INTERNATIONALTELECOMMUNICATIONUNIONITU-T J.83TELECOMMUNICATIONSTANDARDIZATIONSECTOROFITU(04/97)SERIES J: TRANSMISSION OF TELEVISION, SOUNDPROGRAMME AND OTHER MULTIMEDIA SIGNALSDigital transmission of television signalsDigitaI muIti-programme systems for teIevision,sound and data services for cabIe distributionITU-TRecommendationJ.83(PreviouslyCCITTRecommendation)ITU-TJ-SERIESRECOMMENDATIONSTRANSMISSION OF TELEVISION, SOUND PROGRAMME AND OTHER MULTIMEDIA SIGNALSFor further details, please refer to ITU-T List of Recommendations.General Recommendations J.1J.9General Recommendations concerning sound-programme transmissions J.10J.19Performance characteristics of sound-programme circuits J.20J.29Characteristics of equipment and lines used for setting up sound-programme circuits J.30J.39Characteristics of equipment for coding analogue sound-programme signals J.40J.49Digital transmission of sound-programme signals J.50J.59Characteristics of circuits for television transmissions J.60J.69Systems for television transmission over metallic lines and interconnection with radio-relaylinksJ.70J.79Digital transmission of television signals J.80J.89Specific Recommendations for television transmission J.90J.99Transmission of signals with multiplexing of video, sound and data, and signals of newsystemsJ.100J.109Interactive services J.110J.119Recommendation J.83 (04/97) iFOREWORDTheITU-T(TelecommunicationStandardizationSector)isapermanentorganoftheInternationalTelecommunicationUnion(ITU).TheITU-Tisresponsibleforstudyingtechnical,operatingandtariffquestionsandissuingRecommen-dations on them with a view to standardizing telecommunications on a worldwide basis.TheWorldTelecommunicationStandardizationConference(WTSC),whichmeetseveryfouryears,establishesthetopics for study by the ITU-T Study Groups which, in their turn, produce Recommendations on these topics.TheapprovalofRecommendationsbytheMembersoftheITU-TiscoveredbytheprocedurelaiddowninWTSCResolution No. 1 (Geneva, October, 1997).ThissecondeditionofITU-TRecommendationJ.83waspreparedbyITU-TStudyGroup9(1997-2000)andincorporatesAmendment1andAmendment2approvedundertheWTSCResolutionNo.1procedureonthe17thofOctober 1996 and the 22nd of April 1997 respectively.___________________NOTEIn this Recommendation, the expression "Administration" is used for conciseness to indicate both a telecommunicationadministration and a recognized operating agency.ITU1997All rights reserved. No part of this publication may be reproduced or utilized in any form or by any means, electronic ormechanical, including photocopying and microfilm, without permission in writing from the ITU.ii Recommendation J.83 (04/97)CONTENTSPage1 Scope.............................................................................................................................................................. 12 References ...................................................................................................................................................... 13 Terms and definitions..................................................................................................................................... 14 Symbols and abbreviations............................................................................................................................. 24.1 Symbols ............................................................................................................................................ 24.2 Abbreviations.................................................................................................................................... 25 Digital multi-programme systems for cable distribution................................................................................ 3AnnexADigital multi-programme System A..................................................................................................... 5A.1 Introduction ...................................................................................................................................... 5A.2 Cable system concept........................................................................................................................ 5A.3 MPEG-2 transport layer.................................................................................................................... 7A.4 Framing structure.............................................................................................................................. 7A.5 Channel coding................................................................................................................................. 8A.6 Byte to symbol mapping................................................................................................................... 10A.7 Modulation........................................................................................................................................ 11A.8 Baseband filter characteristics .......................................................................................................... 13AnnexBDigital multi-programme System B..................................................................................................... 14B.1 Introduction ...................................................................................................................................... 14B.2 Cable system concept........................................................................................................................ 15B.3 MPEG-2 transport layer.................................................................................................................... 15B.4 MPEG-2 transport framing............................................................................................................... 15B.5 Forward error correction................................................................................................................... 20B.6 Modulation and demodulation.......................................................................................................... 33AnnexCDigital multi-programme System C..................................................................................................... 35C.1 Introduction ...................................................................................................................................... 35C.2 Cable system concept........................................................................................................................ 35C.3 MPEG-2 transport layer.................................................................................................................... 37C.4 Framing structure.............................................................................................................................. 37C.5 Channel coding................................................................................................................................. 38C.6 Modulation........................................................................................................................................ 39AnnexDDigital multi-programme System D..................................................................................................... 43D.1 Introduction ...................................................................................................................................... 43D.2 Cable system concept........................................................................................................................ 43D.3 MPEG-2 transport layer.................................................................................................................... 44D.4 Framing structure.............................................................................................................................. 44D.5 Channel coding................................................................................................................................. 46D.6 Modulation........................................................................................................................................ 51D.7 16-VSB cable receiver ...................................................................................................................... 52D.8 Other VSB modes ............................................................................................................................. 52AppendixIBibliography..................................................................................................................................... 60Recommendation J.83 (04/97) iiiSUMMARYThis Recommendation "Digital multi-programme systems for television, sound and data services for cable distribution"coversthedefinitionoftheframingstructure,channelcodingandmodulationfordigitalmulti-programmesignalsfortelevision, sound and data services distributed by cable networks.This Recommendation has four Annexes (A, B, C and D), that provide the specifications for the four digital televisioncablesystemssubmittedtotheITU-T.ThisreflectsthefactthatstandardizationofdigitalcabletelevisionsystemsisbeingaddressedforthefirsttimebytheITU-Tandthatanumberofsystemshadbeendevelopedandprovisionallyimplemented when this standardization effort was undertaken by the ITU.ThisRecommendationrecommendsthatthoseimplementingnewdigitalmulti-programmeservicesonexistingandfuturecablenetworksshoulduseoneofthesystemswhoseframingstructure,channelcodingandmodulationarespecified in Annexes A, B, C and D.INTRODUCTIONThedevelopmentofnewdigitaltechnologyisnowreachingthepointatwhichitisevidentthattheyenabledigitalsystemstooffersignificantadvantages,incomparisonwithconventionalanaloguetechniques,intermsofvisionandsoundquality,spectrumandpowerefficiency,serviceflexibility,multimediaconvergenceandpotentiallylowerequipment costs. Moreover, the use of cable distribution for the delivery of video and audio signals to individual viewersandlistenersiscontinuallygrowing,andhasalreadybecomethedominantformofdistributioninmanypartsoftheworld. It is also evident that these potential benefits can best be achieved through the economies of scale resulting fromthewidespreaduseofdigitalsystemsdesignedtobeeasilyimplementableonexistinginfrastructureandwhichtakeadvantage of the many possible synergies with related audiovisual systems.AdministrationsandprivateoperatorsplanningtheintroductionofdigitalcabletelevisionservicesareencouragedtoconsidertheuseofoneofthesystemsdescribedinAnnexesA,B,CandD,andtoseekopportunitiesforfurtherconvergence, rather than developing a different system based on the same technologies.This second edition of this Recommendation incorporates Amendment 1 and Amendment 2. These amendments broughtthe following changes with respect to the first edition of the Recommendation:a) In Annex B there is now a specification for 256-QAM;b) InAnnexB,twodistinctoperatingmodesofinterleavingcapabilityarespecified,calledlevel1andlevel 2.Level1isspecifiedfor64-QAMtransmissiononlyandthismodealreadyexistedinthefirstedition of Annex B. Level 2 encompasses 64-QAM and 256-QAM transmission, and for both modulationschemes is capable of supporting variable interleaving.c) In the first edition of Annex D, 24 bits were identified which determined the VSB mode for the data in theframeandtwosuchmodesweredefined:16-VSBCableand8-VSBTerrestrial(trelliscoded).Inthissecond edition, three other VSB modes are defined, i.e. 2-VSB, 4-VSB and 8-VSB.Table1/J.83hasbeenupdatedtotakeaccountoftheseextensions.Inaddition,anewAppendixIcontainingashortBibliography has been added.Recommendation J.83 (04/97) 1Recommendation J.83Recommendation J.83 (04/97)DIGITALMULTI-PROGRAMMESYSTEMSFORTELEVISION,SOUNDANDDATASERVICESFORCABLEDISTRIBUTION(Geneva, 1995; revised in 1997)1 ScopeThe scope of this Recommendation is the definition of the framing structure, channel coding and modulation for digitalmulti-programmetelevision,soundanddatasignalsdistributedbycablenetworks(e.g.CATVsystems)possiblyinfrequency-divisionmultiplex.AseparateRecommendationdefinesthetransmissioncharacteristicsfordigitalmulti-programme signals distributed through SMATV networks.NOTE The system input is specified to be the MPEG-2 transport layer; this provides some ancillary data capacity in theforwardchannel,whichcanbeusedtoaccommodatetheneedsofinteractiveservices(adescriptionoftheprovisionandcharacteristics of the return channel is outside the scope of this Recommendation).Beinghighlyflexible,theMPEG-2transportlayercanbeconfiguredtodeliveranydesiredmixoftelevision,soundanddata signals (with sound either related or unrelated to the video signal content, and at various possible levels of quality). The transportlayercanevenbetotallydevotedtothedeliveryofsoundprogramming,althoughitmaynotnecessarilybeoptimisedforthisapplication.ThespecificcaseofthedeliveryofamultiplexonlycontainingsoundsignalsmaybeaddressedinafutureRecommendation.This Recommendation is intended to ensure that the designers and operators of cable distribution (e.g. CATV) networkscarryingmulti-programmesignalswillhavetheinformationtheyneedtobeabletoestablishandmaintainfullysatisfactorynetworks.Italsoprovidestheinformationneededbythedesignersandmanufacturersofequipment(including receivers) for digital multi-programme signals distributed by cable networks.2 ReferencesThefollowingITU-TRecommendationsandotherreferencescontainprovisionswhich,throughreferenceinthetext,constituteprovisionsofthisRecommendation.Atthetimeofpublication,theeditionsindicatedwerevalid.AllRecommendationsandotherreferencesaresubjecttorevision;allusersofthisRecommendationarethereforeencouragedtoinvestigatethepossibilityofapplyingthemostrecenteditionoftheRecommendationsandotherreferences listed below. A list of currently valid ITU-T Recommendations is regularly published.[1] ITU-RRecommendationBO.1211(1995),Digitalmulti-programmeemissionsystemsfortelevision,soundanddata services for satellites operating in the 11/12 GHz frequency range.[2] ITU-TRecommendationH.222.0(1995)|ISO/IEC13818-1:1996,InformationtechnologyGenericcodingofmoving pictures and associated audio information: Systems.3 Terms and definitionsNo unconventional terms or definitions are used in this Recommendation.2 Recommendation J.83 (04/97)4 Symbols and abbreviations4.1 SymbolsThis Recommendation uses the following symbols. Roll-off factorAk, BkMost Significant Bits at the output of the Byte to m-tuple converterbyte Eight bitsf0Channel centre frequencyfNNyquist frequencyg(x) RS code generator polynomialG(256)RS primitive field generator polynomialG(16)Randomizer generator polynomialI Interleaving depth (bytes)I, Q In-phase, Quadrature phase components of the modulated signalj Branch indexk Number of bytes mapped into n symbolsm Power of 2m-level QAM: 4,5,6 for 16-QAM, 32-QAM, 64-QAM, respectivelyM Convolutional interleaver branch depth for j = 1, M = N/Ims millisecondn Number of symbols mapped from k bytesN Error protected frame length (bytes)p(x) RS field generator polynomialPN(x) Pseudorandom sequence, identified by the number following the symbolrmIn-band ripple (dB)R Randomized sequenceRsSymbol rate corresponding to bilateral Nyquist bandwidth of modulated signalRuUseful bit rate after MPEG-2 transport multiplexerRuBit rate after RS outer coderq Number of bits: 2,3,4 for 16-QAM, 32-QAM, 64-QAM, respectivelyT Number of bytes which can be corrected in RS error-protected packetTsSymbol period4.2 AbbreviationsThis Recommendation uses the following symbols.ATM Asynchronous Transfer ModeBB BaseBandBER Bit Error Ratiobps Bits per secondCATV Community Antenna TelevisionC/N Carrier to Noise ratioRecommendation J.83 (04/97) 3DTVC Digital Television by CableFEC Forward Error CorrectionFIFO First In First OutHEC Header Error ControlHEX HexadecimalIF Intermediate FrequencyIRD Integrated Receiver DecoderLSB Least Significant BitMMDS Multichannel Multipoint Distribution SystemMPEG Motion Picture Experts GroupMSB Most Significant BitMUX MultiplexP ParityPDH Plesiochronous Digital HierarchyPN Pseudorandom Noiseppm Parts per millionPRBS PseudoRandom Binary SequenceQAM Quadrature Amplitude ModulationQEF Quasi Error FreeRF Radio FrequencyRS Reed-SolomonSMATV Satellite Master Antenna TelevisionSNR Signal-to-Noise Ratiosps Symbols per secondSync Synchronizing signalTBD To Be DeterminedTDM Time Division MultiplexTS Transport StreamVLSI Very Large Scale IntegrationVSB Vestigial SideBandXOR Exclusive OR2-VSB 2 level VSB4-VSB 4 level VSB8-VSB 8 level VSB16-VSB 16 level VSB5 Digital multi-programme systems for cable distributionIt is recommended that those implementing new digital multi-programme services on existing and future cable networksshould use one of the systems whose framing structure, channel coding and modulation are specified in Annexes A, B, Cand D. The specifications are compared in Table 1, indicating common features.4 Recommendation J.83 (04/97)Table 1/J.83 Comparison of specifications in summary form indicating common featuresItem Annex B Annex A Annex C Annex DInput signals Modified MPEG-2 transportstream. A parity checksum issubstituted for the sync byte,supplying improved packetdelineation functionality,and error detectioncapability independent of theFEC layer. (See B.4.)MPEG-2 transport Stream(See A.3, C.3, D.3.)Framing structure An FEC frame consists ofa 42- or 40-bit sync trailerfollowing 60 or 88 RSblocks, with each blockcontaining 128 symbols.An RS symbol consists of7 bits. Thus, there is atotal 53 802 or 78 888 bitsin an FEC frame for 64- or256-QAM respectively.(See B.5.3.)The framing organization is based on theMPEG-2 transport packet structure.(See A.4, C.4, D.4.)Randomization The 3-word polynominal forthe PRS:x3 + x + 3 over GF 128.(See B.5.4.)The 15-bit polynominal for the PRBS:1 + x14 + x15(See A.5.1, C.5.1.)The 16-bit poly-nominal for thePRBS:1 + x + x3 + x6+ x7 + x11 + x12+ x13 + x16.(See D.5.1.)ChannelcodingFEC Concatenated coding, RS(128, 122) GF 128 withconvolutional coding.(See B.5.)RS (204, 188) GF 256(See A.5.2, C.5.2.)RS (207, 187)GF 256(See D.5.2.)Interleaving Convolutional interleavingdepth:I 128,64,32,16,8J 1,2,3,4,5,6,7,8,16.(See B.5.2.)Convolutional interleaving, depth:I 12.(See A.5.3, C.5.3.)Convolutionalinterleaving,depth: I 52.(See D.5.3.)Byte to symbolmappingSee B.5.5. See A.6, C.6.1. See D.6.1.Differential coding See B.5.5. See A.6, C.6.2. NoneTrellis coding See B.5.5. NoneBandwidth 6 MHz 8 MHz 6 MHzModu-lationConstellation 64- or 256-QAMFigure B.18 or B.1916-, 32-, 64-QAMFigure A.764-QAMFigure C.72-, 4-, 8-, 16-VSBRoll-off factor 18% or 12% for 64- or256-QAM respectively.See B.6.1.15%See A.713%See C.6.411.5%See D.6.3Baseband filtercharacteristicsTable B.2 Figure A.8 Figure C.8 Figure D.11Recommendation J.83 (04/97) 5AnnexADigital multi-programme System AA.1 IntroductionThisAnnexderivesfromworkdoneondigitaltelevisionsatellitebroadcastinginEurope;itdescribestheframingstructure,channelcodingandmodulation(denoted"theSystem"forthepurposesofthisAnnex)fordigitalmulti-programme television distribution by cable. This System can be used transparently with the modulation/channel codingsystemusedfordigitalmulti-programmetelevisionbysatellite(seeReference[1]).TheSystemallowsforfurtherevolution as technology advances.The System is based on MPEG-2 (see Reference [2] as regards source coding and transport multiplexing. It is based onQuadratureAmplitudeModulation(QAM).Itallowsfor16-,32-,or64-QAMconstellationsandpermitsfutureextension to higher constellations, such as 128-QAM and 256-QAM.TheSystemFECisdesignedtoimprovetheBitErrorRatio(BER)from104toarangeof1010 to1011,ensuring"Quasi Error Free" (QEF) operation with approximately one uncorrected error event per transmission hour.A.2 Cable system conceptThe cable system shall be defined as the functional blockofequipmentperformingtheadaptationofthebasebandTVsignalstothecablechannelcharacteristics(seeFigureA.1).Inthecablehead-end,thefollowingTVbasebandsignalsources can be considered: satellite signal(s); contribution link(s); local programme source(s).The following processes shall be applied as shown in Figure A.1.A.2.1 Baseband interfacing1) and syncThis unit shall adapt the data structure to the format of the signal source. The framing structure shall be in accordancewith MPEG-2 transport layer including sync bytes.A.2.2 Sync 1 inversion and randomizationThisunitshallinverttheSync1byteaccordingtotheMPEG-2framingstructure,andrandomizesthedatastreamforspectrum shaping purposes.A.2.3 Reed-Solomon (RS) coderThisunitshallapplyashortenedReed-Solomon(RS)codetoeachrandomizedtransportpackettogenerateanerror-protected packet. This code shall also be applied to the Sync byte itself.A.2.4 Convolutional interleaverThisunitshallperformadepthI12convolutionalinterleavingoftheerror-protectedpackets.Theperiodicityofthesync bytes shall remain unchanged.A.2.5 Byte to m-tuple conversionThis unit shall perform a conversion of the bytes generated by the interleaver into QAM symbols.A.2.6 Differential encodingIn order to get a rotation-invariant constellation, this unit shall apply a differential encoding of the two Most SignificantBits (MSBs) of each symbol.1)Interfaces are not part of this Recommendation.6Recommendation J.83 (04/97)T0902590-95/d01Baseband and interface to:Local MPEG-2 programme sources,Contribution links,Remultiplexers, etc.Data(Note)ClockBBPhysicalinterfaceSync 1inversion&randomi-zationReed-SolomonCoder(204,188)Convol.Inter-leaver1 =12 bytesClock & sync generatorRFphysicalinterface&QAMdemodu-latorMatchedfilter& EqualizerDifferen-tialdecoderSymboltobytemappingConvo-lutionaldeinter-leaverReed-SolomondecoderSync 1inversion&EnergydispersalremovalBBPhysicalinterfaceFrom RFcablechannelCarrier & clock & sync recoveryNOTE MPEG-2 transport MUX packets.Figure A.1/J.83 Conceptual block diagram of elements at the cable head-end and receiving site8 8mm 8 8 8 8Cable head-end8 m mData (Note)ClockBytetom-tupleconver-sionTo RFcablechannelQAMModulator&IFphysicalinterfaceDifferentialencodingFIGURE A.1/J.83...[D01] A LITALIENNERecommendation J.83 (04/97) 7A.2.7 QAM modulation and physical interfaceThis unit performs a square-root raised cosine filtering of the I and Q signals prior to QAM modulation. This is followedby interfacing the QAM modulated signal to the Radio Frequency (RF) cable channel.A.2.8 Cable receiverA System receiver shall perform the inverse signal processing, as described for the modulation process above, in order torecover the baseband signal.A.3 MPEG-2 transport layerThe MPEG-2 transport layer is defined in Reference [2]. The transport layer for MPEG-2 data is comprised of packetshaving188bytes,withonebyteforsynchronizationpurposes,threebytesofheadercontainingserviceidentification,scrambling and control information, followed by 184 bytes of MPEG-2 or auxiliary data.A.4 Framing structureTheframingorganizationshallbebasedontheMPEG-2transportpacketstructure.TheSystemframingstructureisshown in Figure A.2.T0902600-95/d02187 BytesSync1 byteR187 BytesSync 2R187 BytesR187 BytesR187 Bytes204 bytesR187 BytesRS (204,188, 8)203 BytesSync 1orSync n203 BytesPRBS period = 1503 bytesa) MPEG-2 transport MUX packetb) Randomized transport packets: Sync bytes and Randomized Sequence Rc) Reed-SoIomon RS (204,188, T = 8) error-protected packetd) InterIeaved Frames; InterIeaving depth I = 12 bytesFigure A.2/J.83 Framing structureSync 1Sync 1orSync nSync 1orSync nSync 1orSync nSync 8Sync 1Not randomized complemented sync byte.Sync nNot randomized sync byte, n = 2, 3, ..., 8.Sync 1FIGURE A.2/J.83...[D02] = 14.8 CM8 Recommendation J.83 (04/97)A.5 Channel codingToachievetheappropriateleveloferrorprotectionrequiredforcabletransmissionofdigitaldata,anFECbasedonReed-Solomonencodingshallbeused.IncontrasttotheBaselineSystemforsatellitedescribedinReference[1],noconvolutional coding shall be applied to cable transmission. Protection against burst errors shall be achieved by the useof byte interleaving.A.5.1 Randomization for spectrum shapingTheSysteminputstreamshallbeorganizedinfixedlengthpackets(seeFigureA.2),followingtheMPEG-2transportmultiplexer. The total packet length of the MPEG-2 transport MUX packet is 188 bytes. This includes 1 sync-word byte(i.e. 47HEX). The processing order at the transmitting side shall always start from the MSB (i.e. 0) of the sync word-byte(i.e. 01000111).In order to comply with the System for satellite (see Reference [1]) and to ensure adequate binary transitions for clockrecovery,thedataattheoutputoftheMPEG-2transportmultiplexshallberandomizedinaccordancewiththeconfiguration depicted in Figure A.3.The polynomial for the PseudoRandom Binary Sequence (PRBS) generator shall be:1+x14+x15Loading of the sequence "100101010000000" into the PRBS registers, as indicated in Figure A.3, shall be initiated at thestart of every eight transport packets. To provide an initialization signal for the descrambler, the MPEG-2 sync byte ofthe first transport packet in a group of eight packets shall be bit wise inverted from 47HEX to B8HEX.The first bit at the output of the PRBS generator shall be applied to the first bit of the first byte following the invertedMPEG-2syncbyte(i.e.B8HEX).Toaidothersynchronizationfunctions,duringtheMPEG-2syncbytesofthesubsequent7transportpackets,thePRBSgenerationcontinues,butitsoutputshallbedisabled,leavingthesebytesunrandomized. The period of the PRBS sequence shall therefore be 1503 bytes.The randomization process shall be active also when the modulator input bit stream isnon-existent,orwhenitisnon-compliant with the MPEG-2 transport stream format (i.e. 1 sync byte + 187 packet bytes). This is to avoid the emissionof an unmodulated carrier from the modulator.A.5.2 Reed-Solomon codingFollowing the energy dispersal randomization process, systematic shortened Reed-Solomon encoding shall be performedon each randomized MPEG-2 transport packet, with T 8. This means that 8 erroneous bytes per transport packet can becorrected. This process adds 16 parity bytes to the MPEG-2 transport packet to give a codeword (204, 188).NOTE RS coding shall also be applied to the packet sync byte, either non-inverted (i.e. 47HEX) or inverted (i.e. B8HEX).Code Generator Polynomial: g(x)(x+0)(x+1)(x+2)...(x+15);where:02HEXField Generator Polynomial: p(x)x8+x4+x3+x2+1The shortened Reed-Solomon code shall be implemented by appending 51 bytes, all set to zero, before the informationbytes at the input of a (255, 239) encoder; after the coding procedure these bytes are discarded.Recommendation J.83 (04/97) 91 2 3 4 5 6 7 8 9 10 11 12 13 14 151 1 1 1 0 0 0 0 0 0 0 0 0 0 0T0902610-95/d03AND0 0 0 0 0 0 1 1 ....EnableClear/randomizeddata inputRandomized/de-randomizeddata outputInitialization sequenceXORFigure A.3/J.83 Scrambler/descrambler schematic diagramXORData input (MSB first) 1 0 1 1 1 0 0 0PRBS sequence ::........x1x1x0x0x0x0x0x0FIGURE A.3/J.83...[D03] = 10.8 CMA.5.3 Convolutional interleavingFollowing the scheme of Figure A.4, convolutional interleaving with depth I 12 shall be applied to the error-protectedpackets [see Figure A.2 c)]. This results in an interleaved frame [see Figure A.2 d)].TheconvolutionalinterleavingprocessshallbebasedontheForneyapproachwhichiscompatiblewiththeRamseytype IIIapproach,withI12.TheInterleavedFrameshallbecomposedofoverlappingerror-protectedpacketsandshall be delimited by MPEG-2 sync bytes (preserving the periodicity of 204 bytes).The interleaver may be composed of I 12 branches, cyclically connected to the input byte-stream by the input switch.EachbranchshallbeaFirstInFirstOut(FIFO)shiftregister,withdepth(Mj)cells(whereM17N/I,N204error-protected frame length, I 12 interleaving depth, j branch index). The cells of the FIFO shall contain 1 byte,and the input and output switches shall be synchronized.For synchronization purposes, the sync bytes and the inverted sync bytes shall be always routed in the branch "0" of theinterleaver (corresponding to a null delay).NOTE Thede-interleaverissimilar,inprinciple,totheinterleaver,butthebranchindexesarereversed(i.e.j0corresponds to the largest delay). The de-interleaver synchronization can be carried out by routing the first recognized sync byte in the"0" branch.10 Recommendation J.83 (04/97)T0902620-95/d04Sync word routeSync word route1 byte perposition 1 byte perpositionFIFO shift registerDe-interleaver I = 12Interleaver I = 12Figure A.4/J.83 Conceptual diagram of the convolutional interleaver and de-interleaver89101110980 0012311012317 = M17 217 317 1117 1117 317 217 = M11 = I 1 11 = I 1FIGURE A.4/J.83...[D04] = 7.2 CMA.6 Byte to symbol mappingAfter convolutional interleaving, an exact mapping of bytes into symbols shall be performed. The mapping shall rely onthe use of byte boundaries in the modulation system.In each case, the MSB of symbol Z shall be taken from the MSB of byte V.Correspondingly, the next significant bit of the symbol shallbetakenfromthenextsignificantbitofthebyte.Forthecase of 2m-QAM modulation, the process shall map k bytes into n symbols, such that:8 knmThe process is illustrated for the case of 64-QAM (where m 6, k 3 and n 4) in Figure A.5:T0902630-95/d05b7 b6b1 b0b7 b6 b5 b4b3 b2 b1 b0b5 b4 b3 b2 b1 b0b5 b4 b3 b2 b1 b0b5 b4 b3 b2 b1 b0b5 b4 b3 b2 b1 b0b5 b4 b3 b2 b1 b0b7 b6 b5 b4 b3 b2From interleaveroutput (bytes)Byte V Byte V + 1 Byte V + 2Symbol Z Symbol Z + 1 Symbol Z + 2 Symbol Z + 3To differentialencoder(6-bit symbols)MSB LSBNOTE 1 b0 shall be understood as being the Least Significant Bit (LSB) of each byte or m-tuple.NOTE 2 In this conversion, each byte results in more than one m-tuple, labelled Z, Z + 1, etc. with Z beingtransmitted before Z + 1.Figure A.5/J.83 Byte to m-tuple conversion for 64-QAMFIGURE A.5/J.83...[D05] = 7.6 CMRecommendation J.83 (04/97) 11The two most significant bits of each symbol shall then be differentially coded in order to obtain a /2 rotation-invariantQAM constellation. The differential encoding of the two MSBs shall be given by the following expression:I = (AB ) A I A B A Qk k k k k k k k k + ( ) ( ) ( )1 1Q = (AB ) B Q A B B Ik k k k k k k k k + ( ) ( ) ( )1 1Figure A.6 gives an example of implementation of byte to symbol conversion.T0903190-95/d06Bytetom-tupleconversionDifferentialencodingMappingFromconvolutionalinterleaver2 for 16-QAMq= 3 for 32-QAM 4 for 64-QAMFigure A.6/J.83 Example of an implementation of the byte to m-tuple conversionand the differential encoding of the two MSBsq bits (bq 1, ..., b0)8IQQkIkAk = MSBBk = bqFIGURE A.6/J.83...[D06] = 5.7 CMA.7 ModulationThemodulationoftheSystemshallbeQuadratureAmplitudeModulation(QAM)with16,32,or64pointsintheconstellation diagram.The System constellation diagrams for 16-QAM, 32-QAM and 64-QAM are given in Figure A.7.As shown in Figure A.7, the constellation points in Quadrant 1 shall be converted to Quadrants 2, 3 and 4 by changingthe two MSB (i.e. Ik and Qk) and by rotating the q LSBs according to the rule given in Table A.1.12 Recommendation J.83 (04/97)I IQQQI000000 000001 000101 000100000010 000011 000111 000110001010 001011 001111 001110001000 001001 001101 001100100000 100010 101010100011 101011100101 100111 101111100100 100110 101110100001110101 110001 010000110111 110011 110010111111 111011 111010111101 111001 111000010010 011010 011000010001 010011 011011 011001010101 010111 011111 011101010100 010110 011110 011100110000 110100101001101101101100110110111110111100101000T0902640-95/d070000 10001100 010000011001110101010011111110110111001010101110011000000 00001 0001100111 00101 0010000110 00010 10011 1011110010 10101 1000110110 10100 1000011011 11001 1100011111 11101 1110011010 11110 01011 0111101001 01101 0101001110 01100 01000IkQk= 10 IkQk = 00IkQk= 11 IkQk= 01IkQk= 10 Ik Qk= 00IkQk= 11 IkQk= 01IkQk= 10 IkQk = 00IkQk= 11 Ik Qk= 00NOTE Ik Qk are the two MSBs in each quadrant.Figure A.7/J.83 Constellation diagrams for 16-QAM, 32-QAM and 64-QAM32-QAM 16-QAM64-QAMFIGURE A.7/J.83...[D07] = 21.7 CMRecommendation J.83 (04/97) 13Table A.1/J.83 Conversion of constellation points of quadrant 1 to other quadrantsof the constellation diagram given in Figure A.7Prior to modulation, the I and Q signals shall be square-root raised-cosine filtered. The roll-off factor shall be 0.15.The square-root raised cosine filter shall have a theoretical function defined by the following expression:H(f) = 1 for f | fN| ( ) < 1 H(f) =12 ff f |for f f fNNN N+

1]1'; +12 21 11 2sin|( ) | | ( )/ H(f) = for f | fN0 1 | ( ) > + where:fTRNss 12 2 is the Nyquist frequency and roll-off factor 0.15.The transmitter filter characteristic is given in A.8.A.8 Baseband filter characteristicsThe template given in Figure A.8 shall be used as a minimum requirement for hardware implementation of the Nyquistfilter. This template takes into account not only the design limitations of the digital filter, but also the artefacts comingfrom the analogue processing components of the System (e.g. D/A conversion, analogue filtering, etc).The value of in-band ripple rm in the pass-band up to (1 ) fN as well as at the NyquistfrequencyfNshallbelowerthan 0.4 dB. The out-of-band rejection shall be greater than 43 dB.The filter shall be phase-linear with the group delay ripple 0.1 Ts (ns) up to fN,where:TRss1 is the symbol period.NOTE The values for in-band ripple and out-of-band rejection given in this Annex are subject to further study.Quadrant MSBs LSBs rotation1 002 10 +/23 11 +4 01 +3/2NOTE Receivers shall support at least 64-QAM modulation.14 Recommendation J.83 (04/97)T0902650-95/d08FrequencyOut-of-band rejection 43 dB3.01 dBfNyquist frequencyNIn-band ripple rm 0.4 dBFigure A.8/J.83 Half-Nyquist baseband filter amplitude characteristicsf0fN(1 ) fN(1 + )fNrmrmH(f)0 dBFIGURE A.8/J.83...[D08] = 10 CMAnnexBDigital multi-programme System BB.1 IntroductionThisAnnexdescribestheframingstructure,channelcoding,andchannelmodulationforadigitalmulti-servicetelevisiondistributionsystemthatisspecifictoacablechannel.Thesystemcanbeusedtransparentlywiththedistribution from a satellite channel, as many cable systems are fed directly from satellite links. The specification coversboth64-and256-QAM.Mostfeaturesofbothmodulationschemesarethesame.Wheretherearedifferences,thespecific details for each modulation scheme will be covered.Thedesignofthemodulation,interleavingandcodingisbasedupontestingandcharacterizationofcablesystemsinNorthAmerica.ThemodulationisQuadratureAmplitudeModulationwitha64-pointsignalconstellation(64-QAM)andwitha256-pointsignalconstellation(256-QAM),transmitterselectable.TheForwardErrorCorrection(FEC)isbasedonaconcatenatedcodingapproachthatproduceshighcodinggainatmoderatecomplexityandoverhead.Concatenated coding offers improved performance over a block code, at a similar overall complexity. The system FEC isoptimized for quasi error free operation at a threshold output error event rate of one error event per 15 minutes.The data format input to the modulation and coding is assumed to be MPEG-2 transport. However, the method used forMPEGsynchronizationisdecoupledfromFECsynchronization.Forexample,thisenablesthesystemtocarryATMpacketseasilywithoutinterferingwithATMsynchronization.Infact,ATMsynchronizationmaybeperformedbydefined ATM synchronization mechanisms.Therearetwomodessupported:Mode1hasasymbolrateof5.057Msymbols/sandMode2hasasymbolrateof5.361 Msymbols/s.Typically,Mode1willbeusedfor64-QAMandMode2willbeusedfor256-QAM.Thesystemwill be compatible with future implementations of higher data rate schemes employing higher order QAM extensions.Recommendation J.83 (04/97) 15B.2 Cable system conceptChannel coding and transmission are specific to a particular medium or communication channel. The expected channelerror statistics and distortion characteristics are critical in determining the appropriate error correction and demodulation.The cable channel, including optical fiber, is primarily regarded as a bandwidth-limited linear channel, with a balancedcombinationofwhitenoise,interference,andmulti-pathdistortion.TheQuadratureAmplitudeModulation(QAM)techniqueused,togetherwithadaptiveequalizationandconcatenatedcodingiswellsuitedtothisapplicationandchannel.ThebasiclayeredblockdiagramofcabletransmissionprocessingisshowninFigureB.1.Thefollowingsubclausesdefine these layers from the "outside" in, and from the perspective of the transmit side.T0903400-96/d09Transmitter ReceiverChannelMPEG framingFEC encoderQAM modulatorQAM demodulatorFEC decoderMPEG framingMPEG-2 transportMPEG-2 transportFigure B.1/J.83 Cable transmission block diagramFIGURE B.1/J.83...[D09] = 7.5 CMB.3 MPEG-2 transport layerThe MPEG-2 transport layer is defined in Reference [2]. The transport layer for MPEG-2 data is comprised of packetshaving188bytes,withonebyteforsynchronizationpurposes,threebytesofheadercontainingserviceidentification,scrambling and control information, followed by 184 bytes of MPEG-2 or auxiliary data.B.4 MPEG-2 transport framingThe MPEG transport framing is the outermost layer of processing. It is provided as a robust means of delivering MPEGpacketsynchronizationtothereceiveroutput.ThisprocessingblockreceivesanMPEG-2transportdatastreamconsisting of a continuous stream of fixed length 188 byte packets. This data stream is transmitted in serial fashion, MSBfirst. The first byte of a packet is specified to be a sync byte having a value of 47HEX.Thesyncbyteisintendedforthepurposeofpacketdelineation.Thecabletransmissionsystemhasincorporatedanadditional layer of processing to provide an additional functionality by utilizing the information bearing capacity of thissync byte. A parity checksum which is a coset of an FIR parity check linear block code is substituted for this sync byte,supplying improved packet delineation functionality, and error detection capability independent of the FEC layer.16 Recommendation J.83 (04/97)Theparitychecksumiscomputedovertheadjacent187bytes,whichconstitutetheimmediatelyprecedingMPEG-2packet contents (minus sync byte). It is then possible to support simultaneous packet synchronization and error detection.The decoder computes a sliding checksum on the serial data stream, using the detection of a valid code word to detectthe start of packet. Once a locked alignment condition is established, the absence of a valid code word at the expectedlocation will indicate a packet error. The error flag of the previous packet may optionally be set as the data is passed outof the decoder. The normal sync word must be re-inserted in place of the checksum to provide a standard MPEG-2 datastream as an output.Thesyndromeiscomputedbypassingthe1496payloadbitsthroughaLinearFeedbackShiftRegister(LFSR)asdescribed by the following equation:f(x)=[1+x1497b(x)]/g(x)where:g(x)1+x+x5+x6+x8;andb(x)1+x+x3+x7.This computational structure is illustrated in Figures B.2 and B.3. All addition operations are assumed to be modulo 2.Foranencodeoperation,theLFSRisfirstinitializedsothatallmemoryelementscontainzerovalue.The1496bitswhich constitute the MPEG-2 transport stream packet payload are then shifted into the LFSR. The encoder input is set tozeroafterthe1496databitsarereceived,andeightadditionalshiftsarerequiredtosequentiallyoutputtheeightcomputedsyndromebits.This8-bitresultmustthenbepassedthroughanadditionalFIRfilteringfunctiong(x)(initialized to an all-zeros state prior to introduction of the 8 syndrome bits) to generate an encoder checksum. An offsetof67HEXisaddedtothischecksumresultforimprovedautocorrelationproperties,andcausesa47HEXresulttobeproduced during a syndrome decode operation when a valid code word is present.The final 8-bit checksum with addedoffset is transmitted MSB first following the 1496 payload bits to implement a systematic encoder.Aparitycheckmatrixmaybeusedbythedecodertoidentifyavalidchecksum.Asyndromegenerator,asshowninFigureB.3,mayalsobeemployedforthispurpose.Thecodehasbeendesignedsuchthatwhentheappropriate188 bytesofthemodifiedMPEG-2transportstreampacket(whichincludestheassociatedchecksum)aremultipliedwith the parity check matrix, a valid code word is indicated when the calculated product produces a 47HEX result. Eachofthe8 columnsoftheparitycheckmatrix"P"includesa1497bitvector,hereafterreferredtoas"C".Thisvectorisdefined in Figure B.4.Proceeding from the leftmost column of the matrix "P", the 1497-bit column "C" is duplicated in subsequent columns ofthe matrix "P", shifted down by one bit position. The bit positions unoccupied by the column data are filled with zeros,as illustrated in Figure B.5.Notethatthechecksumiscalculatedbasedontheprevious187bytesandnotthe187bytesyettobereceivedbytheMPEG-2 sync decoder. This is in contrast to the conventional notion of an MPEG packet structure, in that the sync byteis usually described as the first byte of a received packet.The received vector "R" is the MPEG-2 data consisting of 187 bytes followed by the checksum byte, yielding a total of1504 bits. This "R" vector is multiplied (modulo 2) by the parity check "P" matrix, yielding an "S" vector whose lengthis 8-bits, as illustrated in Figure B.6.Recommendation J.83 (04/97) 17Z1B 0B0A1 1 0 1 1 1 0 0b0T0903410-96/d10Z1Z1Z1Z1Z1Z1Z1Z1497Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1Z1b1b2b3b4b5b6b7InputSwitch position A first 1496 shiftsSwitch position B last 8 shifts67HEX offset, MSB first(LSB) (MSB)Encoder checksum outputFigure B.2/J.83 Checksum generator for the MPEG-2 sync byte encoderFIGUREB.2/J.83...[D10] = 11.5T0903420-96/d11Z1Z1Z1Z1Z1Z1Z1Z1Z1497Z1Z1Z1Z1Z1Z1Z1InputDecodersyndromeoutputFigure B.3/J.83 Syndrome generator for the MPEG-2 sync decoderFIGUREB.3/J.83...[D11] = 5.518 Recommendation J.83 (04/97)b0f3857f97a50ddbeba0caa358c12da9a7ee67b2103926275688a47c05c778b361e70aff2f4a1bb7d7419546b1825b534fdccf6420724c4ead1148f80b8ef166c3ce15fe5e94376fae832a8d6304b6a69fb99ec840e4989d5a2291f0171de2c2879c2bfcbd286edf5d06551ac6096d4d3f733d9081c9313ab44523e02e3bc59b0f3857f97a50ddbeba0caa358c12da9a7ee67b2103926275688a47c05c778b361e70aff2f4a1bb7d7419546b1825b534fdccf6420724c4ead1148fT0903430-96/d12C = 1497 1 =C = 1497 1 =All entries are in hexadecimal format except where otherwise noted.1binaryFigure B.4/J.83 "C" column vector (replicated inside the parity check matrix)C0,0C1,0C2,0C1494,0C1495,0C1496,0FIGUREB.4/J.83...[D12] = 12Recommendation J.83 (04/97) 190000000000000000000000000000000000 00 00000000000000000000CCCCCCCCT0903440-96/d13"P" parity check matrix= 1504 8 = P1497 rows(bits)7 rows(bits)8 columnsFigure B.5/J.83 Structure of the parity check matrix "P"FIGUREB.5/J.83...[D13] = 14T0903450-96/d14=R "Vector"(Alignment window)1 1504P "Matrix"(Parity check)1504 8S "Vector"(Received checksum)1 8S = [0100 0111] = 0 47Figure B.6/J.83 Received MPEG-2 vector and parity check matrix multiplicationFIGUREB.6/J.83...[D14] = 5.520 Recommendation J.83 (04/97)A valid checksum is indicated when S [0100, 0111] 47HEX.ForcarriageoftransportprotocolsotherthanMPEG-2Transport,e.g.ATM,thisouterlayerisremovedorbypassed.The FEC layer accepts and delivers data without any constraints on protocol. The framing section could be replaced withone appropriate to the alternative transport protocol if required by an application. All other portions of this specification(modulation,coding,interleaving)areimplementedasdescribedbelow.ForthecaseofATM,noframinglayerisrequired. The ATM HEC typically provides adequate packet framing and error detection. Isochronous ATM streams aretherefore carried transparently without overhead for MPEG or quasi-MPEG packet encapsulation.B.5 Forward error correctionThe Forward Error Correction (FEC) definition is composed of four processing layers, as illustrated in Figure B.7. TherearenodependenciesoninputdataprotocolinanyoftheFEClayers.FECsynchronizationisfullyinternalandtransparent. Any data sequence will be delivered from the encoder input to decoder output.T0903460-96/d15FEC encodingFEC decodingReed- Solomon encoderInter-leaverRan-domizerTrellisencoderChannelTrellisdecoderDe-ran-domizerDe-interleaverReed-SolomondecoderTrellis layerRandomization layerInterleaving layerReed-Solomon layerFigure B.7/J.83 Layers of processing in the FECFIGUREB.7/J.83...[D15] = 7.9The FEC section uses various types of error correcting algorithms and interleaving techniques to transport data reliablyover the cable channel. Reed-Solomon(RS)CodingProvidesblockencodinganddecodingtocorrectuptothreesymbolswithin an RS block. Interleaving Evenly disperses the symbols, protecting against a burst of symbol errors from being sentto the RS decoder. RandomizationRandomizesthedataonthechanneltoalloweffectiveQAMdemodulatorsynchronization. TrellisCodingProvidesconvolutionalencodingandwiththepossibilityofusingsoftdecisiontrellisdecoding of random channel errors.The following subclauses define these 4 layers.B.5.1 Reed-Solomon codingThe MPEG-2 transport stream is Reed-Solomon (RS) encoded using a (128, 122) code over GF(128). This code has thecapabilityofcorrectinguptot3symbolerrorsperRSblock.ThesameRScodeisusedforboth64-QAMand256-QAM. However, the FEC frame format is different for each modulation type, as described in a later subclause.Recommendation J.83 (04/97) 21The Reed-Solomon encoder implementation is described in this subclause. A systematic encoder is utilized to implementat3,(128,122)extendedReed-SolomoncodeoverGF(128).TheprimitivepolynomialusedtoformthefieldoverGF(128) is:p(x) x7+x3+1where:p() 0.The generator polynomial used by the encoder is:g(x)=(x + )(x + 2)(x + 3)(x + 4)(x + 5)x5+52 x4+116 x3+119 x2+61 x+15The message polynomial input to the encoder consists of 122, 7-bit symbols, and is described below:m(x)m121 x121+m120 x120+ ... +m1 x+m0This message polynomial is firstmultipliedbyx5,thendividedbythegeneratorpolynomialg(x)toformaremainder,described by the following:r(x)r4 x4+ r3 x3+r2 x2+r1 x+r0This remainder constitutesfiveparitysymbolswhicharethenaddedtothemessagepolynomialtoforma127-symbolcode word that is an even multiple of the generator polynomial.The generated code word is now described by the following polynomial:c(x)m121 x126+m120 x125+m119 x124+ ... +r4 x4+r3 x3+r2 x2+r1 x+r0A valid code word will have roots at the first through fifth powers of .An extended parity symbol (c_ ) is generated by evaluating the code word at the sixth power of .c_c(6)ThisextendedsymbolisusedtoformthelastsymbolofatransmittedReed-Solomonblock.Theextendedcodewordthen appears as follows: ^cxc(x)+c_m121 x127+m120 x126+ ... +m1 x7+m0 x6+r4 x5+r3 x4+r2 x3+r1 x2+r0 x+c_The structure of a Reed-Solomon block which illustrates the order of transmitted symbols output from the RS encoder isshown below:m121m120m119...m1m0r4r3r2r1r0c_ (order sent is left to right)B.5.2 InterleavingInterleaving is included in the modem between the RS block coding and the randomizer to enable the correction of burstnoise induced errors. In both 64-QAM and 256-QAM a convolutional interleaver is employed.22 Recommendation J.83 (04/97)Convolutional interleaving is illustrated in Figure B.8. At the start of an FEC frame defined in a subsequent subclause,theinterleavingcommutatorpositionisinitializedtothetop-mostbranchandincrementsattheRSsymbolfrequency,with a single symbol output from each position. With a convolutional interleaver, the RS code symbols are sequentiallyshiftedintothebankofIregisters(thewidthofeachregisteris7bitswhichmatchestheRSsymbolsize).EachsuccessiveregisterhasJsymbolsmorestoragethantheprecedingregister.Thefirstinterleaverpathhaszerodelay,the second has a J symbol period of delay, the third 2*J symbol periods of delay, and so on, upto the Ith path which has(I 1)*J symbol periods of delay. This is reversed for the de-interleaver in the Cable Decoder such that the net delay ofeach RS symbol is the same through the interleaver and de-interleaver. Burst noise in the channel causes a series of badsymbols. These are spread over many RS blocks by the de-interleaver such that the resultant symbol errors per block arewithin the range of the RS decoder correction capability.JJ JJ JJ JJ J J JJ JJJJ JJ JJ JJ J JJ JJJ JJ2131 2 I-31 2213I-2 I-1I-1II-2I-1II-3 I-2 I-1T0903470-96/d16I-2De-interleaver7 bitsChannel7 bitsCommutatorSymbol delay(I,J) = (128,1), (64,2), (32,4), (16,8), (8,16)(reduced interleaving modes)I = 128, J = 1 to 8(enhanced interleaving modes)InterleaverCommutator CommutatorFigure B.8/J.83 Interleaving functional block diagramCommutatorFIGUREB.8/J.83...[D16] = 9.1Withregardtointerleavingcapability,twodistinctoperatingmodesarespecified,hereafterreferredtoaslevel1andlevel 2.Level 1 is specified for 64-QAM transmission only. This mode accommodates the installed base of legacy 64-QAM-onlydigital set tops. While operating in level 1, a single interleaving depth will be supported; namely I 128, J 1.Level2shallencompass64-QAMand256-QAMtransmission,andwillforbothmodulationschemesbecapableofsupporting variable interleaving. This will include both enlarged and reduced interleaving depths relative to the nominal64-QAM (level 1) configuration. Four data bits are transmitted in-band during the FEC frame sync interval to convey theinterleaving parameters to the receiver for a given channel.TableB.1describestheinterleaverparametersforlevel1operation,withassociatedlatencyandburstprotection.Table B.2describesthedecodingofthe4-bitin-bandcontrolwordintotheIandJinterleavingparametersforlevel2operation, also with associated burst protection and latency.Recommendation J.83 (04/97) 23Table B.1/J.83 Level 1 interleavingTable B.2/J.83 Level 2 interleavingB.5.3 Frame synchronization sequenceThe frame synchronization sequence trailer delineates the FEC frame, providing synchronized RS coding, interleaving,andrandomization.Additionally,trellisgroupsfor256-QAMonlyarealignedwiththeFECframe.TheFECframingdoes not perform MPEG packet of trellis decoder synchronization. The RS block and 7-bit symbol structures are alignedwith the end of the frame for both 64- and 256-QAM.For 64-QAM, an FEC frame consists of a 42-bit sync trailer which is appended to the end of 60 RS blocks, with each RSblock containing 128 symbols. Each RS symbol consists of 7 bits. Thus, there is a total of 53 760 data bits and 42 framesynctrailerbitsinthisFECframe.Thefirst47-bitsymbolsoftheframesynctrailercontainthe28-bit"unique"synchronization pattern (1110101 0101100 0001101 1101100) or (75 2C 0D 6C)HEX. The remaining 2 symbols (14 bits)are utilized as follows: first 4 bits for interleaver mode control, and 10 bits are reserved and set to zero. The frame synctrailer is inserted by the encoder and detected at the decoder. The decoder circuits search for this pattern and determinethelocationoftheframeboundaryandinterleaverdepthmodewhenfound.TheFECframefor64-QAMisshowninFigure B.9.Controlword(4 bits)I (# of taps) J (increment)BurstprotectionLatencyxxxx 128 1 95 s 4.0 msControlword(4 bits)I (# of taps) J (increment)Burstprotection64-QAM/256-QAMLatency64-QAM/256-QAM0001 128 1 95 s /66 s 4.0 ms/2.8 ms0011 64 2 47 s /33 s 2.0 ms/1.4 ms0101 32 4 24 s /16 s 0.98 ms/0.68 ms0111 16 8 12 s /8.2 s 0.48 ms/0.33 ms1001 8 16 5.9 s /4.1 s 0.22 ms/0.15 ms1011 Reserved1101 Reserved1111 Reserved0000 128 1 95 s /66 s 4.0 ms/2.8 ms0010 128 2 190 s /132 s 8.0 ms/5.6 ms0100 128 3 285 s /198 s 12 ms/8.4 ms0110 128 4 379 s /264 s 16 ms/11 ms1000 128 5 474 s /330 s 20 ms/14 ms1010 128 6 569 s /396 s 24 ms/17 ms1100 128 7 664 s /462 s 28 ms/19 ms1110 128 8 759 s /528 s 32 ms/22 ms24 Recommendation J.83 (04/97)1110101 0101100 0001101 1101100 0000000000T0903480-96/d17TimeFEC frame(contains both "A" and "B" information)6 RS symbols sync trailer (42 bits)Reed-Solomonblock # 1Reed-Solomonblock # 2Reed-Solomonblock # 60122 symbols 122 symbols 122 symbols6 RS parity symbols 6 RS parity symbols6 RS parity symbolsUnique sync. pattern(75 2C 0D 6C)HEXFSYNC word2 RS symbols4-bitcontrol word10 reserved bitsFigure B.9/J.83 Frame packet format for 64-QAMFIGUREB.9/J.83...[D17] = 8For 256-QAM, an FEC frame consists of a 40-bit sync trailer which is appended to the end of 88 RS blocks, with eachRSblockcontaining128symbols.EachRSsymbolconsistsof7bits.Thus,thereisatotalof78 848databitsand40 frame sync trailer bits in this FEC frame. The 40-bit frame sync trailer is divided as follows: 32 bits are the "unique"synchronization pattern (0111 0001 1110 1000 0100 1101 1101 0100) or (71 E8 4D D4)HEX, 4 bits are a control wordwhich determine the size of the interleaver employed, and 4 bits are a reserved word which is set to zero. The FEC framefor 256-QAM is shown in Figure B.10.NotethatthereisnosynchronizationrelationshipbetweenthetransmittedRSblockandtransportdatapackets.Thus,MPEG-2 transport stream packet synchronization is obtained independently from RS frame synchronization. This keepsthe FEC and transport layers de-coupled and independent.0100 1101 1101 0100 1000 1110 0001 0111 0000T0903490-96/d18TimeFEC frameReed-Solomonblock # 1Reed-Solomonblock # 2Reed-Solomonblock # 8840-bit framesync trailer6 RSparity symbols6 RSparity symbolsReserved bitsUnique word(71 E8 4D D4)Frame sync trailer4-bitcontrolword122 symbols 122 symbols 122 symbols6 RSparity symbolsFigure B.10/J.83 Frame packet format for 256-QAMFIGURE B.10/J.83...[D18] = 9.8 CMRecommendation J.83 (04/97) 25B.5.4 RandomizationTherandomizershowninFigureB.11isthethirdlayerofprocessingintheFECblockdiagram.Therandomizerprovidesforevendistributionofthesymbolsintheconstellation,whichenablesthedemodulatortomaintainproperlock.TherandomizeraddsaPseudorandomNoise(PN)sequenceof7-bitsymbolsoverGF(128)(i.e.bit-wiseexclusive-OR) to the symbols within the FEC frame to assure a random transmitted sequence.For both 64- and 256-QAM, the randomizer is initialized during the FEC frame trailer, and is enabled at the first symbolafterthetrailer.Thusthetraileritselfisnotrandomized,andtheinitializedoutputvaluerandomizesthefirstdatasymbol.Initializationisdefinedaspre-loadingtotheallones statefortherandomizerstructureshowninFigureB.11.Therandomizer uses a linear feedback shift register specified by a GF(128) polynomial defined as follows:f (x)x3+x+3where:7+3+10.37 77T0903500-96/d19Z1Z1Z1f(x) = x3 + x + 3Data inThe Randomizer polynomialData outFigure B.11/J.83 Randomizer (7-bit symbol scrambler)FIGUREB.11/J.83...[D19] = 6.8B.5.5 Trellis coded modulationAs partoftheconcatenatedcodingscheme,trelliscodingisemployedfortheinnercode.ItallowstheintroductionofredundancytoimprovethethresholdSignal-to-NoiseRatio(SNR)byincreasingthesymbolconstellationwithoutincreasing the symbol rate. As such, it is more properly termed "trellis coded modulation."B.5.5.1 64-QAM modulation modeFor 64-QAM, the input to the trellis coded modulator is a 28-bit sequence of four, 7-bit RS symbols, which are labelledin pairs of Asymbols, and B symbols. A block diagram of a 64-QAM trellis coded modulator is shown in Figure B.12.All 28 bits are assigned to a trellis group, where each trellis group forms 5-QAM symbols, as shown in Figure B.13.26 Recommendation J.83 (04/97)Of the 28 input bits that form a trellis group, each of two groups of 4 bits of the differentially pre-coded bit streams in atrellisgroupareseparatelyencodedbyaBinaryConvolutionalCoder(BCC).EachBCCproduces5codedbits,asshown in Figure B.12. The remaining bits are sent to the mapper uncoded. This will produce an overall output of 30 bits.Thus, the overall code rate for 64-QAM trellis coded modulation is 14/15.The trellis group is formed from RS symbols as follows: For the "A" symbols, the RS symbols are read, from MSB toLSB, A10, A8, A7, A5, A4, A2, A1 and A9, A6, A3, A0, A13, A12, A11. The four MSBs of the second symbol are input totheBCC,onebitatatime,LSBfirst.Theremainingbitsofthesecondsymbolandallthebitsofthefirstsymbolareinputtothemapper,uncoded,LSBfirstonebitatatime.ThefourbitssenttotheBCCwillproduce5codedbitslabelled, U1, U2, U3, U4, U5. The same process is done for the "B" bits. The process can be seen in Figure B.12. With64-QAM,4RSsymbolsconvenientlyfitintoonetrellisgroup,andinthiscasethesyncwordmayoccupyeverybitposition within a trellis group.U5, U4, U3, U2, U1A9, A6, A3, A0B9, B6, B3, B0A13, A11, A8, A5, A2 A13, A10, A7, A4, A1 B13, B11, B8, B5, B2 B13, B10, B7, B4, B1 V5, V4, V3, V2, V1 C3C0C5C4C2C1WZXYT0903510-96/d20UncodedTime(1/2)Binaryconvolutionalcoder with(4/5 puncture)(1/2)Binaryconvolutionalcoder with(4/5 puncture)Differentialpre-coderBufferParserQAMmapperMSBsof "A"MSBsof "B"LSBof "A"LSBof "B"Every 4-bit sequential input yields a 5-bit sequential outputThe overall rate is 14/15Data stream from randomizer28 bits64-QAMoutputCodedFigure B.12/J.83 64-QAM trellis coded modulator block diagramFIGUREB.12/J.83...[D20] = 13.8Recommendation J.83 (04/97) 27T0903520-96/d21T0T1T2T3T4B2B5B8B11B13B1B4B7B10B12A2A5A8A11A13A1A4A7A10A12B0B3B6B9A3A6A9A10A8A7A5A4A2A1A9A6A3A0A13A12A11B10B8B7B5B4B2B1B9B6B3B0B13B12B11A0Time28 bitsRS symbol to Trellis Group bit orderingOrder ofRS symbolsMSB LSBBits inputto BCCQAMsymbolsLSB MSB LSB MSB LSB MSBFigure B.13/J.83 64-QAM trellis groupRS0RS1RS2RS3FIGUREB.13/J.83...[D21] = 13.1B.5.5.2 256-QAM modulation modeFor256-QAM,ananalogoustrelliscodingisemployedusingthesameBCCas64-QAM,withthesamerate1/2generator and the same 4/5 puncture matrix. The 256-QAM trellis coded modulator is shown in Figure B.14. In this casealltheFECframesyncinformationisembeddedonlyinthetrellisgroupconvolutionallyencodedbitpositionsofatrellis group as shown in Figure B.15.There are two distinct types of trellis groups in 256-QAM: hereafter referred to as a non-sync group and a sync group.Each trellis group generates 5-QAM symbols at the modulator, the non-sync group contains 38 data bits while the syncgroupcontains30databitsand8syncbits.FigureB.15showsbothanon-synctrellisgroupandasynctrellisgroup.Sincethereare88RSblocksplus40framesyncbitsperFECframe,therewillbeatotalof2076trellisgroupsperframe. Of these trellis groups, 2071 are non-sync trellis groups and5aresynctrellisgroups.The5synctrellisgroupscome at the end of the frame. The frame sync trailer is aligned to the trellis groups. In the encoder, the trellis group isfurther divided into two groups: one uncoded bit stream and one coded bit stream. The MSB of the first RS symbol inthe FEC frame is assigned to the first bit in the first non-sync trellis group, as shown in the ordering in Figure B.15. Theoutput from each BCC is the five parity bits labelled U1 through U5 and V1 through V5, respectively, as shown in FigureB.14.28 Recommendation J.83 (04/97)U5, U4, U3, U2, U1 B12, B8, B4, B0A16, A13, A9, A5, A1 B18, B15, B11, B7, B3B17, B14, B10, B6, B2 B16, B13, B9, B5, B1 V5, V4, V3, V2, V1 C4C0C5C3C2C1WZXYT0903530-96/d22A18, A15, A11, A7, A3 A17, A14, A10, A6, A2 C7C6(S6, S4, S2, S0)(S7, S5, S3, S1)A12, A8, A4, A0Data streamfrom randomizer38 bitsDataformatterCoded(1/2)Binaryconvolutionalcoder with (4/5 puncture)(1/2)Binaryconvolutionalcoder with (4/5 puncture)Differentialpre-coderUncodedTimeQAMmapper256-QAMoutputMSBsof AMSBsof BLSBof ALSBof BThe overall rate is 19/20Every 4-bit sequential inputyields a 5-bit sequential outputFigure B.14/J.83 256-QAM trellis coded modulator block diagramFIGUREB.14/J.83...[D22] = 14.4Recommendation J.83 (04/97) 29T0903540-96/d23B3B7B11B15B18B2B6B10B14B17B1B5B9B13B16T0T1T2T3T4A3A7A11A15A18A2A6A10A14A17A1A5A9A13A16B0B4B8B12A0A4A8A12B3B7B11B15B18B2B6B10B14B17B1B5B9B13B16T0T1T2T3T4A3A7A11A15A18A2A6A10A14A17A1A5A9A13A16S1S3S5S7S0S2S4S6A0B0A1B1A2A3B2B3A4B4A5A6A7B5B6B7A8B8A9A10A11B9B10B11A12B12A13A14A15B13B14B15A16A17A18B16B17B18A1B1A2A3B2B3A5A6A7B5B6B7A9A10A11B9B10B11A13A14A15B13B14B15A16A17A18B16B17B18S0S1S2 S3S4S5S6S7Sync trellis group Non-sync trellis group38 bitsNon-sync trellis group bit orderSync trellis group bit orderTimeA0 is assigned to theMSB of the first RS symbol in the FEC frameSyncbitsQAMsymbolsQAM symbols Figure B.15/J.83 256-QAM sync and non-sync trellis groupsFIGUREB.15/J.83...[D23] = 15.6ToformtrellisgroupsfromRScodewords,theRScodewordsareserializedbeginningwiththeMSBofthefirstsymbol of the first RS code word following the frame sync trailer. Bits are placed into trellis group locations fromRSsymbols in the order: A0 B0 A1 ... B3 A4 B4...B16B17B18asshowninFigureB.15.Forsynctrellisgroups,thebitsfromserializedRSsymbolsareplacedbeginningatlocationA1insteadofA0.ThelastfivetrellisgroupsinanFECframe each contain 8 of the 40 sync bits, S0 S1 ... S7 in the frame sync trailer shown in Figure B.10.Of the 38 input bits that form a trellis group, each of two groups of 4 bits of type differentially pre-coded bit streams in atrellisgroupareseparatelyencodedbyaBinaryConvolutionalCoder(BCC).EachBCCproduces5codedbits,asshown in Figure B.14. The remaining bits are sent to the QAM mapper uncoded. This produces a total output of 40 bitsper trellis group. Thus, the overall code rate for 256-QAM trellis coded modulation is 19/20.30 Recommendation J.83 (04/97)B.5.5.3 Rotationally invariant pre-codingThedifferentialpre-codershowninFigureB.16performsthe90rotationallyinvarianttrelliscoding.Rotationallyinvariant coding is employed for both 64- and 256-QAM modulation. The key for robust modem design is to have veryfast recovery from carrier phase slips. Non-rotationally invariant coding requires resynchronization of the FEC when thecarrier phase tracking changes quadrant alignment, leading to a burst of errors at the FEC output.The differential pre-coder allows the information to be carried by the change in phase, rather than by the absolute phase.For64-QAM,the3rdandthe6thbitsofthe6-bitsymbolsaredifferentiallyencoded,andfor256-QAM,the4thand8th bits are differentially encoded. If you mask out the 3rdandthe6thbitsin64-QAMasinFigureB.18(labelledC3and C0) and the 4th and 8th bits in 256-QAM as in Figure B.19 (labelled C4 and C0), the 90 rotational invariance of theremaining bits is inherent in the labelling of the symbol constellation.xJ = WJ + xJ1 + ZJ(xJ1 + YJ1)YJ = ZJ + WJ + YJ1 + ZJ(xJ1 + YJ1)T0903550-96/d24WJZJxJYJDifferentialpre-coderDifferential pre-coder equationsFigure B.16/J.83 Differential pre-coderFIGUREB.16/J.83...[D24] = 5.9B.5.5.4 Binary Convolutional CoderThetrelliscodedmodulatorincludesapuncturedrate1/2binaryconvolutionalencoderthatisusedtointroducetheredundancy into the LSBs of the trellis group. The convolutional encoder is a 16-state non-systematic rate 1/2 encoderwiththegenerator:G1010101,G2011111(25,37octal),orequivalentlythegeneratormatrix[1D2D4,1DD2D3D4].Atthebeginningofatrellisgroup,theBCCcommutatorisinitiallyintheG1position.Foreachinputbitpresentedtothetappeddelayline,twobits(G1,followedbyG2)aresubsequentlyproducedattheoutputinaccordancewiththeassociatedsetofgeneratorcoefficients.Foreachtrellisgroup,4inputbitsproduce8 convolutionallyencodedbits.Thistimeoutputoftheencoderisselectedaccordingtoapuncturematrix:[P1,P2][0001;1111]("0"denotesNOtransmission,"1"denotestransmission),whichproducesasingleserialbitstream.Thepuncture matrix essentially converts the rate 1/2 encoder to rate 4/5, since only 5 of the 8 encoded bits are retained afterpuncturing. The internal structure of the punctured encoder is illustrated in Figure B.17.Recommendation J.83 (04/97) 31T0903560-96/d251 0 1 0 11 1 1 1 10 0 0 11 1 1 1Z1Z1Z1Z1G2 = 37(octal) G1 = 25(octal) CommutatorPuncture matrixFor every 4-bit sequential inputyields a 5-bit sequential output(1/2) Binary Convolutional Coder16 stateBinary Convolutional Coder Structure:1) 16 state.2) Rate 1/2 binary convolutional coder.3) Generating code: G1 = [010101], G2 = [011111] (25,37octal)or Generating Matrix of [1(+)D2(+)D4, 1(+)D(+)D2(+)D3(+)D4]where D is equal to Z1.4) Punctured matrix [P1;P2] = [0001;1111].frompre-coderNOTE 1 0 denotes NO transmission. 1 denotes transmission.NOTE 2 (+) denotesXOR operation.Figure B.17/J.83 Punctured Binary Convolutional CoderTo QAMmapperFIGUREB.17/J.83...[D25] = 15.3B.5.5.5 QAM constellation mappingFor 64-QAM, the QAM mapper receives the coded and uncoded 3-bit Aand Bdata from the trellis coded modulator. Ituses these bits to address a look-up table which produces the 6-bit constellation symbol. The 6-bit constellation symbolis then sent to the 64-QAM modulator where the signal constellation illustrated in Figure B.18 is generated.For 256-QAM, the QAM mapper receives the coded and uncoded 4-bit A and B data from the trellis coded modulator.Itusesthesebitstoaddressalook-uptablewhichproducesthe8-bitconstellationsymbol.The8-bitconstellationsymbol is then sent to the 256-QAM modulator where the signal constellation illustrated in Figure B.19 is generated.32 Recommendation J.83 (04/97)T0903570-96/d26011,011 010,111 111,011 110,111QI011,000 010,100 111,000 110,100001,011 000,111 101,011 100,111001,000 000,100 101,000 100,100111,111 110,101 101,111 100,101111,010 110,000 101,010 100,000011,111 010,101 001,111 000,101011,010 010,000 001,010 000,000001,001 000,011 011,001 010,011001,100 000,110 011,100 010,110101,001 100,011 111,001 110,011101,100 100,110 111,100 110,110101,101 100,001 001,101 000,001101,110 100,010 001,110 000,010111,101 110,001 011,101 010,001111,110 110,010 011,110 010,010C5 C4 C3, C2 C1 C0Figure B.18/J.83 64-QAM constellationFIGUREB.18/J.83...[D26] = 13.4Recommendation J.83 (04/97) 33T0903580-96/d271110,10111111,11011110,11111111,10011110,01111111,01011110,00111111,00010000,11110011,11110100,11110111,11111000,11111011,11111100,11111111,11111100,11101101,11001100,10101101,10001100,01101101,01001100,00101101,00000000,11000011,11000100,11000111,11001000,11001011,11001100,11001111,11001010,11111011,11011010,10111011,10011010,01111011,01011010,00111011,00010000,10110011,10110100,10110111,10111000,10111011,10111100,10111111,10111000,11101001,11001000,10101001,10001000,01101001,01001000,00101001,00000000,10000011,10000100,10000111,10001000,10001011,10001100,10001111,10000110,11110111,11010110,10110111,10010110,01110111,01010110,00110111,00010000,01110011,01110100,01110111,01111000,01111011,01111100,01111111,01110100,11100101,11000100,10100101,10000100,01100101,01000100,00100101,00000000,01000011,01000100,01000111,01001000,01001011,01001100,01001111,01000010,11110011,11010010,10110011,10010010,01110011,01010010,00110011,00010000,00110011,00110100,00110111,00111000,00111011,00111100,00111111,00110000,11100001,11000000,10100001,10000000,01100001,01000000,00100001,00000000,00000011,00000100,00000111,00001000,00001011,00001100,00001111,00001110,00011110,00101110,01011110,01101110,10011110,10101110,11011110,11101101,00011010,00011001,00010110,00010101,00010010,00010001,00010000,00010001,00110000,01010001,01110000,10010001,10110000,11010001,11111101,00101010,00101001,00100110,00100101,00100010,00100001,00100010,00000011,00100010,01000011,01100010,10000011,10100010,11000011,11101101,01011010,01011001,01010110,01010101,01010010,01010001,01011101,01101010,01101001,01100110,01100101,01100010,01100001,01101101,10011010,10011001,10010110,10010101,10010010,10010001,10011101,10101010,10101001,10100110,10100101,10100010,10100001,10101101,11011010,11011001,11010110,11010101,11010010,11010001,11011101,11101010,11101001,11100110,11100101,11100010,11100001,11100100,00010101,00110100,01010101,01110100,10010101,10110100,11010101,11110110,00000111,00100110,01000111,01100110,10000111,10100110,11000111,11101000,00011001,00111000,01011001,01111000,10011001,10111000,11011001,11111010,00001011,00101010,01001011,01101010,10001011,10101010,11001011,11101100,00011101,00111100,01011101,01111100,10011101,10111100,11011101,11111110,00001111,00101110,01001111,01101110,10001111,10101110,11001111,1110IQ C7C6C5C4C3C2C1C0Figure B.19/J.83 256-QAM constellationFIGUREB.19/J.83...[D27] = 17.6B.6 Modulation and demodulationB.6.1 QAM characteristicsThe cable transmission format is summarized in Table B.3 for 64-QAM and 256-QAM. Table B.4 contains a summaryof the pertinent characteristics of the variable interleaving modes.34 Recommendation J.83 (04/97)Table B.3/J.83 Cable transmission formatTable B.4/J.83 Variable interleaving modesB.6.2 QAM modulator RF outputThe 64-QAM and 256-QAM modulator RF output specifications for a 75 cable impedance are shown in Table B.5.Table B.5/J.83 QAM modulator RF outputParameter 64-QAM format 256-QAM formatModulation 64-QAM, rotationally invariantcoding256-QAM, rotationally invariantcodingSymbol size 3 bits for "I" and 3 bits for "Q"dimensions4 bits for "I" and 4 bits for "Q"dimensionsTransmission band 54 to 860 MHz (Note) 54 to 860 MHz (Note)Channel spacing 6 MHz (Note) 6 MHz (Note)Symbol rate 5.056941 Msps t 5 ppm (Note) 5.360537 Msps t 5 ppm (Note)Information bit rate 26.97035 Mbps t 5 ppm (Note) 38.81070 Mbps t 5 ppm (Note)Frequency response Square root raised cosine filter(Roll-off 0.18)Square root raised cosine filter(Roll-off 0.12)FEC framing 42-bit sync trailer following 60 RSblocks(see B.5.3)40-bit sync trailer following 88 RSblocks(see B.5.3)QAM constellationmapping6 bits per symbol (see B.5.5) 8 bits per symbol (see B.5.5)NOTE Thesevaluesarespecificto6MHzchannelspacing.Additionalsetsofvaluesfordifferingchannel spacing are under study.Level 1 Level 2QAM format 64-QAM (see Table B.3) 64- or 256-QAM(see Table B.3)Interleaving Fixed interleaving (see B.5.2)I128J1Variable interleaving (see B.5.2)I128,64,32,16,8J1,2,3,4,5,6,7,8,16Parameter SpecificationInput power level range 9 to+16 dBmVI/Q Phase offset