

2πb sin θ ur = −2U0 sin λ

sin ω0 t.

(8.27)

270

8 Direction Finding of Radio Emission Sources

For a small base between the antennas (i.e., b/λ << 1), the approximate relation can be used:

2πb sin θ sin λ

≈

2πb sin θ . λ

(8.28)

Hence, Equation (8.27) can be reduced to: ur ≈ Ur sin θ sin ω0 t

(8.29)

where Ur = 2U0 (2π b/λ) is the resulting voltage amplitude at the output of the antenna pair. Thus, the bearing pair with a small base composed of antiphase connected elements will have the figure-eight pattern, as for the frame antenna. The antenna system with two bearing pairs with mutually-perpendicular bases is called the Adcock antenna. In the first half of the twentieth century, Adcock antennas were widely used in practice. The Adcock antenna is known as the antenna with a small base, since the geometrical distance between the spike antennas is significantly less than the wavelength. The direction finder with an Adcock antenna system has fewer direction-finding errors caused by the radio waves reflected from the ionosphere, compared with the direction finders with usual frame antennas. If the direction finder has three radio receiving channels and three antenna sections, this direction finder will be mono-pulse. Such a direction finder ensures the direction finding of signals with minimal duration, has the simplest technical implementation, small weight and sizes. The shortcoming of the direction finder with a small base is the fact that the antenna with a small base leads to errors, in the case of multi-path radio propagation. At present, Adcock antenna systems having in its structure more than two bearing pairs are used. Such antenna systems may have the increased base and it decreases the bearing errors caused by the multipath radio waves propagation. Modern direction finders do not use EBT for displaying the voltage ratio at antenna output, but provide the digital signal processing at the intermediate frequency. The good selectivity on the adjacent channel is achieved by the digital processing, while the bearing is calculated numerically and is displayed by PC with the help of graphical interface.

Doppler and Quasi-Doppler Direction Finders A Doppler direction finder is as a phase direction finder that extracts information on electromagnetic wave propagation direction from the spatial location of lines or surfaces with an equal phase. Its operation is based on the Doppler effect. Essentially, the Doppler effect consists in the fact that relative (mutual) displacement of receiver or transmitter leads to the change of the received oscillations fre-

Doppler and Quasi-Doppler Direction Finders

271

quency (and, hence, phase). The received signal frequency becomes different from the emitted frequency. The frequency increment caused by the antenna rotation in the form of induced electromotive force is negative in the time interval, when the antenna moves away from the transmitter, and positive, when the antenna arrives. It is equal to zero when the antenna moves in perpendicular direction to the direction of electromagnetic wave propagation. Let the nondirectional (in the horizontal plane) receiving antenna (for instance, the vertical vibrator) rotate with radian frequency along the circle with R radius in the field created by the remote transmitter of the electromagnetic oscillations with ω0 frequency. By analogy with (8.26), the phase difference between the rotating antenna and the point corresponding to rotation center is equal: ϕA =

2πR sin (t + θ ). λ

(8.30)

Thus, the phase of the signal induced in the antenna, will be modulated in accordance with sine law, and the instantaneous frequency deflection from the nominal value ω0 ωA =

dϕA 2πR = cos (t + θ ) dt λ

(8.31)

will also vary with frequency , and the initial phase of the frequency-changing will be defined by the radio wave incidence angle θ . In practice, instead of the rotating antennas, systems of fixed antennas located along the circle, which are connected in turn to receiver input with the frequency , in some way, are often used. This direction finder is referred to as a quasi-Doppler one. In a quasi-Doppler direction finder, the adjacent antenna-switching is executed in a non-immediate manner but with varying-in-time weighting factors, for example, varying in accordance with linear law. In the first time moment, let the first antenna element be completely connected, and the second one be completely disconnected from the output of the antenna array. With time, the weighting factor of the first antenna connection is gradually (in the ideal case – linearly) reduced to zero till the moment of complete connection of the second antenna, which corresponds to a “smooth” antenna turn in the Doppler direction finder. To extract the initial phase and, hence, to determine the azimuth to the transmitter, the signal from the antenna switch output passes to the phase detector, where it is compared by phase to the reference voltage. As the reference voltage, we can use the following: u0 (t) = U0 cos t.

(8.32)

The phase of this voltage is equal to zero, at the moments of antenna element connection, which corresponds to zero azimuth.

272

8 Direction Finding of Radio Emission Sources

The considered direction finder can define the direction to RES of any type (modulated as well as non-modulated). Actually, in the case of amplitude modulation, the amplitude of induced electromotive force in the antennas will be a changing variable that does not influence in any way on the result. In the case of angular modulation, the lowest frequency Flow of modulating signal is usually some times more than the angular “rotation” frequency of the antenna (for instance, for the standard phone channel Flow = 300 Hz, and is chosen near 100–150 Hz), which can neglect the influence of the angular modulation on the bearing estimation. To obtain unambiguous direction-finding results, the distance between separate antenna elements should be less than half the wavelength of the received emission. In practice, one usually selects a distance of about 1/3 of the minimally-possible wavelength. Direction finders using this approach are mainly suitable for operation with narrow-band sources with continuous modulation types. At that, the Doppler method’s serious shortcoming is the necessity of exact tuning to the signal carrier frequency, because, at operation on the frequency response slopes, the frequency variations for FM are converted into amplitude modulation. Another shortcoming of this method is the long bearing-taking time: a minimally-required one cycle of antenna scanning. At a typical rotation frequency of 150 Hz, one cycle takes approximately 7 ms. Figure 8.25 shows the automatic quasi-Doppler radio direction finder, which was manufactured in Russia for several years in the twentieth century, but was later discontinued due to the above-mentioned shortcomings.

Fig. 8.25 ARK-PK-P quasi-doppler direction finder

Phase and Correlation Interferometers

273

Phase and Correlation Interferometers Direction finders that calculate bearing on the basis of the signal phase difference from the antenna system elements are referred to as interferometers. Interferometers can be divided into two types: phase and correlation interferometers. Usually, not less than two, coherent, reception channels are used in interferometers and thus, interferometry can be considered as the use of multi-channel direction finders. In interferometers, nondirectional wide-band antenna elements are used as the antenna elements. Two receiving channels, in combination with the phase detector, allow the phase delay of the analyzed radio signal received by two different elements of the antenna array to be determined. In especially important cases, to reduce the response time, mono-pulse direction finders are used, in which the number of receiving channels is equal to the number of used antenna elements. In phase interferometers, the bearing is calculated directly on the basis of measured phase differences of the antenna array elements. Consequent comparison is made of the dataset containing the measured phase differences between the elements of the antenna array versus the data array containing the differences values calculated theoretically at the different angles-of-arrival. The comparison is fulfilled by means of square error calculation or the correlation factor. As a final result, the azimuth value is taken, for which the correlation factor is maximal. The general structural diagram of an interferometer with N−channel digital receiver is shown in Fig. 8.26.

Antenna buffer 1

Receiver 1

ADC 1

Antenna buffer 2

Receiver 2

ADC 2

...

...

Antenna element 2

... Antenna element N

Antenna buffer N

Receiver N

Calibration generator

Frequency synthesizer

Antenna array

...

N-channel receiver

Fig. 8.26 Phase interferometer with N− channel receiver

ADC N

Digital signal processing unit

Antenna element 1

Analog-digital signal processing unit

PC

274

8 Direction Finding of Radio Emission Sources

For the electromagnetic signal field, the electrical component strength in the center of the antenna array is: e(t) = E0 cos [ω0 t + ϕ(t) + ϕ0 ]

(8.33)

where E0 is the amplitude, ω0 is the central radio signal frequency, ϕ(t) is the phase, depending on modulation on receiving signal, ϕ 0 is the initial radio signal phase in the center of the antenna array. The electromagnetic field strength generated by the signal in phase center of n-th element is en (t) = E0 cos [ω0 t + ϕn (t) + ϕno ] = = E0 cos [ω0 (t + τdel ) + ϕ(t + τdel ) + ϕ0 ]

(8.34)

where τdel is the delay time caused by the propagation difference of the electromagnetic wave front from n− th element to the center of the antenna array. We shall assume that the distances between the antenna’s array elements and its center are commensurable with the carrier wavelength, and that the signal ek (t) is a narrow-band one. In this case, the phase ϕ(t) changing rate will be negligible compared to the changing rate of function ω0 t, and the Equation (8.34) can be simplified as: en (t) = E0 cos [ω0 t + ω0 τdel + ϕ(t) + ϕ0 ].

(8.35)

Using the Equation (8.23), we get: en (t) = E0 cos [ω0 t +

2πn + ϕ(t) + ϕ0 ]. λ

(8.36)

We assume that the space is three-dimensional. To calculate the propagation difference n , it is convenient to use the formula for the scalar product of two vectors. Let r0 vector to be the unit vector coinciding with the direction to the RES, i.e., n vector is the radiusperpendicular to the wave phase front, as shown in Fig. 8.27. R vector defined the position of n−th element of the antenna array. As is well known, n can be considered as the product the scalar product of the vector r0 to the vector R of the |r0 | vector magnitude to the Rn vector projection to the r0 vector direction [4]. Let both vectors come out from the origin of the coordinates, i.e., from the n vector projection to the unit vector r0 will be equal antenna array center. The R to the wave propagation difference from n−th element of the antenna array to its center: n n = r0T R

(8.37)

where r0T = [x0 ,y0 ,z0 ]T is the unit vector coinciding with the direction to the RES and defined by its projections the coordinate axis of the right-side Cartesian coordi n = [xn ,yn ,zn ]T is the radius-vector of n−th element of the antenna nate system; R array defining its position.

Phase and Correlation Interferometers

275

Fig. 8.27 Determining the complex amplitude of electromagnetic field strength in n− th element of the antenna array

Z

r0 β0 0

x0 α n

β

y0 y n

Θ

Y

Θ0

xn

Rn

X So,

2π T n + ϕ0 en (t) = E cos ω0 t + r R λ 0

(8.38)

and the complex amplitude of the electromagnetic wave strength at the point of the phase center location of the n−th antenna array element is: ( 2π T ˙En = E0 exp j r Rn + ϕ(t) + ϕ0 . λ 0

(8.39)

Having expanded the scalar product, we have: ( 2π ˙En = E0 exp j (x0 xn + y0 yn + z0 zn ) + ϕ(t) + ϕ0 . λ

(8.40)

Let us transfer from the unit vector r0 = [x0 ,y0 ,z0 ]T representation in the Cartesian coordinate system to its representation in the spherical coordinate system. Then, E˙ n = E0 exp j 2π λ (xn sin β0 cos θ0 + yn sin β0 sin θ0 * + zn cos β0 ) + ϕ(t) + ϕ0

(8.41)

276

8 Direction Finding of Radio Emission Sources

where θ 0 is the azimuth direction to the RES, counted out from the x axis counterclockwise; β 0 is the elevation, counted out from the z axis clockwise, as agreed in the spherical coordinate system. In direction-finding practice, the angle θ is usually used for azimuth designation, counted out from the y axis clockwise, and the angle β for elevation designation, counted out from the azimuth plane. The angles θ and β are related to the spherical coordinate system θ 0 and β 0 by the simple relations: θ=

π π − θ0 , β = − β0 . 2 2

(8.42)

To simplify the mathematical expressions, we assume, for RES azimuth and elevation designations in the following computations, as a rule, to use the spherical coordinate system angles θ 0 and β 0 . If all elements of the antenna array are located in the x0y plane, the projection of the antenna element radius-vector on the z axis will be equal to zero (z n = 0) and ( 2π E˙ n = E0 exp j (xn cos θ0 + yn sin θ0 ) sin β0 + ϕ(t) + ϕ0 . λ

(8.43)

Using the last equation, we define the phase shift between two elements of the direction finder antenna array n1,n2 = arg (E˙ n1 ) − arg (E˙ n2 ) = =

2π λ [(xn1

− xn2 ) cos θ0 − (yn1 − yn2 ) sin β0 ]

(8.44)

The last equation contains two unknown variables: azimuth θ 0 and elevation β 0 . To determine these variables, we need two equations. This assumes the presence of two, phase difference values between the elements of the antenna array. Therefore, the minimal number of the antenna array elements should be equal to three. Let the antenna system contain three elements, and the phase shifts between the first and the second 1,2 and between the first and the third 1,3 are measured. Then, the azimuth θ 0 and the elevation β 0 can be calculated from the following system of two nonlinear equations: 2π [(xn1 − xn2 ) cos θ0 − (yn1 − yn2 ) sin θ0 ] sin β0 = 1,2 ; λ 2π [(xn1 − xn2 ) cos θ0 − (yn1 − yn2 ) sin θ0 ] sin β0 = 1,3 λ

(8.45)

If the antenna elements are located in the plane at the apex of the isosceles rectangular triangle with leg length B, as shown in Fig. 8.28, the equation system (8.45) simplifies:

Phase and Correlation Interferometers

277

(x2,y2)

Fig. 8.28 Three element antenna system of phase interferometer

B 90°

(x1,y1)

B

2π B sin θ0 sin β0 = 1,2 ; λ 2π B cos θ0 sin β0 = 1,3 λ

(x3,y3)

(8.46)

Dividing the first equation by the second, we get tan θ0 =

1,2 . 1,3

(8.47)

Having raised both parts of the equations to the second power, and having summed the transformed equation, we get:

2

2π B λ

( sin2 θ0 + cos2 θ0 ) sin2 β0 = 21,2 + 21,3 .

(8.48)

From Equations (8.47) and (8.48), we get the final formulas for calculation of azimuth and elevation, for a three-element phase interferometer: 1,2 θ0 = arctan 1,3 ⎛/ ⎞ 2 1,2 + 21,3 ⎠ β0 = arcsin ⎝ 2πB

(8.49)

(8.50)

λ

At some values of azimuth and B/λ ratio, the phase shift on the antenna array elements may exceed 360◦ . As a result, bearing measurement ambiguity occurs. So, for phase incursion equal to, for instance, 20 and 380◦ , the phase meter will indicate similar values. Therefore, for unambiguous bearing, it is necessary that the phase difference does not exceed ±180◦ . Hence, the distance between the elements of the antenna array should not exceed the half-wavelength of the analyzed signal: B < λ/2. The larger the antenna system base is, the more accurately the phase difference on its elements can be measured, but we may increase this base up to B < λ/2 value.

278

8 Direction Finding of Radio Emission Sources

Thus, in wide-band phase direction finders, multi-base antenna systems are used, in which ambiguity elimination is provided with the help of additional rough meters with smaller bases. The “refinement method” is known as an approach to direction finder structure, when the unambiguous – but not accurate – bearing is determined with the help of a measuring instrument with a small base. After that, consequent “refinement” is provided by a measuring system with a large base. Another approach to multi-base phase interferometer construction is based on the statistical optimization of bearing calculation on the whole sample of measured phase shifts. Such optimization ensures the elimination of ambiguity, as well. The theoretical basis of this approach is the principle of maximal likelihood. In practice, three-element antenna configuration is often intensified by additional antenna elements that allow the possibility of antenna usage in wide frequency range. The most frequently applied structure of such antennas is the isosceles rectangular triangle or the circle array. The application of triangle matrices is restricted by frequency range up to 30 MHz [1]. At higher frequencies, the application of circle arrays is recommended because they ensure the same interconnection between antenna elements, and minimal antenna connection with the supporting mast. Due to central symmetry, we can provide the characteristics independent from the direction. When using the multi-element interferometer, the following possibilities appear: the usage of filled antenna groups: phase difference between the adjacent elements is always less than 180◦ , thus, we can avoid ambiguity; the usage of so-called “thinned out” antenna groups: at least, one element pair can have a phase difference larger than 180◦ .

Peculiarities of Correlation Interferometer The most effective way to eliminate the ambiguity of thinned out antenna arrays is the method used in the correlation interferometer. The basis of its operation is the comparison of measured phase differences between the antenna array elements with phase differences of the reference spatial signal (RSS), which are calculated theoretically at a given angle-of-arrival. The comparison is fulfilled by the calculation of RMS error, or the correlation factor, for two datasets: one obtained by measurement means and the other theoretical. The RSS theoretical dataset is necessary for all possible radio wave arrival directions. The bearing direction is considered an RSS direction for which the correlation factor is maximal for the measured data. The correlation interferometric meter (CIM) calculates the bearing value, based on the signal ensemble obtained from the similar elements of the antenna array. At the initial stages of direction-finding development, difficulties of implementation and characteristics checking of the identical receiving channels necessary for direction finding led to the creation of the single-channel (N = 1), bearing measuring systems. They are multi-channel (N = 1) by operation principle, but

Peculiarities of Correlation Interferometer

279

the physical channel division is changed in this configuration by the frequency or time multiplexing. However, the simplicity of eliminating bearing error related to non-identical receiving channels is accompanied by the appearance of another disadvantage. For example, at channel frequency multiplexing, single channel width inevitably decreases, while, at time multiplexing, the measurement time increases, and the bearing results will depend on the signal parameter varying for the measurement period. An interferometer with a digital receiver, whose number N of coherent channels is equal to the number M of antenna elements, realizes the mono-pulse method of direction finding. Such a direction finder ensures the highest rate of bearing calculation, but is complicated and expensive in manufacture and adjustment. The important problem, which is necessary to solve, is the fulfillment of identity requirements for the amplitude-frequency and phase-frequency responses of the receiving channels. Usually, this problem is solved by means of periodic calibration, via a low power, testing, probing signal acting in the receiving channels. The structural diagram of a N−channel correlation interferometer is similar to the structural diagram of a N−channel phase interferometer, which is shown in Fig. 8.26. For stationary and mobile direction-finding stations, the structure with two receiving channels has the widest distribution, at present. The structural diagram of such an interferometer is shown in Fig. 8.29. The main elements of CIM are: the antenna array, the antenna switch, the double-channel coherent receiver, and the analog-digital processing unit. The double-channel receiver has two inputs. The first one is referred to as signal, the second one – reference. The antenna switch consequently switches to the doublechannel receiver and inputs the pair of elements of the antenna array, which are

Antenna buffer 1

Receiver 1

ADC 1

Receiver 2

ADC 2

... Antenna element N

Antenna buffer 2

... Antenna buffer N

Antenna switch

Antenna element 2

Frequency synthesizer

Digital signal processing unit

Antenna element 1

Calibration generator

Antenna array with switch

Doublechannel receiver

Analog-digital signal processing unit

Fig. 8.29 Correlation interferometer with two receiving channels

PC

280

8 Direction Finding of Radio Emission Sources

selected in accordance with the direction-finding algorithm. To ensure coherent signal-receiving, the similar high-frequency voltage, formed by the frequency synthesizer (FS), passes to both channel mixers. The main functions of the double channel receiver are: frequency conversion of the received signal and primary filtering on the spurious receiving channels, i.e., preparation of the received-signal for conversion into digital form. In the analog-digital processing (ADP) unit, main computation operations are executed in accordance with the digital processing algorithm. The PC included in the interferometer structure controls the functions and displays the results. The functioning principle of the interferometer is based on the comparison of field phase in spatially spaced points, by determining the equal phase surface orientation, which is unambiguously related to the propagation direction from the RES. The increase in the number of pairs of spatially spaced points, differing by the distance of “diversity” or the base and by the angle orientation in space, allows expanded information about the structure of the received wave and, respectively, an increase in the formation quality of the angular spatial spectrum of the radio signal. Antenna elements, which are the electromagnetic field sensors, are located, as a rule, in the bearing plane (azimuth plane), which is related to the technical implementation possibilities of the identical phase centers of each antenna element. To ensure unambiguous direction finding in the circle zone, it is necessary to have, at least, three antenna elements located in the bearing plane. Taking into consideration the need for 360◦ scanning with similar measurement quality, the antenna array should be symmetrical with respect to its phase center. Equidistant single or multi-circular antenna arrays – with the central antenna element or without it – can be classified as the symmetrical plane antenna array type. The shortcoming of CIM with the double-channel receiver, compared to the N−channel version, is the larger time for bearing calculations (when providing the same instrumental accuracy and sensitivity). Nevertheless, the response time (for direction finding) of the double-channel CIM, as a rule, is completely acceptable for most applications. As will be shown below, application of the double-channel receiver with the common local oscillator, and with the simultaneous connection of antenna element pair to the channel pair, practically excludes the influence of direction-finding accuracy on the mutual difference of amplitude-frequency and phase-frequency responses of the receiving channels. Perfection of the main CIM characteristic can be achieved by increasing the size of the antenna array, along with an appropriate growth in the number of antenna elements. This cannot always be executed, however, for known reasons such as: the necessity to provide a high rate of space survey at wide-angle scan, the presence of restrictions on weight and size, radio system parameters, and so on. In this situation, the problem of optimal allocation of a restricted number of antenna elements, which will allow maximal information on electromagnetic field structure in antenna aperture to be obtained, arises. For the correlation interferometer, optimal allocation of antenna elements on the plane should provide: first, the formation of the maximal number of bearing pairs (BP) with different bases; second, different angular BP orientation with the same bases; and, third, the variation of BP bases

Algorithm of Correlation Interferometer Measuring System

281

with similar angular orientation, in accordance with geometric series law. Fulfillment of these mentioned conditions allows for unambiguous direction finding at the distance between the antenna elements, which is larger than half a wavelength of a radio signal.

Algorithm of Correlation Interferometer Measuring System Let consider the operation algorithm of a correlation interferometer measuring system for electromagnetic wave arrival [5, 6]. For the case of the circle antenna array, we present the complex amplitude (8.43) in equivalent form ( R E˙ n = E0 exp j 2π cos (θ0 − αn ) sin β0 + ϕ(t) + ϕ0 λ

(8.51)

n is the radius of the circle, on which the antenna elements are where Rn = R located; α n is the location angle of the n−th antenna array element counted out counter-clockwise from the x axis. In the general case, the antenna array may contain several circles of antenna elements. Radio signal superposition from the different RES is received by the antenna array elements and passes to the inputs of the antenna switch, which runs the signals from the selected antenna element pair through two panoramic receiver inputs. Radio signals, acting at the receiver inputs, are converted in the receiver to the intermediate frequency and, if necessary, to the video frequency. The bandwidth of the panoramic receiver’s simultaneous survey greatly exceeds the signal spectrum width of the separate RES. Therefore, a large number of radio channels may fall in the frequency band received by the radio receiver, as shown in Fig. 8.30. In the general case, the signals in these channels may have different spectrum width and can be classified to the different emission classes. k = 0 k = 1k = 2

Fig. 8.30 Radio signal spectra simultaneously received by the panoramic receiver fmin

k=3

Frequency band of simultaneous survey df

k = k max

fmax

Signals at the intermediate frequency pass from the receiver outputs to ADC inputs, where they are synchronously converted into digital signals with the length of Nn samples. Using the discrete Fourier transform (DFT), one obtains Nn complex spectral samples for each signal. In the future, to simplify the processing, Nn /2 complex samples of each spectrum are used only, and the other Nn /2 samples, corresponding

282

8 Direction Finding of Radio Emission Sources

to the negative frequencies, can be assumed as zero. The signal spectrum in the k−th radio channel corresponds to the signal of the k−th radio emission source. Taking into consideration the following relation [1]: E = U/he

(8.52)

where E is the electromagnetic field strength; U is the voltage measured at antenna output; and he is the effective length of the measuring antenna, the complex amplitudes of the k−th radio channel signal for the signal Zs (n1 ,t) and the reference Zref (n2 ,t) sections are equal to, respectively: Zs (n1 ,t) = he E0 K sin β0 × ( Rn1 × exp j 2π cos (θ0 − αn1 ) sin β0 + ϕ(t) + ϕ0 + ϕ λ

(8.53)

Zref (n2 ,t) = he E0 K sin β0 × ( Rn2 × exp j 2π cos (θ0 − αn2 ) sin β0 + ϕ(t)ϕ0 + ϕ λ

(8.54)

where he is the effective height of the antenna element for the k−th radio channel reduced to receiver input; E0 is the amplitude of the electromagnetic field strength for the radio signal in the k−th radio channel; K, K’ and ϕϕ’ are the gain factors and phase delays of the signal and reference receiver sections, respectively, for the k−th radio channel; n1 ,n2 are the numbers of antenna array elements (n2 = n1 ); αn1 ,αn2 are the location angles of the antenna array elements. The central frequency values fk of radio channels in the bandwidth df of simultaneous analysis, and the radio channel width dF, are known, as a result of fulfillment of the detection operation. The width of analysis df is defined by hardware implementation. Radio channel numbers {k} (1 ≤ k ≤ kmax , kmax = df /dF), in which the radio signals are detected, are stored. Each of these numbers corresponds to the values of the radio channel boundaries, which are recounted into the spectrum component numbers, taking into account the analysis bandwidth df, the volume Nn and width dF of the radio channel: for instance, at df=4 MHz and dF = 25 kHz, we get kmax = 160. If Nn = 4,000, we get 2,000 pairs of complex spectrum samples following through the interval F = Ndf = 2 kHz, as a result of disn /2 crete Fourier transform. For the single channel, the complex samples happen to be q = (dF/F + 1) = 13, and each of them consists of real and imaginary parts, or of magnitude and phase. Having assigned serial numbers to radio channel spectrum components, we get the sequences of spectrum samples numbers corresponding to the frequency bands of radio channels. For each radio channel, in which the signal is detected, the complex samples of the signal spectrum for the signal section Ss (k,i,n1 ) are multiplied by complex conjugated samples of signal spectrum for the reference section Sref (k,i,n2 ). The obtained products are added:

Algorithm of Correlation Interferometer Measuring System

A˙ n1,n2 =

∗ (k,i,n2 ), S˙ s (k,i,n1 )S˙ ref

283

(8.55)

i

where k is the radio channel number; 1 ≤ k ≤ kmax ; i is the spectrum sample number in the channel, i = 0,1,...,q − 1. As a result of the summing of the spectral sample products of the same name for the k−th radio channels, the spectral component is formed corresponding to the non-modulated carrier frequency of the radio signal. Taking into consideration Equations (8.53), (8.54), we get from (8.55) the interference signal vector: A˙ n1 ,n2 = (hE0 sin β0 )2 KK exp [j(n1,n2 + ϕ − ϕ )]

(8.56)

where the phase shift n1,n2 =

2π [Rn1 cos (θ0 − αn1 ) − Rn2 cos (θ0 − αn2 )] sin β0 . λ

(8.57)

Phase shift value n1,n2 depends on the direction of signal arrival, on the orientation angle γn1,n2 of the bearing pair, and on the base bn1,n2 between the n 1 −th and n 2 −th antenna elements. One of the necessary conditions for measurement accuracy is to provide the identity of complex transfer coefficients (CTC) of the receiving channels. In modern radio electronics development, the creation of the wide-band double-channel radio receiver with the similar channel CTC has been a complicated technical problem. In this connection, it seems expedient to use the double-channel radio receiver in CIM with signal and reference channels and with the common local oscillator. At that, to reduce the direction finding and the electromagnetic field strength dependence on the difference in frequency responses of signal and reference receiver channels, we suggest to apply the algorithm using the third antenna element not included in the panoramic receiver. This algorithm consists of three stages. At the first stage, the interference vector between the n 1 −th and n 3 −th antenna elements is determined: 0 S˙ s (k,i,n1 )S˙ ref (k,i,n3 ) = ∗ i Sref (k,i.n3 ) = (hE0 sin β0 )K exp [j(n1,n3 + ϕ − ϕ )]

A˙ n1,n3 =

∗

(8.58)

At the second stage, the interference vector between the n2 −th and n3 −th antenna elements is determined: 0 S˙ s (k,i,n2 )S˙ ref (k,i,n3 ) = ∗ i Sref (k,i.n3 ) = (hE0 sin β0 )K exp [j(n2,n3 + ϕ − ϕ )]

A˙ n2,n3 =

∗

(8.59)

And, finally, at the third stage, the resulting interference vector is calculated:

284

8 Direction Finding of Radio Emission Sources

A˙ n1,n2 = A˙ n1,n3 A˙ ∗n1,n3 = (hE0 K sin β0 )2 exp [j(n1,n2 )].

(8.60)

Application of such measurement procedure allows, at the first stage, the usage of one antenna array element as the antenna element connected to the reference channel of the receiver that simplifies its structure; at the second stage, it is possible to execute the calibration of the signal channel of the receiver only during the measurement of electromagnetic field strength. Weakening the requirements for the identity receiver channels allows the creation of the double-channel radio receiver with large dynamic range and high sensitivity, at less expense. In accordance with (8.60), the amplitudes of the interference vectors are proportional to the field strength square, and phases are defined by the base of appropriate bearing pairs and the spatial orientations of the lines connecting the n2 −th and n1 −th antenna elements, with respect to the front plane of the incident wave. Determination of the phase delays n1,n2 between the signals received by the antenna array elements allows the formation (synthesis) of a directed pattern of antenna array, the main lobe of which orients to the direction of radio signal receiving, which ensures the best conditions for radio signal receiving from various directions and the most accurate strength measurements of its electromagnetic fields. ˙ p of the bearing pair are the patterns of each Partial two-dimensional patterns D antenna pair formed by means of consecutive multiplication of the measured interference vector by the vector of the reference spatial signal (RSS) varying on the spatial angular coordinates. In our case, these patterns are defined as: ˙ p (Lθ dθ ,Lβ dβ) = (hE0 K sin β)2 × D 2π rn1 ( sin β0 cos (θ0 − αn1 ) − sin (Lβ dβ) cos (Lθ dθ − αn1 ))− × exp j λ ( −rn2 ( sin β0 cos (θ0 − αn2 ) − sin (Lβ dβ) cos (Lθ dθ − αn2 ))

(8.61)

where p is the partial pattern number for the panoramic receiver’s n1 −th and n2 −th antenna elements; p = 1,2,...P; p is the number of partial patterns of the antenna array; dθ = 2π/Lθ max is the step for RSS azimuth calculation; Lθ and Lθ max are current values and total number of points for RSS azimuth calculation; 0 ≤ Lθ ≤ Lθ max − 1; dβ = π/(2Lθ max ) is the step for calculation of RSS elevation; Lβ and Lβ max are current values and total number of points for RSS elevation calculation; 0 ≤ Lβ ≤ Lβ max − 1. Figure 8.31 shows the magnitude, and the real and imaginary parts of the partial pattern of the panoramic receiver. Antenna elements of the panoramic receiver are located on the x axis; the distance between them is equal to 2r. The direction of wave arrival counts out counter-clockwise from the y axis and is equal to θ = 60◦ . We can see from the figure that the magnitude and the real part of the pattern function take the maximal value for the angles corresponding to the direction to the RES, and at that, the imaginary part is equal to zero. The pattern has several maximums, which are evidence of the insufficiency of two antenna elements for unambiguous

Algorithm of Correlation Interferometer Measuring System

285

2 r = 0.25 λ 2 r = 0.5 λ 2r = 0.75 λ

0

270

90

180

Dp

–60

( )

0

60

2r = 0.25 λ

120

2r = 0.5 λ

180

240

Re Dp

–60

0

60

120

180

2r = 0.75 λ

θ, degree

θ, degree

240

( )

Im Dp

–60

0

60

120

180

240

θ, degree

Fig. 8.31 Partial pattern. Direction θ = 60◦

bearing determination. When decreasing the wavelength, the ambiguity of direction finding grows. To eliminate the ambiguity of determining the direction of signal arrival, one uses the combination of several patterns. At that, the summing of the partial panoramic receiver patterns is often used, which corresponds to the known approach of pattern direct synthesis. The multiplication of the partial panoramic receiver patterns is used

286

8 Direction Finding of Radio Emission Sources

as well as the method contained in the summing of the partial panoramic receiver patterns following its multiplication. Resulting patterns of the antenna arrays for the first, second and third synthesis methods are defined by the following equations: ˙ A (Lθ dθ ,Lβ dβ) = D

P

˙ p (Lθ dθ ,Lβ dβ); D

(8.62)

˙ p (Lθ dθ ,Lβ dβ); D

(8.63)

p=1

˙ B (Lθ dθ ,Lβ dβ) = D

P

p=1

˙ C (Lθ dθ ,Lβ dβ) = D

Pn N

˙ pn (Lθ dθ ,Lβ dβ), D

(8.64)

n=1 pn =1

where pn’ is the number of the partial panoramic receiver pattern of the n −th antenna array. The estimation of RES azimuth and elevation reduces to the maximum search ˙ B , or D ˙ C , depending on the syn˙ A , D in two-dimensional array of patterns D thesis method. The dependence of output signal power on the angular position (Lθ dθ , Lβ dβ) of the reference direction is used as the output function. Figure 8.32 shows the variation of the synthesized pattern form of the four-element antenna array (three antenna elements along the circle, one in the center) versus RES elevation. The ratio of antenna array circle radius to wavelength is r/λ = 0.45, and elevation is 0, 30, and 60◦ . Increasing the element number of the antenna array leads to the pattern main lobe narrowing. Figure 8.33 shows the synthesized pattern images of the correlation interferometer with an eight-element antenna array. The array circle radius is r = 0.5 m, seven elements are located along the circle and one is in the center. Azimuth of signal arrival was fixed: θ = 0, elevation possessed the values 0, 30◦ , and 60◦ , the frequency was 250 and 500 MHz. As seen in Fig. 8.34, with frequency growth an increase of the side lobe level of the correlation curve takes place, which may be a reason for erroneous direction finding in the upper part of the operation frequency range of the direction finder. Moreover, with frequency growth the side lobe number increases, due to which the probability of bearing overthrow to the erroneous direction is increased. But at the same time, with frequency increase the correlation curve’s main lobe narrows, which leads to a reducing of the bearing calculation error. The magnitude of the resulting pattern ˙ C (Lθ dθ ,Lβ dβ) q(Lθ dθ ,Lβ dβ) = D

(8.65)

is used for the determination of radio wave arrival direction. Its maximal value corresponds to this direction:

Algorithm of Correlation Interferometer Measuring System Fig. 8.32 Pattern form of the four-element antenna array versus RES elevation: (a) β = 0; (b) β = 30◦ ; (c) β = 60◦

a)

b)

c)

287

288

8 Direction Finding of Radio Emission Sources

a)

b)

c) Fig. 8.33 Pattern form of eight-element antenna array versus RES elevation for f = 250 MHz (left) and f = 500 MHz (right): (a) β = 0 , (b) β = 30◦ , (c) β = 60◦

S(θ ,β) = arg Q = arg{max q(Lθ dθ ,Lβ dβ)}

(8.66)

To implement the CIM, for determination of the electromagnetic wave arrival direction, it is necessary to execute the following operations: 1. Obtaining of synchronous-in-time samples of the analyzed radio signal received by two antenna elements.

Single-Channel Measuring System on the Basis of a Correlation Interferometer

289

Fig. 8.34 Vector diagram of signals

2. Representation of obtained signal samples from the time domain to the frequency domain (providing the Fourier transform operation on the obtained samples). 3. Calculation of the complex convolutions among the spectrum components with the same numbers for antenna element pairs, and obtaining the interference vector of signal pair by formula (8.60). 4. Repeating the steps 1–3 for all bearing pairs. 5. Calculation of the partial patterns of all bearing pairs in accordance with (8.66). 6. Synthesis of antenna array pattern with the help of (8.62), (8.63) and (8.66). 7. Calculation of the RES bearing in accordance with (8.66). Since the partial patterns depend on the electromagnetic field strength, the estimation procedure for the propagation direction can be combined with the measurement of its strength. A similar measuring system is considered in [7].

Single-Channel Measuring System on the Basis of a Correlation Interferometer At present, the CIM structure with a N−channel (N ≥ 2) coherent receiver is the most distributed piece of equipment for stationary and mobile radio monitoring stations. Possessing high noise immunity, wide operation range, high accuracy and operation rate, such CIM are rather complicated by their multi-channel analog section, and have relatively large weights and sizes, which causes trouble for their use in portable and handheld equipment. The implementation difficulties of identical receiving channels, necessary for direction finding, have led to the development of single-channel direction finders. They are also multi-channel, on the operation

290

8 Direction Finding of Radio Emission Sources

principle (N = 1), but physical channel separation is changed in its structure by the frequency or time multiplexing. Let us consider the operation algorithm of the direction finder with a single radio receiving section, which allows the calculation of the phase shifts n2,n1 on the bearing pairs of the antenna array elements, by means of measuring the receiving signal’s amplitudes and its combinations [8]. We assume that antenna array elements have the same properties. In this case, the voltage un (t) acting at its output corresponds to the field strengths en (t) in the phase center of the n−th element: un (t) = Un (t) cos [ωt + n + ϕ(t) + ϕ0 ].

(8.67)

Now we add the oscillations un1 and un2 from the outputs of antenna array elements. Since the frequencies of these oscillations are the same, the amplitude A1 of the resulting signal un1 + un2 , in accordance with the vector diagram shown in Fig. 8.34, is equal to: A1 =

/ 2 + U 2 − 2U U cos (π + + ) Un1 n1 n2 n1 n2 n2 (8.68)

/ 2 + U 2 + 2U U cos ( = Un1 n1 n2 n1,n2 ) n2

where n2,n1 is the phase shift between the signals on the antenna array elements. From Equation (8.68) we get the value of phase shift between the signals acting on the antenna array elements, with accuracy up to the sign:

n2,n1

2 − U2 A21 − Un1 n2 = ± arccos 2Un1 Un2

.

(8.69)

To eliminate the ambiguity of phase shift determination, let us apply the Gilbert transform to the signal: uˆ n1 (t) = H[un1 (t)] = H{Un1 (t) cos [ωt + n1 + ϕ(t) + ϕ0 ]}.

(8.70)

Since H[ cos ωt] = sin ωt and the slowly-varying multiplier Un1 (t) can be taken under the sign of the Gilbert transform, we get: uˆ n1 (t) = H[un1 (t)] = Un1 (t) sin [ωt + n1 + ϕ(t) + ϕ0 ].

(8.71)

After adding uˆ n1 (t) and un2 (t), the summed oscillation amplitude / 2 + 2U U cos ( U 2 + Un2 n1 n2 n1,n2 + π/2) / n1 2 + U 2 − 2U U sin ( = Un1 n1 n2 n1,n2 ) n2

A2 =

From the last equation, we obtain

(8.72)

Single-Channel Measuring System on the Basis of a Correlation Interferometer

291

2 2 2Un1 Un2 sin (n2,n1 ) = Un1 + Un2 − A22

(8.73)

and then sin (n2,n1 ) =

2 + U 2 − A2 Un1 n2 2 . 2Un1 Un2

(8.74)

Equations (8.69) and (8.74) allow unambiguous determination of the required phase difference n2,n1 . To determine two possible phase values n2,n1 and −n2,n1 , we use Equation (8.69), the value of which does not depend on the phase turn error, which can occur at technical implementation of the Gilbert transform. Since sinus is the odd function, the formula can be rewritten in such form: 2 − U2 A21 − Un1 n2 2 2 + Un2 − A21 ) (8.75) sgn(Un1 n2n1 = arccos 2Un1 Un2

1, x ≥ 0, −1x < 0. After the n2,n1 calculation to define the direction of electromagnetic wave arrival, the algorithm CIM of angle-of-arrival is used. Structural diagram of the direction finder is shown in Fig. 8.35. where sgn(x) =

Antenna element 1 Antenna buffer 1

Antenna buffer N Calibration generator

Antenna array with switch

Phase shifter Switch 2 Switch 3

Switch 5

... Antenna element N

Antenna switch

...

Antenna buffer 2

Switch 4

Receiver

Analog-digital converter

Analog-digital signal processing unit

Switch 1

Antenna element 2

PC

Frequency synthesizer

Summation unit

Signal formation unit

Single-channel receiver

Analog-digital signal processing unit

Fig. 8.35 Structural diagram of direction finder with single receiving section

The direction finder has, in its structure, the antenna array with the identical nondirectional antenna elements and the switch, signal generation unit (SGU), the single-channel receiver and the single-channel ADP unit. The calculation algorithm for the interference vector of the bearing pair reduces to the following operation sequence: 1. Measurement of amplitude Un1 of the n1 −th antenna array element is executed. For this, the signal from the antenna switch output passes to switch 1, then to

292

2.

3.

4.

5.

8 Direction Finding of Radio Emission Sources

switch 2, then to switch 3 and switch 5. From the output of switch 5, the signal passes to the input of the single-channel panoramic receiver. Measurement of amplitude Un2 of the n2 −th antenna array element is executed. For this, the signal from the second antenna switch output passes to switch 5 and from its output to the input of the panoramic receiver. The summed signal amplitude of the antenna element pair is determined. For this, the signal from the n1 − th antenna element passes through the antenna switch, and switches 1 and 3, to the one input of the summation unit, while the signal from the n2 − 5. th antenna element passes to the second summation unit input through the second antenna switch output and then through switch 4. The summed signal passes to the input of the panoramic receiver. The summed signal amplitude from the n2 −th antenna element and the signal from the n1 −th antenna element, which passed through the phase shifter (PS) for 90◦ is determined. For this, the signal of the n1 −th antenna element from the output of the antenna switch passes to switch 1, then to the PS, then via switches 2 and 3 to the summation unit input. The phase difference n2,n1 is calculated by formula (8.75) and the interference vector is formed A˙ n2,n1 = Un2 Un1 exp (jn2,n1 ).

(8.76)

In a similar manner, the interference vectors for the other pairs of antenna elements are calculated. On the basis of obtained interference vectors, the partial patterns are formed and then the antenna array pattern is synthesized, which is used for calculation of the radio signal arrival direction. Let us compare the considered radio direction finder – with the single receiving section – to the CIM, having two coherent receiving channels. To calculate the bearing, we use the algorithm with the summing of partial patterns: ˙ θ dθ ) = D(L

P

˙ p (Lθ dθ ) D

(8.77)

p=1

˙ is the resulting pattern, D ˙ p is the pattern of the bearing pair, P is the number where D of partial patterns of the antenna array, dθ = 2π/Lθ max is the step of azimuth calculation, and Lθ and Lθ max are the current value and the total point number for azimuth calculation. The comparison is provided by two indices: RMS error of bearing σ and the probability of anomalous error Pan . RMS error can be calculated by the formula σ =

/

(θm − θ0 )2

(8.78)

where ... signifies the operation of averaging on the simulation iteration number; θm is the azimuth estimation obtained by the calculation results; θ0 is the true azimuth value. Estimation of the probability of anomalous bearing determination is

Single-Channel Measuring System on the Basis of a Correlation Interferometer

293

represented by Pan = Nan /N, where Nan is the number of anomalous bearing calculations; N is the total number of iterations. Here, the anomalous bearing calculation means the bearing result, when, due to interference action, the side lobe level of the antenna array pattern exceeds the main lobe level, and the bearing estimation corresponds to the side lobe. Figure 8.36 shows the bearing RMS error versus the ratio r/λ of the antenna array radius r to wavelength λ of the received radio signal. The curves 1–3 are plotted for the CIM with the double, coherent, radio receiver sections, and curves 4–6 represent the CIM with the single radio receiver section. The element number of the circle antenna array N = 5. When simulating, the signal arrival azimuth was changed from 0 to 359◦ with 1◦ -step. For each direction, the bearing estimation was provided 1,000 times. The signal/noise ratio in the signal bandwidth was changed from 10 to 20 dB. SNR was changed from 0.03 to 2. RMS error, degree

40

30

6 5

4 3

20 2 10 1

0

0.25

0.75

1.25

1.75

r

λ

Fig. 8.36 RMS bearing error for circle antenna array with five antenna elements: 1, 2 and 3 – CIM with double-channel receiver at SNR 20, 15 and 10 dB; 4, 5 and 6 – CIM with single-channel receiver at SNR 20, 15 and 10 dB

As follows from the obtained curves, the direction finder with the single radio receiving section is more exposed to the interference action and has less operating frequency range than the double-channel interferometer. Thus, at SNR = 10 dB, RMS error for the single-channel direction finder exceeds 10◦ in the whole range of r/λ ratio-varying. At SNR growth, the accuracy of the direction finder operation is improved. At SNR = 15 dB, RMS error does not exceed 5◦ in the range of r/λ values from 0.27 to 0.49, and, at SNR = 20 dB, the direction finder range is even greater: from 0.25 to 0.77. The double-channel direction finder r/λ range, in which RMS error does not exceed 5◦ , is wider still.

294

8 Direction Finding of Radio Emission Sources

One way to increase bearing accuracy and widen the operating range is via the growth of the antenna array element number. This is confirmed in Fig. 8.37, where the curves are plotted for the antenna array with the element number N = 7, similar to Fig. 8.34. At SNR = 15 dB, RMS error does not exceed 5◦ , even in the r/λ range from 0.26 to 1.25. Thus, the addition of two antenna elements leads to a more than four-fold widening of the operating frequency range. RMS error, degree

6 5

20 4 3 2 10

0

1

0.25

0.75

1.25

1.75

r

λ

Fig. 8.37 RMS bearing error for circular antenna array with seven antenna elements: 1, 2 and 3 – CIM with double-channel receiver at SNR 20, 15, and 10 dB; 4, 5 and 6 – CIM with single-channel receiver at SNR 20, 15, and 10 dB

Figure 8.38 shows the probability of anomalous bearing determination Pan versus r/λ ratio for the antenna array at N = 5. Comparison of figures shows that the probability of anomalous error Pan (r/λ) curves repeats on its form the curves of RMS error σ (r/λ); hence, the degradation of the direction finding quality of the single-channel direction finder, with SNR decreasing, is caused by the anomalous bearing presence. The operation algorithm of the direction finder with the single radio receiving section is based on the fact that the radio signal phase distribution received by the antenna array elements, necessary for the antenna array pattern formation, is defined by the measurement of signal amplitude, received by antenna elements, and their combinations. For fulfillment of necessary operations upon the signals, it is not required to execute the synchronous transformations of signal pairs with the further direct measurement of phase differences, and, relatively, there is no need to use the complicated (at minimum, double-channel) receiver with the common local oscillator, for the receiving and signal frequency conversion. The shortcoming of the considered direction finder, compared to the correlation interferometer with the double-channel receiver, is a four-fold time expense, increasing per one bearing cycle, and less operating frequency range at operation under interference conditions. Nevertheless, the construction simplification, reduction of

ARTIKUL-M4 Foldable Correlation Interferometer

295

P an 0.18

0.14

0.1 1 0.06

2

3

4

0.02 0

0.25

0.75

1.25

r

λ

Fig. 8.38 Probability of anomalous bearing determination for circular antenna array with five antenna elements: 1, 2 and 3 – CIM with double-channel receiver at SNR 10, 15, and 20 dB; 4 – CIM with single-channel receiver at SNR 10 dB

equipment weight and size, and power consumption reduction make expedient the application of the mentioned single-channel method in cheap, portable, and handheld equipment.

ARTIKUL-M4 Foldable Correlation Interferometer For stationary direction-finding posts, the ARTIKUL-M4 radio direction finder is used in the ARK-POM1 system. This direction finder can be also used in ARKPOM2 mobile systems, for operation in parking places (without movement). The radio direction finder consists of the antenna-receiver unit (ARU), ADP unit, power source unit, and PC. The direction finder provides the radio signal bearing in the range from 25 to 3,000 MHz. The functional diagram of the radio direction finder is shown in Fig. 8.39. The external view of the antenna-receiver unit with foldable antenna elements is shown in Fig. 8.40. The antenna-receiver unit contains the antenna switch and tuners unit. Two circle antenna arrays of the first and second ranges are included in the ARU structure. The antenna array of the first range operates in the frequency range of 25–1,000 MHz, the antenna array of the second range – in the frequency range of 1,000–3,000 MHz. Antenna elements of the first range are mounted on the fold-down cross-arm, while the antenna array of the second range is under the plastic radome in the upper part of the ARU.

296

8 Direction Finding of Radio Emission Sources

Frequency converter ARK-PS5

Antenna element1 Antenna element2 ... Antenna elementN

Antenna switch, range 1

Frequency converter ARK-PS5

Signal processing unit

Antenna element1

PC

Input switch

Antenna element2 ... Antenna elementN

Protection and synchronization unit

Antenna switch, range 2

Control unit Heater

Control unit Antennas switching unit Tuner unit

Analogdigital signal processing unit

Fig. 8.39 Functional diagram of ARTIKUL-M4 direction finder

Fig. 8.40 Antenna-receiver unit of ARTIKUL-M4 direction finder: 1 – Antenna array 1,000–3,000 MHz; 2 – Fold-down cross-arm; 3 – Antenna element 22.5–1,000 MHz; 4 – ARU case; 5 – Mounting strap

Antenna element hinging and the rotated cross-arms accelerate deployment of the antenna system. Essentially, deployment consists of attaching the ARU landing

ARTIKUL-M4 Foldable Correlation Interferometer

297

socket to the mast tang until it matches the transverse hole, then fastening the antenna on the mast, by the stopping pin, and turning the antenna cross-arm by 90◦ until it is fixed on the unit base. At that, the antenna elements always maintain the vertical position. To mount the direction finder, we recommend using the ARK-MT3 telescopic mast, which has small weight and a reliable two-loop raising system to avoid dropping the ARU if/when one rope becomes damaged. The mast can be fixed on the ground, on a building roof, or attached to a car, with the help of special brackets (see section “Electric Power Supply Systems”). The ARTIKUL-M4 radio direction finder mounted on a mast is shown in Fig. 8.41. The ARK-CT3 or ARK-CT6 tuner unit (see sections below) is located inside the ARU case. Placing the analog part of the digital radio receiver (DDR) inside the ARU increases the direction finder’s sensitivity in the upper part of the frequency range, at the expense of minimizing the high-frequency cable length, and eliminates the antenna effect. Moreover, since the IF signal from the DDR output has a, relatively, low frequency, it becomes possible to use a down-lead cable of some hundred meters in length. AU efficiency is best in temperatures ranging from –50 to +55◦ C. At below zero temperature, the automatic built-in heater inside the ARU case is activated, while, at positive temperatures, the mandatory ventilation system is activated. The double-channel signals, filtered, amplified, and converted by the tuners, pass at an intermediate frequency of 41.6 MHz into the ADP unit. The analog-digital processing unit of this direction finder is shown in Fig. 8.42. The main part of this unit is the double-channel module for ADP, constructed on the basis of the ARK-CO2 module and the ARK-C5 double-channel specific calculator. Communication with the controlling PC is executed via a USB 2.0 serial interface. The data transmission rate from the ADP unit to the external PC is not less than 40 Mbyte/s. The technical parameters of the ARTIKUL-M4 direction finder are listed in Table 8.3. Curves representing the accuracy and sensitivity of the ARTIKUL-M4 direction finder versus frequency are shown in Figs. 8.43, and 8.44. These curves are plotted as a result of the testing of several units, mounted on a mast with a height of 9 m. The average time on each frequency does not exceed 50 ms. As we see from the figures, RMS bearing error for the ARTIKUL-M4 mobile direction finder is on average 1–1.5◦ . The field sensitivity depends on the frequency of the received signal. At the lower boundary of the operating frequency range, the sensitivity is 6 μV/m, then, in the 100–300 MHz range, the sensitivity improves to 1 μV/m; at the upper boundary frequency of the first antenna range, the sensitivity becomes 6 μV/m. When transferred to the second antenna frequency range, the sensitivity is improved up to 2 μV/m, then its value monotonically becomes worse, and, on the upper frequency boundary, it is near 12 μV/m.

298

Fig. 8.41 ARTIKUL-M4 mounted on a mast

8 Direction Finding of Radio Emission Sources

ARTIKUL-M1 Mobile Direction Finder

299

Fig. 8.42. Analog-digital processing unit: 1 – Unit for signal processing; 2 – Control unit for ARU; 3 – Protection and synchronization unit; 4 – Toggle switch for power supply; 5 – Connector for synchro-signal connection; 6 – Fuse holder; 7 – Connector for power source connection; 8 – Grounding clamp; 9, 10 – Connectors for connection with ARU; 11 – Clamp for unit extraction; 12 – USB connectors for external connections; 13 – USB connector for the controller; 14 – Connector for the head-phones; 15 – Audio output RMS error, degree

8 6 4 2

0

500

1500

2500 Frequency, MHz

Fig. 8.43 RMS bearing error of ARTIKUL-M4 radio direction finder versus frequency

ARTIKUL-M1 Mobile Direction Finder For operation while moving, on mobile direction-finding posts, we recommend the ARTIKUL-M1 direction finder. The ARTIKUL-M4 direction finder provides the direction finding of radio signals in the frequency range from 25 to 3,000 MHz. Its antenna system also has two circular antenna arrays. The antenna array for the first frequency range operates from 25 to 1,000 MHz, the array of the second range operates in 1,000–3,000 MHz. The antenna system of this direction finder may have two implementation versions: the AS-MK1M, in local, removable, radio-transparent radome, with the possibility of urgent mounting on the car roof (Fig. 8.45) and the

300

8 Direction Finding of Radio Emission Sources Table 8.3 Technical parameters of ARTIKUL-M4 direction finder

Parameter

Value

Operating frequency range, MHz Receiving section field sensitivity in the band 12.5 kHz, not worse μV/m: In range 25–100 MHz In range 100–1,000 MHz In range 1,000–3,000 MHz Instrumental RMS error of bearing measurement, not more, degrees: In range 25–100 MHz In range 100–1,000 MHz In range 1,000–3,000 MHz Survey rate without the direction finding, MHz/s Survey rate with the direction finding, not less, MHz/s Selectivity on the adjacent and spurious channels, not less, dB Dynamic range on the third order intermodulation, not less, dB Bandwidth of instantaneous survey of frequency range, MHz DC power voltage, V Consumed current, not more, A Weights, not more, kg

25–3,000

Fig. 8.44 Sensitivity of ARTIKUL-M4 direction finder versus frequency

15 5 20

3 2 2 3,000 300 70 75 5 27 (+3, –4.5) 7 40

Field strength, μV/m

16 12 8 4 0

500

1500

2500 Frequency, MHz

AS-MK6, in non-removable, radio-transparent radome, as part of the car structure (Fig. 8.46). The main features of the ARTIKUL-M1 radio direction finder are listed in Table 8.4. Curves representing the accuracy and sensitivity of the ARTIKUL-M1 radio direction finder versus frequency are shown in Figs. 8.47 and 8.48. These curves are plotted as a result of the testing of some units, mounted on a minivan. The degradation of bearing accuracy near the 100 MHz frequency range is caused by the vehicle body influence, since the radio deviation correction was not executed. As we see from the figures, RMS bearing error of the ARTIKUL-M1 direction finder is, on average, 0.5% worse than for the ARTIKUL-M4 direction finder; it also

ARTIKUL-P Portable Foldable Direction Finder

301

Fig. 8.45 AS-MP1 antenna system of ARTIKUL-M1 direction finder mounted on the car roof

Fig. 8.46 ARTIKUL-M1 mobile radio direction finder with the antenna system in non-removable, radio-transparent radome

has a small loss in sensitivity. Mentioned degradations are caused by the antenna array’s compact structure and the vehicle body influence.

ARTIKUL-P Portable Foldable Direction Finder Portable equipment for direction finding can be created on the basis of the ARTIKUL-P foldable radio direction finder, intended for mounting on a folding mast. The operation frequency range of this direction finder is from 25 to 1,300 MHz.

302

8 Direction Finding of Radio Emission Sources Table 8.4 Main features of ARTIKUL-M1 radio direction finder

Parameter

Value

Operating frequency range, MHz

25–3,000

Receiving section field sensitivity in the bandwidth 12.5 kHz, not worse μV/m: In range 25–100 MHz In range 100–1,000 MHz In range 1,000–3,000 MHz Instrumental RMS error for bearing measurement, not more, degrees: In range 25–100 MHz In range 100–1,000 MHz In range 1,000–3,000 MHz Survey rate without the direction finding, MHz/s Survey rate with the direction finding, not less, MHz/s Selectivity on adjacent and spurious channels, not less, dB Dynamic range on third order intermodulation, not less, dB Bandwidth of instantaneous survey of frequency range, MHz DC power source voltage, V Consumed current, not more, A Weight, not more, kg

25 10 30

5 2 3 3,000 300 70 75 5 27(+3, –4.5) 2 40

RMS error, degree 9 7 5 3 1 0

500

1500

2500 Frequency, MHz

Fig. 8.47 RMS bearing error versus frequency for ARTIKUL-M1 direction finder

During transportation, the antenna system is folded and is housed in a rigid cover equipped with carrying belts (Fig. 8.49). In its collapsed state, the antenna system diameter is equal to 48 cm. Before 2005, this direction finder was completed by the addition of the ARKCT2 receiver. The external view of this receiver is shown in Fig. 8.50. The case is equipped with the mandatory ventilation and heating systems to provide for receiver operation in the temperature range from –50 to +50◦ C. The receiver, in its protective cover, is mounted on the first part of the mast, under the antenna system. The signal on the radio frequency is transferred via the down-lead cable from the antenna system to the receiver. In turn, the amplified signal from the receiver passes to the ADP

ARTIKUL-P Portable Foldable Direction Finder Fig. 8.48 Sensitivity of ARTIKUL-M1 radio direction finder versus frequency

303

Field strength, μV/m

18 14 10 6 2 0

500

1500

Fig. 8.49 Antenna system packed in the cover for transportation

2500 Frequency, MHz

304

8 Direction Finding of Radio Emission Sources

unit (see Fig. 8.51) on the relatively low intermediate frequency, which reduces the losses in the connecting cable, decreases the noise factor of the receiving section, and diminishes the antenna effect of the cable.

Fig. 8.50 ARK-CT2 receiver in protective cover

Fig. 8.51 ARK-ACO-MK7 analog-digital processing unit

Moving the receiver location away from the antenna is possible using the cable at a length of up to several hundred meters. The ARK-ACO-MK7 ADP unit included in the post structure is shown in Fig. 8.51. The ADP unit can operate in two modes: in automatic mode under PC control, and in autonomous mode without the PC. In automatic mode, the PC control is executed via parallel interface, and the SMO-PPK software controls the equipment operation. If the PC is not connected to the ADP unit, after activation of the power supply, the direction finder transfers to autonomous operation. The possible regimes in the autonomous mode are listed in Table 8.5. In autonomous mode, the processing results, the bearing, spectrum, correlation curve, and other parameters are indicated on the LCD. As an example, Figs. 8.52 and 8.53 show the LCD view in “Spectrum in 0.5 MHz” and “Bearing” modes. As we see, the tuning frequency of the equipment is equal to 337 MHz, and the demodulator of the signal with narrow-band frequency modulation (NFM) is activated; for finding direction at selected band 25 kHz, the azimuth to the RES is equal to 182◦ . Beginning from 2005, instead of the ARK-CT2 radio receiver, the directionfinding post is completed by the ARGAMAK double-channel radio receiver (see section “ARGAMAK-I Panoramic Measuring Receiver”). The receiver’s small size accommodates it together with a reserved rechargeable battery with capacity 7 A · h,

ARTIKUL-P Portable Foldable Direction Finder

305

Table 8.5 Operation regimes of ARTIKUL-P radio direction finder without PC Regime number

Regime name

0 1 2 3 4 5 6 7 8 9

Adjustment Spectrum in 2 MHz band Averaged spectrum in 2 MHz band Peak spectrum in 2 MHz band Spectrum in 0.5 MHz band Averaged spectrum in 0.5 MHz band Peak spectrum in 0.5 MHz band Bearing (in accordance with chosen algorithm) Regime of diagnostics Radio modem (optional)

Fig. 8.52 “Spectrum in 0.5 MHz band” mode

Fig. 8.53 “Bearing” mode

inside of a shock-proof, damp-proof, case, with the sizes 268 × 123 × 247 mm. At that, the case weight, together with the battery, does not exceed 5 kg. The power of the rechargeable battery is enough for 3 h of continuous operation of the direction finder. The main parameters of the ARTIKUL-P direction finder are listed in Table 8.6. Figures 8.54 and 8.55 show the accuracy and the sensitivity of the ARTIKUL-P direction finder versus frequency. As we see, in the operating frequency range, this direction finder has parameters that are highly competitive with the parameters of the ARTIKUL-M4 equipment.

306

8 Direction Finding of Radio Emission Sources

Table 8.6 Main parameters of ARTIKUL-P direction finder (on the basis of the ARGAMAK receiver) Parameter

Value

Operating frequency range, MHz

25–1,300

Receiver section field sensitivity in the band 12.5 kHz, not worse μV/m: In range 25–100 MHz

25 10 20

In range 100–1,000 MHz In range 1,000–1,300 MHz Instrumental RMS error of bearing measurement, not more, degrees: In range 25–100 MHz

5 2 1.5 140 50 70 70 5 27 (+3, –4.5) 1.5 25 16

In range 100–1,000 MHz In range 1,000–1,300 MHz Survey rate without direction finding, MHz/s Survey rate with the direction finding, MHz/s Selectivity on adjacent and spurious channels, dB Dynamic range on the third-order intermodulation, not less, dB Instantaneous survey band of frequency range, MHz DC power supply voltage, V Consumed current, not more, A Weight, not more, kg Weight of transportation cover, not more, kg

Fig. 8.54 RMS bearing error of ARTIKUL-P versus frequency

RMS error, degree 9 7 5 3 1 0

500

1000 Frequency, MHz

ARTIKUL-P11 Portable Foldable Direction Finder The most recent modification of the portable direction finder is the ARTIKUL-P11. This new direction finder antenna system is combined constructively with a telescopic mast, which essentially decreases the time for its deployment. In spite of the fact that the mast is added to the antenna structure, the weight of the cover required for transportation is reduced. Specially-designed plane receiver elements are developed in the antenna system, which are attached to the base case, with the

Direction Finding Error Correction in Mobile Systems Fig. 8.55 ARTIKUL-P direction finder sensitivity versus frequency

307 Field strength, μ V/m

14 10 6 2 0

500

1000 Frequency, MHz

help of turned cross-arms. The plane elements allow the reduction of the antenna’s diameter – in folded form – up to 320 mm. In deployed form, the height of the ARTIKUL-P11 antenna system is 4 m, and the transportation cover is used as a mast base. Figure 8.56 shows the antenna system in deployed form. Figure 8.57– in folded form for the transportation. Near for example the ARTIKUL-P antenna system is shown. The ARGAMAK-T2 double-channel radio receiver is mounted directly in the antenna system case. In addition to improved sensitivity, this reduces the time for antenna system deployment, since there is no need to mount the receiver to the mast base and to set up the cable connections between the receiver and the antenna system. In antenna system structure, there is the possibility of additional easy-off module-mounting for the second range, expanding the direction finder operating frequency range up to 3 GHz. Thus, the usage of the ARTIKUL-P11 antenna system with the ARGAMAK-T2 built-in double-channel coherent receiver allows the frequency range to be expanded up to 3 GHz for the portable direction finder, to decrease its weight and size and reduce the time for its deployment. The main parameters of the ARTIKUL-P11 direction finder are listed in Table 8.7.

Direction Finding Error Correction in Mobile Systems If the antenna system is located on the vehicle roof, as a result of the backward emission at radio wave reflection from its parts, the antenna system pattern distorts, which leads to the growth of direction-finding errors, which may achieve 15–20◦ at some azimuth and frequencies. Vehicle body influence on direction-finding errors can be reduced to the problem of electromagnetic field diffraction on the conducting object. However, if the object has a complicated form, the theoretical solution of this problem leads to very bulky calculations. Also, the theoretical calculation does not always yield true

308

8 Direction Finding of Radio Emission Sources

Fig. 8.56 ARTIKUL-P11 antenna system in deployed state

results, since sometimes it is impossible to take into consideration all individual peculiarities of the form and conductive properties of the vehicle, for instance, the size and exact form of its windows, doors, the roof peculiarities, etc. The situation becomes more complicated when the carrier has elements that may cause resonance phenomena, for example, the mounting brackets for the antenna array, the ventilation hatches, and the communication system antennas. Direction-finding errors at resonance frequencies may exceed 30◦ . That is why, under conditions of strong vehicle body influence, methods based on experimental data application – measured for the specific antenna system on the specific car – are more preferable. Sometimes, to increase the accuracy of the mobile direction finder, we may use the direction-finding errors versus signal frequency. One may insert the amendments into the bearing value, on the basis of the obtained error values. Unfortunately, this approach does not always lead to good results since the direction-finding errors for the signals arrived from the closed azimuth, often have opposite signs, have the bearing dependence on frequency, were obtained for different RES azimuth, perhaps even intersected, and, therefore, further correction becomes problematic. Let us analyze Fig. 8.58, which shows the bearing values versus signal frequency, obtained experimentally for one real mobile system. As we see, the bearing curves

Direction Finding Error Correction in Mobile Systems

309

Fig. 8.57 ARTIKUL-P11 antenna systems in folded form

are not only displaced with respect to true azimuth – sometimes up to 10◦ , but, in addition, even intersect for some directions in the range of 76–86 MHz, which leads to the impossibility of unambiguous bearing correction on the basis of these curve applications. Therefore, the presence of even measured-in-depth curves does not always ensure the true correction of direction-finding errors. Hence, the resulting bearing curves should not be used as the initial data for correction, rather, for this purpose, one should use the initial information on which the bearing values are calculated in the bearing algorithm. For direction-finding methods using the antenna arrays, the multi-dimensional, amplitude-phase, field distri-

310

8 Direction Finding of Radio Emission Sources Table 8.7 Main parameters of ARTIKUL-P11 direction finder

Parameter

Value

Operating frequency range, MHz

25–3,000

Receiver section field sensitivity in the band 12.5 kHz, not worse μV/m: In range 25–100 MHz In range 100–1,000 MHz In range 1,000–3,000 MHz Instrumental RMS error of bearing measurement, not more, degrees: In range 25–100 MHz In range 100–1,000 MHz In range 1,000–3000 MHz Survey rate without direction finding, MHz/s Survey rate with direction finding, MHz/s Selectivity on adjacent and spurious channels, not less, dB Dynamic range on third-order intermodulation, not less, dB Instantaneous survey band of frequency range, MHz DC power supply voltage, V Consumed current, not more, A Weight (with transportation cover), kg

15 5 20

3 2 2 3,000 300 70 75 5 27 (+3, –4.5) 1.5 2.4

Fig. 8.58 Bearing values versus signal frequency and angle-of-arrival

butions on the antenna array elements or the matrices of interference vectors can be used as such initial information. The correlation interference direction-finding method used in the ARTIKUL-M mobile system consists of the correlation estimation between the N−dimensional amplitude-phase signal distribution, received by the antenna array, and the basic N−dimensional distributions calculated analytically with small angular steps, for directions of signal arrival from 0 to 360◦ . That azimuth, the basic distribution of which had the maximum correlation with accepted distribution, is considered the bearing value.

Direction Finding Error Correction in Mobile Systems

311

To eliminate carrier body influence in the ARTIKUL-M1 and ARTIKUL-M6 mobile direction finders, we recommend a practical method based on the fact that, at bearing calculation in frequency ranges for which error values exceed the permissible limits, the calibrated amplitude-phase distributions, which were obtained experimentally for the given carrier and saved in the memory of the bearing calculator, are used as basic distributions. Database accumulation for calibrated amplitude-phase distributions for mobile panoramic direction-finding complexes is executed in several steps: • Conduction of estimate measurements in operating frequency range; reveal of frequency bands where radio deviation and the carrier body resonance are observed • Elimination of the highest resonance of carrier body • Conduction of detailed measurements with registration of the calibrated distribution values • Elimination of erroneous results • Calculation of missing values by means of interpolation on frequency and azimuth, distribution smoothing, and averaging • Database file formation for calibrated distributions. The measurements are executed in open flat areas without trees, brooks, ponds, buildings, phone and electric lines. Area sizes should not be less than 1,000 × 1,000 meters. From the area center, several directions (azimuth) are defined, with the steps not more than 6◦ . Wooden pegs are placed along these directions, with the distance approximately 150 m from the center. The first peg corresponds to the northern direction. The vehicle is placed in the area center. A scanning generator should be used to generate the signals, for instance, the ARK-TG1 software-controlled test generator. The external view of generator with the battery and the external control panel is shown in Fig. 8.59. At the first stage, estimate measurements of direction-finding accuracy in the operating frequency range are executed. During the estimate measurements, the bearing values versus frequency are obtained for the azimuth selected, with rela-

1

3

2

4

5

Fig. 8.59 ARK-TG1 test generator: 1 – CD-disk with software; 2 – External control panel; 3 – Bag for transportation; 4 – Power supply with rechargeable battery; 5 – ARK-TG1 test generator

312

8 Direction Finding of Radio Emission Sources

Fig. 8.60 View of SMO-KPK software window

tively large angular steps, usually from 15 to 30◦ . The retuning step of the scanning generator is defined in the limits of from 2 to 5 MHz. During the measurements, the generator is automatically retuned to frequency, and the amplitude-phase distribution values to the antenna array elements are saved in computer memory with the help of the SMO-TESTMK process software, which simultaneously controls the direction-finder equipment retuning. After that, with the help of the SMO-KPK (correction of bearing curves) customized software, intended for indication of the bearing curves versus frequency and angle, and for calibrated distribution formation, the obtained curves are represented in the form of diagrams, which are analyzed in order to detect the possible resonance and to select the frequency range, for which the correction is required. Figure 8.60 shows the window of the SMO-KPK software, in which the bearing values versus frequency and angle are shown. These curves are obtained during the testing of one mobile direction finder. From this figure, we can see that the system has resonance at frequency near 80 MHz. If, during the estimate measurements, the strong resonance effects are observed, a search of the carrier structural elements, which caused these effects, is carried

Direction Finding Error Correction in Mobile Systems

313

out. Then, if it is possible, the resonance elimination is executed, for instance, by creation of several resonated element connections with the carrier chassis. At the third, and most labor-intensive, stage, the detailed measurements of multidimensional, amplitude-phase, signal distributions of the antenna array are carried out for frequency bands, in which the direction-finding errors exceed the permissible limits. The measurements are conducted for each azimuth, located along the circle, with the angular step not more than 6◦ . Frequency step is chosen depending on bearing curve smoothness, but as a rule, not more than 2 MHz. To avoid the area influence, we recommend the execution of several full cycles of measurements, varying the angular vehicle orientation. At the fifth stage, the obtained files of amplitude-phase distributions are processed by SMO-KPK software. False samples, caused, for instance, by outside radio station operation, are eliminated. Then, the cubic spline-interpolation of the interference vector values is fulfilled, first on frequency, then on azimuth. As a result, the amplitude-phase distributions are formed with small frequency and angular steps. To reduce the influence of random factors, the obtained values are smoothed and averaged. At the final stage, the calibration file is generated, in which the basic values of the amplitude-phase distributions are saved in binary form. The calibration file is placed into controller memory, which controls the system equipment operation and, if the bearing correction option is activated, the controller will automatically use it to calculate the bearing. Let us consider, as an example, the calibration results of an ARTIKUL-M1 specific mobile direction finder, set on a Russian minivan. The peculiarity of this system is the fact that besides the antenna array, two antenna systems of data transmission line, which insert the additional errors, are located on the right side of the roof. Figure 8.61 shows RMS bearing error versus frequency, obtained during estimate measurements of the direction-finding system accuracy.

RMS error, degree 18 14 10 6 2 0

40

200

400 Frequency, MHz

Fig. 8.61 RMS error of bearing determination versus signal frequency

If we calculate RMS errors for frequency bands 40–100 MHz, 100–250 MHz, 250–500 MHz on the basis of the Fig. 8.62 data, we get the following values:

314

8 Direction Finding of Radio Emission Sources RMS error, degree 9 7 5 3 1 0

80

160

240

320 Azimuth, degree

Fig. 8.62 Bearing error versus the direction of signal arrival

! " L1 "1 # σ20...100 = σi2 = 8.90 L1 i=1 ! " L2 "1 σ10...250 = # σi2 = 6.60 L2 i=1 ! " L3 "1 σi2 = 3.60 . σ250...500 = # L3

(8.79)

(8.80)

(8.81)

i=1

When frequency increases, the bearing errors decrease, but, at frequencies more than 250 MHz, the direction-finding accuracy becomes almost acceptable for the solution of practical problems. Figure 8.60 shows RMS bearing error versus azimuth of the signal arrival. If the complex is used for an urgent RES search, on the basis of orientation to the object, bearing accuracy for the axis direction is the most important, in practice. From Fig. 8.62, we see that minimal bearing errors are observed exactly in the cases when signal arrival takes place from the directions close to the vehicle’s longitudinal axis (0 and 180◦ ) and does not exceed 2.5◦ , which is completely acceptable for practical operation. However, if the complex is also used from the stationary position, it is necessary to have good direction-finding accuracy for all directions from 0 to 360◦ . But, error values essentially increase and become more than 6◦ , for the angles corresponding to the vehicle diagonals. Thus, the estimate measurements of the direction-finding system’s accuracy have confirmed the necessity of its calibration. During system calibration for the frequency range 40–250 MHz, experimental amplitude-phase distributions were obtained. The measurements were carried out with the frequency step of 2 MHz for 60 azimuths uniformly distributed from 0 to 360◦ . As a result of obtained data processing, a calibration file of the amplitudephase distribution values with a 500 kHz frequency step and 1◦ angular step was

Conclusion

315 RMS error, degree 16 12 8 4 0

60

100

140

180

220 Frequency, MHz

Fig. 8.63 RMS bearing error versus the direction of signal arrival (mode with correction)

RMS error, degree 9 7 5 3 1 0

80

160

240

320 Azimuth, degree

Fig. 8.64 Bearing error versus the direction of the signal arrival (mode with correction)

formed. The repeated RMS bearing error measurements with the activated error correction are presented in Fig. 8.63. RMS bearing errors for frequency bands 40–100 MHz, and 100–250 MHz now have the values σ40...100 = 2.8◦ , σ100...250 = 2.1◦ . Therefore, RMS bearing error for the frequency band 40–250 MHz decreases more than three times, confirming the effectiveness of the considered correction method. Figure 8.64 shows RMS bearing error versus the azimuth of the signal arrival. Comparison of this picture with the curve in Fig. 8.63 shows that, as a result of the correction, the complex ensures acceptable direction-finding accuracy for practically all directions in the range of 0–360◦ . That is why this method of direction-finding error correction is effective and reduces the instrumental error of the mobile direction finder.

Conclusion In this chapter, the main concepts used in radio direction finding, and a brief historical review of the development of direction-finding techniques is given. Also, the classifications of a direction finder and a list of its main technical parameters are discussed.

316

8 Direction Finding of Radio Emission Sources

The operating principles of the main types of direction finders used at present are described, including: systems based on the rotating directional antenna, doublechannel automatic direction finders (Watson-Watt, Adcock), quasi-Doppler systems, phase interferometers, and correlation interferometer measuring systems. The preferred solution, in the form of correlation interferometer systems and mono-pulse direction finders, is proved, for application in automatic radio monitoring problems. The operating principles of the correlation interferometer with double- and single-channel receiving sections are discussed. Examples of modern direction finders, their structures, technical parameters, and applications of handheld and several automatic direction finders, manufactured now in Russia, are described. In the final section, a method of radio deviation correction for the mobile direction finder is offered, and an example of its application is discussed. As is shown, this method can be expanded to the direction finders of various bases, manufactured by different manufacturers.

References 1. Introduction into Theory of Direction Finding. Radio Monitoring and Radio Location 2000/2001. Rohde & Schwarz GmbH & Co. KG Editor: Gerhard Kratschmer, HW-UKD, pp. 85–101. 2. Krivitskiy, B.Kh. (ed.), Reference Book on Radio Electronic Systems (in Russian). In 2 volumes. Vol. 2, Moscow, Energia Publisher, 1979, 368 pp. 3. Spectrum Monitoring Handbook, ITU-R, Geneva, 2002. 4. Korn, G.A., and Korn, T.M., Mathematical Handbook for Scientists and Engineers, 2nd Edition, New York, Dover, 2000. 5. Patent of RF 2096797, MKI G 01 S 3/74. Method of Radio Signal Bearing and Multi-Channel Direction Finder (in Russian): Rembovsky, A.M. and Kondraschenko, V.N., 8 pp. 6. Patent of RF 2144200, MKI G 01 S 3/14. Method of Radio Signal Bearing and Multi-Channel Direction Finder (in Russian): Ashikhmin, A.V., Vinogradov, A.D., Kondraschenko, V.N., and Rembovsky, A.M., 13 pp. 7. Patent of RF 2184980, MKI G 01 R 29/08. Method of Electromagnetic Field Strength Measurement of Radio Signals and Device for Its Implementation (in Russian): Ashikhmin, A.V., Vinogradov, A.D., Litvinov, G.V., Kondraschenko, V.N., and Rembovsky, A.M., 21 pp. 8. The method for radio signal direction finding and the direction-finder for its implementation. Patent 2201599 of Russia, Class G01S 3/14/Ashikhmin, A.V., Vinogradov, A.D., Litvinov, G.V., Kondraschenko, V.N., and Rembovsky, A.M., 21 p.

Chapter 9

Radio Monitoring Systems and Determination of Radio Emission Sources Location

Introduction As was mentioned in Chapter 2, each radio monitoring system should have an obligatory function set, without which it is impossible to solve problems of radio environment monitoring. First, the following functions can be considered universal: • Real-time panoramic analysis with maximal rate and resolving capacity • Fast search of any “new” emission and measurement of its parameters, determination of its degree of danger (value) to the user, by means of comparison with the database • Creation, replenishment and flexible adjustment of databases • Radio channel monitoring, listening and recording of demodulated signals, technical analysis of radio signals. Moreover, the radio monitoring systems may possess the additional functionality for on-site solving of the various problems. In these cases, they should provide the following: • Measurement of radio signal field strength of the regular radio equipment and new RES • Direction finding and determination of RES location. The above-mentioned functions are necessary for solving the problems of radio electronic intelligence, which is an important activity for the military and other law enforcement services, as well as for the problems of radio monitoring, which are typically solved by the civil and federal agencies charged with regulating radio frequency spectrum usage. The main task of radio intelligence consists in the detection of radio emission, the determination of the geographic coordinates of its sources, the estimation of modulation type and measurement of its parameters, and the interception of open, concealed and coded radio transmissions, as well. At the same time, in accordance with ITU recommendations [1], the main tasks of the civil services regulating the radio frequency spectrum load, are: A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_9,

317

318

9

Radio Monitoring Systems and Determination of RES Location

• Emission monitoring of the permission conformity for the frequency assigned • Monitoring the used frequency band and measurement of the frequency channel occupation • Reveal of radio interference sources • Illegal RES activity suppression. Thus, almost all problems solved during the radio monitoring done by military and civil services have similar character. In fact, radio monitoring systems and RES direction finding and direction-location that operate in the interests of government or civil agencies should be able to recognize RES, determine its location and its main parameters. At that, we determine the following general measurement tasks: • • • • • •

Frequency measurement Bandwidth measurement Determination of modulation type and its parameter measurements Spectrum occupation measurement Radio direction finding and coordinate determination Field strength and power flow density measurements.

The difference between radio electronic intelligence and radio frequency spectrum regulation consists in the goals and the solution depth of the executed tasks. For instance, if interception is of secondary importance for the civil services, then monitoring is mainly necessary for RES recognition. Law enforcement agencies, however, due to the specific character of their activities, would consider the radio interception problem to be one of the most important. Radio interception has its own peculiarities and we do not consider them in this book. The measurement problems of the modulation parameters, the carrying frequency, the radio signal technical analysis and demodulation were discussed in Chapter 6. Theoretical aspects of radio direction finding and equipment review are given in Chapter 8. The present chapter is devoted to the structure of radio monitoring and RES location determination systems, and also to the peculiarities of the stationary, mobile and portable radio monitoring equipment.

Requirements for Radio Monitoring and Location Determination Systems Progress in the fields of digital computing, space navigation systems, radio communication equipment, and other areas of high technologies greatly influenced the equipment and system development necessary to coordinate the measurement of radio emission sources. As a result, the appearance of automatic geographically-distributed radio monitoring and RES location determination systems occurred. The basic requirements for such systems are the following: • Obligatory presence of the universal radio monitoring functions • Error of RES location determination inside the system coverage zone should be minimal

Requirements for Radio Monitoring and Location Determination Systems

319

• In the limits of the operating frequency range, the system must determine the location of any type of RES (with arbitrary spectrum width of occupied frequencies and modulation type) • System should ensure the field strength and the power flow density measurement • System should consist of the minimal possible number of observation posts • System observation posts should be joined in a unified network, to ensure the automatic determination of RES location • Expenses on manufacture, deployment and operation of such system should be minimal, as much as possible. As a rule, the operating frequency range must cover the range from 9 kHz to 3 GHz, and, in several cases, up to 18 or even 40 GHz. The required number of radio monitoring stations is determined by the size of the area, the district topography, financing possibilities and affiliations. In the ideal case, any point of the monitored territory should be in the operation zone of, at least, two radio direction finders, to provide the determination of RES coordinates. However, in this scenario, the number of posts and the system cost are unacceptably increased. Therefore, the more preferable approach to radio monitoring systems is when there are stationary stations – for action zones in more densely populated regions, mobile stations – mounted on the ground, airplane or water transport carriers, and portable stations, which can be deployed quickly in the required regions, including in the points difficult to access. The systems satisfying the mentioned requirements may use two known methods for determination of RES coordinates: • Goniometrical method (with the help of direction finders) • Difference-distance-measuring method (hyperbolic systems). Goniometric systems for RES location determination contain direction finders as part of their structure. Since the specific RES location is unknown, two or more direction finders are used, located in different points of the system activity zone. The doubtless advantage of the goniometric systems is the fact that, at direction finding of quasi-continuous RES, none of the observation posts demands exact synchronization in time with the other posts. The rather small information volume, transmitted from post to post, which determines RES coordinates, is another advantage of such systems. Such systems ensure increased persistence: full efficiency is maintained, down to a minimum of three observation posts, while partial efficiency till two. The main shortcomings of these systems link with the appropriate shortcomings of the direction finders: • Errors in determining RES coordinates depend on the mutual location of the direction finder and the RES in the limits of the system coverage zone • Relatively high cost of the direction finders, especially with wide-range antennas with a frequency coverage factor more than 10.

320

9

Radio Monitoring Systems and Determination of RES Location

Additionally, the modulation type and RES spectrum width can essentially affect on direction-finding accuracy. For multi-channel interferometers this influence is practically absent, while for quasi-Doppler radio direction finders this influence is considerable. In difference-distance-measuring (DDM) systems, RES coordinates are calculated on the basis of the measured values of the time delays during the radio wave propagation from RES to the observation posts. For such systems, the independence of results on the type of antennas used and on the radio wave polarization type is typical. Application of a nondirectional antenna usually gives the best results. This DDM property is the main advantage of goniometric systems. The delay difference between two observation posts defines the locus, in which the radio emission source may be located: the hyperbola with the focuses in the points of observation posts locations. That is why DDM systems are also referred to as hyperbolic. Thus, the post number in DDM should be, at minimum, three, but, in this case, at RES location in some spatial zones, the appearance of two (or more) solutions of the hyperbolic equation system is possible. The introduction of the fourth observation post only adds to the required adequacy of the system and allows elimination from the ambiguity. The essential DDM disadvantage is its impossibility to determine the location of the non-modulated signal source. Moreover, the accuracy of RES location determination depends on its modulation. The best results for RES can be achieved with cuspate and fast-falling auto-correlation function of the modulating signal. In contrast to goniometric systems for RES location determination, for DDM systems, we need time synchronization between all system posts, with the accuracy up to 10–8 s. The implementation of such synchronization makes the equipment more complicated. Finally, in contrast to a radio direction finder, the operation results of which may be the azimuth value, the result of a single observation post in the DDM system is the signal sample. Signal samples are transmitted from all observation posts to the total post for coordinate calculation, where the appropriate delays are determined, and then the radio emission source location. Therefore, in the DDM system, the volume of information transmitted from the observation posts to the post of coordinate calculation may be larger than in the goniometric system. Thus, goniometric systems on the basis of direction finders are preferable for the implementation of a geographically-distributed system for the location determination of RES with the arbitrary modulation types.

Structure of the Radio Monitoring System and Determination of RES Location We shall discuss the organization of the automatic radio monitoring system and the determination of RES location using the following systems as examples: stationary ARK-POM1, mobile ARK-POM2, portable ARK-POM3, and combined ARK-POM.

Structure of the Radio Monitoring System and Determination of RES Location

321

ARK-POM1 System This system is intended for operation in large towns and industrial centers. It consists of the ARK-SST (“Archa”) stationary stations, whose antenna systems are deployed at prevailing heights and on the roofs of high buildings. The interaction of the central station with the peripheral ones is provided via the wire, fiber-optical or radio channels with high carrying capacity. The structural diagram of the system is shown in Fig. 9.1. Stationary peripheral stations, as a rule, are remotely controlled and do not require a permanent operator presence.

Fig. 9.1 ARK-POM1 System for determination of radio emission sources

• • • • •

In the ARK-POM1 system, the following units are included: ARK-SST (“Archa”) stationary central station ARK-SST (“Archa”) stationary peripheral stations of the same structure that is in the central station or reduced ARK-MS1 (“Argument”) mobile peripheral stations Communication and data transmission equipment ARK-RP3, ARK-RP4 handheld direction-finding equipment, for more precise determination of RES location on-site or in buildings.

322

9

Radio Monitoring Systems and Determination of RES Location

ARK-POM2 System This system consists of ARK-MS1 (“Argument”) mobile central and peripheral stations mounted on mobile carriers. Control of the mobile peripheral stations is provided from the central mobile station via a radio channel with low and medium carrying capacity. Due to the accommodation of the radio equipment on the mobile carrier, the ARK-POM2 system can quickly change its operation zone, ensuring the urgent fulfillment of radio monitoring actions and the determination of RES location. In the ARK-POM2 system structure (Fig. 9.2), the following units are included: • • • •

ARK-MS1 (“Argument”) central station ARK-MS1 mobile peripheral stations Communication and data transmission equipment ARK-RP3, ARK-RP4 handheld direction-finding equipment, for more precise determination of RES location on-site or in buildings.

Fig. 9.2 ARK-POM2 mobile system

ARK-POM3 Geographically-Distributed System This system is intended for the automatic determination of radio transmitter location and for radio monitoring, and consists of quickly deployed posts. The radio equipment of these posts is categorized in the portable multi-functional equipment family, designed for hand transportation by several operators. Similar equipment

Structure of the Radio Monitoring System and Determination of RES Location

323

can be used at stationary or temporal posts equipped by electric power sources, and at an open site. The ARK-POM3 system is designed for radio monitoring, simultaneous direction finding, and for determination of RES location. It consists of the central post, and one or more peripheral posts on the basis of the ARTIKUL-P portable direction finders. In the ARK-POM3 system, the following units are included (Fig. 9.3): • “Arena” central portable station • “Arena” peripheral portable stations • Equipment for communication and data transmission between central and peripheral stations • Handheld direction-finding equipment, for more precise determination of RES location on-site or in buildings.

Fig. 9.3 ARK-POM3 portable system

Combined ARK-POM System It is possible to combine the three above-mentioned systems into a single system. At that, the combined system may include the stationary posts, the mobile posts, and the deployable posts, as shown in Fig. 9.4.

324

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.4 ARK-POM combined system

The main purpose of this system is radio monitoring in the operating frequency range, the simultaneous or synchronous RES direction finding, the determination of its location, the radio control of given channels. The AK-POM system ensures the following: • • • • • • • • •

Quick search, detection, and location determination of the radio emission sources Data storing and providing the databases on the sources Technical analysis and measurement of the RES modulation parameters Interception and registration of the informational messages Single-channel and multi-channel direction finding Simultaneous (synchronous) direction finding Scanning, receiving, and multi-channel radio monitoring Estimation of the electromagnetic field strength Other functions including the registration of the analyzed range occupation.

The software of the ARK-POM system supports an arbitrary number of stationary, deployable, and mobile direction-finding posts, and the mobile posts can operate during movement.

Structure of the Radio Monitoring System and Determination of RES Location

325

The lower boundary of the operating frequency range for detection and automatic direction finding traditionally lies in the band of 20–25 MHz, which relates to the necessity of stations monitoring at the 27 MHz band. The upper boundary of automatic direction finding is defined by the practical needs of the user and may be 1,000, 2,000, 3,000, or 8,000 MHz. Further increase of the upper frequency boundary of direction finding, up to 18 GHz, is carried out on the basis of the amplitude method of direction finding, using the ARK-KNV4 external radio signal frequency converter, mounted on the mast with a turning device, remotely controlled from the vehicle cabin. The real instrumental accuracy of direction finding is equal to 0.5–3◦ , the location determination accuracy is about 1.5–2◦ of the distance to the RES. Practically achieved sensitivity at direction finding over the range is in the limits of 0.5–30 μ, which corresponds to the values of the best foreign complexes. The spectral analysis rate at receiver bandwidth 5 MHz, and for the double-channel receiver, is about 300 MHz/s. The main features of some radio monitoring and direction-finding stations of the ARK-POM model are listed in Table 9.1.

Table 9.1 Comparative station parameters of ARK-POM system “Archa” stationary station

“Argument” mobile station

“Arena” portable station

Receiver type Operating range, MHz: Basic set Operating range, MHz: Full set Spectral analysis rate, MHz/s Spectrum discreteness, kHz Input sensitivity in band 12 kHz, μV Operating range, MHz: Basic set Operating range, MHz: Full set

ARGAMAK 25–3,000

ARGAMAK 25–3,000

ARGAMAK 24–1,300

0.009–18,000 (post 2)

25–3,000

–

3,000

3,000

1,200

6.25

6.25

3.125

1

1

1

25–3,000

25–3,000

25–8,000

Direction-finding rate, MHz/s

300

25–8,000 (ant. on car roof); 25–18,000 (antenna on mast) 300

Antenna on mast: 25–1,300 25–3,000

Fuction

Parameter

Panoramic analysis (posts 1, 2)

Direction-finding (post 1)

150

326

9

Radio Monitoring Systems and Determination of RES Location Table 9.1 (Continued)

Fuction

Multi-channel radio monitoring

“Archa” stationary station

“Argument” mobile station

“Arena” portable station

Signal bandwidth, MHz Field sensitivity, μV/m

Arbitrary

Arbitrary

Arbitrary

o.5–25

0.5–3

Instrumental accuracy, (RMS error), degrees

0.5–3

Channel number Sensitivity in 12 kHz band, μV Spectral analysis rate, MHz/s

8 2

1–30 (ant. on car roof) 0.5–25 (ant. on mast) 1–30 (ant. on car roof) 0.5–25 (ant. on mast) 4,8 2

24,000 64,000

24,000 64,000

Parameter

0.5–3

– –

–

Control Arrangement in the System Data Exchange between Stationary Posts In the ARK-POM system, the peripheral direction-finding posts include the equipment necessary for communication with the system’s central post. The data exchange between the system’s stationary posts is provided through high-speed radio channels, wire or fiber-optical lines. The modern high-speed radio system of data transmission operates in the microwave range and ensures the data transmission rate from 10 Mbit/s and higher. The communication system between the ARK-POM system posts has two levels: software and hardware. At the software level, unification of the exchange interface between the applications executing on the PCs included in the systems is provided. The hardware level defines the physical implementation of the communication channel between system posts. The software application interaction between them is executed through a unified component, specially developed for the ARK-POM system and for similar ones. It is included in all applications and realizes the protocol of connection-disconnection, and data exchange between the applications. The component is built over TCP/IP (Transport Control Protocol/Internet Protocol) protocol, which allows the applications to interact in a standard way between themselves, irrespective of their physical location. The applications may be on a single computer, in the local or wide area of the computer network, exchanging the information via the cable or radio channel. When in operation on the single computer, the component allows full imitation of

Control Arrangement in the System

327

operation – without the loading of networks means and TCP/IP protocol – due to the data exchanges of the files represented in the memory. The system configuration adjustment (names, addresses, interacting application ports) is executed through the ini.files, which permits the system to readjust flexibly. The standardized exchange protocols of the upper level allow system creation from the already existing applications. The combined text-binary exchange protocol is used for data transmission. In this protocol, the data part is transmitted from application to application, in the form of text messages suitable for the observation and monitoring of network traffic. The other data are transmitted in the form of binary packages associated with the appropriate text messages. Such a data exchange method allows the essential reduction of network loading, compared to the purely text protocol. All stationary stations of the ARK-POM system are connected into the local area network (LAN) with TCP/IP protocol. At the physical level, communication with the remote stationary posts can be established on the basis of several standards, for example, on the basis of radio-Ethernet operating in the range of 2,400–2,483.5 MHz in accordance with IEEE 802.11b standard. The exchange rate of this protocol is from 2 to 11 Mbit/s. The communication distance for this protocol – defined by different factors and at using the directional antennas – is 5–15 km without the additional amplifiers and up to 40 km with the additional amplifiers. Configuration of the radio communication equipment depends on the number of stationary posts and the coverage zone. The relatively simple system consisting of three stationary posts (one central and two remote) may use the wireless channels of the “point-point” type. At such a communication arrangement, two adapters of the wireless network are connected to the PC of the control post. Each of these adapters allows the PC of the control post to support a wireless connection with PC of the direction-finding post, equipped by a similar adapter. In this case, directional (parabolic) antennas are used both at the central post and the peripheral one. The advantages of such a system are its relatively low cost, and low radiated power because of the application of a directional antenna at both sides of the communication channel. The shortcomings are the necessity of two adapter installations in the control post PC, and the irrational usage of high-cost radio frequency resources. In the systems consisting of three or more posts, it is expedient to use the channels of the “point-multi-points” type, for communication. In this case, the communication equipment of the remote posts remains unchanged, while the central post is equipped either by the point-of-multiple-access to the wireless network adapter, with the control post PC assigned as the radio access server, or by the wireless bridge or the router. The advantage of the radio access server is its relatively low cost, and the shortcoming is the loading increase on the control post PC. The advantage of the wireless bridge is the higher operation rate of the wireless network. When using the “point-multi-point” channel, the central post antenna should be nondirectional, which requires the higher transmitter power.

328

9

Radio Monitoring Systems and Determination of RES Location

Data Exchange with the Mobile and Deployed Posts The above-mentioned high-rate systems are factually inapplicable for data exchange with the mobile station because of the absence of straight visibility and the impossibility of directional antenna usage. Therefore, the low-rate radio communication systems are applied here: the autonomous narrow-band radio systems for data transmission, with data transmission rate from 9,600 to 40,000 bit/s, or the radio modems of the cellular radio communication systems. For instance, one may offer the ParagonPD+ basic station for the stationary post and the GeminiPD+ mobile modems for the mobile stations, as the version of stable radio communication implementation between the central stationary and mobile posts. When using a nondirectional antenna, which is necessary for mobile station operation, the operation distance of such a communication system is close to the distance of straight visibility. At the stationary post, the antenna height is 30 m, and, for the mobile station, the antenna height is 3 m; the radio communication distance is near 20–30 km. Many radio stations for commercial and military purposes have external modems for digital data transmission. As an example, we may select the Russian radio stations of the Akvedook P168E family, whose built-in modems are able to transmit digital data at a rate up to 16,000 bit/s. The assortment of dedicated narrow-band low-speed radio modems suitable for portable system application is also rather wide, for example, for this purpose, the IntegraTR or T-96SR radio modems are suitable. In this case, the typical distance between the posts is near 3–10 km, depending on the power of the radio modems used, and, as a rule, it is restricted by the direct visibility between the posts. Figure 9.5 shows the deployed antenna on the foldable mast for the radio modem on the peripheral post. To increase the operation distance, the directional antenna is used and the TR-965SR radio modem in the protective cover (Fig. 9.6) is attached to the mast base, which ensures the minimal length of RF cable, and hence, the minimal signal losses.

Peculiarity of the Low-Speed Radio Channel Application Narrow-band radio modems, as a rule, use the simplex or semi-duplex transmission method, in which it is impossible for data transmission and receiving to happen simultaneously. Therefore, an additional speed decrease occurs due to the time of the communication session, at the data transmission direction change from the central post to the peripheral post and vice versa. Typical time of the communication session for narrow-band radio modems is 30–250 ms. In order to execute data transmission with the help of a narrow-band radio modem, the protocol of packaged data transmission, using the selective repeat (SR) method, is applied. In this protocol, the following main peculiarities of narrow-band radio lines are taken into account:

Control Arrangement in the System

329

Fig. 9.5 Antenna of radio modem

Fig. 9.6 Radio modem in protective cover

• • • •

Lines ensures serial data transmission Simplex or semi-duplex transmission is used Communication session and the transmission direction change take essential time Distortion of the separate bits is possible at the receiving side.

Data transmission by frames consisting of a large number of packages of limited length is a peculiarity of the protocol. For each package, its number is transmitted together with its CRC (Cyclic Redundance Check) value. The package check value is defined again by the received data. If the calculated CRC value does not coincide with the initial value, then this package is rejected. After frame-finishing, the transmission direction changes: the receiving side transmits, in the return frame, the receiving acknowledgement, which contains the numbers of the packages with

330

9

Radio Monitoring Systems and Determination of RES Location

errors. In the same frame, other data are placed for the transmitting side. Due to such integration, the number of communication sessions decreases. Then, the transmission of wrongly received packages repeats. If there are no errors, the frame transmission is considered complete. At the absence of frame receiving acknowledgement during the given time interval (time-out), the transmitting side transmits the given frame again. At operation in the network, the frame transmitted by the central post contains additionally the marker permitting the response to the radio modem with the given number. The selective repeat protocol increases the carrying capacity of the radio line. For instance, at the implementation of a 5 km long radio line with the help of the T-96SR radio modems, the bearing transmission rate achieves 5–8 bearings per minute, i.e., grows approximately twice compared to the block protocol, in which one package only can be transmitted without the acknowledgement. The modem transmitter output power was 5 W, transmission rate was 19,200 bit/s, and directional antennas with the gain of 6 dB were used at the receiving and transmitting sides. To arrange the data exchange using narrow-band radio modems, the radio network with the “star” topology is used. It is shown in Fig. 9.7. The peripheral posts may exchange data with their own central post only. The SMO-PPK customized mathematical software for the panoramic direction-finding

Peripheral posts Central post 8/0

ADP unit

Central post PC Programcontroller

Programcontroller TCP/IP

ADP unit Radio modem

SMS-RMS

TCP/IP SMS PBC (manager) Peripheral post PC

TCP/IP SMS-RMS

Radio modem

ADP unit

Programcontroller TCP/IP

Radio modem

Fig. 9.7 Peripheral post control via narrow-band radio modems

SMS-RMS

Control Arrangement in the System

331

complex operating on the central post PC has the managing functions of the remote radio equipment post control on the basis of the TCP/IP protocol. The software (SMO-RMC customized mathematical software for radio modem communication) converts the data of the TCP/IP protocol into the protocol suitable for data transmission via the narrow-band radio line. The central post PC controls the system operation, and represents and processes the information coming from the other posts. The program-controller of the peripheral post controls its radio equipment operation and transmits the data to the PC-manager, on the basis of the commands coming from the central post.

Usage of Radio Modems of the Cellular Communication Systems When the operation region of the ARK-POM system is located in the coverage zone of a cellular radio communication system, the radio exchange between posts can be executed with the help of that cellular system. This solution has evident advantages: • No need to obtain approval for radio frequency range usage and to purchase expensive narrow-band modems • As a rule, very small weight and sizes of radio equipment • The operating distance is defined by the coverage zone of the cellular radio communication system. Of course, shortcomings exist: • The need to pay the cellular communication operator for the service • Dependence of the ARK-POM3 system on the efficiency of the cellular radio communication system and its coverage zone. At present, cellular radio communication systems of the GSM standard are the most widespread. In the simplest case, the mobile phone, having the function of binary data transmission, may be used as the modem. This phone is connected to the PC or directly to the ADP unit via the RS-232 cable, or through infrared interface. However, the application of special GSM modems is more preferable. Such modems have no “unnecessary” elements, such as the keyboard, color display and photocamera, but support also the standard set of AT-commands, which allow their usage for data transmission, as the usual mobile phones do. In spite of the active development of GPRS (General Package Radio Service) technology, at present, the data transmission mode with channel-switching remains the more reliable version of data transmission via the GSM network. GRPS shortcomings include an inconstant average data transmission rate, which depends on the workload level of the cellular system. In practice, the data transmission rate usually varies in the limits of 900–6,000 bit/s, and sometimes falls to zero. Moreover, the presence of a server is necessary, with a known network address for connection via TCP/IP protocol.

332

9

Radio Monitoring Systems and Determination of RES Location

The data transmission mode with channel-switching allows the arrangement of real time “point-point” type transmission between two objects. At that, the data transmission rate, if the connection has been established, equals the carrying capacity of the voice channel: 9,600 bit/s. Compared to the application of narrow-band radio modems, the GSM modem has the following advantage: it implements the duplex transmission method, at which transmission and receiving can be realized simultaneously, which twice increases the carrying capacity of the channel. As a possible choice, we suggest equipping the station with two data transmission systems: on the basis of cellular modems and on the basis of autonomous radio modems. This choice combines the advantages of the cellular radio modem application with the autonomy of system operation, necessary in case of failure or blockage of the cellular radio communication system.

Data Exchange Implementation in Combined ARK-POM System In the ARK-POM system, there are three ways of implementing radio exchanges between the posts: Central station ARCHA High-rate radio modem ARKMP4 Post1

PC

ARK-D1TR or “Argamak-I

High-rate radio modem

Post1

Hub

Low-rate channel

Post 2

ARK-MP4 or ARK-MP1 ARK-D1TR or “Argamak-I Post 2 ARK-RD8M

Hub ARK-RD8M

Post3

Router

Peripheral station ARCHA High-rate radio channel

Post 3 Wire modems

Basing station of network radio communicati on system

Printer Post 4

Fiber-optical channels

Radio modem

Portable station ARENA Low-rate radio channel

Wire lines

Radio modem

ARKMP4 Post 1

Hub

ARK-D1TR or “Argamak-I

ARK-MP4 or ARK-MP1

Wire modem ARKMP4

Post 2

ARK-P7 or ARK-P11

Post 1

Hub

ARK-D1TR or “Argamak-I

ARKRD8M

ARK-RD8M Post 3

Peripheral station ARCHA Fiber-optical channels

Peripheral station ARCHA Wire channel

Fig. 9.8 The control in the combined ARK-POM system

Mobile station ARGUMENT Low-rate radio channel

Control Arrangement in the System

333

• With the help of high-speed radio modems, wire or fiber-optical lines with the data transmission rate more than 1,000 kbit/s • With the help of low-speed radio modems operating in free-running mode • With the help of radio modems of the cellular radio communication systems. The best choice, for communication between stationary posts, is via highspeed communication channels, while, for mobile or deployable posts, communication executed via radio channels with low-speed data transmission is preferable. Figure 9.8 shows the generalized structural diagram of the combined ARK-POM system. The peripheral stationary stations are controlled from the central stations via three types of communication channels: wire, fiber-optical, and high-speed radio channels. The mobile and portable stations are controlled with the help of the cellular system radio modems or the free-running radio modems. The control system shown in the figure has the “star” topology, i.e., all peripheral stations are connected to the central station only. Other, more complicated, topologies are possible; for example, Fig. 9.9 shows another configuration of the similar system. Finally, the control organization in the system is defined by specific conditions, including the territory, the district topography, where the system is deployed, the degree of infrastructure development, the tasks that the system should solve, the system cost, etc.

Fig. 9.9 The control on the basis of the branched topology

334

9

Radio Monitoring Systems and Determination of RES Location

“Archa” Stationary Station The “Archa” stationary station consists of several posts: • Post No. 1. ARTIKUL-M4 system for radio monitoring and direction finding, with the stationary antenna system for mounting on the mast. • Post No. 2. The panoramic radio receiver: the ARGAMAK single-channel, or the ARK-RD8M multi-channel, or (when necessary to measure the regular radio equipment parameters) the ARK-D1TP or ARGAMAK-I panoramic measuring radio receiver with the measuring antenna set. • Post No. 3. The ARK-RD8 multi-channel radio monitoring system. • Post No. 4. ARK-KN equipment for cartography and navigation. • Handheld equipment for measuring and parameter-monitoring of the regular radio equipment: the ARK-NK3I handheld measuring system on the basis of the ARGAMAK-I panoramic measuring receiver. • Handheld equipment for more precise determination of RES location on-site or in the buildings: the ARK-RP3, ARK-RP4 handheld direction finders. • Communication and data transmission equipment. • System-wide equipment. The “Archa” peripheral stations may have the same components as the central station, or reduced components. The central and peripheral stations, as a rule, differ by the mathematical software for the processing sub-system, for cartography and data registration, and the system-wide equipment. The structural diagrams of the stationary and peripheral “Archa” stations are shown in Fig. 9.8. Here, the structure of the central station, consisting of the four posts, and the structures of the peripheral stations, consisting of three posts, are shown. Let us examine the main components of this station. Only two types of direction finders correspond to the strict requirements related to the direction finding of wide-band signals with complicated modulation types and short-term signals: the double-channel direction finders and the mono-pulse correlation interferometers. As previously mentioned, the implementation of mono-pulse direction finders is accompanied by large material and technical expenses; therefore, the basic type of direction finders used in the “Archa” and “Argument” stations is the correlation interferometers. In the standard configuration, the “Archa” stationary stations of the ARK-POM1 system provide the solution to radio monitoring problems in the range of 25– 3,000 MHz and the determination of RES location with an accuracy up to 1.5–2% of the distance. Expansion of the received frequency range for radio monitoring is possible: down to 9 kHz and up to 18 GHz. The direction finding in these additional ranges is provided by the ARK-RP3 (0.3–3,000 MHz), ARK-RP4 (1–8 GHz) handheld direction finders and on the basis of the ARK-KNV4 frequency converter, which has the built-in directed antenna system in the range of 3–18 GHz. Post No. 1 is the station designed on the basis of the ARTIKUL-M4 direction finder with the stationary external antenna system on the mast, installed on

“Archa” Stationary Station

335

a building roof or on a high place. The height of the antenna system installation defines the radius of the electromagnetic accessibility (EMA) zone and the reliability of the RES monitoring. The double-channel DRR of the ARTIKUL-M4 direction finder is constructively combined with the circle antenna system. Therefore, the radio signals received from the outputs of the two channels with the coherentlyrelated local oscillators are transmitted via RF cables on a relatively low frequency. This makes it possible to use cables with lengths up to some hundred meters. The signals on IF pass to the input of the double-channel ADP unit mounted on Post No. 1 in the station room. Figures 9.10 and 9.11 show examples of antenna system accommodation for the ARTIKUL-M4 radio direction finder of the “Archa” stationary station. Figure 9.12 shows an example of the operator workplace of Post No. 1, combined with Post No. 4, on the “Archa” central station. The workplace for the operator is equipped with two PCs. The ARTIKUL-M4 equipment is connected to the first PC. The second PC is used for solving the problem of RES location determination and for representation of the district electronic map. The operator of Post No. 1 controls the radio direction-finding equipment. If the station does not have a Post No. 4, then the operator of Post No. 1 determines the RES location, with the help of the digital map. Post No. 2 is constructed on the basis of the ARGAMAK panoramic receiver, which provides fast panoramic analysis for the search of “new” RES, or those RES that are of special interest to the operator. In these cases, when the measurement of field strength and the radio signal parameters is required, the ARK-D1TR or

Fig. 9.10 AS-MK4 antenna system on the roof of a 10-storey building

336

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.11 AS-MK4 antenna system on the mast

Fig. 9.12 Post No. 1 operator workplace combined with Post No. 4

ARGAMAK-I certified panoramic measuring receivers are used with the measuring antenna sets. Post No. 3 is intended for multi-channel automated radio monitoring. The ARKRD8M multi-channel panoramic radio receiver is included into its structure. These

“Archa” Stationary Station

337

multi-channel radio receivers were discussed in section “SMO-MCRM Customized Software Package”. Post No. 4 is the post for cartography and navigation. It is used for the calculation of the coordinates of RES being found, and for its representation on the electronic map. Moreover, we can represent the distribution of the electromagnetic field strength on the electronic map, as well as the reliable receiving zone. The map can also show the location of the stationary stations and trace the mobile station movement of the ARK-POM system. It should be noted that, at the combining of Posts No.4 and No.1, the software of the SMO-DF direction-finding post and of the SMO-CN cartography and navigation post are able to operate on separate PCs, (see Fig. 9.12) as well as on the single high-speed PC. All radio receivers have outputs at the analog intermediate frequency of 10.7 MHz or 41.6 MHz, for connection of the additional demodulation or decoding equipment. In the minimal configuration, each of the “Archa” stations contains only one post: Post No. 1. The operator of this post solves the problems of all posts, including RES location calculation and its representation on the cartographic background. The important feature of the “Archa” station software is the possibility of the remote control of its posts equipment via TCP/IP protocol, which allows organization of the peripheral stations on the basis of “unmanned” technology. Under remote control, the central station task of the ARK-POM system is operation management of the remote stations-controllers via the computing network, i.e., the manager functions with respect to the controllers. Controller initialization on the direction-finding posts, spectrum analysis task transmissions to the controllers, signal sampling and its copying to the central station, and controller rerun in case of failure are all included into the manager tasks. Figure 9.13 shows the window of the SMO-DF Dispatcher application, which controls the operation of three stations with the direction-finding posts, two of which are remote and unattended, and one which is included in the system central post. For final solution of the RES location determination task – with the required accuracy for the practical tasks (up to the house, the apartment, etc.) – the ARKRP3 handheld direction finders may be included in the stationary station structure. If the measurement of radio equipment parameters is required upon departure to the location region, the ARK-NK3I handheld measuring complex can play the role of the additional equipment. Means of communication and data transfer are necessary equipment for the “Archa” station. The number and assortment of these means are defined by the ARK-POM system control topology. The antenna-mast devices, electric power supply system, illumination and airconditioning systems, the lightning and fire guard systems should be classified as system-wide equipment. We shall consider the peculiarities of lifting-mast devices and power supply systems for stationary, mobile and portable radio monitoring and direction-finding systems in sections “Navigation Systems for Radio Monitoring Stations” and “Electric Power Supply Systems”. The other systemwide equipment usually used on the stations is described in the referenced

338

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.13 The window of the dispatcher application

books (for example, in [1]). We recommend the interested reader to consult such books.

“Argument” Mobile Station As mentioned earlier, mobile equipment essentially widens the task range on radio monitoring and RES location determination, and makes executable tasks that are difficult or even impossible to realize by stationary equipment. Examples of such tasks are: RES detection and spurious emission level on the boundaries of the monitored zone, the determination of the operation zone of mobile radio communication systems, the exact location-finding of RES under conditions of range overloading, RES signal detection and parameter monitoring, which uses directed transmitting antennas. The mobile station radio equipment structure is defined by the solved tasks, and depends on the size of the vehicle-carrier. If the station is installed in a minivan, then, on account of its small cabin area, it is permissible to have two or three posts in the system structure. An external view of the stations is shown in Fig. 9.14, and the typical equipment accommodation in the technical compartment is shown in Fig. 9.15.

“Argument” Mobile Station

339

a

b

c

Fig. 9.14 “Argument” mobile station: (a) with the ARTIKUL-M basic system having two antenna systems; (b) with the AS-MP6 antenna system under the radome and the horn antenna on the dielectric mast; (c) with single antenna system on the miniven roof; (d) with two antenna systems

340

9

Radio Monitoring Systems and Determination of RES Location

d

Fig. 9.14 (continued) Fig. 9.15 Equipment accommodation in the technical compartment: (1) external antenna system; (2) power supply unit from 220 V; (3) air condition evaporator (in cover); (4) air condition compressor; (5) ARK-PSST power supply; (6) ARK-UPS12 power supply; (7) fire extinguisher of automatic fire fighting system; (8) barrel of telescopic mast; (9) electric petrol generator; (10) coil with the ground wire

In the maximal configuration, the “Argument” mobile station may contain the following equipment: • Post No. 1. The ARTIKUL-M basic complex with two antenna systems: AC-MP1 system under radio transparent radome (Fig. 9.14a,c), or AC-MP6 on the whole vehicle roof (Fig. 9.14b) and the AS-MP4 or AS-PP17 deployable antenna system

“Argument” Mobile Station

•

• • • • • •

341

for mounting on the mast (Fig. 9.14a,d). At station displacement, the deployable antenna system is accommodated in the rear vehicle compartment. Post No. 2. The panoramic radio receiver: the ARGAMAK single-channel or ARK-RD8M multi-channel (see Chapter 3 and Chapter 5); when necessary to measure the regular radio equipment parameters, the ARK-D1TP or ARGAMAK-I certified measuring panoramic radio receivers with the ARKKNV4 certified radio signal frequency converter and measuring antenna on the dielectric mast, installed on vehicle body or in the area near the car. Post No. 3. ARK-RD8 multi-channel radio monitoring complex. Post No. 4. The ARK-CN1 equipment for cartography and navigation. Carrying equipment for measurement and parameter monitoring of the regular radio equipment: carrying the ARK-NK3I measuring system on the basis of the ARGAMAK-I panoramic receiver. Carrying equipment for more precise RES location determination on-site or in the buildings: the ARK-RP3, ARK-RP4 handheld direction finders. The communication and data transmission equipment. System-wide equipment.

The possible organization of the station consisting of three posts is shown in Fig. 9.16 . The first post is intended for solving the tasks of panoramic detection, multichannel and single-channel direction finding, signal technical analysis and database generation, and estimation of the electromagnetic field strength in the frequency band of 25–3,000 MHz. The ARTIKUL-M mobile complex with two antenna systems is the basis of this post. The second post of the complex is based on the ARK-D1TP digital panoramic measuring receiver, or on the ARGAMAK-I, or on the ARK-RD8M multi-channel panoramic radio receiver with increased performance. This second post executes the tasks of panoramic detection, signal technical analysis, measurement of field strength, and the recording of signals on the intermediate frequency – and the demodulated signals – on the PC hard drive, in the frequency range from 9 kHz to 18 GHz (with the appropriate antennas and the ARK-KNV4 frequency converter). For accurate measurement of the field strength, the measuring antenna is used, mounted on the external tripod or on the dielectric telescopic mast with the turning system. A second option, telescopic mast installation directly on the vehicle body, is also possible (see section “Navigation Systems for Radio Monitoring Stations”). The third post of the complex contains the ARK-RD4 or ARK-RD8 multichannel radio monitoring equipment, and is intended for the automatic multichannel radio monitoring of signals in the given sections of the operating frequency range, and for recording of the demodulated signals to the PC hard drive. The radio receivers of all posts have analog intermediate frequency outputs of 10.7 or 41.6 MHz, for connecting additional demodulating or decoding equipment. PCs of all post are united in the local area network (LAN). The second post’s equipment can operate via operator actions. In this case, control of the measuring

342

9

Radio Monitoring Systems and Determination of RES Location Post of radio monitoring and direction finding

Control

Antenna system ARK-MK1M

Antenna switch ARK-AK-A

Signal 10002020 kHz

Signal Converter ARK-PS5

Signal

Navigation equipment ARK-CN1

ADP unit ARK-CO5

Audio PC

Control/Data

Control

Control

PC network hub

Post for radio signal monitoring and parameter measuring

Measuring antenna P6-59

Measuring antennas P5-61, P5-62

Antenna ARK-A7A

Signal up to 1.000 MHz

Power supply filter ARK-PSF

Control

Converter ARK-KNV4 1…18 GHz

Signal 25-1000 kHz

Measuring receiver ARKD1TR or ARGAMAK

PC Audio Control/Data

Post for multi-channel intercept

Antenna ARK-A7A

Multi-channel receiver ARK-RD8 or ARK-RD8M

PC Audio Control/Data

Radio modem antenna

Radio modem

Fig. 9.16 Structural equipment diagram of the “Argument” mobile system

receiver will be conducted by the SMO-DF application, executing at the PC of the first post, which has the necessary network functions to accomplish this. Task redistribution between the posts is possible in such a way that the navigation tasks are executed on the first post PC, together with representation of the electronic map. The transfer of tasks on direction finding and interception from Post No. 2 to Posts No. 1 and No. 3 is provided.

“Argument” Mobile Station

343

In the reduced configuration, two posts are included in the station structure: Post No. 1 and Post No. 2. At that, the ARK-KN1 equipment for cartography and navigation and the electronic map are installed at the first post. In this case, in addition to the direct radio signal processing and its direction finding, Post No. 1 solves the navigational and cartographic tasks on the RES coordinates calculation, the representation, on the electronic map, of the complex’s movement-tracing, RES location, and the field strength distribution. In the minimal configuration, only Post No. 1, combined with Post No. 4, is included in the station structure.

System-Wide Car Equipment The effectiveness and quality of the solutions to radio monitoring problems are defined not only by the technical and tactical features of radio receiving, and the measuring and direction-finding equipment, but by the system-wide car equipment, as well. The success of radio monitoring problem-solving depends, in many respects, on the car: how well the vehicle is arranged, how reliable is the electric power source system, whether the air conditioning and ventilation systems exist or not, how suitable the operator workplaces are, whether places for relaxation exist. Therefore, it is necessary to focus serious attention on the system-wide car equipment. Let us consider, as an example, the possible accommodation of the “Argument” station on the basis of a typical minivan. The equipment accommodation scheme is shown in Fig. 9.17.

Fig. 9.17 Equipment accommodation scheme of the “Argument” mobile station

344

9

Radio Monitoring Systems and Determination of RES Location

The vehicle of the “Argument” mobile radio monitoring station is, physically, a van, with an all-metal body and a flat, metallic, antenna platform on the roof, covered by a radio transparent radome. The vehicle’s internal space is divided into three compartments: driver cabin, operator compartment (three workplaces), and the technical compartment. The driver cabin is separated from the operator compartment by a lightproof curtain, which can be moved to the side, if necessary. The vehicle has additional thermal and audio isolation, the floor and the ceiling are reinforced, and air conditioning and heating systems are added. These systems operate during movement and upon parking. The technical compartment is separated from the operator compartment by a solid partition wall. In the motor and technical vehicle compartments, there is an automatic fire-fighting system. The side door of the compartment opens wide (similar to a car door). Compared to a sliding door, which is typical for minivans, the open-wide door is more preferable for radio monitoring stations, since, firstly, it has less size compared to the sliding one, which makes it possible to use the compartment more effectively; and, secondly, at its opening and operator exiting, the vehicle compartment is not visible, which is important for the solution of concealed tasks. The following equipment is installed inside the operator compartment: an angle table, for two workspaces; an armchair, with an adjustable seat back, and set on a turning pedestal; a small sofa, a drop table for one workspace, a convertible seat; an additional shelf for documents and small items, a clothes peg, and a shelf for the documents. The angle table is mounted in the operator compartment, to the left of the driver saloon. It is attached to the vehicle body by bolts, through a rubber, shock-absorbing saddle. The turning armchair is mounted in the front part of the compartment near the angle table. The sofa is mounted in the rear part of the compartment, near the partition wall of the technical compartment. It has luggage space and is intended for the sitting of several persons or as a place for sleeping. In the operator compartment, at the bottom to the left along the vehicle side, there is a barrel for transportation of the telescopic mast. The barrel hole is open into the technical compartment. Radio equipment is mounted in a special pole, to the left, at the bottom of the first workspace, under the angle table. For PC mounting, special plan-tables are used, installed on the workspace tables (see Fig. 9.18). The PC is kept on the table with the help of the plan-table, at that, its screen is strictly fixed under the given angle. For user convenience, the plan-table is equipped with local illumination. The vehicle is equipped with two air conditioners, their airflow distributed towards the workspaces. There is an additional compartment heater and an electric heater, a vehicle refrigerator, and local illumination lamps. The glass in the operator compartment is tinted, to ensure non-transparency from the outside at daytime. In order to ensure concealed operation at night, nontransparent blinds curtain the operator compartment windows. The drop-table is located near the door on the vehicle’s right side. In the opened position, it is fixed, with the help of an inclined stop. The additional convertible seat is installed in the

“Argument” Mobile Station

345

Fig. 9.18 Operator compartment, Post No. 1

right front part of the operator compartment. In the collapsed position, it is attached to the wall by a belt. Under the convertible seat, the vehicle refrigerator is mounted. Small access for the cable entry allowing the external antenna system connection (Fig. 9.19) is in the vehicle body, along the left side of the driver saloon. The cable entry has a collapsible apron, to protect the connector against atmospheric precipitation. On the right side, the vehicle body has a small access with a lock, for connection of an external power supply source 220 V, 50 Hz. In the technical compartment, the following units are accommodated: control unit of a 220 V power supply; autonomous electric station; secondary power source

a

b

Fig. 9.19 Small access for external antenna connections: (a) the protective apron is lifted; (b) the protective apron is in the lowered position

346

9

Radio Monitoring Systems and Determination of RES Location

consisting of the ARK-UBP12 and ARK-BPCT units; terminal box; reserve rechargeable battery; climate system with 220/12 V transformer and rectifier; external antenna system for direction finding; the pin and the reel for ground wire; the box with the measuring antennas on the shelf. For operation convenience, under darkened conditions it is possible to connect a handheld lamp in the technical compartment. In the technical compartment, there is also a place to the right, for transportation of the AS-MP4 deployable direction finder antenna. At the bottom to the left, is the barrel hole for transportation of the telescopic mast 9–12 m in length.

“Arena” Portable Station The “Arena” portable station for automatic radio monitoring can be classified to the portable multi-functional equipment class, which is intended for handheld transportation by one, two or three operators. Similar facilities can be used on stationary or temporal posts equipped, or not equipped, with an electric supply, and on the open site also. Portable stations can be quickly deployed in inaccessible or difficult to access places for transport carriers: in the mountains, under lack of good road conditions. Additionally, they may be used in urban conditions as temporary stations, where their antenna systems can be quickly deployed on building roofs or on other constructions.

“Arena” Station Structure The following units are included in the potable station structure: • The ARTIKUL-P or the ARTIKUL-P11 portable panoramic radio direction finder consisting of the deployable antenna system, the double-channel digital radio receiver and the analog-digital processing unit (see section “ARTIKUL-M1 Mobile Direction Finder”); • The data transmission system, including the radio modem with antenna • The electric power system • Laptop PC with the software and the electronic digital district map • The ARK-KN1 navigation device, to provide the determination of the post geographic coordinates. The structural diagram of the station is presented in Fig. 9.20. The portable station can operate under operator control, as well as in the remote control mode via narrow-band radio communication lines. There are two ways to organize remote controlled posts. The first is when the peripheral post structure coincides with the central post structure. In this case, the post operates under PC control and is able to execute all possible radio monitoring and direction-finding functions, similar to the central post. In the second case, the PC is not included in the peripheral post structure. Therefore, the narrow-band radio modem is connected directly to the ADP unit. The post operates under the control of

Mast Devices for Radio Monitoring Stations Fig. 9.20 Structural diagram of the “Arena” portable station

347 Panoramic direction finding system

Antenna system for radio monitoring and direction finding

Double-channel digital radio receiver

Analog-digital signal processing unit

Data transmission system Radio modem antenna

Radio modem

PC

Cartography and navigation unit Antenna of device for cartography and navigation

Cartography and navigation unit

Electric power supply system

the unit’s internal micro-application support. This solution has its own advantages and shortcomings. The following advantages are: • Complexity of the peripheral post operation decreases, since there is no PC in its structure, operating, as a rule, in relatively narrow temperature environment and humidity range • Electric power consumed by the post equipment reduces • Transportation and post deployment simplifies • Requirements for qualified service personnel decrease • When the PC is not included in the peripheral post structure, the shortcomings are: • Restricted number of fulfilled functions, since the equipment is controlled by the micro-application support • Necessity of narrow-band radio modem application controlled via RS-232 protocol • In case of central post failure, the peripheral post will not be able to take upon itself the fulfillment of the central post functions. Which peripheral post structure is preferable depends on the problem specificity, which should be solved by the system in the specific case.

Mast Devices for Radio Monitoring Stations To lift the antenna systems of the radio receivers and the transmission modems, we need the mast devices. Let us examine the features of three similar devices: the ARK-MT1, ARK-MT2 telescopic dielectric masts and the ARK-MT3 telescopic metal mast.

348

9

Radio Monitoring Systems and Determination of RES Location

a

c

b

d

Fig. 9.21 The ARK-MT dielectric mast1: (a) with ARK-KNV4 converter and the P6-61 measuring antenna; (b) with the measuring antenna mounted on the ground; (c) the turning mechanism; (d) in the stick of the “Argument” station

The ARK-MT1 telescopic dielectric mast is intended for mounting of the measuring antennas or other radio equipment, for instance, the ARK-KNV4 radio signal frequency converter. This mast is 8 m high. It can be installed on the ground or on the transport carrier and has a turning mechanism around the vertical axis (Fig. 9.21). The mast has a winch with a manual lifting drive. If the mast is mounted on the transport carrier body, then, with the upper section dropped, we may use it, at speeds up to 10 km/h. In the disabled state, the mast goes into the stick of the “Argument” mobile station. The application of dielectric material for the mast case reduces the errors of electromagnetic field strength measurement. The ARK-MT2 telescopic dielectric mast with distant control is intended for the measuring antenna or other radio equipment mounting. This mast is shown in Fig. 9.22 . This mast is 6 m high. The mast’s peculiarity is an electric, distantly-controlled drive for mast lifting and turning. This mast does not require the use of bracing wires and may be mounted on the vehicle body. In collapsible form, it allows for vehicle movement. The mast is controlled with the help of a distant panel or from a PC. High accuracy of the electromagnetic field strength measurement is ensured, due to

Navigation Systems for Radio Monitoring Stations

349

Fig. 9.22 ARK-MT2 dielectric distantly-controlled mast

the upper dielectric section. The mast can be used within the structure of automatic direction finders with rotating antennas. The ARK-MT3 telescopic dielectric mast is intended for installation of the radio direction finder antenna. Depending on the section number, its length is 9, 12, or 15 m. The mast has a three-section or four-section duralumin case with the manual lifting drive. The mast may be mounted on the ground as well as on the vehicle body, as shown in Figs. 9.23 and 9.24. The main technical parameters of mast devices are listed in Table 9.2 .

Navigation Systems for Radio Monitoring Stations For solving the problems of radio emission monitoring, one of the constituent tasks is the task of coordinate and angular orientation determination. This task in referred to as the navigational task. The hardware-software complex for this problem solution is called the navigational system or the navigational complex. Errors of the navigational task solution can lead to errors in the main task solution. For example, errors in determining the platform coordinate and azimuth

350

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.23 ARK-MT3 mast installed on the ground

directly influence the accuracy of calculated bearing-plotting during the determination of RES location on the basis of its direction finding. Another error source should be described. Navigational systems generate navigational messages, with the definite frequency and delay defined by their constructive peculiarities. Errors caused by the finite rate of navigational message generation are called discretization errors. For instance, the variation of the course angle (azimuth) of the vehicle platform may achieve 20◦ per second, so the error of bearing plotting caused by the unaccounted delay 0.1 s may be 2◦ .

Features of Modern Navigation Systems The problem of coordinate and object orientation determination is the classical navigation problem. It is solved with the help of the various navigation systems (inertial, long-wave ground-based, space-based, etc.). A full description and classification of navigation systems exceed the bounds of the present book and we shall only briefly consider the modern sensor types suitable for creating a navigation system for radio monitoring stations. They are the magnetic field sensors (magnetic compasses), the equipment of the satellite radio navigation systems (SRNS)

Navigation Systems for Radio Monitoring Stations

351

Fig. 9.24 ARK-MT3 mast installed on the vehicle body

of the NAVSTAR, GLONASS, and GALILEO types, and the inertial navigation system (INS). Magnetic compasses have been used in navigation systems, up to now. At present, three-dimensional compasses defining the magnetic field lines in space Table 9.2 Main technical parameters of masts Name

ARK-MT1

ARK-MT2

ARK-MT3

Maximal lifting height, m Overall length in transport state, m Basic case diameter, mm Turn angle around vertical axis, degrees Position accuracy, degrees Maximal weight of mounting equipment, kg Minimal weight of equipment, kg Handle force, kg, not more Deployment time, minutes Mast mass without bracing wares and anchors, kg, not more Electric drive supply voltage, V

5.5 2.87 110 360 1 6 3 3 15 12.4

6 3.1 – 360 1 12 3 – 1.5 14.4

9, 12, 15 3.13 85 – – 32 8 4 20 21.7; 29

–

12

–

352

9

Radio Monitoring Systems and Determination of RES Location

are used, or two-dimensional ones finding the projection of the direction to the device plane. The cost of a digital compass suitable for navigation purposes strongly depends on the accuracy class, and can amount to from several tens of U.S. dollars per one module, at an accuracy of about 2◦ , to several hundreds and even thousands of U.S. dollars for systems of high accuracy, with errors of about several minutes. The advantages of magnetic compasses are: the application simplicity, the autonomy, and the arbitrarily high rate of information obtaining. Its shortcomings are the influence of magnetic field abnormality (permanent and temporal) on the readings and, as a consequence, the necessity of creating abnormality maps, and the complexity of magnetic compass application on metallic carriers, namely, the vehicles, in particular. The receivers of satellite radio navigation systems are beyond comparison among modern navigation systems, by their cost-quality ratio, their operation zone, and the simplicity of their integration. At present, the equipment for operation with the U.S. SRNS NAVSTAR, often called GPS (Global Position System), is the most widespread. Work on the reconstruction of the satellite group of the Russian system GLONASS has been conducted. At present, the new European SRNS GALILEO is under development. The operation principles of all three SRNS are similar, and the difference among them consists in some technical features, while the final equipment is practically identical from the user point of view. That is why our further discussion will be oriented on SNRS NAVSTAR. The full SRNS description exceeds the bounds of this book and we consider only some necessary design and operation principles of such systems [2]. SNRS consists of three sub-systems: the space apparatus (SA) sub-system; command-measuring complex sub-system; user equipment (UE) sub-system. The uniform distribution of 24 SA is assumed in orbit, but the actual number of SA depends on the current system budget, on requirements from the U.S. Ministry of Defense and other factors. 31 SAs are included in the NAVSTAR space group, as of the beginning of 2005. Additionally, each SA consists of a transmitter with the system of navigation-information signal formation and the high-stable frequency standard for time scale formation. The command-measuring complex sub-system consists of the network of the command-measuring posts and the system control center. In particular, this subsystem generates the amendments, which compensate for the scale deviation of the on-board frequency standards for the whole SA group. Coordinate determination via user equipment is based on the measurement of so-called pseudo-distances, with the visibility zone being from the user equipment to the SA. Without details, we should note that the pseudo-distance represents the distance from the UE antenna to the SA – with some unknown displacement, which is the same for all SAs. To determine three coordinates (attitude, longitude, and altitude), it is necessary, as a minimum, to measure four pseudo-distances, because one should include the mentioned displacement into the vector of unknown parameters. Besides the pseudo-distance measurement, modern receivers are able to measure the carrier phase of SA navigation signals.

Navigation Systems for Radio Monitoring Stations

353

The advantages of SRNS are their high accuracy, independence on magnetic, gravitational and other abnormalities, and the short time for equipment activation. Their shortcomings are the low rate of generating navigation information, rather low noise immunity, and possible signal losses (shading), which are especially typical for operation under conditions of compact urban development. GPS receivers of the NAVSTAR system differ greatly, by cost and features. The price of the simplest modules is within the limits of 100 USD; the price of single-frequency boards that support phase measurement and movement velocity and direction determination is 1.5–2 times more. The cost of high-accuracy double-frequency boards, intended for operation with geodesic accuracy, may achieve up to $10,000 or more U.S. dollars. Figure 9.25 shows the single-frequency board that supports phase measurements and is able to execute velocity determination. Figure 9.26 shows two types of GPS antennas.

Fig. 9.25 GPS receiver

The introduction of cheap inertial sensors opens new possibilities for the application of corrected inertial navigation systems in areas where there was previously no possibility to use such systems, due to cost, weight, size and other restrictions. One such area is INS application in the orientation systems of mobile ground objects. The classical INS includes the inertial measuring system unit with three accelerometers and three gyroscopes (angular velocity sensors), and the computing device. The typical diagram of sensor accommodation is shown in Fig. 9.27. The sensor is oriented in such a way as to feel the carrier displacements along three spatial coordinates. Since the accelerometers and the angular velocity sensors in such a system are strictly related with the carrier, to determine their orientation, a special computing

354

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.26 GPS antennas

Fig. 9.27 Group of sensors with three accelerometers (1) and three gyroscopes (2) mounted on the common rigid base

Fig. 9.28 Module of inertial-navigation system: (1) digital control processor; (2) group of gyroscopes and accelerometers; (3) power supply

unit is used, which defines their axis direction, and which coordinates – by means of movement – equation integration, using the given inertial sensors (Fig. 9.28). INS application allows the determination of carrier coordinates and orientation, with an arbitrarily high rate. The shortcoming of any inertial system, in turn, is the

Navigation Systems for Radio Monitoring Stations

355

error accumulation, which requires the correction from another navigation system, for example, SRNS. INS cost depends on its accuracy, and is defined first by the cost of the used inertial sensors: the accelerometers and the gyroscopes, which is usually some hundreds of USD for the low accuracy systems and has practically no limit for the precision systems (tens and hundreds of thousands USD). First of all, though, the gyroscope drift defines the system class: 10◦ per hour for low accuracy systems, angular minutes and seconds for the precision systems.

Navigation Systems for Mobile Stations The choice of the specific construction scheme of the navigation system is defined by the requirements for the accuracy of the navigation task solution, the rate of the navigation parameter generation, the restrictions on the system budget. The modern SNRS receivers, together with the coordinates, generate the receiver movement velocity and direction. Velocity determination is based on the use of pseudo-distance increments over some interval. At that, there are some shortcomings: • • • •

It is impossible to obtain information on the orientation of the fixed platform It is difficult to ensure the high frequency of information generation We can determine the movement direction only, but no object orientation The velocity vector determination, even more than the coordinate determination, suffers from the possible SNRS signal loss, which is typical for operation under urban conditions.

Therefore, the direct application of SNRS receivers does not ensure the required features of the orientation-navigation task solution and thus, their data is expedient to use together with the data from the inertial navigation systems. At mutual application of the data from INS and SRNS, the possibility appears to create integrated systems, in which the advantages are kept, and the influence of the shortcomings of each system reduces to a great extent. In such systems, INS is corrected by the high-accuracy data of SRNS, and, at SNRS signal miss, the INS unit of the integrated system is able to generate navigation data with the acceptable accuracy during some time, depending on the features of the used sensors. The most widespread integrated INS with correction by SNRS data includes INS, the SRNS module, which provides the generation of the correcting positional information, and the computing device. However, the low accuracy features of the cheap inertial devices, first of all, the angular velocity sensors, do not allow the application of the classical initialization procedures of INS on the fixed base, using the Earth’s rotational effect. Therefore, at usage of low accuracy sensors, the classical scheme of correctable INS construction will be unable to work in enough important informational situations, for fixed objects or those moving with low velocity.

356

9

Radio Monitoring Systems and Determination of RES Location

One possible solution is the application of an additional magnetic compass. Unfortunately, as mentioned earlier, the usage of cheap magnetic compasses of low accuracy leads to unsatisfactory results. Moreover, the application of any magnetic compass restricts the system application area. Another technically and economically valid approach to solve the problem of platform orientation determination is the application of satellite phase measurements. The interferometer principle lies in the basis of this method of orientation determination: the phase differences on the carrying frequency are measured, which are received from the satellites on the spaced antennas. In this difference caused by the different distances from the antennas to the satellites, the information on the angles between the vector of direction on the satellite and the vector formed by the spaced antennas is contained. As restrictions of this scheme application, we may note the more rigid requirements for SRNS signal quality. Thus, the correcting information generated by the SNRS unit, together with the classical positional information, should include the information on the object orientation. In such systems, there is either the multi-antenna SRNS module (2–4 antennas) with the possibility of phase measurement of signals receiving by different antennas (or its differences), or several independent receivers with the possibility of signal phase measurement. The first method is technically more preferable, but it is realized in the very expensive special boards, while the second method is realized in the standard equipment; the additional efforts on frequency synchronization of the separate modules are required. The quality of the solution is defined by the quality of the used components and the number of antennas in the SRNS module. In the INS-SRNS operation mode, the error of coordinate determination is in units of meters, and the error of orientation angle determination is about 0.5/l, where l is the antenna space in meters, in the projection on the planes orthogonal to the turn vector.

Electric Power Supply Systems Requirements for Electric Power Sources The radio monitoring station equipment covers the wide range of technical facilities, and the digital radio receiver (DRR) is their basis. In addition to one or several DDR, this station includes several antennas with feeder and a PC. Irrespective of the specific station purpose, there is a power source (PS) of some kind in its structure. It can be built-in in the DRR voltage regulator, for power supply from a chemical current source, the power supply for operation from an AC net, or an external or internal universal power supply, with the possibility of operation from the reserve rechargeable battery, etc. It should be noted that the reliability of the station operation, to a great extent, is defined by the reliability of the power supply, since, at malfunction or damage, the equipment becomes completely useless. At the same time, at defects or failure

Electric Power Supply Systems

357

in other units, or at software error presence, the equipment may function even with the restricted possibility set. We should consider electromagnetic compatibility as the most considerable demand to the station power supply. The strict demands concerning the radio emission level result from the fact that the PS is directly in the equipment structure intended for receiving and radio signal analysis. Therefore, for exactly this reason, in most of the measuring and communication equipment, the application of highfrequency impulse regulators and supply voltage converters has been restricted up to now, in spite of their advantages in efficiency and other parameters. In the radio station equipment, units and blocks of different purpose are included, for which the supply voltage, consuming power, and requirements for supply voltage quality strongly differ. In other words, there is a hierarchic structure of power sources in the complex structure – with the power range from tens of mW to tens or even hundreds of watts. The necessity of using thermal-stable micro-power voltage regulators often occurs, to stabilize the operation point, for example, of microwave field-effect-transistors (FET) in the input stage of radio receivers or the stages of automatic gain control. Another “extreme” point is the regulating voltage converters for the PC, including in the ARM complex structure, which are located on the transport carrier. In this PS, the main point is not the voltage regulation accuracy in the limits of percent parts, but in keeping the output voltage with rather high tolerance (up to 20%) at the on-board voltage fluctuations up to 50%, and at its temporal miss. In high-power on-board PS, it is necessary to provide the emergency supply from the reserve rechargeable battery of the complex and its automatic charge from the on-board network.

Electric Power Sources for Radio Equipment Radio monitoring equipment, as any other radio electronic equipment, gets its operating power supply from an external (primary) source. In radio monitoring stations, various primary electric supply sources are used, which is caused by the wide application of the solved problems. Let us consider briefly the most frequently used sources. AC Mains AC mains are usually used for electric power supply for stationary installed equipment. A tolerance of ±20% on the long deflections of the voltage value from nominal should be provided in the technical tasks. There are always consumers, using the electric mains of the general public, generating powerful pulse interference. For example, the typical domestic refrigerator, which has the consuming power of 100–150 W only, in its steady-state condition, at compressor start, may consume 4–10 kW. Electric mains voltage-decreasing at start moment has a pulse character with a value up to 100 V. During the transition to the steady-state condition and at compressor switching off, damped oscillating processes of

358

9

Radio Monitoring Systems and Determination of RES Location

various frequency and duration, with the peak voltage of tens and hundreds of volts, occur. The short-term voltage-decreasing is possible also at computer and other radio electronic equipment switching-on (for example, due to the charge of the input capacities of the power supply, or the operation of a demagnetization loop for a TV tube). It follows from the above that, at development of the secondary power electric supply operating from AC mains, not only the variation of RMS or the amplitude voltage value should be taken into consideration, but also the presence of pulse interference with high energy. Vehicle On-Board Network An on-board net with 12 V voltage is the historically-formed world standard for the on-board voltage of automobiles and small trucks. It appeared in connection with the application of lead accumulators on transport means, which have good electric parameters with low cost, relative construction simplicity and easy maintenance. The main advantage of lead accumulators on transport means is the large short-term permissible current, necessary for the engine start, with the starter’s help. The conventional 12 V value represents the nominal voltage of the lead accumulator, consisting of six connected series of elements, under the loading. It is close to the potential sum (12.6 V) of six electric-chemical elements, in the most balanced condition. At vehicle engine operation, the voltage of the on-board network is defined mainly by the characteristics of the used power generator unit. The maximal voltage of the on-board network is defined by the potential of the fully-charged accumulator, and it is 14–14.5 V. In other words, the output generator voltage is limited by this value, which can be automatically corrected with the variation of the environment temperature. If the accumulator was strongly discharged, at the initial stage of its charging from the generator, the on-board network voltage may be restricted to 10–12 V, by the accumulator potential value and its low internal resistance. Decreased to 10–12 V, the on-board network voltage may be observed at the working generator, in case of increased load, for example, at night and at bad weather, when many regular electric consumer devices (the car’s heater, window wipers, headlights, rear-window defroster, etc.) are switched-on. At the non-operating engine, operation of the on-board electric network with 12 V is defined by the accumulator features, on-board circuit resistance, and by the consuming current value. The extreme accumulator voltages are: 14–14.5 V at the completely charged accumulator at the first minutes after charger switching-off; 9–10 V at full discharge (the potential without the load or at low load); 7–8 V at short-term engine starting. The on-board network voltage may exceed the mentioned limits only at risk of damage to the vehicle’s electric equipment. Moreover, as in all wire networks, pulse voltage decreasing and increasing are possible due to operation of various consumers, for instance, the stop-signal.

Electric Power Supply Systems

359

The on-board 24 V network does not differ principally, with respect to the onboard 12 V network, except that it has twice the voltage. This value is used for circuit current decreasing on large trucks and heavy transport, where there are more electric energy consumers and the total consuming power is higher. We should note that, in modern automobile design, the transition to an on-board voltage of 36 V has begun. The voltage growth allows essentially a decrease in the weight of the copper wires, due to consuming current decrease at the same power. In seriallymanufactured automobiles with such on-board network voltage, for the transition period, an additional network of 12 V standard is arranged, intended mainly for additional devices like radio cassette players, etc. On-Board Network of Aircraft In accordance with the standard, on-board network DC voltage of aircraft may lie in the limits of 21–31.5 V (except the emergency mode). The short-term decreasing of this voltage is fixed separately up to 13 V at the autonomous engine start, and overshoots up to 53.5 V at power equipment switching (duration up to 200 ms). For time intervals about 10–100 ms, the voltage overshoots and decreases may have even larger values. For RMS, AC voltage with the nominal value of 115 V and the frequency of 400 Hz, the steady-state voltage value should be in the limits of 100–127 V. Chemical Current Sources A considerable part of radio monitoring equipment is intended for autonomous operation from built-in or external chemical energy sources. In this device class, rechargeable batteries are usually used. Lead unattended accumulators with gel-like acid electrolyte form the main part of these batteries. They have several advantages compared to the other types of rechargeable batteries, primarily the low cost and their ability to generate large current without capacity loss. This property allows the use of a small capacity battery without problems, as the emergency buffer, to supply high-power consumed equipment. It can be used especially often in the structure of uninterrupted power sources for the PC, when, at network voltage failure, it is necessary to support the system efficiency during at least several minutes, for emergency data-saving. Also, the lead accumulators have no memory effect to the charge-discharge mode and allow the ensuring of long cyclic battery operation without complex circuit solutions. Until recent times, in autonomous equipment of relatively small consuming power (0.1–5 W), the application of leak-free, nickel-cadmium, rechargeable batteries was predominate, which was caused by the technological difficulties of manufacturing the portable lead accumulator. Now, the more compact and power-intensive rechargeable batteries of the nickel-metal-hydride and lithium-ion type are available. The last is preferable for application in communication, due to the absence of memory effect.

360

9

Radio Monitoring Systems and Determination of RES Location

Secondary Electric Supply Sources All types of voltage converters, stabilizers, and regulators, providing electric energy with normalized characteristics for radio equipment component parts, are considered as secondary electric supply sources. We review below the main classes of secondary electric supply sources, together with their peculiarities, advantages and shortcomings. Linear Voltage Regulators and Stabilizers There is no radio-engineering device – including radio-monitoring equipment – that can operate practically without application of the linear voltage regulator or the voltage stabilizer in its structure. Thus, linear voltage regulators are in a privileged position, since they do not generate electromagnetic interference. Linear voltage regulators have several indisputable advantages: • Electromagnetic emission absence is, as mentioned above, the principal and main advantage, with respect to pulse and other types of regulators • Absence of internal origin interference through the power supply circuits (input and output) • Circuit and construction simplicity • Small sizes of PS on the integrated circuits, without taking into account the heatremoving elements (radiators, ventilators, etc.) • Low cost • Minimal expenses for development. At the same time, the linear stabilizers and regulators do have shortcomings, and the main among them is low efficiency, especially at large tolerances on the input supply voltage. The second shortcoming is the principal impossibility of the voltage increase and the absence of galvanic coupling between the output and input circuits. Constructive and technology problems of heat-removing from the regulating element and ensuring the maximal operation time from the autonomous power supply result directly from the first shortcoming property of the linear PS. For example, the device operating time can be less than 50% from the theoretically possible value with the loss-free stabilizer for the equipment supply from the chemical current source, the voltage at which varies by 30% during the discharge (typical value). This is caused by not only the direct energy losses, in the form of heat on the regulating element of the linear regulator, but by the dependence of the chemical source capacity on the consuming current value. For many electric chemical systems, the capacity reduction with twice current growth may be up to 20%. Pulse Voltage Stabilizers and Converters As was mentioned earlier, the main factor limiting the application of pulse supply units (uninterrupted power supply – UPS) for radio monitoring equipment is

Electric Power Supply Systems

361

the interference generated by them in the radio frequency range. This interference is directly emitted by the circuit elements into the environment and is also transmitted from the power supply to other units of the device and to the supply network through the power wires and the control circuits. This interference spectrum from modern UPS operating at frequencies up to 1 MHz may extend up to 200–300 MHz. However, the conversion frequency increase in UPS led to the paradoxical, at first glance, fact, which consists in some simplification of interference control. Its first reason consists in simplification of the control of the interference magnetic component. With the conversion frequency growth, the electromagnetic field penetration into the shield material decreases, due to the skin-effect. As a result, to suppress the interference emission, we do not need the thick-walled perm-alloy shields, as we do for UPS operating at the frequencies of about 100–10,000 Hz. At stable operation mode, the interference spectrum from UPS does not contain the sub-harmonics and begins from the fundamental conversion frequency (most frequently 50–100 kHz) extending exclusively up on frequency. The frequency growth of energy conversion allows also the decreasing of choke and filters sizes in input and output UPS circuits, without reduction of filtering quality. Taking into account the fact that, in modern electronics, the UPS element base has received the widest development, and also due to the necessity of creating equipment with decreased energy consumption (especially for autonomous equipment), it would be unreasonable to neglect the above-mentioned fact. In this connection, the developers of many manufacturers prefer UPS application for all units of produced radio equipment. The cited argumentation does not mean that it is necessary to fully refuse the application of linear PS. For example, it makes no sense to build in the regulated UPS into the antenna amplifier unit mounted directly near the receiving antenna vibrators, although it consumes the current of several tens of milliampere. The share of this current in the total consumption is small, and UPS application would not only create interference directly near the sensing elements, but may insert distortion into the antenna pattern. In a number of cases, especially when the device is supplied purely from 220 V AC mains and the consuming power does not exceed 1–2 W, the application of the linear integrated voltage stabilizer may distinctly reduce the device cost. But, in a number of cases, it is rather difficult to operate without UPS application. For instance, in the typical minivan, the nominal voltage of the on-board network is 12.6 V, but the real voltage is equal to 10–14 V under normal conditions, with the overshoots up to 16–18 V during electromagnetic device switching and with the decreasing up to 7–8 V at engine start. Under such conditions, it is practically impossible to provide stable operation of the device with a 9 V internal supply voltage, using the linear voltage regulator. If the power consumed by the device (unit) is about 20 W, then the additional 10 W, dissipated by the voltage stabilizer, are added to the useful consumed power. Under the strict climate requirements for the equipment, application of the linear voltage stabilizer, as we can see, creates the constructive problem of additional heat withdrawal.

362

9

Radio Monitoring Systems and Determination of RES Location

The following example with the on-board 24 V network is no less visual. In the equipment structure, there are six radio-receiving units, with the nominal supply voltage of 12 V. The consuming power of each unit is 20 W. The on-board voltage may fluctuate in the limits of 18–28 V. Taking into account the good reserve on the input voltage, we can apply the linear voltage stabilizer, and there will be no problems with interference. But, as a result, we have the consuming current from the on-board network near 10 A, independent from the voltage, which is not a small value (corresponding to not more than 10 h of autonomous operation at the reserve rechargeable battery capacity of 100 A h). An additional shortcoming is the heat dissipation on the voltage stabilizer: 10A·(28 V – 12 V)=160 W (the peak value). At a temperature drop of 20◦ C, to dissipate such power, a radiator with 0.8 m2 area will be required, without application of forced ventilation. At the same time, the application of the step-down UPS with efficiency 90% allows: • Almost twice the reduction of the current consumed from the on-board network • Reduction of the dissipated power, approximately to 12 W • Approximately twice the increase of the time of autonomous operation. In this example, which is rather typical for radio monitoring stations, the UPS advantages are evident. We should note that the pulse step-down voltage regulators satisfying the requirements of the given example are rather well represented by the specific micro-circuits. Integrated circuits of this purpose may have the minimal quantity of discrete elements. Experience of UPS application in radio monitoring equipment shows that the guarantee of a successful fight with spurious emissions is, in the first place, the correct approach to the design of the unit’s printed circuits and to the construction of its shields, following from the general physical laws of electromagnetic energy propagation. We would like to attract attention to the interesting fact that the interference from the pulse PS is substantial, for equipment operating in the frequency range of 100–200 MHz. For the higher frequency equipment, more serious problems are created by the emission from its digital units. These include the ADC circuits, all types of signal and controlling processors, IC of programmable and hard logic, and also the units of representation and processing, for instance, the PC. All these equipment components operate at clock frequency, from units to hundreds of megahertz, and the spectrum of harmonics and noise generated by them extends to several gigahertz. It is more difficult to control the interference from the digital units than it is with the interference from the UPS, in spite of the smaller range of voltages and currents of the fundamental frequency. This is related to the very high switching rate of the modern digital IC, which is significantly higher (by orders) than for the power UPS elements.

Example of Pulse Power Supply of Low Power As an example of the practical implementation of UPS of small power, we can consider the power supply for the panoramic technical analysis (PTA) unit, which

Electric Power Supply Systems

363

represents the amplification and intermediate frequency filtering section with its conversion. The internal supply voltages are: +5 V, and –5 V at consuming current 100 mA. The external non-stabilized supply voltage is 8–16 V. The section frequency bandwidth is 2 MHz, the central input frequency is 10.7 MHz, and the central output frequency is 1.6 MHz. The external view of the unit is shown in Fig. 9.29. Fig. 9.29 PTA unit. The shield is not mounted. The elements of the pulse power supply are located in the left part of the unit

Fig. 9.30 Spectrum of non-modulated signal. The power supply shield is not installed. We can easily see the spectral components of interference from the power supply

Figures 9.30 and 9.31 show the signal spectrum at PTA unit output for the signal from a G4–164 signal generator with RMS value –50 dBV connected to PTA input.

364

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.31 Spectrum of non-modulated signal. The board is completely mounted. There are practically no spectral components from interference

Figure 9.30 shows the signal spectrum for power supply shields that have not been installed, while Fig. 9.31 shows the same signal spectrum for installed shields. We may see that, in the second case, there is no interference from the power supply to be practically observed.

Multi-Channel Pulse Power Source As another example, we consider now the ARK-PS power source, oriented on the supply of mobile radio monitoring stations for aircraft. The external view of this power supply is shown in Fig. 9.32. The power source represents a multi-channel device, which ensures the stabilized voltage for up to eleven loads and which has nine outputs of non-stabilized supplies, switched by electronic switchers. The reserve rechargeable battery is connected to this unit. The unit automatically switches to reserve supply when the main supply net fails or misses. To support the efficiency of the reserve battery, an automatic charging device is included into the unit’s structure. In the ARK-PS structure, the sub-units of power supply, remote control panel, and cables for connecting the panel and the hooter are provided. The main technical parameters of the ARK-PS unit are listed in Table 9.3.

Electric Power Supply Systems

365

Fig. 9.32 ARK-PS power source Table 9.3 Main technical parameters of ARK-PS power source Parameter

Value

Input voltage, V Output number of stabilized voltage Output number of non-stabilized voltage Outputs 1, 2

21–31.5 11 9

Outputs 3, 4, 5 Outputs 6, 7, 8, 9 Output 10 Output 11 Outputs 12–20 Weight, not more, kg Sizes of the basic unit, not more, mm Reserve accumulator Operating temperature range Consuming power at full load, not more, W

Note

12 ± 0.25 V, 12 A

Two outputs for radio modem supply 27 ± 0, 5 V, 6 A Three outputs 12 ± 0.25 V, 15 ± 0.25 V, Four outputs with regulated 19 ± 0.25 V, 5 A voltage to PC supply 5 ± 0.2 V, 2 A One output for supply of the navigation equipment One output for supply of the 12 ± 0.25 V, 2 A network concentrator (hub) 21–31.5 V, 1 A Nine end-to-end channels 20 485 × 200 × 380

12 V, 17 A·h From +10◦ C to +40◦ S 1450

Two accumulators

The unit provides the electric supply for the radio equipment of three posts, four usual PCs, the navigation equipment, and the radio modems for data transmission.

ARK-UPS12 Universal Power Supply Unit Another UPS example is the ARK-UPS12 power supply unit, developed for mobile radio monitoring stations on a vehicle base. This power supply unit represents a

366

9

Radio Monitoring Systems and Determination of RES Location

multi-channel device, which provides the stabilized voltage for four loads and the non-stabilized voltage for three loads. The peculiarity of this power supply unit is the presence of a reserve rechargeable battery with the capacity of 190 Ah. The unit’s external view is shown in Fig. 9.33. Control of the ARK-UPS12 power supply unit is provided with the help of the remote control panel, which is shown in Fig. 9.34. This power supply unit has a rigid case, in which the following functional modules are installed: the control and automatic unit, four modules of voltage converters; three modules of end-to-end supply; the accumulator charger unit, and module of electronic switchers. The power supply provides: • Automatic transition to the supply from the reserve rechargeable battery, at fails or misses of the main supply voltage • Automatic charge of the reserve rechargeable battery (in the “background” or “operating” modes)

Fig. 9.33 ARK-UPS12 power supply unit

Fig. 9.34 Remote control panel for ARK-UPS12

Electric Power Supply Systems

367

• Protection from false switching (on and off), under conditions of large vibrations • Emergency sound signaling at deviations from the nominal value of any output stabilized voltages, or at missing the non-stabilized voltages • Visual indication of unit operating mode, and the measurement of the output voltage and current on any input or output. The main technical parameters of the ARK-UPS12 power supply unit are listed in Table 9.4. Table 9.4 Main technical parameters of ARK-UPS12 Parameter

Value

Input voltage, V Stabilized voltage output number Non-stabilized voltage output number Output 1 Outputs 2, 3, 4

9.5–16.5 4 6

Outputs 6, 7, 8 Consuming power, W, not more Weight, kg, not more PS sizes, mm, not more Remote control panel sizes, mm, not more

23.5–27.5 V, 7 A 15 ± 0.5; 19 ± 0.5 and 21 ± 0.5 V, 8 A 9.5–16.5 V, 12 A 500 10 485 × 200 × 380 135 × 120 × 75

Note

Three outputs for PC supply

Three end-to-end channels

Four voltage converters, six supply voltage switchers (end-to-end channels), the control and automatic unit, the accumulator charger unit, the remote control panel, and the module of switch control, are included in the power supply unit. The input supply voltage passes to the control and automatic unit, which analyses this voltage value and also the charging level of the reserve rechargeable battery. At power supply unit activation, the input voltage passes via the appropriate electronic switch to four converting cells, to the charging device, and to end-to-end supply modules. If the voltage on the reserve rechargeable battery decreases below the given value, the control unit automatically generates the command to activate the charging mode. In cases when the input voltage fails, the automatic unit connects the reserve rechargeable battery via the electronic switch. The unit’s structural diagram is shown in Fig. 9.35. The power supply unit is installed in the vehicle’s technical compartment, and its mode control is provided with the help of the remote control panel, removed to the operator compartment, as shown in Fig. 9.18. The remote control panel (see Fig. 9.34) has the indicators “Net”, “On-board” and “Acc”, which indicate input voltage presence and their values. The “Net” indicator shows a 220 V voltage presence from the gasoline generator, or from the external electric mains connected

368

9 From power supply + 12 V

From on-board net

Rechargeable battery

Radio Monitoring Systems and Determination of RES Location

Converter1

Electronic switch

Converter2

Electronic switch

Converter3 Electronic switch Converter4 Accumulator charger unit

27 V/7A 15/19/21 V/5A 15/19/21 V/5A 15/19/21 V/5A

End-to-end supply module 1

10...16 V

End-to-end supply module 2

10...16 V

End-to-end supply module 3

10...16 V

Electronic switch From remote control panel

Control and automatic unit

Fig. 9.35 Structural diagram of ARK-UPS12 power supply unit

to the vehicle. The “On-board” indicator shows the presence of on-board voltage from the additional electric generator, while the “Acc” indicator shows the condition of the reserve rechargeable battery. Depending on the correspondence between the acting input voltage and the normal value, the indicators light in green or red color. If several input voltages are connected to the power supply unit at a given moment, the unit automatically chooses the input voltage, in order of priority: mains, on-board, reserve battery. If the charge of the reserve battery is activated at this moment, the indicator “Charge” lights. The remote control panel has an audio indication of emergency modes and, if necessary, the audio indicator can be switched-off by the sound-switching on-off button; the indicator lights in green color if the sound signal is allowed. The operator can select the voltage supply, which should be switched on or off, with the buttons U1–U9. In the case when the output voltage of at least one activated module does not correspond to normal – and the sound signal is allowed – the sound signal will be generated, signaling the emergency situation. On the numeric indicator “I, A”, the total current value, being consumed by all activated modules, is displayed, and on the indicator “U, V”, the voltage value at input of these modules is displayed. When pressing any button: “Net”, “On-board” or “Acc”, the indicator “U, V” shows the voltage and the current for the appropriate circuit. During the operation, the operator may switch the voltage at module outputs on and off, by choice, with the help of the U1–U9 buttons (all conditions are saved in nonvolatile memory, after switching-off the supply). In the electric supply of the “Argument” station, the charge mode of the rechargeable battery is realized at switched-off power supply. Battery charging may be provided at vehicle movement, from the vehicle generator or from the AC 220 V mains, upon parking. The charge activation is provided automatically at engine start,

Electric Power Supply Systems

369

at electric gasoline generator start, or at connection of the external voltage from the AC 220 V mains. For example, Table 9.5 illustrates the correspondence of the remote control panel buttons and the equipment for the “Argument” mobile station, connected to the power supply, and consisting of three posts.

Table 9.5 Interconnection of the remote control panel (RCP) buttons and the station equipment RCP button

Module number

Connect. number

Voltage, V

Addressing

U1 U2 U3 U4

5 4 3 6

1 1 1 1

15, 19, 21 15, 19, 21 15, 19, 21 12

U5

6

2

12

U6 U7 U8

7 7 8

1 2 1

12 12 12

U9

2

1

27

Post 1, PC supply Post 2, PC supply Post 3, PC supply Post 1, ARK-D1TP supply Post 2. Radio modem for data transmission supply Post 3. ARK-RD4 supply Supply of PC LAN hub Supply of ARK-KN1 (GPS) Supply of ARK-MK1M or ARK-MK4

Thus, the ARK-UPS12 power supply unit provides the electric supply for all equipment of the “Argument” mobile station, including the equipment for navigation and data transmission.

Autonomous Electric Station Usage For the supply of electricity to mobile and portable radio monitoring stations when in operation while the vehicle is parked, it is necessary to use autonomous electric generators. Autonomous gasoline electric generators (GEG) and solar batteries are the most widespread autonomous electric sources. Let us consider the application peculiarities of these autonomous sources as they apply to radio monitoring stations, in which the total electricity consumption of the radio equipment and the PCs does not exceed 400–500 W. A review of the catalogues of the main autonomous GEG manufacturers shows that there are no models with nominal load power less than 650 W. This is obviously caused by the fact that there is no profitable market for low-power generators. Moreover, their cost is commensurable with more powerful assemblies. GEG with power from 650 W to 1.5 kW can be divided into two groups:

370

9

Radio Monitoring Systems and Determination of RES Location

• Those with classical synchronous or asynchronous single-phase generators with engine rotation frequency of 3,000 rpm • Those with voltage generated by the inverter. In both the first and second GEG group, the four-stroke and double-stroke gasoline engines find application. GEG with classical synchronous and asynchronous single-phase generators and the engine rotation frequency of 3,000 rpm weigh, as a rule, from 19 to 22 kg (with an empty gasoline tank). GEG with 220 V network voltage, generated by the inverter, weigh 13 kg, slightly less than classical ones. This is achieved due to the fact that, upon increasing the rotation frequency of the generators, the size of both the engine and the generator decreases, and due to the formation of 220 V network voltage by the electronic converter. With high probability, the application of electronic converters (inverters) causes the presence of radio frequency interference, which may make similar GEG unsuitable for operation within the radio monitoring station structure. Thus, it should be noted that classical synchronous generators in collector-free implementation would also create high frequency interference. Therefore, application of the classical electric generator in radio monitoring stations, especially portable ones, is limited by the weight-size factor. But, the application of lightweight generators of the inverter type requires their testing, as far as convenience, from the point of view of possible radio interference. Which engine is preferable: the double-stroke or the four-stroke? The four-stroke engine starts easier than the double-stroke one and operates on pure gasoline, without the addition of special oil. Actually, it requires oil changing in the crank-case, after 25–100 working hours. The engine oil may be of the type for general application, hence non-expensive. Double-stroke engines are more lightweight compared to the four-stroke ones, but have one main shortcoming: they work on gasoline with the addition of special motor oil, usually in proportion of 1/30–1/50 of fuel volume. The oil for double-stroke engines (these engines are imported in Russia, even for Russian assemblies) can be bought easily in large cities only. The only native substitution is castor oil, which is applied in the same proportion. Thus, GEG with four-stroke engines are more preferable, due to operation reliability and the availability of expendable (lubricant) materials. At that, one should remember that the application of lightweight four-stroke assemblies with generators of the inverter type requires the obligatory generator nature testing on the radio interference absence. The cost range on GEG of the considered class is 250–1,000 USD. Electric generators of the lowest cost category are usually implemented on the basis of maximally-simplified double-stroke engines. They are quite suitable for the occasional reserve of AC 220 V, but they cannot be recommended for long operations. It should be noted that the actual engine resource of small electric generators, before any major repairs, as a rule, does not exceed 1,000 h (the resource of the vehicle’s engine, as a rule, is not more than 10,000 h). From this fact, one can conclude that, at operation during 12 h per day, the period before major repairs is not more than three months.

Electric Power Supply Systems

371

We can expect 2–3 times more engine resource from electric generators of the highest cost category, but, as is true for the low-cost models, strict observation of the operation rules must be followed. These rules include the application of the appropriate lubricant materials, the fuel, and the well-timed changing of air and fuel filters, as for the vehicle. At selection of the electric generator for station supply, it is necessary to take into consideration the possible operating modes. Besides the weight of the generator itself, we need to account for the weight and sizes of the necessary consumed materials. It should be remembered that the fuel expense at one-hour, even at free running, will exceed 0.3 l. The actual expense at the expected load of 150–500 W will be about 0.5 l/h. At round-the-clock work, this will be approximately 10 l/24 h, or 10 kg of additional weight per each 24 h of autonomous operation, taking into account the package weight. At possible autonomous operation in this mode for longer than 4 days and nights, it is necessary to change the engine oil. The maximal changing interval is 100 h, whereas for the vehicle, we would need to change the oil after each 10,000 km. We should note that the oil filtration in the electric generator engines, as a rule, is absent, in contrast to the vehicle engines. To ensure continuous operation, the reserve rechargeable battery should be included into the autonomous station. Its minimal capacity is defined by the duration of GEG refueling and its maintenance procedure and may be 3–10 A·h. At long autonomous operation, it is preferable to use a rechargeable battery of larger capacity in the station, enough for 10–12 h of continuous operation. If the station consists of a single post of radio monitoring and direction finding, the radio equipment of which consumes 20–50 W and the PC of which consumes 50–80 W, it is necessary to have a 12 V rechargeable battery with 120–190 A·h capacity, for 12 h of continuous work. Other modes are possible: for instance, when the station operates 4 h from the rechargeable battery and 4 h from the GEG, and, in the last mode, the battery charges from the GEG. In this case, the preferable operation mode would be when the system operates first from the battery, then from the GEG with the battery charge, and then again from the GEG, etc., since the GEG efficiency at full load is not large. The engine rotates itself and compensates for the losses on friction in the generator. Application of GEG simultaneously, for accumulator charge and to provide the electric supply of the complex, slightly increases the fuel hour expense, but allows GEG-stopping during long intervals. Activity factor can achieve 50% and less, allowing a decrease in the fuel reserve – approximately from 70 kg up to 35 kg for weekly autonomous operation, and that quite compensates for the accumulator weight with 110 A·h capacity and 12 V voltage. Moreover, the GEG lifespan does not exceed 100 h for one week, which allows for the increasing of the interval between oil changes, to a period of about 10 days. Another GEG alternative, especially for autonomous operation during more than one week, may be solar panels. According to the FARNELL catalogue, the solar battery produced by SAVER PRO Kit of ICP Co. with nominal power 30 W weighs 8.3 kg and provides an average energy output of 840 W·h per week. It consists of

372

9

Radio Monitoring Systems and Determination of RES Location

two modules with the sizes of 965 × 25 × 330 mm and is intended for operation with lead accumulators. As per the catalogue, single copies cost near 200 USD. Silent operation is the valuable property of solar batteries. Five to ten such batteries can provide for all of the complex’s operation, with average consuming power of 150 W, and with the simultaneous charging of the accumulator at daytime, for operation at night. The weight of ten batteries is near 80 kg, which, for weekly operation, is less than the total weight of the GEG with the fuel tank. In this situation, the power reserve is ensured, since the solar panels are connected in parallel with the accumulator and, at malfunction of their part, the complex’s efficiency will be saved. The more expensive solar batteries (3,600 USD for 10 sets) may be considered the prime competition to GEG, due to the absence of consumed materials. The cost of gasoline only, per one working hour, is about 0.3 USD. For the expected GEG resource (1,000 h), the gasoline expenses will be not less than 300 USD plus about 50 USD for the lubricant materials. Moreover, after 1,000 h, or even after 42 days of continuous work, GEG of the lowest cost category should be written off or, at a minimum, should be expected to incur large repairs. At the same time, the solar panels have a lifespan of not less than 2–3 years for the flexible panels, and up to 25 years for solid silicon panels.

Special Software Support and Operation Modes of Stations Software Support Structure All posts of the ARK-POM system use the same basic software set. At minimum, the software set consists of three application packages of customized mathematical software (SMO): SMO-PPK (SMO-PA, SMO-PAI), SMO-RMC and SMO-KN. The SMO-PPK package (customized mathematical software for panoramic direction-finding complex) includes the main application (SMO-PPK) and the application-controller (driver) for the equipment. The software package provides the execution of all typical functions of radio monitoring, including single-channel and multi-channel direction finding. For radio monitoring post equipment without the direction-finding function, a reduced version of the SMO-PA software (customized mathematical software for panoramic analysis) is provided, which has no direction-finding mode. The SMO-PAI software is the extended version of SMO-PA and provides the execution of measurement tasks, on equipment certified as measuring equipment. The SMO-RMC package (communication via radio modems) provides the data transmission between the posts, with the help of low-rate radio modems, and also the modems of CDMA or GSM cellular radio networks. In the case of high-rate data transmission lines, using the exchange is fulfilled directly by the TCP/IP protocol.

Special Software Support and Operation Modes of Stations

373

The SMO-KN package (cartography and navigation) is intended for the electronic mapping of the area. It provides the bearing-laying and the automatic determination of RES location on the basis of the direction-finding data, which is received from the SMO-PPK application or entered manually. SMO-ASPD software for post-processing and analysis (software for analysis of spectral bearing data) and SMO-STA technical analysis application (signal technical analysis) can be added to the software set. The software’s peculiarity consists in the fact that, if necessary, any of the system posts may operate autonomously or play the role of the central post of the whole system. Of course, the variant, when the stationary post of direction finding is the system’s central post, having the maximal height of antenna system installation and connecting with other stationary posts via high-rate communication lines and with mobile and portable stations by low-rate lines, is preferable. The software package supports the modes: “Spectrum”; “Search”; “Background review”; “Panorama”; “Playback”; “Measurement”; “Technical analysis”; “Review”; “Bearing”; “Multi-channel direction finding”; “Electronic map”; and the result post-processing modes.

“Spectrum” Mode The “Spectrum” mode is intended for representation of the radio signal spectral panoramas in the given frequency bands. To analyze the spectrum, we can use the equipment of the central post or any peripheral post. When using the central post equipment, all functions of the “Spectrum” mode are available, including the automatic search of new signals, listening and recording of the radio broadcasts. When using the peripheral post equipment, spectrum representation in the given ranges is available only. At that, the representation rate is defined by the pass-band capacity of the low-rate line, and it is substantially smaller than the spectrum panorama rate for the central post equipment. The software allows the representation on a single diagram of the whole operating frequency range of the equipment. Figure 9.36 shows the window of the SMO-PA application in “Spectrum” mode. In this case, the operator defined three frequency ranges: 106–108 MHz, 338–339 MHz, and 871–875 MHz. The “Spectrum” window is situated in the lower part of the screen. It displays the current and accumulated signal spectra and also the frequency-time diagram. The current spectra are shown in the dark color, the accumulated spectra in white color. The accumulated spectrum plot consists of spectral components with the maximal amplitude, which appeared in the current spectrum during the analysis time interval. Such plot type registers all parts of the analyzed frequency ranges, where the radio station emission was observed at least once. In the middle part of the “Spectrum” window, the frequency-time diagram (FTD) for radio-ranges loading is presented. The FTD plot represents the dynamics of radio equipment operation. The time variations of the “signal” spectral component levels that exceeded the detection thresholds are registered. The frequency is marked

374

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.36 SMO-PA application window in “Spectrum” mode

along the FTD horizontal axis, while time is noted along the vertical one. The spectral component levels are represented in the form of regions, the colors of which correspond to the color-marking of levels on the current spectrum’s ordinate axis. If the signal level is less than the level of the detection threshold, the area is painted over by the color corresponding to the minimal possible amplitude value. The FTD is recorded into the file of spectral and bearing data (SBD), for further analysis with the SMO-ASPD application. In “Spectrum” mode, any range section can be observed in the bandwidth 2, 5, and 10 MHz, depending on the equipment. Higher spectral resolutions can be obtained in “Measurement” mode, when there is the spectral lens mode. As an example, in Fig. 9.37 the signal spectrum of a trunking system is shown, for observation bands 2 MHz and 25 kHz.

“Search” Mode In “Spectrum” mode, the “Search” window is displayed on the PC monitor over a FTD, which is intended for the search and detection of new radio emission sources. The channel is considered active if its spectrum exceeds the detection threshold. The channel bandwidth is defined as the zone inside of which the spectral components are larger than the threshold.

Special Software Support and Operation Modes of Stations

375

a

b

Fig. 9.37 “Altay” system signal spectrum: (a) in 2 MHz band; (b) in 250 kHz band; (c) in 25 kHz band

Search frequency ranges are defined by the fulfilled task. The found channel parameters are saved in a database and represented in a “Search” window table. During the spectrum analysis, the parameter values of the found channels are corrected. At that, the maximal bandwidth and the maximal amplitude of the spectral components, which were observed in this channel, are registered.

376

9

Radio Monitoring Systems and Determination of RES Location

For automatic search of the active channels, the software has the possibility of using several algorithms: • • • •

With fixed threshold With floating threshold On accumulated spectrum On exceeding the reference spectrum.

The algorithm “with fixed threshold” uses the detection threshold representing on the monitor in the form of a red horizontal line. The channel is considered active if the spectral samples of the current spectrum exceed the detection threshold. The channel bandwidth is defined as the frequency range, where the spectral component amplitude exceeds the threshold. As soon as the spectrum sample exceeds the threshold, the channel’s left boundary is fixed. As soon as the further spectrum sample, located higher on the frequency, becomes smaller than the threshold, the channel’s right boundary is fixed. The algorithm “with floating threshold” uses the detection threshold calculated separately for each frequency interval, which is equal to the receiver bandwidth. For the ARK-D1TP panoramic measuring receiver, this interval is equal to 2 MHz. The threshold value is calculated based on the current spectrum as the sum of the value estimation of the noise spectrum components in the bandwidth and the selected detection factor. The advantage of the algorithm with the fixed threshold is obvious. The user himself can set the detection threshold to the necessary position. However, if the search is executed in the wide range, in its separate regions, the noise spectral components may have different values. Therefore, the settled fixed threshold value for some regions may be overestimated (one will observe the omission of weak stations), while for the other regions they may be understated (noise spikes will be set as the station signals). Thus, for the ranges with non-uniform spectrum of the noise components, it is expedient to apply the algorithm with the floating threshold, as in this algorithm the detection threshold is adapted to the noise value. However, we need to be attentive, when selecting the detection factor. The larger the detection factor value is, the smaller the algorithm sensitivity is, and vice versa: the smaller the value is, the larger the sensitivity is, which can lead to false detection. Usually the detection factor value is set in the limits of 10–15 dB. The algorithm “on accumulated spectrum” for the channel search uses not the current, but the accumulated spectrum. In this algorithm, the fixed detection threshold is used as in the algorithm with fixed threshold. The application of the accumulated spectrum leads to the fact that the bandwidths of the channels found during the search can increase only, which sometimes leads to adjacent channel merging. This algorithm is useful for entry into the database of all stations, a channel that even once emits a radio signal. The algorithm “on exceeding the reference spectrum” uses the exceeding of the current spectrum over the reference spectrum as a criterion of signal detection. As a reference spectrum, we may use the accumulated spectrum saved in the file, or the current accumulated spectrum (the application will offer to use it, if, at the moment

Special Software Support and Operation Modes of Stations

377

of search activation, the reference spectrum was not loaded from the file). As a detection threshold, the sum of reference spectrum values at this frequency and the detection factor is used. The detection threshold curve is represented on the spectrum plot. The given algorithm is expedient to use for the new station search. To correct the reference spectrum, there is a reference spectrum editor inside the application. During the search, the channel’s central frequencies and bandwidths are dynamically corrected. The algorithm of recurring averaging of left and right channel boundary values is used for the correction. If the adjacent frequency channels overlap, they are considered, in the future, as a single channel. At the detection time of such channels, the detection time of the channel found firstly is assumed. In all search modes, the “Background review” mode is available. If “Search” mode is executed and “Background Review” is activated, at the next new channel detection, the search process is interrupted, the equipment switches into the scanning mode and executes the “Review” mode once, over the frequency of the found channel. At “Background Review” mode, as the detected source response, we can set the record of the detected source spectrum, the audio record during the given time interval, and the signal time sample record, for further technical analysis. After single scanning of the found channel, the channel search continues. If the “Cyclic review of active channels” option is activated, the “Search” mode will cyclically repeat for the channels exceeding the detection threshold. We should note, though, that if there is a single receiver in the system, its permanent transition to the scanning mode will decrease essentially the search rate for new channels. In “Review” mode, the channels from the “Search” table can be imported to the frequency table of the tasks for “Review”. When the ARK-POM system is working in “Background Review” mode, the task distribution among the posts is executed as follows. The equipment of the central post, in accordance with the given algorithm, provides the search of new or active-at-the-moment radio communication channels. At regular channel detection, the search process is temporarily interrupted and the “Review” mode is executed over the frequency for the detected channel. At that, as a response, the equipment of the central post sends the bearing value, a photo picture of the detected source spectrum, the audio sample, and the signal time sample on the intermediate frequency, to be defined for further technical analysis. To reduce the volume of the transmitted data in case of low-rate data transmission lines, the peripheral posts return the bearing value and the signal amplitude value only. After single scanning, the search process continues. The bearings obtained were transmitted to the cartographic application, to determine RES location.

“Bearing” Mode “Bearing” mode is used for obtaining the synchronous bearings at the given frequency, from all system posts, and also for listening and recording radio broadcasts. In this mode, the post equipment executes RES direction finding at a single frequency. Figure 9.38 shows the SMO-PPK application window in “Bearing” mode.

378

9

Radio Monitoring Systems and Determination of RES Location

Fig. 9.38 “Bearing” mode window

The figure represents the direction-finding process of a basic DAMPS standard station by two posts of the ARK-POM system. At that, the current spectrum and bearing values are received from the system’s central post, while, from the peripheral post, the bearing and signal amplitude values only are received. The spectrum and bearings are represented on the PC monitor. The bearing history and the signal amplitude history are represented as well, which allows the estimation of the dynamics of RES parameter time variations. It is possible to represent the spectrum from any peripheral controller, but, in that case, the direction-finding rate decreases. “Bearing” mode ensures RES direction finding at the given frequency with the maximal operation speed. The bearing number from the peripheral post depends on the data transmission rate via radio line. At a data transmission rate of 19,200 bit/s from the peripheral post, one can obtain up to 5–10 bearings per second.

“Measurement” and “Technical Analysis” Modes “Measurement” mode, the window of which is shown in Fig. 9.39, ensures the automatic measurement of modulation parameters of radio signals, and measurement of peak values, power, and electromagnetic field strength.

Special Software Support and Operation Modes of Stations

379

Fig. 9.39 “Measurement” mode window

For further signal technical analysis, for its decoding and demodulation – including with the help of other customized software applications – there is the possibility of received signal recording on the receiver intermediate frequency. At that, permanent radio signal recording at the given bandwidth is supported. In Table 9.6, information is shown for the determination of the required PC hard disk capacity, depending on the frequency bandwidth of the recorded signal. Table 9.6 The required volume of free disk space depending on the bandwidth of the recorded signal Signal bandwidth, kHz

Discretization frequency, MHz

Signal sample type

Sample capacity, bit

Required disk volume for 1 s interval, Mbyte

2,000 250 100 50 0.025 0.0125

6.4 0.25 0.1 0.05 0.025 0.0125

Real Real Complex Complex Complex Complex

16 16 32 32 32 32

12 2 0.8 0.4 0.2 0.1

Automatic measurement of radio signal parameters and technical analysis of radio signals were described in Chapter 6 of this book. In SMO-PPK (PA) software,

380

9

Radio Monitoring Systems and Determination of RES Location

calculation of the field strength in the given bandwidth can be fulfilled on the basis of spectral data and directly on the time sample, with the help of analogues of the peak, quasi-peak, and RMS detectors. The results of the field strength measurement are shown in the PC monitor, as Fig. 9.40 illustrates. In the left part of the “Measurement” window, the histogram of the field strength measurement distribution is displayed, and the current strength value is represented in the information window (in the upper screen, to the right).

Fig. 9.40 Field strength measurements of CDMA signal

In SMO-PAI application, the recording of the field strength measurement results in the file format. When the GPS receiver is connected to a PC, this file will also contain the geographical coordinates of the mobile station. This allows a protocol to be formed for measurement of the coverage zone by the transmitter signal, with the fixing of the geographical coordinate. In the text file of the measurement protocol, the following is included: measurement number, date, time, the latitude, the longitude, the authenticity of the geodesic data, the velocity and direction of movement, the signal frequency and its bandwidth (in MHz), as well as the most probable signal level value. If the SMO-KN cartographic application is activated, the field strength measurement results can be transmitted to this application through the network – with TCP/IP protocol – for processing and representation of the electromagnetic field strength distribution on-site.

Special Software Support and Operation Modes of Stations

381

“Review” Mode Automatic signal monitoring of revealed RES is executed in “Review” mode. This mode is intended for the automatic observation of the radio frequency group (sources), and it ensures the following possibilities: • Detection of the station appearance, the registration of activation time and signal amplitude in the database • Storing the bearing value to the database • Storing the image (photo picture) of the spectrum panorama to the database, at the moment of station detection • Saving the signal time sample in a data file • Recording the non-modulated radio transmission of the detected source • Observation of the database of the earlier detected sources • Calculation of statistical parameters on the review results. “Review” mode operation is fulfilled in accordance with the task, which consists of the frequency list, along with the indication of the detection parameters and sources registration. The frequency list may be formed manually or automatically by the frequencies found during the search, in “Spectrum” mode. For the formation of a summarized statistical table with review results, the "Statistics” page of the “Settings” window is selected. It is shown in Fig. 9.41. The mentioned parameters will be tabulated in the table of the “Statistics” window.

Fig. 9.41 “Statistics” page of the “Settings” window

382

9

Radio Monitoring Systems and Determination of RES Location

A database review is executed by means of browsing in the “Scanning Result” table, with the help of the mouse or the keyboard. After review activation, the detected radio stations will appear in the response table. In this table, the following is indicated: the identifier (the number) of record in the database; controller number, which obtained the given result; date and time of signal detection; the frequency value, in accordance with the task; signal amplitude and bearing. In the last table column, the response components (results) are presented, obtained from the controller: the signal spectrum photo picture, the sound, and the signal time sample. In “Review” mode, there is the possibility of automatic determination of the radio signal modulation type and parameters – by the short time samples obtained during the review, as in “Measurement” mode. Moreover, there is the possibility of statistical results processing, including by the time samples, with the purpose of the necessary parameter calculation, for example, the current frequency deviation from the nominal value and determination of radio-range loading. While in operation in the ARK-POM system structure, in “Review” mode, we can acquire only those responses obtained from the peripheral posts and relayed to the central post via low-rate radio channels: namely, the bearing values and signal amplitudes. Upon use of the high-rate channels, the spectral diagrams, signal time samples on IF with the given bandwidth and duration, and audio files from the demodulator output can all be used as responses.

“Multi-Channel Direction Finding” Mode Special attention should be paid to the “Multi-channel direction finding” mode in the ARK-POM system, since it allows the execution of POFT signal direction finding. This mode is supported in “Spectrum” mode, as a variety of the active channel search algorithm. This mode is intended for automatic direction finding of active RES in the given frequency range (or in frequency ranges). The equipment of the posts is consequently frequency-retuned from the lower task boundary to the upper boundary. The retuning step is equal to the bandwidth of the receiving section (band of instantaneous review), which is 2.5 MHz or 10 MHz depending on the receiver modification. The receiver’s signal processor calculates the complex spectra of the signals over all bandwidths, at each tuning frequency of the receiver. On the basis of the calculated complex spectra, the detection and direction finding of active source signals are executed. The algorithm with the floating threshold is used for signal detection. The threshold value, given as the exceeding value over the noise level, is adaptively adjusted under the level of noise components in the given band, which increases the true detection probability and decreases the signal miss probability. The central frequency, the bandwidth, and the angles-of-arrival in horizontal and vertical planes are automatically estimated. Thus, in “Multi-channel direction finding” mode, the spectrum panorama formation, the source detection, and the bearing calculation for each source are executed at each tuning frequency, for the whole bandwidth at once. Compared to the single

Special Software Support and Operation Modes of Stations

383

direction-finding mode, where the receiver consecutively tunes on each RES frequency, this method considerably quickens the time of bearing calculation. After the upper task boundary is achieved, the receiver automatically returns to the lower boundary and the multi-channel direction-finding process will continue. The speed at which the bearing panorama is obtained depends on the operation speed of the radio receiver and the bearing algorithm. For the basic equipment of the ARGAMAK receiver with a bandwidth of 5 MHz and a nine-element antenna array at signal detection on the receiver’s sensitivity level, it is about 300 MHz/s. There may be 200 found sources with spectrum width 25 kHz, in the single band of 5 MHz. It is practically impossible to transmit the whole volume of obtained data from the peripheral posts through low-rate radio lines. The large volume of information being analyzed makes inexpedient its saving in the RAM of the equipment. Therefore, the multi-channel direction-finding results at the posts are recorded in compressed form in special files of spectral-bearing data (SPD-files). At that, transmission of the bearing result from the system’s peripheral posts to the central post is provided at the request of the central post only. There are several request formats: for instance, a specific frequency request, in which the time interval is indicated and, for this interval, the bearings should be transmitted, and the signal bandwidth. This request is used upon determination of the location of short-term operating RES, whose frequency is already known. For requests related to sector, the angle of the sector is given, and a request is made to return the bearings, the central frequencies, RES signal levels and spectral widths located in this sector.

Peculiarities of the Direction Finding of POFT Stations At implementation of the correlation-interference bearing method on the basis of the double-channel receiver, the bearing time of a single RES is defined by the time of receiver tuning to the given frequency and by the interrogation time for the antenna array pairs, in accordance with the bearing algorithm. When using the ARK-PR5 ARGAMAK receiver, the tuning time to the given frequency does not exceed 2 ms. When using the nine-element antenna array, at sample duration 0.32 ms, the minimal bearing time is equal to about 5 ms. At such bearing time, it is possible to find RES direction, with a retuning speed of 200 jumps per second. If RES emission duration is less than the time required for interrogation of all antenna array pairs, POFT signal direction finding becomes impossible, without modernization of the bearing algorithm. Meanwhile, at present, the situation, where the communication radio stations have the retuning speed of 300 jumps per second and larger, is not rare. The situation is worsened by the fact that the tuning range F of RES with POFT signals, as a rule, is significantly larger than the receiver bandwidth of instantaneous review. As was previously mentioned, at multi-channel direction finding, the post equipment retunes consecutively from the lower task boundary to the upper task boundary. The retuning step is equal to the bandwidth of instantaneous review,

384

9

Radio Monitoring Systems and Determination of RES Location

which is 2.5 MHz or 10 MHz depending on receiver modification. Direction finding becomes possible only in the case, when the frequency position (FP) of the POFT signal falls into the receiver’s instantaneous review. The fundamental way of decreasing direction-finding time is the application of mono-pulse direction finders, in which the receiving channel number is equal to the antenna array element number. On the basis of the ARGAMAK digital radio receiver, multi-channel coherent radio receivers were developed. Modification of the correlation-interference algorithm is possible, which does not require the receiving section number increase, but giving the bearing possibility for RES with high-rate POFT, due to the bearing time increasing. In essence, this consists in the consecutive multi-pass aggregation of amplitude-frequency distributions from the antenna array elements. In this algorithm, for each frequency position (FP), the bearing is taken independently, i.e., the signal vectors are averaged and accumulated in the limits of the single channel of the signal with POFT. Therefore, after long bearing of the range, when sufficient number of the signal vectors will be collected for each FP, the direction finder generates N bearings, where N is a FP number in the observation range. This algorithm allows direction finding for several active transmitters, which use the various frequencies of the mutual range for their operation. The considered algorithm is realized in the ARK-MP4, ARK-MP1, and ARK-P7 equipment software. Experimental testing of the algorithm showed that it ensures the direction finding of both the signals with POFT, and the signals having a high off-duty factor, e.g., the mobile phone signals of GSM standard. Figure 9.42 shows the SMO-PPK application windows in “Multi-channel direction finding” mode. We can see the table of detected channels on the screen, together with the bearing values, the spectral and time-frequency signal diagrams in the range of 363–403 MHz, the circular bearing diagram, the spectral diagram, the bearing and the correlation curve of one of the detected signals. In the lower part of the figure, the accumulated spectrum plot is presented in white color, and the instantaneous spectrum plot in the more dark (blue) color. The accumulated spectrum plot consists of the spectral components with maximal amplitude, which appeared in the current spectrum during the analysis time. Such type of plot fixes on all sections of the analyzed frequency ranges, where the radio station emission was observed at least once. The frequency is this plot argument and the instantaneous spectrum plot, while the maximal values of the spectral component amplitudes are marked along the vertical axis. The plot of the time-frequency diagram (TFD) is presented over the accumulated spectrum plot. The time-frequency diagram plot represents the loading of the frequency ranges by the radio wave sources. The time variation of the “signal” spectrum component levels exceeding the detection thresholds is fixed on the TFD. At that, all spectrum sections, where the spectral component level exceeds the detection threshold at least once, are shown. The frequency is shown along the horizontal axis of the TFD, while the time is shown along the vertical axis. The spectral component levels are represented in the form of the areas, the color of which corresponds to the color marking the levels of the current spectrum ordinate axis. If the signal level is

Special Software Support and Operation Modes of Stations

385

Fig. 9.42 Multi-channel direction finding of POFT signals

less than a threshold, the TFD section is painted with the color corresponding to the minimal possible amplitude value. On the right-hand side of the spectral diagram, the limb of the bearing circular panorama is located, with the results of the multi-channel direction finding. The found RES are presented in the form of points. The angular position of the point corresponds to the bearing value (the angle is counted out clockwise), and the radial position corresponds to the RES frequency. The frequency marker and the azimuth marker are on the circular panorama. The frequency marker has the form of a circle, the azimuth one has the form of a beam. The point where the frequency and the azimuth markers intercept corresponds to the RES, on the position of which there is an indicator in the detected signal table. As we can see from the accumulated spectrum plot, there are sixteen RES with the similar emission level, and it follows from the TFD and from the instantaneous spectrum diagram that, at each time moment, one of the sources is presented; moreover, all source bearings have the same azimuth, which is equal to 23◦ . All mentioned features are evidence of the fact that direction finding is executed for RES signal with jumping frequency variation.

“Electronic Map” Mode The following main requirements are set for cartographic system functioning, within the structure of the radio monitoring system:

386

9

Radio Monitoring Systems and Determination of RES Location

• Possibility to use both vector and raster maps; this requirement is related to the lack of vector map market development at present • Minimal resources for operation, including the main memory and processor time. This demand is related to the necessity of the system to operate on a single computer simultaneously with several applications, intensively using the memory and computing resources • Operation speed at repainting during the scaling process or the map rolling • Presence of the development and distribution means of cartographic applications, including API and/or component for Delphi/C++ Builder, for development, and DLL/ActiveX, for distribution • Independence of the district map and user data • Low licensing cost for the workplace.

At present,“ARCView”,“MapInfo”,“GeoConstructor”, GISToolkit “Panorama”, and GWX ActiveX Control are the most known geo-information systems (GIS) present in the Russian market. After preliminary selection, trial projects in Russia were fulfilled on the basis of “GeoConstructor” and the GISToolkit “Panorama”. The high operation speed and low demands on the memory of the application based on GISToolkit “Panorama”, and also the possibility of transfer to the Linux platform, were the critical arguments in favor of its selection. Another advantage of GISToolkit “Panorama” is its possibility to transform various vector (MIF, DXF, S57) and raster (PCX, BMP, TIFF) formats to the main format of data representation, with the help of the “Map 2000” application package. Additionally, the fact that the basic exchange format of data representation of GISToolkit “Panorama” (format SXF) was developed in 1992 by the experts of the Topographic Agency of Russia, and was approved in 1993 as the main format of digital information on the district, at the Armed Forces and for a series of Federal Agencies, made it possible to directly use the already existing cartographic resources. Lastly, the possibility to represent several user maps (layers), in addition to the main district map, is also the doubtless advantage of the GIS “Panorama”. Taking into consideration the demands to the systems of RES determination on the basis of GIS “Panorama”, the GIS SMO-KN was developed, which is intended for application in the structure of ARK-POM systems. It includes the sub-systems for location determination and for RES representation. In SMO-KN, the user data are saved in the working map. For each operation session, individual working maps may be created. The working maps of radio emission are used for aggregating the information about registered sources and the sources found during the various operation sessions. The district maps depend on the working maps. The distinctive peculiarity of the object representation system for the given district is the necessity to show the bearings from several points, displaced on tens of kilometers. Moreover, the system should give the possibility of detailed district investigation of these objects’ supposed location. Therefore, in GIS SMO-KN, the lens window is realized and the observation point storing is fulfilled, including the initial scale.

Special Software Support and Operation Modes of Stations

387

In map-drawing system structure, a specific database is included, which allows for the saving and future use of the information on the bearings, the measured field strength values and the detected emission sources. When mobile stations are used in the structure of radio monitoring systems, the dynamically-varying navigation information is taken into consideration and represented, i.e., the data on the location and azimuth of the mobile station. The application calculates and represents RES location, its movement routes (in case of mobile sources), fixes – in the history file – the radio monitoring station location, the received bearings, and the measured values of field strength. The system can operate in the multi-post mode. Figure 9.43 shows the SMO-KN application window with the results of the location determination of the stationary RES by the ARK-POM system consisting of three posts.

Fig.9. 43 SMO-KN application window

The possibility of data aggregation and its comprehensive analysis is the necessary condition for the effective operation of the location determination system. In the SMO-KN application package, all the bearing and navigation data entered into the application are saved in the history database. The following possibilities exist, to analyze the aggregated data: • Bearing selection for frequency calculations • Selection for the bearing calculations, passed through the selected area

388

9

Radio Monitoring Systems and Determination of RES Location

• Bearing selection for the calculation of the arbitrary history interval, by means of allocation • Rolling the direction finder movement route with graphical and text information representation • Complete imitation of the operation session, when the aggregated data passes to the application input from the history file • Calculation of RES location, taking into account the bearing selection criterion and the representation of calculated points • Calculation of RES location for all frequencies at once, including in the database. The analysis of the aggregated data, the calculation of RES location and its representation on the map, and the imaging of the movement paths of the mobile stations are executed both in real-time scale and in the further (post) processing mode.

Post-processing Mode For the post-processing of spectral and bearing data accumulated in the system posts, the SMO-ASPD application is used. The initial “material” for this application operation are the files of the spectral (bearing) data (SPD) recorded by the SMO-PPK (PA) application. The SMO-ASPD application represents the spectral and bearing data, in the form of diagrams with regulated frequency, time and angle sector resolution. With the application’s help, one can execute a search of the communication sessions, a RES search on frequency or in the given angle sector, and estimate radio communication effectiveness, as well as the loading of the various radio frequency ranges. Fragmentation of the frequency range into radio channels is the core process of statistical data processing from SPD files. Information on the observed data in each channel in the communication sessions is gathered during the analysis. This allows the calculation of such radio channel indices as the workload and intensity of the channel usage, as well as the average duration of the communication sessions. Thanks to the fact that processed SPD files, as a rule, contain the results of the wide-band monitoring for the radio environment, SMO-ASPD application allows the parallel estimation of the statistical properties of hundreds and thousands of radio channels. The frequency grid defining the frequency axis fragmentation into radio channels should be known before starting the analysis procedure. Information on the frequency axis fragmentation into radio channels and the processing indices of the different channels is stored in the application database. This information includes: • Data on centering and width of radio channels, to define the frequency axis fragmentation into the separately analyzed frequency sub-ranges • Minimal possible signal duration, to allow the suppression of random spikes of spectral activity, which occur due to the inaccuracy of threshold setting or due to transients in the equipment

Special Software Support and Operation Modes of Stations

389

• Minimal pause duration, to avoid the registration of temporal signal-missing (e.g., due to fading) at the completion of the previous, and start of the new, communication session. Statistical processing is executed in two stages. At the first stage, determination of the communication session number and their time determination are executed. To increase the reliability of communication session determination, a priori information on the possible duration of the signals and pauses in the radio channels is used. At the second stage, the final indices calculation is executed for the radio channels. The first processing stage begins just after the loading of the new SPD file. Following the given frequency axis fragmentation into the radio channels, the application executes the time-variation analysis of the signal intensity in each channel, with the aim of determining the most reliable distribution of signals and pauses. The stage of time boundary determination of the channel activity is completed with the generation of the detected signal table. SMO-ASPD application is oriented on the wide-band analysis of the radio environment, therefore, the detailed report on the properties of all analyzed radio channels may be excessively large. In connection with this, at the second stage, the application offers the following two main forms of the analysis result representation: • Graphical information diagrams presenting the most general properties of the analyzing radio channel ensemble • Detailed tabular reports, formed for the given-in-advance list of the monitored frequencies (radio channels).

Fig. 9.44 SMO-ASPD application window for the file containing bearing data processing

390

9

Radio Monitoring Systems and Determination of RES Location

In the lower part of the window, the current spectrum is presented for the selected time moment, while the middle of the window contains the time-frequency diagram and the upper part of the window shows the accumulated spectrum diagram and the plot of average radio-channel loading are plotted. With the help of the SMO-ASPD application, one may execute the bearing data processing. As an example, Fig. 9.44 shows the case, when the current spectrum, TFD, and the current bearings are represented in the application window, for a section of the radio range.

Conclusion This chapter presents the structure, construction, management organization and functions of the software for the systems of multi-position radio monitoring, and for the automatic determination of emission sources location. It has been shown that similar systems can be constructed on the basis of three types of stations: stationary, mobile, and portable (deployable). The presence of additional handheld direction finders for the ranges of 0.3–25 MHz and 3–8 (18) GHz is necessary. The quite significant factor defining the efficiency of the multi-position system is the presence of a reliable sub-system of data transmission. To exchange the data between the stationary posts, it is expedient to use high-rate data transmission channels, while to organize the communication with the mobile station and with the deployable posts – the low-rate systems. In the last case, depending on the requirements and system purpose, it is possible to use autonomous narrow-band radio modems or radio modems of the cellular radio communication systems. The system’s peripheral posts may operate in the remote control mode. Equipment included in the discussed system versions has a unified construction. The equipment’s base is the ARGAMAK portable digital panoramic radio receiver. Its technical features ensure the system’s effectiveness as a whole. In spite of the portable implementation of several units, the ARK-POM system has the high-rate review and direction finding parameters, as well as sufficient sensitivity and large dynamic range. The stations of the geographically-distributed system should have navigation equipment, the accuracy of which should be matched with the direction-finding equipment accuracy. It is shown that the most preferable type is the combined navigation equipment developed with the use of satellite navigation systems and navigation systems of the inertial type, and, in several cases, with magnetic navigation devices added. The reliability of the equipment’s power supply system is a necessary condition for the uninterrupted operation of the complex. In this chapter, the problems of radio equipment power supply are considered, both during movement and upon parking, the advantages of pulse power supply application are discussed, and recommendations on the autonomous power stations for the supply of mobile and portable stations are given.

References

391

The unification of the radio equipment for the ARK-POM system leads to the usage of unified software packages at all posts. This accelerates the system post installation, simplifies the operator training and the work. The software of the ARKPOM system comprises all necessary functions for direction finding, determination of location, recognition, RES parameter measurement for arbitrary transmission types – including radio stations with frequency jumps, and maintains the distant control of the system posts both via the high-rate and the low-rate communication channels. In the latter case, transmitted data volume reduction is used, due to data compression. Hardware-software complexes included in the system have the mode of radio signal technical analysis. It is executed both in real time and in the mode of future (post) processing. The post-processing of the recorded radio signal time samples, and the spectral and bearing diagrams ensure the application demodulation in the selected channels, the search of communication sessions, RES search on frequency or in the given angle sector, the radio communication intensity estimation and the loading of the various radio ranges. The system software implements a valuable function set allowing the solution of most radio monitoring problems. The processing and saving of the essential information array obtained from the geographically-distributed posts, the operation with cartographic, speech, spectral and time data in the modes of real time and for postprocessing, and also the friendly interface ensure the effective application of the ARK-POM system, for the purposes of radio monitoring, and radio engineering intelligence.

References 1. Spectrum Monitoring Handbook, ITU-R, Geneva, 2002. 2. Grewal, M.S., Weill, L.R., and Andrews, A.P., Global Position Systems, Inertial Navigation, and Integration. John Wiley & Sons, Inc., 2001, 409 pp.

“This page left intentionally blank.”

Chapter 10

Radio Emission Source Localization Using Mobile Stations and Field Strength Measurement

Introduction The determination of RES location using the mobile direction finder, under urban conditions, is not a trivial problem. The complexity of this problem is explained by the fact that, under urban conditions, the electromagnetic field received by the direction finder includes the additional random component ensemble caused by the reflections from local objects. Among them, we may mention irregularities of the topography, buildings and other structures, made of various materials, the electric transmission lines, automobiles and other objects, as well as diffraction. Since these components have random amplitudes and phases, the summary field will also randomly change. At station movement, the varying interference pattern will be observed, when the received signal level changes, relative to some average value. The situation is compounded by the fact that, in most cases, there is no straight visibility, under urban conditions, between the antenna of the found RES and the mobile direction finder. We consider the direction-finding results for RES under urban conditions, by the “Argument” mobile station. The bearing error for this station, at frequencies higher than 100 MHz, has the value of 3◦ , at that, with frequency increase, the error reduces, and at frequency 900 MHz does not exceed 1.5◦ . In Fig. 10.1, the movement route is shown by the white curve. The mobile complex was moving from point 1 to point 2. The movement velocity was approximately 40 km/h. During time of motion, RES direction finding was executed, and the current values of bearings and direction finder coordinates were saved in a database. A radio emission source of small power was located in one of the buildings and operated at frequencies of 126, 353, and 900 MHz. In Fig. 10.1, RES location is denoted by point 4. After route passing, the probability estimations of the bearing value deviations were calculated, with respect to the true direction. The calculation results are presented in Fig. 10.2 in the form of probability P of bearing α obtaining versus the bearing error = |αtrue − α|, where αtrue is the true azimuth from the mobile station to the emission source.

A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_10,

393

394

10 RES Localization Using Mobile Stations and Field Strength Measurement

Fig. 10.1 Movement route of the mobile station and RES location

P 0.7 0.6 3

0.5

2 0.4 0.3

1

0.2 0.1 0 2.5

5

7.5

10

15

22.5

Fig. 10.2 Probabilities of true bearing calculations versus the direction-finding error and the signal frequency f

Curve 1 is plotted on the bearing calculation results at a frequency of 900 MHz, curve 2 – at a frequency of 353 MHz, and curve 3 – at a frequency of 126 MHz. As we see from the presented curves, at all three frequencies, the probability of bearing determination with an error less than 2.5◦ does not exceed one tenth, while for an error less than 5◦ it does not exceed two tenths. When frequency increases,

Introduction

395

this probability evidently reduces, which can be explained by the over-reflection number growth. Therefore, RES direction finding under urban conditions relates to the serious difficulties caused by multipath signal propagation and by the possible absence of straight visibility, and reliable estimation-obtaining for RES coordinates is impossible without using the perfect methods of statistical processing of the direction finding results.

Methods of RES Localization Using the Mobile Station In the “Argument” mobile station, several of the following methods for RES location determination are used [1]: • Drive method (homing method) • Quasi-stationary method • Method of calculating RES coordinates at motion. Additionally, combined methods are possible, when two or three abovementioned methods operate simultaneously.

Drive Method The drive method is based on mobile direction finder movement into the RES location zone, along the line of the bearing. When the distance to the RES decreases, the amplitude of the found signal increases, which is the additional sign of direction finder motion to the true direction. When using the drive method, the single-channel mode of RES direction finding is usually used. Figure 10.3 shows the SMO-PPK application window in the singlechannel direction finding mode, during RES search by the drive method. In the single-channel direction-finding mode, there are features that facilitate a RES search for the operator, by the drive method. These are the amplitude history plot (located to the left upward) and the bearing history plot (located more right). The history plots have the mutual ordinate, which is used for the time. The abscissa of the amplitude history plot is the level of the found signal in dB, while the abscissa of the bearing value history is the bearing value in degrees. The history plots allow the tracking in time for variation dynamics of the amplitude signal values and bearing values from RES. Moreover, the bearing distribution density curve on direction (bearing histogram), the maximum of which corresponds to the most probable bearing arrival, is represented on a limb, together with the current bearing values. The maximum direction varying of the bearing distribution curve is fixed on the bearing history plot. Under conditions of strong interference, the maximum direction shows the preferable direction of the direction finder movement.

396

10 RES Localization Using Mobile Stations and Field Strength Measurement

Fig. 10.3 SMO-PPK application window in “Bearing” mode

The amplitude history plots and bearing history plots are shown in this window (Fig. 10.3), and were obtained while the station was moving from point 4 to point 2 (see Fig. 10.1). The signal amplitude increasing from –50 dB/mV to –30 dB/mV and the bearing values in the front sector are evidence of the fact that, at first, the vehicle appeared headed to the RES and, after that, the RES was at the left side of the movement direction, since the bearings move to the rear left sector and the amplitude began to decrease. Now, for the exact orientation to the RES, the vehicle should turn around 180◦ and return back. The advantages of the drive method are the relatively small time for search and the reduced influence of errors, caused by the surrounding objects, since the complex is moving all the time. The method’s shortcomings are the necessity of RES periodical emission during the search and the possible miss of the fulfilled operation concealment, due to the station arrival to RES, within small distances.

Quasi-Stationary Method This method is based on the obtaining of several separate bearing measurements from the fixed points, which are situated at a considerable distance from the object (Fig. 10.4).

Introduction

397

Fig. 10.4 Quasi-stationary method of RES location determination

At this approach, the place of the mobile direction finder location is selected, and it might be far from any systematic interference sources: high buildings, electric transmission lines, tram ways, metallic fences and similar objects. It is expedient that the direction finder is located at the prevailing eminence. The mode of aggregation and formation of bearing histogram is activated. During some interval, the bearing aggregation is executed. The bearing directions are formed on the map based on one or several bearing histogram maximums. Then the direction finder is moved to the next position and the direction finding repeats. The calculation of RES location is fulfilled on the basis of the bearing formation results from the several stationary positions. If necessary, the operator may select the positions and bearings for the calculation from the database. For direction finding, the quasi-stationary method allows the use of the external antenna system deployed on the mast, which provides high sensitivity and directionfinding accuracy compared with the antenna on the vehicle roof, and ensures a larger radius of the activity zone. The advantages of the quasi-stationary method are the found results obviousness, the possibility of operation at large distance from RES, which facilitates the concealed execution of the direction-finding procedure. The method’s shortcomings

398

10 RES Localization Using Mobile Stations and Field Strength Measurement

are the necessity of “good” position selection for RES direction finding and the deployment of the antenna system on the mast, which, as a rule, causes great time expenses, and the possibility of erroneous bearing readings due to the influence of the objects surrounding the position. For effective application of this method, it is expedient to execute, in advance, the search of positions, which are suitable for direction finding in the given district, and to work through the quickest route of movement from one position to another.

Method of Automatic Calculation of RES Coordinates During Movement At mobile station movement on the route, continuous RES direction finding is executed, and the whole bearing group from the session beginning is used for RES location calculation. During the operation session, as bearing aggregation with the big base is executed, the calculated RES locations are recurrently defined more exactly. The visualization of not only final, but the intermediate, results is often required in interactive environments, in order to make it possible for the operator to monitor the automatic algorithm operation and make decisions in complicated situations. So, the instantaneous bearings from all direction finders of the system may be represented on the map of the stationary station besides the calculation points of the RES locations. Sometimes, all bearings during the definite time interval are drawn, to estimate the bearing distribution. There is no sense in representing all the bearings obtained at the route, which, as a rule, has the form of a closed graph around the RES, since it is a flat small-informative picture. In the software package for cartography and navigation (SMO-KN), several algorithms for RES coordinate calculation are used (see section “Conclusion”): • Maximum-likelihood method • Matrix algorithm • Cluster algorithm. RES coordinate calculation method by means of the maximum-likelihood method is explained in [2]. The essence of this method is in the fact that, by means of non-linear equation solution, the maximum of the likelihood function is found out, considering the geometric location of the stations with the direction finders and RES, the distance difference between the posts and RES, and the RMS error of the bearings are obtained. This estimation method has the following advantages: • Taking into account the geometrical location of radio monitoring stations and RES • Taking into account the distance differences from radio monitoring station to RES • The possibility of calculation and taking into account the RMS errors of the measured bearings.

Introduction

399

Application of the maximum-likelihood method for coordinate calculation in the structure of the ARK-POM system has proven its high effectiveness. Its shortcoming is the fact that it works consistently at the coordinate calculation of a single RES only. Matrix method is based on the grid covering the operating zone, and this grid step is selected depending on the operating zone size and on the required accuracy. The grid is considered as the raster one, the bearings are drawn in it by the raster algorithm [3] and are summed on the amplitude in the grid cells. As a result, the three-dimensional surface is found out, formed by the bearing lines, the amplitude of which will be larger in those grid cells, where the bearings collide with each other more frequently and the analyzed signal has the larger significance. The bearing surface is represented on the map in the form of a figure with the amplitude distinguished by color, giving the operator the evident pattern of bearing distribution and their intersections. The example of the bearing surface representation is shown in Fig. 10.5. At simultaneous operation of several radio stations at the same frequency, we can observe, on the bearing surface, two or more maximums comparable to the amplitude, as is illustrated by Fig. 10.6. For automatic determination of the maximums, the algorithm of line-by-line filling is used [4]. As a result, the isolated regions are found out, the amplitude of which

Fig. 10.5 Example of the bearing surface representation

400

10 RES Localization Using Mobile Stations and Field Strength Measurement

Threshold

max1

a

max2

Fig. 10.6 The bearing surface section

Fig. 10.7 The bearing surface formed by two RES operating at the same frequency

exceeds the threshold value, and then the maximum coordinates are determined in the limits of each isolated region. In the issue, two or more RES can be revealed operating at the same frequency, as is shown in Fig. 10.7. The cluster algorithm is the further modification of the matrix algorithm. It also uses the matrix grid covering the operating zone on the map. But, in the grid cells, no bearing lines are aggregated, rather the interception points of the bearing lines are aggregated there. These points are solutions of the triangulation problem. As

Introduction

401

a result, we have a three-dimensional surface formed by the bearing interception points, and the amplitude will be larger in those grid cells, where the bearings intercept each other more frequently. The shortcoming of the cluster method is the large volume of computation; the advantage is the possibility to determine the location of an arbitrary number of RES operating at the same frequency. RES permanent direction finding in motion on a route, under urban conditions, leads to a large number of false bearings; therefore, the high speed of the bearing calculation is the important condition for accurate RES location determination. At fixed vehicle velocity, the high speed of direction finding provides more frequent bearing taking, which decreases the probability of missing “good” route parts, at which the bearings have values that are close to true. “Argument” station equipment is capable of calculating from 20 to 100 and more bearings per second for each of the monitored sources. This means that at vehicle motion, with an average velocity of 40 km/h, the bearings will be calculated in every 0.07–0.5 m of path. The method of RES coordinate calculation at mobile station movement has the following advantages: the possibility of system operation in automatic mode without operator participation, the high concealment of station operation, the continuity of bearing taking, which allows RES localization that operates very rarely. The main condition of a successful operation is the fact that the selected motion route should ensure the necessary base for each found RES. As the route variant, it is possible to suggest the closed curve around the zone of the RES’ possible location.

Peculiarities of Multi-Channel Direction Finding In real functioning it is often required to determine simultaneously the location of not one, but several, or all, RES operating in the given frequency range or ranges. For this purpose, the “Multi-channel direction finding” operation mode of the SMOPPK application is intended (see Section 8.12). As an example, Fig. 10.8 shows the SMO-PPK application window in this mode. Direction finding was executed in the frequency band allocated for the basic stations of the NMT-450 cellular communication standard. In the right lower corner of the figure there is a limb, at which the found sources are shown in the form of points. The point angular position corresponds to the bearing value (the angle is counted clockwise); the radial position corresponds to the source frequency. We can see clearly the azimuths at which the basic stations are situated. Thus, during multi-channel direction finding, the bearing array measured at various frequencies is sent to SMO-KN application input. Parameters and frequencies of the arrived bearings are saved in the bearing history file and are registered in the frequency table. In order to speed up the computation process of RES coordinates and to simplify the operator’s work, the bearings represented on the map can be filtered by frequency. At automatic RES search in SMO-PPK application, the calculated frequency, together with RES signal bandwidth are permanently made more exact. In SMO-KN

402

10 RES Localization Using Mobile Stations and Field Strength Measurement

Fig. 10.8 Multi-channel direction finding of the basic stations of the NMT-450 cellular system

application, at the selection of the bearings corresponding to the specific RES, the filtering is executed, taking into account its frequency bandwidth; at that, to increase the operation speed of the bearing identification procedure belonging to one RES, algorithms on the basis of “red-black trees” are used [4].

Simultaneous Direction Finding At determination of the permanently- or periodically-operating RES location, we can manage the single mobile direction finder, which (at motion) uses any of three methods of direction finding: the drive method, the quasi-stationary method or the method of automatic RES determination in motion. Direction finding of the shortterm operated RES is significantly more difficult. The operation sessions of such RES do not exceed several seconds, and they can often operate once. To determine such RES location, it is necessary to have the mode of simultaneous or synchronous direction finding, which is realized in the systems of direction finding and location determination of the ARK-POM system. For synchronous direction finding, the single-channel and multi-channel modes are used. In the single-channel mode of direction finding, all system stations are tuned to the same frequency. As soon as the signal amplitude at the tuning frequency exceeds the threshold, the bearing calculation is executed (see section “Conclusion”). Since

Electromagnetic Field Strength Measurement

403

the speed of the bearing taking is up to 100 and more bearings per second, simultaneous direction finding is ensured for those RES having short-term operation sessions. In the multi-channel direction-finding mode, all stations of a geographicallydistributed system operate in the same frequency ranges. At that, the direction finding results are transmitted from the peripheral posts to the central post by the requests from the central post. RES direction finding synchronism is also achieved at the expense of the speed of panoramic bearing taking, which is 300 MHz/s and more.

Electromagnetic Field Strength Measurement At planning and operation of the radio networks of mobile radio communication, radio, and TV broadcasting systems, at solving the problems of electromagnetic compatibility, at search of the technical channels of information leakage (Technical Surveillance Countermeasures – TSCM), and at fulfillment of the sanitary norms, the most important stages are the theoretical calculation of electromagnetic field strength generated by RES, and execution of experimental measurements of the field strength and the result processing. To measure the field strength is practically more suitable for the electric component of the electromagnetic field, therefore, in most cases, when speaking about the field strength, one usually assumes the electric component E. Thus, further in the text we understand the magnitude (module) of electric component strength E as the field strength. The measurement unit for field strength is Volt per meter (V/m). The field strength can be expressed in decibels, reduced to the level of 1 μV/m, or to another level, conditionally taken for the unit. One of the main purposes of field strength measurement is to determine the coverage area of the source signal receiving. At present, a large number of approaches have been developed, which allow the theoretical calculation of the coverage zone. However, the calculations based on these approaches have the estimation nature, since the models laying in the basis of these approaches cannot principally consider all specific peculiarities of radio wave propagation in the analyzed zone. Therefore, experimental field strength measurements remain the necessary stage at the planning and operation of the radio networks of mobile radio communication, radio and TV broadcasting systems. All modern technique for the calculation and determination of coverage zones is based on digital cartography. But a series of factors decrease the effectiveness of digital cartography at solving practical problems, including the construction of field strength distribution on-site: • Digital maps of SXP format with the scale 1:200,000 do not ensure high accuracy of the cartography object representation: the accuracy of object locations on plane is 90–140 m, on height is 16–40 m. The digital map accuracy corresponds to the accuracy of the primary printed cartographic materials (the nomenclature lists of

404

10 RES Localization Using Mobile Stations and Field Strength Measurement

the appropriate scale), taking into account the accuracy of instruments used for the automated sheet digitizing. • Digital cartographic information has now partially become out of date, as it was digitized on the basis of the sheets of the 1980s, therefore, some cartographic objects are absent on the map, or its configuration does not correspond to the reality; in particular, for strength determination problems, the features of urban building up, the forest-park zones, and the district topography are important. • The list and the features digital maps do not completely satisfy the need to have the necessary information for the existing model using the radio wave propagation in conformity with the real routes (the information on the building up, the forests, the underlying surface, is not presented fully enough). The theoretical model choice for the radio wave propagation at calculation of the electromagnetic field strength is also limited by the completeness and accuracy of district digital maps. Therefore, the on-site calculation and construction of the field strength should be provided in several stages – in which strength measurement execution is one of the obligatory stages – with the help of stationary or mobile radio monitoring stations together with the strength theoretical calculation, for generation of the full field distribution. Using the mobile station essentially increases the accuracy of the distribution construction, since it ensures the measurements in various district points.

Main Mathematical Relations All methods of field strength measurements are based on the following relation [1]: E = U/he

(10.1)

where E is the field strength in V/m; U is the electromotive force induced in antenna in V; he is the effective height of the measuring antenna in m. For frequencies higher than 1 GHz, the power flow density is used instead of the field strength. This value is expressed in Watts per square meter: P = E2 / (120π )

(10.2)

From (10.1), it follows that the antenna and the voltmeter should be necessary elements of any measuring system for the radio wave field strength. It is accepted to consider the open-circuit potential of the non-loading antenna in this field strength ratio as the effective height or length of the receiving antenna. This means that the load resistance is equal to infinity (in this case the antenna output voltage is equal to the electromotive force induced in antenna by this field). If this condition is not satisfied (usually exactly this situation takes place because,

Electromagnetic Field Strength Measurement

405

to achieve the matching, the equality of antenna input resistance and the load resistance is required), the voltage on antenna output can be calculated by the formula: V0 = URload / (Rant + Rload ) .

(10.3)

where Rload is the load resistance; Rant is the antenna output resistance. Thus, we should use the following variable instead of the known antenna effective height: he calc = he / (1 + Rant /Rload ) .

(10.4)

To calculate the field strength, then we should use the formula: E = U0 /hact calc

(10.5)

Effectiveness or the calibration factor Keff of the receiving antenna is the electric field strength E of the plane wave divided by the voltage at the antenna output U0 at its nominal load resistance, which is usually equal to 50 : Keff = E/U0 .

(10.6)

From comparing (10.5) and (10.6), we can see that Keff = 1/hact calc .

(10.7)

The antenna gain coefficient G with respect to the isotropic antenna is often specified instead of the antenna effectiveness. The isotropic antenna gain coefficient G versus the antenna effectiveness Keff is defined from the formula: 1 keff

1 = √ λ G

f [MHz] 4πZ0 9.73 = √ = √ Rload λ G 30.81 G

(10.8)

where Z0 = 377 , Rload = 50 . Since the voltage and the field strength values are measured in dBμV and in dB( μV/m), the antenna effectiveness values are reduced to the logarithmic form. If keff = 20 lg Keff and g = 10 lg G, the antenna effectiveness is specified in dB/m as keff = −29.77 [dB] − g + 20 lg (f [MHz])

(10.9)

and, hence, the field strength level E can be determined from the antenna output voltage level U0 as E dBμV/m = v0 dBμV/m + keff dB/m .

(10.10)

406

10 RES Localization Using Mobile Stations and Field Strength Measurement

Sometimes, the calibration coefficient keff does not include the attenuation as of the feeder cable between the antenna and the measuring receiver. In this case, the above equation should be expanded (in this case U0 is the voltage on the measuring receiver input) to E dB (μV/m) = v0 dB (μV/m) + keff dB/m + as [dB] .

(10.11)

For instance, at 100 MHz frequency the antenna with the gain of 6.5 dB has the effectiveness of 3.7 dB. At input voltage U0 = 33.4 dBμV and for the cable attenuation as = 1.1 dB, the field strength is 38.2 dB(μV/m). In case of automated systems, the calculations are executed automatically; the tables of the calibration coefficients of the measuring antennas versus frequency are saved in the form of the files and are used in the SMO-PPK, SMO-PA, and SMOPAI software.

Peculiarities of the Field Strength Distribution Estimation In the structure of a radio signal received by the antenna of a moving object, there is usually a fast fading component caused by multipath propagation, and a slowly varying component, depending on the route micro-structure. Maximums and minimums of the interleaving frequency of the received signal will be proportional to the velocity of the moving object, and their depth is mainly defined by the district’s character and the object’s peculiarities. In the simplest case of the singlereflection wave, the standing-wave maximums will follow one after the other, in λ/2 intervals, where λ is the radio signal wavelength. For example, at 1,000 MHz frequency the space interval of the maximum consecution is 0.15 m. In accordance with the recommendations from [1], for the detailed estimation of the field features, the measurements during vehicle motion should be executed at lengths less than the wavelength. In order to estimate the second (slowly varying) component of the field strength, it is necessary to select such spatial observation interval D that the fast fading will not influence on the local average estimation, but, on the other hand, the attenuation caused by the distance should not be noticeable. Evidently, as far as the observation interval D decreases, the obtained average value will increasingly more differ from the value obtained at the averaging over the infinite length interval. Calculations and measurements show that, at choosing of the interval D = 40λ, the average estimation error will not exceed 1 dB. On the other hand, at a distance of 1 km to the source, and for the additional removal by 300λ (λ << 1km), the average estimation error due to the attenuation will be quite insignificant. Therefore, it is recommended to choose the observation interval value D for the field’s slow fading estimation, from the relation 40λ ≤ D ≤ 300λ. At a signal frequency of 1,000 MHz, the averaging interval should be in the limits of 12–90 m.

Electromagnetic Field Strength Measurement

407

Field Strength Measurement In the “Archa” stationary station and in the “Argument” mobile station, field strength measurements are executed with the help of the ARK-D1TR or the ARGAMAK panoramic measuring receivers, which are included in the second post structure. The receiver control is executed with the help of the SMS-PAI software loaded in the second post’s PC. If necessary, control of the second post’s equipment can be provided via the network by the SMO-PPK software operating on the first post’s PC. The field strength calculation is carried out in the spectral domain or by the time sample. It is possible to execute estimate measurements of field strength with the help of the first post’s equipment on the basis of the ARK-CT2 double-channel panoramic receiver. Power spectra of the signal or the time sample values are corrected, taking into account the calibration curve of the measuring antenna, and the losses in the coupling cable and in the antenna switch. The reference points of the calibration curves of the measuring antenna are saved in special files. Before executing the measurement, the values are read off from this file, which corresponds to the used antenna at the moment, while the missing data of the calibration curve are compensated for by means of spline-interpolation. The ARK-D1TR and ARGAMAK-I panoramic receivers support the singlechannel mode of strength measurement. At that, the signal level measurement error, which is sent to the receiver input along with the additional section calibration, does not exceed 1.5–2 dB. In SMO-PAI software, field strength calculation in the given bandwidth can be executed both on the spectrum and on the time sample, with the help of the digital peak, quasi-peak and RMS detectors (see section “Conclusion”). Figure 10.9 shows the SMO-PAI application window in the mode of field strength measurement. This measurement is provided for the spectrum section selected by the operator. Thus, Fig. 10.9 shows the case of field strength measurement for the radio station at the frequency of 72.1 MHz. The obtained value is 45 dB μV/m. In the single-channel measurement mode, the field strength distribution histogram, plotted on the basis of the calculated value samples, is shown in the plot in the left part of the window. During complex motion, the operator can detect the presence of interference by the histogram form. At application of the “Argument” mobile station, we can use several methods of measurements: • Field strength measurement when moving • Field strength measurement at parking • The combined approach. The combined approach assumes measurement execution at both parking and in motion. The certified measuring antennas mounted on the turned dielectric telescopic mast, attached to the vehicle body, are used for operation.

408

10 RES Localization Using Mobile Stations and Field Strength Measurement

Fig. 10.9 Single-channel measurement of field strength

On-Site Calculation of Field Strength Distribution The field strength in the receiving point, which is created by RES, is defined as the free space field multiplied by the attenuation over the route of radio wave propagation. The signal field strength E in V/m in the receiving point is defined by the formula: E=

30Ps Gs L/d0

(10.12)

where Ps is the signal power of the radio station transmitter in W; Gs is the antenna gain coefficient of the transmitting radio station; d0 is the route length in m; L is the additional multiplier for the attenuation (losses) over the radio communication route. The formula expressed in decibels, with respect to 1 μV/m, is the following: E dBμV/m = 134.77 + 10 lg (Ps Gs ) − 20 lg d0 + L [dB] .

(10.13)

Calculation of the additional multiplier for the radio communication route attenuation is a complicated problem. We can nominate the following main factors, as essential influences on radio wave propagation:

Electromagnetic Field Strength Measurement

409

• The district topography • Urban build-up • The vegetation. These factors can be taken into account by three ways: deterministic, statistical and combined. The first refers to the methods concerning mainly the geometrical approximations of diffraction theory. They allow the calculation provision for field strength, but raise high demands for authenticity of the mathematical model of the radio wave propagation environment. Statistical methods consider the random character of the irregularity distribution influencing on the process of radio wave propagation. They allow the prediction of some average values of the signal attenuation. The combined methods are the combination of the first two methods. The initial information contained in the digital district maps, as a rule, is insufficient to execute the calculations by the deterministic approach. Therefore, it is expedient to use the combined method of calculation. Information regarding district topography, urban build-up, and forest areas should be combined with the statistical estimations of propagation effects related to the underlying surface, the field distribution inside the residential areas, and forest areas. Additionally, one should carry out the calculation correction on the basis of the natural measurement results in the separate district points.

District Topography Consideration of district topography is an important point for route calculations under conditions of cross-country radio monitoring. The approximated formulas given in ITU-R reports are the most frequently used to estimate district topography [5]. They are based on the statistical processing of real measurements for definite relief types, but do not consider the features of the specific propagation routes. At the same time, the presence of information regarding district topography in digital maps and the performance of modern computers allow the solution of complex enough problems on the basis of calculation of the separate radio communication routes. On the other hand, the numerous electrodynamic method applications having high accuracy are often limited by the small information value and the reliability of the digital maps. That is why, recently, the most widely distributed approach combines consideration of the topographical features of the specific routes with the results of statistical processing of real measurements. So, to calculate diffraction losses – the important component of the attenuation multiplier along the surface radio communication routes – the following method is used. Typically, along routes that pass over medium cross-country and highland areas, the diffraction angles do not usually exceed 5◦ , and at that, the curvature radius of each obstacle is much less than the Earth’s radius. The obstacles are usually

410

10 RES Localization Using Mobile Stations and Field Strength Measurement

defined as a sphere. Under these conditions, the diffraction losses LD (in dB) with respect to the free space, can be calculated (for a single obstacle along the route) by the formula [6]: / 2θ 2 θ √ LD = −6.4 − 20 lg x + 1 + 1.41 x − (10.14) 18.3θ for θ > 0 0.75 1.5 −KH 6.6x y − 11.7x0.25 y1.5 θ for θ < 0 where 1/3 3.10−4 d0 f 1/3 1/3 x= , y = 14.9R f , KH = exp −0,5 fda db R

(10.15)

d0 is the route length in km; da and db are the distances from the end points to the interception of the obstacle tangents in km; f is the frequency in GHz; θ is the diffraction angle in radians; R is the curvature radius of the obstacle; KH is the additional multiplier considering the radio wave dispersion on the vertex’s rough edges. Even for routes with a single obstacle, the results of the calculations and the measurements are often dispersed. As a rule, the calculated attenuation is more than the measured one. The dispersion increases with the frequency growth, achieving large values, in centimeter range. Probably, this relates to the small irregularities of the obstacle peaks and its non-spherical form. Especially, this difference takes place for the highlands, where the peaks are more often wedge-shaped rather than rounded. Comparison of the calculation results and the numerous measurements led to the introduction of the additional multiplier KH in (10.14). In [6], the semi-empirical model is presented for the calculation of the diffraction radio wave attenuation on the arbitrary number of the real obstacles along the route. The attenuation inserted by a i -th obstacle at the positive diffraction angle (θ > 0) is calculated by (10.14). The total diffraction attenuation along the route with several obstacles is formed by the attenuation on each obstacle. Of course, one should consider the mutual influence of the adjacent obstacles. However, the acceptable estimations of such an influence degree are absent for the real obstacles. From the general considerations, it is clear that this influence increases with the decreasing distance between the obstacles and it decreases with the diffraction angle growth. In [7], the following semiempirical formula for the calculation of the attenuation at diffraction on the several obstacles is introduced, which considers their mutual influence: LD = LD1 + LD2 + LD3 + ...−

0.37 ( (LD1 ,LD2 )0 (d1 d2 ) − −0.64 min (LD1 ,LD2 ) arctan 0.72 min 0.74 min (LD1 ,LD2 ) d1,2 (10.16) 1.5 0.37 ( (d1 d2 ) min (LD2 ,LD3 )0 −0.64 min (LD2 ,LD3 ) arctan 0.72 min (LD2 ,LD3 ) − ... 0.74 d2,3

Electromagnetic Field Strength Measurement

411

Urban Build-Up The urban environment creates specific conditions for radio waves propagation. Dead spots, multiple reflections, and wave dispersions form fields with complicated interference structures and drastic spatial variations of signal level. Under urban conditions, we may single out the following main elements that influence radio wave propagation: • Guide structures (avenues, streets, river sections, contact lines of urban electric transport, etc.) • Separate building or building group • Earth’s surface and the obstacles on it (vehicles, columns, fences, etc.) • Vegetation areas (parks, public gardens, conservation land, etc.) • District topography. For urban conditions, at present there is a whole series of mathematical models enabling the calculation of the average value of the received power, depending on the various parameters. Most of them are almost completely empirical. Historically, empirical plots were first obtained by J. Okamura [8], to enable the determination of the signal median value under conditions of the statistically uniform city, and also to consider, to one degree or another, these or those urban peculiarities or the separate urban regions. With the help of the plots obtained by Okamura, various experts derived analytical expressions for the field calculation. The model offered by K. Olsbrook and J. Parsons allows loss predictions at the propagation [9]. / L = Lfree +

Lpr − Lfree

2

2 +L +γ + LD B

(10.17)

where lfree are the losses in free space, in dB, which can be calculated as: Lfree = 32.45 + 20 lg f + 20 lg d0

(10.18)

f is the operating frequency, in MHz; d0 is the distance between the transmitting and receiving antennas, in km; Lpr is the propagation loss over the plane ground, in dB, which can be calculated as follows: Lpr = 120 − 20 lg hm − 20 lg hb + 40 lg d0

(10.19)

where hm ,hb are the heights of the receiving and the transmitting antennas, in m, relatively; LD is the diffraction loss, in dB, caused by the character of the district topography; LB is the loss caused by the presence of urban build-up, in dB, calculated as h0 − hm + 16; (10.20) LB = 20 lg √ dλ

412

10 RES Localization Using Mobile Stations and Field Strength Measurement

λ is the wavelength, in m; d is the effective street width, in m, at which the receiving antenna is situated; h0 is the average building height near the receiving antenna, in m; hm is the height of the receiving antenna, in m; γ is correcting factor depending on frequency, namely, γ = 0 at f< 200 MHz, and, for f > 200 MHz, the value of γ is defined by the special plot [9]. The averaged empirical formulas for calculation of the propagation losses for the town, in decibels, allow for attenuation determination with accuracy up to 7–17 dB [10]. Figure 10.10 shows examples of the radio wave propagation routes, for which it is possible to make the attenuation calculations more exact, by using deterministic methods on the basis of digital map application.

Vegetation Influence The following factors have the most considerable influence on the radio wave propagation conditions in forest-park zones: types of trees; their location and density; the height and the form of the crown. Of course, the density of the leaves in the tree crown depends on the season of the year, and the electric features of wood essentially depend on the weather conditions. In fact, to account for all defining conditions of radio wave propagation in forest-park zones is impossible without the statistical method application. At receiving antenna shading by the forest tract, the additional attenuation inserted by the forest, can be determined as Lfor = αfor Rfor ,

(10.21)

where αfor is the linear signal attenuation in the forest, in dB/m; Rfor is the path length in the forest, in m (see Fig. 10.11). The attenuation inserted by the forest increases up to the definite value only, and then remains approximately at the same level, due to the receiving of the wave, which rounded the forest. In accordance with the rough estimate, the attenuation in decimeter range may achieve near 30 dB. The described calculation method, strictly speaking, can be applied to the case, when both antennas (of the transmitter and the receiver) are located inside the forest. Along the route presented in Fig. 10.12, the additional losses occur at the wave “entering” into the forest tract. There are a number of models for the linear attenuation calculation and for the experimental data obtained for the various conditions. An example of the plots of the linear attenuation versus the signal frequency and polarization for the routes in the mixed forest is presented in Fig. 10.12. For the field with the vertical polarization, the attenuation is slightly higher than for the horizontal polarization, which can be explained by the influence of the vertical tree trunk.

Electromagnetic Field Strength Measurement

413

a

b

c

Fig. 10.10 Examples of the main routes of radio wave propagation in the city: (a) the open propagation route from the beginning of the residential area build- up, and the calculation of the residential area inside attenuation (at the receiving antenna height below the average house height); (b) the closed route with radio wave diffraction on the building roof edges (buildings are changed by the wedge-shaped semi-transparent obstacles); (c) the route with the open and closed parts, when the signal is the sum of diffraction and reflection components

h1 forest h2

Fig. 10.11 Calculation of the additional attenuation due to the forest

Rf

414

10 RES Localization Using Mobile Stations and Field Strength Measurement

Fig. 10.12 The linear attenuation in the forest versus the frequency and polarization: (1) horizontal polarization; (2) vertical polarization

a, dB/m 0.25

0.15 2

1

0.05 0 50

200

800

f, MHz

Calculation of Field Strength in the SMO-KN Application When calculating signal field strength, in the SMO-KN application, the influences of the following factors should be considered – both separately and in any combination: • • • • • •

District topography Residential area build-up Re-reflections from the residential area build-up Vegetation sections Transmitter antenna pattern Transmitter antenna height.

In SMO-KN, intensities of color are used to represent the field strength distribution of the map background, and, for that purpose, the geodesic color palette is used. With the help of the threshold displacement, we can represent, on the map, the zones, for which the field strength value exceeds the given value. Figure 10.13 shows the results of the theoretical calculations of the coverage zone of sound accompanying the 37th TV channel (vertical polarization), taking into account the antenna pattern; at that, the reflection threshold 62 dBμV/m has been chosen for the coverage area visualization. These results have good coincidence with the results of field strength measurements carried out by the services of the Radio Frequency Center, in accordance with the acting guidelines. In Fig. 10.14, theoretical calculation results for the signal field strength are shown, taking into account the district topography and the underlying surface only, but without consideration of the building up and vegetation influence, while Fig. 10.15 shows the calculation results that take into account all of the factors. As we see, the results of these two calculations differ essentially. Figure 10.16 shows the theoretical results calculation of signal field strength without the account of and with the re-reflections from the urban residential area

Electromagnetic Field Strength Measurement

415

Fig. 10.13 Field strength distribution on the basis of the theoretical calculation results (restricted by the coverage area)

building up. At re-reflection calculation only, those residential areas are considered, which are located along the propagation route, as giving the most contribution. The side re-reflections (outside of the route) are not considered because of large algorithmic and computing difficulties. Nevertheless, the result comparison for the cases of reflection consideration and non-consideration proves the importance of such consideration, since on some sections the signal level varies by 2–5 dB due to re-reflections.

Processing of Field Strength Measurements All arrived data about the field strength and the navigation are saved in the history database in the SMO-KN application. The following possibilities exist for the analysis of aggregated data:

416

10 RES Localization Using Mobile Stations and Field Strength Measurement

Fig. 10.14 Calculation results of field strength, taking into account district topography (without building up and vegetation)

• • • • • • • • •

Value selection for calculation, for the given frequency (source) Data selection for calculation of the field strength distribution in the selected area Data selection for calculation, for the arbitrary time interval Rolling of the movement route of the mobile station with graphical and textual data representation Full imitation of the operation session, when the aggregated data is received from the history file Calculation of the field strength distribution on-site, with account taken of the selected affecting factors Calculation and representation of the RES coverage area, with account taken of strength measurements Determination of RES location and their supposed features Formation of reports with the measurement results.

Electromagnetic Field Strength Measurement

417

Fig. 10.15 Calculation results of field strength, taking into account district topography, building up and vegetation

To construct the field distribution diagram on-site on the basis of measurement results, the Shepard method can be used, known also as the method of the inverse weighting distance. In this method, the value for each matrix point can be found, at interpolation, as the linear combination of values in the basic points and the weight of each basing point: E (x,y) =

n

wi (x,y) Ei ,

(10.22)

i=1

where n is the number of the basic points; E (x,y) is the desired strength value in the point with coordinates (x,y); Ei is the strength value in the basing points, and wi (x,y) is the weight of the basing point defined by the formula: ⎛ ⎞−1 n 1 ⎠ 1 , wi (x,y) = p ⎝ p di d j j=1

(10.23)

418

10 RES Localization Using Mobile Stations and Field Strength Measurement

a)

b)

Fig. 10.16 Example of the signal field strength calculation without re-reflection consideration and with it, for the re-reflections from the urban residential area building up

where p is the degree parameter characterizing the weight decreasing of the basing point with distance growth, and di (x,y) is the distance between the i- th basic point and the point (x,y) being interpolated. A modification of the method of the inverse weighting distance is used in SMOKN application, in which the basic point weight is defined by the formula: wi (x,y) =

R−di Rdi

2

n 0 R−dj 2 j=1

(10.24)

Rdj

where R is the distance to the most far basic point, i.e., R = max d. 1
The “Argument” mobile radio monitoring station, which is capable of providing field strength measurements when moving, allows the problem of recovering so-called “electromagnetic district relief” to be solved, for the given band of the frequency range, taking into account all registered transmitters, industrial interference, unapproved sources. At that, the number of measurements, their density and regularity determine calculation accuracy.

Electromagnetic Field Strength Measurement

419

In some cases (e.g., when assuming the presence of a single dominating source in some local region), one may essentially simplify the problem on the basis of theoretical method application.

Determination of RES Location Usage of the mobile measuring station allows the solution of RES location determination problems and its supposed features on the basis of field strength measurements without direction finding. The method of RES location determination is based on RMS deviation minimization of the calculated field strength values at the measurement results, in case of source location, in the supposed matrix point: (x,y) = arg min

(xm ,ym )

n

2 Ei − En xm, ym ,

(10.25)

i−1

where (x,y) is the unknown value of RES coordinates; (xm ,ym ) are the coordinates of the source search point; n is the number of measurements (basic points); Ei is the field strength in the basic points; Eti (xm ,ym ) is the theoretical (calculated) strength value in the basic points, obtained with the help of the used model of radio wave propagation at the supposed source location in the point (xm ,ym ). RES location determination can be provided both after the measurement session and directly during the measurement, when the station is moving along the route. We recommend the following, for RES location determination: • To move several kilometers in one direction, to form the initial data on the field strength gradients, and then to move the same distance in a perpendicular direction • To use the theoretical model of radio wave propagation at the initial stage, without taking into account the district relief and other parameters (in this case the calculations are executed fast, in real time, for any measurement quantity) • Having oriented on the district of RES possible location, to choose the route encircling this district, trying to drive into it from the other side of the RES, simultaneously decreasing the distance to it (the route exhibits the form of a snail) • After RES localization with an accuracy of 1–2 km, to decrease the measurement area down to 3–4 km around the source, leaving 20–30 measurements inside of this area, and to use now the radio wave propagation model, with account taken of the district topography, building up, and vegetation. The above-mentioned procedure of a RES search while moving is illustrated by the frame sequence presented in Fig. 10.17. In this example, the accuracy of RES location determination using the radio wave propagation model without taking into account the district topography, building up and vegetation is less than 100 m.

420

10 RES Localization Using Mobile Stations and Field Strength Measurement

a)

b)

c)

d)

e)

f)

Fig. 10.17 Determination of RES location by the field level

Electromagnetic Field Strength Measurement

421

Checking Transmitters for Announced Parameters With the help of the SMO-KN application, one can solve the problems of more precise definition of announced (licensed) RES parameters (e.g., the transmitter power, the antenna pattern parameters in vertical and horizontal planes, antenna height and its coordinates) on the basis of the field strength measurement results. At that, we can use the measurements executed at various times under various weather conditions, by the various measurement equipment, differing by the antenna height, because the algorithm of reducing the results to the unified condition is provided. For more precise definition of various parameters, the various volumes of experimental data are required. For instance, for precise determination of the transmitter power, it is enough to execute some tens measurements of electromagnetic field

Fig. 10.18 Theoretical field strength distribution of the source signal

422

10 RES Localization Using Mobile Stations and Field Strength Measurement

level (approximately 20–30), while for precise determination of antenna pattern parameters it is necessary to increase the measurement number by two or even three orders. In the last case, the measurements should be performed along the circular routes of measuring complex movement around RES and at various distances from it. Figure 10.18 shows the theoretical field strength distribution of the transmitter signal, in accordance with the announced power equaled to 30 W, antenna height and its pattern. The problem is solved on the basis of automatic discrete power value-searching from the range defined by operator, with the purpose of searching such power value, which ensures the maximal coincidence of the measurement results and the theoretical calculations. Figure 10.19 shows the results of more precise determinations of the signal field strength by the measurement results, at that, the more precise transmitter power is 6 W only.

Fig. 10.19 More precise signal field strength distribution on the basis of measurement results

Electromagnetic Field Strength Measurement

423

To check the calculation correctness, let us return to the problem of signal field strength recovering on the basis of measurement results for the case, when we can assume the presence of one dominating RES in some local area. We shall solve this problem in such way: on the basis of the above-mentioned theoretical methods and algorithms, using the field strength measurements we determine RES location and its predicted parameters, and then knowing the source location and its power, we execute the theoretical calculation of the field strength. The obtained field distribution practically coincides with the theoretical one shown in Fig. 10.19, and the field, which was made more precise on the basis of the measurement results shown in Fig. 10.16.

Calculation of Electromagnetic Compatibility The most important problem of the measurements and calculations is the calculation of RES electromagnetic compatibility indices, namely, the mutual interference level. Figure 10.20 shows the calculation results of the conflict signal receiving from

Fig. 10.20 Calculation of electromagnetic compatibility

424

10 RES Localization Using Mobile Stations and Field Strength Measurement

two transmitters (their location is shown by the arrows), calculated on the basis of the protective relations – extreme ratios of useful and undesired signal strength for provision of the stable receiving conditions. The coverage areas from the Radio station 1 and Radio station 2 are shown by the dark color, the conflict receiving zones – by the light color.

Conclusion In this chapter, the problems of determining radio emission sources locations on the basis of the direction-finding results and the electromagnetic field strength measurement results, and also the measurement result processing are considered. Several methods of single radio wave source localization or source group localization are described: the methods of drive, quasi-stationary and RES location determination in motion. At that, the single-channel and multi-channel procedures for direction finding and location determination can be used. The source locations are represented on the district electronic map in the SMO-KN application. “Argument” mobile stations, included in the system structure, are capable of measuring the field strength both in express-analysis mode when moving, and at parking. The measurements are executed with the help of the measuring antenna. This makes it possible to construct the full pattern of field strength spatial distribution with the further calculations of RES coordinates, checking the transmitter powers, the measurement of factual pattern of their antennas, the determination of coverage areas of broadcasting and communication systems, the estimation of electromagnetic compatibility of the radio electronic systems. At field strength determination, the influences of district topography, urban buildup, vegetation, antenna pattern, and the transmitter antenna height are considered. In SMO-KN application, all arrived data on the field strength and on navigation are saved in the history database. The analysis of aggregated data can be provided on the separate frequency or on the frequency group, for the spatial area and the time response.

References 1. Spectrum Monitoring Handbook, ITU-R, Geneva, 2002. 2. Glaznev, A.A., Kozmin, V.A., Litvinov, G.V., and Shadrin, I.A., Multi-station Radio Monitoring Systems for Determination of Radio Emission Localization (in Russian). Special technologies. 2002. Special Edition, pp. 20–29. 3. Rogers, D.F., and Earnshaw, R.A., Computer Graphics Techniques: Theory and Practice. Springer, 2001, 542 pp. 4. Cormen, T.H., Leiserson, C.E., Rivest, R.L., and Stein, C., Introduction to Algorithms, 2nd Edition, The MIT Press, 2001, 1184 pp. 5. CCIR. Report 715-2, Propagation by Diffraction. – Recommendation and reports, XVII Plenary assembly, Dusseldorf, 1990. 6. Larin, E.A., Calculation of Radio Wave Diffraction Attenuation on the Surface Traces over the Cross-country and Highlands (in Russian). Elektrosviaz, No. 1, 1997, pp. 17–20.

References

425

7. Hata, M., Empirical Formula for Propagation Loss in Land Mobile Radio Service, IEEE Trans. Veh. Technol., Vol. VT-29, No. 3, 1980, pp. 317–325. 8. Okamura, J. et al., Field Strength and Its Variability in VHF and UHF Land Mobile Radio Services, Rev. Inst. Elec. Eng., Vol. 16, No. 9–10, 1968, pp. 825–873. 9. Delise, G.Y., Propagation Loss Prediction: A Comparative Study with Application to the Mobile Radio Channel, IEEE Trans. Veh. Technol., Vol. VT-34, No. 2, 1985, pp. 86–95. 10. Kvartirkin, E.M. (ed.), Ultra-High-Waves Propagation in Towns (in Russian). Collection of papers. VINITY. Science and Technology Resume. Series Radio Tekhnika, Vol. 42, 1992, pp. 196.

“This page left intentionally blank.”

Chapter 11

Detection and Localization of Technical Channels of Information Leakage

Introduction A technical channel of information leakage (TCIL) is a physical channel (acoustic, electromagnetic, electrical), which allows access to be obtained to an information object with the help of technical means [1, 2]. Protection against information leakage via technical channels (usually called Technical Surveillance Countermeasures) is a complicated multi-pronged problem requiring the fulfillment of the organizational and engineering action complex. An important stage of similar actions is the search of electromagnetic channels of information leakage in radio wave range, which propagate outside the monitored area. This chapter focuses on the methods of radio channel search for acoustic information leakage, which are the most distributed. Electronic devices transmitting information via radio channels will be further referred to as radio microphones. Radio microphones can be implemented in the form of a separate device with small sizes, or may be built-in into the objects of everyday life: phone, cigarette lighter, wristwatch, tie pin, etc. It is technically possible to implement radio microphone operation in practically any radio wave range, however, the meter and decimeter ranges are used more frequently. Various types of modulation are used in radio microphones, ranging from the simplest amplitude or angular modulations to the complicated digital modulation types. To increase operation concealment, in expensive radio microphones, the stages of information aggregation and its transmission are separated. In similar devices, audio signal compression and recording into the internal memory is executed during a definite time interval and then the data transmission is fulfilled during a rather short time period. The simplest detectors of the radio microphone emission are the electromagnetic field indicators, which inform by audio or light signals about the presence of the electromagnetic field with the strength higher than a threshold in its receiving antenna location point. The more complicated radio microphone detectors are so-called interceptors, which automatically tune to the frequency of the most powerful signal and execute its detection. The sensitivity of such detectors is rather small, and therefore, they allow radio microphone detection at small distance only. Radio receivers with automatic tuning and spectrum analyzers have high sensitivity. Besides emission detection, these devices sometimes allow the determination of A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_11,

427

428

11

Detection and Localization of Technical Channels of Information Leakage

the modulation type and its parameters. For radio microphone searches, including the switched-off searches, non-linear radar can be used, the operation principle of which consists in the powerful signal radiation and the receiving of its second or third harmonic, caused by the non-linearity of the radio microphone’s electronic components. To increase the reliability of radio microphone emissions detection and location determination, automated hardware-software radio monitoring systems are used. In the simplest case, such systems are implemented on the basis of modified scanning receivers or spectrum analyzers and portable computers. In more complicated systems, digital panoramic radio receivers with wide bandwidths of instantaneous survey are used, which ensure high operation speed and make possible the detection of radio microphones of intermediate aggregation or with the complex modulation type. The search of electromagnetic emissions with the help of hardware-software systems is usually executed inside a protected object. But since the electromagnetic oscillations from the sources, which cause it, pass outside the object, its detection is possible from the outside also. For this purpose, the application of mobile radio monitoring stations is possible, and is especially effective when using the special software developed for radio microphone detection problems outside of buildings and structures. The problem of fast detection of unauthorized radio microphones inside secure locations, at minimal operator participation, is of current importance. The solution of such a problem becomes essentially complicated due to the fact that the detection should be provided in a very wide frequency range and under complex radio conditions, which can be characterized by a number of various interferences, including, in this case, for example, the signals of radio and TV broadcasting stations. In addition, due to known reasons, a priori data about the emission parameters of the unknown radio microphone are completely absent.

Main Search Stages for Electromagnetic Channels of Information Leakage TCIL detection is a complicated process, which in simplified form can be represented by the several stages [3] shown in Fig. 11.1. The current monitoring of the range loading (the first stage) supposes the aggregation and analysis of the data on frequencies, levels, and character of radio signals in the operating range, in conformity with the monitored premises (MP). The ensemble of the aggregated spectral data for a large enough time interval, obtained at the providing of current check, is referred to as “standard” panorama. At that, we assume that the “dangerous” signals are absent. The purpose of the second stage is the detection of “new” emissions generated inside the MP, and therefore, potentially dangerous from the point of view of information leakage. For effective realization of this stage, one should use the

Main Search Stages for Electromagnetic Channels of Information Leakage STAGE

RESULT OF STAGE EXECUTION

Current check of range loading, obtaining the standard panoramas

Standard panoramas

New emission detection: - standard panorama using - reference antenna using - reference signal using

List of “new” emissions

Estimation of radio signal danger. Identification of detected radio signals, execution of special investigation on CEE: - audition check; - acoustic testing; - introduction of special testing signals; - technical analysis of digital signals, parameter measurement, determination of modulation type

Position localization of the detected sources

Counteraction to taking down (leakage) of information

429

List of identified radio emissions

Technical parameter ensemble for digital radio signals List of frequencies of the detected potential technical channels of the information leakage

Coordinates of RES position inside the monitored premise

Generation of spot jamming on the frequencies of detected RES Reduction of spurious emission levels of checked technical equipment

Fig. 11.1 Process of detection of technical channels for information leakage

double-channel or multi-channel digital panoramic radio receivers of the 3rd–5th generation, which ensure the high speed of the panoramic spectral analysis. The ARK-D11, ARK-D7KM, and (for many premises) the ARK-D13 units can be classified, for example, as systems of such types. The essence of the multi-channel search is the following: to one receiver input, the “reference” antenna is connected, which is located outside the receiving zone of signals from the MP, but which ensures the reliable receiving of all external signals. Other inputs are connected to the receiving antennas situated in the MP. The comparison of radio signals received from the “reference” channel and other “signal” channels allows the radio emission reveal generated inside of the MP. These “new” signals represent the objects of more detailed investigation and testing, which are executed at the third stage. The purpose of the third stage is the danger checking of the found radio emissions, from the point of view of information leakage. Those emissions, the testing of which shows that they contain information regarding the processes in the MP, we shall call identified. It should be noted that complete automation of the

430

11

Detection and Localization of Technical Channels of Information Leakage

determination process of the found emission danger degree is hardly possible in the near future due to its great complexity. Nevertheless, there are a series of sufficiently-effective automated testing methods, which permit accelerated decision making in the most simple-enough situations. Similar approaches based on the emission of specific acoustic signals in the MP will be discussed in this chapter, and methods of special CEE investigation execution will be described in Chapter 12. Unfortunately, the above-mentioned methods do not guarantee the unambiguous identification of all dangerous emissions, which explains the presence in Fig. 11.1 of several “expert” testing methods oriented toward operator participation, since operator experience may help to make the correct decision in very complicated situations. RES position localization (the fourth stage) is possible with operator participation only. This stage can be executed in automated mode for radio microphone localization without the covering, and using handheld direction-finding equipment. Finally, the fifth stage related to the counteraction of information leakage can be implemented by various approaches, depending on the results of previous four stages. For example, if, during some meeting, the fact of radio microphone usage (activation) was detected with the help of the searching equipment – and its used frequency was determined – then, for the period of this meeting, spot jamming could be generated on this frequency, with the parameters ensuring the radio microphone suppression. The technical aspects of this method are considered below. If the found emission is the unpremeditated result of any equipment operation used in the MP, then actions to reduce these spurious emissions are taken: one can use the shielding, and the noisiness of the premises, in which this equipment operation takes place. The methods and equipment for the creation of noisiness are not considered in this book.

Detection of Radio Signals Emitted in Monitored Premise Radio Signal Intensity in Near-Field and Far-Field Regions As is well known from electrodynamics, electromagnetic field strength in near-field and far-field regions varies with distance, in different ways. The components of the electric field strength vector p emitted by the electric dipole are defined in the spherical coordinates as: Er =

1 2π ε

Eθ =

1 4π ε

1 r3

−

jk r2

1 r3

−

jk r2

cos θ |p| exp (−jωt) ;

−

k2 r

sin θ |p| exp (−jωt) ;

(11.1)

Eϕ = 0. where r, θ and ϕ are the spherical coordinates; Er , Eθ and Eϕ are the electric field strength components in the spherical coordinates; ε is the electric permeability of the free space; ω is the emission radian frequency, and k is the wave vector.

Detection of Radio Signals Emitted in Monitored Premise

431

The dipole moment |p|2 is related to the emitted power W by the equation W=

ω4 √ μ εμ |p|2 , 12π

(11.2)

where μ is the magnetic permeability of the free space. of the electric field is defined by the following equaThe strength magnitude E tion, taking into account Equations (11.1) and (11.2): E ∗E = Er∗ Er + E∗ Eθ = = E θ 1 =

|p|2 (2π ε)2

k2 r4

+

1 r6

cos2 θ +

|p|2 (4π ε)2

1 r3

−

k2 r

2

+

k2 r4

(11.3) sin2 θ

Using Equations (11.2) and (11.3), the equation for the electric field strength magnitude can be rewritten in the form: E =

1

1 12πW √ μ εμω4

1 1 K1 cos2 θ + K2 sin2 θ, (2π ε)2 (4π ε)2

(11.4)

where K1 =

1 k2 + 4 ; K2 = 6 r r

1 k2 − 3 r r

2 +

k2 . r4

(11.5)

The equation for the magnitude maximum Emax (r) of the electric field strength, for all possible directions θ at given r, is E =

1

12πW √ μ εμω4

1

1 1 1 K1 − K2 cos2 θ + K2 . 2 2 (2π ε) (4π ε) (4π ε)2

(11.6)

This equation analysis shows the following: 1. If

1 (2π ε)2

K1 >

1 (4π ε)2

K2 , i.e., the first term of the radicand is more than zero, the

field strength maximum is provided for cos2 θ = 1 and is equal to (2π1ε)2 K1 . 2. If the first term of the radicand is less / than zero,/the field strength maximum is 1 2 √ W provided for cos θ = 0 and is equal to μ12π εμω4 (4π ε)2 K2 . Thus, the maximal value of the field strength is defined by the equation 1 Emax (r) =

1 ( 12πW 1 1 K ; K max √ 1 2 . μ εμω4 (2π ε)2 (4π ε)2

(11.7)

432

11

Detection and Localization of Technical Channels of Information Leakage

Figure 11.2 shows the curves of the electric field strength maximal values Emax generated by the rather low-power (100 μW and 1 mW) RES with frequencies of 30 and 300 MHz inside the MP versus distance R from 1 to 10 m. These figures show also the similar curves for the powerful RES (100 W), for instance, a radio broadcasting station displaced from the MP up to 2 km. The analysis of obtained curve character shows that, in the near-field region (1–8 m), the emission level from the low-power RES, as expected, exceeds the level of the powerful but remote sources. Relying on the obtained character of the emission intensity variation, we can offer the structural diagram of the system for radio signal detection emitted inside the monitored premise, which will be considered below. (a) Emax, V/m 0.20 1

0.10 2 3

0

1

3

5

7

9

R,m

5

7

9

R,m

(b) Emax, V/m 0.20

1

0.10 2

3 0

1

3

Fig. 11.2 Emax versus the distance for RES with frequency of 30 MHz (a) and 300 MHz (b) at power 1 mW (1), 100 μW (2) and 100 W (3) (the source remote to 2 km)

Generalized Structure of Equipment for TCIL Detection The above considered analysis shows that the equipment for TCIL detection should ensure the monitoring of the signal intensity variations inside the MP, which implies

Detection of Radio Signals Emitted in Monitored Premise

433

the usage of several receiving antennas in this premise. When using the multichannel receiver, the antennas used are connected directly to DDR inputs. If the antennas number exceeds the receiver input number, these antennas’ signal processing can be provided with the help of an antenna switch, which in turn connects the various antennas to the receiver input. Practical application shows that it is usually enough to use 2–4 similar antennas for monitoring the premise with an area up to 100 m2 . For correct comparison of the observed intensities of radio emission, we recommend the application of antennas with ulterior polarization having the quasi-isotropic pattern, since the polarization type of radio microphones is not standardized. Besides the above-considered peculiarities, we should take into account the following: • The fast search of radio signals in the wide frequency band requires that the equipment ensure a high rate of panoramic spectral analysis • The danger testing for the found radio emissions, executed at the third stage of processing – from the point of view of information leakage, can be most effectively provided using the special testing acoustic signals, which requires the presence of the acoustic system and the microphone. So, for effective solution of problems of detection and TCIL localization, the searching system must contain the following [4, 5]: • Set of wide-band antennas with ulterior polarization and quasi-isotropic pattern distributed inside the MP. It is desirable also that one of these antennas (“reference”) is situated outside the MP • Controlled antenna switch • Controlled radio receiver with the simultaneous analysis bandwidth (on IF) not less than several megahertz • Acoustic loud-speakers and microphone for test execution • Analog-digital processor (FFT processor) • PC.

Comparison Technique for Signal Intensities The modulation types used in usual radio microphones and their generated signals are, as a rule, the relatively narrow-band ones. So, the detection problem for radio microphone signals is similar to the problem of narrow-band radio signal detection, which was discussed earlier in Chapter 4. The initial data for this analysis in this case will be the power spectrum samples of the observed signals. Let us consider the sample set of the averaged power spectrum XRj (n) =

R 2 1 cj,r (n) , R r=1

(11.8)

434

11

Detection and Localization of Technical Channels of Information Leakage

where cj,r (n) is the samples of the discrete Fourier transform of the r−th registered sample of the processed wide-band random process obtained via the j−th channel (from the j−th antenna). The samples XRj (n) characterize the power distribution of the observed process over the frequencies. If the sample set with the numbers from nmin to nmax correspond to the signal u(t), the average power of this signal will be proportional to the sum of these samples. If we wish to express the power of this signal in decibels – with respect to some level, then the equation for the signal u(t) power Pˆ uj (in dB) estimation will take the form: Pˆ uj = 10 lg

n max

XRj (n) + μj ,

(11.9)

n=nmin

where μj is the correction factor defined by the calibration features of the antenna and the equipment of the j−th ARM channel. Let us now require the comparison of the signal u(t) power estimation at its observation via the j−th channel and the signal of another channel, which will be referred to as “reference” from now on. Because of non-identical characteristics, the correction factors μ of these channels will differ, so the exceeding of the observed power level in the j−th channel over the reference one, expressed in decibels, will be defined by the following relation: Pj = 10 lg

n max n=nmin

2 XRj (n)

n max

XR0 (n) + (μj − μ0 ),

(11.10)

n=nmin

where the values with index j correspond to the j−th channel, and values with index 0 to the reference channel. Practical application of Equation (11.10) becomes complicated due to the fact that the correction factors μ of each channel may vary during the measurements, for example, due to AGC system action. One can consider this factor in two ways. The first (deterministic) approach is based on the calibration of each channel and the AGC system disconnecting – or the estimation of its influence by means of the appropriate value re-calculation. If it is impossible to check the current operation mode of the channel and to re-calculate μ factor values, we should estimate the difference (μj − μ0 ) directly on the measurement results. For this, for instance, we may monitor the observed intensity of the additive noise, on the background of which the signal search is executed, and this intensity satisfies the relation: Pˆ n = 10 lg [X˜ R0 (n∗ )] + μ0 [dB]

(11.11)

where X˜ R0 (n∗ ) is the smoothed energy spectrum of the reference channel (see Equation 4.25), and n∗ = arg min X˜ R0 (n∗ ) is the minimum coordinate of this funcn tion corresponding to the “noisy” section of the observed random process spectrum. Assuming the equal intensity of the external noise for all receiving channels, for the difference of correction factors μ, we get the equation:

Detection of Radio Signals Emitted in Monitored Premise

μj = μj − μ0 = 10 lg{X˜ R0 (n∗ )/X˜ Rj (n∗ )}, dB

435

(11.12)

which is more convenient for practical application, than Equation (11.10).

Detection Algorithm for Radio Signal Sources in Monitored Area Most radio microphones use relatively narrow-band signals for information transmission. Therefore, at search of the dangerous radio emissions, as a rule, we do not need to analyze mutually the radio signals, which use frequency diversity. In this connection, we shall be limited by the procedure of signal processing in the current analysis bandwidth F. Practical realization of the dangerous radio emission detection in the bandwidth F depends on: • Number of antennas located inside of the MP • Presence of “reference” antenna, which locates outside of MP limits • Number of receiver channels. However, in spite of the differences related to these factors, signal processing at detection reduces to the following stage sequence: 1. For the current analysis bandwidth F, the “standard” panorama is defined (the specific variant of influence of the “standard” panorama obtaining on algorithm features will be discussed later). 2. For signals from all antennas located inside of the MP, the energy spectrum XRj (n) is calculated and, on this basis, in accordance with algorithms described in Chapter 4, the detection of narrow-band signal is executed. 3. If necessary, by means of the adaptive correction of μ factors for each channel, the calibration correction μj of the given channel with respect to the “standard” panorama is calculated. 4. For each of the found signals u(t), a power comparison in various channels is executed, and the maximal exceeding of the observed power level with respect to the “standard” panorama is determined Pmax (u) = max Pj , dB j

(11.13)

where pj is calculated in accordance with (11.10). 5. RES class corresponding to the signal u(t) is determined in accordance with the rule ⎧ forPmax (u) > thr2 ; ⎨ ρMP (11.14) = not defined forthr1 ≤ Pmax (u) ≤ thr2 ; ⎩ forPmax (u) < thr1 ρext

436

11

Detection and Localization of Technical Channels of Information Leakage

where ρMP unites the RES ensemble belonging to the monitored premise, and ρext is the ensemble of remote radio emission sources; thr1 is the threshold of decision making of the fact that the signal u(t) is formed outside of the monitored zone, i.e., that its source belongs to the remote RES; thr2 is the threshold of decision making of the fact that the RES for the signal u(t) is situated inside of the MP. The following factors affect on the detection features: 1. The choice of thresholds thr1 and thr2 , which may be corrected in each specific case, with account taken of the monitored zone (premise) properties, the peculiarities of receiving antenna location, and also of radio environment properties on the various sections of frequency range. 2. The applied technique of the “standard” panorama obtaining. Practical application of this algorithm has shown its effectiveness and capacity for work.

Detection Effectiveness Dependence on the Equipment and the Ways of “Standard” Panorama Obtaining As was mentioned earlier, in situations of complicated radio environments and when there is need for a fast rate of analysis execution, the type of equipment used and the ways of “standard” radio signal panorama obtaining become essential. To detect short-term radio microphone signals and signals with dynamic timefrequency distribution, and to make a fast decision about the presence or absence of radio microphones inside the monitored premise, e.g., during important meetings, high performance ARM equipment is necessary. When using a single-channel receiver, the signal power level measurements from various antennas can be executed consecutively in time only. Because of modulation presence, and taking into account that the signal bandwidth is also unknown and, at measurement, we should take its estimation, considerable error occurs in the power estimation, which leads to an increased probability of false detection or signal missing. Moreover, longer signal-processing time is the natural consequence of single-channel equipment usage. The double-channel method ensures a higher detection probability of short-term and pulse signals, under these conditions (see Chapter 5). Application of the doublechannel receiver, the channels of which are synchronously tuned to the same analysis bandwidth, reduces the measurement duration and increases the accuracy of signal power estimation. The increasing accuracy can be achieved due to the fact that the time sample obtaining is synchronously fulfilled on both channels, and the power level estimation is executed under equal conditions. Since comparison of the relative variables takes place at decision-making, the errors related to the power estimation algorithm errors do not practically affect on the result [6].

Identification and Localization of Radio Microphones

437

The most effective solution of the above-mentioned problems can be achieved with the multi-channel receiver application, which allows simultaneous measurement execution both outside and within several points inside of the MP. Irrespective of the channel number in the used receiver, essential improvement of the detection algorithm characteristics is observed when using the “reference”, receiving channel, basing on the antenna located outside of the MP. This channel presence allows the correction of the “standard” panorama, in accordance with the changing radio environment.

Identification and Localization of Radio Microphones In addition to solving the problem of radio emission source detection inside of the MP, it is necessary to determine the danger degree of the found emission. To do so, it is necessary, first of all, to identify it, i.e. to re-check whether the found emission belongs to the radio microphone installed inside of the MP. In cases when the radio microphone is really dangerous and the “radio-game” cannot be supposed, i.e., the transmission of wittingly false information does not occur, we should localize the radio microphone, i.e., to determine its location. Let us consider the approaches for radio microphone identification and localization [4, 6], which are suitable for a large number of modulation types, including AM, narrow-band and wide-band FM, and continuous modulation types with the spectrum inversion and with the frequency mosaic. For identification, acoustic testing signals of various types are used. After receiver tuning to the frequency of one of the “new” found RES, acoustic sounding in the MP is provided with the help of the usual sounds (fast rhythmic music, continuous speech) or of the special signals formed by TCIL detection equipment. By various demodulators, the received radio signal is demodulated, and correlation processing with the modulating signal is executed for each demodulated signal in time and spectral domains. On the basis of the correlation coefficient value, a decision is made on the received signal source belonging to the radio microphone located inside the premise. To maintain the effectiveness of this approach when using simple concealment methods, such as the spectrum inversion and the frequency mosaic, a doublefrequency signal or chirp-signal is used as the testing signal, and the processing is provided in the spectral domain. The decision regarding radio microphone presence is made on the basis of the presence of the spectral components of the testing signal in the demodulated signal, over the time interval T, which is equal to the sounding duration of the testing signal, and of its absence outside of this interval. The mutual unambiguity of the acoustic signal-transform lies in the basis of the algorithm construction for detection of radio microphones with static waveform scrambling in the frequency domain, i.e., the possibility to recover unambiguously the initial signal basing on the transformed signal. This means that, at multiple repeating of any acoustic signal, the emitted signal will be also repeated. Similar repetition allows the determination of the radio microphone presence.

438

11

Detection and Localization of Technical Channels of Information Leakage

Radio microphone localization is executed on the basis of measurement of the delay duration of the testing signal as it arrived to the radio microphone, radiated by the acoustic loudspeaker, on the basis of the triangulation method. In the simplified case, the algorithm is realized using two acoustic loudspeakers. At that, the calculation ambiguity is eliminated by the fact that the acoustic loudspeakers are situated near the walls of the monitored premise. Beginning in 1996, similar approaches to radio microphone identification and localization began to be implemented in searching equipment, including the ARKD1, ARK-PK-3KU, ARK-PK-5KU, ARK-D1T and all further modifications having the single-channel DRR on the modern element base. In this equipment, the bandwidth of simultaneous analysis is equal to 2 MHz, and the panoramic analysis rate is 150 MHz/s for the 3rd generation equipment and 800 MHz/s for the 5th generation equipment, in the whole operating frequency range. Using a system of several antennas with quasi-isotropic pattern distributed inside the MP, one can ensure the sensitivity, which makes it possible to detect the radio microphone with 100μW of power in the premise of 10 × 10 m (100 m2 ).

Distant Radio Monitoring Systems of Remote Premises Construction Principles of Remote Radio Monitoring Systems If it is necessary to check TCIL absence in several premises, it is not always possible to install separate radio monitoring systems in each separate MP, since in this case it is difficult to conceal the operation. Moreover, this solution translates to large financial expenses. The problem of constructing remote radio monitoring systems (RRMS) for several premises in a building with one central post (CP) has its own specific peculiarities, and attempts to solve it by means of the simple combining of autonomous equipment (for the monitoring of one premise) and additional equipment not adapted specially for this purpose do not lead to positive results [7]. The results discussed further are correct for the typical situation, when RRMS location is carried out inside of an already existing building, as well as for the case when RRMS presence was considered at the building’s design, i.e., it is coordinated with the building’s structure, its wire and power communications. A possible way to ensure the security of several premises consists in the use of one high-speed complex, which supplies simultaneously several MP. Such a complex can be installed in a specially-allocated operating room, with external antennas connected with the operating room by the cable lines attached to the monitored premise. In this case, the problem of concealed system operation can be solved much easier, since the external antennas may be formed for the camouflaged installation. The cost of such a camouflaged system will be less compared to the cost of several autonomous systems. Figure 11.3 shows the example of RRMS for the radio monitoring of three premises. The central module of the system, together with the antenna switch, is installed in the special operating room. Antenna systems are taken into the

Distant Radio Monitoring Systems of Remote Premises Room 1 Antenna 1

439 Room 2

Antenna 2 Reference antenna

Room 3 Antenna 3

Antenna switch

Central module

Operator room

Fig. 11.3 System of remote radio monitoring with external antennas

monitored premises. The reference antenna is installed outside of the limits of the monitored zone, for example, on the building roof. The operation speed of the central module, the antenna switch, and the permissible length of the cable lines all limit the quantity of monitored premises. We would like to note that the structure presented in Fig.11.3 is used in practice, but it has several disadvantages. The first shortcoming consists in the fact that the antennas are used only inside of monitored premises. Such an approach allows for radio microphone detection of a “new” emission source with accuracy up to the premise only, but does not provide an exact answer to the question of whether the newly-detected source belongs exactly to the facilities of unapproved information receiving. Also, it becomes impossible to determine in which place within the limits of the premise this source is situated. In order to answer these questions, a microphone and acoustic loudspeakers should be mounted additionally in each MP in such way that the identification and location of the radio microphone can be provided. Moreover, inside some most important premises, it is expedient to use special remote-controlled generators, which can create spot-jamming at the frequencies of the detected RES. The second shortcoming consists in the fact that the frequency range, in which the unapproved facility will operate, is not a priori known. Substantially, the radio microphone radiation frequency may be in the limits from tens of kilohertz to tens of gigahertz. Use of a single receiving antenna in each premise is insufficient since it is difficult or practically impossible to ensure such a great frequency coverage coefficient for the single antenna. It is expedient to have an antenna set, each of which has its own operating frequency sub-range, for example: • • • •

9 kHz–30 MHz 25–3,000 MHz 3,000–8,000 MHz Higher than 8,000 MHz.

The necessity of fragmenting the monitored frequency area into sub-ranges is caused by the differences in antenna design and in switching devices. This fragmentation is rather conditional. The sub-system operating range boundaries may

440

11

Detection and Localization of Technical Channels of Information Leakage

vary; however, the mentioned approach as a whole is the most expedient under the “effectiveness – cost” criterion. The application of wide-band antenna groups provides the system sensitivity growth, however, it relates to the additional difficulties caused by the signal transmission from each antenna, also. The monitoring at frequencies lower than 9 kHz, assuming the wire communication observation, AC net, etc., oversteps the limits of this book and represents a separate analysis. The third shortcoming consists in the fact that the application of the long radio frequency cables connecting the monitored premises with the operating room leads to degradation of the system’s potential sensitivity. In actuality, as was shown in section “Digital Radio Receivers”, the coaxial cable introduces the attenuation and, as any device with losses, has a noise factor approximately equal to the signal attenuation at room temperature. If the coaxial cable introduces the attenuation only, it is possible for that to be compensated for by the amplifier, at its output, however, this amplifier will also amplify the noise, which is caused by the cable. In Chapter 3, Fig. 3.16 showed the linear attenuation (for 1 m of length) versus the frequency for several types of Russian coaxial cables. As we can see from the figure, the signal attenuation value in the cable and hence, its noise factor increases with the growth of the transmitted signal frequency. The less attenuation the cable has, the larger (as a rule) its diameter and the permissible bending radius are, and the more complicated cable laying is. To reduce losses and noise parameters we should aspire to decrease the length of the cable line and to lay the cable along the shortest distance from the monitored premise to the operating room. However, this recommendation, as a rule, remains little more than a good intention only. In practice, the cable is laid not along the straight lines connecting the premises, but along the walls and partitions, starting from the cable channels disposed in the building, which essentially increases the cable length. Growth of the cable length leads to an increase of the antenna effect, at which signals from strange sources are induced on the cable. As a result, at cable output, a mixture of the useful signal, received by antenna, and the induced unauthorized signal is observed. To reduce the cable influence on the system’s sensitivity, it is necessary to use low-noise preliminary amplifiers at cable input (see section “Digital Radio Receivers”), i.e., factually, the antenna system should be active. The same measure decreases the antenna effect at the expense of the transmitted signal level increase and the matching improvement. We would note that a decreasing of the antenna effect is achieved also if the differential circuits of symmetric cable connection and the double shielding are used. But the most efficient way to reduce the cable influence is by preliminary frequency conversion of the signal received by the antenna into the lower range. The application of the external modules is a possible solution, aimed at improving the operation of the system shown in Fig. 11.3. These modules are the devices installed inside the monitored premises, to which the groups of wide-band antennas are connected, together with the acoustic loudspeakers, the microphone, and the spot-jamming generator. The similar RRMS structure with the external modules is presented in Fig. 11.4.

Distant Radio Monitoring Systems of Remote Premises Room 1 External module 1

Loudspeaker Microphone

Room 3 External module 3

Loudspeaker

441 Room 2

External module 2

Loudspeaker Microphone

Control unit of external modules

Module of reference antennas

Microphone

Central module

Operator room

Fig. 11.4 System of remote radio monitoring with the external modules

In this structure, the external module may have the antenna switch, the secondary power supply for the active antennas, the control unit, for the spot-jamming generator, the signal formation unit for the acoustic loudspeakers, and the controlling microprocessor. The high-frequency signals from the selected antenna and the acoustic signals from the microphone are transmitted from the external module via the coaxial cable. For the transmission of control signals, we may use the additional low-frequency cable, the cost of which is not high.

Examples of Remote Radio Monitoring Systems Until 1999, in practice, RRMS of the 2nd generation were used: ARK-D3K using ARK-D1, ARK-PK-3KU, ARK-PK-5KU, the single-channel systems developed on the basis of modified AR-3000A, AR-5000 imported receivers. Since 1999, due to the development of the Russian ARK-CT1 DRR, which has more efficient characteristics than for the mentioned imported receivers, these systems were discontinued and RRMS of the 3rd–5th generation – ARK-D3T, ARK-D9 and ARK-D13 came to take their place. The core of each of these systems is the central module, which is the automated radio monitoring and TCIL detection system. Its dynamic range, the operation rate, the resolution, the automation degree of the signal processing and the performance are the governing factors for the whole system. In the ARK-D3T RRMS, the ARKD1T single-channel system with the ARK-CT1 DRR, or the ARK-D1TM with the ARK-PR5 DRR is the central module (see Chapter 3). In the ARK-D9 and ARKD13 systems, the ARK-D7K and ARK-D11 double-channel panoramic systems are used, with the panoramic analysis rate in each channel, relatively, 1,000 and 1,500 MHz/s. The distinctive peculiarities of these systems are: • High rate of panoramic spectral analysis, which ensures the fast detection of any radio microphones, including the short-term periodic emission ones, the wideband transmitters, and devices with digital modulation types

442

11

Detection and Localization of Technical Channels of Information Leakage

• The presence of several antennas in each monitored premise on the appropriate operating frequency ranges, in combination with its discrete switching. This ensures the obtaining of high sensitivity of the searching system; • DRR dynamic range on the mutual intermodulation of the 2nd and 3rd order is not less than 70 dB, which provides the system operation in the complicated electromagnetic environment peculiar to large towns. The ARK-D3T RRMS provides the automatic panoramic search with a 2 MHz step, at the real re-tuning rate of not less than 100 MHz/s (600 MHz/s for ARK-D1TM). The ARK-D9 and ARK-D13 systems possessing the double-channel central module ensure the automatic panoramic search with a 5 MHz step, executing synchronously on two channels at the real re-tuning rate of not less than 1,000 MHz (1,500 MHz for ARK-D13). Together with the hardware, the software of the complex realizes practically all presently known approaches for detection and identification of radio microphones (both active and passive) including: • Division of RES into external and internal, by means of alternating (ARKD3T) and synchronous (ARK-D9, ARK-D13) comparison of radio signal levels received from the selected (by the definite algorithm) antenna pair • Synchronous monitoring of radio sources on the frequencies, which are the multipliers of the detected frequency (test by harmonics) • Correlation search between the acoustic influence (environment background or the special testing signal) and the received radio signal parameters • Recording on PC hard disk of the radio signal on IF • Real-time technical radio signal analysis and post-processing • Determination of modulation type and demodulation of the registered digital signals • Measurement of the radio signal parameters with various modulation types • Aggregation in database of the information concerning RES and its parameters, obtained with the help of technical analysis, for further application. Besides the particular searching function (i.e., detection of unauthorized radio emissions), these systems ensure: • Registration of the radio environment in “frequency-level-time” coordinates, with the formation of files with the results on each monitored premise • Fixation of operating frequency-range loading with the aggregated results on each premise • Saving the obtained information in database, with the possibility of further deferred processing by the special program package • Adaptation to the industrial noise and interference level • Report generation during the defined operation period • Remote monitoring and administration via TCP/IP network.

Distant Radio Monitoring Systems of Remote Premises

443

Peculiarities of ARK-D3T Remote Radio Monitoring System The ARK-D3T remote radio monitoring system is based on the ARK-D1T and ARK-D1TM single-channel radio monitoring systems intended for wide-band automated radio monitoring and TCIL detection. The ARK-CT1 DRR is used in ARKD1T system with the frequency range from 9 kHz to 2,020 MHz. The ARK-PR5 DRR with the operating frequency range from 9 kHz to 3,000 MHz is used in the ARK-D1TM system. The central module of the ARK-D3T system is shown in Fig. 11.5 and it has the case view. On the module faceplate, there are eight inputs of antenna switch, the input for the microphone connection, the loudspeaker, sockets for the acoustic loudspeakers, connectors for connection with a PC.

Fig. 11.5 Central module of ARK-D1T complex

The ARK-A5 (20–3,000 MHz), ARK-A8N (10 kHz–6,000 MHz) or ARKA12 (20–3,000 MHz) antennas of external implementation are used. Inside the premises, the following antennas can be used: ARK-A2 (20–1,000 MHz); ARK-A6 (20–1,000 MHz); ARK-A10 (1,000–2,000 MHz); ARK-A11 (2,000–8,000 MHz); ARK-A8V (10 kHz–6,000 MHz). Main parameters of the mentioned antennas are listed in Table 11.1. The ARK-KPS control unit of wire networks with an active ARK-ASP2 and passive ARK-PCP2 network probe is a part of the equipment. The ARK-KNV3 (3–8 GHz) or the ARK-KNV4 (3–18 GHz) radio signal converter, and the ARK-SPM (40–1,000 MHz) unit for spot-jamming generation, can

444

11

Detection and Localization of Technical Channels of Information Leakage

Table 11.1 Antennas used in remote radio monitoring systems

External view and implementation

Operating frequency range, MHz

SWR

Pattern

Sizes, mm

20–3,000

In assembled view 680 × SWR Quasi520 × 540; in ≤3 isotropic packed – 520 × 320 × 70

20–2,000

Antenna with SWR one≤3 sided pattern

0.01–6,000

SWR Quasi260 × 260 ≤3 isotropic × 12

Fig. 11.6 ARK-A5 (outside the premises)

320 × 320 × 15

Fig. 11.7 ARK-A6 (outside and inside the premises)

Fig. 11.8 ARK-A8V (inside the premises), ARK-A8N (outside the premises)

Distant Radio Monitoring Systems of Remote Premises

445

Table 11.1 (continued)

External view and implementation

Operating frequency range, MHz

SWR Pattern

Sizes, mm

1,000–2,000

SWR Quasi90 × 90 ≤3 isotropic × 63

2,000–8,000

SWR Quasi90 × 90 ≤3 isotropic × 45

20–3,000

SWR Quasi522 × 470 ≤3 isotropic × 96

Fig. 11.9 ARK-A10 (inside the premises)

Fig. 11.10 ARK-A11 (inside the premises)

Fig. 11.11 ARK-A12 (outside the premises)

be included in the system structure. The technical parameters for the ARK-D3T system’s basic structure are listed in Table 11.2. The ARK-D3T system is manufactured in two versions. In version 1, the system has the single external module intended only for the signal switching of the

446

11

Detection and Localization of Technical Channels of Information Leakage

Table 11.2 Technical parameters of the ARK-D3T system’s basic structure Parameter Main operating frequency range, MHz, not less For ARK-D1T For ARK-D1TM Additional operating frequency ranges, GHz: At presence of ARK-KNV3 converter At presence of ARK-KNV4 converter Panoramic analysis rate, MHz/s, not less For ARK-D1T For ARK-D1TM Resolve capacity, kHz, not more Dynamic range, dB, not less Receiving section sensitivity in bandwidth of 12 kHz for panoramic analysis, μV, not more Relative accuracy of the central frequency measurement, not worse Operating frequency range of the ARK-SPM spot-jamming formation unit, MHz Minimal re-tuning step of ARK-SPM, kHz Output power of ARK-SPM, mW Operating frequency range of ARK-KPS wire networks monitoring unit, kHz, not less Maximal input voltage of the ARK-PSP2 probe, V Maximal input voltage of the ARK-ASP2 active probe: At frequencies less 60 Hz, V At frequencies from 60 Hz to 20 kHz, V At frequencies from 20 kHz to 5 MHz, V Input resistance: Of ARK-ASP2 probe, MOhm, not less Of ARK-PSP2 probe, kOhm, not less Supply voltage of ARK-D1T: AC, V DC, V Consumed power, VA, not more Overall sizes of the central module, mm, not more Weight, kg, not more

Nominal value

Notes

9 kHz–2,020 GHz 9 kHz–3,000 GHz 2,000–8,000 2,000–18,000 140 600 7 70 3

For operation with a single antenna

+10–6

In frequency range higher 20 MHz

40–1,000

12.5 100–150 0,4–30,000 400

400 50 10 1 1 90–240 13 ± 3 55 465 × 427 × 175 15

reference antennas. Antenna systems installed in the monitored premises are connected directly to the antenna switch of the central unit. In version 2, the control unit for the external modules and four external modules installed in the monitored premises are added to the system structure. Figure 11.12 shows the connection diagram for the ARK-D3T RRMS with the external modules. The external view of the external module is shown in Fig. 11.13, and Fig. 11.14 shows its structural diagram.

Distant Radio Monitoring Systems of Remote Premises

447

Antenna

Antenna

Antenna

Antenna

ARKA11

ARKA11

ARKA11

ARKA11

Antenna

Antenna

Antenna

Antenna

Antenna

Antenna

Antenna

Antenna

ARKA2A1

ARKA11

ARKA2A1

ARKA11

ARKA2A1

ARKA11

ARKA2A1

ARKA11

"20-1000 "1-2 GHz "2-6 GHz" MHz"

"20-1000 "1-2 GHz" "2-6 GHz" MHz"

External module "Output"

"Input1"

"Control"

"Control1"

"20-1000 "1-2 GHz" "2-6 MHz" GHz"

External module

"20-1000 "1-2 GHz" "2-6 GHz" MHz"

External module

External module

"Output"

"Control"

"Output"

"Control"

"Input2"

"Control2"

"Input3"

"Control3"

"Control"

"Output"

"Input4"

"Control4"

To external modules

Control unit for external modules "Output""Case " "=10...16 V"

"~220V"

Antenna

ARK-MA1

Power supply unit

Power supply unit

Passive indicator Antenna

Measuring antenna

ARKA7A

Active indicator

Jammer

Antenna

ARKA2A1

"Ant.4" "D3"

"Net "Supply =10...16 V" ~220V"

"Supply of "SW ant.1" SW ant."

"SW ant.2" "Add. port"

Antenna "Ant.3"

ARKA8

"Ant.2"

Central module ARK-D1T "To PC"

Antenna "Ant.1"

ARKA2M

"Ant"

"LPT" "COM"

Antenna

ARKA5

"Control"

Antenna amplifier

"To PC"

Antenna to

Antenna

ARK-KHV1

ARK-A2

Fig. 11.12 Connection diagram of ARK-D3T (Version 2)

The ARK-KNV3 radio signal converter transforms signal frequency from the range of 2–8 GHz to the range down to 1 GHz. The RF switch connects one of three antennas or the microphone to the central unit. The acoustic unit controls the loudspeaker operation and the external microphone. The control unit serves to communicate with the central unit or with the control unit of the external modules and defines the operation mode of the RF switch, the acoustic unit, and the frequency converter.

448

11

Detection and Localization of Technical Channels of Information Leakage

Fig. 11.13 ARK-VM-K2 external module From central module

ARK-A11K

Converter KNV3

Control unit

Acoustic unit

To loudspeaker From microphone

RF switch

Antenna amplifier

To central module

ARK-A2A ARK-A11A

Fig. 11.14 Structural diagram of the external module ARK-VM-K2

The power supply of all devices included in the system (except the PC and the control unit of the external modules) is provided from the secondary power supply unit built into the central module. The primary power supply is provided from AC networks of 90–250 V (50/60 Hz) or from the vehicle on-board network with the voltage of 10–16 V. The emergency power supply is stipulated in the system, namely rechargeable battery installed inside the case of the central module and intended for shortterm operation, mainly for emergency operation completion, i.e., for data saving in the PC.

Peculiarities of the ARK-D9 Remote Radio Monitoring System The ARK-D9 remote radio monitoring system is designed on the basis of the ARKD7 double-channel radio monitoring and TCIL detection system. At its core is the ARK-CT3 double-channel DRR, which has better technical parameters compared to the ARK-CT1, especially in the rate of panoramic analysis. Up to 16 external modules can be included in the ARK-D9 system. Figure 11.15 shows the external view of the central unit: the ARK-D7 radio monitoring and TCIL detection system.

Distant Radio Monitoring Systems of Remote Premises

449

Fig. 11.15 ARK-D7K central unit

The main technical parameters of ARK-D9 are listed in Table 11.3. In contrast to the ARK-D3T system, the ARK-D9 RRMS is able to fulfill the double-channel synchronous TCIL search and detection, which increases the detection probability of the short-term RES together with the high rate of frequency re-tuning. At present, the ARK-D9 RRMS is out of production, after appearing in the modules of the ARGAMAK series and the panoramic DRR on its basis, and it has now been perfected into the new ARK-D13 system.

Table 11.3 Main technical parameters of ARK-D9 system Parameter Panoramic analysis and fast signal search Number of monitored premises Operating frequency range of the basic set, MHz Operation range in full set, MHz Rate in the operation range, MHz/s At single-channel search At double-channel search Dynamic range on 2nd and 3rd order intermodulation, dB Radio signal recording, technical analysis and parameter measurement Bandwidth of processed frequencies/resolution

Value

Up to 16 25–3,000 Up to 8,000 3,000 1,500 75

5 MHz/15 kHz, 250 kHz/500 Hz, 120 kHz /240 Hz, 50 kHz/100 Hz, 9 kHz/20 Hz, 6 kHz/12 Hz

450

11

Detection and Localization of Technical Channels of Information Leakage

Peculiarities of the ARK-D13 Remote Radio Monitoring System The main technical parameters of the ARK-D13 system are listed in Table 11.4. This system provides the monitoring of up to 64 remote premises by a single set of the ARK-D11 double-channel radio monitoring equipment installed in the central post. This expansion of the monitored premise number became possible owing to the usage of the unified antenna switches with eight inputs, which ensures treelike branching among them. Thus, it is possible to connect eight other switches to a single switch. The treelike structure essentially reduces the coaxial cable expense and simplifies its wiring. Table 11.4 Basic specifications of ARK-D13 Parameter

Value

Panoramic analysis and fast signal search Number of monitored premises Operating frequency range, MHz In basic integration In full integration Bandwidth of simultaneous analysis, MHz Panoramic spectral analysis rate in the range 25–3,000 MHz at bandwidth 5 MHz and FFT discreteness 6 kHz, MHz/s At single-channel search At double-channel search Dynamic range on 2nd and 3rd order intermodulation, not less, dB Integral sensitivity of the system (transmitter power in the premise with 10 × 10 m area, detected with the probability 0.99), μW Radio signal recording, technical analysis and parameter measurements Bandwidth of the processed frequencies/resolution

Up to 64 25– 3,000 0.009–18,000 5

3,000 1,500 75 70

5 MHz/15 kHz, 250 kHz/500 Hz, 120 kHz/240 Hz, 50 kHz/100 Hz, 9 kHz/20 Hz, 6 kHz/12 Hz

In the ARK-D13 system, perfected external modules are used, in which the ARKPS5 radio signal converters of the ARGAMAK family are built in. The module’s structural diagram is presented in Fig. 11.16. Signals from antennas are converted From central module

ARK-A11K

ARK-A2A

Converter KNV3

Control unit

Acoustic unit

To loudspeaker From microphone

Signal converter PS5

IF buffer

To central module

Fig. 11.16 Structural diagram of the external module with the ARK-PS5 radio signal converter

Software for Remote Radio Monitoring Systems

451

into an IF of 41.6 MHz, which reduces the losses and the connecting cable noise factor to a minimum, and essentially decreases the antenna effect influence. Additionally, control signal transmission, together with IF output signal, is implemented in this module through the single cable.

Software for Remote Radio Monitoring Systems The problem of remote radio monitoring is solved by the developed package of customized software. The main problems solved using the software packages, together with the equipment, are listed in Table 11.5. Table 11.5 List of problems solved by the hardware-software complex The problem solved Automatic detection of radio microphone emission, determination of its location in the monitored premise Monitoring of wire networks in the presence of transmitter voltages providing the acoustic information leakage in the monitored premise and its transmission via monitored network Interference generation for signal receiving from radio microphones in the area of its electromagnetic accessibility at the fixed frequencies (for the force structures only) Real-time panoramic analysis including detection and measurement of emission parameters, representation of the radio signal spectral structure on the monitor screen and recording of information about the detected emissions in the database, registration of the radio environment for further post-processing Post-processing of the results of the radio environment recording Recording of the radio signal fragments in vector form on the PC hard disk, including in the bandwidths of 5 MHz, 2 MHz, 250 kHz and 15 kHz Continuous radio signal recording in vector form on the PC hard disk, in the bandwidths up to 250 kHz Real-time technical radio signal analysis and its post-processing Demodulation of received AM, NFM, WFM, LSB, USB signals, with the possibility of its recording on the PC hard disk

Additional equipment, software package SMO-DX

SMO-DX

Unit ARK-SPM, SMO-DX package

SMO-PA

SMO-ASPD SMO-PA, SMO-STA

SMO-STA SMO-STA Any of the software packages

The SMO-PA and SMO-STA software packages were described in the previous chapters, so now we concentrate on the SMO-DX software package, in more detail.

452

11

Detection and Localization of Technical Channels of Information Leakage

Purpose and Possibilities of SMO-DX Application The SMO-DX software package is a component of the customized mathematical software of both single-channel and double-channel automated systems for radio monitoring and radio microphone detection. The DX program provides the execution of the following functions: • • • • • • • • • • • • • •

Controlling of the hardware complex Urgent estimate of radio environment Real-time spectral radio signal analysis Operation in accordance with the various tasks defined by the user Urgent examination of obtained panorama (simultaneously with its renewing) Recording of aggregated spectrum files into database Examination of earlier aggregated spectrum panorama and operation with it Detection of new RES Estimation of the received low-frequency signal correlation with the acoustic oscillation inside the premise, using the various acoustic testing signals including on the basis of the natural acoustic background Detection of radio microphones of various types including scrambled, and using the modulation of complicated type Detection of radio microphone location in the limits of premise Saving new and identified RES to the parameters database Work with database of the registered sources Generation of reports on the operation results.

The following functions are available in the presence of the ARK-SPM spotjamming generation unit: • • • • •

Formation of the task for spot-jamming generation Setting the modulation type for jamming (carrier, NFM, WFM) Modulation by the external audio signal Setting of the tuning order to frequencies Setting the transmission time on each frequency. The following functions are available at wire line analysis:

• • • •

Analysis of RF signal spectrum in AC network and in other wire lines Detection of RF signals from the microphones in wire lines Detection of LF signals from the microphones in wire lines Search of microphone location connected to wire lines.

The program operates under the Windows 2000 or Windows XP operating systems, in multi-task mode, both in local and in network versions. SMO-DX application maintains the operation with various equipment types (with the appropriate program-drivers).

Software for Remote Radio Monitoring Systems

453

SMO-DX application ensures a unified user interface for operation with various equipment types. Interface parameters are dynamically adjusted in accordance with the used hardware system. The interface frame of SMO-DX does not address the equipment directly but interacts with it via the equipment controller. Data exchange between SMO-DX applications and the equipment controller may be executed both via the mutual memory and by means of network interaction using TCP/IP protocol. This allows the creation of the distributed systems, in which the operator controls remotely the remote hardware system. The example of such system implementation is shown in Fig. 11.17. Operator PC DX program

Remote PC Network with protocol TCP/IP

Technical analysis program STA

Driver-program for equipment

Hardware complex

Fig. 11.17 Structure of distributed system

Network exchange support can be provided in various ways, for instance, with the help of network cards via coaxial or twisted-pair lines, via phone-pair on the basis of modems, and wirelessly with the help of radio modems. If necessary, the operation in local mode is possible using a single PC.

Peculiarities of Radio Microphone Detection The above-mentioned algorithms for detection of “new” emissions, radio microphone identification and localization, harmonic analysis, and correlation checking of the acoustic signal inside the premise and the received signal are implemented in the application. Implementation of the detection algorithm in SMO-DX provides for its separate as well as combined usage (in any combination), depending on the stated problem. In this application, adaptation to the electromagnetic environment is anticipated, which results in the generation of a file containing the amplitude-frequency loading of the operating frequency range called the standard panorama. At sufficient aggregation time in this panorama, stations that emit periodically will also be registered, for instance, mobile system stations. Comparison with the standard panorama allows the detection of new radio emission sources based on the earlier aggregated data.

454

11

Detection and Localization of Technical Channels of Information Leakage

The algorithm of the radio signal spatial selection is anticipated in the application [3], for which the distributed antenna system is necessary. The spatial selection is based on the relatively-fast attenuation of the radio signals in the near-field zone and on its attenuation by the constructive elements of the object (walls, grates, partitions). To distinguish radio signals emitted inside the premise and the external radio signals, an external reference antenna is used. In most instances, radio microphones have a strongly pronounced bandwidth, in which the main radio emission energy is concentrated. For reliable receiving, the signal level of the radio microphone should essentially exceed the jamming level. The noise level estimation is anticipated in the software for the radio signal search, exceeding the jamming level by the value mentioned in the task. As a rule, radio microphones emit a radio signal not only on the fundamental frequency but on its harmonics also. The higher harmonic presence allows the distinction of the radio microphone emission and the signals from external stations, to which parameters strong restrictions are applied. In SMO-DX, an analysis algorithm for the list of “new” frequencies is provided, for higher harmonic presence detection. At that, each frequency from the list of “new” frequencies is consecutively analyzed and its higher harmonics are searched. In the case when the radio receiver is tuned to the radio microphone frequency and the appropriate demodulator type is activated, the low frequency signal at the demodulator output is correlated with the acoustic oscillations inside the premise. At that, for scrambled radio microphones (concealed with the help of the inverse spectrum or the frequency mosaic), the correlation character is more complicated. To determine the correlation between the signals, the program has correlation checking between the natural acoustic background of the premise and the demodulated signal (passive test), and also between the specially-emitted audio tone signals and the signals with linear frequency modulation (active test). The main advantage of the passive test is the fact that, for its execution, we do not need the emission of special acoustic testing signals, which decamouflage the detection system, but cannot detect the scrambled radio microphones. The best indices of the passive test are provided in presence of natural acoustic background in the form of fast rhythmic music.

Joint Usage of the Various Detection Algorithms The “threshold level” concept is introduced in SMO-DX, for implementing the use of the various detection criteria. The threshold level is calculated separately for each antenna. A new frequency search is also provided separately for each antenna. This improves the system operation, when antennas are situated in various receiving conditions (e.g., in the various premises, at various losses in the cable, or at using the various types of antennas, or antennas with different patterns).

Software for Remote Radio Monitoring Systems

455

Radio microphone Localization Inside of Monitored Premises The problem of determining radio microphone localization is solved at a later stage, after its identification. The location is determined on the basis of delayed measurement of the demodulated signal, with respect to the acoustic probe signal and locating of the acoustic source positions and the place of the probable location, in accordance with the premise plan, the results of which are represented in Fig. 11.18.

Fig. 11.18 SMO-DX application window at determination of the radio microphone location

Equipment Operation in the Remote Radio Monitoring System Several tasks may be saved to a database of the program, but only one can be active at a time. The task list structure is shown in Fig. 11.19. In the “Used antennas” task section, those antennas are presented, which are used for the receiving. In the “Used rooms” sections, those rooms are presented, in which the execution of the acoustic tests is permitted. Moreover, there is an “AntennaRooms” association in the task sections for each antenna, which allows for more accurate system adjustment, for a multi-room operation and, hence, improves operation quality. The need for this association consists in the fact that a signal can be

456

11

Detection and Localization of Technical Channels of Information Leakage File 1

Tasks list

Setting for antennas - antenna ensemble - attenuator 1; - attenuator 2; - amplifier; - exceeding over panorama; - exceeding over noise; - exceeding over reference antenna; - reference antenna number; - usage option of reference antenna

File 2 .....

Task 1 - used antennas - used rooms - list of associations “Antennas - rooms”

File M

Range 1 - antenna group - lower frequency - upper frequency - average

Task 2

Setting for antennas 2

Range 2 .....

Task N

.....

.....

Range K

Setting for antennas L

Fig. 11.19 Task list structure

induced on the antenna from different rooms and, at detection of the signal induced in this antenna is impossible to decide unambiguously, in which room the radio microphone is located. To eliminate this ambiguity, we may provide the acoustic test consequently, in all rooms associated with this antenna. Figure 11.20 shows the example of an antenna system location inside the monitored premise. By dashed-line curve, the radius of the antenna action is indicated. For this location, it is necessary to arrange the following associations: • Antenna 1 should be associated with the room 1 • Antenna 2 should be associated with rooms 2, 4 • Antenna 3 should be associated with rooms 3, 4.

Room 1

Room 2

Antenna 1

Antenna 2

Antenna 3 Room 3

Room 4

Fig. 11.20 Possible antenna location

In the task sections, there is the “Test the antenna with the maximal signal only” option, which is useful for single-room system use. At that, if the signal is detected from the outputs of several antennas, the antenna with the maximal signal only is selected for further testing. This reduces the total testing time, because, in the case

Detection of TCIL Sources by the Mobile Station

457

of non-exact setting of the “Antenna-Rooms” associations, the maximal signal can be induced in the antenna where the signal source is situated, regardless of the room, thereby avoiding radio microphone missing.

Detection of TCIL Sources by the Mobile Station This section considers the application features of the mobile radio monitoring and direction-finding system, with regard to object checking on the presence of TCIL sources. For this purpose, the following types of radio monitoring and directionfinding mobile systems can be used: “Argument” and the mobile direction finders of regular structure, e.g., ARTIKUL-M1. It is not necessary for the system structure to have navigation and cartographic systems and the equipment for digital data transmission, in order to search and to localize RES in a specific object. At the same time, the specificity of the problem of CEE detection produces a number of additional requirements for the system. First of all, ensuring the concealment of the system’s operation is the important requirement. The external vehicle view should not attract the attention of those around it or tell about its purpose. Antenna array (AA) mounted on a vehicle should not decamouflage it; however, it is not necessary that the antenna system provide equal high accuracy of direction finding for all possible directions of signal arrival, from 0 to 360◦ . It is enough that the antenna system should have good accuracy parameters for several survey sectors only, for example, from the sides of the vehicle, since, under conditions of urban build-up, in most cases, the vehicle assumes position exactly by the side of the monitored object. We should not completely disregard the antenna system characteristics in other sectors, though, because the environment around the object can prevent the vehicle from taking the optimal position for the equipment operation. Moreover, it is necessary that the direction of radio wave arrival be estimated in the vertical plane, to localize RES position on elevation, e.g., in the higher building. Additionally, it is necessary to equip the vehicle with a concealed digital photography or video camera. An image of the monitored object is required in order to fix the supposed RES location to the specific object region. The necessity of long equipment operation with the vehicle engine switched off requires the presence of a noiseless power supply with increased capacity. Application of a perfect autonomous generator of electric voltage may hardly be expedient, since the generated noise may attract undesired attention. Moreover, comfortable operating conditions for the operators maintaining the system are important requirements for the successful system operation. It is desirable to have an air conditioner, refrigerator, and autonomous heater inside the vehicle. The availability of special software to consider the solved-problem peculiarities and to implement fast and reliable search methods is the governing condition for the successful operation of the mobile system intended for solving the problems of RES detection inside objects.

458

11

Detection and Localization of Technical Channels of Information Leakage

Antenna System Selection Antenna system sizes are limited by the useful area under the radio transparent radome on the vehicle roof, and are usually approximately 3 × 1 m. We have considered the following types of antenna systems: 1. Circular AA with 0.5 m radius, with a single supporting structure situated in the circle center and seven elements located on the circle, at the same space from each other. 2. V-type AA, with nine elements of equal space. 3. V-type AA, with nine elements logarithmically located. Configuration of the circular and V-type antenna systems, with the logarithmic element location, is shown in Fig. 11.21.

0

0

⎯0

⎯0 90

270

180

90

270

Antenna elements

180

Fig. 11.21 Antenna system variants located on the vehicle roof

The investigation results show that, at large signal/noise ratio (SNR), the V-type antenna array with the logarithmic element location demonstrates the better parameters compared to the circular AA. However, at small SNR, irregular errors begin to essentially influence on the direction-finding accuracy, and the system with the V-type AA has worse characteristics at frequencies higher than 300 MHz compared to the circular AA. Relying on the mentioned data, we can conclude that, at signal direction finding, the arrival direction of which does not exceed the limits of the AA opening angle and when the SNR is large, the V-type configuration of the antenna system in the form of a sector with the logarithmic step of the antenna element location is preferable.

Detection of TCIL Sources by the Mobile Station

459

However, if parking the mobile system vehicle by the side of the observed object is difficult or always impossible, then this advantage of the V-type antenna system comes to naught. At the same time, the circular antenna system, at large SNR, yields on accuracy only slightly to the V-type, but excels at small SNR, which is very important for the search of weak signal sources. Moreover, it should be taken into consideration that the circular AA has a number of positive properties, such as smaller sizes, its location comfort inside the local radio transparent radomes, and the possibility of AA application on different carriers. So, in the mobile systems intended for direction finding and RES searching, the application of the circular AA should be considered as more fair.

Methods of RES Detection The most evident methods of RES detection in lengthy objects, with the help of mobile systems located outside of these objects, are: 1. Comparison of the signal levels 2. Direction finding. Comparison of the signal levels consists in the comparison of the amplitudes of the signal spectra obtained in direct closeness to the checked objects and at sufficient distance from it. If the source is located inside the object, then, as a rule, the signal amplitude at distances on the order of several tens of meters from it essentially exceeds the signal amplitude at distances more than several hundreds of meters. At that, the signal levels from the strange emission sources do not practically change. To increase the true detection probability, signal receiving should be executed from the different sides of the object both nearby and at a distance from it. For each system position, with respect to the object, we need to measure and to save the spectral panoramas, which will be further analyzed for RES detection, and whose signal amplitudes, at distance from the object, are mostly decreased. As an example, Fig. 11.16 shows the spectra from a RES with the power near 60 mW and the emission frequency of 300.25 MHz, and which is located inside a brick building. The spectrum in Fig. 11.22a is obtained at a system distance of 50 m from the building, and the spectrum in Fig. 11.22b at a system distance of about 1,000 m.

Fig. 11.22 Signal spectra: (a) near RES, (b) at RES distance of 1,000 m

460

11

Detection and Localization of Technical Channels of Information Leakage

The comparison of spectra confirms that the RES signal amplitude located inside the building decreases by approximately 20 dB, while the signal amplitude at frequency 300.8 MHz, belonging to the remote source, is not practically changed, which coincides with the conclusions in section “Detection of Radio Signals Emitted in Monitored Premise”. The direction-finding method used at search reduces to the problem of determining RES location on the basis of bearings, but is solved conversely (and vice versa). With the usual approach, the source frequency is known in advance, and on the basis of the bearings calculated from the various positions, it is required to determine the RES location. Here, the fact itself of RES presence is unknown, its frequency is also unknown, but the possible location is restricted by the object’s boundaries. Therefore, it is required to detect the fact of RES presence inside the object, to determine its emission frequency, and, after that, to localize its position inside the object. As when comparing signal levels, direction finding with the help of a mobile station should be executed from several positions located at the various sides of the object. For each position, we must keep in mind the angular object location with respect to the mobile system, then to calculate the bearings for all sources, which should also be kept in mind. On the basis of the bearing values obtained from the various positions, the source frequencies, for which the angle-of-arrival coincides with the angular object position for this position, are determined. It results in the formation of a list of suspicious frequencies falling inside the object. Then, more precise determination of the direction finding and the frequency (from the list) listening results is executed. As was already mentioned in Chapter 10, direction finding under urban conditions is a probabilistic problem, due to multipath radio waves propagation and local object influence. The correctness of the obtained bearings will be defined in many respects by the mobile station’s position. It often happens that displacement of the station’s position – even by one to two meters – leads to variation of the bearing value. At position choice, it is necessary that straight visibility of the object is ensured and that there are no high structures or large metal objects near the mobile station. Unfortunately, it is hardly possible to give more detailed recommendations as to how we should select station position under dense building-up conditions, but, in any case, an increase in the number of mobile station positions, as a rule, increases RES detection reliability in the monitored object. The possible RES frequency range spreads from hundreds of kilohertz to tens of thousands of megahertz, and strong loading of the radio frequency range takes place for large towns. For instance, in the range of 25–3,000 MHz, the number of simultaneously-observed RES may exceed several hundreds. Therefore, the operation effectiveness of the mobile complex will be defined by the rate of spectra and bearings calculation and also by the operation speed of the algorithms dealing with the obtained database.

Detection of TCIL Sources by the Mobile Station

461

Equipment Structure of ARTIKUL-M6 Mobile Direction Finder Similar to the ARTIKUL-M1 mobile direction finder, the ARTIKUL-M6 system, intended for operation in the angular sector, is based on the radio receiving equipment for panoramic detection, parameter measurement and radio signal direction finding. In the equipment’s structure, the circular AA and the double-channel digital receiver with the operating frequency range from 25 to 3,000 MHz are included. The antenna array is installed under the radio transparent radome, the form of which repeats the vehicle roof. To expand the frequency range up to 8(18) GHz upward on the frequency scale, inclusion of the additional ARK-KNV3 (3–8 GHz) or ARKKNV4 (3–18 GHz) radio signal converters is anticipated. To detect the radio signal direction, the correlation-interference method is used, giving the possibility of bearing the sources having both the narrow-band and wideband modulation types. The system equipment ensures the fast panoramic analysis of signals with rates of more than 2,000 MHz per second, the multi-channel RES direction finding with the rate of 150–300 MHz/s, the detailed spectral analysis and measurement of the main signal parameters, listening and recording of radio broadcasts. The external view of the ARTIKUL-M6 system, with the antenna system under the radio transparent radome, is shown in Fig. 11.23.

Fig. 11.23 ARTIKUL-M6 mobile complex

462

11

Detection and Localization of Technical Channels of Information Leakage

The double-channel panoramic digital receiver on the basis of the ARGAMAK family modules, and connected to a PC, is installed in the vehicle cabin. Radio receiving equipment and the PC receive their power supply from the secondary electric power source. For a switched-off engine, at the secondary power supply unit is automatically connected: the emergency rechargeable battery. A necessary element of the system is the digital video camera installed on the turning tripod and connected to the PC.

Software Structure and Search Procedure Implementation Two applications of customized mathematical software, namely, the controller program for the panoramic direction-finding system (SMO-PPK) and the program for data saving and RES finding (SMO-SECTOR), are included into the mobile system software. During system operation, the PC executes these applications simultaneously, as two Windows-applications, which change the data between themselves. SMO-PPK is the regular application for the panoramic direction-finding system for radio monitoring. It is used to obtain the initial data, namely, the spectral and bearing panoramas necessary for solution of the RES search problem. The peculiarities of this software package were discussed in detail in section “Conclusion”. SMO-SECTOR is the specific application developed for solution of the problem of RES detection and location determination in lengthy monitored objects. SMOSECTOR software ensures the following: • Importation of the results of multi-channel direction finding from SMO-PPK, and the storing of those results to the database • Representation of the spectra panorama, the frequency list, the bearings and the bearing curves • Images saving of the monitored object, obtained with the help of the digital video camera • Integration of several digital object images into the mutual “panoramic” image • Formation and saving of the operation sessions and frames for the object • Visual setting of the object boundaries on the digital image, for the source search by the direction finder • Using the azimuths and amplitudes of the detected signals, for the search of the “suspicious” frequencies • Generation of “suspicious” frequency list, which may correspond to the emission source inside the object • Translation of the results of single-channel direction finding from SMO-PPK to the object image, with snapping to the angular coordinates • Exporting of the frequency list to SMO-PPK, for its detailed analysis • Support of the database containing both the initial information obtained from SMO-PPK, and the results of the RES search for each of the monitored objects.

Detection of TCIL Sources by the Mobile Station

463

The initial data are the spectral and bearing panoramas imported from the SMOPPK application, and also the images of the monitored objects obtained with the help of the digital video camera. The list of “suspicious” frequencies, which may belong to emission sources located inside the object, is the result of the application operation. The application represents the places of the probable RES location on the object’s color photo picture, and allows the export of the “suspicious” frequency list into SMO-PPK, for its listening and the detailed analysis. In SMO-SECTOR, the database is stored, which saves in full volume the data obtained from SMO-PPK and the results of the source search for each of the monitored objects. The database consists of the general database and the autonomous databases of objects, created for each object in a separate folder. The general database contains the object table and the table of prohibited frequencies. The table of prohibited frequencies contains the frequencies, which can be excluded from the analysis; usually they are the broadcasting, communication or any other frequencies. In the object database, full information for the monitoring sessions executed with it, at different times, is saved. In its turn, each monitoring session consists in frames, as is shown in Fig. 11.24. The frame corresponds to the definite position of the mobile system with respect

Fig. 11.24 Operation session of the mobile section consists in several frames

464

11

Detection and Localization of Technical Channels of Information Leakage

to the monitored object, and contains the following initial data: the spectral and bearing panoramas and also the color object images obtained at this position. The RES search session is executed by the ARTIKUL-M6 mobile direction finder, in several stages: • Aggregation of the initial data frames • Frame processing and generation of the “suspicious” frequency list • Checking the frequencies from the list, and more precise RES location determination.

Aggregation of the Initial Data Frames The task of RES search, at the first stage, is the formation of the initial data bank, which is necessary to make a decision. At the first stage, the mobile direction finder consequently takes several positions at near distance from the monitored object and several remote positions. At each position, a single frame of the initial information is generated. The selection of positions should be provided in such way that they are, if possible, from the other sides of the object. To obtain the spectral and bearing panoramas, the “Multi-channel direction finding” mode of the SMO-PPK application is used. In this mode, the generation of spectra panorama, RES detection and direction finding are fulfilled at each frequency of DRR tuning, simultaneously for all pass-bands of the digital section. Compared to the single-channel direction-finding mode, where the receiver is consequently tuned on the frequency of each detected signal, the above method essentially decreases the time for bearing panorama obtaining. For DRR of the ARGAMAK family, at the multi-channel bearing rate of 150 MHz/s, the period for the frequency range passing from 25 to 3,000 MHz does not exceed 20 s. After the upper task boundary is achieved, the receiver automatically returns to the lower boundary and the multi-channel bearing process will continue. During the aggregation of the spectral and bearing data with the help of the SMO-PPK application, the system operator takes the object photos by the digital video camera, mounted on the turning tripod. The tripod structure provides for the video camera to turn on fixed angles in the horizontal and vertical planes. Therefore, if the angular object sizes exceed the field of video camera vision, the operator has the possibility to form the full object image using several photo pictures. The operator enters data about the direction of the video camera and the positions of photo-picture taking. After that, the object’s image is imported to the database of the SMO-SECTOR program. For the frames formed at large distance from the object, we should enter the sign “D” (distant position) to make possible the source search by means of level comparison. After finishing the frequency range passing, the spectral panorama and the list of detected frequencies from SMO-PPK are imported, in accordance with operator

Detection of TCIL Sources by the Mobile Station

465

command, into the database of the SMO-SECTOR application. The list of detected frequencies includes: the values of the central frequency, the bandwidth, the amplitude and the bearing in the horizontal and vertical planes, and the resulting pattern of the antenna array. The digital object image, the spectral panorama and the list of detected frequencies are represented in the SMO-SECTOR window. The frequency, which is indicated by the marker in the frequency table, distinguishes on the spectral panorama with the help of a vertical marker. In the spectral lens window, the detailed spectrum image is given, and the bearing values in the horizontal and vertical planes are shown on the dial limb, together with the antenna array pattern. The bearing – in the form of intercepted lines – is shown in the digital photo picture (Fig. 11.25).

Fig. 11.25 Object image for the first frame

Frame Processing and Generation of “Suspicious” Frequency List After the formation of not less than 3–5 frames of initial data, the operator – with the help of the mouse – draws the monitored object’s image by color contour onto the color photo pictures of the object represented in the SMO-SECTOR window, as is shown in Fig. 11.25. In that way, the angular boundaries of the object are defined, which is necessary for the search by direction finding. After activation of the “Filtering by frame” option in the application, only those signals in which the horizontal and vertical angles-of-arrivals are situated inside the object boundaries defined by the frames are used for the calculation. After defining the angular boundaries, the operator issues the command for the initiation of the sources detection procedure. As a result of this procedure, a new frame is formed in the object database. This frame contains the integrated

466

11

Detection and Localization of Technical Channels of Information Leakage

frequencies obtained from the frame frequencies of the initial data. Frequency integration is carried out in the following way. At band interception for the several frequencies, a new frequency is generated, the band value of which is considered as equal to the integrated band, and the central frequency value corresponds to the integrated band center. For instance, the frequency E is formed in the integrated frame as a result of frequencies A, B, C, D integration (Fig. 11.26). Fig. 11.26 Formation of integrated frequency

A B C D E

For checking, information on the initial frequency number falling into the integrated frequency and information on the number of initial data frames, in which this frequency has been observed, is included into the resulting table of the integrated frame. The operator may set the filter on the basis of the frame number, in which the frequency was observed, and also to observe all initial frequencies for the integrated frequencies. The superposed frequencies contained in both one and in various frames are integrated into a single frequency containing all range of superposed frequencies. After the frequency integration procedure, the list of “suspicious” frequencies is represented in the integrated frame. This list contains the signals, which are observed in not less than two frames and which direction-of-arrival coincides with the angular location of the monitored object. The frames with a sign “D” (distant position) do not take part in the formation of the integrated frame. After forming the integrated frame from the “nearest” frame (with a sign “N”), the amplitudes of frequencies contained in the integrated frame are compared with the amplitudes of frequencies from the frames with a sign “D”. If the amplitude of the signal for the frame near the object is larger for a given value than the amplitude of the signal observed for the frames with a sign “D”, the frequency in the integrated frame is marked by an exclamation mark. This exclamation mark indicates the high probability of the attribute of the given frequency to the source inside the object, since the positive result for this frequency is obtained at direction finding and at level comparison. After finishing the detection procedure, the list of the “suspicious” frequencies is represented in the integrated frame, with the signals observed in not less than two frames, and which angle-of-arrival coincides with the angular location of the monitored object, and also with the frequencies detected by the level comparison. As an example, Fig. 11.27 shows the SMO-SECTOR window after finishing the RES detection procedure. Detection is executed by six frames of initial data. Data documenting for three frames took place in immediate proximity to the object, under the condition of straight visibility, and the digital object images obtained from these mobile system

Detection of TCIL Sources by the Mobile Station

467

Fig. 11.27 View of the SMO-SECTOR application window after finishing the search procedure

positions are shown in the application window. For another three frames, data documenting was executed from positions outside of the straight visibility, with the distance near 800–1.500 meters. During the processing, the frequency of 300.25 MHz was detected, the signal of which was observed in all six frames of initial data, and the electromagnetic emission direction-of-arrival corresponded to the angular object position. From Fig. 11.27, it follows that the probable place of RES location is the premise of the second floor of the building, which corresponds to the fourth and the fifth window from the left side of the object.

Checking the Frequencies from the List and More Precise RES Detection Under urban conditions, interference is often observed at the receiving and direction finding steps, which can lead to essential errors at the formation of the “suspicious” frequency list. Therefore, each frequency from the obtained list needs detailed checking on the proximity of RES, situated inside the object. To execute this checking, the operator marks the necessary frequencies in the integrated frame table, then exports them to the “Search” window of SMO-PPK by the special command for its listening, single-channel detection and detailed spectral analysis. At single-channel direction finding of signals, the bearing values from the SMOPPK application are translated into the SMO-SECTOR application, where they are

468

11

Detection and Localization of Technical Channels of Information Leakage

transferred on the object image obtained from the video camera. This allows visual estimation of the RES location inside of the monitored object. The single-channel direction finding is executed from several positions of the mobile system, in order to decrease the possible influence of interference. The detailed signal spectral analysis allows, within its structure, the detection of spurious spectral components confirming the insignificant distance to the RES. If the elevation calculation is activated in the SMO-PPK application, the yellow “cross” location will be defined in the horizontal plane by the azimuth, and in the vertical plane by the RES elevation. It should be noted that, in SMO-PPK, the elevation value is calculated with less accuracy than the azimuth. As a rule, acceptable results are obtained if RES frequency exceeds 200 MHz. If it is stated that RES is situated inside the object and its approximate position is detected, then the search is executed directly inside the premise. Since the approximate position and RES frequency are already known, the further search, as a rule, does not offer a serious problem. After formation of the “suspicious” frequency list, we need to execute additional checking on its attribute to the RES, situated inside the object.

Conclusion In this chapter, the problems of RES detection and localization are considered, for the technical channels of information leakage. It is shown that TCIL detection is a complicated process, which may be presented by several stages. The main tasks, which should be solved in this case, are: • Current radio-range checking • Detection of dangerous radio emissions, under complicated radio environment conditions • Identification of detected RES attribute to the radio microphone class • Localization of identified radio microphone position. The important factor defining the success of solving these tasks is the presence of high-rate ARM equipment with highly integrated sensitivity. Methods of solving these tasks by single-channel and double-channel detection equipment are given. A considerable part of this chapter is devoted to the structure peculiarities of distant radio monitoring systems for several building premises, which are only slightly discussed in technical literature. The given results are true for both the situation that is the most typical for the majority of users, when the RRMS installation is executed in an already existing building, and in cases when RRMS creation was developed in advance at the building’s construction, i.e., was coordinated in advance with the building construction, its wire and power communications. The ways of RRMS sub-system development

References

469

and its components parts, namely, the central post, the peripheral equipment, are shown. The main features of the ARK-D13, ARK-D9, and the ARK-D3T systems are discussed, which found their application among various users, in the period from 1999 to 2005. The purpose, structure and technical features of the ARK-D3T, ARK-D9, and ARK-D13 basing on the ARK-D1T, ARK-D1TM single-channel ARM systems and the ARK-D7K and ARK-D11 double-channel ARM systems are described. Functions of the SMO-DX software package in the automated radio monitoring and radio microphone detection systems are considered, and also the peculiarities of its interaction with the equipment of single-channel and double-channel technical means. In the concluding sub-section, the features of TCIL sources detection in lengthy monitored objects, with the help of the ARTIKUL-M6 mobile, direction finder and the SMO-SECTOR software package, are discussed.

References 1. Khorev, A.A., Methods and Means for Search of the Electronic Devices for Information Interception. Moscow, Russian Ministry of Defense Publisher, 1998, 224 pp. 2. Clark, L., and Algaier, W.E., Surveillance Detection, The Art of Prevention. Cradle Press LLC, 2007, 197 pp. 3. Rembovsky, A.M., Detection of Technical Channels of Information Leakage: Methods, Structure and Facility Features (in Russian). Vestnik of MSTU (Bauman Moscow State Technical University). Instrumentation. No 3, 2003, pp. 83–107. 4. Russian Patent 2099870, MKI3 H 04 I 1/46. Radio Microphone with Transmitter Detection Method and the Device for Its Implementation (in Russian). Rembovsky, A.M., and Ashikhmin, A.V., 11 pp. 5. Ashikhmin, A.V., and Rembovsky, A.M., Detection of Technical Channels for Information Leakage: Methods, Structure and Facility Features (in Russian). Special technologies. 2002. Special Edition, pp. 42–48. 6. Russian Patent 2206101, MKI3 H 04 B 1/46. Method for Detection of Electromagnetic Emission Sources in the Limits of Monitored Area and the Device for Its Implementation (in Russian). Ashikhmin, A.V., Bykovnikov, V.V., Vinogradov, A.D., and Rembovsky, A.M., 23 pp. 7. Rembovsky, A.M., Radio Monitoring Complexes and Systems (in Russian). Encyclopedia “Weapon and Technologies of Russia. XXI Century”. Moscow, Oruzhie i Tekhnologii Publisher, Vol. «Information Security», 2003, pp. 103–132.

“This page left intentionally blank.”

Chapter 12

Methods and Equipment for Protection Against Information Leakage Via CEE Channels

General Information Operation of radio and electric equipment inevitably leads to electromagnetic field (EMF) emissions into the surrounding environment. For most equipment, in particular, for computers, scanners, printers, etc., EMF emission appearance is the spurious and undesirable result of its operation. A similar spurious EMF is referred to as a compromising electromagnetic emanation (CEE). Also, technical devices create pick-up on surrounding conductive objects. The existence of CEE and CEE and pick-up (CEEP) makes possible the interception of information, which is processed by the equipment, with the help of sensitive receivers, even for large (tens of meters) distances. Sources of information CEEP can be the units and conductors of the devices, the signals of which are directly related to the processed data. Such devices include: the video amplifier of the monitor; the cable on which the video signal passes from the video adapter to the monitor; connections via which the signals from the keyboard controller are transmitted to the input-output port on the mother-board. The circuits executing the auxiliary functions, for instance, the internal circuits of power supply, and the circuits forming the synchronization signals, produce fields, which do not contain sense information. Moreover, such non-informative CEE in some cases may play a positive part, being the jamming for the interception. Several publications are devoted to the research of CEEP as a technical channel of information leakage (for example, see [1, 2]). At the same time, the devices used for information processing improve along with the permanent improvements of the interception equipment, and as the reference-methodical documents (RMD), which define CEEP checking, are relatively corrected. Therefore, the need to improve the methods and equipment for information protection checking is of current importance [3].

Special Investigation Types and Information Security Index When the possibility of information leakage via a CEEP channel is investigated, the following computing equipment is checked: workstations, servers, data cables of the A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_12,

471

472

12

Protection Against Information Leakage Via CEE Channels

LAN and other technical devices used for confidential information processing. Due to pick-up, the information leakage can be realized through the power circuits and grounding buses, as well as through the conductive line circuits, fire and guard signaling, and other conducting lines and structures, for example, the heating systems, and the water-supply systems having an exit out of the limits of the monitored zone of the information object. We can distinguish three types of special investigations: • Laboratory (bench-top) special investigation of computing equipment oriented to the estimation of the necessary radius of the monitored zone • Tests at information security checking on the informational objects • Investigations executed for the effectiveness estimation of the taken information security measures. During checking, the following features are liable: 1. Spurious informative electromagnetic emissions of computing equipment in the frequency range from 3 kHz to 18 GHz, measured with the help of electrical antennas, and in the frequency range from 3 kHz to 30 MHz, where the measurements are carried out with the help of magnetic antennas. 2. Informative signal, in the frequency range from 3 kHz to 300 MHz, is induced on the supply and grounding circuits of the computing equipment, and also on the output circuits, which exit outside the limits of the object monitored zone. Measurements of the induced voltages are executed with the help of special active and passive probes. 3. Informative signal, in frequency range from 3 kHz to 300 MHz, is induced on the lumped and distributed random antennas located in the objects. The distributed random antennas are the wires, the cables, and the conductive objects, passing near the computing facilities, but which have no galvanic connection with the computing facilities – but do have an exit outside the monitored zone. The lumped random antennas are the auxiliary technical facilities, the lines of which have no galvanic connection with the object computing facilities, but have an exit outside the limits of the monitored zone. In each of the above-mentioned cases, the investigation of information security, which is processed by the computing facilities, is executed in two stages. The purpose of the first stage is to determine the frequencies of CEE informative components. The aim of the second stage is to measure the intensity of the detected information components and calculate the final security indices. The information security index is the ratio of RMS values of informational signal Es and the noise En . This ratio should not exceed the defined-by-RMD maximal permissible ratio δ of the informative signal and the noise, at which it is still impossible to intercept the protected information: = Es /En < δ.

(12.1)

Calculation of Information Security Index

473

In cases when the observed ratio exceeds the permissible, δ in order to ensure the information security, we can use the system of active noisiness (SAN).

Calculation of Information Security Index To search CEE informative components, the separate computing facilities’ units (blocks) are consequently converted in a specially-arranged testing operation mode. In this testing mode, the information signals take the form of a pulse train sequence, which leads to signal power concentration inside the narrow frequency bands and simplifies the detection of CEE spectral components. Almost always, the cause of the appearance of spurious electromagnetic emissions from the equipment that processes the information has a complicated basis, therefore it is more convenient to fulfill the search and analysis of CEEP components in the spectral domain. The signals used in the testing modes are periodic ones, therefore, their power is concentrated in the ensemble of the narrow-band spectral components displaced one from another on the frequency axis by the clock frequency of the testing signal: 3 Fcl = 1 T Hz

(12.2)

where T is the period of the clock pulses, measured in seconds. Let us consider the periodic signal with the clock pulse duration τ seconds and the off-duty factor = T/τ , as shown in Fig. 12.1. The average power of this signal can be calculated in time or frequency domains. The average signal power in the time domain is Pav =

Pp Ens = Ens Fcl = Pp τ Fcl = T

(12.3)

where Ens is the energy of each pulse; Pav is the average power of the clock pulse sequence; Pp is the average pulse power. The average signal power in the frequency domain is:

Stest(t), V S0

τ−Τ

0

t

T

τ+Τ

2T

t, μ s

E j ,V 1 T

Fig. 12.1 Testing signal and its spectrum

0

1/τ

2/τ

F, MHz

474

12

Protection Against Information Leakage Via CEE Channels

Pav =

1 2 Ej , 2

(12.4)

j

where Ej is the amplitude of the signal spectral component at frequency Fj Having equated the right parts of Equations (2.2) and (2.3), we get, for the average pulse power: Pp =

1 2 2 Ej = Ej · 2Fcl τ 2 j

(12.5)

j

As a result, the RMS value of the electromagnetic field strength of the single pulse can be written in the form:

Eav

! " " =#

1 2 Ej · 2Fcl τ

(12.6)

j

The RMS value of the electromagnetic field strength of normalized noise En in the pass-band F = 1/τ is then ! " " En = #

2 (f )df , Enn

(12.7)

F

where Enn (f) is the spectral power density of the normalized noise value, the value of which is defined by RMD. Thus, RMS ratio of the informative signal and the interference is defined by the following equation: ! 2! " Eav " 1 2 " " =# Ej = # En 2Fcl τ j

2 (f )df · Enn

(12.8)

f

This relation is the basis for the calculation of all information security indices. The above-mentioned Equation (12.8) assumes the presence of information on the clock frequency Fcl and the off-duty factor of the testing signal, however, in practice, the exact values of these parameters are unknown and should be more exactly defined during the investigation. As a rule, the determination of these parameters is executed in two stages. First, from a priori information about the testing mode, one calculates the approximate values of the mode parameters, and then, on the basis of the measured intensity of CEE informative components, these values can be more exactly defined. Let us consider the determination of these parameters on an example of LCD and CRT monitors.

Estimation of the Testing Mode Parameters for a CRT Monitor

475

Estimation of the Testing Mode Parameters for a LCD Monitor Let us consider the case when the operator selects the vertical scanning frequency Fver = 60 Hz and the screen resolution is Phor × Pver = 1280 × 1024 pixels. At that, the frequency of the horizontal scanning is equal to Fhor = Pver × Fver = 1024 × 60 ≈ 60 kHz. In “point through point” representation mode, the number of dark vertical lines (dark squares of chessboard) in each image line is equal to m = Phor /2 = 1280/2 = 640 (see Fig. 12.2). As a result, the periodic sequence of the image elements, with the clock frequency, is Fcl = mFhor = 640 × 60 = 38,400 kHz = 38.4 MHz.

a)

b)

Fig. 12.2 Images for the “vertical lines” (a) and “chessboard” (b) active modes

In a LCD monitor, the video signal form corresponds to the generated image and, in the considered case, represents the sequence with the off-duty factor = 2. Thus, τ = 1/F = 1/(Fcl ) = 1/76.8 MHz = 0.013 μ s.

Estimation of the Testing Mode Parameters for a CRT Monitor Let the CRT monitor operate in a mode with the resolution of Phor × Pver = 1024 × 768 pixels and the vertical scanning frequency of Fver = 70 Hz. At that, the horizontal scanning frequency will be equal to Fhor = Pver × Fver = 768 × 70 ≈ 54 kHz. When representing the vertical lines or the chessboard with 1 pixel size of elements, each line will contain m = Phor /2 = 1024/2 = 512 periods of image elements, and the clock frequency of the generated pulse sequence is Fcl = mFhor = 512 × 54 = 27,640 kHz = 27.64 MHz. Determination of the formed pulse duration for CRT monitor signals may be the specific problem, because, to increase the image legibility, the video signal pulses have a duration that is less than the “duration” of the represented pixels. Hence, the off-duty factor of the generated pulse sequence is more than two and should be defined directly on the basis of the results of the spectral investigations, and the pulse duration is less than the one for the LCD monitor: τ < 1/F = 1/(Fcl ) =1/55.28 MHz = 0.0181 μs.

476

12

Protection Against Information Leakage Via CEE Channels

More precise determination of the testing mode parameters for a CRT monitor is executed in two stages. At the first stage, on the basis of the radio measurement results of the periodicity of CEEP component distribution on the frequency axis, the clock frequency Fcl value is defined more precisely. At the second stage, also on the basis of the radio measurement results, the frequency fhar of the nearest harmonic of the clock frequency, which takes zero meaning and the prime number v ∈ {2,3,5,7,11,13} , is determined by means of an enumerative technique, for which the following condition are satisfied: 3Fcl ≤ fhar /v ≤ 4Fcl ·

(12.9)

τ = v/fhar or 1/τ = fhar /v,

(12.10)

= fhar /(vFcl )·

(12.11)

In this case

and the off-duty factor is

If to select v in such manner that the condition (12.9) is impossible to satisfy, then the selection should be repeated, being oriented on the alternative condition 2Fcl ≤ fhar /v ≤ 4Fcl ·

(12.12)

Methods of Detection of CEE Informative Components It should be noted that there is a principal difference between the detection of the spurious electromagnetic emission as a whole and the search of CEE informative components. To solve the first task with acceptable quality, we can fulfill it on the basis of a simple comparison of the average spectra. In order to do so, it is necessary to execute the averaging of observed spectral estimations in the monitored frequency range, firstly at passive operation mode of the testing equipment, and, after that, to repeat the spectral data aggregation, switching the testing equipment into the active (testing) mode. It is difficult to reveal CEE informative components by means of such a procedure, especially if the testing is executed in the insufficiently-shielded (against external interference) premise. The problem is that the data aggregation both in passive and in active modes takes a rather large time interval, during which the “external” radio environment will almost surely change. This generates the problem of distinguishing the variations caused by the mode-changing of the tested equipment and the variations caused by external reasons. The correlation methods successfully used for detection of weak electromagnetic emissions of the complicated form are hardly applied to the search of CEE informative components. This relates to the fact that we should use the intrinsic signals of

Methods of Detection of CEE Informative Components

477

the tested equipment as the standard signals, which are known only approximately, have low intensity, and may essentially change from one piece of equipment to another. The investigation of CEE signal self-descriptiveness is essentially simplified during the testing, due to the possibility to change repeatedly the current operation mode of the testing equipment. At such an approach, the detection of interrelation between the operation mode of the testing equipment and the observed power distribution of the spectral components over the frequencies can be used as a criterion of self-descriptiveness. The specific methods of this interrelation detection depend on the characteristics of the radio monitoring system, which is used for testing. For equipment with low spectral resolution, it is possible to use the double-stage algorithm described below. If the radio monitoring system is able to ensure the high resolution on frequency, the search and self-descriptiveness testing of detected CEE components can be combined. This approach will be considered in detail below. At present, the double-stage algorithm of detection of CEE informative components is the most frequently used. The first stage serves to detect all CEE of the checked equipment, and it is based on the comparison of the aggregated spectra. At first, in the analyzed frequency range, averaging of the observed spectral estimations is executed at the passive operation mode of the equipment under testing. After that, the spectra aggregation is repeated, switching the testing equipment into the testing mode. As a result of comparison of the obtained data, we can extract the frequency ensemble for CEE components, which should be checked for their self-descriptiveness. We should note that to avoid missing a weak CEE component at this stage, we should consider even low (units of decibel) intensity deviations of the spectra samples. At the same time, usage of the similar, low detection, threshold inevitably leads to the essential growth of frequency numbers falling into the list for self-descriptiveness checking. The purpose of the second stage is the self-descriptiveness checking of all suspicious frequencies. The checking can be “orally” executed by the operator, using this or that demodulator set, or automatically at the expense of the alternate narrow-band processing of CEE components that have fallen into the list. The selfdescriptiveness criterion is the detection of the interaction between the operation mode of the testing equipment and the observed power distribution of the spectral components over frequencies, and the mathematical basis of the checking execution we may find, for instance, in [1]. The effectiveness of the second stage depends on the demodulator set used at the narrow-band processing. At that, it is rather difficult to determine, in advance, exactly which demodulator will be effective for the self-descriptiveness determination. To increase the reliability, it is desirable to use all demodulators available, and, during many cycles, to check the presence or the absence of interaction between the operation mode of the testing equipment and the observed power distribution over the frequencies, at the output of each demodulator. The described double-stage procedure is quite efficient and ensures the reliable detection of informative CEE. At the same time, it is distinguished by the low operation rate. Moreover, the necessary time expenses for the processing may be changed,

478

12

Protection Against Information Leakage Via CEE Channels

within very wide limits, so, even at the presence of high-speed equipment, the checking of a specific device may last several days. At the same time, for radio monitoring equipment with low spectral resolution, the same approach is probably the only one possible. In the future, we shall call this technique the TDS algorithm (testing and detection separate). If the radio monitoring system ensures high (up to hundreds of hertz) resolution on frequency, then the search and testing of the self-descriptiveness of the detected CEE components can be executed in a single stage. We shell refer to this algorithm as the TDM algorithm (testing and detection mutual). In order to explain the operation of the TDM algorithm, it is preliminarily necessary to concentrate on the spectral estimation properties used at processing in ARM systems.

Probabilistic Features of Periodogram Samples The initial data available for analysis in radio monitoring systems are the time samples of the observed signals sr (k) where r is the sequence number of the sample, k is the sample number in the sample (0 ≤ k ≤ N − 1, N is the sample volume). Each sample sr (k) can be presented in the frequency domain using DFT (4.2). The random character of the analyzed signals places obstacles in the way of the complex spectra integration (4.2) obtained on the various samples, as they differ by random phase amendments. At the same time, the spectra samples magnitudes are characterized by considerable variance. Therefore, the average periodogram (4.3) is used instead of spectra samples magnitudes, for problems requiring spectral data stability. When spectral data stability is not critical, the non-averaged periodograms may use Xr (n) = |cr (n)|·

(12.13)

For the computing facilities, analysis of the statistical properties of CEE spectra shows that the sample distribution density Xr (n) for CEE informative components can be approximated with acceptable accuracy by the normal law W(x;n) ≈ √

1 exp − 2 (x − an )2 2σn 2π σn 1

(12.14)

where an is a parameter characterizing the intensity of the signal component at n−th frequency, σn is the RMS deviation defining the intensity of the noise component, for the sample. During equipment testing, for informative CEE, parameter an will change when changing from the passive mode to the testing mode. For non-informative CEE this parameter remains constant. For samples, which do not contain the signal component, the approximation (12.14) is inaccurate; however, this is not critical for the informative component search.

TDM Algorithm

479

The normalization of periodogram samples is determined by the following: • The spurious emissions have small intensity, therefore, at their receiving, one tries to locate the receiving antenna of the radio monitoring system maximally close to the equipment under testing. The observed electromagnetic field is the result of the simultaneous action of all units (component parts) of the computing facilities, and has a rather complicated structure. • Maximum permissible resolution on frequency is restricted by the maximal permissible analysis time and by the technical features of the applied equipment. The typical spectral resolution is from hundreds of hertz to several kilohertz per sample, therefore, each periodogram sample (12.13) really reflects the averaged intensity of several narrow-band components closely situated.

TDM Algorithm Let us proceed to aggregate Ry periodograms corresponding to the active (testing) mode, at the expense of consecutive switching of the computing facility unit under test from the active to the passive state and vice versa, together with Rz periodograms corresponding to the passive mode. Because the signal periodicity in the testing modes ensures the CEE signal concentration in the narrow spectral bands [3] and the samples in the periodogram are weakly correlated between each other, in order to detect CEE informative components, we shall analyze the samples of the various frequencies separately from each other. From the values of the n–th sample obtained in the different periodograms for testing the operation mode of the equipment under test, we shall form the vector y = {y1 (n),y2 (n),...,yRy (n)}. The similar values obtained for the passive operation mode of the equipment under test we shall combine into vector z = {z1 (n),z2 (n),...,zRz (n)}. If there is no CEE informative component (hypothesis H0 ) at the frequency of the n−th sample, the samples of vectors y and z should follow the same distribution with an0 and σn0 parameters. If the hypothesis H1 about CEE informative component at the frequency of the n−th sample is true, the parameters any and σn1 of vector y distribution and the parameters anz and σn1 of vector z distribution will differ from each other. Therefore, the likelihood functions have the form: L0 (y,z) =

√

−(Ry +Rz ) 2π σn0 × 4 Ry Rz 0 0 1 2 2 × exp − 2 {yr (n) − an0 } + {zr (n) − ano } ; 2σn0

L1 (y,z) =

r=1

r=1

√

−(Ry +Rz ) 2π σn1 × 4 R Rz 0y 0 1 2 2 × exp − 2 {yr (n) − any } + {zr (n) − anz } ; 2σn1

r=1

(12.15)

r=1

(12.16)

480

12

Protection Against Information Leakage Via CEE Channels

Upon changing the unknown parameters included into these functions by their maximal likely estimations, the likelihood ratio of H1 and H0 hypotheses can be transformed to the form: ∗ Ry +Rz σn0 L1 (y,z) , (12.17) = l(y,z) = ∗ L0 (y,z) σn1 ∗ and σ ∗ are the maximal likely estimates of RMS deviations of distribuwhere σn0 n1 tions. As a result, the optimal (on the maximal likelihood criterion) algorithm of decision making on the self-descriptiveness component, at the frequency of the n–th sample of the observed periodograms, assumes the comparison with the threshold of decision making by statistics Ry 0

Qyz (n) =

r=1 Ry 0 r=1

y2r (n)+

Rz 0

r=1

y2r (n)− R1y

1 z2r (n)− Ry +R z

Ry 0

r=1

2 yr (n)

+

Ry 0

yr (n)+

r=1

Rz 0

r=1

Rz 0

2 zr (n)

r=1

z2r (n)− R1z

Rz 0

2 zr (n)

H0 <

>

H1

(12.18)

r=1

Application of (12.18) statistics allows the simultaneous detection of all CEE informative components observed in some range (the band of the simultaneous spectral analysis), on the basis of the periodogram ensemble with high resolution on frequency. This ensures a gain in the analysis rate for the information security, which is processed by the computing facilities, compared with the usual approach of separately checking the self-descriptiveness of each CEE suspicious component. It is difficult to express analytically the statistics’ (12.18) properties, but the performed experimental research shows that the threshold of decision making, by statistics, can be calculated by the following empirical formula = K1 +

K2 , R − K3

(12.19)

where K1 − K3 coefficients are defined by the required probability of false detection and by the sample number in the periodogram. In particular, at the simultaneous analysis band of 2 MHz and the interval between the samples being 3.125 kHz, the recommended coefficient values are K1 = 1.02; K2 = 13; K3 = 3. If the higher resolution on frequency is applied (the interval between the samples being 390 Hz), then the recommended coefficient values are K1 = 1.02; K2 = 9; K3 = 6. Practical application of the TDM algorithm shows that the reliability of informative component detection depends on the quantity of processed periodograms, on their resolution on frequency and the intensity of CEE detected components. For reliable detection of CEE informative components, the quantity of peridograms aggregated in each monitored frequency band should be not less than twenty. At that, it is desirable to provide a frequency resolution not worse than several hundreds hertz. The features of the TDM method are shown in Figs. 12.3 and 12.4.

TDM Algorithm Fig. 12.3 Detection probability of CEE informative components for the TDM method: 1 – = 0 dB, f = 390 Hz; 2 – = 4 dB, f = 3.125 Hz; 3 – = 3 dB, f = 3.125 Hz

481 Pdet 0.9

1 2

0.7

3 0.5 0.3

0.1 30

0

60

90

120

R

Figure 12.3 shows the probability of CEE information component detection versus the number of processed periodograms. Parameter characterizes the intensity of the CEE detected component and it numerically equals the difference of the spectral sample values (expressed in decibels), which are registered at the presence and at the absence of CEE informative components. On the basis of Fig. 12.4, we can compare the quality of CEE information component detection when using the TDM technique and the typical double-stage approach. The detection thresholds recommended above ensure close to unit detection probability of CEE informative components, in the presence of which the exceeding over the panorama is 5 and more decibels. The weaker CEE components are also detected, but evidently with smaller probability. Due to the additional periodogram aggregation and usage, this probability can be increased; however, it is much more effective to use the periodograms with the increased frequency resolution. When using high resolution, significantly less of the noise power is acting per each periodogram sample, which allows the detection of more weak signals. The price for

Pdet 0.9

Fig. 12.4 Comparison of probability features of the TDM method and the typical TDS technique: 1 – detection probability of CEE informative components for TDM approach (R = 60, f = 390 Hz); 2 – detection probability of CEE informative components on the basis of the typical technique

0.7 0.5 0.3

1 2

0.1 0

2

4

6

8

Δ ,dB

482

12

Protection Against Information Leakage Via CEE Channels

this improvement is the essential increasing of time and hardware resources spent for data processing. As we can see from Fig. 12.4, the detection features ensured by the TDM approach are rather close to the features of the typical technique, exceeding those for the weak CEE components. At the same time, the list of checking frequencies usually contains tens of CEE components, related to the same wide-band periodogram. With the typical technique, these components are tested alternatively, but when using the TDM approach, they can be tested together with each other, which, as a rule, allows a several times decrease in testing time.

Application of ARK-D1TI Measuring Complex An effective piece of equipment for solving the problem of information security estimation is the ARK-D1TI multi-functional portable radio monitoring system, which was certified by the Russian Standard Agency as measuring equipment. This system is a completely Russian development. It has wide measuring and functional features and is intended for solving the problem of radio engineering control. This system allows real-time spectral analysis of radio signals with resolution varying from some tens of kilohertz to tens of hertz, and is capable of detecting and analyzing signals in the operating frequency range from hundreds of hertz to 2 GHz. When using the ARK-KNV4 radio signal converter, which was also certified as measuring equipment, the operating range of the system is expanded up to 18 GHz. The high sensitivity of the receiver, and the large dynamic range for the measured signal levels in the wide-band section (not less than 70 dB on 2nd and 3rd order intermodulation in the pass-band of 2 MHz) allow effective solving of all problems of information security monitoring, which is processed by the computing facilities. The software support of the ARK-D1TI system, in conformity with the problems of information security monitoring, represents a package of customized mathematical software, which consists of the following interacting programs: • SMO-TESTER, for organization of testing operation modes for equipment under test (operates in the automated mode together with the SMO-DX application) • SMO-DX, for execution of the electromagnetic environment analysis including the wire networks, and the detection of spurious electromagnetic emissions from the computing facilities • SMO-PRIZ, for calculation of the information security indices. In 2005, in connection with new RMD, the software support for this system was renewed: the modified version of SMO-DX application for the execution of the electromagnetic environment analysis and the SMO-PRIZ customized software for calculation of the information security were included in the new software version. In accordance with the new RMD requirements, SMO-PRIZ is intended for the security estimation of the technical equipment and the systems of information processing, and has a certificate. It allows the fulfillment of the following calculations:

Application of ARK-D1TI Measuring Complex

483

• Radius of the monitored zone of the computing facilities, necessary for the avoidance of information leakage via CEE channel • Indices of information security, which are processed by the computing facilities, to prevent information leakage via CEE channel and CEE pick-up on the auxiliary technical means and systems • Estimation of effectiveness of the used security measures against the leakage via CEE channel. The application saves the measurement and calculation results in the internal database and allows the formation (in accordance with RMD requirements) of the protocols with the investigation results, saved in HTML and RTF formats. Now we consider the peculiarities of mutual operation of the application included into the package, upon execution of laboratory special investigations of the computing facilities. The purpose of such investigations is to determine the following: • Maximal possible zone sizes for interception of the spurious electromagnetic emissions (zone 2) • Extreme distance to the auxiliary technical facilities and systems and their cable communications having exit outside the monitored zone (zone 1). To fulfill testing, the tested computing facilities are installed on the turning table (Fig. 12.5) in the measuring area, satisfying RMD requirements. The SMOTESTER application is installed in these computing facilities and their COM-ports are connected to the control output of the ARK-D1TI system. Under control of the ARK-D1TI system, alternate testing for the blocks and units of the monitored computer begins. The first testing stage for each block is oriented on detection of Measuring antenna Complex ARK-D1TI

Analyzed radio emission R0 =1m

Controlling PC Programs SMO-DX and SMO-PRIZ

COM-port COM-port

Computing facility under test Program SMOTESTER

Control via 0-modem cable Turning table (views from above)

Fig. 12.5 Determination of the monitored zone radius for the computing facilities

484

12

Protection Against Information Leakage Via CEE Channels

its CEE informative components and is executed with the help of the SMO-DX application.

Search of CEE Informative Components The search of CEE components for a specific block of the computing facilities is executed with the help of SMO-TESTER and SMO-DX applications, in the following way: 1. The computing facilities under test are switched into testing mode of the appropriate block with the help of the SMO-TESTER application. At that, we recommend the use of testing signals, ensuring maximal clock repetition frequency of the informative pulse (pulse trains). Thus, at testing the computer monitors, the testing of information representation should be conducted in the mode of vertical strips representation with widths of 1 pixel (the “point through point” mode) or in the “chessboard” mode with the element size of 1 pixel. 2. The measuring antenna of the system is installed at minimal distance (several tens of centimeters) from the analyzed computing facility (emission source) for the most effective detection of even weak CEE components. 3. On the control PC of the ARK-D1TI system, the SMO-DX is initiated. The operator sets the testing boundaries on frequency, corresponding to appropriate properties of the computing facilities, and activates the “Panorama” mode for the preliminary estimation of the radio environment. When operating in this mode, the computing facilities under test are automatically transferred into the passive state, ensuring the minimal emission level from the block under test (for instance, when testing the monitor, its screen is turned off), and the information aggregation on observed radio emissions (not belonging to the analyzed block) in the frequency band is executed. This allows for the time of CEE informative component search to be essentially decreased in the future. 4. The SMO-DX application controlling the ARK-D1TI system is switched to “Detection” mode, in which the search of CEE components is fulfilled for the monitored block. At that, the computing facilities under test are automatically switched to the active operation mode. This allows the detection of CEE components that exceed the earlier aggregated panorama of radio emissions. All radio emission components exceeding the threshold level set in the task are exposed to self-descriptiveness checking. 5. The extraction of informative components in the obtained CEE component ensemble is executed by the ARK-D1TI system automatically. For this, the dependence of the computing facility under test to switch into active/passive modes, based on the parameters variation of the emitted signal, is estimated. The detection of such dependence indicates that the analyzed signal may be dangerous for information leakage. As a result of self-descriptiveness checking, the detected CEE components can be divided into non-informative and informative radio emissions.

Application of ARK-D1TI Measuring Complex

485

Fig. 12.6 SMO-DX application window in the mode of CEE Informative component detection and the result transmission to SMO-PRIZ application

6. The search of CEE informative components finishes by the intensity measurement of the detected components and by the transmission of the measurement results to the database of the SMO-PRIZ application. The window of SMO-DX, at transmission of the measurement results to SMO-PRIZ, is shown in Fig. 12.6.

Measurement of CEE Informative Component Intensity The procedure of CEE component intensity measurement and its transmission to the SMO-PRIZ application consists in the following actions: 1. The computing facility under test is switched into active (testing) operation mode. For each frequency fj of CEE informative components, the direction of the most intensive emission is determined at the expense of the turning table rotation and the position selection of the measuring antenna. In the detected direction, for the distance d = 1m from the computing facility, measurement of the electromagnetic field strength Emj [dB] levels, emitted by the m−th block of the computer facilities at the j−th frequency, is fulfilled. The linearity of the ARK-D1TI measuring section allows the determination of the spectral component levels in the whole frequency range, with an accuracy of 1–2 dB.

486

12

Protection Against Information Leakage Via CEE Channels

2. The monitored block (or the computing facility as a whole) is switched off and, for each frequency earlier detected, the measurement of strength Enj [dB] levels created by the natural electromagnetic background at these frequencies is fulfilled. 3. The measurement results are automatically transmitted to the database of the SMO-PRIZ application. The operator should make only the more precise determination of the clock frequency Fcl , the pulse duration τm and the offduty factor m of the testing signal of the computing facility’s m−th block, in accordance with the recommendations of section “Methods of Detection of Cee Informative Components”.

Calculation of the Monitored Zone Radius by SMO-PRIZ Application SMO-PRIZ application realizes the finishing stage of the laboratory investigations of the computing facilities, executing the calculation of the monitored zone radius on the basis of the transmitted data. The structural diagram of the algorithm for the radius determination is presented in Fig. 12.7. In this diagram, the following designations are used: • Esmj is the strength (measured in μV/m) of the signal component of the CEE informative component emitted by the m−th testing block of the computing facilities at frequency fj . It can be calculated on the basis of the measured levels of electromagnetic field strength Emmj and Enj , in accordance with the rule Esmj =

100.1Emmj − 100.1Enj ;

(12.20)

• m is the off-duty factor of the testing signal of the th block of the computing facility; • K0j (r) is the coefficient of the electromagnetic field attenuation in the free space; • Knm is the capacity factor of the m–th block of the computing facility (for the parallel codes Knm = n/2, where n is the number of bit circuits of the analyzed block; for the serial codes Knm = 1); • Enn (f ) is the field strength of the normalized noise, corresponding to the current type of radio monitoring equipment and calculated in accordance with RMD; • Fi is the frequency interval to which the CEE testing components belong and the width of which is defined by the pulse duration τm of the testing signal for the computing facility under test Fi = 1/τm ;

(12.21)

• δperm is the maximal permissible (for the current object category) signal/ interference ratio in the point of possible location of the equipment for information interception.

Application of ARK-D1TI Measuring Complex

487

Start

Calculation of value ensemble of informative signal strength Esmj on the basis of equation (12.20 ) for all frequencies f j and all computing facility blocks. Setting of initial calculated radius value R 2 = 0

Yes

R 2 < 10.0

R 2 = R 2 + 1.0

No R 2 = R 2 + 5.0

Calculation of the maximal possible signal / interference ratio 2 ⎧ ⎫ ⎛ Ec m j ⎞ ⎪ Qm ⎪ ⎟ Kn m ⋅ ∫ ( E (f))2df ⎬ ⋅ ∑⎜ ⎨ Δ max(R2 ) = max ⎟ ⎜ nn 2 K (R ) m, i ⎪ j ⎝ oj 2 ⎠ ⎪ ΔFi ⎩ ⎭

Finish

Yes

Δ max (R 2 ) > δ perm

No

Fig. 12.7 Structural diagram of the algorithm for the monitored zone radius R2 calculation when estimating computing facility security against information leakage through CEE channels

In Fig. 12.7, we can see that the radius R2 of the monitored zone is defined, in an iterative way, as the distance at which, for each computing facility’s blocks and for all frequency intervals Fi , the signal/noise ratio does not exceed the maximal permissible by RMD value δperm . Laboratory investigation of the computing facilities includes also the calculation of zone radii r1 and r1 , representing the minimal permissible distances from the computing facility to the auxiliary technical means and systems and its cable connections having exit outside the monitored zone. The radius r1 is calculated with respect to the lumped random antennas, while the radius r1 with respect to the distributed random antennas. The calculations of these variables are executed similarly to the calculation of zone radius R2 , but with the substitution of the field strength of normalized noise Enn (f ) with the sensitivity levels of lumped and distributed random antennas, which are defined by RMD. Since SMO-PRIZ application repeats the recalculation of the security indices directly, as far as the data arrival, the final value of the R2 radius is available just after completing the data entry concerning the blocks ensemble of the analyzed computing facility. In this connection, the operator can check the variation of information security indices directly during the data entry, and the testing protocol is ready for printing just after finishing the information entry.

488

12

Protection Against Information Leakage Via CEE Channels

Thus, the software support of the ARK-D1TI system ensures the full-scale investigation of the computing facility properties, in the automated mode. The calculation results and the protocols formed as a result of the investigations completely correspond to the requirements of RMD enacted at present.

Information Security Monitoring The purpose of both calibration testing and effectiveness estimation of enacted security measures is to determine RMD correspondence of the spurious electromagnetic emission and pick-up levels on the boundary of the monitored zone. Security against information leakage through CEE channels is considered acceptable if all SNR obtained at the analysis of the computing facilities’ blocks do not exceed the limit value corresponding to the category of the testing information object. All investigations of information security on information objects are really executed on the basis of the algorithm, whose structural diagram is presented in Fig. 12.8. The differences consist in the calculation details for the information security indices only. A peculiarity of calibration tests is their execution directly on the information object and the necessity to consider the object’s location features with respect to the potential places of radio monitoring equipment location, as well as its real observed signal attenuation at electromagnetic field propagation to the boundary

Start

1. Calculation of signal component intensity for CEE informative components for all frequencies f j and all blocks of the computing facilities 2. Calculation of propagation and attenuation parameters for signals 3. Selection for the calculation as initial the computing facility block m = 0.

Calculation of maximal observed SNR Δ m for signals of m − th block of the computing facility in the places of possible location of interception equipment and its comparison with the permissible limit δ

Finish

No

Have another blocks?

Yes

m = m+1

Fig. 12.8 Structural diagram of information security estimation algorithm processed by the computing facility

Information Security Monitoring

489

of the monitored zone, etc. The theoretical consideration of such peculiarities is rather complicated and insufficiently accurate [4], therefore, the calculation of security indices during calibration tests is fulfilled by the determination and usage of real signal attenuation factors ⎧ ⎪ ⎨

Edj /ERm j Kpj = Edj K0j (R) ⎪ ⎩ E K (R ) Rm j 0j m

at Rm = R, (12.22)

at Rm < R,

where Edj is the signal field strength corresponding to the frequency fj and formed by the auxiliary generator near the computing facility; ERm j is the field strength measured at distance Rm from the computing facility; R is the distance to the place of possible location of the interception equipment. At estimation of the information’s security against leakage through a CEE channel, calculation of the information security index m is fulfilled in accordance with the equation ⎧! ! ⎪ 2 ⎨" " " m Esmj 2 Knm " m = max # # i ⎪ 2 Kpj ⎩ j

Fi

[Enn (f )]2 df

⎫ ⎪ ⎬ ⎪ ⎭

(12.23)

where Esmj is the signal CEE component strength (measured in μV/m) radiated by the m–th testing block of the computing facility at the frequency fj , and calculated in accordance with (12.20); m is the off-duty factor of the testing signal for the m–th block of the computing facility; Knm is its capacity factor; Enn (f ) is the field strength of the normalized noise, corresponding to the current type of interception equipment; Fi is the frequency interval (12.21), to which the tested CEE components belong. At investigation of information leakage danger due to CEE pick-up, calculation of the signal component intensities of the induced voltages is executed in accordance with the rule Usmj =

/

2 − U2 Uimj nmj

(12.24)

where Uimj is the voltage induced in the testing line at frequency fj , at operation of the m−th block of the computing facility in the active mode;Unmj is the voltage created in the same line by the natural electromagnetic field, at a switched-off computing facility. The rule for calculation of information security index m takes the form: ⎫ ⎧! ! ⎪ ⎪ 2 2 ⎬ ⎨" " " m Usmj " 2 # Knm # [he (f )Enn (f )] df m = max (12.25) ⎪ i ⎪ 2 Klj ⎭ ⎩ j Fi

490

12

Protection Against Information Leakage Via CEE Channels

where Klj is the real signal attenuation factor in the testing line; the effective height he characterizes the properties of the tested line as a random antenna (this function is approximated on the basis of the measurement results); parameters m ,Enn (f ) and Fi remain the same as for the analysis of the CEE channel. In cases when, for improvement of information security at the information object, the system of active noisiness (SAN) is used, the rule (12.25) for calculation of the security index m takes the following form: ⎧! ⎫ ! " 2 ⎨" " Enk 2 ⎬ " m Esmj 2 Kn Knm # , m = max # i ⎩ 2 Kpj Kpnk ⎭ j

(12.26)

k

where Enk is the noise field strength value created by the SAN at the k–th frequency; Kpnk is the real noise attenuation, created by the SAN generator; Kn is the factor of SAN noise quality. Other parameters, including the rule (12.20) for calculation of CEE signal component strength Esmj , completely correspond to the earlier discussed cases. Lastly, the rule for the calculation of information security index m , mentioned in the algorithm in Fig. 12.8, takes the following form at effectiveness estimation in conformity with the pick-up channel: ⎧! ⎫ 2 1 ⎨" ⎬ " m Usmj 2 2 Unlk Kn Knm m = max # , ⎭ i ⎩ 2 Knj j

(12.27)

k

where Unlk is the RMS value of the noise voltage created in the line by the SAN generator at the k–th frequency; the calculation of informative component intensity for the induced signals is fulfilled as per the rule (12.24), and all other designations correspond to the earlier discussed cases.

SMO-PRIZ Application Operation for Information Security Monitoring Figures 12.9 and 12.10 show the SMO-PRIZ application windows for the information security-monitoring mode, in cases when the SAN system is used on the information object and cases when it is not. As earlier, the initial columns of the table presented in the left part of the window contain the results of the intensity measurements for CEE informative components. Since the example in Fig. 12.9 illustrates the pick-up analysis, the information on the intensity of the induced signals is presented now by the measured voltages, and the tested line is characterized by the effective antenna height and by the attenuation factor: unified for all types of radio monitoring equipment. Some table columns represent the calculated (in accordance with Equation (12.24)) level of signal component, the number of frequency interval, to which the

Information Security Monitoring

491

Fig. 12.9 Main window of the SMO-PRIZ application, at information security analysis against leakage through the pick-up channel (SAN system is absent)

Fig. 12.10 Main window of the SMO-PRIZ application, at effectiveness estimation of used information security measures against leakage through the pick-up channel (SAN system is activated)

current CEE component belongs, and, in “SNR” columns, the calculated SNR for these frequency intervals at the boundary of the monitored zone. The SMO-PRIZ operation mode at effectiveness estimation of the used information security measures (see Fig. 12.10) is, in many respects, similar to its operation at information security estimation, and differs only by the additional data account characterizing the operation of the security system. These data are represented in the right part of the main window of the SMO-PRIZ application. The additional tool panel located directly over the data table of the SAN system, and the “SAN” section of the main menu of the application allow the saving of the SAN characteristics on

492

12

Protection Against Information Leakage Via CEE Channels

the hard disk and the earlier saved (or created by the external applications) tables to be involved in the calculation. An example presented in Fig. 12.10 shows that the noisiness system application allows provision of essentially smaller values of the observed SNR (see contents of the “SNR” columns). In that way, the reliable requirement fulfillment of the “Regulations of information security” in all frequency ranges can be realized. Above-mentioned examples illustrate the application of the ARK-D1TI radio monitoring system for solution of the problems of information security estimation, processed by computing facilities. Application of the SMO-TESTER and SMODX applications allows the detection of the CEE informative component ensemble, and the SMO-PRIZ application is capable of calculating all information security indices required by RMD. At that, SMO-PRIZ calculates the information security indices directly during the data acquisition. The operator has complete control over the calculation procedure, and the testing protocol is available for printing, just after the completion of information entry. Moreover, if necessary, the SMOPRIZ application for the information security calculation can be used together with not only the ARK-D1TI system, but with other radio measurement equipment. For automation of the measurement result transmission, the SMO-PRIZ package includes a special module intended for external application data conversion into the format of the SMO-PRIZ database. This makes it possible to use the SMO-PRIZ program for the measurement data processing of other programs different from SMO-DX. At the same time, for the considered software package, the following two shortcomings are typical: • The SMO-DX application uses only the classical double-stage procedure for CEE informative component search (TDS algorithm), which may lead to essential time expenses at analysis execution • The mutual operation of SMO-DX and SMO-PRIZ does not allow, in full, automation of the process of execution and saving of the investigation results. In connection with these mentioned shortcomings, in 2006, a new software package – SMO-THESIS – was developed in Russia, which expands the functions of SMO-DX and SMO-PRIZ applications.

Purposes and Functions of SMO-THESIS Application The SMO-THESIS application for information testing was developed in 2006, and allows the provision of laboratory (bench-top) testing of the computing facilities, as well as attestation of the fact that the information object is processed by the computing facilities. To detect CEE informative components, this application may use both the TDS method and the TDM method. Using the TDM method ensures the essential time reductions necessary for location determination on the frequency axis of all CEE informative components. SMO-THESIS automates also the intensity

Purposes and Functions of SMO-THESIS Application

493

measurement of the noise emission created on the monitored computing facility by the noise generator, at using the system of active noisiness and some other indices anticipated by RMD. SMO-THESIS application is capable of providing most information security investigations in the automated mode. The automated mode (Fig. 12.11) assumes that the SMO-THESIS package takes the testing control on itself. It is enough for the operator in this case to be guided by the requests and program recommendations. A similar approach essentially simplifies the operator actions and ensures a high testing rate. In addition to the automated mode, the SMO-THESIS package offers to the experienced operator the possibility of “manual” rechecking of the obtained data. At any moment, the operator has the possibility to investigate in detail (having stopped the automatic task execution) an interesting section of the frequency range. Having distantly controlled the condition of the equipment under test, the operator can visually monitor the current spectrum variation and also track “orally” the characteristics caused by the testing equipment. The demodulator used for listening can be chosen from a large set of variants, directly during the testing. Lastly, there is the possibility to form the sequence of testing acoustic signals, ensuring the check of acoustic-electrical conversion presence in the monitored equipment.

Fig. 12.11 Investigations in automated mode

494

12

Protection Against Information Leakage Via CEE Channels

At determination of information security indices on the information object, the SMO-THESIS application automates the estimation of the real signal attenuation factor, the signal attenuation factor at propagation pick-up lines, and also the effective heights of random antennas. Using these parameters, for each potential information leakage channel, the value of SNR is calculated at the monitored zone boundary, allowing comparison of the real information security with RMD requirements. Let us consider, in particular, the determination of the signal attenuation factor in the line used for the estimation of information security against leakage via the pickup channel of the information object. To determine the attenuation factor, one can connect two high-frequency voltage probes to the analyzed line, at distance l, each from the other (Fig. 12.12). Into the line at each j–th frequency, the signal from the auxiliary source (the generator of noise signals or the generator of sine signals with the random initial phase) is inserted. Measurement of the induced voltages in the points of connecting the probes allows the determination of the attenuation factor at j–th frequency Klj = Ur1j /Ur2j

(12.28)

where Ur1j is the voltage near the computing facility; Ur2j is the value at the boundary of the monitored zone. Having obtained the data from the equipment, the operator can switch the SMOTHESIS package to the nominal mode intended for the calculation of the information security indices against leakage through the CEE channel. The indices calculation is executed in accordance with the RMD changed in 2005. The SMO-THESIS package allows the calculation of:

Boundary of the monitored zone Computing facility

R

l

r1

B

A

Signal generator

Voltage probe

Complex ARK-D1TI

Fig. 12.12 The circuit for the line attenuation factor measurement

Voltage probe

Purposes and Functions of SMO-THESIS Application

495

• The radius of the monitored zone for the computing facilities, necessary to avoid information leakage through CEE channels • Information security indices, which are processed by the computing facilities, against the leakage through CEE channels on the auxiliary technical facilities and systems • Effectiveness estimation of used information security measures against the leakage through CEE channels. At determination of the monitored zone radius on the basis of the given test ensemble and the selected (by operator) frequency range sets, the SMO-THESIS package provides the following: • Detection of the frequency list of informative CEE • Estimation of detected components’ intensity • Calculation of the monitored zone radii R2 , r1 and r1 , ensuring information security against leakage through the CEE channel and the pick-up channel • Formation of testing protocols corresponding to RMD. In the course of estimating the adopted security measure’s effectiveness, the SMO-THESIS package allows the estimation for the indices of SAN operation and the calculation of the observed (on the monitored zone boundaries) value of the SNR. An example of the SMO-THESIS package’s operation in this mode is shown in Fig. 12.13. In this case, the recorded testing protocol can help to reveal the reasons

Fig. 12.13 Calculation of security indices and saving of investigation results

496

12

Protection Against Information Leakage Via CEE Channels

for the detected discrepancy and to provide the measures for the elimination of these shortcomings of the used system of information active noisiness. The investigation results are saved in the internal database of SMO-THESIS. The information on the various tested objects is saved independently, which allows – without considerable restrictions – the interruption and renewal of testing to be executed for the specific information object and also (if necessary) to use the information obtained during the previous testing.

Conclusion In this chapter, the peculiarities of informative CEE detection in the technical facilities of information processing are considered. Descriptions of the two following algorithms of the informative components are given: the typical double-stage algorithm and the single-stage algorithm of mutual detection and checking of the components with regard to information presence. The single-stage algorithm has the better indices on time expenses, but, for its implementation, ARM equipment with high spectral resolution is necessary. The ARK-D1TI multi-functional radio monitoring system is the modern radio measurement equipment that allows the effective estimation of information security against leakage through CEEP channels. The system software includes a modified version of electromagnetic environment analysis application (SMO-DX) and a special application for the information security calculation (SMO-PRIZ), which permits the execution of the calculation of the monitored zone radius for the computing facilities, the information security indices against leakage through CEEP channels, and the effectiveness estimation of adopted security measures. SMO-PRIZ is certified and fully corresponds to RMD of 2005. To increase the effectiveness of information security investigations, SMOTHESIS was developed, in which the algorithm of mutual detection and property determination of the informative CEE is realized. At investigation of information leakage danger through CEEP channels, the detection of CEE components is provided for the equipment under test, as well as the automatic detection of informative emission components, the measurement of their intensity, and the calculation of information security indices.

References 1. Burmin, V.A., Bykovnikov, V.V., and Tupota, V.I., Multi-Functional Complex for Monitoring of Information Security Effectiveness (in Russian). Special technologies. 2002. Special Edition, pp. 63–75. 2. Kuznetsov, Yu.V., and Baev, A.B., Methods of CEE Measurement: Comparative Analysis (in Russian). Konfident. No. 4–5, 2002, pp. 54–57. 3. Tupota, V.I., Kozmin, V.A., and Tokarev, A.B., Multi-Functional Complex ARK-D1TI Application for Estimation of Information Security Against the Leakage via CEE Channel (in Russian). Special technologies. No 2, 2006, pp. 51–56. 4. Frolov, V.Yu., Pogorelov, A.A., and Petrov, V.V., Calculation of Standard Attenuation Factor of Electromagnetic Field on the Basis of Electromagnetic Field Attenuation Model in Free Space (in Russian). Zaschita informatsii. INSIDE Publisher, No 3, 2005, pp. 82–85.

Conclusion

Respectful reader! The authors of this book hope that, after becoming acquainted with it, you have begun to understand deeply the peculiarities of the problems, which are solved by radio monitoring systems, and that the book has given you a new-found appreciation for the technical features of such systems. The book introduced you to the principles of the digital radio receiver structure, the radio direction-finding of radio emission sources, the localization of the technical channels of information leakage, and the equipment for special investigations of compromising electromagnetic emanation and pickups, as well as the mathematical methods underlying the operation of radio monitoring equipment. There is information in the book related to research, engineering, and applied science, which we feel will be useful for undergraduate and post-graduate students of universities and colleges, who study the fields of radio engineering and electronics, as well as for experts from state and commercial organizations, who deal with radio monitoring and information security. We attempted to describe both the theoretical problems of automated radio monitoring equipment, as well as the specific equipment and the software samples. The authors are active employees of the Russian company, IRCOS, and directly participate in the development and manufacturing of radio monitoring equipment. Due to this, the authors are well versed in its features, and therefore, the descriptions and discussions in this book were based on examples of the equipment manufactured by this and other Russian companies. The long-term successful experience of developing radio monitoring equipment and systems confirms the high potential of Russian experts in this area. The technical solutions used in the equipment and systems are protected by Russian patent, are confirmed by application and independent engineering expertise, are implemented into modern radio engineering systems, and have been honored with various Russian awards. Engineering development occurs very fast and these technical equipment parameters, which today seem state-of-the-art, after several years will be standard for this field. Nevertheless, the principles and structure approaches for the automated radio monitoring equipment described in this book are more conservative, to our A. Rembovsky et al., Radio Monitoring, Lecture Notes in Electrical Engineering 43, C Springer Science+Business Media, LLC 2009 DOI 10.1007/978-0-387-98100-0_BM2,

497

498

Conclusion

mind, than the technological developments that will allow its application in further elaborations to come. Of course, we were not able to address all the current problems of radio monitoring, and various questions will certainly be raised in the mind of the interested reader. Which problems are not included in this book? First, we think that, in a number of cases, we were not able to include enough recommendations regarding the practical applications of the discussed equipment. Unfortunately, we could not include with the book a CD with models of radio signals, radio environment, and equipment. Also, we could not consider in the book a number of new algorithms for signal processing, which increase the effectiveness of radio monitoring problem solutions. We are aware that there are shortcomings in the present book, but it is our hope that they will not affect interest in it from the experts and the profiled institutions. In further editions, these shortcomings can be eliminated with time, and new materials can be added to the book. We think that with the help of experts in this field, from various states and countries, the material in these chapters can be improved, and the authors welcome recommendations and honest evaluations. We thank responding readers for this in advance. The authors would like to express special gratitude to those users of the equipment discussed in the book, who checked its effectiveness both in civil and military fields and whose opinions and suggestions have stimulated the execution of additional investigations and developments and helped in the selection of most of the prospective engineering solutions. The authors anticipate gratefully all future reader remarks and suggestions on the book’s contents, and those comments can be directed to Springer USA or directly to us.

Subject Index

A Accelerometers, 353–355 Acoustic-electrical conversion, 493 Acoustic loud-speakers, 433 Active radio channels, search of, 130 Adcock antenna, 238–239, 270 Adjustable filters, 89 Akvedook P168E, 328 Algorithm “on accumulated spectrum”, 376 cluster, 400 correlation-interference, 384 “on exceeding the reference spectrum”, 376 of radio signal spatial selection, 454 raster, 399 single-channel, 105 testing and detection mutual (TDM), 172, 478–479, 481 testing and detection separate (TDS), 478, 492 “with fixed threshold”, 376 Amplifying-converting path (ACP), 57 Amplitude-frequency response, 27–28 Amplitude modulation, 53, 142, 145, 157–158, 162, 164, 166–167, 203–205, 219–220, 222, 272 Amplitude modulation variety, 142–143 Amplitude-phase, 142, 309 distributions, 311, 313–314 modulation, 142 shift-keying (APSK), 156 signal distribution, 310 Amplitude signal-transfer function, 53 Analog-digital conversion, 57, 72, 139, 242 Analog-digital converter (ADC), 10, 251 Analog-digital processing (ADP), 280 Analog-digital processing unit, 11, 14, 17, 70, 73, 279, 297, 299, 304, 346

Analog-digital processor (FFT processor), 433 Analog modulation, 140–147 amplitude modulation, 140 frequency modulation, 143 phase modulation, 146 Analog switchers (AS), 68 Angular velocity sensors, see Gyroscopes Antenna array (AA), 457 Antenna effect, 12, 239, 297, 304–405, 440, 451 Antenna gain coefficient, 405, 408 Antenna-mast devices, 337 Antenna-receiver unit (ARU), 295 Antenna-switching, 271 Antipodal signals, 152 Aperture sampling, 250 ARCView, 386 ARGAMAK, 67, 80–81, 83–84, 86, 89–90, 93, 125–126, 128, 257, 263, 304, 306–307, 334–336, 341, 383–384, 390, 407, 449–450, 462, 464 “Argument” mobile station, 344, 369, 393, 407, 418, 424 ARK-CO5 DSP unit, structural diagram of, 87 ARK-CT3 digital receiver, 65 ARK-D11, technical specifications of, 127 ARK-D1TI measuring complex, 482–488 CEE informative components, 484 monitored zone radius calculation, 486 ARK-D1TP receiver, 73, 75, 376 ARK-KNV3 radio signal converter, 447 ARK-KNV4 external radio signal frequency converter, 325 ARK-MT1 telescopic dielectric mast, 348 ARK-PR5 DRR technical specifications, 85 ARK-RD8M multi-channel panoramic radio receiver, 128, 341 functional diagram of, 128 technical specifications of, 129

499

500 ARK-SPM spot-jamming generation, 452 ARM station, see “Argument” mobile station Atmospheric duct, 165 Attenuator adjustment data, 80 Attenuator attenuation value, 35 Attenuators, 57, 74, 208 Audio records playback, 131 Audio signal modulation, 201 Autofahrer Rundfunk Data (ARI), 167 Automated operation modes, 132 Automated technical analysis of signals, 219–233 Automatic determination of radio transmitter location, 322 Automatic frequency hopping (FH), 164 Automatic gain control (AGC), 57 Automatic link establishment (ALE), 164 Automatic radio compass, 255, 263–265 direction finder, 266–270 monitoring, mode of, 259 signal-analysis process, 222 signals recognition, 235 Automatic signal analysis, 233–235 Auto request (ARQ) technology, 163 Azimuth calculation, 277 Azimuth estimation, 292 B Band-pass filter (BPF), 67–68, 72, 78, 89, 206–208 Bearing curves, 308–309 Bearing directions, 397 Bearing modes, 304 Bessel function, 120, 144 Bi-Di protocols, 73 Binary phase shift-keying, 221 Boltzmann constant, 36 BPSK signal, see Binary phase shift-keying British TV-Radio Company (BBC), 170 Broadband radio modems, 179 Broadband wireless local loop, 180 Bureau on Radio Communication of ITU, 200 C Calibration coefficient, 406 Calibration curve of measuring antenna, 407 Calibration tests, 488–489 Carrying capacity, 8, 321–322, 330, 332 Cartography, 334, 398 CCIR standard, 166, 205 CEE informative components, 476–477, 486 Cellular radio communication system, 249, 328, 331–333, 390

Subject Index Channel keeping method, 172 Chirp-signal, 146, 437 Citizen’s Band (CB), 158 Clock-frequency synthesizer, 92 CLOVER protocol, 164 Coaxial cable, 40, 440–441, 450 Coded designation of system, 187–188 multiplexing method, 186 transmitted data type, 186 Code division multiple access (CDMA), 176 COFDM signals, 197 Coherently-related local oscillator, 11, 335 Coherent synthesizer, 54 Comb, 14 Combined shift keying, 139 Comb sections, 14 Complex transfer coefficients (CTC), 283 Compromising electromagnetic emanation (CEE), 5–6, 471 Conference European of Post and Telecommunication (CEPT), 136 Continuous dynamic channel selection (CDCS), 178 Control and power supply unit (CPSU), 77, 79 Correlation interference method, 461 Correlation interferometric meter (CIM), 241, 251, 278 CRT monitor testing mode parameters, 475 Crystal receiver, 24 Crystal resonator, 93 Cube law, 48, 52 Cyclic redundance check (CRC), 329 D Dangerous radio emissions, detection of, 468 Data accumulation (processing) time, 113 Data radio channel (DARC), 167 Dead zone, 160 Decay slope, 36 Decoding or demodulating units, 84 Delphi/C++ Builder, 386 Demodulated messages, recording of, 130 Demodulation process, 139 Demodulator/decoder selection, 202 Detection threshold, 107–108, 111, 373–374, 376–377, 384, 481 Difference-distance-measuring (DDM) systems, 320 Diffraction theory, 409 Diffusive dispersion (re-reflection), 165 Digital-analog tracking systems, 55 Digital cartographic information, 404

Subject Index Digital communication protocols, 163 Digital demodulators, 57 Digital enhanced cordless telecommunications (DECT), 178–179 Digital frequency synthesizers, 57 Digital radio mondiale (DRM), 161 Digital radio receiver (DDR), 56–60, 297, 356 ARK-CT1, 67–73 ARK-CT3, 75–79 ARK-KNV4 external remote-controlled converter, 79–81 ARK-PR5 argamak digital radio receiver, 81–93 general principles, 56 types of ARM receivers, 58 Digital signal processing (DSP), 57, 199, 240 Digital spectrum analysis, 63, 128 Digital video broadcasting (DVB-T), 170 Directional antenna, properties of a, 238 Direction finder accuracy, 16, 19, 21, 241, 243, 252–253, 256, 280, 311, 314–315, 320, 397, 458 disadvantages of, 255 mono-pulse, 242, 248, 273, 316, 334, 384 operation accuracy of, 245 sensitivity, 246 Direction-finding (DF) stations, 16 Direction-finding errors, 243, 268, 270, 307–309, 313 correction, 307–315 Direction-finding method, 240, 242–243, 250–252, 256, 267, 309–310, 378, 460 multi-channel, 382–384, 464 Direct sequence spread spectrum (DSSS), 176 Discrete (digital) modulation, 147–158 amplitude-shift keyed signal, 147 Discrete Fourier transformation (DFT), 197, 281 Discretization errors, 350 Discrimination coefficient, 37–38 Dispatcher panel, 208 Distant monitoring of remote premises, 125 Distant radio monitoring, 438–451 ARK-D13 system, 450 ARK-D3T system, 443 ARK-D9 system, 448 construction principles, 438 examples of, 441 DLL/ActiveX, 386 Doppler effect, 161, 240, 270 Doppler shift compensation, 55

501 Double-channel method, 436 Double-channel panoramic digital receiver, 462 Double phase telegraphy, 153 Double relative phase telegraphy (DRPT), 154 Double-sideband AM (DSB), 143 Double-sideband AM with suppressed carrier (DSBSC), 143 Double side-band modulation (DSB), 219 DRM broadcasting formats, 192 DVB-T broadcasting formats, 192 Dynamic AM (DAM), 143 E Electromagnetic accessibility (EMA), 335 Electromagnetic compatibility (EMC), 25, 136 Electromagnetic district relief, 418 Electromagnetic emissions of computing equipment, 472 Electromagnetic environment analysis application (SMO-DX), 496 Electromagnetic field strength measurement, 403–424 calculation of, 408, 414 compatibility calculation, 423 distribution estimation, 406 district topography, 409 location determination, 419 mathematical relations, 404 processing of measurements, 415 transmitters checking, 421 urban build-up, 411 vegetation influence, 412 Electron-beam tube (EBT), 266 Emission parameters, 3, 6, 9, 18, 20, 428 Enhanced data rates for global evolution (EDGE), 176 EPP protocols, 73 Equisignal method, 253–255, 263 Error correction, 164, 185, 315 Euroboard standard, 86, 88 European Broadcast Union (EBU), 166 European Telecommunication Standard Institute (ETSI), 136 F Facsimile, see Image transmission system False-detection probability, 101 False switching, 367 Fast Fourier transform (FFT), 72, 196 Fast task correction during scanning, 131 Field-effect-transistors (FET), 357 Filter fine-tuning, 69 Forward error correction (FEC), 164

502 Frequency converter, 12, 14, 17, 22, 26, 30, 65, 67, 71, 89–90, 125, 128, 334, 341, 348, 447 Frequency correction coefficients, 80 Frequency deflection, 151, 196, 198, 271 Frequency deviation, 205 estimation, 212 Frequency measurement, 195–201 FFT method, 197–198 instantaneous method, 196–197 spectrum width measurement, 199–201 Frequency mixer, 31 Frequency modulation index, 143, 146, 151 Frequency-shift keyed signal, 148 Frequency synthesizer (FS), 54, 280 Frequency-time diagram, 373 Full detection probability, 108 G G4–164 signal generator, 363 Gain coefficient, 24, 37, 405 Gain-transfer characteristic (GTC), 44 Gasoline electric generators (GEG), 369, 371 Gating, 220–221 Gaussian distribution, 101 Gaussian FSK (GFSK) signal, 148 GeminiPD+, 328 General package radio service (GPRS), 176, 331 GeoConstructor, 386 Geodesic line view, 237 Geo-information systems (GIS), 386 Gilbert transform, 197, 290–291 GISToolkit Panorama, 386 Global positioning system (GPS), 168, 352 Global system for mobile communications (GSM), 175 GWX ActiveX control, 386 Gyroscopes, 353–355 H Handheld radio direction finder ARK-RP3, 256–262 ARK-RP4, 262–263 Heterodyne, 26 High-pass filter (HPF), 68 High performance local area network (HiperLAN), 182 Homing method, 20, 395 Horn measuring antenna, 79 Hut model, 165 I IEEE standards, 192 Image transmission system, 162

Subject Index Inertial navigation system (INS), 351 Inertial sensors, 353–355 Information leakage, electromagnetic channels of, 428–430 Information security calculation, 492, 496 Information security index, 471–473, 488–490 calculation of, 473–474 Information security monitoring, 488–492 Information transmission, 1, 3, 220, 435 Input-disturbing signal level, 48 Instantaneous frequency measurement (IFM), 196 IntegraTR or T-96SR radio, 328 Interceptors, 427 Interferometer correlation, 273–281 double-channel, 293 measuring system, 281–289 N–channel correlation, 279 phase, 251, 273–278 principle, 356 single-channel measuring system, 289–295 structural diagram of, 273 Intermediate frequency, 11, 25–26, 31, 33, 55, 57, 78, 80, 89, 198, 221–242, 266–267, 270, 281, 297, 304, 337, 341, 363, 377, 379 Intermediate frequency unit (IFU), 78 Intermodulation characteristics, 46 Intermodulation component (IC), 47, 49, 51 International Consultative Committee on Radio Broadcast, 166 International frequency range distribution (IFRD), 188–191 International Organization of Radio, 166 International Telecommunication Union (ITU), 136 International teleprinter alphabet (ITA2), 163–164 Intrusion protection systems, 189 Inverse operation, 153 Inverse weighting distance method, 417–418 Ionosphere irregularities, 165 K Kaiser-Bessel window, 72 Kotelnikov theorem, 198 L Lacing, 200 LCD monitor testing mode parameters, 475 Likelihood functions, 96, 479 Limiters, 57 Linear interpolation method, 74–75

Subject Index Linear voltage regulators and stabilizers, 360 LO harmonics, 27, 32 Low frequency modulating oscillation, 142 Low-noise amplifier (LNA), 41 M Malfunctions in communication channels, 203 MapInfo, 386 Marker pegs, 243 Mark frequencies, 151 Maximal-likelihood, 99, 398–399, 480 Measurement mode, 221–231 Measuring receiver ARGAMAK-I panoramic receiver, 93 ARK-D1TP digital panoramic receiver, 73–75 Micro-cellular communication systems, 178 Microprocessor-controlled (MPC) unit, 90 Minimal shift-keying (MSK), 220, 151 MobiDARC, 168 Mobile direction finder, 249, 297, 308, 311–313, 315–316, 393, 395, 397, 402, 457, 461, 464 ARTIKUL-M1, 299–301 ARTIKUL-M4, 295–299 ARTIKUL-P, 301–306 ARTIKUL-P11, 306–307 Mobile radio monitoring stations, 289, 364–365, 404, 428 Mobile telegraph, 179 Modern radio electronic signals, 158–188 international system for signal designation, 182–188 SW range signals, 158–164 VHF range signals, 165–182 Modulating signal, 139–143, 146, 167, 199, 204–205, 209, 211, 272, 437 auto-correlation function of, 320 binary, 147–148 coded designation of, 185 filtering of, 150 sinusoidal, 140–141, 211 Modulation parameters, measurement problems of, 318 Modulation theory, 135 Modulation type, determination of, 201, 203, 205 Moore code, 164 Morse code, 182 MPEG-4, 162 Multichannel multipoint distribution systems (MMDS), 169 Multi-channel receivers

503 ARK-D11 double-channel complex, 125–126 ARK-RD8M multi-channel complex, 126–130 panoramic multi-channel receivers, 123–125 Multi-path propagation, 240 Multipath radio waves propagation, 270, 460 Multiple-pass panoramic coverage mode, 60 Multiple signal classification (MUSIC), 241 Multi-position radio monitoring, 390 N Narrow-band components, 95, 98, 101, 479 Narrow-band frequency modulation (NFM), 146, 162, 304 Narrow-band signal threshold algorithm of, 100 double-channel detection of, 117 Navigation software package, 398 Near instantaneous companded audio multiplex, 170 Neiman-Pirson criterion, 99 NICAM, see Near instantaneous companded audio multiplex Noise immunity of receiver, 58 Non-linear law, 52 Non-linear radar, 428 Normal time interval, 175–176 O On-board network of aircraft, 359 One-signal selectivity, 34, 43 Operation concealment, 396, 427 Optimal maximal-likelihood estimate, 101 Orthogonal frequency division multiplexing (OFDM), 161 P PACTOR, 164 Paging communication, see Mobile telegraph Panoramic measurement receiver, 59 Panoramic receiver, 8, 55, 58–59, 93, 257, 281, 283–286, 292, 335, 341, 407 Panoramic spectral analysis, 130–131, 133, 257, 429, 433 Panoramic technical analysis (PTA), 362 ParagonPD+ basic station, 328 Path transfer function, 28, 44, 46 Periodograms, 478–481 Phase crosstalk distortions, 54 Phase deflection amplitude, 146 Phase delays, determination of, 284 Phase detectors, 207–208

504 Phase direction finders, 250 Phase-locked-frame (PLL), 265 Phase-locked-loop methods, 205 Phase shift-keying (PSK), 152 Point-of-multiple-access, 327 Point-point type transmission, 332 Poisson event, 112 Polarization angle, 238, 245 Polarization direction finders, 250 Post-processing signal analysis, 208 Power gain factors, 39 Pre-detection and post-detection amplification, 24 Programmable operating frequency tuning (POFT), 13, 108, 115 signal detection, 115–116, 123 signal direction finding, 382–383 signal-reveal method, 116 signal transmitter, 109, 112 Pseudo-distances, 352 Pseudo-Doppler systems, 240 Pseudo-random sequence, 177 Pulse amplitude, 139 duration, 139, 473, 475, 486 repetition rate, 139 signal polarity, 167 voltage stabilizers, 360 Purely noise, 103 Q Q–factor, 24 Quadrature amplitude modulation (QAM), 143, 157 Quadrature amplitude shift-keying (QASK), 157 Quadrature modulation (QM), 220 Quadrature phase shift-keying with the shift (OQPSK), 220 Quasi-doppler direction finders, 270–272 Quasi-doppler systems, 251, 316 Quasi-optimal methods, 99, 118 Quasi-peak detector, 75 Quasi-stationary method, 396–397 R Radial communication system, 171 Radio amateurs’ zones, 171 Radio amateur standard, 163 Radio communication regulations of Russian Federation, 136 Radio data system (RDS), 167 Radio direction finders, 237, 239–251, 253, 262, 319–320

Subject Index structural diagram and characteristics, 241–242 technical parameters of, 242–250 accuracy of direction finding, 243 being-found signals, 249 cost, 250 deployment time, 249 noise immunity, 247 operating frequency range, 249 operating rate, 248 resolution, 248 sensitivity, 246 Radio direction-finding technique, 238–241 Radio electronic environment (REE), 2 Radio electronic intelligence vs. radio frequency spectrum, 318 Radio electronic means (REM), 1, 5, 9, 136 Radio emission occupied frequency band of, 199 classification, 203 detection, 435 sources (RES) drive method, 395 automatic calculation method, 398 localization methods, 395 peculiarities of multi-channel direction finding, 401 quasi-stationary method, 396 simultaneous direction finding, 402 Radio-Ethernet operating, 327 Radio frequency center, 414 Radio frequency interference, 370 Radio frequency spectrum (RF), 135 administrative division of, 137 Radio-game, 437 Radio microphone detection of, 442 detection problems, 428 identification of, 437–438, 442, 453 localization of, 437–438, 453 polarization type of, 433 Radio-monitoring devices, 11 Radio monitoring equipment characteristics, 16–21 detection systems, 16 manpack ARM equipment, 19–21 portable equipment, 18–19 radio monitoring and RES location stationary and mobile stations, 16–18 classification of, 6–9 design constraints, 8 performance, 8 design philosophy, 9–12

Subject Index technical parameters, 12–15 quality criterion selection, 12–13 Radio monitoring stations, 195, 200, 262, 319, 344, 350, 357, 362, 364, 369–370, 398 Radio monitoring system Archa stationary station, 334–338 Arena portable station, 346–347 Argument mobile station, 338–346 control arrangement in system, 326–333 combined ARK-POM, 332 low-speed radio channel, 328 mobile and deployed posts, 328 radio modems, 331 stationary posts, 326 electric power supply, 356–372 ARK-UPS12, 365 autonomous electric station usage, 369 multi-channel pulse source, 364 pulse power supply of low power, 362 radio equipment sources, 357 requirements for, 356 secondary sources, 360 location determination system, 318–320 mast devices, 347–349 navigation systems for, 349–356 features of, 350 mobile stations navigation, 355 special software support, 372–390 Structure of, 320–326 ARK-POM1, 321 ARK-POM2, 322 ARK-POM3 geographically-distributed system, 322 combined ARK-POM system, 323 Radio-monitoring unit, 59 Radio navigation, 23, 147, 237, 352 Radio network configuration, 172 Radio-operators, 158 Radio-phone communication, 171 Radio receiver parameters, 27–56 amplitude-frequency response, 28 attenuator influence, 51 blockage effect, 52 crosstalk distortions, 53 inherent noise and sensitivity, 36 intercept points, 47, 52 intermodulation noise, 43 intermodulation-free dynamic range, 50 main and spurious channels, 30 multi-signal selectivity, 43 phase noise, 54 pre-amplifiers, 39, 42

505 RR selectivity, 34 voltage standing wave ratio, 29 Radio signal converters, 92, 450, 461 demodulation, 64, 91 detection in monitored premise, 430–437 detection algorithm, 435 detection effectiveness dependence, 436 generalized structure of equipment, 432 near-field and far-field regions, 430 deviation, 167–168 emission, 110 modulation type, 208–218 AM signal, 208 automated radio signal, 219 FFSK signal, 211 FM signal, 211 PSK signal, 216 parameters, automatic measurement of, 379 processing, 343 propagation channel, 140 recording, 125, 449 superposition, 281 transition, 79 Radiotext, 167 Radio transparent radome, 340, 344, 458–459, 461 Radio wave incidence angle, 271 Radio wave propagation in forest-park zones, 412 mathematical model of, 409 Radio wave reflection, 307 Random direction-finding errors, 243–244 Rayleigh distribution, 118–120 Receiving and processing unit (RPU), 257 Red-black trees, 402 Reference-methodical documents (RMD), 471 Reference oscillator (RO), 56 Reference spatial signal (RSS), 278, 284 Refinement method, 278 Reflex receiver, 24 Regenerative and super-regenerative amplifiers, 24 Regional radio broadcast, 160 Remote radio monitoring systems, software for, 451–457 detection algorithms, 454 equipment operation, 455 radio microphone detection, 453 radio microphone localization, 455 SMO-DX application, 452 Repeater (repeater network), 171 RF amplifier (RFA), 23

506 Root-mean-square (RMS) error, 243 Root-mean-square (RMS), 36 RTTY protocol, 163, 233–235 Rubidium oscillator, 198 Russian Arm systems, 60–67 fifth-generation radio receivers, 66 first- and second-generation systems, 60 third and fourth generation, 62 Russian coaxial cables, 440 Russian minivan, 313 Russian Standard Agency, 482 S Satellite radio navigation systems (SRNS), 350 Satellite retransmitter-transponder, 154 Scanning generator, 311–312 Scanning receivers, 58, 60, 428 Scrambling, 170, 202, 437 Secondary power supply (SPS), 90 Selective micro-voltmeter, 28, 58, 93 Sensor array processing, 251 Shepard method, 417 Shift-keying characteristics, determination, 204 Signal amplitude, measurement, 294 Signal attenuation, 35–36, 40, 69, 89, 247, 409, 412, 440, 488–490, 494 Signal demodulation, 90, 206 Signal frequency estimation, 198 Signal generation unit (SGU), 291 Signal inverse shift, 33 Signal-missing probability, 105 Signal-noise ratio (SNR), 13, 37, 89, 458 Signal spectrum, measurement error of, 111 Signal transfer function variation, 52 Single-channel signal detection, 97–105, 108–117 ARM system parameters, 113 characteristics, 105 discrete Fourier transform (DFT), 97 frequency observation time, 109 registered frequencies, 112 separate frequency registration, 111 Single-channel vs. double-channel processing, 119–121 Single side band (SSB), 56, 142–143 Single sideband with suppressed carrier (SSBSC), 143 Sinusoidal carrier oscillation, 142 Sinusoidal modulating oscillation, 143, 146, 204 Sinusoidal signals, 47 Skin-effect, 361

Subject Index Slow frequency hopping (SFH), 176 SMO-KN application package, 373, 380, 387 SMO-MCRM software, 130–133 operation modes, 131 purpose and performance capabilities, 130 Smoothing window, 104 SMO-PPK (PA) software, 304, 372, 379, 407 SMO-PRIZ application, 485–487, 490–492 SMO-SECTOR program, 464 SMO-STA software, 206–208 SMO-TESTER application, 482–484, 492 SMO-THESIS, 492–496 Solar batteries, 9, 369, 372 Space frequencies, 151 Special mathematical software (SMS) program, 11 Spectral and bearing data (SBD), 374 Spectral-bearing data (SPD), 383, 388 Spectral frequency scale, 209 Spectral-power density, 200 Spectrum analyzer (SA), 28, 58, 93, 115, 196, 427–428 modification, character of, 202 transition, 25, 31, 72 Speech transmission method, 171 Spike antenna, 238–239, 241, 270 Spline-interpolation, 313, 407 Spot-jamming, 439–441, 443, 452 SPP protocols, 73 Spurious channel formation, 30 Star topology, 330, 333 Statistical theory of pattern recognition, 204 Stenographing, 131, 133 Sub-carrier Communication Allocation (SCA), 167 Superheterodyne receiver, 23, 25–27, 30, 58, 60, 70, 93 Super-regenerative receiver, 25 Suppression method, 70 Survey mode, 231–233 SXP format digital maps, 403 Synchronization connectors, 88 Synthesis theory of joint optimal algorithms, 96 System of active noisiness (SAN), 473, 490 System-wide car equipment, 343 T TCP/IP, see Transport Control Protocol/Internet Protocol T-DAB broadcasting, 168, 192 Technical channel of information leakage (TCIL)

Subject Index antenna system selection, 458 ARTIKUL-M6 mobile direction finder, 461 detection of sources by mobile station, 457–468 frequencies checking, 467 initial data frames, 464 RES detection, 459 software structure and search procedure, 462 suspicious frequency list, 465 Technical surveillance countermeasures (TSCM), 403, 427 Testing protocol, 487, 492, 495 Thermal-stable micro-power voltage regulators, 357 Thinned out antenna groups, 278 Third register, 163 Threshold displacement, 414 Time division multiple access (TDMA), 175 Time-frequency diagram (TFD), 384 Time multiplexing, 279, 290 TR-965SR radio, 328 Transport Control Protocol/Internet Protocol, 326–327, 331, 337, 372, 380, 453 Trunking (hitcher) systems, 171–172, 192, 374 analog, 173 digital, 173–174 Tuned radio receiver, 23–27 Tuner spectrum components, 32 TV channel frequencies, 190–191 U Ultra high frequency (UHF), 67, 86, 88, 143, 256

507 Universal mobile telecommunications system (UMTS), 178 V Very high frequency (VHF), 67, 89, 165–168, 170 Video signal spectrum, 170 Voice codec, 175 Voltage multiplication operation, 43 Voltage standing wave ratio (VSWR), 27, 29 W Walsh 64-bit sequence, 177 Watson-Watt direction finders, 239, 250 Wave impedance, 30, 239 Weak electromagnetic emissions, detection, 476 Weighting distance, 417 Wide-band signals, 42, 95, 126, 334 Wide frequency modulation (WFM) 146 Window functions (weighting functions), 198 Wireless local area network (WLAN), 180 Wireless metropolitan area network (WAN), 180 Wireless personal area network (WPAN), 182 World Administrative Radio Communication Conference (WARS), 136 World Technical Standard Communication Conference (WTSC), 136 Wullenweber direction finder, 239–240, 242 Z ZOOM FFT, 198

Continued from page ii

Digital Terrestrial Broadcasting Networks Beutler, Roland ISBN 978-0-387-09634-6, Vol. 23 Logic Synthesis for Compositional Microprogram Control Units Barkalov, Alexander, Titarenko, Larysa ISBN: 978-3-540-69283-6, Vol. 22 Sensors: Advancements in Modeling, Design Issues, Fabrication and Practical Applications Mukhopadhyay, Subhas Chandra; Huang, Yueh-Min (Eds.) ISBN: 978-3-540-69030-6 Smart Sensors and Sensing Technology Mukhopadhyay, Subhas Chandra; Sen Gupta, Gourab (Eds.) ISBN: 978-3-540-79589-6 Basic Principles of Fresnel Antenna Arrays Minin, Igor V., Minin, Oleg V. ISBN: 978-3-540-79558-2 Fundamental Numerical Methods for Electrical Engineering Rosloniec, Stanislaw ISBN: 978-3-540-79518-6 RFID Security and Privacy: Concepts, Protocols, and Architectures Henrici, Dirk ISBN: 978-3-540-79075-4 Advances in Mobile and Wireless Communications: Views of the 16th IST Mobile and Wireless Communication Summit Frigyes, István; Bito, Janos; Bakki, Péter (Eds.) ISBN: 978-3-540-79040-2 Informatics in Control Automation and Robotics: Selected Papers from the International Conference on Informatics in Control Automation and Robotics 2006 Andrade Cetto, J.; Ferrier, J.-L.; Pereira, J.M.C.D.; Filipe, J. (Eds.) ISBN: 978-3-540-79141-6 Bandwidth Extension of Speech Signals Iser, Bernd, Minker, Wolfgang, Schmidt, Gerhard ISBN: 978-0-387-68898-5 Proceedings of Light-Activated Tissue Regeneration and Therapy Conference Waynant, Ronald; Tata, Darrell B. (Eds.) ISBN: 978-0-387-71808-8