INTERNATIONAL
`STANDARD
`
`ISO/IEC
`11172-2
`
`First edition
`1993-08-Gl
`
`Information technology - Coding of
`moving pictures and associated audio for
`digital storage media at up to about
`1,5 Mbit/s-
`Part 2:
`Video
`
`Techno/ogres de /'information - Codage de /'image animBe et duson
`associe pour les supports de stcckage numerique jusqu 'a environ
`1.5 MbiVs-
`Partie 2: Video
`
`Reference number
`ISO/I EC 11172-2 1 S93[E)
`
`Page 1 of 124
`
` ZTE EXHIBIT 1004
`
`

`
`ISO/IEC 11172-2: 1993 (E)
`
`Contents
`
`F01eword ......... .
`
`Page
`
`..iii
`
`lntroGuction ................................. .
`
`···············-······-······iv
`
`Sectien 1: Geuenl ...... .
`
`. .............. I
`
`1.1
`
`S::ope .............................................................................. .
`
`. ........ I
`
`1.2 Normative references .......... ; ................................................................ !
`
`Section 2: Techriical elements ........................................................................ .3
`
`2.1 Definitions.
`
`................ 3
`
`2.2
`
`Symbols-~nd abbreviatiDns ................................................................. II
`
`2.3 Method of desciibing bitslream syntax ....
`
`. .. ···-···-······ 13
`
`2.L Requirements .... , .................................................................. .
`
`..... 15
`
`Annexes
`
`A
`
`B
`
`c
`
`D
`
`E
`
`F
`
`8 by 8 [merse discrete cosine trflnsform ......................... , ...... .
`
`. .... 39
`
`Variable: lentth co:ie tables ............................................ . ................. 40
`
`Video buffering verifier ....
`
`.. ················· ................ 49
`
`Guide to eno)ditlg video.
`
`......................................... 5 I
`
`Bibliography .................................................................................. 108
`
`List cf patent holders ........................................................................ 109
`
`© !SO/IEC 19'3
`All rights rese~ved. No part of this pllbfication may be reproduxd or utilized in any fonn or by
`any means, el~ctrorjc or mechanic-a~ including ph::rtocopyiag and microfilm, wittout
`perll"tission in writing from the publisher.
`
`ISOI£EC Copyright Office • Case Postale 56 • CH l2ll Geneve 20 • Switzerland
`
`Printed in Switzerland.
`
`ii
`
`Page 2 of 124
`
`

`
`© ISO/IEC
`
`ISCYIEC 11172-2: 1993 (E)
`
`Foreword
`
`ISO ithe International 0"gonization for Standardization) and IEC (the lnter(cid:173)
`nationa Electrotechnical Commission) fDrm the specialized system for
`worldwde standardization. Natonal bodies that are members of ISO or
`IEC participate i1 the development of ln•emational Standards through
`technical ccrnmittees establshed bv the respective organization to deal
`with particular fields of techn cal actvity_ ISO and IEC technical com(cid:173)
`mittees collaborate in fields of mutual interest. Other international organ(cid:173)
`izations, govemmental and non-governmental, in liaison wth ISO and IEC,
`also take part n the work.
`
`In the field of informa•io1 technology, ISO a1d IEC have established a joint
`technical committee, SO/I=:C JTC 1. D-aft International Standards adopted
`by the ioi:lt te:hnical ccmmi:tee are circulated "to national bodies for vot(cid:173)
`ing. Publica1ion as an lnternetional Standard requires·apwoval by at least
`75% d the national bodies casting a vote.
`
`International Standard ISO/IEC 11172-2 was prepared by Joint Technical
`Committee >SO/I =:c JTC 1, Information technology, Sub-Committee SC 29,
`Coded represent3tiori- of audio, pbture, muftime.dia and hvpermedia infcr(cid:173)
`mati6n.
`
`ISD/'EC 11172 consists oi the following parts, under ti1e general title In(cid:173)
`formation technology - Coding of movJng Pictures 9nd aSsociated audio
`foe digital storage media at up 'a about t5 Mbir(s:
`
`- Part 1: Systems
`
`- Part 2: 'Video
`
`- Part 3: Audio
`
`- Part 4: Compliaoce testing
`
`Annexes A, Band C fo-m an integral part of this part of ISOAEC 11172,
`Annexes D, E and F are fer information onl·{.
`
`iii
`
`Page 3 of 124
`
`

`
`ISO/IEC 11172-2: 1993 (E)
`
`©ISOIIEC
`
`Introduction
`
`Note-· Readers interested iu an overview of the MPEG Video layer sllonld read this Introduction and then
`rroceed to amex D, before returning to clauses 1 and 2.
`
`0.1 Purpose
`
`This pilrt ofiSDIIEC 11172 was de,eloped in response 10 the growing need for a conunon fonnat for
`representing compressed video 011 various digital storage media such as CDs, DATs. Winchester disks and
`q:.tical drives. This part of ISOIIEC Ill n specifies a coded represenlation that can be •sed for
`compressing video sequences 10 bitrntes aro•nd 1,5 Mbitis. The nse of this part of ISO/IEC 11172 means
`that motion video em be manipulated as a form of computer data and can retransmitted and received over
`eKi&ting and futme 11etworks. The coded representation can be usal with OOth 625-Iine and 525-line
`television cmd provides flexibility for use with worbtalion and personal computerdispbys.
`
`This pilrt ofiSD/IEC 11172 was de'eloped 10 opera1e principally from storage media offering a continuous
`tJan.Sfer rate of about 1.5 Mbit/s.. Nevertheless it can be used more widely than this l::ecause the approach
`taken is generic:.
`
`0. 1 . 1 Coding parameters
`
`The intention in_ developing this part of ISOJIEC 1.1 172 has been to define a source coding algorithm with a
`large degree of flexibility that can be used in many different applications. To achieve this goal, a number of
`the parameters defining the characteristics of coded bitstremru and decorers are contained in-t!Je bitstream
`itself.. This allov.s for exmiple, the algorithm to be used for pictures with a variety of sizes and a."J)Cet
`ratios and on cil.annels or devices operating at a wide range of bitrates..
`
`Bemuse of the large range of the characteristic of bitstreams that can be represented by this part of ISO.'IEC
`11172. a rmb-set of the~ coding parameters known. as. the "Constrained Parameters" bas been defined. The
`aim in defming tbe constrained parameters is to offer guidance about a wit:ely useful range of parameters.
`Conforming to this >et of constrain Is is not a requirement of this part of JSOIIEC 1 J 1 i2. A flag in the
`bitstream indicates whether or not it is a Constrained Parametets bitstrearn.
`
`Summary of the Ccmstrained Parameters:
`
`Horizontal victure size
`Vertical picture size
`Picture area
`Pelra1e
`Picture rate
`Motion vector range
`
`Input buffer size riu VB V model)
`Bittate
`
`Less than or eQUal to 768 pels
`Less than or equal to 576lines
`Less than or eQUal to 396 macroblocts
`Less than or eQUal to 396>25 macroblock,;ls
`Less than or equal to 30 Hz
`Less than -64 1D +63,5 pels (usmg half-pel vec10rs)
`[backward f codeanlfonvard f code<;4(see1ableD.7)]
`Less than or eoual to 327 680 bit>
`Less tlnn or equal to 1856 000 bits/s (constm1 bitrrte)
`
`0. 2 Overview of the algorithm
`
`The coded representation defined in this part of ISOIIEC 11172 achieves a high compression ratio while
`preserving good picture quality. Ib; algorithm is not lossless as the exact pel \lalues are 110t preserved
`during coding. The choice of 1he techniques is based on the need to balance a higb pictnreqmlity and
`rompression ratio with the requireiiEnt to make random access to the coded bitstream. Obtaining goOO
`picture quality at the bitrates of interest demands a very bigll compression ratio. which is not achievable
`with intraframe roding alone .. ~need for rarldotn access, however, is best satisted with pure intraframe
`rodin~. Thl; f"..,quires a careful balartce between intra- and interframe coding and between recursive and non(cid:173)
`recursive temp:>r.d redundancy redudion.
`
`iv
`
`Page 4 of 124
`
`

`
`©ISO!IEC
`
`ISO/IEC 11172-2: 1993 (E)
`
`A number of techniques are used to achieve a high compression ratio. The first, which is almost
`independent from this part ofiSOIIEC 11172, i~ to select an awropriale spatial resolution for the signal.
`The algorithm liEn uses block-ffised motion compensation to redoce the temporal redundancy. Motion
`rompensation is used for causal prediction of the Cllllent picture from a previ:ms picture, for non-causal
`predict.ion of lb:ecunwt picture from a future picture, or for interpolative prediction from past m.d future
`pictures. Motioll vectDrs are defined for each ItS-pel by 16-Iine regioa of the picture. The difference signal.
`the prediction error. is furthercrnnpressed using the discre1e cosine trnmforrn (DCT) to remove spatial
`rorrelation before it is quantized in an irre\iersible process that discards tbe.less important infomatim.
`Finally. the motion vectors are oombined with the DCT information, and coded using variab:e leugth codes.
`
`0.2.1 Temporal processing
`
`Because of the conflicting requirements of randan access wd highly efficient compression, three main
`pictllfe. l)pes are clefmed. Intra-roded pictures (!-Pictures) are coded witl:lott. reference to other pictures.
`They provide acress points to the coded sequeuce where decoding can begin, but are coded with only a
`moderate compression r.Iio. Predicthe coded pictures (P~Pictures) are rodedmore efficieutly using motio[l
`COmperlsated Jlrediction :fiom a past intra or predictive coded picture and are generaUy used as a reference fer
`furtherprediction. Bidirectiornlly-predictive axled picttre.< (B-Pictures) provide the highest degree of
`compression but require both past and fttme reference pictures for motion comr:ens.ation. Bidirectionally(cid:173)
`predictive coded pictmes are never used as references for prediction The orgm:tisation of the three picture
`t)'Pes in a sequence is very fle:dble, The choice is left tc the enroder and wil1 depend on tbe requirements of
`the application. Figure 1 illuSI:rates tbe relationship between the tllree different picrure types.
`
`Bi-directional
`Prediction
`
`Figure 1 -- Ex2mple of t~mporal picture structure
`
`Prediction
`
`The fourth picture type dofmed in this part of!SO!IEC 11172, the D-pictme, is provided to al.low a 'imple,
`but limited quality, fast-forward playback mode.
`
`0. 2. 2 Motion representation - macroblocks
`
`The choice of 16 by 16 macroblccks for the ID.)tion-compensation unit is a result of the trade-offbetwren
`increasing the coding efficiency provided by using motion infonnation and the oveihea.d needed to stcre it.
`EaclJ macroblockcan be one of antB.nberofdifferent types. For example, intra-coded, focward-predictive(cid:173)
`cc<led, badcward-predicti'.e cod.xl, Md bidirectionally-prediaive-coded macroblocks are penrJued in
`bidirectionally-predictive cOOed pictures. Depending on the type of the macroblock, motion vector
`information and other side information are stored with the mmpressed prediction error signal in each
`macroblock. The motion vectors are encoded differentially with respect to the last coded motion vector,
`using variable-length rodes. The maximlll11 length of the vectors tl:Jat may be represented can be
`programmed, on a picture-by-pictwe basis, so that the most demandillg applicatious c3ll be met without
`compromising the pt'rlonnance of the system in more normal situations
`
`It is the resp:m~ibility of the encoder to calculate appmpriate motion vectors. This part of ISOIIEC 11172
`does not specify bow this soould be done.
`
`v
`
`Page 5 of 124
`
`

`
`ISO/IEC 11172-2: 1993 (E]
`
`©ISOAEC
`
`0. 2. 3 Spatial redundancy reducti<>n
`
`Both original pictures ani prediction error signals have high spatial redu::tdancy. This part of fSOIIEC
`11172 mesa block-bl!Sed DCT melhcxl with visually weig!Jted quantization and run-lenEth coding. Each 8
`by 8 block of the original picwre for intra-coded ma:roblocts or of the pl"'.Aiiction error for predicti ve-roded
`macroblocks is transformed into the DCT domain where it :is scaled before being quantized. Mter
`quantization many of !he coefficients are zero in value aiJd so two-~mional nm-length and variable
`length coding is used ID encode the remaining coeffici~nts efficiently.
`
`0.3 Encoding
`
`This part of ISO.'IEC 11172 doe; not specify an encodiOE process. It specifie;; the syntax: and serrumtics of
`the bitsfl:eam and the signal processing in the decoder. As a result, many options are left open to encoders
`to trade-off CGst and speed against picture quality and coding efficiency. This claJse is a brief description of
`the functions that need to be perfonned by an encoder. Figure 2 sbows !he marn functiooal blocks.
`
`Picture
`store I
`Predictor
`
`DC1' is discrete cosine transform
`ocr-I is inverse discrete cosine transfonn
`Q is quantization
`Q-1 is dequantization
`VLC is vllriable length coding
`
`Figure 2 -- Simplified videv em:oder block diagram
`
`The input video signal-must te digitized and represented as a Juminanre IDd two colour difference signals
`(Y, Ct>, Cr). This may be followed by preprocessing and fonnat conversion to relectanappropriate
`window, resolution and input format. This pmt of !SOIIEC 11172 requires that the colour difference
`signals (Cb and Cr) are subsampled with respect tu the luminance l>y 2: l in both vertical and horizol1lal
`directions and are reformatted, if necessa-y, as a non-interlaced signal.
`
`The encoder must choose which picture type to use for eacll picture. Having defined the picture types, the
`encoder estimates motion vectors for each 16 by 16 macroblock in the picture.
`fu P-Pictures one vector is
`needed for each non-intra macroblcckand in B-Pictures one or two vectors are needed.
`
`[f B-Pictures are used, some reordering of the pictme reqLience. is necessary before encoding. Because B(cid:173)
`Pictures are coded Lising bidirectional motion compensated prediction, they can only be decoded after the
`subsequent refere11ce picture (ail I or P-Picture) has been decoded. Therefore tl:le pictures are reOidered by the
`
`vi
`
`Page 6 of 124
`
`

`
`©ISO/lEG
`
`ISOIIEC 11172-2: 1993 (E)
`
`encoder so that the pictwesarrive at the decoder in tlle order for decoding. 'J1"e rorrect di<iplayorder is
`recovered by the decoder.
`
`'IlJe basic unit of coding within a picture is the macroblodc Within each picture, macroblocks are encoded
`in sequence, left to right,. top to bottom. Each macro blOck con:;ists of six 8 by 8 blocks: fow blocks of
`Iuminaoce, one block of Cb chrcminance, and one block of Cr chromillance. See figure 3. Note that the
`picture area covered by the four blocks ofllDllinance is the same as t.lle area covered by each of the
`chrominallce blocks. This is due to subsampling of the chrominance information to match the sensitivity of
`tbe human visual system.
`
`[][]I]
`IIJIJ
`y
`
`Cb
`
`Cr
`
`Figure 3 ~~ Macroblock structare
`
`Firstly, for a given m.tcroblodc, the coding mcde is chosen. It depends 011 the picture type, the
`effectiveness of motion compensated predictim in that local region, and the nature of the signal within the
`block. Serondly. depending on lhecoding moie, a motion rompensated prediction of the contents of the
`block based m past and/or future reference pictures is formed. Tills predictioo is subtracted from the actual
`data in tbecurrent macroblod: to form an error signal. Thirdly, this enor sisn.:1l is separated into 8 by 8
`blocks (4 luminance and 2 chrorninance blocks in each macroblock) and a discrete cosine trap;fonn is
`perfonned on each block. Each re"lting 8 by 8 block of DCT coefficients is quantized illld the two(cid:173)
`dimensional block is scanned in a zig-z.ag order to convert it into a one-dimensionol string of quantized OCT
`coefficients. Frurtb.ly, tbe side-infmmation for the macroblock (mode, motion vectors etc) and the
`quantized coefficient data are encoded. For maximum efficiency, a number of vmi.able length code tables are
`defined for the <fifferent data elements. Run-length roding is used for the quantized coefficient data.
`
`A consequence of using diffe~nt picture types and variable length ccxling is that the overall data rate is
`variable. In applications that invohe a ftxed-n.techannel, a FIFO buffer may be uied to match the encoder
`output to the channel. The statm of this buffer may be monitored to control the nllmber of bits generated
`by the encoder. Controlling the quantization process is the most direct way of controlling the titrate. This
`part of ISOIJEC 11172 speci.Ires an abstract model of the buffering system (the Video Buffering Verifier) iit
`order to coostrain the maximum variability in the number of bits that are used for a given picture. This
`emures thatabitstream CaJJ b€ deccxled with a buffer of known size.
`
`At this stage, the coded rep-esentation of the picture has teen generated. The final step in the ell coder is to
`regenerate £-Pictures and P-Pictures by decoding the data so thai trey can reused as reference pictmes for(cid:173)
`subsequentenca:ling. The quantized coeffidents are dequmlized and an invezse 8 by 8 DCf is performed on
`each block The prediction error signal produced is lhen added back to the prediction signal and limitfrl to
`the required range to give a decodxl reference picture.
`
`0.4 Decoding
`
`Decoding is the inver;e of the eucocling operation. lt is considerably simpler than encoding as there is no
`need to perfonnmotion estimation and there are many feweropticns. The decoding process is defined by
`this part of IS01IEC 11172. The descriptim that follows is a 'ery brief overview of one possible way of
`decoding a bitstream. Othe.r .XCOders with different architectures are possible. Figure 4shows the main
`functior1al blocks.
`
`vii
`
`Page 7 of 124
`
`

`
`lSO!IEC 11172·2: 1993 (E)
`
`©ISO/lEG
`
`Motion Ve~.:tors
`L--~~:!!..!~~'------111>-l Picture stl)re
`and
`Predk1or
`
`Rewrutructed
`output pictwes
`
`Where
`
`DCT·l
`Q-l
`MUX-1
`VlD
`
`is inverse discrete cosine transform
`is dequanlizatioo
`is demultiplexing
`is variable lenglh derodin&
`
`Figure 4 -- .Basic video decoder block diagram
`
`For fi~ed-rate applications, the channel fills a FIFO buffer at a constant rate with the ceded bitstream. The
`decoder reads this buffer and deccdes tbe data elements in the bitstream according to the defined syttax.
`
`As the decoder reads the bitstream. ·it identifies the starl of a coded picture and then the type of tile picture.
`It decodes eacl:t macro block in the {icture in tum. The macroblock type and the motion vectors, if present.
`are used to construct a prediction of the currentmacroblock tased on past and future reference pX::tures tbat
`~ave teen stored in !he decoder. The coefficient data are dec<Jded md dequanfized. Eacb 8 l>y 8i>lock of
`coefficient datlt is tramfonned by au inverse DCT (specified in annex A), and the result is ad<£d to the
`prediction sigral and limited Ill the defmed range.
`
`After all the macroblocks in the picblre have been processed, the picture has been recollStrucled If it ism[(cid:173)
`picture or a P-picture it is a reference picture for subsequent pictures and is stored, replacing the oldest stored
`reference piclllre. Before the pictnres are displayed they may need to l>e re-ordered from lhe ceded order to
`their natural display order. Mterreord.."''ing, the pictures are available. in digital form. for post-proce~ing
`and display in any manner that the fiPplication chooses.
`
`o. 5 Structure of the coded video bitstream
`
`This part ofiSOiiEC 11172 specifies a syntax for a coded video l>it>tream. This synta< contains six layers,
`each of which either supports a signal processing or a sYstem function:
`
`L•yers of the synta<
`Se:juenre layer
`Grollp of pictwes layer
`Picture layer
`Slice layer
`Macro I> leek layer
`Block layer
`
`Function
`Random access unh: ronte>it
`Rc!nOOm access unit video
`Primary coding unit
`Resynchro11ization nnit
`Motion compens.:1.lion unit
`DCT unit
`
`0.6 Features supported t>y the algorithm
`
`Applicatious using compessed video on digit21 storage media need 10 be abE to perform a namber of
`operalions in addition to normal forward playback of the reqoer1ce. The ceded bitstream has been designed
`to support a number of these operations.
`
`viii
`
`Page 8 of 124
`
`

`
`© ISO/IEC
`
`ISO/IEC 11172-2: 1993 (E)
`
`0. 6. 1 Random access
`
`Random access is an essential feature fcr \'ideo on a storafie medirm, Rarldom access requires that any
`picture can be decoded in a l:i:niterl amount of time. It implies the existence of access points in the
`bitstream - that is segments of ir.formation that are ident.ifi~ble 211d can be decoded without reference to other
`segments of data A spacing of r:wo random access points (l:rtra-Pictures) per second c:m be achieved
`without significant loss of picture quality.
`
`0. 6. 2 Fast search
`
`Depending on tbe storage medium, it is possible to scan the access points in a coded birstream (with the
`help of an application-specific directory or other knowledge !>eyond the scope of tlti• partof!SOIIEC
`11172) to obtain a fast-forwanl :md fast-reverse playback effecL
`
`0. 6. 3 Reverse playback
`
`Some applications may require the video signal to be played in reverse order, This can be achieved in a
`deaxler by using memory to store entire groups of pictures after they have been decoded tefore being
`displayed in reverse order. An encoder can make this feature easier by redJcing tle Ength of groups of
`pictures.
`
`0. 6. 4 Error robustness
`
`Most digital storage media an! communication channels are not error-f1ee. Appropriate ch:rnnel coding
`schemes should i>e used and are beyood the scope of this part of ISCI'IEC 11172. Nevertbeles; the
`compres:;ion scheme defined in this pan of ISO/JEC-11172 is robust to residual errors. The slice structure
`allows a decoder to recover after a dati error and to resyochronize its decoding. Therefore. bit errors in the
`compressed data wUI cause errors in the deco:led pictures to be limited in area. Deccders may be able to use
`concealment strategies to disguise these enurs.
`
`0.6.5 Editing
`
`There is a ronflict between the requirement for high coding efficiet~cy and elliyediting. The ceding structure
`and S)Titax bave not teen designed with the primary aim of simplifying editing at arty picture. Nevertheles~
`a number of features l:Jave bem inclured that en2ble editing of coded data.
`
`Page 9 of 124
`
`

`
`This page inten1ionall~' left blank
`
`Page 10 of 124
`
`

`
`INTERNATIONAL STANDARD© ISO/lEG
`
`ISDnEC 11172-2: 1993 (E)
`
`Information technology - Coding of moving
`pictures and associated audio for digital storage
`media at up to about 115 Mbit/s
`
`Part 2:
`Video
`
`Section 1: General
`1.1 Scope
`
`This part of ISO'IEC 11172 speciftes the coded re]resentation of video for di)!ital storage media and
`specifies tre deroding process. The representation supports noma] speed forward playback, as wen as
`special fllnction.s such as random access, fast forward play bact, fast reverse playback, normal speed reverse
`playbact, pause and still pictures. Ttris part ofiSOIIEC 11172 is compatible witll standard 525- and 625·
`line television formats, and it provides fie;xibility for use with personal computer and workstation displays.
`
`ISOIIEC 11172 is prim<Uily applicable to digital storage media supporting a continuous transfer rate up to
`about 1,5 Mbit/s, sucll as Compl.CtDisc. Digital Audio Tape, and magnetic hard disks. Nevertheless it can
`be used more wilely than this because of the generic approach taken. The storaf:e medi<'< may be directly
`connected to the deCGder, or via communications means such as busses, LANs, or telecommunications
`links. 1bi~ partofiSD/IEC 11172is intended fornon-iaterloced video formats haviug apprm:.imately 288
`lines of 352 pels and picture rates around 24Hz: to 30Hz.
`
`1 .2 Normative references
`
`The following International Standanl:i rontain provisions which, througt:a reference in this text, constitute
`provisions of this part of ISO/IEC 11172. At the time of publication, the editions indicated were valid.
`All sllllldards are s"bject to revision. and pru1ies to aEJ'eemeniS based on this part of ISO/IEC 11172 are
`encouraged to investigate the possibility of applying the most recent editions of ::he stmdards indiarted
`\>ebw. \iembers ofiEC and ISO maintaill registers of currently valid Internaticnal Standards.
`
`ISO/IEC 11172-1:1993 /'!fOrmation tecltnology- Coding of moving pictures and associated audio [or dJgillll
`uorage media at up 10 abo uti ,5 Mbit/.s - Part I: SyJtems.
`
`ISOIIEC 11172-3:1993/'!frJrmation tecltnology- Coding of moving pict.res and associated audio for dJgillll
`rtora~e media at up 10 about 1.5 Mb1tls- Part J Audio.
`
`CCIR Reconunendation 601-2 Encoding parameters of digitaltelevi.sior.for studios.
`
`CCIR Report 624-4 Characteristics of systems for monocltrom.t and coiour television.
`
`CCIR Reconnnendation M8 Recording oj audio 5ignals.
`
`CCIR Report955-2 Soumi broadcasting by satellite [or portcble and fiWbile receivers, tnduding Annex IV
`Summary de;cription oj Advanced Digital Systtm ll.
`
`CCITI Recommendation J.17 Pre-emphasisu.sedon Sound-Prograrrune Circuits.
`
`1
`
`Page 11 of 124
`
`

`
`ISO/IEC 11172-2: 1993 (E)
`
`© ISO/IEC
`
`IEEE_ Draft Standard Pll80/D21990 SpecifLCaiionfor the implerr.entati011 of 8x S i11verse. disc ret:? cosine
`Jran.sjo nn ".
`
`IEC publication 9013:1987 CD Digitol Audio System.
`
`2
`
`Page 12 of 124
`
`

`
`© ISO/lEG
`
`ISO/lEG 11172-2: 1 993 [E]
`
`Sec1ion 2: Technical elements
`2.1 Definitions
`
`fur the purposes of ISO.IJEC 11172, the following defmitiOIJS apply. If specifi: to a part, this is noted in
`square brnctels.
`
`2.1.1 ac coefficient I video]: Any OCT coefficient foc which the frequency in one or buth dimensions
`is non-zero.
`
`21.2 acce$ •nit !system]: 111 the case of compressed audiu an acress unit is an audio access unit. In
`the case of compressed video an access unil is the coded representaioo of a pictnre.
`
`2.1.3 adaptive segmeJJtation [audio]: A subdivision uf the digital representation of an audio signal
`in variable segments of time.
`
`2.1.4 adaptive bit allocation [audio]: The assignment of bits to subband5 in a time and frequency
`vai")'ing fashion according to a psychoacoustic model.
`
`2.1.5 adaptive noist allocation [ a11dio ]: The assignment of coding noise to frequency bands in a
`time and frequency varying fasllion acccnling to a psychoacoustic mcdel.
`
`2.1.6 alias [alldio]: Mirrored signal component resulting frcm sub-Nyquist sampling.
`
`2.1.7 aoal}sis filtubank [audio!: Filterbank in tbe wcoder that transforms a broadband PCM audio
`signal into a set of mbsampled subband samples.
`
`2.:1.8 auclio access wtit [audio]: For-Layers I and II an audio access unit is defined as the sm21lest
`partofthe encoded bitst.ream which cal be decoded by itself, wheredoccded mems "fully reconstructed
`sound•. For La-yer UI an audio access unit is part of the bff.stream that is decodable with the use of
`previously acquired main information.
`
`2.1.!) audio bllffer [:audio]: A buffer in the system t:aJget decoder for storage of CDmpressed audio data.
`
`2.1.10 alldio sequence [aurlioJ: A non-interrupted series of audic frames in which tbe following
`parameters are not dlanged:
`-ID
`-layor
`- Sampling Frequency
`- filr Layer I and ll: Bitrate index
`
`2.1.11 backward motio11 vector [video}: A motio11 vector thai: is used for motion compensation
`from a reference picture at a later time in display order.
`
`2.1.12 Bark [audio]: Unit of critical bmd rate. The Bark scale is a non·li11ear mapping of the frequency
`seal~ over the audio ranEC closely corresponding with~ frequency selectivity of the human ea- across the
`OOild.
`
`2.1.13 bidirectionally predictive-coded picture; B-pidure [video]: A picture that is cOOed
`using motion compensated prediction from a JXLq_ and' or futrne reference picture.
`
`2.1.14 bitratt:~: The rate at which the compressed bitstrenn is delivered from the storage medilllll to the
`input of a decoder.
`
`2..1.15 block C(lmpaoding (audio]: Normalizing of the digital representation of an audio signal
`within a certain time period.
`
`2.1.H block [video]: i\u 8-row by 8-column orthogonal block of pels.
`
`21.17 b[)und [audio]: The lowest subband in which intensity stereo coding is llied.
`
`3
`
`Page 13 of 124
`
`

`
`ISO/IEC 11172-2: 1993 (E)
`
`©ISOIIEC
`
`2.1.18 hyte aligned: A bit in a coded bitstream is byle-aligned if its position is a m\lltiple of &-bits
`from the first bit in tl1e stream.
`
`2.1.19 byte: Sequence of 8-bits.
`
`2.1.20 cban•el: A d1gital medium that stores or transports an IS Om~: 11 L 72 stteam.
`
`2.1.21 channel [ auc:l.iol: The left and righ1 channels of a stereo signal
`
`2.1.22 cbrominaoct (component) [ yfdeo]: A. matrix, blocl:. or single pel representing one of the
`two colour differer1ce signals related to the primary colours in the manner defined in CCIR Rec 601. The
`symbols used for the colour difference signals rue Cr and Cb.
`
`2.1.23 coded aadio bitstrearn [audio]: A. coded representation of a.Tt audio signal as specified in
`ISO/IEC 11172-3.
`
`Z-.1.24 coderl video bitstream [video I: A coded representation of a series of one or more pictures as
`specified in this part of ISO/lEC 11172.
`
`2.1.25 coded order [videCl]: The order in wb.ich the pictures are stored and decoded. This order is not
`necessarily t.lx'! same as the display order.
`
`2:..1.26 .:oded r~presentation: A data element as represented in its enccdecl fonn.
`
`2.1.27 coding parameteno [video]: The set of user-definable parameters that characterize a coded video
`bitstream. Bitstreams are charncterised by coding parameters. Decoders are clk'lfficterised by the bitstreams
`thllt they are capable of deooding.
`
`2.1.28 co:D:tponent [video]: A matrix, block or single pel from oue of the. three matrices (I11minance
`and two cl:rominance) thai make up a picture
`
`2.1.2!J coDlpression: Reduction in the number of bits used to represent ar1 item of data.
`
`2.1.30 constant bitrate coded video [video]: A compressed video bitstream with a coilstmt
`average bitrate.
`
`2.1.31 constant bitrate: Operation \\>here the titrate is constant from start to finish Gf the compressed
`bi5tream.
`
`2.1.32 constrained parameters [video]: 1be values of the set of coding parameters defir1ed ir1
`2.4.3.2.
`
`2.1.33 constrained system parameter stream (CSPS) [sys1em]: An ISOAEC 11172
`multiplexed stream for w~ich the constraints defined in 2.4.6 of ISO.'IEC 11 172-1 apply.
`
`2.1.34 CRC: Cyclic redundallcy code.
`
`2.1.35 critical band rate [audio]: PsycllOa.coustic function of fre-quency. At a given audible
`frequency it is proportional to the number of critical bands below that frequency. The units of the critical
`baod rale scale are Bailes.
`
`2.1.36 critical b~nd [audio]: Psychoacoustic nrasure in the spectral domain which ronesponds to the
`frequency selectivity of the human ear. This selec6vity is expre&sed in Bark.
`
`2.1..37 data element An item of data as represented before encoding and after decoding.
`
`2.1.38 dc-coefficietJt [video}: The DCT coeffident for which the frequency is zero in OOth
`dimensious.
`
`4
`
`Page 14 of 124
`
`

`
`©lSD/lEG
`
`ISD/IEC 11172-2: 1993 (E)
`
`2.1.39 de-coded picture; D-picture [video]: A picture that is coded using only information from
`itself. Of the OCT coefficients in the ceded representation, only the dc-roefficients are present
`
`2.1.40 OCT coefficient: The amplitude of a specific cosine basis ftiDction.
`
`2.1.41 decoded stream: The decoded reconstruction of a compressed bitstream.
`
`2.1.42 decoder input buffEr [vide()]: Tlle first-in first-out (FIFO) buffer specified in the video
`buffering verifier.
`
`2.1.43 decoder input rate [vide(!}: Tbe data rate spedfied in the video buffering verifier and enccded
`in the coded vi<P.o biiStream.
`
`2.1.44 decoder: An embodiment of a decoding :rrocess.
`
`2.1.45 decoding (process): The proce:;s defined in ISO/IEC 11172 tbat reads ail input cmEd bitstream
`ani produces decoded pictures or audio samples.
`
`2.1.46 de~oding time-stamp; DTS (system I: A :ield :llat may be present in a packet header that
`indic1tes the time·that an access unit is decoded in the sy&tem target decoder.
`
`2.1.47 de-emphasis [~urliol: Filtering applied to an audio siBnal after storage or transmission to undo
`a linear di~tortion due to .em-phasis.
`
`2.1.48 dequantization [video]: The process of r~scaling tree quantized DCT coefFieienB after their
`representation in tbe bitstream has been decoded and "before they are presented to the inverse OCT.
`
`2.L49 digih-1 storage rnerlia; DSM: A digit.:'\..!. storage or t:n.nsmissiml revice or system.
`
`2.1.50 discrett' cosint' transform; OCT [vid.eo]: Either the forv.ard discrete cosii~ transform or the
`inverse discrete cosir;e tr.u1sfom. 'The DCT is an invertible, discrete ort1ogonal transformation. The
`inverse DCT is defmed in anne:x A
`
`2.1.51 display order [video}: The order in which the decoded picture& shollld be displayed. Nonnally
`this is the same order in which. they were JXesented at the input of the encoder
`
`2.1.52 dual channel m