Digital Satellite Communications Second Edition Tri T. Ha Naval Postgraduate SchoolAbout the Author ‘Tri 7. Ha, Ph.D,, ie Professor of Electrical and Computer Engineering at the Naval Postgraduate School, Monterey, California,Other McGraw-Hill Communications Books of Interest “Azevedo + 1S: THE STRATEGIC DIALOG MANAGER ‘Beam » COMMAND, CONTHOL, AND COMMUNICATIONS SYSTEMS ENGINEERING CChorafas * THe.COMPLETE LAN REFERENCE Cooper (Ranade, Ed.) * COMPUTER & COMMUNICATIONS SECURITY Dayton (Ranade, Ed.) * WTEGRATING DIGITAL SERVICES Pols = aie MechAN? LL COMPILATION OF DATA COMMUNICATIONS STANDARDS Fortier « naNoGOK OF LAN TECHNOLOGY Fihenas + MANUALOF SATELLETE COMMUNICATIONS Inglis + ELeCTRONIC COMMUNICATIONS TANDDOOK Knightion * STANDARDS FOR OFEN SYSTEMS CONNECTION“ Lee * MOBILE CELLULAR TELECOMMUNICATIONS SYSTEMS [Lee + MOBILE COMMUNICATIONS ENGINEERING ‘Nemiow » KEEPING THE LINK: ETERNET INSTALLATION & MANAGEMENT (Owen » DIGHEAL TRANSMISSION SYSTEMS Ranade + AOVANCED S¥A NETWORKING FRanade * WeTHOBUCTION TO SNA NETWORKING Ronade * VSAM: PERFORMANCE, DESIGH, AND FINE-TUNING Remade, Rona * SAM: CONCEPTS, PROGRAMMING, AMD DESIGN hee + ERROR CORRECTION CODING THEORY Sebin + sinct-stocaND SYSTEMS AND CRCUTTS Sarch + 1wtEORATING VOICE ANO DATA Sarch; Abbatiello » THLECOMMUNICATIONS A DATA COMMUNICATIONS FACT OOK Tugel, Tuga + OATA TRANSMISSION Ungaro + NETWORKING SOFTWARE UuntueRstono of cantanera pretiorcea ceseccces=~ TMM oorssaz7 Digital Satellite Communications Second Edition Tri T. Ha Naval Postgraduate Schoo! AIS SRY | bow be R26 McGraw-Hill Publishing Company ew Yoo St Louis Sen Fancieo, Axalord, Bogett ‘Stat anburg_Usbon Larson Madi, Menco Iitan Monees! Now Dah Pass 'Sorapen sa Pose, Sages ‘Syaray Toyo. TornDIGITAL SATELLITE COMMUNICATIONS: INTERNATIONAL EDITION Copyright © 1990 [Exclusive rights by McGraw-Hill Book Co - Singapore for manufacture ant export, This book cannot be re-exported from the country to which it is eonsigne: by McGraw-Hill Copyright © 1990 by McGraw-Hill, Ine. All rights reserved. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, oF stored in a database or retrieval system, without the prior written permission of the publisher. First edi \n published by MacMillan in 1986. 34567890KHL 98765 ‘The sponsoring editor for this book was Danicl A. Gonneau and the production supervisor was Suzanne Babcut, It was set in Times Roman by J.M. Post Graphic Corp. Library of Congress Cataloging-in-Publication Datal Ha, Tri Ti, Date. Digital satellite communications /TriT. Ha -- 2nd ed em. (McGraw-Hill communications series) Includes bibliographical references. ISBN 0-07-025389-7 1. Artifical satelite in telecommunication. 2. Digitat communications. I. Title Series. 89-13462 ap Information contained in this work has becn obtained by McGraw-Hill, Inc, | Hom sures ete to be rll However, neither McGraw-Hill nor its authors guarantees the accuracy or completeness of any information published herein and neither McGraw-Hill nor its authors shall be responsible for any cerrors, omissions, or damages arising out of use of this information. This work i published with the understanding that McGraw-Iill and its authors are Supplying information but are not attempting to render engineering or other | professional services. Ifsuch services are required, the assistance of an [ appropriate professional should be sought When ordering this title use ISBN 0-07-100752.0) Printed in Singapore t Ree EL & For my daughter Renee Lan-HuongContents chapter I MW 12 13 13.1 13.2 133 134 135 13.6 14 1S Preface List of Acronyms List of Symbols Elements of Satellite Communication SATELLITE FREQUENCY BANDS SATELLITE SYSTEMS, TRANSMISSION AND MULTIPLEXING Pulse Code Modulation Delta Modulation Time Division Multiplexing-Putse Code Modulation Digital Hierarchy Frequency Division Multiplexing Transmultiplexing MODULATION MULTIPLE ACCESS xvii xix xxiii “4 15 7 7 a 2220 22 23 24 25 26 27 270 272 273 274 Bu 34d 32 Ra Bld INS 3.16 32 32d 3.2.2 3.23 chapter 2 chapter 3 FREQUENCY REUSE BY ORTHOGONAL POLARIZATIONS ADVENT OF DIGITAL SATELLITE COMMUNICATION REFERENCES PROBLEMS Communications Satellite: Orbit and Description ORBITAL PERIOD AND VELOCITY EFFECTS OF ORBITAL INCLINATION AZIMUTH AND ELEVATION COVERAGE ANGLE AND SLANT RANGE ECLIPSE PLACEMENT OF A SATELLITE INA GEOSTATIONARY ORBIT SATELLITE DESCRIPTION ‘Communications Subsystem Telemetry, Command, and Ranging Subsystom Altitude Control Subsystem Electrical Powor Subsysiem REFERENCES PROBLEMS, Earth Station EARTH STATION ANTENNA Antenna Types Antenna Gain Antenna Pointing Loss Etfective Isotropic Radiated Power Antenna Gain-to-Noise Temperature Ratio G/T Measurement HIGH-POWER AMPLIFIER, Redundancy Configurations Carrier Combining Power Combining 25 28 29 29 31 92 38 a 44 46 48 52 59 62 66 67 68 70 70 n 76 78 at 82 88. 93 94 96 7 33 33.1 3.3.2 34 341 34.2 BS 3.5.1 35.2 36 BT 4d 42 420 422 423 424 425 426 427 43 43d 43.2 433 434 chapter 4 LOW-NOISE AMPLIFIER Redundancy Configurations Nonlinearity UPCONVERTER Conversion Process ‘Transponder Hopping, Polarization Hopping. and Redundancy Configuration for Upconverter DOWNCONVERTER Conversion Process Transponder Hopping, Polarization Hopping, and Redundancy Configuration for Downconverter MONITORING AND CONTROL RELIABILITY REFERENCES APPENDIX 3A; THERMAL NOISE SOURCE PROBLEMS Satellite Link BASIC LINK ANALYSIS INTERFERENCE ANALYSIS Carrier-to-Noise lus Interference Ratio Interference into or from Adjacent Satellite Systems Terrestrial Interference Cross-polarization Interference Adjacent Channel Interference intermodulation Interference Intersymbo! Interference RAIN-INDUCED ATTENUATION Prediction of Attenuation Effect of Rain Attenuation on ‘System Noise Temperature Carrier-to-Noise plus interference Ratio including Rain-Induced Attenuation Path Diversity aa 102 104 105 108 108 109 11 112 114 116 120 124 125 128 129 130 136 198 142 181 182 154 156 187 157 187 165 166 167“44 45 46 46.1 46.2 Sa 52 53 54 55 56 57 61 64d 612 613 62 chapter 5 chapter 6 RAIN-INDUCED CROSS-POLARIZATION INTERFERENCE ‘SYSTEM AVAILABILITY SATELLITE LINK DESIGN Link without Frequency Reuse Link with Frequency Reuse REFERENCES PROBLEMS Frequency Division Multiple Access FDM-FM-FOMA SINGLE CHANNEL PER CARRIER FM-FOMA TELEVISION COMPANDED FON-FM-FOMA AND See aM: FOMA INTERMODULATION PRODUCTS Teer FROM AMPLITUDE NONLINEARITY INTERMODULATION PRODUCTS ASS TING FROM BOTH BEPLITUDE AND PHASE NONLINEARITIES opTIMIZED. GARRIER-TO-INTERMODULATION PLUS NOISE RATIO. REFERENCES APPENDIX 5A: LEAST-SQUARES FITTING "APPENDIX 5B: BESSEL FUNCTIONS PROBLEMS Time Division Multiple Access TDMA FRAME STRUCTURE Reference Burst Traltie Burst Guard Time ‘Toma BURST STRUCTURE 170 W7 178 179 183, 186 186 192 194 202 204 205 206 213 218 219 220 221 223 226 226 227 228 228 229 6.2.2 6.23 624 63 64 65 65.1 6.5.2 68.1 6.8.2 69 69.1 6.9.2 69.3 69.4 6.95 6.10 6.10.1 6.10.2 6.10.3 Unique Wora Signaling Channel Tralfic Data TDMA FRAME EFFICIENCY TOMA SUPERFRAME STRUCTURE FRAME ACQUISITION AND ‘SYNCHRONIZATION Receive Frame Acquisition and Receive Frame Synchronization Transmit Frame and/or Burst Acquisition and Transmit Frame ‘and/or Burst Synchronization SATELLITE POSITION DETERMINATION Single-Station Ranging ‘Three-Station Ranging Satellite Position Error: Single-Station Ranging Satellite Position Error: ‘Three-Station Ranging Errors in Dy BURST TIME PLAN CONTROL AND COORDINATION BY THE REFERENCE STATION Burst Position Control Trallic Coordination: Burst Time Plan Change TOMA TIMING Slip Rate in Digital Terrestrial Network ‘TDMA System and Terrestrial Network Interconnection Plesiochronous Interfaces ‘Asynchronous interfaces Synchronous Interfaces TOMA EQUIPMENT TOMA Processor Terrestrial Intertace Module ‘TDMA Monitoring and Control 238 239 240 281 243 247 249 258 258 262 264 265 265 269 270 270 om 279 282 286 288 290 293 295 295 297 298611 a 72 74 74 74.1 74.2 743 744 75 76 2 78 Bl 82 83 a4 85 86 87 chapter 7 chapter 8 ADVANCED TOMA SATELLITE SYSTEMS REFERENCES PROBLEMS Efficient Techniques: Demand Assignment Mulliple Access and Digital Speech Interpolation THE ERLANG 8 FORMULA TYPES OF DEMAND ASSIGNMENTS DAMA CHARACTERISTICS REAL-TIME FRAME RECONFIGURATION Frame and Burst Structures for DA-TOMA Capacity Search for DA-TOMA Repacking On-Going Calls, How Fast Is Frame Reconfiguration? ! DAMA INTERFACES + ‘SCPC-DAMA SPADE DIGITAL SPEECH INTERPOLATION REFERENCES PROBLEMS Satellite Packet Communications PRELIMINARIES MESSAGE TRANSMISSION BY FOMA\ THE M/G/1 QUEUE MESSAGE TRANSMISSION BY TOMA PURE ALOHA: SATELLITE PACKET SWITCHING SLOTTED ALOHA PACKET RESERVATION TREE ALGORITHM 300 302 303 306 308 ait 318 32t 321 326 327 333 335 337 339 343 348 348 352 353 354 358 362 366 378 379 9.2 921 922 923 93 93.1 9.3.2 9.3.3 94 95 95.1 952 96 9.6.1 9.6.2 97 98 99 9.10 9.10. 9.10.2 9.10.3 9.10.4 9.10.5 chapter 9 yeas convents sit REFERENCES PROBLEMS Digital Modulation OPTIMUM COHERENT DEMODULATION PHASE-SHIFT KEYING Probability of Error Phase Ambiguity Resolution by Unique Word Phase Ambiguity Resolution by Differential Encoding QUATERNARY PHASE-SHIFT KEYING Probabilily of Error Phase Ambiguity Resolution by Unique Word Phase Ambiguity Resolution by Differential Encoding M-ARY PHASE-SHIFT KEYING FREQUENCY-SHIFT KEYING Coherent FSK Nonooherent FSK M-ARY FREQUENCY-SHIFT KEYING ‘Coherent MFSK Noncoherent MFSK ‘QPSK AND MSK BAND-LIMITED NONLINEAR SATELLITE CHANNEL COMPUTER SIMULATION OF A SATELLITE CHANNEL DIGITAL MODULATION WITH ERROR-CORRECTION CODING Concept of Error-Correction Coding Linear Block Coding Error Rate with Linear Block Coding Convolutional Coding Trellis-Coded Modulation REFERENCES APPENDIX 9A: GRAM-SCHMIOT ORTHOGONALIZATION 382 382 384 384 392 399 402 404 407 412 415 419 424 427 427 428 432, 433 435 437 442 449 454 487 462 466 468 473 479 480 AR ARAN ARAN “web od bSLBSUGE EL LEH 8 iv CONTENTS chapter 10 10.1 10.1.1 10.12 10.2 10.2.1 10.2.2 10.2.3 10.2.4 10.2.5 103 103.1 10.3.2 10.3.3 10.3.4 10.3.5 104 chapter I] wd id mt? mAs u2 APPENDIX 98: POWER SPECTRAL DENSITY PROBLEMS Cartier and Symbol Timing Synchronization CARRIER RECOVERY FOR MPSK ‘Analysis Performance in Noise PHASE-LOCKED LOOP Pincple of Operation Steady-State Tracking Performance Transient Response Phase ster Due to Noise Hang-up CARRIER RECOVERY CIRCUIT WITH CARAS WGAND BANDPASS FILTER BAD AUTOMATIC FREQUENCY CONTROL LOOP Single Tuned Bandpass Fiter Double-Tuned Bandpass Fier Cycle Siping Interburst Interference Burstto-Burst Frequency Variations sywBOL TIMING RECOVERY Giacuir REFERENCES PROBLEMS Satellite Spread Spectrum Communications DIRECT SEQUENCE SPREAD SPECTRUM SYSTEMS PN Sequence Error Rate Performance in Uniform Jamming Error Rate Performance in Pulsed Jamming DIRECT SEQUENCE CODE DIVISION MULTIPLE ACCESS 481 484 488 488 490 492 499 500 505 505 506 508 509 512 513 514 520 524 531 536 536 539 541 541 544 847 550 «60 1123 24 N25 3 4 ms 6 Rd 12.2 RB RA 24d RAD Rs BA 2 Ba 134 chapter 12 chapter 13 CONTENTS x ‘Sequence-Asynchronous iar SioDs-CDMA. Random Access DS-CDMA DS-CDMA Link Analysis FREQUENCY HOP SPREAD SPECTRUM SYSTEMS FREQUENCY HOP CODE DIVISION MULTIPLE ACCESS: DS ACQUISITION AND ‘SYNCHRONIZATION FH ACQUISITION AND SYNCHRONIZATION SATELLITE ON-BOARD PROCESSING REFERENCES APPENDIX 11A: CHERNOFF BOUND PROBLEMS Very Small Aperture Terminal Networks VSAT TECHNOLOGIES NETWORK CONFIGURATIONS MULTIACCESS AND NETWORKING ! NETWORK ERROR CONTROL Go-Back-N for SCPC or DS-CDMA Channels Link Contro! for Variable Length Packet Aloha/TOM POLLING VSAT NETWORKS REFERENCES PROBLEMS Mobile Satellite Networks OPERATING ENVIRONMENT MSAT NETWORK CONCEPT CDMA MSAT NETWORK STATISTICS OF MOBILE PROPAGATION REFERENCES PROBLEMS 561 564 566 568 572 873 879 581 581 582 583 586 586 688 594 599 602 605 610 613 614 615 61s 617 625 629 631 631Preface ‘The second edition of Digital Satellite Communications is an expanded version ofthe first edition, Within the past few years the progress of satellite technology has resulted in the creation of very small aperture terminal (VSAT) networks and of the forthcoming mobile satellite (MSAT) networks that can extend ve- hicular communications from the urban to rural areas. Both networks can handle voice as well as data traffic, It is therefore necessary to update the first edition to include VSAT (Chap. 12) and MSAT (Chap. 13) networks. Furthermore, some VSAT and MSAT networks are and will be using spread spectrum tech- nology such as direct-sequence code division multiple access (DS-CDMA). Thus, a more detailed discussion of DS-CDMA is incorporated in the second edition (Chap. 11). Another addition is convolutional coding and trellis-coded modulation (Chap. 9). Also for the instructor a solutions manual is available from the publisher on request. 1 would like to take this opportunity to thank my advisor and academic mentor Professor Robert W. Newcomb (University of Maryland), my long time friends Dr. Ferial El-Mokadem and Dr. Gerd Keiser, my colleagues Professors John P, Powers, R. Clark Robertson, end Robert D. Strum, and my family for their moral suppor. 1 also would like to thank my former colleagues at GTE for providing supporting materials for the first edition and Professors Simon Haykin (MeMaster University) and Leon Couch Il (University of Florida) who reviewed the first edition and Dr. Don Torrieri who reviewed the second edition and provided many helpful comments. I also wish to thank Hien T. Ha for typing the second edition manuscript. Finally I dedicate this book to my daughter Renee Lan-Huong who brings so much joy to my life. Monterey, California June 1989 On SB80095 19° TAAAANAAACK ‘acu ‘AREA ‘ARQ ‘AWGN PF Bsc cor cert cer CDMA. conus esc DA-FDMA DA-TDMA DAMA DAU pos bc persk Dic Mc Desk Ds DS-CDMA bst EIRP QL Fcc, DM FDMA FEC FH FH.CDMA acknowledgment antenna conto! unit assisted recive frame acqui sutomatie request for retransmission auditve white Gaussian noise bandpass filter binary symmetric channel Intemational Radio Consultative Commitee Invemational Telegraph and Telephone Consultative Commitee cartier and clock recovery code division multiple access Continental United States common signaling channel demand assignment-frequency division multiple acess ‘einand assignmest-time division mulipte access demand assignment multiple access data acquisition unit iect broadcasting satellite downconvener ferential encoding PSK with coberent detection data link contro! diserete memoryless channet differential encoding PSK with noncoherent detection direct sequence direct sequence-code division multiple access Aigital speech interpolation effective isotopic radiated power equalizer Federal Communication Commission frequency division multiplexing frequency division multiple access Forward error corection Frequency bop frequency hop-cade division multiple accessXx LIST OF ACRONYMS PM TSK PA 1F nD. INTELSAT iu LAN ENA Lo Lec LPP Mac MESK MPSK MSAT MSK MTTF MTTR NaK Nec om cgesk PC PCM PLL PN Ro Rs: PSK Qusk Rov RF REA RES RET su scr. SCPB-DAMA, scPc SCPC-DAMA frequency modulation Arequeney-shil keying high-power amplifier intermetiate fequency independent and identically distributed International Telecommunications Satelite Organization International Teleeommunicaion Union local area network low-noise amplifier local oscilatoe linear preditive coding low-pass fer ‘monitoring and contro! Meaty frequency sift Keying Mary phaseshft Key snabile satelite minimum shift keying meantime to failure ‘mean time t0 repair nepatve acknowledgment network contol center ‘orthogonal mode transducce offset quaternary phase shit keying personal computer pulse code modulation phase-locked loop peudo-noise Irimary reference burst primary reference station phase shit keying ‘quaternary phase-shif Keying, receive butst timing radio frequency , teceive frame acquisition rective frame synchronization receive Fame timing short burst single channel pee burst single channel per burst-demand assignment multiple access single chanel per eaerier single channel per eatrier-deman assignment multiple access SRB SRS. SS-TDMA ssp ssB. SSB-AM-PDMA 1 ™ 3 Ter ToM TDM ‘TDMA TFA TFS. TPT TIM RT 1S Ww TWA veo vsaT ware Ust oF ACRONYMS x11 secondary fees baat secon retention Sseliteched TOMA toretiane sot fig shad nie sidean-mpliode modlaton-srequeny vision mail 1 Stes caer transmit agiston tate bors transit ut tinng teleoded motion te dvs molesting time dion lil aces tsi fame seiton tans frame syteroiaion tans fee ing testi ince module Ting and reterenc ospondr tenet yehonzaon tceien teeing wave tbe ampli Spooner votageconled xin very sal peste termina Wo Sdmistuive Radio Conference VO VAAAAAAR AAAa List of Symbols a bite radius (42,168,2 km), trafic intensity 4 amplitude, area, azimuth angle by baseband signal ® ‘bandwidth aa) blocking probability a 3-dB bandwidth Bom ms bandwidth (10.95) 2 CDMA spread bandwidth BO, input backofl BO, output backoff « Tight velocity (2.997925 % 10° kr) c carrier power G Speviodie autocorrelation of PN sequence € channel eapacity a Hamming distance, Euclidean distance 4 downlink slant range a free distance the distance feo satelite to station 4, reference distance a uplink slant range > antenna diameter Dy transmit fame delay E clevaton angle B nergy per bitaxiy LIST OF syMooLS BR & Ei) Erfolx) tn leg Jog: Nya No 5 ts) p.) a Pe Pe p energy per coded bit energy per symbol expected value of x complementary emer fonction (9.31) frequency force, noise figure universal constant, rectangular pulse antenna gain, channel talc sssymptotic coding gain impulse response orbital alitude (35786.045 km), transfer function inclination angle imerference power modified Bessel function of order n ofthe first kind imaginary number, summation index, product index jamming power jamming density Bessel function of order of the fst kind Boltzmann constant (1.38 % 107 1K), information block length ina codeword, numberof users in DS-CDMA, number of inputs t0 convolutional encoder degree Kelvin, randomized interval, constrain length, ratio of direct-toxtit fuse power ratual logarithm logarithm of base 10 logarithm of base 2 length, power loss, packet length in bits atmospheric attenuation ‘dovwnlink fee-space atenuation uplink free space attenvaton sass, motibvam frequeney reuse factor Mf = 28 in Mary signaling. shit register Teng length of a code word noise power, processing gain, PN sequence period or length effective noise power total encoder memory noise power spectral density tefetve noise power spectral density transition probability, fraction of shadowing probability density function of x packet correct probability given k interusers power Availablity probability probability of bit ertor detection probability false alarm probability robability of symbol error , tii of in convolutional encoder Pett 1 a) @ oe) ® R 5 BEC pean xguecs ae oe esses pees ust oF syMNOLS xxv probability of x successful probability of a packet slocorrelstion Function of random binary sequence number of quantized levels of a coded DMC Gaussian integral (9.29) distance reliability, resistance, channeteapacity bit rate cant radius (6378.155 kim) chip rate symbol rate signal channel throughput ower spectral density signal-to-noise power ratio. cerrorcorecting capability time noise temperature bit duration ambient temperature (290 K) ‘equivalent noise temperature feame length satelite propagation delay symbol duration, system noise temperature unit step fonction teve anowaly ‘rbital velocity, voltage, amplitude sapertare window rosspaarizaton desimination, x coordinate envelope of a burst, ‘yeooadinate 2 random varinble + coordinate coupling coefficient, voice activity factor coupling coefficient, polarization on a multipath reftcetion interference factor ‘maximum root mean squace of panial eross-corelations paril erss-coreation relletion oeticient duty cycle diferent atenuation iferavial phase rms surface error of antenna, threshold of unique word detection ‘antenna efficiency, interbeam interference factor carrier phase, antenna off-axis angle canh station latitude eth station longitade wavelength, a constant in Chernoff bound, latitude, arival rate mean value, mean hang-up rate 3.1416 Aaaad VAAAHAARTIRAAADigital Satellite Communicationschapter Elements of Satellite Communication ature of communications satellites is their ability to simulta thereby providing distance- ions. This capability applies to nul to mobile terminals on and, in the air, and at sapacity can be dynamically allocated to users sures make satellite communications systems unique in design, ‘This chapter serves as an overview of satellite com: res the render for more elaborate study in the rest of “The unique neously link all users on the earth's st insensitive point-to-multipoint communi munication and pre the book, ‘Arthur C. Clarke, author of many famous books on exploration, wrote in IVireless World in 1945 [1] that a satellite with a circular equato- Fial orbit at a correct altitude of 35,786 km would make one revolution every 24 hz that is, it would rotate at the same angular velocity as the earth. An observer looking at such a geostationary satellite would see it wging ata fixed spot in the sky. Clarke showed that three geostationary lites powered by solar energy could provide worldwide com- munications for all possible types of services. Clarke's vision became a reality 20 years later when the futernational Telecommunications Satel- lite Organization (INTELSAT), established in 1964, launched the Early Bird (INTELSAT 1) in Apeil 1965. Many INTELSAT satellites have been Inunched ‘or are in the planning stages, ranging from instruments with a small eapacity (240 voice circuits oF one television channel) to those with a huge capacity TAR NA AAR ASigure 1 INTELSAT sateltes, 3 Ha=30 kite 1010" ‘Very low frequency (VLF) 30-300 kit Jot-10" Low frequency (LF) 300 KH2-3 Mite 1010" Medium frequency (ME) 3-30 MHz 1ot-10 High frequeney (HF) 30-300 Mite 10-1 Very high fequeney (VHF) 300 MM2=3 GHz 1-10 Urahigh frequency (UHF) 3-30 GHz 10-10 Supethigh fequency (SHP) 30-300 GHz to-%10" Extremely high frequency (EHF) 102-10" GHz 3103 x 10-" Infrared, vse Hiht leavolet (40,000 voice circuits for INTELSAT VI) and covering three regions—the At- lantic, Pacific, and Indian oceans (Fig. 1-1). By 1989 hundreds of geostationary satellites were in service. A summary of satellite locations ean be found in (2). 1.1 SATELLITE FREQUENCY BANDS Communications systems employ the electromagnetic frequency spec- trum shown in Table 1.1, The frequencies used for satellite com- munications are allocated ia superhigh-frequency (SHF) and ¢x- tremely high-frequency (EHF) bands which are broken down into sub- bands as stimmarized in Table 1.2. Spectrum management is an important ‘activity that facilitates the orderly use of the electromagnetic frequency spectrum not only for satellite communications but for other telecom- munications applications as well. This is done under the auspices of the ‘Table 1.2 Satellite frequency spectrum ae 12 a4 8 en 3 mis 127 Ka 17-40 Millimeter 440-300 ngxaee4 prorat sarenerti comaruntenTions International Telecommunication Union (ITU) which is a specialized agency of the United Nations (UN). It predates the UN, having come into existence in 1932 as a result of the merging of the Jaternational Tele ‘graph Union (1865-1932) and the Radio Telegraph Union (1903-1932), ‘There are four permanent organs of the ITU: (1) the General Secretariat, headquartered in Geneva and responsible for executive management and technical cooperation: (2) the International Frequency Registration Board (IF RB), responsible for recording frequencies and orbital positions. ‘and for advising member countries on operation of the maximum practical number of radio channels in portions of the spectrum where harmful inter- ference may occur’ (3) the International Radio Consultative Committee (CCIR, from the initial letters in French), responsible for studying tech hical and operational questions relating {0 radiocommunications which results in reports, recommendations, resolutions, and decisions published ‘as-a group in the Green Books every 4 yr following CIR plenary as- Ssemblies: and (4) the Inernational Telegraph and Telephone Consulta tive Committee (CCITT), responsible for studying technical, operational, ‘and tarriff questions refating to telegraphy and telephony and for adopting reports and recommendations. “The ITU has developed rules and guidelines called radio regulations ‘ata series of international radio conferences held since 1903. ‘The 1979 World Administrative Radio Conference (WARC-79) was the most recent in this long series, The frequency bands allocated by WARC-79 for satellite communications involve 17 service eategories (although some fof them represent special subestegories), as listed in Table 1.3. and three geographic regions: region 1 whieh inchides Europe, Affiea, the USSR. and Mongolia: region 2 which inchules North and South America and Greenland: and region 3 which inches Asia (exeept the USSR and Mongolia), Australia, and the Southwest Pacific. Tables tt and 1.S show the WARC-79 frequency allocations for fixed satellite service (FSS) and broadcasting satellite service (BSS). Satellite services Fiot Meteorol Intersattite Space opeation Mutie ‘Amateur Jain mobile Ralndsterinaion avitime mobile Radionavigation Aeronautical mobile Aeranautialradionivgation roadeasting Maritime radionvigation Tian exploration Standard Fequency ss time sit Space reseech ‘Table 1.4K Frequeney range (GH) 25-2505 2535-20655 2655-200 34-42 4548 5725-585 5385-10025 128-775 19-84 107-117 nea ns23 1n7-1295 ass 1416s 14s 148 3-07 7-181 equeney allocation ELEMENTS OF SATELLITE COMMUNICATION 5 for fixed satelite service Frequency Restietonst range (Gilz)_ Restetons* m2 wid In, 2b. 3a wens tn, 2u% 30" Jn 26,30 25-31 " a B7s40s 0d a as-os 4 Iu, 28,38 472-492 a u 492-5020 u 4 S04-si4 ws & 114 ° 14-755 " aie 92-95 ° 120 3 wor-05 u 9-164 « 27 arm ¥ ws75 ow “1, Region 1:2 eyion 2: 3, region 3: w, uplink earth to space): d downline (space to earth): #, not allocated b, Bidirectional for bro Frequeney 042-079 25-269 earn nena pais W327 127-1275 ns wosaas FAG “Television only reception only: 1. region 3. able 1.5 Frequency allocal ig satellite servi ngs (GH) Tinks, Restriction Ns only hz only 2. e oly 3c only 2 Yenly community 1: 2yregion 2:3, TAA AANA ARMA AN AA6 biGiTAL SATELLITE COMMUNICATIONS 1,2 SATELLITE SYSTEMS [A satelite system consists basically of a satellite in space which Hinks ay earth stations on the ground, as shown schematically in Fig, 1. say sor generates the baseband signal which is routed to the earth s- tn eough the terrestrial network. The terrestrial network can be 9 16- phone switch ora dedicated link tothe earth station At the earth 8 reo teseband signal is processed and transmitted by a modulated Fowie esa (RF) cartier tothe satellite. The satellite can be thought of 0s 6 taoperepeaterin space. ILseceives the modulated RF enriers p's uplink (carth-torspace) frequency spectrum from all the earth stavions Mt the (Rover, amplifies these carriers, and retransmits them back ‘0 earth in neMTowalink (epace-to-eath) frequency spectrum, which is él from the uplink frequency spectrum in order to avotd interference, The remiving earth station processes the modulated RF carrier dow 'e the receiving Signal which i sent through the terrestrial network to Ne Hote a See ecial communications satelites today wilize a SOOM anteotthe the pink and a $00-MEz bandwidth on the downlink The vay. used frequency spectrum is the 6/4-GHz, band, wih Mijn of §.725 10 7.078 GHzand a downlink of ¥4t0 48 O12 The 6/4 crt sand for geostationary satelites is becoming overcrowded Dees’ an used by common carriers for terrestrial microwave inks; satel ‘erestial Terrestrial network User fear User igure 12 A basi steite syster LHLEMENTS OF SATRLLITE COMMUNICATION 7 ites are now being operated in the 14/12-GH2. band using an uplink of 12.75 to 14,8 GHz and a downlink of either 10.7 to 12.3 GHz or 12.5 to 127 Gliz. The 14/12-GHz band will be used extensively in the future na is not yet congested, but one problem exists—rain, which attenuates 14/12-GHz signals much more than it does those at 6/4 GHz. The frequency spectrum in the 30/20-GHz band has also been set aside for commercial satellite communications, with a downlink of 18.1 to 21.2 GHz and an uplink of 27.5 to 31 GHz. Equipment for the 30/20-GHz band is still in the experimental stage and is expensive. “the typical S00-MHz, satelite bandwidth at the 6/4 and 14/12-GHlz bands cean be sepmented into many satellite transponder bandwidths. For example, ight transponders can be provided, each with a nominal bandwidth of $4 MHz sin a center-to-center frequency spacing of 61 MHz. Modem communications satellites also employ Frequency reuse 10 increase the number of transponders in the 500 MHz allocated to them. Frequency reuse can be accomplished through ‘nthogonalpolsrizatons where one transponder operates in one polarization (e+ \ertcal polarization) and a cross-polarized transponder operates inthe orthogonal polarization (e-., horizontal polarization). Isolation of the two polarizations can te maintained at 30 dB or more by staggering the center frequencies ofthe cross- polaried transponders s0 that ony sideband energy ofthe RF carers overlaps, Fe shown in Fig, 1-3. With onhogonal polarizations a satelite can double the umber of transponders in the available 500-MHYz. bandwidth, hence double its Capacity. A review of orthogonal polarizations wil be presented in Sec. 1.6. With this brief discussion of a general satellite system we will now take a look at an earth station that transmits information to and receives Jnformation from a satellite. Figure 1.4 shows the functional elements fof a digital earth station. Digital information in the form of binary digits from the tervestrial network enters the transmit side of the earth station find is then processed (bulfered, multiplexed, formatted, ete.) by the thascband eqitipment so that these forms of information can be sent 10 the appropriate destinations. The presence of noise and the nonideal na- ture of any communication channel introduce errors in the information being sent and thus limit the rate at which it ean be transmitted between the source and the destination. Users generally establish an error rate hove which the received information is not usable. If the received infor- imation does not meet the error rate requirement, error-correction coding performed by the encoder can often be used to reduce the error rate 0 the eceptable level by inserting extra digits into the digital stream from the fomipt of the baseband equipment. These extra digits earry no informa: ‘centuate the uniqueness of each information mes- yys chosen so as to make it unlikely that the channel age to destroy its tion but are used 1 sage. They are alwa disturbance will corrupt enough digits in a mes uniqueness,yp & yo wesTeIp Yo0IgfeuonsUN Ft aan vevanico | srenpousgt—el sapcasa at g moony : YS A “OO soydue 3 samod seuenuoody, soreinpow 49po2u; = “UGH 3 asspucnoen pong 35001 £020 Ret Rena) at ne ny ano nc tA man got sel Gen a RE eT Poth ep to ep en oe on an Gh Oe Ee oe Ga Meo ET we we meso ee an 1D On GL EH RE TH OF HOT Ae AL __A8_AS__Aw__ Aenea an Gh So Se ven Ge Gn en om RO AMANADAD AA a sAAAAA‘igits, where M = 2*and each kigit set or symbol is used to select one of the Af waveforms. For example, in one particular binary modulation Scheme called phaxe-shift keying (PSK), the digit 1 is represented by the snaveform sf) = A COS unt and the digit 0 is represented by the waveform Mtn) = TA 605 ext, where wy isthe intermediate frequency. (In this book ihe letter symbols w and J will be used to denote angular frequency and frequency, respectively, and we will refer to both of them as frequency.") “The modulated 1F cartier from the modulator is fed to the upcon verter, where its intermediate frequency ty is translated to the uplink RE frequency w, in the uplink frequency spectrum of the sitellite. “This modulated RF carrier is then amplified by the high-power amplifier (HPA) to suitable level for transmission to the satelite by the antenna, On the receive side the earth station antenna receives the low-level modulated RF cartier in the downlink frequency spectrum of the satelite ‘A jow-noise amplifier (LNA) is used to amplify this low-level RI carrier fo keep the carrier-to-noise ratio at a level necessary to meet the error rate requirement. The downconverter accepts the amplified RF carrier fom the output of the low-noise amplifier and transtates the downlink frequency «wy to the intermediate frequency ay, The reason for downcon- verting the RF frequency of the received carrier wave to the interme, Trequeney is that itis much ensier to design the demodulator to work on 140 Milz than at a downlink frequency of 4 or 12 GHz. The frodulated IF carrier is fed to the demodulator, where the information is Gatracted. The demodulator estimates which of the possible symbols was transmitted based on observation of the received IF carrier. The probably ity that a symbol will be correctly detected «lepencts on the cirvier oy veige ratio of the moxlulated carrier, the characteristics of the satelite ‘and the detection scheme employed. ‘The decoder performs 2 ccupy the same orbital arc “The FCC ruling poses a major challenge to antenna engineers to design a directional feed for controlling the amount of energy received off= ‘xis by the antenna feed, thus reducing interference from an adjacent sat~ allite, This challenge is especially great because the trend in earth stations fetoward smaller antennas, but smaller antennas have a wider beamwidth find thus look at a wider angle in the sky. The FCC ruling specified that, as of July 1, 1984, all new satellite cearth station antennas had to be manufactured to necommodate the spac fing of 2° and that, as of January 1, 1987, all existing antennas must be modified to conform to the new standards. 1.3 TRANSMISSION AND MULTIPLEXING In the above section we took a look at a simplified satellite com- munications system where digital information (a sequence of symbols it Stead of continuous signals) is carried between terrestrial networks. His~ torieally, analog transmission has dominated satellite communications since its inception. Even today many satellite systems still transmit tele phone and (elevision signals using frequency modulation (FM), and this Trend will continue for some time to come because of the large investment 's. With the advent of digital electronics and com- igital transmission 10 in existing earth station: puters, many earth stations have begun to use prove satellite capacity over analog transmission, ‘These digital earth tions ean interconnect digital terrestrial networks or analog terrestrial networks with appropriate analog-to-digital (A/D) conversion equipment. Ack tage of digital transmission is that it permits integration of jnformition in various forms. Such analog information as speech and12 pica, saTELAaTE COMMUNICATIONS LN slllliiull Figure 1.5 Sampling of an analog sign al signals ean be converted (o digital form and thereby combined with, for transmission, switching, processing, and retrieval 1.3.1 Pulse Code Modulation One commonly used technique for converting an analog signal to form is pulse code nodulation (PCM) which requires three operations: sampling, quantizing, and coding. Sampling converts the continuous Analog signal into a set of periodic pulses, the amplitudes of which repre Sent the instantancous amplitudes of the analog signal at the sampling in tant, as shown ia Fig. 1.5. A question naturally arises: What sampling Tate js required in order to reconstruct the signal completely from these samples? Nyquist [3,4] proved that, ifthe analog signal is band-limited 10 fa bandwidth of B hertz, the signal can be completely reconstructed if the fe is at least the Nyquist rate which is 28. For example, tele Jeech is band-limited to 4 kHz and thus requires 8000 samples continuous amplitude range, the ampli= samplin phone sp per second, Since analog signals have Samples are also continuous in amplitude. When the continuous tude samples are transmitted over a noisy channel, the receiver cannot discern the exact sequences of transmitted values. This effect of noise in the system can be minimized by breaking the sample amplitude into liscrete levels and transmitting these levels using a binary scheme. ‘The process of representing the continuous amplitude of the samples by a ft rite set of levels is called quaniizing. 1f 7 quantized levels are employed to renfesent the amplitude range. it will take log: 1” bits to code cach sample, In telephone transmission 256 quantized levels are employed. hhence each sample is coded using log, 256 = 8 bits. and thus the digital bit fate is ROOK SS = 64,000 bits per second (pst PLEMENT OF SATELLETE COMMUNICATION 13 1.3.2 Delta Modulation It has been found that analog signals such as speech and video signals generally have « considerable amount of redundancy; that is, there is a significant correlation between successive samples when these signals are ‘sampled at a rate slightly higher than the Nyquist rate, For example, the frequency spectrum of the human voice is 300 to 3400 Hz but it is sampled at 8000 samples per second in a PCM system. When these correlated samples are coded as in a PCM system, the resulting digital stream contains redundant information. The redundancy in these analog signals makes it possible to predict a sample value from the preceding sample values and to transmit the difference between the actual sample value and the predicted sample value estimated from past samples. This results in a technique called difference encoding. One- of the simplest forms of difference encoding is delta modulation which provides a stair- case approximation of the sampled version of the analog input signal as shown in Fig. 1.6. The difference between the input and the appro: mation is quantized into two levels, +4 and ~A, corresponding to a posi tive and a negative difference, respectively. Thus at any sampling instant the approximation is increased by A or decreased by A depending on whether it is below or above the analog input signal. A digital output of 1 ‘or 0 can be generated according to whether the difference is +4 or ~A. In delta modulation, overloaiding can occur if the amplitude of the analog input signal changes too fast for the encoding to keep up. Increasing the step size 4 will result in poor resolution, and increasing the sampling rate + will lead to higher digital bit rate. A better scheme for avoiding overloading is to detect the overloading condition and to adjust the step size A toa larger value. This is called adaptive delia modulation. Delta ‘modulation has been used to encode speech with good quality at 32,000 bps. Another important approach in digital encoding of analog signals is i fF ase nie alee road 9 0 00000 0 0 Figure 6 Delia sedation e cog "AMABAAANATbines the simplicity of delta modulation anu ee aes BM. and in many applications it can provide good reproduclton sire oe signals comparable to PCM with a considerable eduction in the igital bit rate. To rar eher new strategies to encode speech at lower Bi ates tan Ihe ove nao (12), One etd is linear predictive coding (LPC) which scenes Jpeect compression by estimating a speech signal as «linear Fonche of past Svat of the speech quantizing system, Nearll-qaliy speech voreee (peech, Siting systems) at 4.8 kbps are now Being developed for mobile satellite ccommanieations (13,14), 1.33 Time Division Multiplexing-Pulse Code Modulation Information trafic between a terrestrial network and an cart tae" aanoemuch more than a PCM channel at 64,000 bps. In order (oo any more channels simultaneously over a single transm/ssi facility sivas a wire pair or coaxial cable, multiplexing mus! Be ‘employed. One atthe most widely used multiplexing techniues for telephone do ch sig: one Mision mutiplesing-pulse code modulation (TDM-PEM) ae eg L.2, Here 24 speech signals are fed 1024 contacts fa fo Staynchronized electronic switehes at che transmit and Feces ends. The dee aeirye amplitude ofthe speech signals is repeatedly sAmPle ns the aortrates, Each ofthe 24 speech signals is sampled every 1 as and ic rped to form atime division-multiplexed signal, Each sample of the ine rr vionemaiplexed signal is quantized and converied 19 5 Bebit Wie nerd, The B-bit PCM codeword forms a time sl Sor ponding toa sample from one ofthe 2 speech signs. TAY four time spontfonm a. 125-us frame which consists of 192 bits and an, additional Hn ofr the end ofthe frame thats used for establishing and TANCE ing frame timing. Normally the receiver checks the 193rd bit every frame ir ar gure that it has not lost synchronization. If synchron'ato has teen lost, the receiver can scan for the framing wisi and be resynchronized. Since there are 193 bts/125 #8, the total bitrate is 1.544 Mbps. PS daion tothe voice signal, the frame also carries signaling infor- mation needed to transmit teleplione dial tones a8 ‘well as on-hook and off Freak signats, Every sixth frame, the least significant bi re cighth bit) of oe veice channel is deleted and a signaling bit s inserted Ps place. ais type of TDM-PCM bitstream is employed in the Bell system's TI carrier which is used in North America [5] Switen Switeh speech signal 1 a ee a |______— 125 us ——— Time slot} Time slot 2 Time slot 24 TH I a pamrom earrrem ——e-siT PCM Framing bit igure 1.7 Time division molilexing-pulse code mosiltion ‘An international standard also exists for PCM transmission. The CCITT has a recommendation for a PCM earier at 2.048 Mbps. In this farrier, there are 32 8-bit time slots in each 125-us frame. Thirty of these ime slots are used for speech at a bit rate of 64 kbps, one for synchro. vidion and one for signaling. The 2.048-Mbps PCM carriers are used outside North America and Japan. 134 tal Hierarchy Yo transmit igtzed analog signals such as tlenhone speech and visual signals having different bit rates, and data with a diversified bt rate over thesame transmission channel, higher-order digital multiplexing ora digi-Mux Figure 1.8 Digi HLEMENTS OF SATELLITE COMMUNICATION IT tal hierarchy must be used. Figure 1.8 illustrates the Bell System di hierarchy which consists of four levels. The respective data signals with bit rates of 1.544 Mbps (Tl), 6.312 Mbps (T2), 44.736 Mpbs (T3), and 274.176 Mbps (T4) correspond to levels 1,2, 3, and 4. Level | is the out- put of a DI channel bank which time division-multiplexes and PCM-en- codes 24 speech signals, or one of the four outputs of a D3 channel bank which multiplexes and encodes 96 speech channels, or the output of a data multiplexer which multiplexes cata with bitrates of 2.4, 4.8, 9.6, and 56 kbps. The DS-I data signal is carried by the TI carrier system over a ire pair. Level 2is formed by multiplexing four DS-1 data signals and is carried by aT? carrier system over a wire pair. Level 3 is formed by m tiplexing seven DS-2 data signals and is carried by a T3 carrier system over coaxial cable. Level 4 is formed by multiplexing six DS-3 data sig- nals and is carried by a T4 carrier system over coaxial cable. 1.3.5 Frequency Division Multiplexing Another form of multiplexing that characterizes analog communications is frequency division multiplexing (FDM), as shown in Fig. 1.9. Twelve speech signals, each of which occupies a bandwidth from 300 to 3400 Hz, fare used to modulate 12 separate carriers each 4 KHz apart. The output of the modulator, which is the product of the speech signal and the carrier, consists of a lower sideband and an upper sideband centered around the cartier frequency. The signals are then passed through 4-kH2 bandpass filters that reject the upper sideband and pass only the lower sideband. “This technique is called single-sideband suppressed carrier (SSBSC) gen- ration, ‘Twelve lower sidebands are then combined to form a group that ‘occupies the frequency band from 60 to 108 kHz. Five groups can be ‘multiplexed in a similar fashion to form a supergroup of 60 speech signals that occupies the band from 312 to 552 kHz, and five supergroups form master group of 300 speech channels that occupies the band from 812 10 2044 kHz, 1.3.6 Transmulliplexing, Despite the explosive progress in digital telecommunications technology, a major portion of terrestrial transmission facilities still uses a frequency division multiplexing hierarchy and will do so in the future because of the large investment in existing systems. For a digital earth station to inter- face with such an analog terrestrial network, some means for conversion between a FDM hierarchy and a digital hierarchy is needed. This ean be ichieved by a FDM-TDM converter, for example between two super ‘groups (120 speech channels) and five TI carriers (120 PCM channels) as shown in Fig. 1.10, It consists of a FDM multiplexer or demultiplexer pR2000 VENANANATIAsrauve4g svomsau vows | os] wf owen q : i 2 | xnwaorcon wos) ues 5 wonnquise | 3 | XAMAGIOM) | snoubsodne ee en ee ornare eae Taienio> WOLWOD20 nicrvaL saresste: commUntestiOns [Sian see woos | chal [sea oe sone a ro-10m [oxcnannats| ,, O90 [aseemops | 9A | nk ran | | llc [sccm] “tame” [ssiners| Sin won zechel [isms Figure 11 FDM-TDM transmulileser and PCM channel banks connected back to back. On the transmit side the FDM supergroups are demultiplexed to 120 individual speech channels which are then time division-multiplexed and PCM-encoded by the chan- rel banks into five separate 1.544-Mbps TI carriers. On the receive side, the five separate 1.544-Mbps TI carriers are PCM-decoded and time division-demultiplexed into 120 individual speech channels which are then frequency division-multiplexed into two supergrouips. A four-wire distribution frame is used for the purpose of individual channel manipula tion such as reordering, adding, deleting, and testing, ‘A special type of FDM-TDM converter called a transnutiplexer can interconvert FDM speech signals and TDM-PCM signals at the mul tiplex level and thus avoid breaking the signals down into individual Sheech channels. as shown in Fig. 1.11 where two supergrouns are con Verted to five T1 carriers at 1.S44 Mbps each, With the use of a distribu on frame, individual channels can be reordered. added. deleted. and tested. The fundamental element of a transmultipleser is shown in Fis. 112, FDM-TDM conversion is accomplished by removing any un wanied out-of-band components from the FDM signal with an analog bandpass filter (312 to 582 kHe for two supergroups). The Bltered signal js passed through an A/D converter to produce a digital stream. The ind Vidual channel in this digital stream is processed by the digital signal processor via a real-time signal-processing algorithm. Transmultiplexer performance can be superior to that achievable with the convention@l EDM.TDM conversion equipment shown in Fig. 1.10, with approxi- mately 2:1 and 5:1 reduction advantages in cost per voice channel and in size, respectively. Dour rom Aealon avpanaoia data signal iter converter ae a Figure 1.12 Fuodamental elements ofa transnltinler ELEMENTS OF SATELLITE COMMUNICATION 21 1.4 MODULATION ‘As mentioned in See. 1.2, modulation must be employed 0 transmit baseband information over a bandpass channel. In analog modulation such as frequency modulation, which is extremely popular in satellite communications, the signal-to-noise ratio at the output of the FM demodulator is an intuitive measure of how Well the FM demodulator can recover the analog information signal from the received modulated carrier in the presence of additive white Gaussian noise (AWGN). The ontput signal-to-noise ratio is defined as the ratio of the average power of the analog information signal to the average power of the noise atthe output of the demodulator. In digital modulation, the performance of the demodulator is measured in terms of the average probability of bit error, or the bit error rate as it is often called, The binary information, which consists of sequences of | and 0 digits, can be used to modulate the phase, Frequency, or amplitude of a carrier. Consider the carrier A cos (ax! + @), where dis the carrier amplitude, ais the carrier frequency, and ¢ is the cartier phase. To transmit the binary digit orbit 1, is set to Orad, and to transmit the bit 0, @ is set to a radians. Thus | is represented by the waveform Acosee!, and 0 is represented by the waveform A costuct + 2) =—A cos at. This type of discrete phase modulation is called phase-shift keying (PSK). Similarly, 1 can be transmitted by using the waveform A cos wit and 0 transmitted by using the waveform A cos taut. where «, # ws, This type of digital modulation is ealled frequenc shit keying (FSK), where two waveforms at different carrier frequencies toy and w» are used to convey the binary information, The problem with digital modulation is that sometimes the binary digit | is transmitted but the demodulator decodes it as 0, or vice versa, because of perturbation of the carrier by noise; this results in bit errors in the demodulation ofthe binary information, The average probability ofbit error P, is a convenient measure of the performance of the demodulator and is a function of the ratio ofthe energy. per bit to the noise density, Ey/Ns, where the energy per bit Ey is the energy of the carrier during a signaling interval or bit du- fation Ty and Nol2 is the noise power spectral density. When the baseband information is transmitted ata rate of R bits per second, the bit duration is simply Ty = IIR seconds, and this is also the signaling interval of the waveform that represents a particular bit. For example, in PSK ‘modulation, =A cosmt = OSTST SQ) =A cosa 0 S0STs where si(t) represents 1 and s(t) represents 0. By definition we have PSECOIDIGE FAPODOSTIATMote that 4 = cture mies Note thats gaageearier power C, andthe noise power Nv measored teat cewer noise bandwidth B. By definition, the average carrier power is cng [Petrone (2) aveform during the signaling interval Ty and where s(t) is the carrier w farrier waveforms have identical EL] is the expected value. If all the © energy E, during any signaling interval, then Es 2 a Recall that the power spectral density of noise is Nol2 and that the nese eeeavidth is B. Hence the noise power measured within the noise Dandwidth for both positive and negative frequencies is N=NB aa) c -Therefore itis seen thatthe ratio ofthe energy per bit wo the noise de can be expressed as bch -7,0($) Roe OCR a9 where CIN is the average carrero-noise ratio. In satelite com were CIN itis the quantity CIN that is directly evaluated, as we wit ae Chap. 4. Once the CIN is knovn and the bandwidth of he aesriver ig selected, EylN can be calculated, as well as the average proba: bility of bit error Py which is a function of Ex/Ne 1.5 MULTIPLE ACCESS One advantage of communications satellites over other transmiciot ois is their ability to link all earth stations together, thereby providing pointto-multipoint communications, A satelite transponder be ace aoined by many earth stations and therefore it necessary 10 NOYe tech- tion control subsystem, and an apogee kick motor 2.7.1 Communications Subsystem ions subsystem provides the receive and transmit cover ‘The communicat ip communications antenna and a com age for the satellite. It consists of ae cations repeater. The communications antenna serves as an interface otween the earth stations on the ground and various satelite subsystems tring operation. ‘The main function of the antenna is to provide shane ‘Towatink and uplink beams for transinission and reception of cory peetions signals inthe operating frequency bands (eg. the C or Ky rainy ta addition the antenna may be wsed 10 provide signal link fr the aniice telemetry, command, and ranging subsystem which in coniunt ah ie atitade control subsystem provides beacon tracking signals for precise pointing of the antenna toward the earth coverage ‘rea tore tie ie. 2 LT. To increase the communications capacity of the sate The, Yrequency reuse is often employed through the use of orthogonally polarized beams; hence the communication antenna requires Wwe reflec mn, that can be incorporated into a single phys- eeTsrueture such as that shown in Fig. 2.18, The single structure con” weats of two polarization-selective gridded offset parabolic reflectors se ed one behind the other to facilitate aperture sharing, With this Be, setry, the two reflectors are fed by two independent multihorn feed Grays aehich do not interfere with each other either physically oF eis araiy.‘The polarization selectivity of each reflector is accomplished bY ae vo the desired polarization and bonded 10 tors, one for each potariz a vein etrips pars COMMUNICATIONS SATELLITE: ORNIT AND DESCRIPTION SS Altitude contrat asain ‘Beacon tack signal Uptink = communications signals Communications Communications Downlink repeater antenna oe communiestion signals |_ command-ranging signal Telemetry, command and ranging subsystem Tolemety-ranging signals Figure 217 Fundamentals of « eommusicatons satelite subsystem. ‘a paraboloid shell. The polarization orientation is normally chosen such that the vertical polarization plane is defined as the plane formed by the satelite north-south axis and the center ofthe earth, and the horizontal polarization is orthogonal to the vertical. The isolation between the copolarized and cross-polarized signals incident on the antenna is nor- mally about 33 dB. To provide high gain for the coverage areas, multiple beams may be required by the communications antenna and can be designed using mltihorn feed arrays, Typical antenna coverage of the continental United States (CONUS), Alaska, and Hawaii is shown in Fig. 2.19. These antenna beams must be accurately pointed to avoid maga ‘igure 2.18 Dusl-poarized communications satel a .% ations satelite antenna, (Courtesy of Hughes Aircroft56. pIGITAL. SATELLITE COMMUNICATIONS deviation of the beam boresight from its nominal diveetion, ‘Ty icl 1 evan pointing accuracies are 20.05°for east-west (pitch) and north-south: {Goll beam pointing errors and 10.2° for beam rotation (ran) Ors a pemacond part of the communications, subsystem 78 the cone munications repeater which is an intercanncetion af many channelized transponders, The conununications repeater generally consists of the fake rowing modules, as shown schematically in Pg, 2.201 1 A wideband communications receiverldowncanverter 2, Aa input multiplexer T Shannelized sraveling wave tube amplifiers TWTAS) 4, An output multiplexer -the wideband communications receivefdownconverter is designed 19 oper Toe te ppeal 300-MElz bandwidth allocated for C-band (5.9 10 6-4 GH) wine band Cd to 14.5 GH.) uplink signals and is shown schematically. Fig. 2.21 fora Ku-band uptink “The uplink signals are first filtered by awaveguide bandpass filter with about a 6O0-MEtz bandwidth and then amplified by & pare bandpass Het dette galiom arsenite ld effect transistor (GaAs FET) low er ever witha typeal noise igure of 2c0 4 dB, The plied signals rei arr nconveted (the L710 12.2-Clkz downlink Ku pa 7 19 ¢ 2 ae aca downtink Chand) by a mierweave integrate! crit downeany cee aie eee acomversion, the signals are agin amplified y 2 11.7 © 1 Gite iter er aplifer and passed through feet slaor 10 he inpu mali see rapt multiplexer is employed to separate the SO0-NHZ ‘pant seid ate maiviual transponder chamnets whose tah depends on a's mission. For example, 2 SOO-MHZ fuanbiit 0 be are santo 8 transponder channels with a enter-tevcenter Ween searation of G1 MEKZ, With frequency reuse, there we altogether 16 ersponder channels inthe satelite. The inpot matinlesss normally con ar aput circulators, input filers. group delay equaiZers amplitude SGunlizers-and output circulators. shown in Fig aac anelized TW As amplify the TowTevel sownliak Sens)s 0 & high level for transmission back to earth. Diver ‘amplifiers. are normally weMoved. in front of the high-power TWTAS 10 afew the com crriedtions receiver to be operated inthe linear made, The 8 of the TWA depends on the mission a! is about 15 to 30 W for GI-MHz Teta iransponder. The TWTA establishes the tramspondey ‘output ower and normally operates near saturation to achieve the desired out. re power. Thus it the dominant nonlinear deviee in 8 transponder and cam affect the link signal performance consilerably. The outpat downlink signals from the channelized TWA inc the outpi mliplexe for retransmission 10 easth THe ‘output tplexe provides the required ontat outa had stlennatiOn well e — a Figure 2.19 Multibeam antenna coverage (Courtesy of Hues Airraft Co{omyou fouepunpay [i ~LE ee {exojdyou yest Communications COMMUNICATIONS SATELLITE: ORBIT AND DESCRIPTION 59 tore tech irae 12.6He Lownaise Bandpass Bandpass Guts FET ‘nauter he moor te 149-1450He ~~ saare2¢He Lupting ~ = Dewan 2acue (tom aeito Figure 221 Wideband communications recsiverddownconverte, fs the attenuation necessary to suppress signal harmonics and spurious noise generated by the TWTAs. Variable power dividers may be used at the input of the output multiplexer to provide the necessary power split to select the desired transmit antenna coverage which can be selected by ground command. ’A complete communications subsystems employing frequency reuse and consisting of 16 transponders (8 transponders use the horizontal po- larization and 8 transponders use the vertical polarization) is shown in Fig. 2.23 for the type of antenna coverage shown in Fig. 2.19. The number of odd and even transponders in the east and west beams can be sclected by using the variable power dividers. 2.7.2 Telemetry, Command, and Ranging Subsystem The telemetry subsystem monitors all satellite subsystems and continu ously transmits to the earth sufficient information for determination of the satellite attitude, status, and performance as required for satellite and sub- system control. The telemetry transmitter also serves as the downlink transmitter for the ranging tones. The primary telemetry data mode is nor- mally pulse code modulation. In normal on-station operation, telemetry ddata is transmitted via the communications antenna. In the transfer orbit, the telemetry transmitter is connected toa TWTA in the communications repeater selected to provide adequate power for telemetry coverage via the omni antenna, The command subsystem controls the satellite operation through all phases of the mission by receiving and decoding commands from the ground station. It also generates a verification signal and upon receipt of tan execute signal carries out the commands. The command subsystem also serves as an uplink receiver for the ranging signals. Again, the omni ‘antenna is used in the transfer orbit for command and ranging and as an on-station backup, while the communications antenna is used on-station for command and ranging. “The ranging subsystem determines the slant range from the ground62 pIarTaL SATELLITE COMMUNICATIONS Command ‘Communications oe fd [one | Fencoew | igure 2.24 Telemetey, command, aa ranging subsystem any orbit ation to the satellite for precise transfer and geostationary js determined! by transmitting (o the satel 1 which is received! aansmitted by the contol Setemintion The st ng etn Comimand, reeves, demodelted, an el th oom othe ground contol sation where the phase di tclemetry aneaiy measured, During ansation peration ranging is Tee ee camactions tena antennacoverg pero afer onbisprovidedby thorn antenna, A Bick age Aa ans command and ranging subsystem s shown in Fig, 2.24 rae eine command receiver, the tlemety tranamier the Forel ye encoder ate all in a redundancy mode, (Redundancy configurations wil be discussed in detail in Chap. 3.) 2.7.3 Attitude Control Subsyster ‘attitude control subsystem are to maintain ae munications antenna pointing for the tite angler orbit and “The major functions of the curate satellite position and eomt and to control satellite manevers in the mee Gy © ‘oF sve satelite COMMUNICATIONS SATELLITE ORT AND DESCRIPTION 63 uring stationkeeping. The requirements of the attitude contro! system depend on whether the satellite is spin-stabilized or three-axis body- stabilized in a geostationary orbit, Most satellites are spin-stabilized in the transfer orbit, When rotated at a certain speed around the axis of max imum inertia, x satelite attains a high angular momentum anc the spin taxis is fixed in a certain direction in space by preservation of the law of tingular momentum, While in the transfer orbit, the altitude control sub- system must perform the following tasks: |. Ensure nutational sta 2. Determine the spin axis orientation and reorientation control vin the telemetry, command, snd ranging subsystem and ground controt Residual nutation stemming from the launcher separation maneuver has. Jo be damped out to m: is stabilization. For a basically stable {ellite, a simple, efficient damping system such as a pendulum ora ball in curved tube can be used. The resonance frequency and damping cocf- ficient are tuned to the nutational frequency as determined by the satellite ‘moments of inertia basically unstable satellite, the damping has to be performed by an active nutation control loop. Nutation is sensed by 2 linear accelerometer with its sensitive axis parallel to the spin axis, Nuta- tion produces a sinusoidal acceleration at the mutation frequency with a pk value proportional to the mutation amplitude. The accelerometer sig- ndpassfitered to remove both de and high-frequency compo- nents and then fed to a threshold circuit. When the amplitude exceeds the is fred to damp out the autation, “The spin axis orientation or attitude is determined via telemetered {data from enrth and sun sensors in the satellite, The earth sensor is a pa sive infrared device with 1 pencil beam field of view which operates in the 14. to I6-jan wavelength carbon dioxile absorption band. It senses the intraed rays radiating from the earth's surface by scanning the boundary ‘of the horizon between the earth and space. The location of the horizon is inlicated by the shaep increase in the sensor output caused by the temper~ ‘ature difference between space and the earth's horizon. Two earth sen- sors are canted about 5*north and 5° south ofthe spin plane, respectively, ay shown in Fig. 2.25. The sensors provide output radiance pulses as they sean the earth. When the attitude of the satellite is correctly maintained, the outpat pulses of the north earth sensor and the south earth sensor are in-phase, But when the spin axis is inclined at an angle. the output pulses: ‘of the sensors ire out-of-phase. The phase difference is uilized to control the attiude of the satellite, For spin-stabilized satellites in a geo: synchronous orbit with a spin axis nominally parallel to the earth's polar axis, telemetry data from the earth sensors alone is adequate for spin axis altitude determination. ‘This is because of the slow drift rate of the spin sais and the systematic variation in inertial direction to the earth due to an SAAR AAAS(64 iGivat SATELLITE COMMUNICATIONS +800 E snsiser Y?4 South seth tanh fils of view A held of view igure 2.26 Each and sun sensors tit mtg ve crac of £0.02 Additional dat se ima be determined a a ovidd by the sm Sensors, The sn Sensor has # the easter os Mow and operates ithe vial spectral fee Tt fished fo foto cel anda ait apertare fr detecting sl oa eet ie swept ps ie un There are wo fan-shaped Gk Fada tsp the spin asad he lhe cated 352d aoe aoe ae tian othe, Each sensor outputs a sor pulse 8 toatl 35° wit athe eld of view ofthe sn sensor. These pulses toe a ern to obtain sole aspect angle measurement ae ere eatin sla apest angle, the earth sapect angle (Hom The combina othe ain angle between the sun andthe earth der the ar igoosl the apn axis oven. Fed ett the ection contol subsystem thrusters nae onzal modes: pin anisatitade conte orbit conto ee et sai uta consol. A typea arangement of soi ae conto) a in hown in Fg 2.26. Two axial hoses have he Cae np atelitecentex of mass, and wo eal thruster moment am heater entero mas plane Atta coil = Heapproximaely pride by ground corimand. The update never a Peps manerverdepeds on epi axis rit ated © nero beter ese frgues A a geostationary aid, the onl sg exteral ditbane ge tat de to sole atom presse The st Fi ee eigned to minimize torque by rscingthe center of orbital motion. Over a 24-h obs! Conmniustcations SATELLETIS OnUIT AND DESCRIPTION 65 spin cit Rail tnvuster 2 nestor 1 6 axial eta snvuster 2 tnvaster 1 Figure 226 Arrangement of thrasters in a spine stabilzed satelite mass close to the average center of pressure. The drift varies accord ing to the changing solar aspect over the year. Attitude control, which holds the spin axis within a 0.2° range, may be performed at intervals of approximately 6 days. Orbit control consists of east-west stationkeeping longitude control) and north-south stationkeeping (inclination control) The frequency of stationkecping maneuvers depends on the permissible error and the satellite orbital position, For a stationkeeping accuracy of 0.05", east-west maneuvers are required at intervals of about 21 days using a single radial thruster operating in a pulsed mode, and north-south 1 longer interval of about 28 days using two axial maneuvers are required thrusters fired together. ‘or a three-axis body-stabilized satellite, stabilization canbe achieved with a bias-momentum system. The momentum of a high-speed Wheel provides gyroscopic stability for the satelite. Magnetic torquing, wheel speed control, and wheel axis trim automatically correct attitude crrors after initial acquisition. The wheel axis is the satellite pitch axis; thus pitch is controlled by ciusing changes in wheel speed to exchange angular momentum with the satellite body. Magnetic torquing corrects roll and yaw ervors and is used for nutation damping. Both roll and pitch can be modulated about a bias setting by ground observation and programmed into the on-board control logic. When activated, this provides a bias to pitch and roll error signals of the earth sensor and com mands the momentum wheel to pivot about the roll axis. Torquing cur- rents and wheel speed signals respond to the earth sensor error signals. Control results from the interaction between the magnetic fields of the torquer and the earth. The earth sensor detects earth-space transition at about +45° latitude, Roll error is the difference in the length of the two cehords: piteh error is indicated by off-centering of the longer chord.(6 picitat SNFELLETE COMMUNICATIONS the booms that fasten each set of panels to the satellite are normal to the The satellite generates 6 Areay weight (9) 8 a ° 7 300 720 de Power (H) Lon of thveeaxis boy-stabiized and svi eure 2.27 Weghtverus-pomer compris stared steites. eran 76 COMMUNICATIONS SATELLITE: ORMIT AND DESCRIPTION 67 plane of orbit, and itis only necessary to rotate the boom one revolution per day to keep the face of the array pointing toward the sun. Each array ig driven by an independent but redundant motor. Figure 2.27 is a weight-versus-power comparison of three-axis body-stabilized and spin- stabilized satellites. These curves show that at a higher power level a larger power-to-weight ratio can be obtained by using deployed solar arrays. However, the weight and complexity of a deployed solar array drive motor make the spin-stablized type more attractive for low-power satellites. Because geostationary satellites experience 88 eclipses during I year con station with a maximum duration of 70 min/day, batteries must be used to deliver power during an eclipse. Nickel-cadmium batteries are com- monly used, but nickel-hydrogen batteries are replacing them because of the higher energy-to-weight ratio. The batteries are charged regularly by the solar array. Besides the communications subsystems, the telemetry, command, ‘nd ranging subsystem, the attitude control subsystem, and the electrical power subsystem described in the above sections, the satellite also has a reaction control subsystem consisting of thrusters that provide the im- pulse required to perform satelite velocity and attitude control maneu- vers such as the one shown in Fig. 2.26 for a spin-stabilized satellite. The thrust can be created either from gas under pressuré in tanks or by a small rocket motor. The gas hydrazine, which is decomposed with a catalyst i the thrust nozzle, is commonly used. The apogee kick motor is a solid propellant rocke! motor employed to provide the velocity increment required to place the satellite in a geostationary orbit. The apogee kick motor is fred by a ground command, and afer firing the spent motor remains as part of the in-orbit mass of the satelite REFERENCES 1. EM, Gaposchin and K. Lambeck, "1969 Smithonian Standard Barth (1D Spec. Rep, 315, Smithtonisn Asteophysieal Observatary, May 18, 1970. M.E.Ash,"Determination of Eath Satellite Orbis" Teeh, Note 1972-5 Lincoln Lab- coin Laboratory. MIT, May 24, 1973 1, Charyk, "Communications Satelite,” Von Karman Lectureship in Astronauts, ALAA 13th Annual Meeting, Washington, D.C., 1977, Paper No, 77-323 RJ. Rash etal, “INTELSAT V Spacecraft Design Summary." AIAA 7th Com ions Satelite Systems Conference, San Diego. Cali. 1978. pp. 8-20 6. ROC, Davis eta, “Future Trends fn Communications Satelite Systems” deta Ase imauticn, Wok 8.0. 3-4, Mateh- Ape 197, pp. 275-298, 1, GrAbwtale esl, "The COMSTAR Satelite System.” COMSAT Tech, Hes, Vol. No.1 Spring 197, pp. 35-4 oe P2@OQWAAATHANARANHAND et68 11GITAL, SAFELITE COMMUNICATIONS sf. D.C. akeman et at, “The Relative Brits of Three-Axis and Dual Soin Staion Shen for Future Synchronous Communications Suits” AIAA Progress in As Se nd teroneatcs: Communications Saelies forthe 70'~Systems, Ve 26, 971, po 608-653, 9. se Dovgheny eta tex Employing Narrow-eam Antennss: 1971 pp AMHAL 1 iter tsk Momentum Whet hic Axis Atte Control for Synchronous Fe eseetiume Stites" in PL. Barge (). Communications Satelite Tel ooo ylaa, Wol 33. Cambridge, Mass: MIT Pres, 1974. pp. HE 161 Pen eee Mdm Spaceenat Dynamics and Contra. New York: Wiley, 1976 Feared Magnet Attn Con for Gesnyuchronns Seacecrall ALAA Se easaenae gnome Satelite Systems Coaference, San Diego. Cail. 1978, pm ane tke 1a rN Weber “Tetemetry and Comma Sulvystemse—The INTELSAT 1V Space Tn: COMSAT TECH. Rev. Vol 2. No. 2 Fall 1972, 0. 41-288 1a SEL yan Trees, Sauelie Communications. New Yorks VEEE, 1979 Ha Ae Wi Thee eta, Communications Saree Sptems Weide, 1975-1985 Dedham, Mass: Horion-House Microwave, (978 ee tie Conmuicaons Terese Tokys KDD Engineeng and Cons tne, 98 17 Jesper Distal Communi 197, 14) Nein, Communications Satelite Sestems. Engewoos Cis, NJ ~anitade Stabitization of Syncheonous Commnicaions Sal i spuceraf kets, Vol BNO 8, AME ations hy Satz, Enpoosoat Chis, NL: Prete al Prone tal 19. GK Oeil, The Feehnoogy Bale 120, G. Maal and M. Bousquet, Stetie Commins Fe Wa Petcard and FA, Sal, Saetne Cmmaicrons hits, NJ. Premee Hal, 1986 a, WW Wo, Elements of Digi! Satelite Conmaninions {Compt Since Press, 1985. aaa W, Donan, Stlite Communit New You: Wiley, 1986. Bae Ripa cts Fundamental of Aurodwamics, New York: Doves 197 New You: Simo & Schuster, 1943 Synter, New Yok: Wey, 1786 ‘Sens Engincering. Unpewood Vote ant, Rockile, M PROBLEMS aut inthe veloiyofasarelite tthe perigee apogee fits ie tir Stame seminar axis aa the eccentric se What is ine togestpeostationay satelite coun dey? 7 Conier tv qeotatonry ee at ngitdes 75 2nd 75°, Can etn fits ee veh other? 414 Caleulte the station faite in terms of see Sree cast ad wet tits the gzottionry a vise fom ay gen 4) ie 2S The eye pean corse ote stato a than lao «eine Hi of wii fo the geosiaionary #3 Ter the lowest elevation angle Be Ce ronan car sation a ade 48.42°N and Tonite 69.266 se Cateutte the xin ime satelite sie toa er om the cm ing a ert obit he elevation angle, the tation foie, 0d CommunicaTHONS SATELLITES ORI aND DESCRIPTION 69 27 ind the coverage ren of satelite om which its ws a 0" for the flowing crear orbits 1 asostationary (ty = 10.000 4m (6) 11 = 20.000 km, 2k Coser »slite ata alias of 17 ~ 15.000 km, What i de ime pero! during sthichit it pss vce pin onthe cath with a elevation angle exceeding = 20" sum ing the satelite rotten im the sme direction athe eat? 129 A saelit has an otitl period of hours in the ste diection asthe earth's ori tine reed ding which an cath station can communicate with this satelite with a ininionum clvation ale of E = 10" 4b, Find T- Em {L10 Conse msatelitesin the same circular tof ais 10,000 km and an earth station “working with a inion elevation angle of 10" How many suelites would be needed for the earth stlion ta beable communicate all the time? 2.11 “Two geostationary satelites at lngtes O*W and 120°W communicate via an nter- fntcite Hak. Find the roundrip peopagation delay between them. 2112 Find the ssaximu tne-fsightUstnce between two satelites atthe same atc 413 Consider a geostationary satelite at longitude 105°W and two earth stations a lone ques O-W and I20W. respectively, and at ftitdes 3S°N and ASN, eespectively, What Sr the tbl delay in sending 100 Kits over a S-Mbps satelite fink? 1144 Leteryatbe the geoventie evordinaes af an earth station, where the zaxisthe polar faitoienled from south tomomh, the aaa the intersection of the equatorial plane and the ‘tonal planes and the yas completes the right-handed coordinate system, vty in the cocwlinate Cpa) interne f the station longitode ad ied {0) Let he the longtde ofa geostationy satelite. Find the coordinates (x'y'' of the stele Cc) ind the stance R between the station and the sate, 4118 Consider two sations A and 1 vith the distances Rand to geostationaty satel tie et Be the elevation singe f station A relive tothe satelite and et rhe its aimth we enna ete at fine eomnecting the tv sations. Asam tha they are close gh tbe enmsiered in the sme fea horizontal ae.) tay Show that R= Pa? Dud cs c0sB, where the stance between sa- vis A a0 hy Assume that he wngles aan sre messured What is the total ear ein the estinated vale of Ru? Te) Assn that mia (fy) = 12, the angle measurement accuracy is 20.0%, and t= 109 ns what i the maxima eer for a? 4.16 An Ariane racket (bul in Western Europe) launches a satelite into a geostationary Jutitot Kou in French Galan, Kourou’ bude 5.3°N. Find the ineremental velocity teal to correct the erat inclination and 10 achieve orbit cieulrzaion assuming the ferice of the elplical transfer orbit is 450 km 117 Consider an eth station loeated at fongitwde 6O°W and latitude 20% and geost- qhamay cnet fone fe = ROW, Fal the earth station azimuth and elevation anges. LIV Ananth sation focated at longitve S8°W and nid 5°S points its antenna toward a [nvsionary snelite with am elevation angle of = 78.907, Fd the satelite fongtwe ssn it Bens of the ead station, crore having an equal variance 1119 A slits i in velar equatorial orbit at an ata of 17 = 10,000 km. Fi the novi eeipe Sie 2120 Verity equion 24chapter Earth Station tal eneth station 1 functional description of the digit ly divided into eee nial rth, station is basic shown in Fig, 1.4. In practice, an ea two parts: + of an upconverter and a downcon: ‘a low-noise amplifier, and an anton ts of baseband equipment, and demodulator. 1, A RF terminal, which consist verter, a high-power amplifier, 2. A baseband terminal, which consist encoder and decoder, and a modulator Senna ay eet ri aspeoprnte Fines, tn this chapter we wl foes AEM corn athe RE terminal and. wil consider the attention on ema aystem point of view ony because is Sub ase rr ey upaatly ani dea in subsequent chaies Le ean ml bedseussedin Chaps, Gand 7: the mode: The aan rl be considered in Chaps. 9 and 10; and the to Sour and decoder wile covered in Chap. 9 The RF terminal and the bascban. 3.1 EARTH STATION ANTENNA important subsystems of the RF. the modulated RE “The earth station antenna is one oft terminal because it provides a means of transmitting anti STATION 71 carrier to the satellite within the uplink frequency spectrum and receiving the RF carrier from the satellite within the downlink frequency spectrum. ‘The earth station antenna must meet three basic requirements: 1. The antenna must have a highly directive gain; that is, it must focus its radiated energy into a narrow beam to illuminate the satellite an- tenna in both the transmit and receive modes to provide the required uplink and downlink carrier power. Also, the antenna radiation pat- tern must have a low sidelobe level to reduce interference from unwanted signals and to minimize interference into other satellites and terrestrial systems. 2. ‘The antenna must have a low noise temperature so that the effective noise temperature of the receive side of the earth station, which is proportional to the antenna temperature, can be kept low to reduce the noise power within the downlink carrier bandwidth. To achieve a low noise characteristic, the antenna radiation pattern must be controlled in such a way as to minimize the energy radiated into sources other than the satellite. Also, the ohmic losses of the antenna that contrib: tite direetly to its noise temperature must be minimized. This the olimic loss of the waveguide that connects the low-noise to the antenna feed. 3. The antenna must be easily steered so that a tracking system (if required) can be employed to point the antenna beam accurately toward the satelite taking into account the satellite's drift in position. ‘This is essential for minimizing antenna pointing loss 1 Antenna Types The two most popular earth station antennas that meet the above requirements are the paraboloid antenna with a focal point feed and the Cassegrain antenna. ‘A paraboloid antenna witha focal point feed is shown in Fig. 3.1. This type of antenna consists of a reflector which is a section of a surface formed by rotating a parabola about its axis, and a feed whose phase center is located at the focal point of the paraboloid reflector. The size of the antenna is represented by the diameter D of the reflector. The feed is, connected to a high-power amplifier and a low-noise amplifier through an orthogonal mole transducer (OMT) which is a three-port network. ‘The inherent isolation of the OMT is normally better than 40 dB. On the trans- the signal energy from the output of the high-power amplifier is iated at the focal point by the feed and illuminates the reflector which reflects and focuses the signal energy into a narrow beam. On the receive side the signal energy captured by the reflector converges on the Focal point and is received by the feed which is then routed to the input of the VaAAAS AMARAR AR AFrom tigh-powar ampiier Lind feos phase poner - 77 LAV rc cere ance (eeaaeae | Figure 3.4 Paraholot antenaa with a fea pint Fe rowsnoise amplifier. This type of antenna is easily steered and offers rea- rele gain efficiency in the range of 50 to 60%. The disadvantage som gen the antenna points to the satellite at a high elevation ankle eine ease. the feed radiation which spills over the edge ofthe reflector (Colever energy) illuminates the ground whose noise temperate ca We se igh as 200° K and results in a high antenna noise contrib Fe mbploid antennas with a focal point feed are most often employet in the United States for receive-only applications a Censegrain antenna is a dual-relector antenna which consists of paraboloid main reflector, whose focal point is coincident wih ihe virtual Focal point ofa hyperbotoid subreflector, and a feed, whose phase cenit soca petra focal point of the subreflector, as shown in Fig, 3.2. On the i ee, the signal energy fom te oxpt ofthe high-power ampli raeraited atthe real focal point by the feed and illuminates the conse coi ast che subseflector which reflects the signal energy back as i ere incident from a feed whose phase center is located at the commer Theat point of the main reflector and subreflector. The reflected eneTBy erred again by the main cflector to form the antenna beam. On the re, ‘main reflector is directed teive side, the signal energy captured by the coe it focal point, However, the subreflector reflects the sign! enerEY tenet is real focal point where the phase center of the feed is locale sen ced therefore reecves the incoming energy and routes it tthe inet area noise amplier through the OMT. A Cassegrain antes is ‘elector, subreflector, and feed-—to proce an optimum antenna system Coen te isuogninstean eles many advantage oer (he setts lw se tmprsture poms sccurey a eibiy reat sever crrp fom he eed siete foward tas hesceo temper ype festa 30 econ wee eens cons ne temperate sa compared fo ht ofthe veal atena swith fed oeated ear the etx ofthe tcc reste mechni Sabi canbe eve ha Wh he ere ein and ntenra Ts ierested stb permits rrp ihe area Tatate he sre nioe tansion nes connecting he igh 2 ee copier tee, seem wvepc see eraeetesey be employed. a Casein antenna With beam clei ect system aston ing 33 The beam wavepideasem- See ftom mrrssunrrted by shoud and precy Tested eet acter: eee, he tration asa the axa SAAR mio onfgrton ats 8 RF energy fone between he iy Ane Wabstesr anda such, most be desine fo achieve a Ws ah towing ie fee ibe mounted nthe conte sun lene The shoud agteby ace els Paraboiola From ih-power ampiiliar | it om | To ‘on-noise ampiiee Hyperbotoid lector focal point and subreflector virtual Foeal point snrtiecion support and feed phase center gure 2 Casserain ante14 IGITAL SATELLITE COMMUNICATIONS la ere) ReEESS IN en (Courtesy of CTE) ground noise and provides a rig structure which maintains the mounting AMrepity ofthe mirrors when the antenna is subjected (0 wind, therma} oF ther external loading conditions. The lower section of the shroud assem- Shy is supported by the pedestal and rotates about the azimuth axi, The upper section ofthe shroud assembly is supported by the main rece or seprort structure and rotates about the elevation axis. As seen in Fis Fe, the beam waveguide mirror system directs the energy to and from the feed and the reflectors. The configuration utilized is based on optics though a correction is made for diffraction effects by using slightly elip- tient curved mirrors. For proper shaping and positioning of the beam earns station 78 1 br \ 1 ee eae \WRTUAL MAGE OF EEO PASE CENTER ' ! ' pomerls y Feeo pase centen ‘waveguide mirrors, the energy from the feed located in the equipment cane OO sta hs fed pe center appers fo be ae etait inopanton niron ae. CrandD more ee ete Sinaia rotcs Mier sone eeton 2 Mae hen he ares seed ig ceva Se ee sated ans neponn nie main eho Se cet seantnony seis ohn ore er arya wo atpentet sures beta See een a yeni emmy ue an dng Samana:pe aL BID 116 Gira SAFELITE COMMUNICATIONS Va waren SECTION a gure 4 Wistand ortho wansovcertilrer fe. (Comesy of OTE) awity, Figure 3.4 shows a wideband OMT diplexer type of the satellite caps nena that provides horizontal frequency reuse feed for a Cassegrain dnd vertical polurization for a Ku-band eperation. 3.1.2 Antenna Gain mance parameter of an earth station a Cain is perhaps the Key neo ue and downlink carrier power. tenna because it directly affects the uplink For an antenna, the gain is given by ont on 1 at lego) yeaton fewer Ja Hee pas 18s coe ge een (9D ova cnr apart, ifs hat 4 = 70" Hert on9(22) =n 22) 62 antenna diameter (m). where D ts the percentage of the aperture arc A ‘Phe antenna efficiency m represent iption and is a product of that is used effectively in transmission oF rece ees 8 185154 Cal oe ko o EARTIE STATION 77 various efficiency factors that reduce the antenna gain. Typical efficiency faetors for a Cassegrain antenna such as the one shown in Fig. 3.30 are ‘i 21 nenemminsme 63) where my = main reflector illumination efficiency ‘n= spillover efficiency ‘na = phase efficiency R= subreflector and support structure blocking efficiency ed system dissipative efficiency feflector surface tolerance efficiency ‘The illumination efficiency , is determined by the characteristic of the field distribution across the main reflector aperture. If it is uniform over the entire aperture area, then m, =. The spillover efficiency ms repre- ents not only the energy spilled over the edge of the main reflector but also the energy spilled over the edge of the subreflector. To minimize the “pillover loss, 2 feed with low sidelobes in its radiation pattern is desired. Tonchieve this pattern, multiple modes are used in the design of the feed radiation section which is a horn. Furthermore, the feed angle subtended by the subreflector is chosen so that the main beam of the feed radiation teen intersects the subreflector at a low level, minimizing the feed main beam spillover past the subreflector. However, the low edge illumination forthe subreflector normally results in sharply tapered illumination across the main reflector aperture, resulting in a low illuminating efficiency mv. With 2 Cassegrain antenna this condition can be improved substantially ty deliberately altering the shape of the subreflector to distribute the energy essentially uniformly nearly to the edge of the main reflector but then falling off very sharply. An illumination efficiency m, of 0.94 to 0.96 tan be achieved in practice with a main reflector spillover efficiency of as high as 0.99. With good feed design, a subreflector spillover efficiency on the order of 0.98 can be realized, Thus a spillover efficiency of mx = 0.97 team be achieved in a shaped system. Distorting the shape of the subreflector to achieve uniform jllumina- tion across the main reflector results in a phase error being introduced into the main reflector. This phase error results in eneray being radiated in tindesired directions, thus decreasing the gain and increasing the sidelobe fevelof the antenna. The phase efliciency 73 identifies this gain loss. Most ofthis loss can be eliminated, however, by reshaping the main reflector to torreet the phase error. In a well-designed Cassegrain antenna the phase fflieney ny can be on the order of 0.98 and 0.99 at the design frequency nd remain’ on the order of 0.95 over 70% of the operating 500-MHz hand. Blocking of the main reflector aperture by the subreflector and sup- port structure results in an effectively smaller aperture, hence afoss in the dntenna gain, The subreflector blocking efficiency is about 0.97, and that atthe support structure is about 0.95 in a well-designed antenna. The dis118 DIGITAL SATELLITE COMMUNICATIONS also reduces the antenna gain. Depending sipti feed system structure the efficiency 7 can be as high as 0.94 libs atovercited efficiency factors are primarily dependent on the main reflector and subreflector geometries and the feed system s\ructite aa eres ake operating frequency. In practice the main relector and a tector eannat be bail tothe ideal shapes without some surface 10 saree. This results ina scattering of energy in unwanted directions 10 crance: Tyas vo that associated with the phase error. The surface (ole anne sn effect, be considered a special type of phase error which mits the maximum achievable grin Gy in the sense that, for a given 0 firitglerance and antenna diameter, increasing the operating Keaven’y face eee antenna gain until it equals Gy» Further increasing, Ne aerating frequency will derease the antenna grin. The surface 1OCTanC® Crecency naimposes an upper imit on the maximum operating reaver), ficiency Te ipaximam antenna gain, and is, fixed by current mane pence technology. The reflector surface tolerance efficiency me may BE ae ne f-(428)° wee ENT Fo nen) C27 troduced by the surface error ¢. The anténna st writhin current commercial technology is as follows: sipative loss of the feed system enna surface to! 10 = fs 10 psi2m 2xWtsfssxloe 2SmSD=6m G5) lots genx $msDsxm fed Ku-band 20-m Cassegrain antenna ‘The performance of a well-design fem ig shown in Table r peam waveguide frequency reuse feed syst 3. 3.1.3 Antenna Pointing Loss ‘cussed the gain ofthe antenna as given in of the antenna which ean be ‘curately toward the satellite, A js notin fine with the Inthe previous section we have dis in G.2). This is the peak or on-axis ‘achieved if the antenna beam is pointed Toss in gain can occur ifthe antenna pointing vector ARTI STATION 79, ‘Table 3.1 Performance of a Ku-band 20-m Cassegrain antenna amen femetes 9S OH _1425 CUE ‘amination fclney 096 094 Spillover eiency Main eletor 099 099 Svbefetor 096 a9 Phase efeieney 098 098 locking eficiensy ‘Subrefector 097 oo Seppo siuctore 095 09s Feed system dissipative eicieney ‘asi eed 094 093 Diplexer 098 098 Team waveguide 091 036 Surface toleraneeeflency Main selector on 083 Subrelector 097 097 Net antena eiiency 0s? 054 6533 65.82 satellite position vector as shown in Fig, 3.5. The antenna poi an be evaluated from the antenna gain pattern which is a function of the Sitrakis angle @ For a paraboloid reflector with aperture diameter D, the Normalized gain pattern for a uniform aperture distribution (2) is given by (3.60) Mester Ka) Pointing _Y/ /satetite position 9 cor igure 35 Antena pointing ero.“The nulf-power. beamwidth for a vniform aperture, distrioysia twice he value of Oat Gy (0) ~ 4, which is 1.02ND radians or $830 degrees, Fora parabolic aperture distribution, the hall-power beamwidth seeegaID radians or 72.7ND degrees. 1n practice the antenna polniog a aes normaly Kept within one-third ofthe nallspower beam Jamwidth of 20-m Cassegrain antenna For example, the half-power be rot cniform aperture distsibution is 0,062" at 14.25 GHz. Ifthe pointing eat eeet within 0:02, the normalized gain willbe greater than ©» cre 0.757 or, equivalently, the toss in gain will be less than 1) Ne a ion. qNote that the normalized peak guint 0= Ois sion 1 1 scone the earth station antenna i subjected toa wind Yoading Tee! and eee dit in orbit, an antenna tracking system is necessa0)/oF eh vo minimize the pointing error, ‘The antenn a taper ayatem isa coset-foop pointing systems tha iy the anne’ tainting vector which isa funetion ofthe aeimuth an elevator angles is rant irom the received signal- One of the commonly used seat sereapapatems for earth stations isa step trick which derives 16 90 tar pointing vector from the signal strength ofa satelite beacon, signal teri ope of tacking the antenna is caused to move around he ee ofa raat ater in the plane normal 10 the axis of the antenna Rent aa ire 3.6. The center ofthe square ABCD corresponds 10 Ne shown Tete psition. Thus, the satelite poston vector is signe sume Sat can pating vector, the amplitude of the Beacon sant wt ar enc of the four pointing postions A, B, C, and D wit be cee An erzor in elevation results inthe off-axis. ange # fo ves OO 4 cate difering fom the of-axis angle 0 for positions Cand D, Sines the a pc signal amplitude is proportional to the square of the off-axis eae as indicated in (36a) or (3.7), by comparing he aver amplitude aaa ey with that of Cand D,theifference i the offset anes of 6% ae aye error can be determined. Similarly, by comparing he Whe vot teat 4 and D with that a B and Cy the azimuth ale eoFoF be determined. ‘The step track system ean provide & tracking accuracy of ‘hast to Lor the half-power beamwidth, which results ina pointing loss of enween 0.5 and 1.5 dB. For better tracking accuracy, a monopulse track= ing system must be employed (2]. This type of tracking ean provide rryeking necuracy of up to #r of the half-power beamwidth. For a fixed veer ckition with a small antenna (a half-power beamwidth on the order eg. or more), program tracking system can be used. This isan open- tap pointing system where the antenna pointing veetor is derived from the enrth station position and the satellite ephemeris data, The satelite ephemeris data is normally obtained from a felemetry, tacking, and ¢om- eejatation (TT&C) and is stored in the memory of the tracking system. ‘ihe satellite ephemeris data is updated periodically, The updated interval depends on the satelite stationkeeping characteristics. 1.1 Effective Isotropic Radiated Power ro express the transnitted power ofan earth station ora satellite, the ef Jertive isotropic radiated power (EIRP) is normally employed. The earth sition EIRP is simply the power generated by the high-power amplifier Times the gain of the earth station antenna, taking into account the loss in the tansinission Tine (waveguide) that connects the output of the high- power amplifier to the feed of the earth station antenna. 1f we let Pr Menote the input power at the feed ofthe antenna and Gr the transmit an~ tenna gain, the earth station EIRP is simply EIRP = PxG+ (3.8) For example, consider a 2-4W high-power amplifier and a 204n Cas- Sprain antenna whose transmitted gain is 66.82 UB at 14.25 GHz asB82 DIGITAL SATELLITE COMMUNICATIONS given in Table 31. Ift is assumed thatthe oss ofthe waveguide that con: ects the high-power amplifier to the feed is 1 dB, then the earth station EIRP in decibel-watts is (noting that P decibel-watts is equivalent to 10 log Pr watts). = 98,82 dBW 33 466.82 ~ “The uplink carrier power, that i, the power of the enrrier received st the satellite is diectly proportional to the earth station EIRP. This will be discussed in Chap. 4 EIRP 341.5 Antenna Gain-to-Noise Temperature Ratio “The antenna gain-to-noise temperature ratio GIT isa figure of merit com monly used to indicate the performance ofthe earth station antenna Hn ‘her in relation to sensitivity in receiving the downlink cee er from the satellite. The parameter G is the receive antenna gain catered to the input ofthe low-noise amplifier. For example, ifthe input trike low-noise amplifier is connected directly to the output port of the feed system of the 20-m Cassegrain antenna whose performance i ius raed in Table 3.1 then the receive antenna gain at 11.95 GHzis 65:93 io ire piece of waveguide with a 0.53-dB loss is used to connect the ‘pat ofthe Lowenoise amplifier to the output port of the feed system, the i aret se antenna gain referred (othe input of the low-noise amplifier is sine ‘iy 65 dB. The parameter is defined asthe earth station system oe re perature referred also to the input of the low-noise amplifier. We Have aiscussed the antenna gain previously, therefore in this section We vv concentrate on determination of the earth station system noise tem: perature. ‘The treatme of noise called white noise whos the low-noise amplif sat of noise in communications systems is based on a form se power spectral density is NuI2, as Shown in Fig, 3.7, and is fat overa large range of frequencies. White noise cain aterivied as a Gaussian random process with a zero mean, ad i wom motion of electrons in includes thermal noise produced by the ram sacriueting media, solarnoise, and cosmic noise. White noise corrupts the sensived signal in an additive fashion and is normally referred to as additive ‘shite Gaussian noise (AWGN) in the analysis of communications sys wane ne treatment of noise is outside the scope of this book, and itis ver mended thatthe reader consult (3] fora detailed study. (Appendix 3A gives a brief review of thermal noise.) sive tctrical communications systems, the power spectral density of white noise delivered to a matched load from a noise source is cus: Tomarily expressed in watts per hertz, (W/H2) as Bo EARTIL STATION 83) Winite noise power spectral densly Frequency gure 2.7 White nose power special density where k is Boltzmann's constant (1.38 x 10° JIK) and 7, is the noise tempertire af th alse oes measured in kelvins. This means that if his noise source is connected tothe input of an ideal bandpass filter with bandwidth B in hertz (Fig. 3.8) whose input resistance is matched to the source resistance, the output noise power in watts is simply N=NB =k 6.10) Since any pasive or active two-ort ? sive or active tworport system, sch a the waveguide sn te lane ethene ampere area an se lowase amples have eqvalent nose hat contributes fo the wesc trom the tenn we mus ike igo acount ther eet Conier wo ort system wih un © ans neve souree of temperate Ty con- teaed tts input The output nose power ina Bandwidth (Ha) then N=GRTB+Ny oan wer = ° oe 8 2B eure 3 teal Indpass fer, ja) temsfer fonction of hadpass ler: 4, center frequency of fanipass fer, bandwith of bandpass filter in her sReaaan 4= GKBT, + Td) voter where Ne Te oie GB.) rom (312) its seen that Nan be considered tobe produced BY 8. ficti- From (1? sgoree of equivalent noise teniperaure T, connected 16 He eet ofthe system. Therefore we conclude that a noisy Wwo-Por SST ine er tharacteized by its equivalent noise temperature Pe The carameter T, + T, in G.12) is defined as the system noise temperature referred to the input of the two-port system: T=T+T G4) mother words, we can model a noisy two-port system as anoistess I aet gm and account for the increased nite by assigning (6 input port system an Mow temperature T higher than Ty by Te Mote tt ae etree pis simply the noise temperature measured t (he OUT) Cine twoeport system, Another measure ofthe internal generated fy a two-port aystem is the noise gure F, defined 9® the output noise wer of the system divided by the output nose power if the system is roe ot eal the interaal noise sources are absenD, assuming HA the aoiseeurees atthe input is atthe ambient temperature To (F< normally rei soipe 290 K). By this definition the noise figure Fis also the ratio of the signalsto-ncse ratio atthe system input to the sanahsN AEE ratio at qResgatem output. Ths Fis simply the ratio of Vin(3.11) with T,= Te tnd GkToB which is N when Nu GkTyD £N, KT tT 14 es) From (3.15) itis seen that F-1Te 6.16) T. 1s My and Mz in cascade as shown in zed by its gxin Gy and its equivalent Now consider two 2-port system Fig, 3.9. Bach system Mis character rage emperare Ta {= 12 The noise source at the input of the Figure 3.9 Cascio tworport system For equivalent 0 equivalent noise temperature analysis coded system is assumedto have temperature Ty The no temperature Ts The noise power Nat the output of system My is given by (3.12): , N, = GkB(T, + Te) B.17) ‘This x amplifed by Ma and appears at its output as Miz = GiG:kB(T, + Tes) G.18) sn it he me nwa tnd yh nt nar nd aera ele ee a Ny= GakTaB us) “he total output noise power is simply the sum of Nis and Ne N= Nyt Ny = GiGAB(T, + Te) + GokTeaB = o.oaa(n+ Ta +B) (320 ee te eriate menial nen =r Masatezed by ts pa G = GyGs, whieh obvious, and is noise temperature: ands equivalent Ta Tat Ge @2n Fqvation (3.21) demonstrates clearly the contribution of the second sys tem fo te overlie tempers Sen that, if the gain of the fst system M, is large enough (G, >> TealT~), then the second system contributes negligibly to the overall noise temperature, The results of (31) can be easly generalized to a cascade of m systems Ina I, Tot Pt Get tT oe 0) By sing 0.15). the nie figure of n systems in cascade can be expressed Tt. Food TOG Gas 3.23)86 DIGITAL SATELLITE COMMUNICATIONS uation of the earth station system noise temperature referred tothe iaput ofthe low-noise amplifier, we pause here fora moment to make the following observation concerning the ohmic Toss inva transmission line such as a waveguide, coaxial cable, or other levies characterized by power loss rather than power gain. For such fossy two-port system, let L>1 be the power loss the. its. grin G eifL <1) and Ty its ambient temperature; then the output noise is Simply TB. By using (3.12) with T, = Ty and G = I/L we have Before proceeding to an eva ATB = ABT a + To) which yields the equivatent noise temperature of alossy two-port system: T.=(L- DT 3.24) By compating (3.24) and (3.16) it s seen that the noise Sgure of lossy two-port system is Pol 6.25) Now consider the receive side of the earth station, which consists of the antenna, the waveguide that connects the antenna feed to the low. noise amplifier, the low-noise amplifier, and the Jownconverter in cas ane, as shown in Fig. 3.10, The antenna noise temperature is measured sting feed output and is denoted by 7, The waveguide is characterized ayits power loss Ly > 1 (or power gain Gy = I/Ly < I)andits equivalent poise temperature Te: = (La ~ 1)To. The low-noise amplifier is character Jed by is gain Gx and its equivalent noise temperature Tex The equiva: Tent noise temperature of the downconverter is T.. The equivalent noise temperature T, ofthe cascaded low-noise amplifier and downconverter is given by (3.21) as Ta 6) Te=Tat 3.26) Low-noi ‘smpliter Gur i = Waveguide bile te rreure 340 Recsive sf f th eth station for sytem mse eeratere evan ARTI STATION 87 From (3.12) we note that the noise power at the output of the waveguide is given by : N= GAB + To) L ABT, + (La ~ OTe) men (Ts “Thus the noise temperature T,measured at the output of the waveguide is r) 027) Taper 6.28) From (14), (8.26), and (3.28, iis seen that the earth station system pois temperature referred tothe inp ofthe low-noise ample i simply TaTAT, u Tay teal Taps tet nt tas 0.29 ‘ample 3:1 To illstateevaunton of the antenna gainto.noie temperature ratio, Sse 201 Casein nnn whose ecieaean gan is 63.5 dh al 11.95 (Gitras given by Table 1, The cecive she ofthe earth station is characterized by le flowing parameters “Antenna noize terapeature: T= 60 K Waveguide lass: y= 1.072 0 00) ownoivesmpier Travvalet noise temperature: T= 130 K Gains Gy» 10° (60.0) Downconvere equivalent nate fempernure: Ty 11 10" Soubattoting al the above ves into 0.29) yeks the earth station system noise tem perature refered to the inp ofthe low oise amplifier as (T, = 290 K oo, 9002 nim pa hg BE (a0) 4 180+ finan ee race a 255K the antenna gun eferted to the input ofthe low-nave amplifier i G=6559- 6583-03 =6523 8 “Thus the antenna gains-osiae Lempert ratio in decel per Kelvin is @ Sm Gum = 10%" T 36523-2383 ean at > VAAN2, The lower the loss of the waveguiue aie rie ee Late low-noise ampli, the higher the value of GT 5, The lower the equivalent noise temperature of the low-noise amplif- re nigher the value of G7 Also its gn mt be sufficiently ves Te reduce the noise contribution of the downconverter, ‘4 The value of GIT is invariant regardless ofthe reference DOIN We ae atte input of the fow-noise amplifier because this fs the Povnt choose I Jontibution ofthe Tow-noise semper is clearly shown. 3.6 G/T Measurement One of the precision methods for measuring the antenna HAT aS Cmperature ratio G/T ofthe earth station isthe radio te method. This temperement employs the Y factor defined as the ratio of ths nase EON wth the star accurately centered on the receive beam 2708 the cefasound noise power (when the star has ited out of she beam): T+ Toe 7 0.0) ue of star and T= system noise 0 Trae {4] a8 follows: effective noise temperat where Toor = tenna gain G is related temperature. The receive ant IT gan 45 = hux density per I-H2 bandwidth of rao star (W/m TH) {f= measurement frequency (H2) Speed of Hight = 2.997925 X 10" mis t= 1.38 x 10 SK where f= Boltemann’s constant Substituting 2.30) into 3.31) yiekts ¢ at 3.32) Gay ue 11 where there is no atmos condi uation for Equation (3.32) represents the ideal point source, The ea pheric loss and the radio star is hhanideat condition is given bY G fink Gaba “OE a3) war ex were ba = atmospheric loss and £4 = correction factor f tent of radio star on oon oor OOF Gon 061 Hoe 0.023 023) 01k 02 0s O19 016 0104s 033029 0220.8 0 Nak 090062030 vor oa! 052,022 0 tte 992059 02s Shean surlase_ condilons: 21°C, 7.5 gin? HO. sud atm 4 dy ‘The atmospheric loss or attenuation L, can be cnleutated from exist- ing data and is given by (4B) + by (oo =7.5 gli) 62 QUE = Ta) 1, (4B) = EAB) + bak 2 QNC=Te (3,34) sin E nith onesay attenuation for a moderate ion fora moderately humid atmos: there (1 gn surfice water vapor and a surface tempera 1__lmse atl ne Ba gen = iter vapor density coretion colicient, by i give ‘Table 3.3 Nhrise y= temperature vorvecton coefcent: ey given in Table 3:3 fuses wer vapor density (Ei TRE fcr by = 7.5 Haw) accounts forthe difference Between the local surface Seater vapor density and 7.5 gine where Li ‘Table 33 Water vapor density and temper- nure correction coefficients Treqeney Water vapor density Temperature illey _Corection cotrection fe ny 210% 10" 240% 10" Is fad 10 45510 ” Bag 10-7 Ts 108 w 2an x10 1310 0 Drs x10 97108 = Sins x 10° S74 10790 mattat, SATELLITE COMMUBIEATIONS T Figure 3.1 Atmosphere atenuntion, surface temperature (XC), The factor ¢(21°C ~ Ty) accounts for the difference between the loc: surface temperature and “The surface water vapor density py may be found from the ideal gas law: EARTH STATION 91 he, 0 TAIT, F BID where f= relative humidity (e.g, h = 60% =0.6) and e,= saturated par tial pressure of water vapor (N/m*) corresponding to local surface temper ature T9(°C); ¢y 206.43 exp[0.0354 Te("F)]. “The correction factor Ly for the angular extent of the radio star is a function of its brightness distribution and the antenna gain pattern and is plotted in Fig. 3.12 in decibels for Taurus, Cassiopeia A, and Cygnus A. ‘The nominal ux density for Taurus A at 3.95 GHz is 716.9 x 10 Wim/H. This star does not exhibit any flux variation with time and has a spectral index of 0.25, ‘The nominal flux density for Cassiopeia A at 4.08 GHz was 1086 x 10 W/m?/Hz in the year 1964.8 (about mid-September 1964), lus specteal index is 0.74. The flux density of Cassiopeia A decreases at arate of 1.1% per year. The flux for the date of measurement may be ‘computed from the expression S=S.(1- A)" G35) where $= lux density for date of measurement ‘Sy reference flux density taken to be 1086 x 10-" Wim'/Hz for the epoch 1964.8 f= 0.011 y= number of years from 1964.8 to date of measurement oa G02 ask 5 Goldie) z oo Tua (Ooted tna) 00, oo 01 02 03 04 05 Hatt-power Figure 3.12 Correction for for the smith of antenna (degree) anguac extent of rai str PG390 9800004 VARIA{hue nonmnal tu oenatay 14 9 8 Wim*/Hz. This star does not exhibit any ‘and has a spectral index of 1.19. Wien a measurement is made at a frequenty for which the flux den- &-(4y 630 flux density variation with time f reference frequency for where x= spectral index of radio star and fo reference flux density Sw are at setup for GIT measurement is shown schematically in Fis 3.13, and the measurement procedure is outlined as follows the desired measurement frequency. 1. Tune the oscillator to obtain Lightly higher than the anticipated ¥ fac 2, Set the attenuator toa value s tor. 9, Ore the star ephemeris to move the antenna team ont the sta. Peak Tharmeter indication by adjusting the antenna servo controls 44, Whenthe peak meter reading is obtained, record this value and that of the attenuator. 5, ‘asthe sae drifts outside the antenna beam gradually decrease 9° aeinton to Keep the meter reading approximately constant, When Trent is completed, adjust the attenuator until che peak reading o& aoe ary in step 4is reestablished and then record the new valve of oo uator. The dilerence in attenuator readings between sters 4 and 5 is the Y factor in decibels. 6, Repeat the measurement for a number of times and take the average. example Colette aospheri attention ase on the fling mene Frequency: 20 GHE ‘aitode: 480m Relave humisiy: b= 60% = 06 Surface temperatre: Ty 26.7°C = 80°F Fevation ange: E= 47" 1 Determine £5 for 75 en? surface water vapor an 2 21°C surfs temperate fom Table 3.2 02008 a. indihe water var ny The saturated pai ress of A S Ta sSo0 Nim Ths no = 06 50004810067 4279) = 192 ue 4, Betermine by and cy fom Table 3.2 888 33 220035 en 138x107 “amplitior converter .tanuation si coven igure 313 Test sete for GIT surement 4, ind La from (3.340: 02.4027 - 0.009 02 eT SOO? = 0.63 0b 3.2 INGH-POWER AMPLIFIER eet the most wlety use igh gover spits in earth tons the il ave abe unpifer (WEA) shown schema i i 3.14 Ts traeig wave lide employs the pi of wlosty modulation in ior evclng waves Te RE sl sai travels down Ie srr eset ton ea i a eat en pte euler afer delering th caer tte RE fl The hel slows down the propagation velco she RF su the velco ah oh fhe lesteon ben, whieh Monae bythe ae age at he estode Thies n an eration sarees ise a ndced by te RP signal a the elton AF input cathode AF output . ae 7 = wi : Magnets Cottctoe gine 44 Traveling wave the uml94 miotrat, SATELLITE CoMMUNTERTIONS which results in the transfer of eneray from the electron beam to the RE signal causing it tobe amplified. The amplification grows as the RF signal travels down the tube. The traveling wave tube can achieve a bandwidgh ton the order of 10%, hence ean cover the entire 500 MHz allocated for satelite communications. "Another type of high-power amplifier used in earth iystron amplifier which can provide higher gain and better efficiency than the traveling wave tube amplifier but at a much smaller bandwidth on the order of 2%. For low-power applications, Impat diode amplifiers oF GaAs FET amplifiers ean be employed. These are solid-state amplifiers and offer much better efficiency than the above two types of amplifiers. jons is the 3.2.1 Redundancy Configurations Reliability is of utmost importance in satellite communications. When @ single high-power amplifier is used, transmission will stop upon its failure. ‘Therefore the high-power amplifier in earth stations always employs some sort of redundancy configuration. The most basic redundancy configura- tion s the 1:1 redundancy shown in Fig. 3.15. The signal from the output fof the upconverter is split equally by the power divider and fed to the input of HPA 1 and HPA 2. The waveguide switch allows the output sie al of HPA 1 to be transmitted by the antenna while the output signal from HPA 2s dumped to a load which dissipates the power inthe form of heat, When HPA | fails, the switch automatically connects the output of HPA | tothe matched load and that of HPA 2'to the antenna feed (the dotted lines in the switch in Fig. 3.15 are the connections when HPA 1 fails ‘When dual polarization is employed i tion, four high-power amplifiers are needed to provide a 1:1 redun chown schematically in Fig. 3.16. Since the cost ofa high-power ampli feris inthe tens of thousands of dollars, the |:2 redundancy configuration ‘shown in Fig. 3.17 is often employed for cost-effective reasons at the ¢x- a frequency reuse earth st iney Waveguide vat ‘eaten watched from Power To sntenna upconverer——}aivider tees uw LL] neaz | — gure 3. HPA in 11 edundancy eohgwaton TE BELE EARTH STATION 95° way fom roe to entenna nconvenee — Giger toe J aiatzaton 9) neaz Hag cen Power To anteana ‘ wager tees ‘polation 2) was Figure 3.16 HPA in I rsundsney configuration for doa-plariztion operation pense of less reliability. A common load is used for both switches. The 2 redundancy configuration shown in Fig 3.17 i ned when the earth station transmits anly one cartier in each polarization st any instant in os a cose ton 2 oS Gog IAAT= Se aioe canoe? igre 418 HPA n= 1:2 edndacy conf wih exer conning ime, When two carvers are transmitted simnltangonsly ne Se etfon.the 1:2 eedundaney configuration io Fi. 5 18 can be employed: A iratier combiner is employed 10 combine the two Garett ‘When more sree ce carers are transmitted simultaneously, 9 et) elaborate tha ee gat be used to combine the carters, a5 WH] Be discussed in the next section. 3.2.2 Carrier Combining nest ave for ambiing cosets 8 dren sn: cos (aot + #) cos ot = Fos (og ~ alt ~ AL oan + Hos (ue + walt + 4 tous one can sleet the LO frequency ot sch that ea = wr + fm a renee cm and we the ands ler at the outplc ss Met oe anit ow consider the dul-conversion process in Fig. 3.30. The tt sidan othe Fea conan! + 4) by he Gist LO corr cos wr! els CEE EARTH STATION 107 assuming we) > od) cos(wrt + 4) c08 wat = 4c0s (we, — wed = 4) 3.38) + Feos (wn + ole + $1 The it nase selects the upper sideband cos{(we + a)t + 4}, and the second conversion bythe second LO carir cos wa yes the following (assuming wr, > We, + Orb ; cos oe, + eal + g] cos eat = 4 e08 (Wa ~ Wer ~ wo)! ~ 4) + cos (we + 1 + ood + ‘The second bandpass le set the uppers er sles the upper sideband cole, + we + een nee ors mers sly cone stom chosen povided they sist he cosa y+ 010 = 04 ~ We > Wey + ty, and wey > Wo. “ oe andoass| o Mixer Mier? * Sanco Bandpast| oy Atenvates err? Image PF mea = © gore 338 Senate nga of yam ofan upconvet.) Sngle omer (bia com sina. (c) Frequency spectrum of dual-upeoaversion. a eae PQOWAD IA OAAMARAANS AAAsot nies ei ee Soe igen £3, Tare we nse he spc nest shen Teter: fo = ate = 70 1 Lan, 3.9 64 Cie (SDM Salih Chr onder B= 36S tne Ca ney fa = wen este 0 tf 4 fa> SOOM Therefore fo = 930 Mle “rosie eter equeny ofthe fit bana irs ox + fu = (it > 500 Mt fp = Mite an st be designed fo stent eae = fo + BD = 78 Mite to Te sia at fy = 930 Miz a he ing sidebo as ee evel The second LO feeney is eet 9 cas nok fa [Note nat the waantey Wer StU OLLUPES Jame 30 foe ~ Wer +f) = 3.9 GHE fies ~ fa ~ fo = fon ~ Ws + £9 = 44 OH, ‘etc otie ie fequeny spect 391064 Gir. The upconversion proces situated in Fig. 3 300, 3.4.2 Transponder Hopping, Polarization Hopping, and Redundancy Configuration for Upconverter Modern digital earth stations employing time division multiple access often have to transmit more than one traffic burst during one TOMA frame length to a number of designated transponders according to the transmit trafic assignment, or transmit burst time plan as it i often called. The ability of the earth station to hop its carrier from transponder to transponder during each frame is referred to as transponder hopping. ‘Also, if dual polarization is employed, the earth station might have to hop its carrier from one transponder in one polarization to another trans- ponder in the other polarization, This is known as polarization hopping. ‘Transponder hopping and polarization hopping must be achieved within a fraction of the guard time between bursts (Fig. 1.13). In practice this ean then the order of 100 to 200 ns. Also, for reliability reasons, the upcon- verter must employ some sort of redundancy configuration. In the follow- ing discussion we will consider various solutions for an upconversion that can meet the above three requirements, namely, transponder hopping, po- larization hopping, and redundancy. “The first solution is to use frequency agile upconversion where hop- ping of the carrier is achieved by switching the second LO frequency a (Fig, 3.306). This can be done by using as many local oscillators as there are hopping frequencies and a fast switeh (a pin diode switch). This type of Uupconverter is called a hopping LO upconverter. A fast polarization switch (a pin diode switch) is employed at the output of the upconverter to direct the RF uplink carrier to the proper polarization. The LO hopping tupconverter shown in Fig. 3.31 also provides a 1:1 redundancy. Since the amplitude and group delay variations in the transmit side of the earth sta- tion and the satellite contribute to the distortion of the modulated carrier, and the path length differences between upconverters result in various signal delays, equalizers (EQL) must be employed. The number of equal- iners is equal to the number of hopping transponders. A fast IF switch (a prin diode switch) is used at the input of the upconverter to select a desig- fnated equalizer for a hopping carrier. The IF switch, the LO switch, and the polarization switch are controlled by signals from the TDMA haseband equipment. Because of the path length differences between the TDMA baseband equipment and the three above-mentioned switches, an npropriate time delay must be employed for each control signal. TThe second solution for achieving transponder hopping is to use as110 piciTAL SATELLITE COMMUNICATIONS gore 331 LO hopring urconvertet : fers as there are frequencies, Two RF switches are farzation as shown in Pig. 3.32, which also “The switching can also be done at the 1F 3.33. The advantage of IF switching is that tesrvsne fast eich s required foreach chin. Also, the IF swilehes ot te pat in the baseband TDMA terminal, which is an nQvanr when aera tongconnection between the baseband terminal andthe BF et, theres Fnce then ony the time delay due co the electrical pat lent of il tor needs to be taken nto account. But combining at he our i he upconverters may increase the noise output and the spuriou! Se oa Wee athough ot correlated at he, transmit output cov! Pe ral ned atthe satelite if every statin in the network uses he same conriguration and the same second IF frequency inthe upconvertet vaaratgole upconverter provides easier pathlength equalization Won 4LO hopping upeonverter and is ess costly when the numberof KpPie a aneponders is Tess than three, but its expensive when the umber of hopping transponders is large many upconvert ‘employed, one for each pol provides a 1:1 redundancy, frequency as shown in eal a i rom | § -e00}—[ve]}-4® {iran el feet Ei Head = Towa gem somaoes Tapa Sa tron is cebaonn --4- i aq 1 | ! i 4 1 fon | ' enodutator 2! ' i ' ' 1 ! i 1 i ‘ 1 ; ' ‘Hopping conic H (trom TOMA evipment) i rms ey eulomat | gure 332 Multiple upconverter oof iran honing 38 DOWNCONVERTER io with RE sth anspor and polar ‘The downconverter (DC) receives the modulated RF carrier from the low-noise amplifier and translate process or with PeOBIOA, js radio frequency nin the dowalin Irequency spectrum of the satellite to the intermediate frequenc) L ‘ upconversion, downconversion may be. a hieved with a single-conversion dual-conversion process, 28 shown in Fig. 3.34 a-c. ADAAIAIAAG yaadeau ToHPA system (polanaton # | Hee | To Pa system |[ear| (potazaton 2 % fe] fe] Hopping control (trom TOMA equipment) - 4 : i if u 1 fom it H ee \ ae 3 \ el t | 1 \ LH ear i \ wpe ant ! (tom TOMA equipment) randy equipment ie 7 rutin with A sis for tanspander bo igure 339 Mabie unsonventercoguaion with HF fin polaiaton hopin 4.541 Conversion Process "ge dnwnconverson is seldom used neath tations except some A ile down on ly statons. For the dan downconversion in Une pe cterion oi he RF ar cont + #07 second LO carrier cos Wet (assuming Wer S00 Ml, Teefne we ‘The wecond LO fcqeney st be selected shh tha fa= som" ‘othe center Feeney of he fist bandpass er is f= fa = 162 nd he ist LO freuen even by fn = Jo ~ fo ~ fe Jo = 930 te somite cmon The snd rth a en cy at i 2 bh of aa rt arate eget within fs = fn 3 2 1H = J fe ee cao wep tec, Ne hh ge ed eas : aoe eee cigs 2 > 0 = 1 ll. To id alae fe fs ' ea tsar tps ay pl the pt of es Baas ee fier Tae downeonversion process is ilsate ia Fig, 3.34: 8 downconverter 43,5.2 Transponder Hopping, Polarization Hopping, and Redundancy Configuration for Downconverter In TDMA operations, the earth station'may be required to receive onoverlapping trafic bursts from different transponders in different po- oiaarions during a frame. As with upconverters, transponder hopping and polarization hopping can be achieved by a LO hopping downcon: erter, as shown in Fig. 3.35, or by multiple downconverters with IF Switching. as shown in Fig. 3.36. The latter is more cost-effective when the number of hopping transponders is less than three, white the former is more attractive when the number of hopping transponders is large Figure 235 LON us 6399.00 13PBINAD arerenene Q16 DIGITAL SATELLITE COMMUNICATIONS stapler, — ~Tadmoante J Least ro- 4 ' ' : 1 nae Sret—fral— | | Sn 8 TT ELL remote i | | 1 ' 1 1 TL i ‘ oe tian iort336 Mohn dowaconvesterconfigaton with 1 tees fr tampon fe Iantation hoping. 3,6 MONITORING AND CONTROL, sections, except for the antenna all earth sta ower amplifier, the low-noise amplifier verter. must employ some sort of this of utmost importance. ‘As we have seen in previous tion systems namely. the high the upeonverter. and the downcor "sto maintain high reliability whic oneline equipment. The modes and resolving all these failure modes the failed to the redundant system is cal (mao. by amomatis switchover Ned moniraring and control EARTH STATION 17 Ina satellite communications network, which consists of many earth stations and a network control center (NCC), monitoring and control is extremely important and involves many levels such as 1, RF terminal monitoring and control 2. Baseband terminal monitoring and controt 3, Remote monitoring and control of the entire earth station from the nnetwork control center vin the satellite link with a backup terrestrial telephone tine “The monitoring and control system must have the capability t0 Collect status data for classification Convey status data to network operators. Interpret status data Initiate fault isolations Switch over redundant equipment on command Convey control data to the baseband equipment for traffic assignment, an~ tenna pointing (satellite ephemeris for program tracking), and so forth Maintain surveillance of equipment shelter facilities. In this section we concentrate on RF terminal monitoring and control (RFT M&C) and postpone a discussion of the baseband equipment and emote monitoring and control to Chap. 6 where the TDMA system is considered. The overall monitoring and control ofa typical TDMA earth station is shown in Fig, 3.37. RFT M&C typically has the following three main components: 1. Redundancy switching logic for the high-power amplifier, the low- noise amplifier, the upconverter, and the downconverter 2. A data acquisition unit (DAU) 3, Amantenna control unit (ACU) ‘The redundaney switching logic of each subsystem is itself part of the subsystem. Inherent in this concept is the understanding that the subsys- tems being monitored and controlled are redundant and self-sustaining. That is, an automatic switchover from the failed to the redundant system is executed by the subsystem logic. Failure alarms, status indications, and possibly analog values are then collected by the data acquisition unit which inputs them to a M&C ‘minicomputer that may be the central processor of the TDMA equipment (which also handles other TDMA functions besides monitoring and con- trol) or an entirely separate minicomputer that handles only earth station monitoring and control, (In the latter case the M&C computer and theFigere 337 Monitoring and control ofthe ext sta EARTH STATION 119: “TDMA central processor must be appropriately interfaced for the pur- pose of remote monitoring and control by the NCC through a satellite Tink.) The DAU is a microprocessor-based device which continuously scans the inpit status and alarm points, stores them internally after each ‘scan, and reports them to the M&C minicomputer which scans the DAU at a predetermined rate (normally once every few seconds). The M&C minicomputer then adds the points to the data base, checks for alarms. falus changes, and priority and outputs the appropriate message to 2 focal portable cathode ray tube (CRT) terminal and printer at the earth station of transmits it-via a M&C channel through the satellite to the NCC. The DAU also accepts switching control information from the local portable terminal or from the NCC via the M&C minicomputer and inputs control information to the appropriate subsystem logic fora redun- dancy switchover. The DAU is interfaced with the M&C minicomputer via standard RS-232 or RS-422 data links or an IEEE-488 bus. |The antenna control unit is also interfaced with the M&C minicom- puter via standard RS-232 or RS-422 data links or an IEEE-488 bus. ‘Alarm, statis, and tracking information is transmitted to the M&C tminicomputer following a data scan request received by the ACU. When program tracking is employed, the ACU must be able to accept the track- ing table downloaded to it from the M&C minicomputer responsible for generating the tracking information based on satellite ephemeris data received from the NCC. The satellite ephemeris is updated periodically by the NCC based on the latest TT&C data, ‘As mentioned above, the status and alarm data in the M&C minicom- ter is continuously compared to the data received from the previous an. Any changes in state detected are noted, placed in a message queue, nd outputted at the local portable terminal or transmitted to the NCC for Anpropriate aetion, The alarm event normaily assumes two priorties:1 and 2, oF major and minor, which represent the state of alarm from the subsystem; a change in state is immediately brought to the operator's at- tention. The status event generally assumes priority 3 and reflects a state tf the subsystem not considered an alarm and which may be changed to the opposite state via a control signal. When the M&C minicomputer de- tects a status change, the following actions will be considered: If the status change is due to the result of a control signal sent to it recently, the change is interpreted as being authorized and the new state is simply stored in the current data base. [the status change is not due to a recent contro! signal, the change is con- sidered unauthorized; the event priority is then changed from 3 to 1. ind it is treated as a major alarm. Figure 3.38 shows the logic control ofa low-noise amplifier system in 1:1 redundaney configuration. The switching logic can accept commandsded bei vee ee YS 120 plorTaL SATELLITE COMMUNICATIONS rom was ited L_ Sounconverer | uw: |}-— Logie J Figure 238 Loic contol ofa low-ncise amir from individual LNAs of remote commands from the NCC downline foaded to the M&C minicomputer. Typical alarms, status, and controls for the 1:1 redundant subsystems of the RF terminal are given in Table 3:5. It is worth noting that in the worst case of a logic control failure the situation at the RE teeminal is unknown but that the RF terminal is not cout of service, 3.7 RELIABILITY system of the earth station except the an 0 increase its reli |As we know by now, every subs tenna employs some sort of redundancy configuration bility. In general the reliability of hardware is-the probability that it will perform its task without failure under designed conditions and withi Sefined periods. If we assume that failures occur randomly, the probabil ity of the hardware being operated longer than the time interval i given by the exponential distribution RO) xpt—pt) oan) Jc, The average fuilure time, commonly where 0 is the average failure . EARTH STATION 121 ‘Fable 3.5 Alarms, status, and controls for the 1:1 redundant subsystems of the RF terminal Subsystem Deer on Deer off Rin Bower on Rin blower off [Antenna contrat Lighepower amplifer Helix overcurent etic overvetage Hightermperature wide Hightemperature power supply HPA T onctne HPA 1 offtine HPA 2 ondine HPA 2 offine HPA 1 prime power IDA 2 pie power UPA 1 faut reset HPA 2 fault reset Input switch position | Iepet switch poston 2 Outpt switch position 1 ‘Oui switch position 2 noise ample ENA | offing ENA 2 onine ENA offline Tapa site positon Input switch poston 2 Output switch poston t ‘Outpt ewteh position 2 \uqsonverter downconverter Power supa # Power sop 2 Local oeilators CHDC tenting UC HDC 1 ofine UC DC 2 ontine UC UDC 2 ofine Inpet switch position Input switch positon 2 Ghat witch poston F it with positon 2 ‘Asem Peres Stes HK XXH HK Control HOH Xm122 DIGITAL SATELLITE COMMUBICATIONS called the meu time to failure (MTF), is thus given by + 1 rere = [=n Mere = [une = = stmany independent subsystems ate connected in cascade and any cone of thes fail, the cascade system will also fil, Thus the refaility of are raded system fs simply te product of the cliablites ofthe indivi tal subsystem, Rit) = e&B(-P Rad = RUDRA RAD G43) so(- $m oan the mean time to failure of a cascaded system is L “Therefore srt, = f= tan = 3 = Z : Sa STAY (3.45) here MTF) = Hp and the corresponding average failure rate is w= 3 0.46) ‘cascaded systems are the low-noise amplifier-downcon- rer amplifier-uneonverter chain. ted in parallel, the system will ‘Therefore the reliability of the Examples of verter chain and the high-pow ‘When identical subsystems are connec! ‘operate as long as any one system operates, parallel system is given by Rage tJ = Rio] == ROY [I= exp(-pi))" ~ Scar) ewrint Su) enti Ban ~ ()- “Thus the mean time to failure is EARTIE STATION 123 3.48) ind the corresponding average failure rate is 7 (3.49) ey Examples of parallel systems are the 1:1 and 1:2 redundancy configura~ tions discussed in previous sections. It is obvious that a parallel system has a higher reliability then a cascaded system. Once the average failure rate or the mean time to failure of the sub- systems in an earth station is known, the mean time to failure of the earth station can be evaluated by appropriate applications of the results in (3.45) and (3.48) or (3.46) and (3.49), The average probability of availabil- y of the earth station depends not only on its mean time to failure but also on how fast the failed subsystem (or subsystems) can be repaired or replaced, Let MTTF, and MTTR, denote the mean time to failure and mean time to repair, respectively, ofthe transmit side, and let MTTF; and MTTR, denote the mean time to failure and mean time to repair of the re~ ceive side, respectively. Then the earth station availability is given by Pam PasPoa 6.502) where MITF, Pa" STTR MTT R, (2.508) oe 6.50) Pes TTR, + MTTR, Large earth stations are normally designed to have a very high degree of reliability, that is, with an average probability of availability on the order of 0.9990 to 0.9998. jle.35 tn this example the MTTF of the redundant LNA shown in Fig. 3.27 ‘aleulate, Typical parameters are MTF of LNA =3.2 10° MTTF of svteh = 5 % 10" “The 3:pon switch canbe enodled by (wo 2-podt switches each wih the same MTTF. Tc equivalent rednnant scheme ie shown‘ Fig. 3.39-By sing 0.45) the MTT, of 28960095 Rea= =e < : rs] Lefa}d igure 339 Redondant LNA and is equivalence one Te of he parallel system it MTTFe = Sapa 2 ION HS IOI nines The ATE of he pe system can bin fom 8 2 2) NTT vertey (are. (2) So 3 TTR, = 42,8538 ow assuming that the MTTR is 2h the probability of valbity othe paral system is pet TT, + = 099995 < a) REFERENCES 1. J Rute,“Antenna Tolerance Theory—A Review." Pro. EEE, Vol.S4.NO. 4 Apr. 1966, pp. 633-640. 2, Mot. Stet Znro 3. M. Sehwarts, Information Tronsnisrion, McGraw Hil, 1970 4, We Kroutl Jeet ‘Apr. 1978, pr 7 5, RUC. Hansen, "Low Noise Antennas 6, Jibiiketal, "Antenna Noise Temperature,” 1403-1410 1. W. L, Stuteman and G. Thiele, Ancenna Theory. New York: Wil 1 Ri stegens, "Design and Performance of Low-Cost Iterated MI enters for Earth Station Applications.” COMSAT Tech. 1999, 90H sducton to Rader Systems, 24 6, New York: MeGraw- i 1980. ‘Modulrion, nd Noise, 28 ei, New Yoo "Satelite System Measurements," Proc. IEEE, Vol. 6, No.4 Microwave J Sune 1999, pp. 19-24 Prac IEE, Vel. 115,N6. 10, Oct. 1968.90. lr, 1981, Up. and Downcon EARTH STATION 125 APPENDIX 3A THERMAL NOISE SOURCE __arerrrnrereaeerer—Eaeeeos Random motion of free electrons ina resistor whose temperature is above absolute zero causes a noise voltage v(t) to be generated at the terminals “The power spectrum of the noise voltage [3] is aA ol x [h * Spain i cy where R = resistor value (ohms) Planck's constant Boltzmann's constant K = kelvins T, = temperature of resistor Sap 12 10° Jos 38 x 10 N/K ‘Atsoom temperature T, = 293 K and, for frequencies below 1000 GHz, DUAT, < 0.15, hence explhl/VAT,) =I + hlsUAT, and (3.A 1) canbe ap- proximated by SU) = 2RKI (.a2) Iterefore the thermal noise power generated in a resistor R over a Inindpass bandwidth 2 (Fig. 3.8) is r ‘The avaximum thermal noise power that this source delivers to.a matched loud of resistance R (noise-free) is one-fourth of P, divided by R: N=ETD Nis then the available nvise power from the noise source v(1) ant dependent of the source resistance R. The available power spectral den- sity of a thermal noise source is therefore NAT SUN apo RAT Gan PROBLEMS 1M Hd the antenna ga oss dv fo sfce tolerance atthe maximum achievable gain. Is hin Hse acceptable? [M2 Show thatthe normalized gin pattern ofthe mi abe of an antenna ean Be aporoxi= ‘nated by the Gasian funtion og exp (FZ)126 DIoITAL SATELLITE COMMUNICATIONS “where isthe hal power beamwidth of he antenna. What i the error at the Gul level Town fom the pesk ein? Compare the posting loss at = 0,02" with that from (3.7 fora 420m antenna a 14.25 GHz 143 Using the result in Peob. 3.2, fd the maxiqum achievable antenon gain for 9 given pointing eror. What i the corresponding pintng loss? Compare this result with that ia Prob. 234 Consider a 20m Casserain antenna witha parabulic aperture distuation operating 5t 11.25 Gite with a efceney of = 0.54 aaueface tolerance dD = 2 x 10"*ands pointing iro of 002" Assume thatthe lee ofthe waveguide that connects the tpt ofthe ih flower ample othe feed ofthe antenna i 0.7 UB. Wht is the rated power ofthe high power amplifer needed to achieve an EIKP of 98 JO? 13s Consider the eceve side ofthe ath sation showa a Fig. 3.10, Find the system noise iempertue feed fo the fee theses Ge he apt of he waves) and refered tothe downconverter inpot assuming Rat the equipment ambient temperature is “Tea Ke, the antenna noise empertue is T = 60 K, she waveguide lossis Ly = 0.3 dD, Ine cquivalet ols temperate end gat ofthe low-noise amplifier ate Ty, 150 K an G00 dO, rexpesively, andthe equivalent noize temperature ofthe downconverter is Fen Sook. Lethe antenna gain be 65.53 dl Fed the antenna pinsonoise temper tore eli for ench ease 6 Consider the receive side of ert station (Fig. 3.10. Find he system ise temper tore refereed tothe input ofthe low-noise amplifier vsing the following partes ‘Antena gun: $2.6 dB Waveguide lors: 0.1 JB ownoise ali Gain $3 a8 Eauivalent noise temperature: 40 K Dowaconterter eqivalent note lemperature; 219.600 K ‘Ambien temperature: 300 K Main beam: 25.6 K ‘Subreetr spillover: 3.) K Main rector spiover: 0.7 K Blockage: 9.8 K Surface tolerance: 18 K Feed loss: 139 K 17 Let Tube the cosmic note temperature: T, and ne Toss ofthe atmosphere, respectively. Assuming thatthe ise enters the antenna vin the main beam ony. what i he antenna noise temperature? 3 The ois temperatre ofan antenna i given by are the temperature apd absorbing G10-9)7(.9) sn 0d8-40 where 10.) = distribution of temperature overall anges about antenna 610.4) = gain of antenna in direction (2. (0.8) = spherical coordinates pie antenna asuing the average atmosshiri mise Find the nis temperature of ans temperature in kevin canbe approximated by oa oe gu [EARTH STATION 127 ra-ta-2voe osesns TOS) = To) =IH~95.5(90- 0) 7S <0 90" Ten) = 719 = 290 219 Consider the receive se of 99 earth station shove schematically in Fig. 310 and ase ‘ume that the mise conthation by the donmnconverter x nelle. Let he antenna gin be (6.3 uh the waveguide lows be 0.65 dt, snd the swith lor f the redundant low-noise am lifer subsystem (Fp, 3.27) be 02 JB, Ii desired to have GIT margin of 0.3 BK. Find "he required equivalent nase temperature ofthe low-noise amplifier in the redundancy con guration given that GIT = 38.7 UDIK. 3.10 Consider the transmit side of an eanth station with the following peromeers: FIRP = 87.5 dBW; aniena gin = 66.2 UB; the high-power amplifier subsystem employs the 1:1 reundaney configuration shown in Fig 3-13, where he switch lors i 0.2 dB; thei sividaal HPA ternal fss is O18 JD andthe HPA is eequted to operate wth am uot tuckotFo UB from its maximum power. Findthe required maximum outpa power ofthe Alividual HPA, assuming ar EIRP margin of 0.5 JBW i necessary. [AM How many 3-4B couplers ate zeded to combine four carters? Show the combining scheme IF the ss of «3B coupler 0.1 lB, what i the combining lose for each caer? 2.12 Consider the dy Ble combiner shown in Fig. 3.20 and assume tha the iset= tin os of ech hybrid i 1 dl and the insertion loss of ench fier sO. eB What i the ower fss of carer 1, easier 2, nd earir 3? 4.13 Consider the power combining scheme in Fig. 3.29 and atsune that th switch loss is {0.2.9 the combining oss de to pase errors 0-1 JB, and the hybrid foe JB, Find the maximum output power in wats ofthe individual HPA needed to produce a oxtput ower of 30D. Ifthe individval HPA has to aperate with an output power Sack of 2 {what ets required maximum utp power in watt? 2114 nthe theoretical equivalent nose temperature of eryageically cooled parametric amplifier operating nt 12.2 Git assuming the pump fequency ie 60 GH [AS Consider the upconverter showa in Fi. 3.30b, Find the estaints on ey anda i te tower sdebands are selected ‘M6 Consider a dal upconverter with the following parameters Upliakfequeney spectrums: 4 9 145 Gtr Fist inermediate frequency: 140 Miz Career bandwith: 72 Mi [BPE 1 center trequeney (Fi. 3.06) = 1.19 Gz (a) Find theft local oxciltor frequency. (0) Find the range ofthe second lol orator frequency. (6) Find the Fequency spectrum of the unwanted sideband (a) Fin the image spectrum atthe ouput ofthe upconverter, (6) Find the order ofthe Chebyshev bandpass Biter. IDF peeded to provide an attenuation of 100 4B for the st lea! oir. 1L17 Consider a duel upconverter with the fellowing parameters Uplink fequency spect: [4 to 148 GH Fist intermediate frequency: 70 MHz (Carer bandwidth 49 MIT Number of channels 10 Lower edge of image spestrim of first channel: 13.9305 GHz Ole 6 BOO000 ARAN AAA128 DIGITAL SAYELLITE COMMUNICATIONS Lower sieband centr fequency of est channel, 11.679 GH {a) Find the second LO frequency {b) Find the first LO frequensy suas My ne rile upeonvertr consign in Fig 32 10 icrent ens 28 Moti be upon eniets, ad fast RE wices Oe using te hn ar unaeapeonerter coin 1 redundancy confyaten) O1 inl acne py he aie congo an 2 be eee of NYE 2 sneer ugconetereupd wi reacncy seni a ie OF oe ee eet ay cre eqsency inthe st a tally soe eS {en ejcycontgration, were abe mut of o-ine OPORYer sa eran pooner cnguracion it Fig 2.931 incteae 68 HOEY a9 Mes mer upconveteseqnieran fas TF sche ve 9p win emote cofgaraion a» abe! dee of aly (He Ot het slaaton vy uz osonvetr chain 1:1 cedundaneyconfiuralon) Sipe dencanerer noni i, 340, Fine contains on ov andoef Sr Consider a dual downconverter fr 10 channel the folowing parameters Cro a cney spectrum: 11,725 + (a W049 BHI2 = 12-0010 Ps intermediate requency: 70 MIz Second intermedi Cequency: 1173 GH ey Find the fist focal oselistr frauen. oa ng the second lc ecto egueey (Fi 2408 2) Ed ee reueny spect hot ste ji bythe andes Her (oor ty ate second intermediate frequen. sav Consider a dua downconverter wih the following raramees ‘Satta requency spectrom: 1010 1.7 GME Partincermedate (equeney: 140 MERE Fat oct oseilator fequeney: 2.1 GHE Caner bandwidths 72 MH er ant 7 eon lca xen gene i eed om acted utes inthe range 152 50 1707S MT wig a HERAT lle. ‘What isthe moliieative factor? Wt) Find the image frequency sp#s0 sian teng deeceverr equiped wh ewecy sete au Oot TO a emery to ny caietreaoeay it the sytem as SNH CATT, avoily a s conuraon Fig 2.26108 or anda) coe the 1 edn enone Ht: Use wo ail pion 0s te ten edt equion ncnahe RF termi MAC en 2 2 a ni lop coel os taster fab? SHER 9 Mac a eer these problems, Js ou newer by comparing he Ara a0 ae ages to those ofthe system discussed i the Text sania he NTT athe HPA syatems in Fig 3.13 and FE 21% 25 Feats ofthe niin devise For he power ie area 0" forthe wavesuie site 5.26 Find the MTTF for the hooeing 3 pe average let the ios vies 4 Even a ace 0 rh oetnation swith 4% 106 the ao peonverter and 2 10" fr the waves sh exch with a 49-Mibe bani, th 0 tforihe HPA. 110 upconveser conigueation shown i Fit, 3.34 Satellite Link ‘Asin any other communications system, the ultimate goal of a steiig “ystem isto provide a satisfactory transmission quality fr signals relayed between earth stations. We know by now that a satelite channel is a bandpass channel, and modulation must be employed~to- transmit baseband information. In afvaaTog satellite system using frequency mod lation the signal-to-noise ratio ofa voice channel at the FM demodulator ‘output is & measure of the signal's hdelity. This output signal-to-noise fatio is a function of the carrier-to-noise ratio C/N of the satellite Hak {Chap $) In a digital satellite system the performance of the satelite sig- pal received at an earth station is measured in terms of theawerage proba- hilty of bit error, which is a function of the link carrier-to-noise ratio CIN, the information bit duration 7 (oF, equivalently, the information bit rate Thm I/T,)-and the noise bandwidth B of the satellite channel. For given information bit rate, the signal quality results in a trade-off between the type of digital modulation, which we will discuss in Chap. 9, and the car- siete ratio ver he satelite link, which is the main subject of this chapter. ‘We know that a satellite link consists of an uplink and a downlink. nal quality over the uplink depends on how strong the signal is when it leaves the originating earth station and how the satellite receives it. On the dawnfink, the signal quality depends on how strongly the satelite can fetransmit the signal and how the destination earth station receives it, ny130 DIGITAL SATELLITE COMMUNICATIONS Because of the great distances between a geostationary satellite and the earth stations (compared 10 terrestrial communications terminals) and because the power of the radiated signal diminishes as the square of the distance it travels, the uplink signal received by the satelite and the ‘downlink signal received by the earth station are very weak and ean be tasily disturbed by the ever-present AWGN. In audition, the uplink sig tal may be contaminated by signals transmitted by other earth stations (o adjacent satellites, and the downlink signal may be contaminated by sig ‘acent satellites. Furthermore, rain can severely at- is above 10 GHz. and-reduce the isolation between .queney reuse. system. In this Tink will be dis nals coming from adj tenuate satellite signal orthogonally polarized signals in a fret ‘chapter these effects on the performance of a satellite cussed in detail, 4.1 BASIC LINK ANALYSIS own in Fig. 4.1.-The transmit earth sta- Consider a basic satellite ink shi yy the EIRP of the tion transmits the carrier s(1) whose power is simpl carrier given by (3.8): EIRP= PG an carrier power at antenna feed and Gy transmit an assume that the transmission occurs tinder clear-sky conditions and that the only attenuation the carrier s(¢) ‘suffers is the uplink free space loss, the atmospheric attenuation discussed in See, 3.1.6, and the antenna tracking loss. The uplink free space lass is siven by where Py tenna gain, For the time being we own Figure 4.1 asic satelite Hk SATELLITE LINK 131 ad) tn (Y= “ny where d, = uplink slant range (m) snd canbe calculated from (2.22) A. = pink wavelength (=) 4. = wplink cartier Frequency (He) ‘c= speed of light = 2.997925 x 10 m/s Let y(t) be the earrier received atthe satellite and nt) the AWGN with zero mean that contaminates the uplink carriers). Then the received carrier plus noise at the satellite is s4(t) + (1). If the satellite antenna gain is Gy, the uplink carrier power, that is, the power of s,(1), is a= Elst) EIRP) G, TL - ox Coe a (4.3) where E-] denotes the expected valu and Lis the sum of he antenna tracking oss and atmospheric attenuation. The uplink noise power in @.10) as mines Ma= Bt] mee (44) where 7, = satelite system noise temperature (K) and can be calculated from the noise temperature of the satellite antenna (which is about 290 K since the antenna always sees a hot earth) and the equivalent noise temperature of the satellite com- ‘munications repeater shown in Fig. 4.1 ise bandwidth of satellite channel (H2) Boltemann’s constant = 1,38 x 10"? J/K. k ‘The parameter G,/T. is the satellite antenna gain-to-noise temperature fli! From (43) and (44) the uplink eaievtomoise rags (a). (FE) (#) (a) 2 (waz) (7) (ea) 3) Note that EIRP/4zrd3L is simply the carrier power fluc densi pve tinny ply power fluc density (Wim') % 46)132 DIGITAL SATELLITE COMMUNICATIONS ture area is 1 m? as seen from (3-1) when = Tand A= Fm? soft Guar = an Substituting (4.6) into (4.5) yields (i). oe) €) (o) an In summary, the uplink carrier-to-noise ratio can be cateulated by (4.5) if the carrier EIRP is given, and by (4.8)if the power fx density atthe sat- ellite is given «ae we have indicated previously. the received carrier plus noise atthe satelite is sott) + nglt). This carrer plus noise is amplified and downcon- Nerted by the satellite communications receiver-downconverter and then Smnplified again by the satellite TWTA and retransmitted back to earth by fhe satelite antenna. Denote the retransmitted carrier plus noise by Fre) + At) which possesses the same carrier-to-noise ratio (C1). given AU Sy as the received carrier plus noise suit) + nyirit Let EIRP, be the lclite EIRP (or power) of the retransmitted carrier $7, that is, Bs (2) = EIRP,. Then the power ofthe accompanied uplink noise é & (49 & (490 4g, = BIRD. ‘ Nu= CIN ame “The received cartier plus noise at the receive earth station is ite) ht) + nat) where (1) and ie) ate the attenuated versions of 41) And futt, espectively, and na(t) isthe additional independent down/ink SAWGN with zero mean that further contaminates #(1). After taking into aaevant the fre space loss Ly, the antenna tracking loss and the atmospheric Menvation on the dovinlink L', and the receive antenna gain G of the earth ‘Hation, the power of the carier (atthe receive earth station is c= APO _ EIRPDG tal! (EIR ) (aq? (4.10) | nconsierng the Cequency-ransatngianspondes, we nelet he feuency trans on erik and dowalnk an simply convert he former co the later hou 28 ao erter of ging. Therefore C.N, = gCaleNS™ CalNe™ (CIN) SATELLITE LINK 133, = downlink slant range (rn) = downlink carrier frequency (Hz) ‘antenna tracking loss and atmospheric attenuation “The power ofthe accompanied uplink noise (0) appearing on the downlink n EU Tal? {EIRP TLC. went (Ee) (§). & arp The downlink noise power is again given by (3.10) as, Na = BLO) =sTB (4.12) where T earth station system noise temperature (K) and ean be evaltiated from (3.29 Thus the composite noise power at the receive earth station ist N= E (GHG) + ma) = BUCO) + ELH) N+Ny OBR gia) (lieve am he eartier-to-noise ratio of the overall satelite link (uplink and slowalink) is therefore given by (4.10) and (4.13) as follows; (EIRP Ycl4 nf) ?GIL’ CIN CERF) ah FATE [22 Gat) @) Ga)] SR GH) 4 (4.14) Let (a), bby comparing (4.15) to (4.5) itis seen that (C/N) is simply the downlink sige Ho wees sana ANN aces) ad eee dot ale nr em nh sl dE Eln(i)] =0- ner 1 el Qt) + ELC) men} + Ela) OD) =134 DrorTaL SATELLITE COMMUNICATIONS parameter G/T is the antenna gain- ‘Substituting (4.15) carrier-to-noise ratio, and the famili to-noise temperature ratio of the receive earth station. into (4.14) yields the link carrier-to-noise ratio: wl ON] 19 spe quai ste teers tie este be cant fom rages goon 1) pe fname als eae eee alte myrtle lets ie op ne as ely oe gpl ed be coped te dvi, Fm ao Wc (eaten CN = (ON Ise he sna i eereene stance, at CN), << CC woettclton fs shown in Fi. #2. The operating point where the TWEA 80, saturation point teansler cenarcteisic ‘out power nosmaired to saturation Vente £2 Nonlinear characterise ofa satelite TWEA, SATELLITE LINK 135 output power is maximum is called the saturation point. A further increase in the TWTA input power will result in a decrease in the output power, In many cases the TWTA is operated below the saturation point to avoid nonlinear distortions, especially when there are many carriers per transponder (more than one), to reduce the power of the generated in- termodulation products that act as interference signals (we will discuss in- termeidulation interference in Chap. 5). Let BO, and BO, be the input and ‘output buck-offs of the TWTA, respectively, which are defined as >t 7) (4.18) (4.19) where EIRP. = carrier EIRP required to saturate satellite TWTA erate satan BR 1. = 0) an a nonliearueton of BO, Sutin 17 no (45.41 in (48, a9 419 io (4.15 ls © c_\" (Ga) (1 iL (9), -cemmeaa ( CIN, where CIN is the carrier-to-noise ratio, (By a “large” C'/ we mean that itis on the order of 20dB or more and is atleast 3 UB larger than C/N. Such a satelite lnk is sid to be nvise-dominant. When CHI < CIN. the link is said to be inter Jerence-donitianiy Otherwise the replacement of non-Gaussian interfer- ences by Gaussian noise ofthe same power will result in a higher proba- bility of bit error; thatis, the result will be pessimistic. Figures 4.3 and 4.4 show the average probability of bit error ?, for PSK and QPSK earters, respectively, inthe presence of AWGN and one non-Gaussian (PSK and QPSK) imeference signal [1,2]. Consider the case of » QPSK carter in Fig. 4.4: it is seen that, when there is no interference (Cit = ®), a cartiee-to-noise ratio of GON = 13.5 dB is required for Py = 10°. Wien Cit = 20 dB, itis required that CIN = 14.3 dB to achieve P= 10°, Ifthe imesference is treated as AWGN, then according to (4,29a) the total carrier-to-noise ratio willbe 13.26 Alt which according to 9.51), oF the curve labeled Ci = © in Fig. 4.4, yields 2 AP RAAANAAA