SMPTE ST 302M-2007 Television - Mapping of AES3 Data into an MPEG-2 Transport Stream.pdf

1、 Table of Contents Page 1 Scope 1 2 Conformance Notation 1 3 Normative References 2 4 Introduction .2 5 SMPTE AES3 Elementary Streams2 6 Packetized Elementary Stream.5 7 MPEG-2 Transport System.7 Annex A Bibliography (Informative) 8 Revision Notes .9 1 Scope 1.1 This standard specifies the method of

2、 transporting AES3 data in an MPEG-2 transport stream for television applications. For these television applications, the AES3 frame rate is 48 kHz. 1.2 Some applications may require linear PCM (pulse code modulated) digital audio in conjunction with compressed video specified in the MPEG-2 4:2:2 pr

3、ofile. The MPEG audio standard defines compressed audio, but does not define uncompressed audio for carriage in an MPEG-2 transport system. This standard augments the MPEG standards to address the requirement to carry AES3 streams, which may consist of linear PCM audio (at 48 ksample/s), or data car

4、ried within the AES3 format. 2 Conformance Notation Normative text is text that describes elements of the design that are indispensable or contains the conformance language keywords: “shall“, “should“, or “may“. Informative text is text that is potentially helpful to the user, but not indispensable,

5、 and can be removed, changed, or added editorially without affecting interoperability. Informative text does not contain any conformance keywords. All text in this document is, by default, normative, except: the Introduction, any section explicitly labeled as “Informative“ or individual paragraphs t

6、hat start with “NOTE” . Page 1 of 9 pages SMPTE 302M-2007Revision of SMPTE 302M-2002 Copyright 2007 by THE SOCIETY OF MOTION PICTURE AND TELEVISION ENGINEERS 3 Barker Avenue, White Plains, NY 10601 (914) 761-1100 SMPTE STANDARD for Television Mapping of AES3 Data into an MPEG-2 Transport Stream Appr

7、oved October 25, 2007 SMPTE 302M-2007 Page 2 of 9 pages The keywords “shall“ and “shall not“ indicate requirements strictly to be followed in order to conform to the document and from which no deviation is permitted. The keywords, “should“ and “should not“ indicate that, among several possibilities,

8、 one is recommended as particularly suitable, without mentioning or excluding others; or that a certain course of action is preferred but not necessarily required; or that (in the negative form) a certain possibility or course of action is deprecated but not prohibited. The keywords “may“ and “need

9、not“ indicate courses of action permissible within the limits of the document. The keyword “reserved” indicates a provision that is not defined at this time, shall not be used, and may be defined in the future. The keyword “forbidden” indicates “reserved” and in addition indicates that the provision

10、 will never be defined in the future. Unless otherwise specified the order of precedence of the types of normative information in this document shall be as follows. Normative prose shall be the authoritative definition. Tables shall be next, followed by formal languages, then figures, and then any o

11、ther language forms. 3 Normative References The following documents contain provisions which, through reference in this text, constitute provisions of this standard. At the time of publication, the editions indicated were valid. All standards are subject to revision, and parties to agreements based

12、on this standard are encouraged to investigate the possibility of applying the most recent edition of the standards indicated below. AES3-2003, Digital Audio Engineering Serial Transmission Format for Two-Channel Linearly Represented Digital Audio Data SMPTE 337M-2000, Television Format for Non-PCM

13、Audio and Data in an AES3 Serial Digital Audio Interface SMPTE EG 32-1996, Emphasis of AES/EBU Audio in Television Systems and Preferred Audio Sampling Rate ISO/IEC 13818-1:2000, Information Technology Generic Coding of Moving Pictures and Associated Audio Information: Systems 4 Introduction MPEG-2

14、transport streams convey one or more programs of coded data, and may be constructed from one or more elementary coded data streams, program streams, or other transport streams. This standard specifies the method to transport AES3 streams, which may consist of linear PCM digital audio, or data, withi

15、n an MPEG-2 transport stream. The specifications are described in terms of a model which starts with AES3 data, constructs elementary streams (ES) from the AES3 data, then constructs packetized elementary streams (PES) from the elementary streams, and finally constructs MPEG-2 transport streams (MTS

16、) from thepacketized elementary streams. Although this model is used to describe the transport of AES3 streams in MPEG-2 transport streams, the model is not mandatory. MPEG-2 transport streams may be constructed by any method which results in a valid stream. 5 SMPTE AES3 Elementary Streams 5.1 SMPTE

17、 AES3 elementary streams shall consist of AES3 data words, which may be derived from AES3 subframes, together with validity, user, and channel status (V,U,C) bits and a framing (F) bit. The F bit SMPTE 302M-2007 Page 3 of 9 pages replaces the parity (P) bit normally carried in AES3 subframes, and is

18、 used to indicate the start of a block of AES3 channel status (C) and user (U) bit information (see paragraphs 5.5 and 5.6 and figure 3). 5.2 There may be 1, 2, 3, or 4 AES3 data streams conveyed in a single audio elementary stream and corresponding packetized elementary stream. Multiple packetized

19、elementary streams may be used in applications requiring carriage of a larger number of AES3 streams. Where multiple packetized elementary streams are used to convey multiple AES3 streams, the ordering of AES3 subframes within a packetized elementary stream is maintained within the packetized elemen

20、tary stream, while the ordering across different packetized elementary streams can be maintained through the value of channel_identification (see 6.6) in the elementary stream header. 5.3 The data word size shall be 16, 20, or 24 bits per word. All AES3 streams in one SMPTE AES3 elementary stream sh

21、all have the same word size. 5.4 The word rate shall be 48 kHz. This rate shall be locked to the 27-MHz transport clock. All channels in a SMPTE AES3 elementary stream shall have the same word frequency. 5.5 AES3 streams consist of a sequence of AES3 frames. A set of 192 sequential frames forms a bl

22、ock. Individual AES3 frames consist of a pair of AES3 subframes, designated the A subframe and the B subframe. AES3 subframes consist of data for a single data word, as well as additional data including a validity bit (V), a user data bit (U), a channel status bit (C), a parity bit (P), and four aux

23、iliary sample bits (aux). AES3 subframes may include 24-bit data words as shown in figure 1, 20-bit data words as shown in figure 2, or 16-bit data words as shown in figure 3. AES3 subframes which carry 16-bit data words use a 20-bit word with the four least significant bits (LSBs) of the 20-bit wor

24、d set to 0. 5.6 Validity, user, and channel status (V,U,C) bits shall be carried in the MPEG transport stream specified by this standard. 5.7 Framing information signaling the first AES3 subframe of a block is carried by a framing (F) bit that replaces the parity bit normally carried in the AES3 str

25、eam. The framing bit shall be set to 1 for the first A subframe of an AES3 block (indicated by the Z preamble in the AES3 stream) and to 0 for all other subframes (including all B subframes). Note that when multiple AES3 data streams are conveyed within one SMPTE AES3 elementary stream, each individ

26、ual AES3 stream that is carried will have the F bit set to 1 for one A subframe out of every 192 A subframes; the F bit in all B subframes will always be set to 0. It is not necessary for all of the individual AES3 streams conveyed by the SMPTE AES3 data elementary stream have the same phasing of th

27、e 192 frame Blocks. 5.8 Elementary streams derived from AES3 data words and V,U,C,F bits shall be constructed as shown in figure 4. Elementary streams shall be composed of 302M word groups. Each 302M word group shall contain one AES3 data word together with the associated V,U,C and F bits for each s

28、ubframe data channel carried in the elementary stream (i.e., one A subframe or one B subframe from each AES3 stream). The bits of the AES3 data words shall be the LSBs of the 302M data word; V shall be more significant, then U, then C, and the F bit shall be the MSB. The bits of the 302M data words

29、shall be sent in the same order as in the AES3 sub-frame as shown in figures 1, 2, and 3 (i.e., LSB first). The 302M data words within each 302M data word group shall be in the order of their channel number (AES3 stream 1 subframe A, followed by AES3 stream 1 subframe B, followed by AES3 stream 2 su

30、bframe A, etc.). Successive 302M data word groups shall contain successive 302M data words from each of the channels carried in the elementary stream. SMPTE 302M-2007 Page 4 of 9 pages 0 3 4 27 28 29 30 31 Preamble 24-bit data word V U C P LSB MSB Figure 1 24-bit AES3 subframe 0 3 4 7 8 7 27 28 29 3

31、0 31 Preamble Aux 20-bit data word V U C P LSB MSB Figure 2 20-bit AES3 subframe 0 3 4 7 8 11 12 27 28 29 30 31 Preamble Aux 0 0 0 0 16-bit data word V U C P LSB MSB Figure 3 16-bit AES3 subframe X YVU CP VU CPDataPayloadDataPayloadZ YVU CP VUCPDataPayloadDataPayloadFrame 191Frame 0X YVUCP VU CPData

32、PayloadDataPayloadFrame 1PreamblesVUC0AES3 DataWordVU C0AES3 DataWordVU C1AES3 DataWordVUC0AES3 DataWordVUC0AES3 DataWordVU C0AES3 DataWordAES3Stream 1First Group of2xM Data WordsVUC0AES3 DataWordVU C0AES3 DataWordVU C1AES3 DataWordVUC0AES3 DataWordVUC0AES3 DataWordVU C0AES3 DataWordAES3Stream 2VUC0

33、AES3 DataWordVU C0AES3 DataWordVU C1AES3 DataWordVUC0AES3 DataWordVUC0AES3 DataWordVU C0AES3 DataWordAES3Stream MSubframe ASubframe ASubframe ASubframe BSubframe BSubframe BSubframe ASubframe ASecond Group of2xM Data WordsNthGroup of 2xMData WordsTotal of M Channel Pairs of AES3 DataSubframe B(Chann

34、el 2)Subframe ASubframe ASubframe B Subframe BSubframe B Subframe BSubframe A(Channel 1)Subframe A(Channel 1)Subframe A(Channel 1)Subframe B(Channel 2)Subframe B(Channel 2)Figure 4 SMPTE AES3 elementary stream SMPTE 302M-2007 Page 5 of 9 pages 5.9 The 302M data words shall be packed, leaving no unus

35、ed data space (see figure 5). In the case of 20-bit AES3 data word resolution, the AES3 data word plus VUCF bits for each channel are inherently packed since the sum of payload sample word length plus VUCF bits are multiples of 8 bits. In the case of 16- and 24-bit AES3 data word resolution, the AES

36、3 data word plus VUCF bits shall be packed without sample byte alignment. Since AES3 frames carry a pair of AES3 data words, the resulting length of a 16- or 24-bit AES3 data payload plus VUCF for each AES3 frame will be an integer number of bytes (see figure 5). Figure 5 Packing of AES3 data into E

37、S bytes 6 Packetized Elementary Stream 6.1 PES packets for SMPTE AES3 data shall follow the specifications in ISO/IEC 13818-1 and as noted in this standard. 6.2 SMPTE AES3 data PES packets shall have an MPEG-2 PES header as described in ISO/IEC 13818-1, clauses 2.4.3.6 (PES packet) and 2.4.3.7 (sema

38、ntic definition of fields in PES packet). Additional SMPTE AES3 data header information shall appear at the beginning of the MPEG-2 PES packet payload as shown in figure 6. MPEG-2 PES Packet MPEG-2 PES Payload MPEG-2 PES Header SMPTE 302M AES3 data header SMPTE 302M AES3 data payload Figure 6 Audio

39、data PES packets SMPTE 302M-2007 Page 6 of 9 pages 6.3 The SMPTE AES3 data PES packets shall conform to the semantic definition for private_stream_1 in ISO/IEC 13818-1, clause 2 .4 .3 .7 . The semantic definition for private_stream_1 includes presentation time stamps (PTS) and other information. (No

40、te that the semantic definition for private_stream_2, which is an abbreviated header, is not used in this standard.) 6.4 SMPTE AES3 data PES packets shall have PTS_DTS_flags set to XO. 6.5 In SMPTE AES3 data PES packets, the ESCR_flag, ES_rate_flag, DSM_trick_mode_flag, additional_ copy_info_flag, a

41、nd PES_extension_flag shall be set to 0. 6.6 Each SMPTE AES3 data PES packet payload shall include a SMPTE AES3 data header which shall be carried at the beginning of the associated MPEG-2 PES payload. This header shall include audio_packet_size, number_channels, channel_identification, and bits_per

42、_sample fields as described below. This header shall apply to all of the AES3 subframe channels in the PES packet. NOTE The channel_identification may prove useful when multiple packetized elementary streams are used to convey multiple AES3 data streams in separate SMPTE AES data elementary streams.

43、 (If more than four AES3 data streams must be conveyed, it is necessary to employ multiple SMPTE AES3 data elementary streams.) For example, suppose a 12 channel audio program contained in six AES3 streams must be carried in an MPEG-2 transport stream. The first four AES3 streams containing audio ch

44、annels 0-7 could be carried by one 302M stream with channel_identification set to a value of 0, and the final two AES3 streams containing audio channels 8-11 could be carried by a second 302M stream with channel_identification set to a value of 8. 6.7 The SMPTE AES3 data header data shall be as defi

45、ned in table 1. (Note that uimsbf means unsigned integer with most significant bit first, as used in ISO/IEC 13818-1.) Table 1 SMPTE AES3 data elementary stream header audio_packet_size 16 uimsbf 1)SMPTE AES3 payload packet size in bytes exclusive of SMPTE header number_channels 2 uimsbf 00 = 2 data

46、 channels 01 = 4 data channels 10 = 6 data channels 11 = 8 data channels channel_identification 8 uimsbf channel number of first data channel 2)bits_per_sample 2 uimsbf 00 = 16 bits/data word 01 = 20 bits/data word 10 = 24 bits/data word 11 = reserved alignment bits 4 uimsbf SMPTE header byte alignm

47、ent (reserved, set to 0) NOTES 1)Unsigned integer with most significant bit first 2)See 6.6 SMPTE 302M-2007 Page 7 of 9 pages 6.8 SMPTE AES3 data PES packets shall be integer numbers of bytes in length. 6.9 Groups of AES3 data words associated with individual video frames shall be collected into AES

48、3 access units corresponding to the video frames. The payload of each AES3 data PES packet shall contain one AES3 access unit. The PES packet payload length shall be set according to the size of the associated AES3 access unit. In the case of video frame rates which do not have a simple integer rela

49、tionship to the AES3 clock, not all AES3 access units will contain the same number of AES3 data words. For example, ANSI/SMPTE 272M specifies that, at 29.97 video frames per second and with 48-ksamples/s audio, the audio samples per frame of 8008/5 will be in a sequence of 1602, 1601, 1602, 1601, 1602 samples per frame. In the case of 25 video frames per second, an integer number of audio samples per frame results (1920). 6.10 Each PES packet shall carry a presentation time stamp (PTS). The value of the PTS shall matc