idnits 2.17.1 draft-ietf-avt-rtp-rfc3984bis-05.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** There are 25 instances of too long lines in the document, the longest one being 1 character in excess of 72. == There are 1 instance of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. -- The abstract seems to indicate that this document obsoletes RFC3984, but the header doesn't have an 'Obsoletes:' line to match this. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == The document seems to use 'NOT RECOMMENDED' as an RFC 2119 keyword, but does not include the phrase in its RFC 2119 key words list. -- The exact meaning of the all-uppercase expression 'NOT REQUIRED' is not defined in RFC 2119. If it is intended as a requirements expression, it should be rewritten using one of the combinations defined in RFC 2119; otherwise it should not be all-uppercase. == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'MUST not' in this paragraph: The answerer MAY or MAY not include "sprop-parameter-sets", i.e., the answerer MAY use either out-of-band or in-band transport of parameter sets for the stream it is sending, regardless of whether out-of-band parameter sets transport has been used in the offerer-to-answerer direction. When the offer includes "in-band-parameter-sets" equal to 1, the answerer MUST not include "sprop-parameter-sets" and MUST transmit parameter sets in-band. All parameter sets included in the "sprop-parameter-sets", when present, for the accepted payload type in an answer MUST be associated with the accepted level, as indicated by the profile-level-id in the answer for the accepted payload type. -- The document seems to contain a disclaimer for pre-RFC5378 work, and may have content which was first submitted before 10 November 2008. The disclaimer is necessary when there are original authors that you have been unable to contact, or if some do not wish to grant the BCP78 rights to the IETF Trust. If you are able to get all authors (current and original) to grant those rights, you can and should remove the disclaimer; otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (April 22, 2009) is 5483 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Unused Reference: '4' is defined on line 4120, but no explicit reference was found in the text -- Possible downref: Non-RFC (?) normative reference: ref. '1' -- Possible downref: Non-RFC (?) normative reference: ref. '2' -- Possible downref: Non-RFC (?) normative reference: ref. '3' ** Obsolete normative reference: RFC 4566 (ref. '6') (Obsoleted by RFC 8866) ** Obsolete normative reference: RFC 3548 (ref. '7') (Obsoleted by RFC 4648) -- Obsolete informational reference (is this intentional?): RFC 2429 (ref. '11') (Obsoleted by RFC 4629) -- Obsolete informational reference (is this intentional?): RFC 2326 (ref. '27') (Obsoleted by RFC 7826) -- Obsolete informational reference (is this intentional?): RFC 5117 (ref. '29') (Obsoleted by RFC 7667) Summary: 4 errors (**), 0 flaws (~~), 5 warnings (==), 10 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Audio/Video Transport WG Y.-K. Wang 2 Internet Draft Huawei Technologies 3 Intended status: Standards track R. Even 4 Expires: October 2009 Self-employed 5 T. Kristensen 6 Tandberg 7 April 22, 2009 9 RTP Payload Format for H.264 Video 10 draft-ietf-avt-rtp-rfc3984bis-05.txt 12 Status of this Memo 14 This Internet-Draft is submitted to IETF in full conformance with the 15 provisions of BCP 78 and BCP 79. This document may contain material 16 from IETF Documents or IETF Contributions published or made publicly 17 available before November 10, 2008. The person(s) controlling the 18 copyright in some of this material may not have granted the IETF 19 Trust the right to allow modifications of such material outside the 20 IETF Standards Process. Without obtaining an adequate license from 21 the person(s) controlling the copyright in such materials, this 22 document may not be modified outside the IETF Standards Process, and 23 derivative works of it may not be created outside the IETF Standards 24 Process, except to format it for publication as an RFC or to 25 translate it into languages other than English. 27 Internet-Drafts are working documents of the Internet Engineering 28 Task Force (IETF), its areas, and its working groups. Note that 29 other groups may also distribute working documents as Internet-Drafts. 31 Internet-Drafts are draft documents valid for a maximum of six months 32 and may be updated, replaced, or obsoleted by other documents at any 33 time. It is inappropriate to use Internet-Drafts as reference 34 material or to cite them other than as "work in progress." 36 The list of current Internet-Drafts can be accessed at 37 http://www.ietf.org/ietf/1id-abstracts.txt. 39 The list of Internet-Draft Shadow Directories can be accessed at 40 http://www.ietf.org/shadow.html. 42 This Internet-Draft will expire on October 22, 2009. 44 Copyright Notice 46 Copyright (c) 2009 IETF Trust and the persons identified as the 47 document authors. All rights reserved. 49 This document is subject to BCP 78 and the IETF Trust's Legal 50 Provisions Relating to IETF Documents in effect on the date of 51 publication of this document (http://trustee.ietf.org/license-info). 52 Please review these documents carefully, as they describe your rights 53 and restrictions with respect to this document. 55 Abstract 57 This memo describes an RTP Payload format for the ITU-T 58 Recommendation H.264 video codec and the technically identical 59 ISO/IEC International Standard 14496-10 video codec, excluding the 60 Scalable Video Coding (SVC) extension and the Multivew Video Coding 61 extension, for which the RTP payload formats are defined elsewhere. 62 The RTP payload format allows for packetization of one or more 63 Network Abstraction Layer Units (NALUs), produced by an H.264 video 64 encoder, in each RTP payload. The payload format has wide 65 applicability, as it supports applications from simple low bit-rate 66 conversational usage, to Internet video streaming with interleaved 67 transmission, to high bit-rate video-on-demand. 69 This memo obsoletes RFC 3984. Changes from RFC 3984 are summarized 70 in section 18. Issues on backward compatibility to RFC 3984 are 71 discussed in section 17. 73 Table of Contents 75 1. Introduction...................................................4 76 1.1. The H.264 Codec...........................................4 77 1.2. Parameter Set Concept.....................................5 78 1.3. Network Abstraction Layer Unit Types......................6 79 2. Conventions....................................................7 80 3. Scope..........................................................7 81 4. Definitions and Abbreviations..................................8 82 4.1. Definitions...............................................8 83 4.2. Abbreviations............................................10 84 5. RTP Payload Format............................................10 85 5.1. RTP Header Usage.........................................10 86 5.2. Payload Structures.......................................13 87 5.3. NAL Unit Header Usage....................................14 88 5.4. Packetization Modes......................................17 89 5.5. Decoding Order Number (DON)..............................18 90 5.6. Single NAL Unit Packet...................................20 91 5.7. Aggregation Packets......................................21 92 5.7.1. Single-Time Aggregation Packet......................23 93 5.7.2. Multi-Time Aggregation Packets (MTAPs)..............25 94 5.7.3. Fragmentation Units (FUs)...........................29 95 6. Packetization Rules...........................................33 96 6.1. Common Packetization Rules...............................33 97 6.2. Single NAL Unit Mode.....................................34 98 6.3. Non-Interleaved Mode.....................................34 99 6.4. Interleaved Mode.........................................34 100 7. De-Packetization Process......................................35 101 7.1. Single NAL Unit and Non-Interleaved Mode.................35 102 7.2. Interleaved Mode.........................................35 103 7.2.1. Size of the De-interleaving Buffer..................36 104 7.2.2. De-interleaving Process.............................36 105 7.3. Additional De-Packetization Guidelines...................38 106 8. Payload Format Parameters.....................................39 107 8.1. Media Type Registration..................................39 108 8.2. SDP Parameters...........................................56 109 8.2.1. Mapping of Payload Type Parameters to SDP...........56 110 8.2.2. Usage with the SDP Offer/Answer Model...............57 111 8.2.3. Usage in Declarative Session Descriptions...........64 112 8.3. Examples.................................................65 113 8.4. Parameter Set Considerations.............................72 114 8.5. Decoder Refresh Point Procedure using In-Band Transport of 115 Parameter Sets (Informative)..................................74 116 8.5.1. IDR Procedure to Respond to a Request for a Decoder 117 Refresh Point..............................................75 118 8.5.2. Gradual Recovery Procedure to Respond to a Request for a 119 Decoder Refresh Point......................................75 120 9. Security Considerations.......................................76 121 10. Congestion Control...........................................77 122 11. IANA Consideration...........................................77 123 12. Informative Appendix: Application Examples...................78 124 12.1. Video Telephony according to ITU-T Recommendation H.241 125 Annex A.......................................................78 126 12.2. Video Telephony, No Slice Data Partitioning, No NAL Unit 127 Aggregation...................................................78 128 12.3. Video Telephony, Interleaved Packetization Using NAL Unit 129 Aggregation...................................................79 130 12.4. Video Telephony with Data Partitioning..................79 131 12.5. Video Telephony or Streaming with FUs and Forward Error 132 Correction....................................................80 133 12.6. Low Bit-Rate Streaming..................................82 134 12.7. Robust Packet Scheduling in Video Streaming.............83 136 13. Informative Appendix: Rationale for Decoding Order Number....84 137 13.1. Introduction............................................84 138 13.2. Example of Multi-Picture Slice Interleaving.............84 139 13.3. Example of Robust Packet Scheduling.....................86 140 13.4. Robust Transmission Scheduling of Redundant Coded Slices89 141 13.5. Remarks on Other Design Possibilities...................90 142 14. Acknowledgements.............................................91 143 15. References...................................................91 144 15.1. Normative References....................................91 145 15.2. Informative References..................................92 146 16. Authors' Addresses...........................................94 147 17. Backward Compatibility to RFC 3984...........................94 148 18. Changes from RFC 3984........................................96 150 1. Introduction 152 This memo specifies an RTP payload specification for the video coding 153 standard known as ITU-T Recommendation H.264 [1] and ISO/IEC 154 International Standard 14496 Part 10 [2] (both also known as Advanced 155 Video Coding, or AVC). In this memo the name H.264 is used for the 156 codec and the standard, but the memo is equally applicable to the 157 ISO/IEC counterpart of the coding standard. 159 This memo obsoletes RFC 3984. Changes from RFC 3984 are summarized 160 in section 18. Issues on backward compatibility to RFC 3984 are 161 discussed in section 17. 163 1.1. The H.264 Codec 165 The H.264 video codec has a very broad application range that covers 166 all forms of digital compressed video, from low bit-rate Internet 167 streaming applications to HDTV broadcast and Digital Cinema 168 applications with nearly lossless coding. Compared to the current 169 state of technology, the overall performance of H.264 is such that 170 bit rate savings of 50% or more are reported. Digital Satellite TV 171 quality, for example, was reported to be achievable at 1.5 Mbit/s, 172 compared to the current operation point of MPEG 2 video at around 3.5 173 Mbit/s [10]. 175 The codec specification [1] itself distinguishes conceptually between 176 a video coding layer (VCL) and a network abstraction layer (NAL). 177 The VCL contains the signal processing functionality of the codec; 178 mechanisms such as transform, quantization, and motion compensated 179 prediction; and a loop filter. It follows the general concept of 180 most of today's video codecs, a macroblock-based coder that uses 181 inter picture prediction with motion compensation and transform 182 coding of the residual signal. The VCL encoder outputs slices: a bit 183 string that contains the macroblock data of an integer number of 184 macroblocks, and the information of the slice header (containing the 185 spatial address of the first macroblock in the slice, the initial 186 quantization parameter, and similar information). Macroblocks in 187 slices are arranged in scan order unless a different macroblock 188 allocation is specified, by using the so-called Flexible Macroblock 189 Ordering syntax. In-picture prediction is used only within a slice. 190 More information is provided in [10]. 192 The Network Abstraction Layer (NAL) encoder encapsulates the slice 193 output of the VCL encoder into Network Abstraction Layer Units (NAL 194 units), which are suitable for transmission over packet networks or 195 use in packet oriented multiplex environments. Annex B of H.264 196 defines an encapsulation process to transmit such NAL units over 197 byte-stream oriented networks. In the scope of this memo, Annex B is 198 not relevant. 200 Internally, the NAL uses NAL units. A NAL unit consists of a one- 201 byte header and the payload byte string. The header indicates the 202 type of the NAL unit, the (potential) presence of bit errors or 203 syntax violations in the NAL unit payload, and information regarding 204 the relative importance of the NAL unit for the decoding process. 205 This RTP payload specification is designed to be unaware of the bit 206 string in the NAL unit payload. 208 One of the main properties of H.264 is the complete decoupling of the 209 transmission time, the decoding time, and the sampling or 210 presentation time of slices and pictures. The decoding process 211 specified in H.264 is unaware of time, and the H.264 syntax does not 212 carry information such as the number of skipped frames (as is common 213 in the form of the Temporal Reference in earlier video compression 214 standards). Also, there are NAL units that affect many pictures and 215 that are, therefore, inherently timeless. For this reason, the 216 handling of the RTP timestamp requires some special considerations 217 for NAL units for which the sampling or presentation time is not 218 defined or, at transmission time, unknown. 220 1.2. Parameter Set Concept 222 One very fundamental design concept of H.264 is to generate self- 223 contained packets, to make mechanisms such as the header duplication 224 of RFC 2429 [11] or MPEG-4 Visual's Header Extension Code (HEC) [12] 225 unnecessary. This was achieved by decoupling information relevant to 226 more than one slice from the media stream. This higher layer meta 227 information should be sent reliably, asynchronously, and in advance 228 from the RTP packet stream that contains the slice packets. 229 (Provisions for sending this information in-band are also available 230 for applications that do not have an out-of-band transport channel 231 appropriate for the purpose.) The combination of the higher-level 232 parameters is called a parameter set. The H.264 specification 233 includes two types of parameter sets: sequence parameter set and 234 picture parameter set. An active sequence parameter set remains 235 unchanged throughout a coded video sequence, and an active picture 236 parameter set remains unchanged within a coded picture. The sequence 237 and picture parameter set structures contain information such as 238 picture size, optional coding modes employed, and macroblock to slice 239 group map. 241 To be able to change picture parameters (such as the picture size) 242 without having to transmit parameter set updates synchronously to the 243 slice packet stream, the encoder and decoder can maintain a list of 244 more than one sequence and picture parameter set. Each slice header 245 contains a codeword that indicates the sequence and picture parameter 246 set to be used. 248 This mechanism allows the decoupling of the transmission of parameter 249 sets from the packet stream, and the transmission of them by external 250 means (e.g., as a side effect of the capability exchange), or through 251 a (reliable or unreliable) control protocol. It may even be possible 252 that they are never transmitted but are fixed by an application 253 design specification. 255 1.3. Network Abstraction Layer Unit Types 257 Tutorial information on the NAL design can be found in [13], [14], 258 and [15]. 260 All NAL units consist of a single NAL unit type octet, which also co- 261 serves as the payload header of this RTP payload format. The payload 262 of a NAL unit follows immediately. 264 The syntax and semantics of the NAL unit type octet are specified in 265 [1], but the essential properties of the NAL unit type octet are 266 summarized below. The NAL unit type octet has the following format: 268 +---------------+ 269 |0|1|2|3|4|5|6|7| 270 +-+-+-+-+-+-+-+-+ 271 |F|NRI| Type | 272 +---------------+ 274 The semantics of the components of the NAL unit type octet, as 275 specified in the H.264 specification, are described briefly below. 277 F: 1 bit 278 forbidden_zero_bit. The H.264 specification declares a value of 279 1 as a syntax violation. 281 NRI: 2 bits 282 nal_ref_idc. A value of 00 indicates that the content of the NAL 283 unit is not used to reconstruct reference pictures for inter 284 picture prediction. Such NAL units can be discarded without 285 risking the integrity of the reference pictures. Values greater 286 than 00 indicate that the decoding of the NAL unit is required to 287 maintain the integrity of the reference pictures. 289 Type: 5 bits 290 nal_unit_type. This component specifies the NAL unit payload 291 type as defined in Table 7-1 of [1], and later within this memo. 292 For a reference of all currently defined NAL unit types and their 293 semantics, please refer to section 7.4.1 in [1]. 295 This memo introduces new NAL unit types, which are presented in 296 section 5.2. The NAL unit types defined in this memo are marked as 297 unspecified in [1]. Moreover, this specification extends the 298 semantics of F and NRI as described in section 5.3. 300 2. Conventions 302 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 303 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 304 document are to be interpreted as described in RFC-2119 [3]. 306 This specification uses the notion of setting and clearing a bit when 307 bit fields are handled. Setting a bit is the same as assigning that 308 bit the value of 1 (On). Clearing a bit is the same as assigning 309 that bit the value of 0 (Off). 311 3. Scope 313 This payload specification can only be used to carry the "naked" 314 H.264 NAL unit stream over RTP, and not the bitstream format 315 discussed in Annex B of H.264. Likely, the first applications of 316 this specification will be in the conversational multimedia field, 317 video telephony or video conferencing, but the payload format also 318 covers other applications, such as Internet streaming and TV over IP. 320 4. Definitions and Abbreviations 322 4.1. Definitions 324 This document uses the definitions of [1]. The following terms, 325 defined in [1], are summed up for convenience: 327 access unit: A set of NAL units always containing a primary coded 328 picture. In addition to the primary coded picture, an access 329 unit may also contain one or more redundant coded pictures or 330 other NAL units not containing slices or slice data partitions of 331 a coded picture. The decoding of an access unit always results 332 in a decoded picture. 334 coded video sequence: A sequence of access units that consists, 335 in decoding order, of an instantaneous decoding refresh (IDR) 336 access unit followed by zero or more non-IDR access units 337 including all subsequent access units up to but not including any 338 subsequent IDR access unit. 340 IDR access unit: An access unit in which the primary coded 341 picture is an IDR picture. 343 IDR picture: A coded picture containing only slices with I or SI 344 slice types that causes a "reset" in the decoding process. After 345 the decoding of an IDR picture, all following coded pictures in 346 decoding order can be decoded without inter prediction from any 347 picture decoded prior to the IDR picture. 349 primary coded picture: The coded representation of a picture to 350 be used by the decoding process for a bitstream conforming to 351 H.264. The primary coded picture contains all macroblocks of the 352 picture. 354 redundant coded picture: A coded representation of a picture or a 355 part of a picture. The content of a redundant coded picture 356 shall not be used by the decoding process for a bitstream 357 conforming to H.264. The content of a redundant coded picture 358 may be used by the decoding process for a bitstream that contains 359 errors or losses. 361 VCL NAL unit: A collective term used to refer to coded slice and 362 coded data partition NAL units. 364 In addition, the following definitions apply: 366 decoding order number (DON): A field in the payload structure or 367 a derived variable indicating NAL unit decoding order. Values of 368 DON are in the range of 0 to 65535, inclusive. After reaching 369 the maximum value, the value of DON wraps around to 0. 371 NAL unit decoding order: A NAL unit order that conforms to the 372 constraints on NAL unit order given in section 7.4.1.2 in [1]. 374 NALU-time: The value that the RTP timestamp would have if the NAL 375 unit would be transported in its own RTP packet. 377 transmission order: The order of packets in ascending RTP 378 sequence number order (in modulo arithmetic). Within an 379 aggregation packet, the NAL unit transmission order is the same 380 as the order of appearance of NAL units in the packet. 382 media aware network element (MANE): A network element, such as a 383 middlebox or application layer gateway that is capable of parsing 384 certain aspects of the RTP payload headers or the RTP payload and 385 reacting to the contents. 387 Informative note: The concept of a MANE goes beyond normal 388 routers or gateways in that a MANE has to be aware of the 389 signaling (e.g., to learn about the payload type mappings of 390 the media streams), and in that it has to be trusted when 391 working with SRTP. The advantage of using MANEs is that they 392 allow packets to be dropped according to the needs of the 393 media coding. For example, if a MANE has to drop packets due 394 to congestion on a certain link, it can identify and remove 395 those packets whose elimination produces the least adverse 396 effect on the user experience. 398 static macroblock: A certain amount of macroblocks in the video 399 stream can be defined as static, as defined in section 8.3.2.8 in 400 [3]. Static macroblocks free up additional processing cycles for 401 the handling of non-static macroblocks. Based on a given amount 402 of video processing resources and a given resolution, a higher 403 number of static macroblocks enables a correspondingly higher 404 frame rate. 406 default sub-profile: The subset of coding tools, which may be all 407 coding tools of one profile or the common subset of coding tools 408 of more than one profile, indicated by the profile-level-id 409 parameter. 411 default level: The level indicated by the profile-level-id 412 parameter, which consists of three octets, profile_idc, profile- 413 iop, and level_idc. The default level is indicated by level_idc 414 in most cases, and, in some cases, additionally by profile-iop. 416 4.2. Abbreviations 418 DON: Decoding Order Number 419 DONB: Decoding Order Number Base 420 DOND: Decoding Order Number Difference 421 FEC: Forward Error Correction 422 FU: Fragmentation Unit 423 IDR: Instantaneous Decoding Refresh 424 IEC: International Electrotechnical Commission 425 ISO: International Organization for Standardization 426 ITU-T: International Telecommunication Union, 427 Telecommunication Standardization Sector 428 MANE: Media Aware Network Element 429 MTAP: Multi-Time Aggregation Packet 430 MTAP16: MTAP with 16-bit timestamp offset 431 MTAP24: MTAP with 24-bit timestamp offset 432 NAL: Network Abstraction Layer 433 NALU: NAL Unit 434 SAR: Sample Aspect Ratio 435 SEI: Supplemental Enhancement Information 436 STAP: Single-Time Aggregation Packet 437 STAP-A: STAP type A 438 STAP-B: STAP type B 439 TS: Timestamp 440 VCL: Video Coding Layer 441 VUI: Video Usability Information 443 5. RTP Payload Format 445 5.1. RTP Header Usage 447 The format of the RTP header is specified in RFC 3550 [5] and 448 reprinted in Figure 1 for convenience. This payload format uses the 449 fields of the header in a manner consistent with that specification. 451 When one NAL unit is encapsulated per RTP packet, the RECOMMENDED RTP 452 payload format is specified in section 5.6. The RTP payload (and the 453 settings for some RTP header bits) for aggregation packets and 454 fragmentation units are specified in sections 5.7 and 5.8, 455 respectively. 457 0 1 2 3 458 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 459 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 460 |V=2|P|X| CC |M| PT | sequence number | 461 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 462 | timestamp | 463 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 464 | synchronization source (SSRC) identifier | 465 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 466 | contributing source (CSRC) identifiers | 467 | .... | 468 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 470 Figure 1 RTP header according to RFC 3550 472 The RTP header information to be set according to this RTP payload 473 format is set as follows: 475 Marker bit (M): 1 bit 476 Set for the very last packet of the access unit indicated by the 477 RTP timestamp, in line with the normal use of the M bit in video 478 formats, to allow an efficient playout buffer handling. For 479 aggregation packets (STAP and MTAP), the marker bit in the RTP 480 header MUST be set to the value that the marker bit of the last 481 NAL unit of the aggregation packet would have been if it were 482 transported in its own RTP packet. Decoders MAY use this bit as 483 an early indication of the last packet of an access unit, but 484 MUST NOT rely on this property. 486 Informative note: Only one M bit is associated with an 487 aggregation packet carrying multiple NAL units. Thus, if a 488 gateway has re-packetized an aggregation packet into several 489 packets, it cannot reliably set the M bit of those packets. 491 Payload type (PT): 7 bits 492 The assignment of an RTP payload type for this new packet format 493 is outside the scope of this document and will not be specified 494 here. The assignment of a payload type has to be performed 495 either through the profile used or in a dynamic way. 497 Sequence number (SN): 16 bits 498 Set and used in accordance with RFC 3550. For the single NALU 499 and non-interleaved packetization mode, the sequence number is 500 used to determine decoding order for the NALU. 502 Timestamp: 32 bits 503 The RTP timestamp is set to the sampling timestamp of the content. 504 A 90 kHz clock rate MUST be used. 506 If the NAL unit has no timing properties of its own (e.g., 507 parameter set and SEI NAL units), the RTP timestamp is set to the 508 RTP timestamp of the primary coded picture of the access unit in 509 which the NAL unit is included, according to section 7.4.1.2 of 510 [1]. 512 The setting of the RTP Timestamp for MTAPs is defined in section 513 5.7.2. 515 Receivers SHOULD ignore any picture timing SEI messages included 516 in access units that have only one display timestamp. Instead, 517 receivers SHOULD use the RTP timestamp for synchronizing the 518 display process. 520 RTP senders SHOULD NOT transmit picture timing SEI messages for 521 pictures that are not supposed to be displayed as multiple fields. 523 If one access unit has more than one display timestamp carried in 524 a picture timing SEI message, then the information in the SEI 525 message SHOULD be treated as relative to the RTP timestamp, with 526 the earliest event occurring at the time given by the RTP 527 timestamp, and subsequent events later, as given by the 528 difference in SEI message picture timing values. Let tSEI1, 529 tSEI2, ..., tSEIn be the display timestamps carried in the SEI 530 message of an access unit, where tSEI1 is the earliest of all 531 such timestamps. Let tmadjst() be a function that adjusts the 532 SEI messages time scale to a 90-kHz time scale. Let TS be the 533 RTP timestamp. Then, the display time for the event associated 534 with tSEI1 is TS. The display time for the event with tSEIx, 535 where x is [2..n] is TS + tmadjst (tSEIx - tSEI1). 537 Informative note: Displaying coded frames as fields is needed 538 commonly in an operation known as 3:2 pulldown, in which film 539 content that consists of coded frames is displayed on a 540 display using interlaced scanning. The picture timing SEI 541 message enables carriage of multiple timestamps for the same 542 coded picture, and therefore the 3:2 pulldown process is 543 perfectly controlled. The picture timing SEI message 544 mechanism is necessary because only one timestamp per coded 545 frame can be conveyed in the RTP timestamp. 547 Informative note: Because H.264 allows the decoding order to 548 be different from the display order, values of RTP timestamps 549 may not be monotonically non-decreasing as a function of RTP 550 sequence numbers. Furthermore, the value for inter-arrival 551 jitter reported in the RTCP reports may not be a trustworthy 552 indication of the network performance, as the calculation 553 rules for inter-arrival jitter (section 6.4.1 of RFC 3550) 554 assume that the RTP timestamp of a packet is directly 555 proportional to its transmission time. 557 5.2. Payload Structures 559 The payload format defines three different basic payload structures. 560 A receiver can identify the payload structure by the first byte of 561 the RTP packet payload, which co-serves as the RTP payload header and, 562 in some cases, as the first byte of the payload. This byte is always 563 structured as a NAL unit header. The NAL unit type field indicates 564 which structure is present. The possible structures are as follows: 566 Single NAL Unit Packet: Contains only a single NAL unit in the 567 payload. The NAL header type field will be equal to the original NAL 568 unit type; i.e., in the range of 1 to 23, inclusive. Specified in 569 section 5.6. 571 Aggregation Packet: Packet type used to aggregate multiple NAL units 572 into a single RTP payload. This packet exists in four versions, the 573 Single-Time Aggregation Packet type A (STAP-A), the Single-Time 574 Aggregation Packet type B (STAP-B), Multi-Time Aggregation Packet 575 (MTAP) with 16-bit offset (MTAP16), and Multi-Time Aggregation Packet 576 (MTAP) with 24-bit offset (MTAP24). The NAL unit type numbers 577 assigned for STAP-A, STAP-B, MTAP16, and MTAP24 are 24, 25, 26, and 578 27, respectively. Specified in section 5.7. 580 Fragmentation Unit: Used to fragment a single NAL unit over multiple 581 RTP packets. Exists with two versions, FU-A and FU-B, identified 582 with the NAL unit type numbers 28 and 29, respectively. Specified in 583 section 5.8. 585 Informative note: This specification does not limit the size of 586 NAL units encapsulated in single NAL unit packets and 587 fragmentation units. The maximum size of a NAL unit encapsulated 588 in any aggregation packet is 65535 bytes. 590 Table 1 summarizes NAL unit types and the corresponding RTP packet 591 types when each of these NAL units is directly used as a packet 592 payload, and where the types are described in this memo. 594 Table 1. Summary of NAL unit types and the corresponding packet 595 types 597 NAL Unit Packet Packet Type Name Section 598 Type Type 599 --------------------------------------------------------- 600 0 reserved - 601 1-23 NAL unit Single NAL unit packet 5.6 602 24 STAP-A Single-time aggregation packet 5.7.1 603 25 STAP-B Single-time aggregation packet 5.7.1 604 26 MTAP16 Multi-time aggregation packet 5.7.2 605 27 MTAP24 Multi-time aggregation packet 5.7.2 606 28 FU-A Fragmentation unit 5.8 607 29 FU-B Fragmentation unit 5.8 608 30-31 reserved - 610 5.3. NAL Unit Header Usage 612 The structure and semantics of the NAL unit header were introduced in 613 section 1.3. For convenience, the format of the NAL unit header is 614 reprinted below: 616 +---------------+ 617 |0|1|2|3|4|5|6|7| 618 +-+-+-+-+-+-+-+-+ 619 |F|NRI| Type | 620 +---------------+ 622 This section specifies the semantics of F and NRI according to this 623 specification. 625 F: 1 bit 626 forbidden_zero_bit. A value of 0 indicates that the NAL unit 627 type octet and payload should not contain bit errors or other 628 syntax violations. A value of 1 indicates that the NAL unit type 629 octet and payload may contain bit errors or other syntax 630 violations. 632 MANEs SHOULD set the F bit to indicate detected bit errors in the 633 NAL unit. The H.264 specification requires that the F bit is 634 equal to 0. When the F bit is set, the decoder is advised that 635 bit errors or any other syntax violations may be present in the 636 payload or in the NAL unit type octet. The simplest decoder 637 reaction to a NAL unit in which the F bit is equal to 1 is to 638 discard such a NAL unit and to conceal the lost data in the 639 discarded NAL unit. 641 NRI: 2 bits 642 nal_ref_idc. The semantics of value 00 and a non-zero value 643 remain unchanged from the H.264 specification. In other words, a 644 value of 00 indicates that the content of the NAL unit is not 645 used to reconstruct reference pictures for inter picture 646 prediction. Such NAL units can be discarded without risking the 647 integrity of the reference pictures. Values greater than 00 648 indicate that the decoding of the NAL unit is required to 649 maintain the integrity of the reference pictures. 651 In addition to the specification above, according to this RTP 652 payload specification, values of NRI indicate the relative 653 transport priority, as determined by the encoder. MANEs can use 654 this information to protect more important NAL units better than 655 they do less important NAL units. The highest transport priority 656 is 11, followed by 10, and then by 01; finally, 00 is the lowest. 658 Informative note: Any non-zero value of NRI is handled 659 identically in H.264 decoders. Therefore, receivers need not 660 manipulate the value of NRI when passing NAL units to the 661 decoder. 663 An H.264 encoder MUST set the value of NRI according to the H.264 664 specification (subclause 7.4.1) when the value of nal_unit_type 665 is in the range of 1 to 12, inclusive. In particular, the H.264 666 specification requires that the value of NRI SHALL be equal to 0 667 for all NAL units having nal_unit_type equal to 6, 9, 10, 11, or 668 12. 670 For NAL units having nal_unit_type equal to 7 or 8 (indicating a 671 sequence parameter set or a picture parameter set, respectively), 672 an H.264 encoder SHOULD set the value of NRI to 11 (in binary 673 format). For coded slice NAL units of a primary coded picture 674 having nal_unit_type equal to 5 (indicating a coded slice 675 belonging to an IDR picture), an H.264 encoder SHOULD set the 676 value of NRI to 11 (in binary format). 678 For a mapping of the remaining nal_unit_types to NRI values, the 679 following example MAY be used and has been shown to be efficient 680 in a certain environment [14]. Other mappings MAY also be 681 desirable, depending on the application and the H.264/AVC Annex A 682 profile in use. 684 Informative note: Data Partitioning is not available in 685 certain profiles; e.g., in the Main or Baseline profiles. 686 Consequently, the NAL unit types 2, 3, and 4 can occur only if 687 the video bitstream conforms to a profile in which data 688 partitioning is allowed and not in streams that conform to the 689 Main or Baseline profiles. 691 Table 2. Example of NRI values for coded slices and coded slice data 692 partitions of primary coded reference pictures 694 NAL Unit Type Content of NAL unit NRI (binary) 695 ---------------------------------------------------------------- 696 1 non-IDR coded slice 10 697 2 Coded slice data partition A 10 698 3 Coded slice data partition B 01 699 4 Coded slice data partition C 01 701 Informative note: As mentioned before, the NRI value of non- 702 reference pictures is 00 as mandated by H.264/AVC. 704 An H.264 encoder SHOULD set the value of NRI for coded slice and 705 coded slice data partition NAL units of redundant coded reference 706 pictures equal to 01 (in binary format). 708 Definitions of the values for NRI for NAL unit types 24 to 29, 709 inclusive, are given in sections 5.7 and 5.8 of this memo. 711 No recommendation for the value of NRI is given for NAL units 712 having nal_unit_type in the range of 13 to 23, inclusive, because 713 these values are reserved for ITU-T and ISO/IEC. No 714 recommendation for the value of NRI is given for NAL units having 715 nal_unit_type equal to 0 or in the range of 30 to 31, inclusive, 716 as the semantics of these values are not specified in this memo. 718 5.4. Packetization Modes 720 This memo specifies three cases of packetization modes: 722 o Single NAL unit mode 724 o Non-interleaved mode 726 o Interleaved mode 728 The single NAL unit mode is targeted for conversational systems that 729 comply with ITU-T Recommendation H.241 [3] (see section 12.1). The 730 non-interleaved mode is targeted for conversational systems that may 731 not comply with ITU-T Recommendation H.241. In the non-interleaved 732 mode, NAL units are transmitted in NAL unit decoding order. The 733 interleaved mode is targeted for systems that do not require very low 734 end-to-end latency. The interleaved mode allows transmission of NAL 735 units out of NAL unit decoding order. 737 The packetization mode in use MAY be signaled by the value of the 738 OPTIONAL packetization-mode media type parameter. The used 739 packetization mode governs which NAL unit types are allowed in RTP 740 payloads. Table 3 summarizes the allowed packet payload types for 741 each packetization mode. Packetization modes are explained in more 742 detail in section 6. 744 Table 3. Summary of allowed NAL unit types for each packetization 745 mode (yes = allowed, no = disallowed, ig = ignore) 747 Payload Packet Single NAL Non-Interleaved Interleaved 748 Type Type Unit Mode Mode Mode 749 ------------------------------------------------------------- 750 0 reserved ig ig ig 751 1-23 NAL unit yes yes no 752 24 STAP-A no yes no 753 25 STAP-B no no yes 754 26 MTAP16 no no yes 755 27 MTAP24 no no yes 756 28 FU-A no yes yes 757 29 FU-B no no yes 758 30-31 reserved ig ig ig 760 Some NAL unit or payload type values (indicated as reserved in 761 Table 3) are reserved for future extensions. NAL units of those 762 types SHOULD NOT be sent by a sender (direct as packet payloads, or 763 as aggregation units in aggregation packets, or as fragmented units 764 in FU packets) and MUST be ignored by a receiver. For example, the 765 payload types 1-23, with the associated packet type "NAL unit", are 766 allowed in "Single NAL Unit Mode" and in "Non-Interleaved Mode", but 767 disallowed in "Interleaved Mode". However, NAL units of NAL unit 768 types 1-23 can be used in ''Interleaved Mode'' as aggregation units in 769 STAP-B, MTAP16 and MTAP14 packets as well as fragmented units in FU-A 770 and FU-B packets. Similarly, NAL units of NAL unit types 1-23 can 771 also be used in the "Non-Interleaved Mode" as aggregation units in 772 STAP-A packets or fragmented units in FU-A packets, in addition to 773 being directly used as packet payloads. 775 5.5. Decoding Order Number (DON) 777 In the interleaved packetization mode, the transmission order of NAL 778 units is allowed to differ from the decoding order of the NAL units. 779 Decoding order number (DON) is a field in the payload structure or a 780 derived variable that indicates the NAL unit decoding order. 781 Rationale and examples of use cases for transmission out of decoding 782 order and for the use of DON are given in section 13. 784 The coupling of transmission and decoding order is controlled by the 785 OPTIONAL sprop-interleaving-depth media type parameter as follows. 786 When the value of the OPTIONAL sprop-interleaving-depth media type 787 parameter is equal to 0 (explicitly or per default), the transmission 788 order of NAL units MUST conform to the NAL unit decoding order. When 789 the value of the OPTIONAL sprop-interleaving-depth media type 790 parameter is greater than 0, 792 o the order of NAL units in an MTAP16 and an MTAP24 is NOT REQUIRED 793 to be the NAL unit decoding order, and 795 o the order of NAL units generated by de-packetizing STAP-Bs, MTAPs, 796 and FUs in two consecutive packets is NOT REQUIRED to be the NAL 797 unit decoding order. 799 The RTP payload structures for a single NAL unit packet, an STAP-A, 800 and an FU-A do not include DON. STAP-B and FU-B structures include 801 DON, and the structure of MTAPs enables derivation of DON as 802 specified in section 5.7.2. 804 Informative note: When an FU-A occurs in interleaved mode, it 805 always follows an FU-B, which sets its DON. 807 Informative note: If a transmitter wants to encapsulate a single 808 NAL unit per packet and transmit packets out of their decoding 809 order, STAP-B packet type can be used. 811 In the single NAL unit packetization mode, the transmission order of 812 NAL units, determined by the RTP sequence number, MUST be the same as 813 their NAL unit decoding order. In the non-interleaved packetization 814 mode, the transmission order of NAL units in single NAL unit packets, 815 STAP-As, and FU-As MUST be the same as their NAL unit decoding order. 816 The NAL units within an STAP MUST appear in the NAL unit decoding 817 order. Thus, the decoding order is first provided through the 818 implicit order within a STAP, and second provided through the RTP 819 sequence number for the order between STAPs, FUs, and single NAL unit 820 packets. 822 Signaling of the value of DON for NAL units carried in STAP-B, MTAP, 823 and a series of fragmentation units starting with an FU-B is 824 specified in sections 5.7.1, 5.7.2, and 5.8, respectively. The DON 825 value of the first NAL unit in transmission order MAY be set to any 826 value. Values of DON are in the range of 0 to 65535, inclusive. 827 After reaching the maximum value, the value of DON wraps around to 0. 829 The decoding order of two NAL units contained in any STAP-B, MTAP, or 830 a series of fragmentation units starting with an FU-B is determined 831 as follows. Let DON(i) be the decoding order number of the NAL unit 832 having index i in the transmission order. Function don_diff(m,n) is 833 specified as follows: 835 If DON(m) == DON(n), don_diff(m,n) = 0 837 If (DON(m) < DON(n) and DON(n) - DON(m) < 32768), 838 don_diff(m,n) = DON(n) - DON(m) 840 If (DON(m) > DON(n) and DON(m) - DON(n) >= 32768), 841 don_diff(m,n) = 65536 - DON(m) + DON(n) 843 If (DON(m) < DON(n) and DON(n) - DON(m) >= 32768), 844 don_diff(m,n) = - (DON(m) + 65536 - DON(n)) 846 If (DON(m) > DON(n) and DON(m) - DON(n) < 32768), 847 don_diff(m,n) = - (DON(m) - DON(n)) 849 A positive value of don_diff(m,n) indicates that the NAL unit having 850 transmission order index n follows, in decoding order, the NAL unit 851 having transmission order index m. When don_diff(m,n) is equal to 0, 852 then the NAL unit decoding order of the two NAL units can be in 853 either order. A negative value of don_diff(m,n) indicates that the 854 NAL unit having transmission order index n precedes, in decoding 855 order, the NAL unit having transmission order index m. 857 Values of DON related fields (DON, DONB, and DOND; see section 5.7) 858 MUST be such that the decoding order determined by the values of DON, 859 as specified above, conforms to the NAL unit decoding order. If the 860 order of two NAL units in NAL unit decoding order is switched and the 861 new order does not conform to the NAL unit decoding order, the NAL 862 units MUST NOT have the same value of DON. If the order of two 863 consecutive NAL units in the NAL unit stream is switched and the new 864 order still conforms to the NAL unit decoding order, the NAL units 865 MAY have the same value of DON. For example, when arbitrary slice 866 order is allowed by the video coding profile in use, all the coded 867 slice NAL units of a coded picture are allowed to have the same value 868 of DON. Consequently, NAL units having the same value of DON can be 869 decoded in any order, and two NAL units having a different value of 870 DON should be passed to the decoder in the order specified above. 871 When two consecutive NAL units in the NAL unit decoding order have a 872 different value of DON, the value of DON for the second NAL unit in 873 decoding order SHOULD be the value of DON for the first, incremented 874 by one. 876 An example of the de-packetization process to recover the NAL unit 877 decoding order is given in section 7. 879 Informative note: Receivers should not expect that the absolute 880 difference of values of DON for two consecutive NAL units in the 881 NAL unit decoding order will be equal to one, even in error-free 882 transmission. An increment by one is not required, as at the 883 time of associating values of DON to NAL units, it may not be 884 known whether all NAL units are delivered to the receiver. For 885 example, a gateway may not forward coded slice NAL units of non- 886 reference pictures or SEI NAL units when there is a shortage of 887 bit rate in the network to which the packets are forwarded. In 888 another example, a live broadcast is interrupted by pre-encoded 889 content, such as commercials, from time to time. The first intra 890 picture of a pre-encoded clip is transmitted in advance to ensure 891 that it is readily available in the receiver. When transmitting 892 the first intra picture, the originator does not exactly know how 893 many NAL units will be encoded before the first intra picture of 894 the pre-encoded clip follows in decoding order. Thus, the values 895 of DON for the NAL units of the first intra picture of the pre- 896 encoded clip have to be estimated when they are transmitted, and 897 gaps in values of DON may occur. 899 5.6. Single NAL Unit Packet 901 The single NAL unit packet defined here MUST contain only one NAL 902 unit, of the types defined in [1]. This means that neither an 903 aggregation packet nor a fragmentation unit can be used within a 904 single NAL unit packet. A NAL unit stream composed by de-packetizing 905 single NAL unit packets in RTP sequence number order MUST conform to 906 the NAL unit decoding order. The structure of the single NAL unit 907 packet is shown in Figure 2. 909 Informative note: The first byte of a NAL unit co-serves as the 910 RTP payload header. 912 0 1 2 3 913 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 914 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 915 |F|NRI| Type | | 916 +-+-+-+-+-+-+-+-+ | 917 | | 918 | Bytes 2..n of a Single NAL unit | 919 | | 920 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 921 | :...OPTIONAL RTP padding | 922 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 924 Figure 2 RTP payload format for single NAL unit packet 926 5.7. Aggregation Packets 928 Aggregation packets are the NAL unit aggregation scheme of this 929 payload specification. The scheme is introduced to reflect the 930 dramatically different MTU sizes of two key target networks: wireline 931 IP networks (with an MTU size that is often limited by the Ethernet 932 MTU size; roughly 1500 bytes), and IP or non-IP (e.g., ITU-T H.324/M) 933 based wireless communication systems with preferred transmission unit 934 sizes of 254 bytes or less. To prevent media transcoding between the 935 two worlds, and to avoid undesirable packetization overhead, a NAL 936 unit aggregation scheme is introduced. 938 Two types of aggregation packets are defined by this specification: 940 o Single-time aggregation packet (STAP): aggregates NAL units with 941 identical NALU-time. Two types of STAPs are defined, one without 942 DON (STAP-A) and another including DON (STAP-B). 944 o Multi-time aggregation packet (MTAP): aggregates NAL units with 945 potentially differing NALU-time. Two different MTAPs are defined, 946 differing in the length of the NAL unit timestamp offset. 948 Each NAL unit to be carried in an aggregation packet is encapsulated 949 in an aggregation unit. Please see below for the four different 950 aggregation units and their characteristics. 952 The structure of the RTP payload format for aggregation packets is 953 presented in Figure 3. 955 0 1 2 3 956 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 957 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 958 |F|NRI| Type | | 959 +-+-+-+-+-+-+-+-+ | 960 | | 961 | one or more aggregation units | 962 | | 963 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 964 | :...OPTIONAL RTP padding | 965 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 967 Figure 3 RTP payload format for aggregation packets 969 MTAPs and STAPs share the following packetization rules: The RTP 970 timestamp MUST be set to the earliest of the NALU-times of all the 971 NAL units to be aggregated. The type field of the NAL unit type 972 octet MUST be set to the appropriate value, as indicated in Table 4. 973 The F bit MUST be cleared if all F bits of the aggregated NAL units 974 are zero; otherwise, it MUST be set. The value of NRI MUST be the 975 maximum of all the NAL units carried in the aggregation packet. 977 Table 4. Type field for STAPs and MTAPs 979 Type Packet Timestamp offset DON related fields 980 field length (DON, DONB, DOND) 981 (in bits) present 982 -------------------------------------------------------- 983 24 STAP-A 0 no 984 25 STAP-B 0 yes 985 26 MTAP16 16 yes 986 27 MTAP24 24 yes 988 The marker bit in the RTP header is set to the value that the marker 989 bit of the last NAL unit of the aggregated packet would have if it 990 were transported in its own RTP packet. 992 The payload of an aggregation packet consists of one or more 993 aggregation units. See sections 5.7.1 and 5.7.2 for the four 994 different types of aggregation units. An aggregation packet can 995 carry as many aggregation units as necessary; however, the total 996 amount of data in an aggregation packet obviously MUST fit into an IP 997 packet, and the size SHOULD be chosen so that the resulting IP packet 998 is smaller than the MTU size. An aggregation packet MUST NOT contain 999 fragmentation units specified in section 5.8. Aggregation packets 1000 MUST NOT be nested; i.e., an aggregation packet MUST NOT contain 1001 another aggregation packet. 1003 5.7.1. Single-Time Aggregation Packet 1005 Single-time aggregation packet (STAP) SHOULD be used whenever NAL 1006 units are aggregated that all share the same NALU-time. The payload 1007 of an STAP-A does not include DON and consists of at least one 1008 single-time aggregation unit, as presented in Figure 4. The payload 1009 of an STAP-B consists of a 16-bit unsigned decoding order number (DON) 1010 (in network byte order) followed by at least one single-time 1011 aggregation unit, as presented in Figure 5. 1013 0 1 2 3 1014 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1015 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1016 : | 1017 +-+-+-+-+-+-+-+-+ | 1018 | | 1019 | single-time aggregation units | 1020 | | 1021 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1022 | : 1023 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1025 Figure 4 Payload format for STAP-A 1027 0 1 2 3 1028 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1029 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1030 : decoding order number (DON) | | 1031 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1032 | | 1033 | single-time aggregation units | 1034 | | 1035 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1036 | : 1037 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1039 Figure 5 Payload format for STAP-B 1041 The DON field specifies the value of DON for the first NAL unit in an 1042 STAP-B in transmission order. For each successive NAL unit in 1043 appearance order in an STAP-B, the value of DON is equal to (the 1044 value of DON of the previous NAL unit in the STAP-B + 1) % 65536, in 1045 which '%' stands for the modulo operation. 1047 A single-time aggregation unit consists of 16-bit unsigned size 1048 information (in network byte order) that indicates the size of the 1049 following NAL unit in bytes (excluding these two octets, but 1050 including the NAL unit type octet of the NAL unit), followed by the 1051 NAL unit itself, including its NAL unit type byte. A single-time 1052 aggregation unit is byte aligned within the RTP payload, but it may 1053 not be aligned on a 32-bit word boundary. Figure 6 presents the 1054 structure of the single-time aggregation unit. 1056 0 1 2 3 1057 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1058 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1059 : NAL unit size | | 1060 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1061 | | 1062 | NAL unit | 1063 | | 1064 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1065 | : 1066 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1068 Figure 6 Structure for single-time aggregation unit 1070 Figure 7 presents an example of an RTP packet that contains an STAP-A. 1071 The STAP contains two single-time aggregation units, labeled as 1 and 1072 2 in the figure. 1074 0 1 2 3 1075 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1076 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1077 | RTP Header | 1078 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1079 |STAP-A NAL HDR | NALU 1 Size | NALU 1 HDR | 1080 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1081 | NALU 1 Data | 1082 : : 1083 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1084 | | NALU 2 Size | NALU 2 HDR | 1085 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1086 | NALU 2 Data | 1087 : : 1088 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1089 | :...OPTIONAL RTP padding | 1090 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1092 Figure 7 An example of an RTP packet including an STAP-A containing 1093 two single-time aggregation units 1095 Figure 8 presents an example of an RTP packet that contains an STAP-B. 1096 The STAP contains two single-time aggregation units, labeled as 1 and 1097 2 in the figure. 1099 0 1 2 3 1100 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1101 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1102 | RTP Header | 1103 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1104 |STAP-B NAL HDR | DON | NALU 1 Size | 1105 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1106 | NALU 1 Size | NALU 1 HDR | NALU 1 Data | 1107 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + 1108 : : 1109 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1110 | | NALU 2 Size | NALU 2 HDR | 1111 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1112 | NALU 2 Data | 1113 : : 1114 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1115 | :...OPTIONAL RTP padding | 1116 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1118 Figure 8 An example of an RTP packet including an STAP-B containing 1119 two single-time aggregation units 1121 5.7.2. Multi-Time Aggregation Packets (MTAPs) 1123 The NAL unit payload of MTAPs consists of a 16-bit unsigned decoding 1124 order number base (DONB) (in network byte order) and one or more 1125 multi-time aggregation units, as presented in Figure 9. DONB MUST 1126 contain the value of DON for the first NAL unit in the NAL unit 1127 decoding order among the NAL units of the MTAP. 1129 Informative note: The first NAL unit in the NAL unit decoding 1130 order is not necessarily the first NAL unit in the order in which 1131 the NAL units are encapsulated in an MTAP. 1133 0 1 2 3 1134 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1135 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1136 : decoding order number base | | 1137 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1138 | | 1139 | multi-time aggregation units | 1140 | | 1141 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1142 | : 1143 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1145 Figure 9 NAL unit payload format for MTAPs 1147 Two different multi-time aggregation units are defined in this 1148 specification. Both of them consist of 16 bits unsigned size 1149 information of the following NAL unit (in network byte order), an 8- 1150 bit unsigned decoding order number difference (DOND), and n bits (in 1151 network byte order) of timestamp offset (TS offset) for this NAL unit, 1152 whereby n can be 16 or 24. The choice between the different MTAP 1153 types (MTAP16 and MTAP24) is application dependent: the larger the 1154 timestamp offset is, the higher the flexibility of the MTAP, but the 1155 overhead is also higher. 1157 The structure of the multi-time aggregation units for MTAP16 and 1158 MTAP24 are presented in Figures 10 and 11, respectively. The 1159 starting or ending position of an aggregation unit within a packet is 1160 NOT REQUIRED to be on a 32-bit word boundary. The DON of the NAL 1161 unit contained in a multi-time aggregation unit is equal to (DONB + 1162 DOND) % 65536, in which % denotes the modulo operation. This memo 1163 does not specify how the NAL units within an MTAP are ordered, but, 1164 in most cases, NAL unit decoding order SHOULD be used. 1166 The timestamp offset field MUST be set to a value equal to the value 1167 of the following formula: If the NALU-time is larger than or equal to 1168 the RTP timestamp of the packet, then the timestamp offset equals 1169 (the NALU-time of the NAL unit - the RTP timestamp of the packet). 1170 If the NALU-time is smaller than the RTP timestamp of the packet, 1171 then the timestamp offset is equal to the NALU-time + (2^32 - the RTP 1172 timestamp of the packet). 1174 0 1 2 3 1175 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1176 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1177 : NAL unit size | DOND | TS offset | 1178 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1179 | TS offset | | 1180 +-+-+-+-+-+-+-+-+ NAL unit | 1181 | | 1182 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1183 | : 1184 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1186 Figure 10 Multi-time aggregation unit for MTAP16 1188 0 1 2 3 1189 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1190 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1191 : NAL unit size | DOND | TS offset | 1192 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1193 | TS offset | | 1194 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1195 | NAL unit | 1196 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1197 | : 1198 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1200 Figure 11 Multi-time aggregation unit for MTAP24 1202 For the "earliest" multi-time aggregation unit in an MTAP the 1203 timestamp offset MUST be zero. Hence, the RTP timestamp of the MTAP 1204 itself is identical to the earliest NALU-time. 1206 Informative note: The "earliest" multi-time aggregation unit is 1207 the one that would have the smallest extended RTP timestamp among 1208 all the aggregation units of an MTAP if the NAL units contained 1209 in the aggregation units were encapsulated in single NAL unit 1210 packets. An extended timestamp is a timestamp that has more than 1211 32 bits and is capable of counting the wraparound of the 1212 timestamp field, thus enabling one to determine the smallest 1213 value if the timestamp wraps. Such an "earliest" aggregation 1214 unit may not be the first one in the order in which the 1215 aggregation units are encapsulated in an MTAP. The "earliest" 1216 NAL unit need not be the same as the first NAL unit in the NAL 1217 unit decoding order either. 1219 Figure 12 presents an example of an RTP packet that contains a multi- 1220 time aggregation packet of type MTAP16 that contains two multi-time 1221 aggregation units, labeled as 1 and 2 in the figure. 1223 0 1 2 3 1224 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1225 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1226 | RTP Header | 1227 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1228 |MTAP16 NAL HDR | decoding order number base | NALU 1 Size | 1229 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1230 | NALU 1 Size | NALU 1 DOND | NALU 1 TS offset | 1231 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1232 | NALU 1 HDR | NALU 1 DATA | 1233 +-+-+-+-+-+-+-+-+ + 1234 : : 1235 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1236 | | NALU 2 SIZE | NALU 2 DOND | 1237 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1238 | NALU 2 TS offset | NALU 2 HDR | NALU 2 DATA | 1239 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1240 : : 1241 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1242 | :...OPTIONAL RTP padding | 1243 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1245 Figure 12 An RTP packet including a multi-time aggregation packet of 1246 type MTAP16 containing two multi-time aggregation units 1248 Figure 13 presents an example of an RTP packet that contains a multi- 1249 time aggregation packet of type MTAP24 that contains two multi-time 1250 aggregation units, labeled as 1 and 2 in the figure. 1252 0 1 2 3 1253 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1254 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1255 | RTP Header | 1256 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1257 |MTAP24 NAL HDR | decoding order number base | NALU 1 Size | 1258 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1259 | NALU 1 Size | NALU 1 DOND | NALU 1 TS offs | 1260 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1261 |NALU 1 TS offs | NALU 1 HDR | NALU 1 DATA | 1262 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + 1263 : : 1264 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1265 | | NALU 2 SIZE | NALU 2 DOND | 1266 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1267 | NALU 2 TS offset | NALU 2 HDR | 1268 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1269 | NALU 2 DATA | 1270 : : 1271 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1272 | :...OPTIONAL RTP padding | 1273 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1275 Figure 13 An RTP packet including a multi-time aggregation packet of 1276 type MTAP24 containing two multi-time aggregation units 1278 5.7.3. Fragmentation Units (FUs) 1280 This payload type allows fragmenting a NAL unit into several RTP 1281 packets. Doing so on the application layer instead of relying on 1282 lower layer fragmentation (e.g., by IP) has the following advantages: 1284 o The payload format is capable of transporting NAL units bigger 1285 than 64 kbytes over an IPv4 network that may be present in pre- 1286 recorded video, particularly in High Definition formats (there is 1287 a limit of the number of slices per picture, which results in a 1288 limit of NAL units per picture, which may result in big NAL units). 1290 o The fragmentation mechanism allows fragmenting a single NAL unit 1291 and applying generic forward error correction as described in 1292 section 12.5. 1294 Fragmentation is defined only for a single NAL unit and not for any 1295 aggregation packets. A fragment of a NAL unit consists of an integer 1296 number of consecutive octets of that NAL unit. Each octet of the NAL 1297 unit MUST be part of exactly one fragment of that NAL unit. 1298 Fragments of the same NAL unit MUST be sent in consecutive order with 1299 ascending RTP sequence numbers (with no other RTP packets within the 1300 same RTP packet stream being sent between the first and last 1301 fragment). Similarly, a NAL unit MUST be reassembled in RTP sequence 1302 number order. 1304 When a NAL unit is fragmented and conveyed within fragmentation units 1305 (FUs), it is referred to as a fragmented NAL unit. STAPs and MTAPs 1306 MUST NOT be fragmented. FUs MUST NOT be nested; i.e., an FU MUST NOT 1307 contain another FU. 1309 The RTP timestamp of an RTP packet carrying an FU is set to the NALU- 1310 time of the fragmented NAL unit. 1312 Figure 14 presents the RTP payload format for FU-As. An FU-A 1313 consists of a fragmentation unit indicator of one octet, a 1314 fragmentation unit header of one octet, and a fragmentation unit 1315 payload. 1317 0 1 2 3 1318 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1319 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1320 | FU indicator | FU header | | 1321 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1322 | | 1323 | FU payload | 1324 | | 1325 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1326 | :...OPTIONAL RTP padding | 1327 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1329 Figure 14 RTP payload format for FU-A 1331 Figure 15 presents the RTP payload format for FU-Bs. An FU-B 1332 consists of a fragmentation unit indicator of one octet, a 1333 fragmentation unit header of one octet, a decoding order number (DON) 1334 (in network byte order), and a fragmentation unit payload. In other 1335 words, the structure of FU-B is the same as the structure of FU-A, 1336 except for the additional DON field. 1338 0 1 2 3 1339 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 1340 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1341 | FU indicator | FU header | DON | 1342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-| 1343 | | 1344 | FU payload | 1345 | | 1346 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1347 | :...OPTIONAL RTP padding | 1348 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1350 Figure 15 RTP payload format for FU-B 1352 NAL unit type FU-B MUST be used in the interleaved packetization mode 1353 for the first fragmentation unit of a fragmented NAL unit. NAL unit 1354 type FU-B MUST NOT be used in any other case. In other words, in the 1355 interleaved packetization mode, each NALU that is fragmented has an 1356 FU-B as the first fragment, followed by one or more FU-A fragments. 1358 The FU indicator octet has the following format: 1360 +---------------+ 1361 |0|1|2|3|4|5|6|7| 1362 +-+-+-+-+-+-+-+-+ 1363 |F|NRI| Type | 1364 +---------------+ 1366 Values equal to 28 and 29 in the Type field of the FU indicator octet 1367 identify an FU-A and an FU-B, respectively. The use of the F bit is 1368 described in section 5.3. The value of the NRI field MUST be set 1369 according to the value of the NRI field in the fragmented NAL unit. 1371 The FU header has the following format: 1373 +---------------+ 1374 |0|1|2|3|4|5|6|7| 1375 +-+-+-+-+-+-+-+-+ 1376 |S|E|R| Type | 1377 +---------------+ 1379 S: 1 bit 1380 When set to one, the Start bit indicates the start of a 1381 fragmented NAL unit. When the following FU payload is not the 1382 start of a fragmented NAL unit payload, the Start bit is set to 1383 zero. 1385 E: 1 bit 1386 When set to one, the End bit indicates the end of a fragmented 1387 NAL unit, i.e., the last byte of the payload is also the last 1388 byte of the fragmented NAL unit. When the following FU payload 1389 is not the last fragment of a fragmented NAL unit, the End bit is 1390 set to zero. 1392 R: 1 bit 1393 The Reserved bit MUST be equal to 0 and MUST be ignored by the 1394 receiver. 1396 Type: 5 bits 1397 The NAL unit payload type as defined in Table 7-1 of [1]. 1399 The value of DON in FU-Bs is selected as described in section 5.5. 1401 Informative note: The DON field in FU-Bs allows gateways to 1402 fragment NAL units to FU-Bs without organizing the incoming NAL 1403 units to the NAL unit decoding order. 1405 A fragmented NAL unit MUST NOT be transmitted in one FU; i.e., the 1406 Start bit and End bit MUST NOT both be set to one in the same FU 1407 header. 1409 The FU payload consists of fragments of the payload of the fragmented 1410 NAL unit so that if the fragmentation unit payloads of consecutive 1411 FUs are sequentially concatenated, the payload of the fragmented NAL 1412 unit can be reconstructed. The NAL unit type octet of the fragmented 1413 NAL unit is not included as such in the fragmentation unit payload, 1414 but rather the information of the NAL unit type octet of the 1415 fragmented NAL unit is conveyed in F and NRI fields of the FU 1416 indicator octet of the fragmentation unit and in the type field of 1417 the FU header. An FU payload MAY have any number of octets and MAY 1418 be empty. 1420 Informative note: Empty FUs are allowed to reduce the latency of 1421 a certain class of senders in nearly lossless environments. 1422 These senders can be characterized in that they packetize NALU 1423 fragments before the NALU is completely generated and, hence, 1424 before the NALU size is known. If zero-length NALU fragments 1425 were not allowed, the sender would have to generate at least one 1426 bit of data of the following fragment before the current fragment 1427 could be sent. Due to the characteristics of H.264, where 1428 sometimes several macroblocks occupy zero bits, this is 1429 undesirable and can add delay. However, the (potential) use of 1430 zero-length NALU fragments should be carefully weighed against 1431 the increased risk of the loss of at least a part of the NALU 1432 because of the additional packets employed for its transmission. 1434 If a fragmentation unit is lost, the receiver SHOULD discard all 1435 following fragmentation units in transmission order corresponding to 1436 the same fragmented NAL unit. 1438 A receiver in an endpoint or in a MANE MAY aggregate the first n-1 1439 fragments of a NAL unit to an (incomplete) NAL unit, even if fragment 1440 n of that NAL unit is not received. In this case, the 1441 forbidden_zero_bit of the NAL unit MUST be set to one to indicate a 1442 syntax violation. 1444 6. Packetization Rules 1446 The packetization modes are introduced in section 5.2. The 1447 packetization rules common to more than one of the packetization 1448 modes are specified in section 6.1. The packetization rules for the 1449 single NAL unit mode, the non-interleaved mode, and the interleaved 1450 mode are specified in sections 6.2, 6.3, and 6.4, respectively. 1452 6.1. Common Packetization Rules 1454 All senders MUST enforce the following packetization rules regardless 1455 of the packetization mode in use: 1457 o Coded slice NAL units or coded slice data partition NAL units 1458 belonging to the same coded picture (and thus sharing the same RTP 1459 timestamp value) MAY be sent in any order; however, for delay- 1460 critical systems, they SHOULD be sent in their original decoding 1461 order to minimize the delay. Note that the decoding order is the 1462 order of the NAL units in the bitstream. 1464 o Parameter sets are handled in accordance with the rules and 1465 recommendations given in section 8.4. 1467 o MANEs MUST NOT duplicate any NAL unit except for sequence or 1468 picture parameter set NAL units, as neither this memo nor the 1469 H.264 specification provides means to identify duplicated NAL 1470 units. Sequence and picture parameter set NAL units MAY be 1471 duplicated to make their correct reception more probable, but any 1472 such duplication MUST NOT affect the contents of any active 1473 sequence or picture parameter set. Duplication SHOULD be 1474 performed on the application layer and not by duplicating RTP 1475 packets (with identical sequence numbers). 1477 Senders using the non-interleaved mode and the interleaved mode MUST 1478 enforce the following packetization rule: 1480 o MANEs MAY convert single NAL unit packets into one aggregation 1481 packet, convert an aggregation packet into several single NAL unit 1482 packets, or mix both concepts, in an RTP translator. The RTP 1483 translator SHOULD take into account at least the following 1484 parameters: path MTU size, unequal protection mechanisms (e.g., 1485 through packet-based FEC according to RFC 2733 [18], especially 1486 for sequence and picture parameter set NAL units and coded slice 1487 data partition A NAL units), bearable latency of the system, and 1488 buffering capabilities of the receiver. 1490 Informative note: An RTP translator is required to handle RTCP 1491 as per RFC 3550. 1493 6.2. Single NAL Unit Mode 1495 This mode is in use when the value of the OPTIONAL packetization-mode 1496 media type parameter is equal to 0 or the packetization-mode is not 1497 present. All receivers MUST support this mode. It is primarily 1498 intended for low-delay applications that are compatible with systems 1499 using ITU-T Recommendation H.241 [3] (see section 12.1). Only single 1500 NAL unit packets MAY be used in this mode. STAPs, MTAPs, and FUs 1501 MUST NOT be used. The transmission order of single NAL unit packets 1502 MUST comply with the NAL unit decoding order. 1504 6.3. Non-Interleaved Mode 1506 This mode is in use when the value of the OPTIONAL packetization-mode 1507 media type parameter is equal to 1. This mode SHOULD be supported. 1508 It is primarily intended for low-delay applications. Only single NAL 1509 unit packets, STAP-As, and FU-As MAY be used in this mode. STAP-Bs, 1510 MTAPs, and FU-Bs MUST NOT be used. The transmission order of NAL 1511 units MUST comply with the NAL unit decoding order. 1513 6.4. Interleaved Mode 1515 This mode is in use when the value of the OPTIONAL packetization-mode 1516 media type parameter is equal to 2. Some receivers MAY support this 1517 mode. STAP-Bs, MTAPs, FU-As, and FU-Bs MAY be used. STAP-As and 1518 single NAL unit packets MUST NOT be used. The transmission order of 1519 packets and NAL units is constrained as specified in section 5.5. 1521 7. De-Packetization Process 1523 The de-packetization process is implementation dependent. Therefore, 1524 the following description should be seen as an example of a suitable 1525 implementation. Other schemes may be used as well as long as the 1526 output for the same input is the same as the process described below. 1527 The same output means that the resulting NAL units, and their order, 1528 are identical. Optimizations relative to the described algorithms 1529 are likely possible. Section 7.1 presents the de-packetization 1530 process for the single NAL unit and non-interleaved packetization 1531 modes, whereas section 7.2 describes the process for the interleaved 1532 mode. Section 7.3 includes additional de-packetization guidelines 1533 for intelligent receivers. 1535 All normal RTP mechanisms related to buffer management apply. In 1536 particular, duplicated or outdated RTP packets (as indicated by the 1537 RTP sequence number and the RTP timestamp) are removed. To determine 1538 the exact time for decoding, factors such as a possible intentional 1539 delay to allow for proper inter-stream synchronization must be 1540 factored in. 1542 7.1. Single NAL Unit and Non-Interleaved Mode 1544 The receiver includes a receiver buffer to compensate for 1545 transmission delay jitter. The receiver stores incoming packets in 1546 reception order into the receiver buffer. Packets are de-packetized 1547 in RTP sequence number order. If a de-packetized packet is a single 1548 NAL unit packet, the NAL unit contained in the packet is passed 1549 directly to the decoder. If a de-packetized packet is an STAP-A, the 1550 NAL units contained in the packet are passed to the decoder in the 1551 order in which they are encapsulated in the packet. For all the FU-A 1552 packets containing fragments of a single NAL unit, the de-packetized 1553 fragments are concatenated in their sending order to recover the NAL 1554 unit, which is then passed to the decoder. 1556 Informative note: If the decoder supports Arbitrary Slice Order, 1557 coded slices of a picture can be passed to the decoder in any 1558 order regardless of their reception and transmission order. 1560 7.2. Interleaved Mode 1562 The general concept behind these de-packetization rules is to reorder 1563 NAL units from transmission order to the NAL unit decoding order. 1565 The receiver includes a receiver buffer, which is used to compensate 1566 for transmission delay jitter and to reorder NAL units from 1567 transmission order to the NAL unit decoding order. In this section, 1568 the receiver operation is described under the assumption that there 1569 is no transmission delay jitter. To make a difference from a 1570 practical receiver buffer that is also used for compensation of 1571 transmission delay jitter, the receiver buffer is here after called 1572 the de-interleaving buffer in this section. Receivers SHOULD also 1573 prepare for transmission delay jitter; i.e., either reserve separate 1574 buffers for transmission delay jitter buffering and de-interleaving 1575 buffering or use a receiver buffer for both transmission delay jitter 1576 and de-interleaving. Moreover, receivers SHOULD take transmission 1577 delay jitter into account in the buffering operation; e.g., by 1578 additional initial buffering before starting of decoding and playback. 1580 This section is organized as follows: subsection 7.2.1 presents how 1581 to calculate the size of the de-interleaving buffer. Subsection 1582 7.2.2 specifies the receiver process on how to organize received NAL 1583 units to the NAL unit decoding order. 1585 7.2.1. Size of the De-interleaving Buffer 1587 In either Offer/Answer or declarative SDP usage, the sprop-deint-buf- 1588 req media type parameter signals the requirement for the de- 1589 interleaving buffer size. It is therefore RECOMMENDED to set the de- 1590 interleaving buffer size, in terms of number of bytes, equal to or 1591 greater than the value of sprop-deint-buf-req media type parameter. 1593 When the SDP Offer/Answer model or any other capability exchange 1594 procedure is used in session setup, the properties of the received 1595 stream SHOULD be such that the receiver capabilities are not exceeded. 1596 In the SDP Offer/Answer model, the receiver can indicate its 1597 capabilities to allocate a de-interleaving buffer with the deint-buf- 1598 cap media type parameter. See section 8.1 for further information on 1599 deint-buf-cap and sprop-deint-buf-req media type parameters and 1600 section 8.2.2 for further information on their use in the SDP 1601 Offer/Answer model. 1603 7.2.2. De-interleaving Process 1605 There are two buffering states in the receiver: initial buffering and 1606 buffering while playing. Initial buffering occurs when the RTP 1607 session is initialized. After initial buffering, decoding and 1608 playback are started, and the buffering-while-playing mode is used. 1610 Regardless of the buffering state, the receiver stores incoming NAL 1611 units, in reception order, in the de-interleaving buffer as follows. 1612 NAL units of aggregation packets are stored in the de-interleaving 1613 buffer individually. The value of DON is calculated and stored for 1614 each NAL unit. 1616 The receiver operation is described below with the help of the 1617 following functions and constants: 1619 o Function AbsDON is specified in section 8.1. 1621 o Function don_diff is specified in section 5.5. 1623 o Constant N is the value of the OPTIONAL sprop-interleaving-depth 1624 media type parameter (see section 8.1) incremented by 1. 1626 Initial buffering lasts until one of the following conditions is 1627 fulfilled: 1629 o There are N or more VCL NAL units in the de-interleaving buffer. 1631 o If sprop-max-don-diff is present, don_diff(m,n) is greater than 1632 the value of sprop-max-don-diff, in which n corresponds to the NAL 1633 unit having the greatest value of AbsDON among the received NAL 1634 units and m corresponds to the NAL unit having the smallest value 1635 of AbsDON among the received NAL units. 1637 o Initial buffering has lasted for the duration equal to or greater 1638 than the value of the OPTIONAL sprop-init-buf-time media type 1639 parameter. 1641 The NAL units to be removed from the de-interleaving buffer are 1642 determined as follows: 1644 o If the de-interleaving buffer contains at least N VCL NAL units, 1645 NAL units are removed from the de-interleaving buffer and passed 1646 to the decoder in the order specified below until the buffer 1647 contains N-1 VCL NAL units. 1649 o If sprop-max-don-diff is present, all NAL units m for which 1650 don_diff(m,n) is greater than sprop-max-don-diff are removed from 1651 the de-interleaving buffer and passed to the decoder in the order 1652 specified below. Herein, n corresponds to the NAL unit having the 1653 greatest value of AbsDON among the NAL units in the de- 1654 interleaving buffer. 1656 The order in which NAL units are passed to the decoder is specified 1657 as follows: 1659 o Let PDON be a variable that is initialized to 0 at the beginning 1660 of the RTP session. 1662 o For each NAL unit associated with a value of DON, a DON distance 1663 is calculated as follows. If the value of DON of the NAL unit is 1664 larger than the value of PDON, the DON distance is equal to DON - 1665 PDON. Otherwise, the DON distance is equal to 65535 - PDON + DON 1666 + 1. 1668 o NAL units are delivered to the decoder in ascending order of DON 1669 distance. If several NAL units share the same value of DON 1670 distance, they can be passed to the decoder in any order. 1672 o When a desired number of NAL units have been passed to the decoder, 1673 the value of PDON is set to the value of DON for the last NAL unit 1674 passed to the decoder. 1676 7.3. Additional De-Packetization Guidelines 1678 The following additional de-packetization rules may be used to 1679 implement an operational H.264 de-packetizer: 1681 o Intelligent RTP receivers (e.g., in gateways) may identify lost 1682 coded slice data partitions A (DPAs). If a lost DPA is detected, 1683 after taking into account possible retransmission and FEC, a 1684 gateway may decide not to send the corresponding coded slice data 1685 partitions B and C, as their information is meaningless for H.264 1686 decoders. In this way a MANE can reduce network load by 1687 discarding useless packets without parsing a complex bitstream. 1689 o Intelligent RTP receivers (e.g., in gateways) may identify lost 1690 FUs. If a lost FU is found, a gateway may decide not to send the 1691 following FUs of the same fragmented NAL unit, as their 1692 information is meaningless for H.264 decoders. In this way a MANE 1693 can reduce network load by discarding useless packets without 1694 parsing a complex bitstream. 1696 o Intelligent receivers having to discard packets or NALUs should 1697 first discard all packets/NALUs in which the value of the NRI 1698 field of the NAL unit type octet is equal to 0. This will 1699 minimize the impact on user experience and keep the reference 1700 pictures intact. If more packets have to be discarded, then 1701 packets with a numerically lower NRI value should be discarded 1702 before packets with a numerically higher NRI value. However, 1703 discarding any packets with an NRI bigger than 0 very likely leads 1704 to decoder drift and SHOULD be avoided. 1706 8. Payload Format Parameters 1708 This section specifies the parameters that MAY be used to select 1709 optional features of the payload format and certain features of the 1710 bitstream. The parameters are specified here as part of the media 1711 subtype registration for the ITU-T H.264 | ISO/IEC 14496-10 codec. A 1712 mapping of the parameters into the Session Description Protocol (SDP) 1713 [6] is also provided for applications that use SDP. Equivalent 1714 parameters could be defined elsewhere for use with control protocols 1715 that do not use SDP. 1717 Some parameters provide a receiver with the properties of the stream 1718 that will be sent. The names of all these parameters start with 1719 "sprop" for stream properties. Some of these "sprop" parameters are 1720 limited by other payload or codec configuration parameters. For 1721 example, the sprop-parameter-sets parameter is constrained by the 1722 profile-level-id parameter. The media sender selects all "sprop" 1723 parameters rather than the receiver. This uncommon characteristic of 1724 the "sprop" parameters may not be compatible with some signaling 1725 protocol concepts, in which case the use of these parameters SHOULD 1726 be avoided. 1728 8.1. Media Type Registration 1730 The media subtype for the ITU-T H.264 | ISO/IEC 14496-10 codec is 1731 allocated from the IETF tree. 1733 The receiver MUST ignore any unspecified parameter. 1735 Media Type name: video 1737 Media subtype name: H264 1739 Required parameters: none 1741 OPTIONAL parameters: 1743 profile-level-id: 1744 A base16 [7] (hexadecimal) representation of the following 1745 three bytes in the sequence parameter set NAL unit specified 1746 in [1]: 1) profile_idc, 2) a byte herein referred to as 1747 profile-iop, composed of the values of constraint_set0_flag, 1748 constraint_set1_flag,constraint_set2_flag, 1749 constraint_set3_flag, and reserved_zero_4bits in bit- 1750 significance order, starting from the most significant bit, 1751 and 3) level_idc. Note that reserved_zero_4bits is required 1752 to be equal to 0 in [1], but other values for it may be 1753 specified in the future by ITU-T or ISO/IEC. 1755 The profile-level-id parameter indicates the default sub- 1756 profile, i.e. the subset of coding tools that may have been 1757 used to generate the stream or that the receiver supports, and 1758 the default level of the stream or the receiver supports. 1760 The default sub-profile is indicated collectively by the 1761 profile_idc byte and some fields in the profile-iop byte. 1762 Depending on the values of the fields in the profile-iop byte, 1763 the default sub-profile may be the set of coding tools 1764 supported by one profile, or a common subset of coding tools 1765 of multiple profiles, as specified in subsection 7.4.2.1.1 of 1766 [1]. The default level is indicated by the level_idc byte, 1767 and, when profile_idc is equal to 66, 77 or 88 (the Baseline, 1768 Main, or Extended profile) and level_idc is equal to 11, 1769 additionally by bit 4 (constraint_set3_flag) of the profile- 1770 iop byte. When profile_idc is equal to 66, 77 or 88 (the 1771 Baseline, Main, or Extended profile) and level_idc is equal to 1772 11, and bit 4 (constraint_set3_flag) of the profile-iop byte 1773 is equal to 1, the default level is level 1b. 1775 Table 5 lists all profiles defined in Annex A of [1] and, for 1776 each of the profiles, the possible combinations of profile_idc 1777 and profile-iop that represent the same sub-profile. 1779 Table 5. Combinations of profile_idc and profile-iop 1780 representing the same sub-profile corresponding to the full 1781 set of coding tools supported by one profile. In the 1782 following, x may be either 0 or 1, while the profile names 1783 are indicated as follows. CB: Constrained Baseline profile, 1784 B: Baseline profile, M: Main profile, E: Extended profile, 1785 H: High profile, H10: High 10 profile, H42: High 4:2:2 1786 profile, H44: High 4:4:4 Predictive profile, H10I: High 10 1787 Intra profile, H42I: High 4:2:2 Intra profile, H44I: High 1788 4:4:4 Intra profile, and C44I: CAVLC 4:4:4 Intra profile. 1790 Profile profile_idc profile-iop 1791 (hexadecimal) (binary) 1793 CB 42 (B) x1xx0000 1794 same as: 4D (M) 1xxx0000 1795 same as: 58 (E) 11xx0000 1796 same as: 64 (H), 6E (H10), 1xx00000 1797 7A (H42), or F4 (H44) 1798 B 42 (B) x0xx0000 1799 same as: 58 (E) 10xx0000 1800 M 4D (M) 0x0x0000 1801 same as: 64 (H), 6E (H10), 01000000 1802 7A (H42), or F4 (H44) 1803 E 58 00xx0000 1804 H 64 00000000 1805 H10 6E 00000000 1806 H42 7A 00000000 1807 H44 F4 00000000 1808 H10I 64 00010000 1809 H42I 7A 00010000 1810 H44I F4 00010000 1811 C44I 2C 00010000 1813 For example, in the table above, profile_idc equal to 58 1814 (Extended) with profile-iop equal to 11xx0000 indicates the 1815 same sub-profile corresponding to profile_idc equal to 42 1816 (Baseline) with profile-iop equal to x1xx0000. Note that 1817 other combinations of profile_idc and profile-iop (not listed 1818 in Table 5) may represent a sub-profile equivalent to the 1819 common subset of coding tools for more than one profile. Note 1820 also that a decoder conforming to a certain profile may be 1821 able to decode bitstreams conforming to other profiles. For 1822 example, a decoder conforming to the High 4:4:4 profile at 1823 certain level must be able to decode bitstreams confirming to 1824 the Constrained Baseline, Main, High, High 10 or High 4:2:2 1825 profile at the same or a lower level. 1827 If the profile-level-id parameter is used to indicate 1828 properties of a NAL unit stream, it indicates that, to decode 1829 the stream, the minimum subset of coding tools a decoder has 1830 to support is the default sub-profile, and the lowest level 1831 the decoder has to support is the default level. 1833 If the profile-level-id parameter is used for capability 1834 exchange or session setup procedure, it indicates the subset 1835 of coding tools, which is equal to the default sub-profile, 1836 and the highest level, which is equal to the default level, 1837 that the codec supports. All levels lower than the default 1838 level are also supported by the codec. 1840 Informative note: Capability exchange and session setup 1841 procedures should provide means to list the capabilities 1842 for each supported sub-profile separately. For example, 1843 the one-of-N codec selection procedure of the SDP 1844 Offer/Answer model can be used (section 10.2 of [8]). The 1845 one-of-N codec selection procedure may also be used to 1846 provide different combinations of profile_idc and profile- 1847 iop that represent the same sub-profile. When there are 1848 many different combinations of profile_idc and profile-iop 1849 that represent the same sub-profile, using the one-of-N 1850 codec selection procedure may result into a fairly large 1851 SDP message. Therefore, a receiver should understand the 1852 different equivalent combinations of profile_idc and 1853 profile-iop that represent the same sub-profile, and be 1854 ready to accept an offer using any of the equivalent 1855 combinations. 1857 If no profile-level-id is present, the Baseline Profile 1858 without additional constraints at Level 1 MUST be inferred. 1860 max-mbps, max-smbps, max-fs, max-cpb, max-dpb, and max-br: 1861 These parameters MAY be used to signal the capabilities of a 1862 receiver implementation. These parameters MUST NOT be used for 1863 any other purpose. The profile-level-id parameter MUST be 1864 present in the same receiver capability description that 1865 contains any of these parameters. The level conveyed in the 1866 value of the profile-level-id parameter MUST be such that the 1867 receiver is fully capable of supporting. max-mbps, max-smbps, 1868 max-fs, max-cpb, max-dpb, and max-br MAY be used to indicate 1869 capabilities of the receiver that extend the required 1870 capabilities of the signaled level, as specified below. 1872 When more than one parameter from the set (max-mbps, max- 1873 smbps , max-fs, max-cpb, max-dpb, max-br) is present, the 1874 receiver MUST support all signaled capabilities simultaneously. 1875 For example, if both max-mbps and max-br are present, the 1876 signaled level with the extension of both the frame rate and 1877 bit rate is supported. That is, the receiver is able to 1878 decode NAL unit streams in which the macroblock processing 1879 rate is up to max-mbps (inclusive), the bit rate is up to max- 1880 br (inclusive), the coded picture buffer size is derived as 1881 specified in the semantics of the max-br parameter below, and 1882 other properties comply with the level specified in the value 1883 of the profile-level-id parameter. 1885 If a receiver can support all the properties of level A, the 1886 level specified in the value of the profile-level-id MUST be 1887 level A (i.e. MUST NOT be lower than level A). In other words, 1888 a sender or receiver MUST NOT signal values of max-mbps, max- 1889 fs, max-cpb, max-dpb, and max-br that taken together meet the 1890 requirements of a higher level compared to the level specified 1891 in the value of the profile-level-id parameter. 1893 Informative note: When the OPTIONAL media type parameters 1894 are used to signal the properties of a NAL unit stream, 1895 max-mbps, max-smbps, max-fs, max-cpb, max-dpb, and max-br 1896 are not present, and the value of profile-level-id must 1897 always be such that the NAL unit stream complies fully with 1898 the specified profile and level. 1900 max-mbps: The value of max-mbps is an integer indicating the 1901 maximum macroblock processing rate in units of macroblocks per 1902 second. The max-mbps parameter signals that the receiver is 1903 capable of decoding video at a higher rate than is required by 1904 the signaled level conveyed in the value of the profile-level- 1905 id parameter. When max-mbps is signaled, the receiver MUST be 1906 able to decode NAL unit streams that conform to the signaled 1907 level, with the exception that the MaxMBPS value in Table A-1 1908 of [1] for the signaled level is replaced with the value of 1909 max-mbps. The value of max-mbps MUST be greater than or equal 1910 to the value of MaxMBPS for the level given in Table A-1 of 1911 [1]. Senders MAY use this knowledge to send pictures of a 1912 given size at a higher picture rate than is indicated in the 1913 signaled level. 1915 max-smbps: The value of max-smbps is an integer indicating the 1916 maximum static macroblock processing rate in units of static 1917 macroblocks per second, under the hypothetical assumption that 1918 all macroblocks are static macroblocks. When max-smbps is 1919 signalled the MaxMBPS value in Table A-1 of [1] should be 1920 replaced with the result of the following computation: 1922 o If the parameter max-mbps is signalled, set a variable 1923 MaxMacroblocksPerSecond to the value of max-mbps. 1924 Otherwise, set MaxMacroblocksPerSecond equal to the value 1925 of MaxMBPS for the level in Table A-1 [1]. 1927 o Set a variable P_non-static to the proportion of non-static 1928 macroblocks in picture n. 1930 o Set a variable P_static to the proportion of static 1931 macroblocks in picture n. 1933 o The value of MaxMBPS in Table A-1 of [1] should be 1934 considered by the encoder to be equal to: 1936 MaxMacroblocksPerSecond * max-smbps / ( P_non-static * max- 1937 smbps + P_static * MaxMacroblocksPerSecond) 1939 The encoder should recompute this value for each picture. The 1940 value of max-smbps MUST be greater than the value of MaxMBPS 1941 for the level given in Table A-1 of [1]. Senders MAY use this 1942 knowledge to send pictures of a given size at a higher picture 1943 rate than is indicated in the signalled level. 1945 max-fs: The value of max-fs is an integer indicating the maximum 1946 frame size in units of macroblocks. The max-fs parameter 1947 signals that the receiver is capable of decoding larger 1948 picture sizes than are required by the signaled level conveyed 1949 in the value of the profile-level-id parameter. When max-fs 1950 is signaled, the receiver MUST be able to decode NAL unit 1951 streams that conform to the signaled level, with the exception 1952 that the MaxFS value in Table A-1 of [1] for the signaled 1953 level is replaced with the value of max-fs. The value of max- 1954 fs MUST be greater than or equal to the value of MaxFS for the 1955 level given in Table A-1 of [1]. Senders MAY use this 1956 knowledge to send larger pictures at a proportionally lower 1957 frame rate than is indicated in the signaled level. 1959 max-cpb: The value of max-cpb is an integer indicating the 1960 maximum coded picture buffer size in units of 1000 bits for 1961 the VCL HRD parameters (see A.3.1 item i of [1]) and in units 1962 of 1200 bits for the NAL HRD parameters (see A.3.1 item j of 1963 [1]). The max-cpb parameter signals that the receiver has 1964 more memory than the minimum amount of coded picture buffer 1965 memory required by the signaled level conveyed in the value of 1966 the profile-level-id parameter. When max-cpb is signaled, the 1967 receiver MUST be able to decode NAL unit streams that conform 1968 to the signaled level, with the exception that the MaxCPB 1969 value in Table A-1 of [1] for the signaled level is replaced 1970 with the value of max-cpb. The value of max-cpb MUST be 1971 greater than or equal to the value of MaxCPB for the level 1972 given in Table A-1 of [1]. Senders MAY use this knowledge to 1973 construct coded video streams with greater variation of bit 1974 rate than can be achieved with the MaxCPB value in Table A-1 1975 of [1]. 1977 Informative note: The coded picture buffer is used in the 1978 hypothetical reference decoder (Annex C) of H.264. The use 1979 of the hypothetical reference decoder is recommended in 1980 H.264 encoders to verify that the produced bitstream 1981 conforms to the standard and to control the output bitrate. 1982 Thus, the coded picture buffer is conceptually independent 1983 of any other potential buffers in the receiver, including 1984 de-interleaving and de-jitter buffers. The coded picture 1985 buffer need not be implemented in decoders as specified in 1986 Annex C of H.264, but rather standard-compliant decoders 1987 can have any buffering arrangements provided that they can 1988 decode standard-compliant bitstreams. Thus, in practice, 1989 the input buffer for video decoder can be integrated with 1990 de-interleaving and de-jitter buffers of the receiver. 1992 max-dpb: The value of max-dpb is an integer indicating the 1993 maximum decoded picture buffer size in units of 1024 bytes. 1994 The max-dpb parameter signals that the receiver has more 1995 memory than the minimum amount of decoded picture buffer 1996 memory required by the signaled level conveyed in the value of 1997 the profile-level-id parameter. When max-dpb is signaled, the 1998 receiver MUST be able to decode NAL unit streams that conform 1999 to the signaled level, with the exception that the MaxDPB 2000 value in Table A-1 of [1] for the signaled level is replaced 2001 with the value of max-dpb. Consequently, a receiver that 2002 signals max-dpb MUST be capable of storing the following 2003 number of decoded frames, complementary field pairs, and non- 2004 paired fields in its decoded picture buffer: 2006 Min(1024 * max-dpb / ( PicWidthInMbs * FrameHeightInMbs * 2007 256 * ChromaFormatFactor ), 16) 2009 PicWidthInMbs, FrameHeightInMbs, and ChromaFormatFactor are 2010 defined in [1]. 2012 The value of max-dpb MUST be greater than or equal to the 2013 value of MaxDPB for the level given in Table A-1 of [1]. 2014 Senders MAY use this knowledge to construct coded video 2015 streams with improved compression. 2017 Informative note: This parameter was added primarily to 2018 complement a similar codepoint in the ITU-T Recommendation 2019 H.245, so as to facilitate signaling gateway designs. The 2020 decoded picture buffer stores reconstructed samples. There 2021 is no relationship between the size of the decoded picture 2022 buffer and the buffers used in RTP, especially de- 2023 interleaving and de-jitter buffers. 2025 max-br: The value of max-br is an integer indicating the maximum 2026 video bit rate in units of 1000 bits per second for the VCL 2027 HRD parameters (see A.3.1 item i of [1]) and in units of 1200 2028 bits per second for the NAL HRD parameters (see A.3.1 item j 2029 of [1]). 2031 The max-br parameter signals that the video decoder of the 2032 receiver is capable of decoding video at a higher bit rate 2033 than is required by the signaled level conveyed in the value 2034 of the profile-level-id parameter. 2036 When max-br is signaled, the video codec of the receiver MUST 2037 be able to decode NAL unit streams that conform to the 2038 signaled level, conveyed in the profile-level-id parameter, 2039 with the following exceptions in the limits specified by the 2040 level: 2042 o The value of max-br replaces the MaxBR value of the signaled 2043 level (in Table A-1 of [1]). 2045 o When the max-cpb parameter is not present, the result of the 2046 following formula replaces the value of MaxCPB in Table A-1 2047 of [1]: (MaxCPB of the signaled level) * max-br / (MaxBR of 2048 the signaled level). 2050 For example, if a receiver signals capability for Level 1.2 2051 with max-br equal to 1550, this indicates a maximum video 2052 bitrate of 1550 kbits/sec for VCL HRD parameters, a maximum 2053 video bitrate of 1860 kbits/sec for NAL HRD parameters, and a 2054 CPB size of 4036458 bits (1550000 / 384000 * 1000 * 1000). 2056 The value of max-br MUST be greater than or equal to the value 2057 MaxBR for the signaled level given in Table A-1 of [1]. 2059 Senders MAY use this knowledge to send higher bitrate video as 2060 allowed in the level definition of Annex A of H.264, to 2061 achieve improved video quality. 2063 Informative note: This parameter was added primarily to 2064 complement a similar codepoint in the ITU-T Recommendation 2065 H.245, so as to facilitate signaling gateway designs. No 2066 assumption can be made from the value of this parameter 2067 that the network is capable of handling such bit rates at 2068 any given time. In particular, no conclusion can be drawn 2069 that the signaled bit rate is possible under congestion 2070 control constraints. 2072 redundant-pic-cap: 2073 This parameter signals the capabilities of a receiver 2074 implementation. When equal to 0, the parameter indicates that 2075 the receiver makes no attempt to use redundant coded pictures 2076 to correct incorrectly decoded primary coded pictures. When 2077 equal to 0, the receiver is not capable of using redundant 2078 slices; therefore, a sender SHOULD avoid sending redundant 2079 slices to save bandwidth. When equal to 1, the receiver is 2080 capable of decoding any such redundant slice that covers a 2081 corrupted area in a primary decoded picture (at least partly), 2082 and therefore a sender MAY send redundant slices. When the 2083 parameter is not present, then a value of 0 MUST be used for 2084 redundant-pic-cap. When present, the value of redundant-pic- 2085 cap MUST be either 0 or 1. 2087 When the profile-level-id parameter is present in the same 2088 signaling as the redundant-pic-cap parameter, and the profile 2089 indicated in profile-level-id is such that it disallows the 2090 use of redundant coded pictures (e.g., Main Profile), the 2091 value of redundant-pic-cap MUST be equal to 0. When a 2092 receiver indicates redundant-pic-cap equal to 0, the received 2093 stream SHOULD NOT contain redundant coded pictures. 2095 Informative note: Even if redundant-pic-cap is equal to 0, 2096 the decoder is able to ignore redundant codec pictures 2097 provided that the decoder supports such a profile (Baseline, 2098 Extended) in which redundant coded pictures are allowed. 2100 Informative note: Even if redundant-pic-cap is equal to 1, 2101 the receiver may also choose other error concealment 2102 strategies to replace or complement decoding of redundant 2103 slices. 2105 sprop-parameter-sets: 2106 This parameter MAY be used to convey any sequence and picture 2107 parameter set NAL units (herein referred to as the initial 2108 parameter set NAL units) that can be placed in the NAL unit 2109 stream to precede any other NAL units in decoding order. The 2110 parameter MUST NOT be used to indicate codec capability in any 2111 capability exchange procedure. The value of the parameter is 2112 a comma (',') separated list of base64 [7] representations of 2113 parameter set NAL units as specified in sections 7.3.2.1 and 2114 7.3.2.2 of [1]. Note that the number of bytes in a parameter 2115 set NAL unit is typically less than 10, but a picture 2116 parameter set NAL unit can contain several hundreds of bytes. 2118 Informative note: When several payload types are offered in 2119 the SDP Offer/Answer model, each with its own sprop- 2120 parameter-sets parameter, then the receiver cannot assume 2121 that those parameter sets do not use conflicting storage 2122 locations (i.e., identical values of parameter set 2123 identifiers). Therefore, a receiver should buffer all 2124 sprop-parameter-sets and make them available to the decoder 2125 instance that decodes a certain payload type. 2127 The "sprop-parameter-sets" parameter MUST only contain 2128 parameter sets that are conforming to the profile-level-id, 2129 i.e., the subset of coding tools indicated by any of the 2130 parameter sets MUST be equal to the default sub-profile, and 2131 the level indicated by any of the parameter sets MUST be equal 2132 to the default level. 2134 sprop-level-parameter-sets: 2135 This parameter MAY be used to convey any sequence and picture 2136 parameter set NAL units (herein referred to as the initial 2137 parameter set NAL units) that can be placed in the NAL unit 2138 stream to precede any other NAL units in decoding order and 2139 that are associated with one or more levels lower than the 2140 default level. The parameter MUST NOT be used to indicate 2141 codec capability in any capability exchange procedure. 2143 The sprop-level-parameter-sets parameter contains parameter 2144 sets for one or more levels which are lower than the default 2145 level. All parameter sets associated with one level are 2146 clustered and prefixed with a three-byte field which has the 2147 same syntax as profile-level-id. This enables the receiver to 2148 install the parameter sets for one level and discard the rest. 2149 The three-byte field is named PLId, and all parameter sets 2150 associated with one level are named PSL, which has the same 2151 syntax as sprop-parameter-sets. Parameter sets for each level 2152 are represented in the form of PLId:PSL, i.e., PLId followed 2153 by a colon (':') and the base64 [7] representation of the 2154 initial parameter set NAL units for the level. Each pair of 2155 PLId:PSL is also separated by a colon. Note that a PSL can 2156 contain multiple parameter sets for that level, separated with 2157 commas (','). 2159 The subset of coding tools indicated by each PLId field MUST 2160 be equal to the default sub-profile, and the level indicated 2161 by each PLId field MUST be lower than the default level. All 2162 sequence parameter sets contained in each PSL MUST have the 2163 three bytes from profile_idc to level_idc, inclusive, equal to 2164 the preceding PLId. 2166 Informative note: This parameter allows for efficient level 2167 downgrade in SDP Offer/Answer and out-of-band transport of 2168 parameter sets, simultaneously. 2170 use-level-src-parameter-sets: 2171 This parameter MAY be used to indicate a receiver capability. 2172 The value MAY be equal to either 0 or 1. When the parameter 2173 is not present, the value MUST be inferred to be equal to 0. 2174 The value 0 indicates that the receiver does not understand 2175 the sprop-level-parameter-sets parameter, and does not 2176 understand the "fmtp" source attribute as specified in section 2177 6.3 of [9], and will ignore sprop-level-parameter-sets when 2178 present, and will ignore sprop-parameter-sets when conveyed 2179 using the "fmtp" source attribute. The value 1 indicates that 2180 the receiver understands the sprop-level-parameter-sets 2181 parameter, and understands the "fmtp" source attribute as 2182 specified in section 6.3 of [9], and is capable of using 2183 parameter sets contained in the sprop-level-parameter-sets or 2184 contained in the sprop-parameter-sets that is conveyed using 2185 the "fmtp" source attribute. 2187 Informative note: An RFC 3984 receiver does not understand 2188 sprop-level-parameter-sets, use-level-src-parameter-sets, 2189 or the "fmtp" source attribute as specified in section 6.3 2190 of [9]. Therefore, during SDP Offer/Answer, an RFC 3984 2191 receiver as the answerer will simply ignore sprop-level- 2192 parameter-sets, when present in an offer, and sprop- 2193 parameter-sets, when conveyed using the "fmtp" source 2194 attribute as specified in section 6.3 of [9]. Assume that 2195 the offered payload type was accepted at a level lower than 2196 the default level. If the offered payload type included 2197 sprop-level-parameter-sets or included sprop-parameter-sets 2198 conveyed using the "fmtp" source attribute, and the offerer 2199 sees that the answerer has not included use-level-src- 2200 parameter-sets equal to 1 in the answer, the offerer gets 2201 to know that in-band transport of parameter sets is needed. 2203 in-band-parameter-sets: 2204 This parameter MAY be used to indicate a receiver capability. 2205 The value MAY be equal to either 0 or 1. The value 1 2206 indicates that receiver discards out-of-band parameter sets in 2207 sprop-parameter-sets and sprop-level-parameter-sets, therefore 2208 the sender MUST transmit all parameter sets in-band. The 2209 value 0 indicates that the receiver utilizes out-of-band 2210 parameter sets included in sprop-parameter-sets and sprop- 2211 level-parameter-sets. However, in this case, the sender MAY 2212 still choose to send parameter sets in-band. When in-band- 2213 parameter-sets is equal to 1, use-level-src-parameter-sets 2214 MUST NOT be present or MUST be equal to 0. When the parameter 2215 is not present, this receiver capability is not specified, and 2216 therefore the sender MAY send out-of-band parameter sets only, 2217 or it MAY send in-band-parameter-sets only, or it MAY send 2218 both. 2220 packetization-mode: 2221 This parameter signals the properties of an RTP payload type 2222 or the capabilities of a receiver implementation. Only a 2223 single configuration point can be indicated; thus, when 2224 capabilities to support more than one packetization-mode are 2225 declared, multiple configuration points (RTP payload types) 2226 must be used. 2228 When the value of packetization-mode is equal to 0 or 2229 packetization-mode is not present, the single NAL mode, as 2230 defined in section 6.2 of RFC 3984, MUST be used. This mode 2231 is in use in standards using ITU-T Recommendation H.241 [3] 2232 (see section 12.1). When the value of packetization-mode is 2233 equal to 1, the non-interleaved mode, as defined in section 2234 6.3 of RFC 3984, MUST be used. When the value of 2235 packetization-mode is equal to 2, the interleaved mode, as 2236 defined in section 6.4 of RFC 3984, MUST be used. The value 2237 of packetization-mode MUST be an integer in the range of 0 to 2238 2, inclusive. 2240 sprop-interleaving-depth: 2241 This parameter MUST NOT be present when packetization-mode is 2242 not present or the value of packetization-mode is equal to 0 2243 or 1. This parameter MUST be present when the value of 2244 packetization-mode is equal to 2. 2246 This parameter signals the properties of an RTP packet stream. 2247 It specifies the maximum number of VCL NAL units that precede 2248 any VCL NAL unit in the RTP packet stream in transmission 2249 order and follow the VCL NAL unit in decoding order. 2250 Consequently, it is guaranteed that receivers can reconstruct 2251 NAL unit decoding order when the buffer size for NAL unit 2252 decoding order recovery is at least the value of sprop- 2253 interleaving-depth + 1 in terms of VCL NAL units. 2255 The value of sprop-interleaving-depth MUST be an integer in 2256 the range of 0 to 32767, inclusive. 2258 sprop-deint-buf-req: 2259 This parameter MUST NOT be present when packetization-mode is 2260 not present or the value of packetization-mode is equal to 0 2261 or 1. It MUST be present when the value of packetization-mode 2262 is equal to 2. 2264 sprop-deint-buf-req signals the required size of the de- 2265 interleaving buffer for the RTP packet stream. The value of 2266 the parameter MUST be greater than or equal to the maximum 2267 buffer occupancy (in units of bytes) required in such a de- 2268 interleaving buffer that is specified in section 7.2 of RFC 2269 3984. It is guaranteed that receivers can perform the de- 2270 interleaving of interleaved NAL units into NAL unit decoding 2271 order, when the de-interleaving buffer size is at least the 2272 value of sprop-deint-buf-req in terms of bytes. 2274 The value of sprop-deint-buf-req MUST be an integer in the 2275 range of 0 to 4294967295, inclusive. 2277 Informative note: sprop-deint-buf-req indicates the 2278 required size of the de-interleaving buffer only. When 2279 network jitter can occur, an appropriately sized jitter 2280 buffer has to be provisioned for as well. 2282 deint-buf-cap: 2283 This parameter signals the capabilities of a receiver 2284 implementation and indicates the amount of de-interleaving 2285 buffer space in units of bytes that the receiver has available 2286 for reconstructing the NAL unit decoding order. A receiver is 2287 able to handle any stream for which the value of the sprop- 2288 deint-buf-req parameter is smaller than or equal to this 2289 parameter. 2291 If the parameter is not present, then a value of 0 MUST be 2292 used for deint-buf-cap. The value of deint-buf-cap MUST be an 2293 integer in the range of 0 to 4294967295, inclusive. 2295 Informative note: deint-buf-cap indicates the maximum 2296 possible size of the de-interleaving buffer of the receiver 2297 only. When network jitter can occur, an appropriately 2298 sized jitter buffer has to be provisioned for as well. 2300 sprop-init-buf-time: 2301 This parameter MAY be used to signal the properties of an RTP 2302 packet stream. The parameter MUST NOT be present, if the 2303 value of packetization-mode is equal to 0 or 1. 2305 The parameter signals the initial buffering time that a 2306 receiver MUST wait before starting decoding to recover the NAL 2307 unit decoding order from the transmission order. The 2308 parameter is the maximum value of (decoding time of the NAL 2309 unit - transmission time of a NAL unit), assuming reliable and 2310 instantaneous transmission, the same timeline for transmission 2311 and decoding, and that decoding starts when the first packet 2312 arrives. 2314 An example of specifying the value of sprop-init-buf-time 2315 follows. A NAL unit stream is sent in the following 2316 interleaved order, in which the value corresponds to the 2317 decoding time and the transmission order is from left to right: 2319 0 2 1 3 5 4 6 8 7 ... 2321 Assuming a steady transmission rate of NAL units, the 2322 transmission times are: 2324 0 1 2 3 4 5 6 7 8 ... 2326 Subtracting the decoding time from the transmission time 2327 column-wise results in the following series: 2329 0 -1 1 0 -1 1 0 -1 1 ... 2331 Thus, in terms of intervals of NAL unit transmission times, 2332 the value of sprop-init-buf-time in this example is 1. The 2333 parameter is coded as a non-negative base10 integer 2334 representation in clock ticks of a 90-kHz clock. If the 2335 parameter is not present, then no initial buffering time value 2336 is defined. Otherwise the value of sprop-init-buf-time MUST 2337 be an integer in the range of 0 to 4294967295, inclusive. 2339 In addition to the signaled sprop-init-buf-time, receivers 2340 SHOULD take into account the transmission delay jitter 2341 buffering, including buffering for the delay jitter caused by 2342 mixers, translators, gateways, proxies, traffic-shapers, and 2343 other network elements. 2345 sprop-max-don-diff: 2346 This parameter MAY be used to signal the properties of an RTP 2347 packet stream. It MUST NOT be used to signal transmitter or 2348 receiver or codec capabilities. The parameter MUST NOT be 2349 present if the value of packetization-mode is equal to 0 or 1. 2350 sprop-max-don-diff is an integer in the range of 0 to 32767, 2351 inclusive. If sprop-max-don-diff is not present, the value of 2352 the parameter is unspecified. sprop-max-don-diff is 2353 calculated as follows: 2355 sprop-max-don-diff = max{AbsDON(i) - AbsDON(j)}, 2356 for any i and any j>i, 2358 where i and j indicate the index of the NAL unit in the 2359 transmission order and AbsDON denotes a decoding order number 2360 of the NAL unit that does not wrap around to 0 after 65535. 2361 In other words, AbsDON is calculated as follows: Let m and n 2362 be consecutive NAL units in transmission order. For the very 2363 first NAL unit in transmission order (whose index is 0), 2364 AbsDON(0) = DON(0). For other NAL units, AbsDON is calculated 2365 as follows: 2367 If DON(m) == DON(n), AbsDON(n) = AbsDON(m) 2369 If (DON(m) < DON(n) and DON(n) - DON(m) < 32768), 2370 AbsDON(n) = AbsDON(m) + DON(n) - DON(m) 2372 If (DON(m) > DON(n) and DON(m) - DON(n) >= 32768), 2373 AbsDON(n) = AbsDON(m) + 65536 - DON(m) + DON(n) 2375 If (DON(m) < DON(n) and DON(n) - DON(m) >= 32768), 2376 AbsDON(n) = AbsDON(m) - (DON(m) + 65536 - DON(n)) 2378 If (DON(m) > DON(n) and DON(m) - DON(n) < 32768), 2379 AbsDON(n) = AbsDON(m) - (DON(m) - DON(n)) 2381 where DON(i) is the decoding order number of the NAL unit 2382 having index i in the transmission order. The decoding order 2383 number is specified in section 5.5 of RFC 3984. 2385 Informative note: Receivers may use sprop-max-don-diff to 2386 trigger which NAL units in the receiver buffer can be 2387 passed to the decoder. 2389 max-rcmd-nalu-size: 2390 This parameter MAY be used to signal the capabilities of a 2391 receiver. The parameter MUST NOT be used for any other 2392 purposes. The value of the parameter indicates the largest 2393 NALU size in bytes that the receiver can handle efficiently. 2394 The parameter value is a recommendation, not a strict upper 2395 boundary. The sender MAY create larger NALUs but must be 2396 aware that the handling of these may come at a higher cost 2397 than NALUs conforming to the limitation. 2399 The value of max-rcmd-nalu-size MUST be an integer in the 2400 range of 0 to 4294967295, inclusive. If this parameter is not 2401 specified, no known limitation to the NALU size exists. 2402 Senders still have to consider the MTU size available between 2403 the sender and the receiver and SHOULD run MTU discovery for 2404 this purpose. 2406 This parameter is motivated by, for example, an IP to H.223 2407 video telephony gateway, where NALUs smaller than the H.223 2408 transport data unit will be more efficient. A gateway may 2409 terminate IP; thus, MTU discovery will normally not work 2410 beyond the gateway. 2412 Informative note: Setting this parameter to a lower than 2413 necessary value may have a negative impact. 2415 sar-understood: 2416 This parameter MAY be used to indicate a receiver capability 2417 and not anything else. The parameter indicates the maximum 2418 value of aspect_ratio_idc (specified in [1]) smaller than 255 2419 that the receiver understands. Table E-1 of [1] specifies 2420 aspect_ratio_idc equal to 0 as "unspecified", 1 to 16, 2421 inclusive, as specific Sample Aspect Ratios (SARs), 17 to 254, 2422 inclusive, as "reserved", and 255 as the Extended SAR, for 2423 which SAR width and SAR height are explicitly signaled. 2424 Therefore, a receiver with a decoder according to [1] 2425 understands aspect_ratio_idc in the range of 1 to 16, 2426 inclusive and aspect_ratio_idc equal to 255, in the sense that 2427 the receiver knows what exactly the SAR is. For such a 2428 receiver, the value of sar-understood is 16. If in the future 2429 Table E-1 of [1] is extended, e.g., such that the SAR for 2430 aspect_ratio_idc equal to 17 is specified, then for a receiver 2431 with a decoder that understands the extension, the value of 2432 sar-understood is 17. For a receiver with a decoder according 2433 to the 2003 version of [1], the value of sar-understood is 13, 2434 as the minimum reserved aspect_ratio_idc therein is 14. 2436 When sar-understood is not present, the value MUST be inferred 2437 to be equal to 13. 2439 sar-supported: 2440 This parameter MAY be used to indicate a receiver capability 2441 and not anything else. The value of this parameter is an 2442 integer in the range of 1 to sar-understood, inclusive, equal 2443 to 255. The value of sar-supported equal to N smaller than 2444 255 indicates that the reciever supports all the SARs 2445 corresponding to H.264 aspect_ratio_idc values (see Table E-1 2446 of [1]) in the range from 1 to N, inclusive, without geometric 2447 distortion. The value of sar-supported equal to 255 indicates 2448 that the receiver supports all sample aspect ratios which are 2449 expressible using two 16-bit integer values as the numerator 2450 and denominator, i.e., those that are expressible using the 2451 H.264 aspect_ratio_idc value of 255 (Extended_SAR, see Table 2452 E-1 of [1]), without geometric distortion. 2454 H.264 compliant encoders SHOULD NOT send an aspect_ratio_idc 2455 equal to 0, or an aspect_ratio_idc larger than sar-understood 2456 and smaller than 255. H.264 compliant encoders SHOULD send an 2457 aspect_ratio_idc that the receiver is able to display without 2458 geometrical distortion. However, H.264 compliant encoders MAY 2459 choose to send pictures using any SAR. 2461 Note that the actual sample aspect ratio or extended sample 2462 aspect ratio, when present, of the stream is conveyed in the 2463 Video Usability Information (VUI) part of the sequence 2464 parameter set. 2466 Encoding considerations: 2467 This type is only defined for transfer via RTP (RFC 3550). 2469 Security considerations: 2470 See section 9 of RFC xxxx. 2472 Public specification: 2473 Please refer to RFC xxxx and its section 15. 2475 Additional information: 2476 None 2478 File extensions: none 2480 Macintosh file type code: none 2482 Object identifier or OID: none 2483 Person & email address to contact for further information: 2484 Ye-Kui Wang, yekuiwang@huawei.com 2486 Intended usage: COMMON 2488 Author: 2489 Ye-Kui Wang, yekuiwang@huawei.com 2491 Change controller: 2492 IETF Audio/Video Transport working group delegated from the 2493 IESG. 2495 8.2. SDP Parameters 2497 8.2.1. Mapping of Payload Type Parameters to SDP 2499 The media type video/H264 string is mapped to fields in the Session 2500 Description Protocol (SDP) [6] as follows: 2502 o The media name in the "m=" line of SDP MUST be video. 2504 o The encoding name in the "a=rtpmap" line of SDP MUST be H264 (the 2505 media subtype). 2507 o The clock rate in the "a=rtpmap" line MUST be 90000. 2509 o The OPTIONAL parameters "profile-level-id", "max-mbps", "max- 2510 smbps", "max-fs", "max-cpb", "max-dpb", "max-br", "redundant-pic- 2511 cap", "use-level-src-parameter-sets", "in-band-parameter-sets", 2512 "packetization-mode", "sprop-interleaving-depth", "sprop-deint- 2513 buf-req", "deint-buf-cap", "sprop-init-buf-time", "sprop-max-don- 2514 diff", "max-rcmd-nalu-size", "sar-understood", and "sar-supported", 2515 when present, MUST be included in the "a=fmtp" line of SDP. These 2516 parameters are expressed as a media type string, in the form of a 2517 semicolon separated list of parameter=value pairs. 2519 o The OPTIONAL parameters "sprop-parameter-sets" and "sprop-level- 2520 parameter-sets", when present, MUST be included in the "a=fmtp" 2521 line of SDP or conveyed using the "fmtp" source attribute as 2522 specified in section 6.3 of [9]. For a particular media format 2523 (i.e., RTP payload type), a "sprop-parameter-sets" or "sprop- 2524 level-parameter-sets" MUST NOT be both included in the "a=fmtp" 2525 line of SDP and conveyed using the "fmtp" source attribute. When 2526 included in the "a=fmtp" line of SDP, these parameters are 2527 expressed as a media type string, in the form of a semicolon 2528 separated list of parameter=value pairs. When conveyed using the 2529 "fmtp" source attribute, these parameters are only associated with 2530 the given source and payload type as parts of the "fmtp" source 2531 attribute. 2533 Informative note: Conveyance of "sprop-parameter-sets" and 2534 "sprop-level-parameter-sets" using the "fmtp" source attribute 2535 allows for out-of-band transport of parameter sets in 2536 topologies like Topo-Video-switch-MCU [29]. 2538 An example of media representation in SDP is as follows (Baseline 2539 Profile, Level 3.0, some of the constraints of the Main profile may 2540 not be obeyed): 2542 m=video 49170 RTP/AVP 98 2543 a=rtpmap:98 H264/90000 2544 a=fmtp:98 profile-level-id=42A01E; 2545 packetization-mode=1; 2546 sprop-parameter-sets= 2548 8.2.2. Usage with the SDP Offer/Answer Model 2550 When H.264 is offered over RTP using SDP in an Offer/Answer model [8] 2551 for negotiation for unicast usage, the following limitations and 2552 rules apply: 2554 o The parameters identifying a media format configuration for H.264 2555 are "profile-level-id" and "packetization-mode", when present. 2556 These media format configuration parameters (except for the level 2557 part of "profile-level-id") MUST be used symmetrically; i.e., the 2558 answerer MUST either maintain all configuration parameters or 2559 remove the media format (payload type) completely, if one or more 2560 of the parameter values are not supported. Note that the level 2561 part of "profile-level-id" includes level_idc, and, for indication 2562 of level 1b when profile_idc is equal to 66, 77 or 88, bit 4 2563 (constraint_set3_flag) of profile-iop. The level part of 2564 "profile-level-id" is downgradable, i.e. the answerer MUST 2565 maintain the same or a lower level or remove the media format 2566 (payload type) completely. 2568 Informative note: The requirement for symmetric use applies 2569 only for the above media format configuration parameters 2570 excluding the level part of "profile-level-id", and not for 2571 the other stream properties and capability parameters. 2573 Informative note: In H.264 [1], all the levels except for 2574 level 1b are equal to the value of level_idc divided by 10. 2575 Level 1b is a level higher than level 1.0 but lower than level 2576 1.1, and is signaled in an ad-hoc manner, due to that the 2577 level was specified after level 1.0 and level 1.1. For the 2578 Baseline, Main and Extended profiles (with profile_idc equal 2579 to 66, 77 and 88, respectively), level 1b is indicated by 2580 level_idc equal to 11 (i.e. same as level 1.1) and 2581 constraint_set3_flag equal to 1. For other profiles, level 1b 2582 is indicated by level_idc equal to 9 (but note that level 1b 2583 for these profiles are still higher than level 1, which has 2584 level_idc equal to 10, and lower than level 1.1). In SDP 2585 Offer/Answer, an answer to an offer may indicate a level equal 2586 to or lower than the level indicated in the offer. Due to the 2587 ad-hoc indication of level 1b, offerers and answerers must 2588 check the value of bit 4 (constraint_set3_flag) of the middle 2589 octet of the parameter "profile-level-id", when profile_idc is 2590 equal to 66, 77 or 88 and level_idc is equal to 11. 2592 To simplify handling and matching of these configurations, the 2593 same RTP payload type number used in the offer SHOULD also be 2594 used in the answer, as specified in [8]. An answer MUST NOT 2595 contain a payload type number used in the offer unless the 2596 configuration is exactly the same as in the offer or the 2597 configuration in the answer only differs from that in the offer 2598 with a level lower than the default level offered. 2600 Informative note: When an offerer receives an answer, it has 2601 to compare payload types not declared in the offer based on 2602 the media type (i.e., video/H264) and the above media 2603 configuration parameters with any payload types it has already 2604 declared. This will enable it to determine whether the 2605 configuration in question is new or if it is equivalent to 2606 configuration already offered, since a different payload type 2607 number may be used in the answer. 2609 o The parameters "sprop-deint-buf-req", "sprop-interleaving-depth", 2610 "sprop-max-don-diff", and "sprop-init-buf-time" describe the 2611 properties of the RTP packet stream that the offerer or answerer 2612 is sending for the media format configuration. This differs from 2613 the normal usage of the Offer/Answer parameters: normally such 2614 parameters declare the properties of the stream that the offerer 2615 or the answerer is able to receive. When dealing with H.264, the 2616 offerer assumes that the answerer will be able to receive media 2617 encoded using the configuration being offered. 2619 Informative note: The above parameters apply for any stream 2620 sent by the declaring entity with the same configuration; i.e., 2621 they are dependent on their source. Rather than being bound 2622 to the payload type, the values may have to be applied to 2623 another payload type when being sent, as they apply for the 2624 configuration. 2626 o The capability parameters ("max-mbps", "max-smbps", "max-fs", 2627 "max-cpb", "max-dpb", "max-br", ,"redundant-pic-cap", "max-rcmd- 2628 nalu-size", "sar-understood", "sar-supported") MAY be used to 2629 declare further capabilities of the offerer or answerer for 2630 receiving. These parameters can only be present when the 2631 direction attribute is sendrecv or recvonly, and the parameters 2632 describe the limitations of what the offerer or answerer accepts 2633 for receiving streams. 2635 o An offerer has to include the size of the de-interleaving buffer, 2636 "sprop-deint-buf-req", in the offer for an interleaved H.264 2637 stream. To enable the offerer and answerer to inform each other 2638 about their capabilities for de-interleaving buffering in 2639 receiving streams, both parties are RECOMMENDED to include "deint- 2640 buf-cap". For interleaved streams, it is also RECOMMENDED to 2641 consider offering multiple payload types with different buffering 2642 requirements when the capabilities of the receiver are unknown. 2644 o The "sprop-parameter-sets" or "sprop-level-parameter-sets" 2645 parameter, when present (included in the "a=fmtp" line of SDP or 2646 conveyed using the "fmtp" source attribute as specified in section 2647 6.3 of [9]), is used for out-of-band transport of parameter sets. 2648 However, when out-of-band transport of parameter sets is used, 2649 parameter sets MAY still be additionally transported in-band. If 2650 neither "sprop-parameter-sets" nor "sprop-level-parameter-sets" is 2651 present, then only in-band transport of parameter sets is used. 2653 An offer MAY include either or both of "sprop-parameter-sets" and 2654 "sprop-level-parameter-sets". An answer MAY include "sprop- 2655 parameter-sets", and MUST NOT include "sprop-level-parameter- 2656 sets". 2658 If the answer includes "in-band-parameter-sets" equal to 1, then 2659 the sender MUST transmit parameter sets in-band. 2661 Otherwise, the following applies. 2663 o When an offered payload type is accepted without level 2664 downgrade, i.e. the default level is accepted, the following 2665 applies. 2667 o When there is a "sprop-parameter-sets" included in the 2668 "a=fmtp" line of SDP, the answerer MUST be prepared to 2669 use the parameter sets included in "sprop-parameter- 2670 sets" for decoding the incoming NAL unit stream. 2672 o When there is a "sprop-parameter-sets" conveyed using 2673 the "fmtp" source attribute as specified in section 6.3 2674 of [9], and the answerer understands the "fmtp" source 2675 attribute, it MUST be prepared to use the parameter 2676 sets included in "sprop-parameter-sets" for decoding 2677 the incoming NAL unit stream, and it MUST include 2678 either "use-level-src-parameter-sets" equal to 1 or the 2679 "fmtp" source attribute in the answer. 2681 o When there is a "sprop-parameter-sets" conveyed using 2682 the "fmtp" source attribute as specified in section 6.3 2683 of [9], and the answerer does not understand the "fmtp" 2684 source attribute, the sender MUST transmit parameter 2685 sets in-band, and the answerer MUST NOT include "use- 2686 level-src-parameter-sets" equal to 1 or the "fmtp" 2687 source attribute in the answer. 2689 o When "sprop-parameter-sets" is not present, the sender 2690 MUST transmit parameter sets in-band. 2692 o The answerer MUST ignore "sprop-level-parameter-sets", 2693 when present (either included in the "a=fmtp" line of 2694 SDP or conveyed using the "fmtp" source attribute). 2696 o When level downgrade is in use, i.e., a level lower than the 2697 default level offered is accepted, the following applies. 2699 o The answerer MUST ignore "sprop-parameter-sets", when 2700 present (either included in the "a=fmtp" line of SDP or 2701 conveyed using the "fmtp" source attribute). 2703 o When "use-level-src-parameter-sets" equal to 1 and the 2704 "fmtp" source attribute are not present in the answer 2705 for the accepted payload type, the answerer MUST ignore 2706 "sprop-level-parameter-sets", when present, and the 2707 sender MUST transmit parameter sets in-band. 2709 o When "use-level-src-parameter-sets" equal to 1 or the 2710 "fmtp" source attribute is present in the answer for 2711 the accepted payload type, the answerer MUST be 2712 prepared to use the parameter sets that are included in 2713 "sprop-level-parameter-sets" for the accepted level, 2714 when present, for decoding the incoming NAL unit stream, 2715 and ignore all other parameter sets included in "sprop- 2716 level-parameter-sets". 2718 o When no parameter sets for the accepted level are 2719 present in the "sprop-level-parameter-sets", the sender 2720 MUST transmit parameter sets in-band. 2722 The answerer MAY or MAY not include "sprop-parameter-sets", i.e., 2723 the answerer MAY use either out-of-band or in-band transport of 2724 parameter sets for the stream it is sending, regardless of 2725 whether out-of-band parameter sets transport has been used in the 2726 offerer-to-answerer direction. When the offer includes "in-band- 2727 parameter-sets" equal to 1, the answerer MUST not include "sprop- 2728 parameter-sets" and MUST transmit parameter sets in-band. All 2729 parameter sets included in the "sprop-parameter-sets", when 2730 present, for the accepted payload type in an answer MUST be 2731 associated with the accepted level, as indicated by the profile- 2732 level-id in the answer for the accepted payload type. 2734 Parameter sets included in "sprop-parameter-sets" in an answer 2735 are independent of those parameter sets included in the offer, as 2736 they are used for decoding two different video streams, one from 2737 the answerer to the offerer, and the other in the opposite 2738 direction. The offerer MUST be prepared to use the parameter 2739 sets included in the answer's "sprop-parameter-sets", when 2740 present, for decoding the incoming NAL unit stream. 2742 When "sprop-parameter-sets" or "sprop-level-parameter-sets" is 2743 conveyed using the "fmtp" source attribute in as specified in 2744 section 6.3 of [9], the receiver of the parameters MUST store the 2745 parameter sets included in the "sprop-parameter-sets" or "sprop- 2746 level-parameter-sets" for the accepted level and associate them 2747 to the source given as a part of the "fmtp" source attribute. 2748 Parameter sets associated with one source MUST only be used to 2749 decode NAL units conveyed in RTP packets from the same source. 2750 When this mechanism is in use, SSRC collision detection and 2751 resolution MUST be performed as specified in [9]. 2753 Informative note: Conveyance of "sprop-parameter-sets" and 2754 "sprop-level-parameter-sets" using the "fmtp" source attribute 2755 may be used in topologies like Topo-Video-switch-MCU [29] to 2756 enable out-of-band transport of parameter sets. 2758 For streams being delivered over multicast, the following rules apply: 2760 o The media format configuration is identified by the same 2761 parameters as above for unicast (i.e. "profile-level-id" and 2762 "packetization-mode", when present). These media format 2763 configuration parameters (including the level part of "profile- 2764 level-id") MUST be used symmetrically; i.e., the answerer MUST 2765 either maintain all configuration parameters or remove the media 2766 format (payload type) completely. Note that this implies that the 2767 level part of "profile-level-id" for Offer/Answer in multicast is 2768 not downgradable. 2770 To simplify handling and matching of these configurations, the 2771 same RTP payload type number used in the offer SHOULD also be 2772 used in the answer, as specified in [8]. An answer MUST NOT 2773 contain a payload type number used in the offer unless the 2774 configuration is the same as in the offer. 2776 o Parameter sets received MUST be associated with the originating 2777 source, and MUST be only used in decoding the incoming NAL unit 2778 stream from the same source. 2780 o The rules for other parameters are the same as above for unicast. 2782 Table 6 lists the interpretation of all the 20 media type parameters 2783 that MUST be used for the different direction attributes. 2785 Table 6. Interpretation of parameters for different direction 2786 attributes. 2788 sendonly --+ 2789 recvonly --+ | 2790 sendrecv --+ | | 2791 | | | 2792 profile-level-id C C P 2793 packetization-mode C C P 2794 sprop-deint-buf-req P - P 2795 sprop-interleaving-depth P - P 2796 sprop-max-don-diff P - P 2797 sprop-init-buf-time P - P 2798 max-mbps R R - 2799 max-smbps R R - 2800 max-fs R R - 2801 max-cpb R R - 2802 max-dpb R R - 2803 max-br R R - 2804 redundant-pic-cap R R - 2805 deint-buf-cap R R - 2806 max-rcmd-nalu-size R R - 2807 sar-understood R R - 2808 sar-supported R R - 2809 in-band-parameter-sets R R - 2810 use-level-src-parameter-sets R R - 2811 sprop-parameter-sets S - S 2812 sprop-level-parameter-sets S - S 2814 Legend: 2816 C: configuration for sending and receiving streams 2817 P: properties of the stream to be sent 2818 R: receiver capabilities 2819 S: out-of-band parameter sets 2820 -: not usable, when present SHOULD be ignored 2822 Parameters used for declaring receiver capabilities are in general 2823 downgradable; i.e., they express the upper limit for a sender's 2824 possible behavior. Thus a sender MAY select to set its encoder using 2825 only lower/less or equal values of these parameters. 2827 Parameters declaring a configuration point are not downgradable, with 2828 the exception of the level part of the "profile-level-id" parameter 2829 for unicast usage. This expresses values a receiver expects to be 2830 used and must be used verbatim on the sender side. 2832 When a sender's capabilities are declared, and non-downgradable 2833 parameters are used in this declaration, then these parameters 2834 express a configuration that is acceptable for the sender to receive 2835 streams. In order to achieve high interoperability levels, it is 2836 often advisable to offer multiple alternative configurations; e.g., 2837 for the packetization mode. It is impossible to offer multiple 2838 configurations in a single payload type. Thus, when multiple 2839 configuration offers are made, each offer requires its own RTP 2840 payload type associated with the offer. 2842 A receiver SHOULD understand all media type parameters, even if it 2843 only supports a subset of the payload format's functionality. This 2844 ensures that a receiver is capable of understanding when an offer to 2845 receive media can be downgraded to what is supported by the receiver 2846 of the offer. 2848 An answerer MAY extend the offer with additional media format 2849 configurations. However, to enable their usage, in most cases a 2850 second offer is required from the offerer to provide the stream 2851 property parameters that the media sender will use. This also has 2852 the effect that the offerer has to be able to receive this media 2853 format configuration, not only to send it. 2855 If an offerer wishes to have non-symmetric capabilities between 2856 sending and receiving, the offerer should offer different RTP 2857 sessions; i.e., different media lines declared as "recvonly" and 2858 "sendonly", respectively. This may have further implications on the 2859 system. 2861 8.2.3. Usage in Declarative Session Descriptions 2863 When H.264 over RTP is offered with SDP in a declarative style, as in 2864 RTSP [27] or SAP [28], the following considerations are necessary. 2866 o All parameters capable of indicating both stream properties and 2867 receiver capabilities are used to indicate only stream properties. 2868 For example, in this case, the parameter "profile-level-id" 2869 declares only the values used by the stream, not the capabilities 2870 for receiving streams. This results in that the following 2871 interpretation of the parameters MUST be used: 2873 Declaring actual configuration or stream properties: 2875 - profile-level-id 2876 - packetization-mode 2877 - sprop-interleaving-depth 2878 - sprop-deint-buf-req 2879 - sprop-max-don-diff 2880 - sprop-init-buf-time 2882 Out-of-band transporting of parameter sets: 2884 - sprop-parameter-sets 2885 - sprop-level-parameter-sets 2887 Not usable(when present, they SHOULD be ignored): 2889 - max-mbps 2890 - max-smbps 2891 - max-fs 2892 - max-cpb 2893 - max-dpb 2894 - max-br 2895 - redundant-pic-cap 2896 - max-rcmd-nalu-size 2897 - deint-buf-cap 2898 - sar-understood 2899 - sar-supported 2900 - in-band-parameter-sets 2901 - use-level-src-parameter-sets 2903 o A receiver of the SDP is required to support all parameters and 2904 values of the parameters provided; otherwise, the receiver MUST 2905 reject (RTSP) or not participate in (SAP) the session. It falls 2906 on the creator of the session to use values that are expected to 2907 be supported by the receiving application. 2909 8.3. Examples 2911 An SDP Offer/Answer exchange wherein both parties are expected to 2912 both send and receive could look like the following. Only the media 2913 codec specific parts of the SDP are shown. Some lines are wrapped 2914 due to text constraints. 2916 Offerer -> Answerer SDP message: 2918 m=video 49170 RTP/AVP 100 99 98 2919 a=rtpmap:98 H264/90000 2920 a=fmtp:98 profile-level-id=42A01E; packetization-mode=0; 2921 sprop-parameter-sets= 2922 a=rtpmap:99 H264/90000 2923 a=fmtp:99 profile-level-id=42A01E; packetization-mode=1; 2924 sprop-parameter-sets= 2925 a=rtpmap:100 H264/90000 2926 a=fmtp:100 profile-level-id=42A01E; packetization-mode=2; 2927 sprop-parameter-sets=; 2928 sprop-interleaving-depth=45; sprop-deint-buf-req=64000; 2929 sprop-init-buf-time=102478; deint-buf-cap=128000 2931 The above offer presents the same codec configuration in three 2932 different packetization formats. PT 98 represents single NALU mode, 2933 PT 99 represents non-interleaved mode, and PT 100 indicates the 2934 interleaved mode. In the interleaved mode case, the interleaving 2935 parameters that the offerer would use if the answer indicates support 2936 for PT 100 are also included. In all three cases the parameter 2937 "sprop-parameter-sets" conveys the initial parameter sets that are 2938 required by the answerer when receiving a stream from the offerer 2939 when this configuration is accepted. Note that the value for "sprop- 2940 parameter-sets" could be different for each payload type. 2942 Answerer -> Offerer SDP message: 2944 m=video 49170 RTP/AVP 100 99 97 2945 a=rtpmap:97 H264/90000 2946 a=fmtp:97 profile-level-id=42A01E; packetization-mode=0; 2947 sprop-parameter-sets= 2948 a=rtpmap:99 H264/90000 2949 a=fmtp:99 profile-level-id=42A01E; packetization-mode=1; 2950 sprop-parameter-sets=; 2951 max-rcmd-nalu-size=3980 2952 a=rtpmap:100 H264/90000 2953 a=fmtp:100 profile-level-id=42A01E; packetization-mode=2; 2954 sprop-parameter-sets=; 2955 sprop-interleaving-depth=60; 2956 sprop-deint-buf-req=86000; sprop-init-buf-time=156320; 2957 deint-buf-cap=128000; max-rcmd-nalu-size=3980 2959 As the Offer/Answer negotiation covers both sending and receiving 2960 streams, an offer indicates the exact parameters for what the offerer 2961 is willing to receive, whereas the answer indicates the same for what 2962 the answerer accepts to receive. In this case the offerer declared 2963 that it is willing to receive payload type 98. The answerer accepts 2964 this by declaring an equivalent payload type 97; i.e., it has 2965 identical values for the two parameters "profile-level-id" and 2966 "packetization-mode" (since "packetization-mode" is equal to 0, 2967 "sprop-deint-buf-req" is not present). As the offered payload type 2968 98 is accepted, the answerer needs to store parameter sets included 2969 in sprop-parameter-sets= in case the offer 2970 finally decides to use this configuration. In the answer, the 2971 answerer includes the parameter sets in sprop-parameter- 2972 sets= that the answerer would use in the 2973 stream sent from the answerer if this configuration is finally used. 2975 The answerer also accepts the reception of the two configurations 2976 that payload types 99 and 100 represent. Again, the answerer needs 2977 to store parameter sets included in sprop-parameter-sets= and sprop-parameter-sets= in case 2979 the offer finally decides to use either of these two configurations. 2980 The answerer provides the initial parameter sets for the answerer-to- 2981 offerer direction, i.e. the parameter sets in sprop-parameter- 2982 sets= and sprop-parameter-sets=, for payload types 99 and 100, respectively, that it will use 2984 to send the payload types. The answerer also provides the offerer 2985 with its memory limit for de-interleaving operations by providing a 2986 "deint-buf-cap" parameter. This is only useful if the offerer 2987 decides on making a second offer, where it can take the new value 2988 into account. The "max-rcmd-nalu-size" indicates that the answerer 2989 can efficiently process NALUs up to the size of 3980 bytes. However, 2990 there is no guarantee that the network supports this size. 2992 In the following example, the offer is accepted without level 2993 downgrading (i.e. the default level, 3.0, is accepted), and both 2994 "sprop-parameter-sets" and "sprop-level-parameter-sets" are present 2995 in the offer. The answerer must ignore sprop-level-parameter- 2996 sets= and store parameter sets in sprop- 2997 parameter-sets= for decoding the incoming NAL 2998 unit stream. The offerer must store the parameter sets in sprop- 2999 parameter-sets= in the answer for decoding the 3000 incoming NAL unit stream. Note that in this example, parameter sets 3001 in sprop-parameter-sets= must be associated 3002 with level 3.0. 3004 Offer SDP: 3006 m=video 49170 RTP/AVP 98 3007 a=rtpmap:98 H264/90000 3008 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3009 packetization-mode=1; 3010 sprop-parameter-sets=; 3011 sprop-level-parameter-sets= 3013 Answer SDP: 3015 m=video 49170 RTP/AVP 98 3016 a=rtpmap:98 H264/90000 3017 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3018 packetization-mode=1; 3019 sprop-parameter-sets= 3021 In the following example, the offer (Baseline profile, level 1.1) is 3022 accepted with level downgrading (the accepted level is 1b), and both 3023 "sprop-parameter-sets" and "sprop-level-parameter-sets" are present 3024 in the offer. The answerer must ignore sprop-parameter- 3025 sets= and all parameter sets not for the 3026 accepted level (level 1b) in sprop-level-parameter-sets=, and must store parameter sets for the accepted level 3028 (level 1b) in sprop-level-parameter-sets= for 3029 decoding the incoming NAL unit stream. The offerer must store the 3030 parameter sets in sprop-parameter-sets= in the 3031 answer for decoding the incoming NAL unit stream. Note that in this 3032 example, parameter sets in sprop-parameter-sets= must be associated with level 1b. 3035 Offer SDP: 3037 m=video 49170 RTP/AVP 98 3038 a=rtpmap:98 H264/90000 3039 a=fmtp:98 profile-level-id=42A00B; //Baseline profile, Level 1.1 3040 packetization-mode=1; 3041 sprop-parameter-sets=; 3042 sprop-level-parameter-sets= 3044 Answer SDP: 3046 m=video 49170 RTP/AVP 98 3047 a=rtpmap:98 H264/90000 3048 a=fmtp:98 profile-level-id=42B00B; //Baseline profile, Level 1b 3049 packetization-mode=1; 3050 sprop-parameter-sets=; 3051 use-level-src-parameter-sets=1 3053 In the following example, the offer (Baseline profile, level 1.1) is 3054 accepted with level downgrading (the accepted level is 1b), and both 3055 "sprop-parameter-sets" and "sprop-level-parameter-sets" are present 3056 in the offer. However, the answerer is a legacy RFC 3984 3057 implementation and does not understand "sprop-level-parameter-sets", 3058 hence it does not include "use-level-src-parameter-sets" (which the 3059 answerer does not understand, either) in the answer. Therefore, the 3060 answerer must ignore both sprop-parameter-sets= and sprop-level-parameter-sets=, and 3062 the offerer must transport parameter sets in-band. 3064 Offer SDP: 3066 m=video 49170 RTP/AVP 98 3067 a=rtpmap:98 H264/90000 3068 a=fmtp:98 profile-level-id=42A00B; //Baseline profile, Level 1.1 3069 packetization-mode=1; 3070 sprop-parameter-sets=; 3071 sprop-level-parameter-sets= 3073 Answer SDP: 3075 m=video 49170 RTP/AVP 98 3076 a=rtpmap:98 H264/90000 3077 a=fmtp:98 profile-level-id=42B00B; //Baseline profile, Level 1b 3078 packetization-mode=1 3080 In the following example, the offer is accepted without level 3081 downgrading, and "sprop-parameter-sets" is present in the offer. 3082 Parameter sets in sprop-parameter-sets= must 3083 be stored and used used by the encoder of the offerer and the decoder 3084 of the answerer, and parameter sets in sprop-parameter- 3085 sets=must be used by the encoder of the 3086 answerer and the decoder of the offerer. Note that sprop-parameter- 3087 sets= is basically independent of sprop- 3088 parameter-sets=. 3090 Offer SDP: 3092 m=video 49170 RTP/AVP 98 3093 a=rtpmap:98 H264/90000 3094 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3095 packetization-mode=1; 3096 sprop-parameter-sets= 3098 Answer SDP: 3100 m=video 49170 RTP/AVP 98 3101 a=rtpmap:98 H264/90000 3102 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3103 packetization-mode=1; 3104 sprop-parameter-sets= 3106 In the following example, the offer is accepted without level 3107 downgrading, and neither "sprop-parameter-sets" nor "sprop-level- 3108 parameter-sets" is present in the offer, meaning that there is no 3109 out-of-band transmission of parameter sets, which then have to be 3110 transported in-band. 3112 Offer SDP: 3114 m=video 49170 RTP/AVP 98 3115 a=rtpmap:98 H264/90000 3116 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3117 packetization-mode=1 3119 Answer SDP: 3121 m=video 49170 RTP/AVP 98 3122 a=rtpmap:98 H264/90000 3123 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3124 packetization-mode=1 3126 In the following example, the offer is accepted with level 3127 downgrading and "sprop-parameter-sets" is present in the offer. As 3128 sprop-parameter-sets= contains level_idc 3129 indicating Level 3.0, therefore cannot be used as the answerer wants 3130 Level 2.0 and must be ignored by the answerer, and in-band parameter 3131 sets must be used. 3133 Offer SDP: 3135 m=video 49170 RTP/AVP 98 3136 a=rtpmap:98 H264/90000 3137 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3138 packetization-mode=1; 3139 sprop-parameter-sets= 3141 Answer SDP: 3143 m=video 49170 RTP/AVP 98 3144 a=rtpmap:98 H264/90000 3145 a=fmtp:98 profile-level-id=42A014; //Baseline profile, Level 2.0 3146 packetization-mode=1 3148 In the following example, the offer is also accepted with level 3149 downgrading, and neither "sprop-parameter-sets" nor "sprop-level- 3150 parameter-sets" is present in the offer, meaning that there is no 3151 out-of-band transmission of parameter sets, which then have to be 3152 transported in-band. 3154 Offer SDP: 3156 m=video 49170 RTP/AVP 98 3157 a=rtpmap:98 H264/90000 3158 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3159 packetization-mode=1 3161 Answer SDP: 3163 m=video 49170 RTP/AVP 98 3164 a=rtpmap:98 H264/90000 3165 a=fmtp:98 profile-level-id=42A014; //Baseline profile, Level 2.0 3166 packetization-mode=1 3168 In the following example, the offerer is a Multipoint Control Unit 3169 (MCU) in a Topo-Video-switch-MCU like topology [29], offering 3170 parameter sets received (using out-of-band transport) from three 3171 other participants B, C, and D, and receiving parameter sets from the 3172 participant A, which is the answerer. The participants are 3173 identified by their values of CNAME, which are mapped to different 3174 SSRC values. The same codec configuration is used by all the four 3175 participants. The participant A stores and associates the parameter 3176 sets included in , , 3177 and to participants B, C, and D, respectively, 3178 and uses for decoding NAL units carried in 3179 RTP packets originated from participant B only, uses for decoding NAL units carried in RTP packets originated from 3181 participant C only, and uses for decoding NAL 3182 units carried in RTP packets originated from participant D only. 3184 Offer SDP: 3186 m=video 49170 RTP/AVP 98 3187 a=ssrc:SSRC-B cname:CNAME-B 3188 a=ssrc:SSRC-C cname:CNAME-C 3189 a=ssrc:SSRC-D cname:CNAME-D 3190 a=ssrc:SSRC-B fmtp:98 3191 sprop-parameter-sets= 3192 a=ssrc:SSRC-C fmtp:98 3193 sprop-parameter-sets= 3194 a=ssrc:SSRC-D fmtp:98 3195 sprop-parameter-sets= 3196 a=rtpmap:98 H264/90000 3197 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3198 packetization-mode=1 3200 Answer SDP: 3202 m=video 49170 RTP/AVP 98 3203 a=ssrc:SSRC-A cname:CNAME-A 3204 a=ssrc:SSRC-A fmtp:98 3205 sprop-parameter-sets= 3206 a=rtpmap:98 H264/90000 3207 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3208 packetization-mode=1 3210 8.4. Parameter Set Considerations 3212 The H.264 parameter sets are a fundamental part of the video codec 3213 and vital to its operation; see section 1.2. Due to their 3214 characteristics and their importance for the decoding process, lost 3215 or erroneously transmitted parameter sets can hardly be concealed 3216 locally at the receiver. A reference to a corrupt parameter set has 3217 normally fatal results to the decoding process. Corruption could 3218 occur, for example, due to the erroneous transmission or loss of a 3219 parameter set NAL unit, but also due to the untimely transmission of 3220 a parameter set update. A parameter set update refers to a change of 3221 at least one parameter in a picture parameter set or sequence 3222 parameter set for which the picture parameter set or sequence 3223 parameter set identifier remains unchanged. Therefore, the following 3224 recommendations are provided as a guideline for the implementer of 3225 the RTP sender. 3227 Parameter set NALUs can be transported using three different 3228 principles: 3230 A. Using a session control protocol (out-of-band) prior to the actual 3231 RTP session. 3233 B. Using a session control protocol (out-of-band) during an ongoing 3234 RTP session. 3236 C. Within the RTP packet stream in the payload (in-band) during an 3237 ongoing RTP session. 3239 It is recommended to implement principles A and B within a session 3240 control protocol. SIP and SDP can be used as described in the SDP 3241 Offer/Answer model and in the previous sections of this memo. 3242 Section 8.2.2 includes a detailed discussion on transport of 3243 parameter sets in-band or out-of-band in SDP Offer/Answer using media 3244 type parameters "sprop-parameter-sets", "sprop-level-parameter-sets", 3245 "use-level-src-parameter-sets" and "in-band-parameter-sets". This 3246 section contains guidelines on how principles A and B should be 3247 implemented within session control protocols. It is independent of 3248 the particular protocol used. Principle C is supported by the RTP 3249 payload format defined in this specification. There are topologies 3250 like Topo-Video-switch-MCU [29] for which the use of principle C may 3251 be desirable. 3253 If in-band signaling of parameter sets is used, the picture and 3254 sequence parameter set NALUs SHOULD be transmitted in the RTP payload 3255 using a reliable method of delivering of RTP (see below), as a loss 3256 of a parameter set of either type will likely prevent decoding of a 3257 considerable portion of the corresponding RTP packet stream. 3259 If in-band signaling of parameter sets is used, the sender SHOULD 3260 take the error characteristics into account and use mechanisms to 3261 provide a high probability for delivering the parameter sets 3262 correctly. Mechanisms that increase the probability for a correct 3263 reception include packet repetition, FEC, and retransmission. The 3264 use of an unreliable, out-of-band control protocol has similar 3265 disadvantages as the in-band signaling (possible loss) and, in 3266 addition, may also lead to difficulties in the synchronization (see 3267 below). Therefore, it is NOT RECOMMENDED. 3269 Parameter sets MAY be added or updated during the lifetime of a 3270 session using principles B and C. It is required that parameter sets 3271 are present at the decoder prior to the NAL units that refer to them. 3272 Updating or adding of parameter sets can result in further problems, 3273 and therefore the following recommendations should be considered. 3275 - When parameter sets are added or updated, care SHOULD be taken to 3276 ensure that any parameter set is delivered prior to its usage. 3277 When new parameter sets are added, previously unused parameter set 3278 identifiers are used. It is common that no synchronization is 3279 present between out-of-band signaling and in-band traffic. If 3280 out-of-band signaling is used, it is RECOMMENDED that a sender 3281 does not start sending NALUs requiring the added or updated 3282 parameter sets prior to acknowledgement of delivery from the 3283 signaling protocol. 3285 - When parameter sets are updated, the following synchronization 3286 issue should be taken into account. When overwriting a parameter 3287 set at the receiver, the sender has to ensure that the parameter 3288 set in question is not needed by any NALU present in the network 3289 or receiver buffers. Otherwise, decoding with a wrong parameter 3290 set may occur. To lessen this problem, it is RECOMMENDED either 3291 to overwrite only those parameter sets that have not been used for 3292 a sufficiently long time (to ensure that all related NALUs have 3293 been consumed), or to add a new parameter set instead (which may 3294 have negative consequences for the efficiency of the video coding). 3296 Informative note: In some topologies like Topo-Video-switch- 3297 MCU [29] the origin of the whole set of parameter sets may 3298 come from multiple sources that may use non-unique parameter 3299 sets identifiers. In this case an offer may overwrite an 3300 existing parameter set if no other mechanism that enables 3301 uniqueness of the parameter sets in the out-of-band channel 3302 exists. 3304 - In a multiparty session, one participant MUST associate parameter 3305 sets coming from different sources with the source identification 3306 whenever possible, e.g. by conveying out-of-band transported 3307 parameter sets, as different sources typically use independent 3308 parameter set identifier value spaces. 3310 - Adding or modifying parameter sets by using both principles B and 3311 C in the same RTP session may lead to inconsistencies of the 3312 parameter sets because of the lack of synchronization between the 3313 control and the RTP channel. Therefore, principles B and C MUST 3314 NOT both be used in the same session unless sufficient 3315 synchronization can be provided. 3317 In some scenarios (e.g., when only the subset of this payload format 3318 specification corresponding to H.241 is used) or topologies, it is 3319 not possible to employ out-of-band parameter set transmission. In 3320 this case, parameter sets have to be transmitted in-band. Here, the 3321 synchronization with the non-parameter-set-data in the bitstream is 3322 implicit, but the possibility of a loss has to be taken into account. 3323 The loss probability should be reduced using the mechanisms discussed 3324 above. In case a loss of a parameter set is detected, recovery may 3325 be achieved by using a Decoder Refresh Point procedure, for example, 3326 using RTCP feedback Full Intra Request (FIR) [30]. Two example 3327 Decoder Refresh Point procedures are provided in the informative 3328 Section 8.5. 3330 - When parameter sets are initially provided using principle A and 3331 then later added or updated in-band (principle C), there is a risk 3332 associated with updating the parameter sets delivered out-of-band. 3333 If receivers miss some in-band updates (for example, because of a 3334 loss or a late tune-in), those receivers attempt to decode the 3335 bitstream using out-dated parameters. It is therefore RECOMMENDED 3336 that parameter set IDs be partitioned between the out-of-band and 3337 in-band parameter sets. 3339 8.5. Decoder Refresh Point Procedure using In-Band Transport of 3340 Parameter Sets (Informative) 3342 When a sender with a video encoder according to [1] receives a 3343 request for a decoder refresh point, the encoder shall enter the fast 3344 update mode by using one of the procedures specified in Section 8.5.1 3345 or 8.5.2 below. The procedure in 8.5.1 is the preferred response in 3346 a lossless transmission environment. Both procedures satisfy the 3347 requirement to enter the fast update mode for H.264 video encoding. 3349 8.5.1. IDR Procedure to Respond to a Request for a Decoder Refresh Point 3351 This section gives one possible way to respond to a request for a 3352 decoder refresh point. 3354 The encoder shall, in the order presented here: 3356 1) Immediately prepare to send an IDR picture. 3358 2) Send a sequence parameter set to be used by the IDR picture to be 3359 sent. The encoder may optionally also send other sequence 3360 parameter sets. 3362 3) Send a picture parameter set to be used by the IDR picture to be 3363 sent. The encoder may optionally also send other picture parameter 3364 sets. 3366 4) Send the IDR picture. 3368 5) From this point forward in time, send any other sequence or 3369 picture parameter sets that have not yet been sent in this 3370 procedure, prior to their reference by any NAL unit, regardless of 3371 whether such parameter sets were previously sent prior to 3372 receiving the request for a decoder refresh point. As needed, 3373 such parameter sets may be sent in a batch, one at a time, or in 3374 any combination of these two methods. Parameter sets may be re- 3375 sent at any time for redundancy. Caution should be taken when 3376 parameter set updates are present, as described above in Section 3377 8.4. 3379 8.5.2. Gradual Recovery Procedure to Respond to a Request for a Decoder 3380 Refresh Point 3382 This section gives another possible way to respond to a request for a 3383 decoder refresh point. 3385 The encoder shall, in the order presented here: 3387 1) Send a recovery point SEI message (see Sections D.1.7 and D.2.7 of 3388 [1]). 3390 2) Repeat any sequence and picture parameter sets that were sent 3391 before the recovery point SEI message, prior to their reference by 3392 a NAL unit. 3394 The encoder shall ensure that the decoder has access to all reference 3395 pictures for inter prediction of pictures at or after the recovery 3396 point, which is indicated by the recovery point SEI message, in 3397 output order, assuming that the transmission from now on is error- 3398 free. 3400 The value of the recovery_frame_cnt syntax element in the recovery 3401 point SEI message should be small enough to ensure a fast recovery. 3403 As needed, such parameter sets may be re-sent in a batch, one at a 3404 time, or in any combination of these two methods. Parameter sets may 3405 be re-sent at any time for redundancy. Caution should be taken when 3406 parameter set updates are present, as described above in Section 8.4. 3408 9. Security Considerations 3410 RTP packets using the payload format defined in this specification 3411 are subject to the security considerations discussed in the RTP 3412 specification [5], and in any appropriate RTP profile (for example, 3413 [16]). This implies that confidentiality of the media streams is 3414 achieved by encryption; for example, through the application of SRTP 3415 [26]. Because the data compression used with this payload format is 3416 applied end-to-end, any encryption needs to be performed after 3417 compression. A potential denial-of-service threat exists for data 3418 encodings using compression techniques that have non-uniform 3419 receiver-end computational load. The attacker can inject 3420 pathological datagrams into the stream that are complex to decode and 3421 that cause the receiver to be overloaded. H.264 is particularly 3422 vulnerable to such attacks, as it is extremely simple to generate 3423 datagrams containing NAL units that affect the decoding process of 3424 many future NAL units. Therefore, the usage of data origin 3425 authentication and data integrity protection of at least the RTP 3426 packet is RECOMMENDED; for example, with SRTP [26]. 3428 Note that the appropriate mechanism to ensure confidentiality and 3429 integrity of RTP packets and their payloads is very dependent on the 3430 application and on the transport and signaling protocols employed. 3431 Thus, although SRTP is given as an example above, other possible 3432 choices exist. 3434 Decoders MUST exercise caution with respect to the handling of user 3435 data SEI messages, particularly if they contain active elements, and 3436 MUST restrict their domain of applicability to the presentation 3437 containing the stream. 3439 End-to-End security with either authentication, integrity or 3440 confidentiality protection will prevent a MANE from performing media- 3441 aware operations other than discarding complete packets. And in the 3442 case of confidentiality protection it will even be prevented from 3443 performing discarding of packets in a media aware way. To allow any 3444 MANE to perform its operations, it will be required to be a trusted 3445 entity which is included in the security context establishment. 3447 10. Congestion Control 3449 Congestion control for RTP SHALL be used in accordance with RFC 3550 3450 [5], and with any applicable RTP profile; e.g., RFC 3551 [16]. An 3451 additional requirement if best-effort service is being used is: users 3452 of this payload format MUST monitor packet loss to ensure that the 3453 packet loss rate is within acceptable parameters. Packet loss is 3454 considered acceptable if a TCP flow across the same network path, and 3455 experiencing the same network conditions, would achieve an average 3456 throughput, measured on a reasonable timescale, that is not less than 3457 the RTP flow is achieving. This condition can be satisfied by 3458 implementing congestion control mechanisms to adapt the transmission 3459 rate (or the number of layers subscribed for a layered multicast 3460 session), or by arranging for a receiver to leave the session if the 3461 loss rate is unacceptably high. 3463 The bit rate adaptation necessary for obeying the congestion control 3464 principle is easily achievable when real-time encoding is used. 3465 However, when pre-encoded content is being transmitted, bandwidth 3466 adaptation requires the availability of more than one coded 3467 representation of the same content, at different bit rates, or the 3468 existence of non-reference pictures or sub-sequences [22] in the 3469 bitstream. The switching between the different representations can 3470 normally be performed in the same RTP session; e.g., by employing a 3471 concept known as SI/SP slices of the Extended Profile, or by 3472 switching streams at IDR picture boundaries. Only when non- 3473 downgradable parameters (such as the profile part of the 3474 profile/level ID) are required to be changed does it become necessary 3475 to terminate and re-start the media stream. This may be accomplished 3476 by using a different RTP payload type. 3478 MANEs MAY follow the suggestions outlined in section 7.3 and remove 3479 certain unusable packets from the packet stream when that stream was 3480 damaged due to previous packet losses. This can help reduce the 3481 network load in certain special cases. 3483 11. IANA Consideration 3485 The H264 media subtype name specified by RFC 3984 should be updated 3486 as defined in section 8.1 of this memo. 3488 12. Informative Appendix: Application Examples 3490 This payload specification is very flexible in its use, in order to 3491 cover the extremely wide application space anticipated for H.264. 3492 However, this great flexibility also makes it difficult for an 3493 implementer to decide on a reasonable packetization scheme. Some 3494 information on how to apply this specification to real-world 3495 scenarios is likely to appear in the form of academic publications 3496 and a test model software and description in the near future. 3497 However, some preliminary usage scenarios are described here as well. 3499 12.1. Video Telephony according to ITU-T Recommendation H.241 Annex A 3501 H.323-based video telephony systems that use H.264 as an optional 3502 video compression scheme are required to support H.241 Annex A [3] as 3503 a packetization scheme. The packetization mechanism defined in this 3504 Annex is technically identical with a small subset of this 3505 specification. 3507 When a system operates according to H.241 Annex A, parameter set NAL 3508 units are sent in-band. Only Single NAL unit packets are used. Many 3509 such systems are not sending IDR pictures regularly, but only when 3510 required by user interaction or by control protocol means; e.g., when 3511 switching between video channels in a Multipoint Control Unit or for 3512 error recovery requested by feedback. 3514 12.2. Video Telephony, No Slice Data Partitioning, No NAL Unit 3515 Aggregation 3517 The RTP part of this scheme is implemented and tested (though not the 3518 control-protocol part; see below). 3520 In most real-world video telephony applications, picture parameters 3521 such as picture size or optional modes never change during the 3522 lifetime of a connection. Therefore, all necessary parameter sets 3523 (usually only one) are sent as a side effect of the capability 3524 exchange/announcement process, e.g., according to the SDP syntax 3525 specified in section 8.2 of this document. As all necessary 3526 parameter set information is established before the RTP session 3527 starts, there is no need for sending any parameter set NAL units. 3528 Slice data partitioning is not used, either. Thus, the RTP packet 3529 stream basically consists of NAL units that carry single coded slices. 3531 The encoder chooses the size of coded slice NAL units so that they 3532 offer the best performance. Often, this is done by adapting the 3533 coded slice size to the MTU size of the IP network. For small 3534 picture sizes, this may result in a one-picture-per-one-packet 3535 strategy. Intra refresh algorithms clean up the loss of packets and 3536 the resulting drift-related artifacts. 3538 12.3. Video Telephony, Interleaved Packetization Using NAL Unit 3539 Aggregation 3541 This scheme allows better error concealment and is used in H.263 3542 based designs using RFC 2429 packetization [11]. It has been 3543 implemented, and good results were reported [13]. 3545 The VCL encoder codes the source picture so that all macroblocks (MBs) 3546 of one MB line are assigned to one slice. All slices with even MB 3547 row addresses are combined into one STAP, and all slices with odd MB 3548 row addresses into another. Those STAPs are transmitted as RTP 3549 packets. The establishment of the parameter sets is performed as 3550 discussed above. 3552 Note that the use of STAPs is essential here, as the high number of 3553 individual slices (18 for a CIF picture) would lead to unacceptably 3554 high IP/UDP/RTP header overhead (unless the source coding tool FMO is 3555 used, which is not assumed in this scenario). Furthermore, some 3556 wireless video transmission systems, such as H.324M and the IP-based 3557 video telephony specified in 3GPP, are likely to use relatively small 3558 transport packet size. For example, a typical MTU size of H.223 AL3 3559 SDU is around 100 bytes [17]. Coding individual slices according to 3560 this packetization scheme provides further advantage in communication 3561 between wired and wireless networks, as individual slices are likely 3562 to be smaller than the preferred maximum packet size of wireless 3563 systems. Consequently, a gateway can convert the STAPs used in a 3564 wired network into several RTP packets with only one NAL unit, which 3565 are preferred in a wireless network, and vice versa. 3567 12.4. Video Telephony with Data Partitioning 3569 This scheme has been implemented and has been shown to offer good 3570 performance, especially at higher packet loss rates [13]. 3572 Data Partitioning is known to be useful only when some form of 3573 unequal error protection is available. Normally, in single-session 3574 RTP environments, even error characteristics are assumed; i.e., the 3575 packet loss probability of all packets of the session is the same 3576 statistically. However, there are means to reduce the packet loss 3577 probability of individual packets in an RTP session. A FEC packet 3578 according to RFC 2733 [18], for example, specifies which media 3579 packets are associated with the FEC packet. 3581 In all cases, the incurred overhead is substantial but is in the same 3582 order of magnitude as the number of bits that have otherwise been 3583 spent for intra information. However, this mechanism does not add 3584 any delay to the system. 3586 Again, the complete parameter set establishment is performed through 3587 control protocol means. 3589 12.5. Video Telephony or Streaming with FUs and Forward Error Correction 3591 This scheme has been implemented and has been shown to provide good 3592 performance, especially at higher packet loss rates [19]. 3594 The most efficient means to combat packet losses for scenarios where 3595 retransmissions are not applicable is forward error correction (FEC). 3596 Although application layer, end-to-end use of FEC is often less 3597 efficient than an FEC-based protection of individual links 3598 (especially when links of different characteristics are in the 3599 transmission path), application layer, end-to-end FEC is unavoidable 3600 in some scenarios. RFC 5109 [18] provides means to use generic, 3601 application layer, end-to-end FEC in packet-loss environments. A 3602 binary forward error correcting code is generated by applying the XOR 3603 operation to the bits at the same bit position in different packets. 3604 The binary code can be specified by the parameters (n,k) in which k 3605 is the number of information packets used in the connection and n is 3606 the total number of packets generated for k information packets; i.e., 3607 n-k parity packets are generated for k information packets. 3609 When a code is used with parameters (n,k) within the RFC 5109 3610 framework, the following properties are well known: 3612 a) If applied over one RTP packet, RFC 5109 provides only packet 3613 repetition. 3615 b) RFC 5109 is most bit rate efficient if XOR-connected packets have 3616 equal length. 3618 c) At the same packet loss probability p and for a fixed k, the 3619 greater the value of n is, the smaller the residual error 3620 probability becomes. For example, for a packet loss probability 3621 of 10%, k=1, and n=2, the residual error probability is about 1%, 3622 whereas for n=3, the residual error probability is about 0.1%. 3624 d) At the same packet loss probability p and for a fixed code rate 3625 k/n, the greater the value of n is, the smaller the residual error 3626 probability becomes. For example, at a packet loss probability of 3627 p=10%, k=1 and n=2, the residual error rate is about 1%, whereas 3628 for an extended Golay code with k=12 and n=24, the residual error 3629 rate is about 0.01%. 3631 For applying RFC 5109 in combination with H.264 baseline coded video 3632 without using FUs, several options might be considered: 3634 1) The video encoder produces NAL units for which each video frame is 3635 coded in a single slice. Applying FEC, one could use a simple 3636 code; e.g., (n=2, k=1). That is, each NAL unit would basically 3637 just be repeated. The disadvantage is obviously the bad code 3638 performance according to d), above, and the low flexibility, as 3639 only (n, k=1) codes can be used. 3641 2) The video encoder produces NAL units for which each video frame is 3642 encoded in one or more consecutive slices. Applying FEC, one 3643 could use a better code, e.g., (n=24, k=12), over a sequence of 3644 NAL units. Depending on the number of RTP packets per frame, a 3645 loss may introduce a significant delay, which is reduced when more 3646 RTP packets are used per frame. Packets of completely different 3647 length might also be connected, which decreases bit rate 3648 efficiency according to b), above. However, with some care and 3649 for slices of 1kb or larger, similar length (100-200 bytes 3650 difference) may be produced, which will not lower the bit 3651 efficiency catastrophically. 3653 3) The video encoder produces NAL units, for which a certain frame 3654 contains k slices of possibly almost equal length. Then, applying 3655 FEC, a better code, e.g., (n=24, k=12), can be used over the 3656 sequence of NAL units for each frame. The delay compared to that 3657 of 2), above, may be reduced, but several disadvantages are 3658 obvious. First, the coding efficiency of the encoded video is 3659 lowered significantly, as slice-structured coding reduces intra- 3660 frame prediction and additional slice overhead is necessary. 3661 Second, pre-encoded content or, when operating over a gateway, the 3662 video is usually not appropriately coded with k slices such that 3663 FEC can be applied. Finally, the encoding of video producing k 3664 slices of equal length is not straightforward and might require 3665 more than one encoding pass. 3667 Many of the mentioned disadvantages can be avoided by applying FUs in 3668 combination with FEC. Each NAL unit can be split into any number of 3669 FUs of basically equal length; therefore, FEC with a reasonable k and 3670 n can be applied, even if the encoder made no effort to produce 3671 slices of equal length. For example, a coded slice NAL unit 3672 containing an entire frame can be split to k FUs, and a parity check 3673 code (n=k+1, k) can be applied. However, this has the disadvantage 3674 that unless all created fragments can be recovered, the whole slice 3675 will be lost. Thus a larger section is lost than would be if the 3676 frame had been split into several slices. 3678 The presented technique makes it possible to achieve good 3679 transmission error tolerance, even if no additional source coding 3680 layer redundancy (such as periodic intra frames) is present. 3681 Consequently, the same coded video sequence can be used to achieve 3682 the maximum compression efficiency and quality over error-free 3683 transmission and for transmission over error-prone networks. 3684 Furthermore, the technique allows the application of FEC to pre- 3685 encoded sequences without adding delay. In this case, pre-encoded 3686 sequences that are not encoded for error-prone networks can still be 3687 transmitted almost reliably without adding extensive delays. In 3688 addition, FUs of equal length result in a bit rate efficient use of 3689 RFC 5109. 3691 If the error probability depends on the length of the transmitted 3692 packet (e.g., in case of mobile transmission [15]), the benefits of 3693 applying FUs with FEC are even more obvious. Basically, the 3694 flexibility of the size of FUs allows appropriate FEC to be applied 3695 for each NAL unit and unequal error protection of NAL units. 3697 When FUs and FEC are used, the incurred overhead is substantial but 3698 is in the same order of magnitude as the number of bits that have to 3699 be spent for intra-coded macroblocks if no FEC is applied. In [19], 3700 it was shown that the overall performance of the FEC-based approach 3701 enhanced quality when using the same error rate and same overall bit 3702 rate, including the overhead. 3704 12.6. Low Bit-Rate Streaming 3706 This scheme has been implemented with H.263 and non-standard RTP 3707 packetization and has given good results [20]. There is no technical 3708 reason why similarly good results could not be achievable with H.264. 3710 In today's Internet streaming, some of the offered bit rates are 3711 relatively low in order to allow terminals with dial-up modems to 3712 access the content. In wired IP networks, relatively large packets, 3713 say 500 - 1500 bytes, are preferred to smaller and more frequently 3714 occurring packets in order to reduce network congestion. Moreover, 3715 use of large packets decreases the amount of RTP/UDP/IP header 3716 overhead. For low bit-rate video, the use of large packets means 3717 that sometimes up to few pictures should be encapsulated in one 3718 packet. 3720 However, loss of a packet including many coded pictures would have 3721 drastic consequences for visual quality, as there is practically no 3722 other way to conceal a loss of an entire picture than to repeat the 3723 previous one. One way to construct relatively large packets and 3724 maintain possibilities for successful loss concealment is to 3725 construct MTAPs that contain interleaved slices from several pictures. 3726 An MTAP should not contain spatially adjacent slices from the same 3727 picture or spatially overlapping slices from any picture. If a 3728 packet is lost, it is likely that a lost slice is surrounded by 3729 spatially adjacent slices of the same picture and spatially 3730 corresponding slices of the temporally previous and succeeding 3731 pictures. Consequently, concealment of the lost slice is likely to 3732 be relatively successful. 3734 12.7. Robust Packet Scheduling in Video Streaming 3736 Robust packet scheduling has been implemented with MPEG-4 Part 2 and 3737 simulated in a wireless streaming environment [21]. There is no 3738 technical reason why similar or better results could not be 3739 achievable with H.264. 3741 Streaming clients typically have a receiver buffer that is capable of 3742 storing a relatively large amount of data. Initially, when a 3743 streaming session is established, a client does not start playing the 3744 stream back immediately. Rather, it typically buffers the incoming 3745 data for a few seconds. This buffering helps maintain continuous 3746 playback, as, in case of occasional increased transmission delays or 3747 network throughput drops, the client can decode and play buffered 3748 data. Otherwise, without initial buffering, the client has to freeze 3749 the display, stop decoding, and wait for incoming data. The 3750 buffering is also necessary for either automatic or selective 3751 retransmission in any protocol level. If any part of a picture is 3752 lost, a retransmission mechanism may be used to resend the lost data. 3753 If the retransmitted data is received before its scheduled decoding 3754 or playback time, the loss is recovered perfectly. Coded pictures 3755 can be ranked according to their importance in the subjective quality 3756 of the decoded sequence. For example, non-reference pictures, such 3757 as conventional B pictures, are subjectively least important, as 3758 their absence does not affect decoding of any other pictures. In 3759 addition to non-reference pictures, the ITU-T H.264 | ISO/IEC 14496- 3760 10 standard includes a temporal scalability method called sub- 3761 sequences [22]. Subjective ranking can also be made on coded slice 3762 data partition or slice group basis. Coded slices and coded slice 3763 data partitions that are subjectively the most important can be sent 3764 earlier than their decoding order indicates, whereas coded slices and 3765 coded slice data partitions that are subjectively the least important 3766 can be sent later than their natural coding order indicates. 3767 Consequently, any retransmitted parts of the most important slices 3768 and coded slice data partitions are more likely to be received before 3769 their scheduled decoding or playback time compared to the least 3770 important slices and slice data partitions. 3772 13. Informative Appendix: Rationale for Decoding Order Number 3774 13.1. Introduction 3776 The Decoding Order Number (DON) concept was introduced mainly to 3777 enable efficient multi-picture slice interleaving (see section 12.6) 3778 and robust packet scheduling (see section 12.7). In both of these 3779 applications, NAL units are transmitted out of decoding order. DON 3780 indicates the decoding order of NAL units and should be used in the 3781 receiver to recover the decoding order. Example use cases for 3782 efficient multi-picture slice interleaving and for robust packet 3783 scheduling are given in sections 13.2 and 13.3, respectively. 3784 Section 13.4 describes the benefits of the DON concept in error 3785 resiliency achieved by redundant coded pictures. Section 13.5 3786 summarizes considered alternatives to DON and justifies why DON was 3787 chosen to this RTP payload specification. 3789 13.2. Example of Multi-Picture Slice Interleaving 3791 An example of multi-picture slice interleaving follows. A subset of 3792 a coded video sequence is depicted below in output order. R denotes 3793 a reference picture, N denotes a non-reference picture, and the 3794 number indicates a relative output time. 3796 ... R1 N2 R3 N4 R5 ... 3798 The decoding order of these pictures from left to right is as follows: 3800 ... R1 R3 N2 R5 N4 ... 3802 The NAL units of pictures R1, R3, N2, R5, and N4 are marked with a 3803 DON equal to 1, 2, 3, 4, and 5, respectively. 3805 Each reference picture consists of three slice groups that are 3806 scattered as follows (a number denotes the slice group number for 3807 each macroblock in a QCIF frame): 3809 0 1 2 0 1 2 0 1 2 0 1 3810 2 0 1 2 0 1 2 0 1 2 0 3811 1 2 0 1 2 0 1 2 0 1 2 3812 0 1 2 0 1 2 0 1 2 0 1 3813 2 0 1 2 0 1 2 0 1 2 0 3814 1 2 0 1 2 0 1 2 0 1 2 3815 0 1 2 0 1 2 0 1 2 0 1 3816 2 0 1 2 0 1 2 0 1 2 0 3817 1 2 0 1 2 0 1 2 0 1 2 3819 For the sake of simplicity, we assume that all the macroblocks of a 3820 slice group are included in one slice. Three MTAPs are constructed 3821 from three consecutive reference pictures so that each MTAP contains 3822 three aggregation units, each of which contains all the macroblocks 3823 from one slice group. The first MTAP contains slice group 0 of 3824 picture R1, slice group 1 of picture R3, and slice group 2 of picture 3825 R5. The second MTAP contains slice group 1 of picture R1, slice 3826 group 2 of picture R3, and slice group 0 of picture R5. The third 3827 MTAP contains slice group 2 of picture R1, slice group 0 of picture 3828 R3, and slice group 1 of picture R5. Each non-reference picture is 3829 encapsulated into an STAP-B. 3831 Consequently, the transmission order of NAL units is the following: 3833 R1, slice group 0, DON 1, carried in MTAP,RTP SN: N 3834 R3, slice group 1, DON 2, carried in MTAP,RTP SN: N 3835 R5, slice group 2, DON 4, carried in MTAP,RTP SN: N 3836 R1, slice group 1, DON 1, carried in MTAP,RTP SN: N+1 3837 R3, slice group 2, DON 2, carried in MTAP,RTP SN: N+1 3838 R5, slice group 0, DON 4, carried in MTAP,RTP SN: N+1 3839 R1, slice group 2, DON 1, carried in MTAP,RTP SN: N+2 3840 R3, slice group 1, DON 2, carried in MTAP,RTP SN: N+2 3841 R5, slice group 0, DON 4, carried in MTAP,RTP SN: N+2 3842 N2, DON 3, carried in STAP-B, RTP SN: N+3 3843 N4, DON 5, carried in STAP-B, RTP SN: N+4 3845 The receiver is able to organize the NAL units back in decoding order 3846 based on the value of DON associated with each NAL unit. 3848 If one of the MTAPs is lost, the spatially adjacent and temporally 3849 co-located macroblocks are received and can be used to conceal the 3850 loss efficiently. If one of the STAPs is lost, the effect of the 3851 loss does not propagate temporally. 3853 13.3. Example of Robust Packet Scheduling 3855 An example of robust packet scheduling follows. The communication 3856 system used in the example consists of the following components in 3857 the order that the video is processed from source to sink: 3859 o camera and capturing 3860 o pre-encoding buffer 3861 o encoder 3862 o encoded picture buffer 3863 o transmitter 3864 o transmission channel 3865 o receiver 3866 o receiver buffer 3867 o decoder 3868 o decoded picture buffer 3869 o display 3871 The video communication system used in the example operates as 3872 follows. Note that processing of the video stream happens gradually 3873 and at the same time in all components of the system. The source 3874 video sequence is shot and captured to a pre-encoding buffer. The 3875 pre-encoding buffer can be used to order pictures from sampling order 3876 to encoding order or to analyze multiple uncompressed frames for bit 3877 rate control purposes, for example. In some cases, the pre-encoding 3878 buffer may not exist; instead, the sampled pictures are encoded right 3879 away. The encoder encodes pictures from the pre-encoding buffer and 3880 stores the output; i.e., coded pictures, to the encoded picture 3881 buffer. The transmitter encapsulates the coded pictures from the 3882 encoded picture buffer to transmission packets and sends them to a 3883 receiver through a transmission channel. The receiver stores the 3884 received packets to the receiver buffer. The receiver buffering 3885 process typically includes buffering for transmission delay jitter. 3886 The receiver buffer can also be used to recover correct decoding 3887 order of coded data. The decoder reads coded data from the receiver 3888 buffer and produces decoded pictures as output into the decoded 3889 picture buffer. The decoded picture buffer is used to recover the 3890 output (or display) order of pictures. Finally, pictures are 3891 displayed. 3893 In the following example figures, I denotes an IDR picture, R denotes 3894 a reference picture, N denotes a non-reference picture, and the 3895 number after I, R, or N indicates the sampling time relative to the 3896 previous IDR picture in decoding order. Values below the sequence of 3897 pictures indicate scaled system clock timestamps. The system clock 3898 is initialized arbitrarily in this example, and time runs from left 3899 to right. Each I, R, and N picture is mapped into the same timeline 3900 compared to the previous processing step, if any, assuming that 3901 encoding, transmission, and decoding take no time. Thus, events 3902 happening at the same time are located in the same column throughout 3903 all example figures. 3905 A subset of a sequence of coded pictures is depicted below in 3906 sampling order. 3908 ... N58 N59 I00 N01 N02 R03 N04 N05 R06 ... N58 N59 I00 N01 ... 3909 ... --|---|---|---|---|---|---|---|---|- ... -|---|---|---|- ... 3910 ... 58 59 60 61 62 63 64 65 66 ... 128 129 130 131 ... 3912 Figure 16 Sequence of pictures in sampling order 3914 The sampled pictures are buffered in the pre-encoding buffer to 3915 arrange them in encoding order. In this example, we assume that the 3916 non-reference pictures are predicted from both the previous and the 3917 next reference picture in output order, except for the non-reference 3918 pictures immediately preceding an IDR picture, which are predicted 3919 only from the previous reference picture in output order. Thus, the 3920 pre-encoding buffer has to contain at least two pictures, and the 3921 buffering causes a delay of two picture intervals. The output of the 3922 pre-encoding buffering process and the encoding (and decoding) order 3923 of the pictures are as follows: 3925 ... N58 N59 I00 R03 N01 N02 R06 N04 N05 ... 3926 ... -|---|---|---|---|---|---|---|---|- ... 3927 ... 60 61 62 63 64 65 66 67 68 ... 3929 Figure 17 Re-ordered pictures in the pre-encoding buffer 3931 The encoder or the transmitter can set the value of DON for each 3932 picture to a value of DON for the previous picture in decoding order 3933 plus one. 3935 For the sake of simplicity, let us assume that: 3937 o the frame rate of the sequence is constant, 3938 o each picture consists of only one slice, 3939 o each slice is encapsulated in a single NAL unit packet, 3940 o there is no transmission delay, and 3941 o pictures are transmitted at constant intervals (that is, 1 / 3942 (frame rate)). 3944 When pictures are transmitted in decoding order, they are received as 3945 follows: 3947 ... N58 N59 I00 R03 N01 N02 R06 N04 N05 ... 3948 ... -|---|---|---|---|---|---|---|---|- ... 3949 ... 60 61 62 63 64 65 66 67 68 ... 3951 Figure 18 Received pictures in decoding order 3953 The OPTIONAL sprop-interleaving-depth media type parameter is set to 3954 0, as the transmission (or reception) order is identical to the 3955 decoding order. 3957 The decoder has to buffer for one picture interval initially in its 3958 decoded picture buffer to organize pictures from decoding order to 3959 output order as depicted below: 3961 ... N58 N59 I00 N01 N02 R03 N04 N05 R06 ... 3962 ... -|---|---|---|---|---|---|---|---|- ... 3963 ... 61 62 63 64 65 66 67 68 69 ... 3965 Figure 19 Output order 3967 The amount of required initial buffering in the decoded picture 3968 buffer can be signaled in the buffering period SEI message or with 3969 the num_reorder_frames syntax element of H.264 video usability 3970 information. num_reorder_frames indicates the maximum number of 3971 frames, complementary field pairs, or non-paired fields that precede 3972 any frame, complementary field pair, or non-paired field in the 3973 sequence in decoding order and that follow it in output order. For 3974 the sake of simplicity, we assume that num_reorder_frames is used to 3975 indicate the initial buffer in the decoded picture buffer. In this 3976 example, num_reorder_frames is equal to 1. 3978 It can be observed that if the IDR picture I00 is lost during 3979 transmission and a retransmission request is issued when the value of 3980 the system clock is 62, there is one picture interval of time (until 3981 the system clock reaches timestamp 63) to receive the retransmitted 3982 IDR picture I00. 3984 Let us then assume that IDR pictures are transmitted two frame 3985 intervals earlier than their decoding position; i.e., the pictures 3986 are transmitted as follows: 3988 ... I00 N58 N59 R03 N01 N02 R06 N04 N05 ... 3989 ... --|---|---|---|---|---|---|---|---|- ... 3990 ... 62 63 64 65 66 67 68 69 70 ... 3992 Figure 20 Interleaving: Early IDR pictures in sending order 3994 The OPTIONAL sprop-interleaving-depth media type parameter is set 3995 equal to 1 according to its definition. (The value of sprop- 3996 interleaving-depth in this example can be derived as follows: Picture 3997 I00 is the only picture preceding picture N58 or N59 in transmission 3998 order and following it in decoding order. Except for pictures I00, 3999 N58, and N59, the transmission order is the same as the decoding 4000 order of pictures. As a coded picture is encapsulated into exactly 4001 one NAL unit, the value of sprop-interleaving-depth is equal to the 4002 maximum number of pictures preceding any picture in transmission 4003 order and following the picture in decoding order.) 4005 The receiver buffering process contains two pictures at a time 4006 according to the value of the sprop-interleaving-depth parameter and 4007 orders pictures from the reception order to the correct decoding 4008 order based on the value of DON associated with each picture. The 4009 output of the receiver buffering process is as follows: 4011 ... N58 N59 I00 R03 N01 N02 R06 N04 N05 ... 4012 ... -|---|---|---|---|---|---|---|---|- ... 4013 ... 63 64 65 66 67 68 69 70 71 ... 4015 Figure 21 Interleaving: Receiver buffer 4017 Again, an initial buffering delay of one picture interval is needed 4018 to organize pictures from decoding order to output order, as depicted 4019 below: 4021 ... N58 N59 I00 N01 N02 R03 N04 N05 ... 4022 ... -|---|---|---|---|---|---|---|- ... 4023 ... 64 65 66 67 68 69 70 71 ... 4025 Figure 22 Interleaving: Receiver buffer after reordering 4027 Note that the maximum delay that IDR pictures can undergo during 4028 transmission, including possible application, transport, or link 4029 layer retransmission, is equal to three picture intervals. Thus, the 4030 loss resiliency of IDR pictures is improved in systems supporting 4031 retransmission compared to the case in which pictures were 4032 transmitted in their decoding order. 4034 13.4. Robust Transmission Scheduling of Redundant Coded Slices 4036 A redundant coded picture is a coded representation of a picture or a 4037 part of a picture that is not used in the decoding process if the 4038 corresponding primary coded picture is correctly decoded. There 4039 should be no noticeable difference between any area of the decoded 4040 primary picture and a corresponding area that would result from 4041 application of the H.264 decoding process for any redundant picture 4042 in the same access unit. A redundant coded slice is a coded slice 4043 that is a part of a redundant coded picture. 4045 Redundant coded pictures can be used to provide unequal error 4046 protection in error-prone video transmission. If a primary coded 4047 representation of a picture is decoded incorrectly, a corresponding 4048 redundant coded picture can be decoded. Examples of applications and 4049 coding techniques using the redundant codec picture feature include 4050 the video redundancy coding [23] and the protection of "key pictures" 4051 in multicast streaming [24]. 4053 One property of many error-prone video communications systems is that 4054 transmission errors are often bursty. Therefore, they may affect 4055 more than one consecutive transmission packets in transmission order. 4056 In low bit-rate video communication, it is relatively common that an 4057 entire coded picture can be encapsulated into one transmission packet. 4058 Consequently, a primary coded picture and the corresponding redundant 4059 coded pictures may be transmitted in consecutive packets in 4060 transmission order. To make the transmission scheme more tolerant of 4061 bursty transmission errors, it is beneficial to transmit the primary 4062 coded picture and redundant coded picture separated by more than a 4063 single packet. The DON concept enables this. 4065 13.5. Remarks on Other Design Possibilities 4067 The slice header syntax structure of the H.264 coding standard 4068 contains the frame_num syntax element that can indicate the decoding 4069 order of coded frames. However, the usage of the frame_num syntax 4070 element is not feasible or desirable to recover the decoding order, 4071 due to the following reasons: 4073 o The receiver is required to parse at least one slice header per 4074 coded picture (before passing the coded data to the decoder). 4076 o Coded slices from multiple coded video sequences cannot be 4077 interleaved, as the frame number syntax element is reset to 0 in 4078 each IDR picture. 4080 o The coded fields of a complementary field pair share the same 4081 value of the frame_num syntax element. Thus, the decoding order 4082 of the coded fields of a complementary field pair cannot be 4083 recovered based on the frame_num syntax element or any other 4084 syntax element of the H.264 coding syntax. 4086 The RTP payload format for transport of MPEG-4 elementary streams [25] 4087 enables interleaving of access units and transmission of multiple 4088 access units in the same RTP packet. An access unit is specified in 4089 the H.264 coding standard to comprise all NAL units associated with a 4090 primary coded picture according to subclause 7.4.1.2 of [1]. 4091 Consequently, slices of different pictures cannot be interleaved, and 4092 the multi-picture slice interleaving technique (see section 12.6) for 4093 improved error resilience cannot be used. 4095 14. Acknowledgements 4097 Stephan Wenger, Miska Hannuksela, Thomas Stockhammer, Magnus 4098 Westerlund, and David Singer are thanked as the authors of RFC 3984. 4099 Dave Lindbergh, Philippe Gentric, Gonzalo Camarillo, Gary Sullivan, 4100 Joerg Ott, and Colin Perkins are thanked for careful review during 4101 the development of RFC 3984. Randell Jesup, Stephen Botzko, Magnus 4102 Westerlund, Alex Eleftheriadis, Thomas Schierl, and Tom Taylor are 4103 thanked for their valuable comments and inputs during the development 4104 of this memo. 4106 This document was prepared using 2-Word-v2.0.template.dot. 4108 15. References 4110 15.1. Normative References 4112 [1] ITU-T Recommendation H.264, "Advanced video coding for generic 4113 audiovisual services", November 2007. 4115 [2] ISO/IEC International Standard 14496-10:2008. 4117 [3] ITU-T Recommendation H.241, "Extended video procedures and 4118 control signals for H.300 series terminals", May 2006. 4120 [4] Bradner, S., "Key words for use in RFCs to Indicate Requirement 4121 Levels", BCP 14, RFC 2119, March 1997. 4123 [5] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson, 4124 "RTP: A Transport Protocol for Real-Time Applications", STD 64, 4125 RFC 3550, July 2003. 4127 [6] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session 4128 Description Protocol", RFC 4566, July 2006. 4130 [7] Josefsson, S., "The Base16, Base32, and Base64 Data Encodings", 4131 RFC 3548, July 2003. 4133 [8] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model with 4134 Session Description Protocol (SDP)", RFC 3264, June 2002. 4136 [9] Lennox, J., Ott, J., and Schierl, T., "Source-Specific Media 4137 Attributes in the Session Description Protocol", draft-ietf- 4138 mmusic-sdp-source-attributes-02 (work in progress), October 4139 2008. 4141 15.2. Informative References 4143 [10] Luthra, A., Sullivan, G.J., and T. Wiegand (eds.), Special 4144 Issue on H.264/AVC. IEEE Transactions on Circuits and Systems 4145 on Video Technology, July 2003. 4147 [11] Bormann, C., Cline, L., Deisher, G., Gardos, T., Maciocco, C., 4148 Newell, D., Ott, J., Sullivan, G., Wenger, S., and C. Zhu, "RTP 4149 Payload Format for the 1998 Version of ITU-T Rec. H.263 Video 4150 (H.263+)", RFC 2429, October 1998. 4152 [12] ISO/IEC IS 14496-2. 4154 [13] Wenger, S., "H.26L over IP", IEEE Transaction on Circuits and 4155 Systems for Video technology, Vol. 13, No. 7, July 2003. 4157 [14] Wenger, S., "H.26L over IP: The IP Network Adaptation Layer", 4158 Proceedings Packet Video Workshop 02, April 2002. 4160 [15] Stockhammer, T., Hannuksela, M.M., and S. Wenger, "H.26L/JVT 4161 Coding Network Abstraction Layer and IP-based Transport" in 4162 Proc. ICIP 2002, Rochester, NY, September 2002. 4164 [16] Schulzrinne, H. and S. Casner, "RTP Profile for Audio and Video 4165 Conferences with Minimal Control", STD 65, RFC 3551, July 2003. 4167 [17] ITU-T Recommendation H.223, "Multiplexing protocol for low bit 4168 rate multimedia communication", July 2001. 4170 [18] Li, A., "RTP Payload Format for Generic Forward Error 4171 Correction", RFC 5109, December 2007. 4173 [19] Stockhammer, T., Wiegand, T., Oelbaum, T., and F. Obermeier, 4174 "Video Coding and Transport Layer Techniques for H.264/AVC- 4175 Based Transmission over Packet-Lossy Networks", IEEE 4176 International Conference on Image Processing (ICIP 2003), 4177 Barcelona, Spain, September 2003. 4179 [20] Varsa, V. and M. Karczewicz, "Slice interleaving in compressed 4180 video packetization", Packet Video Workshop 2000. 4182 [21] Kang, S.H. and A. Zakhor, "Packet scheduling algorithm for 4183 wireless video streaming," International Packet Video Workshop 4184 2002. 4186 [22] Hannuksela, M.M., "Enhanced concept of GOP", JVT-B042, 4187 available http://ftp3.itu.int/av-arch/video-site/0201_Gen/JVT- 4188 B042.doc, anuary 2002. 4190 [23] Wenger, S., "Video Redundancy Coding in H.263+", 1997 4191 International Workshop on Audio-Visual Services over Packet 4192 Networks, September 1997. 4194 [24] Wang, Y.-K., Hannuksela, M.M., and M. Gabbouj, "Error Resilient 4195 Video Coding Using Unequally Protected Key Pictures", in Proc. 4196 International Workshop VLBV03, September 2003. 4198 [25] van der Meer, J., Mackie, D., Swaminathan, V., Singer, D., and 4199 P. Gentric, "RTP Payload Format for Transport of MPEG-4 4200 Elementary Streams", RFC 3640, November 2003. 4202 [26] Baugher, M., McGrew, D., Naslund, M., Carrara, E., and K. 4203 Norrman, "The Secure Real-time Transport Protocol (SRTP)", RFC 4204 3711, March 2004. 4206 [27] Schulzrinne, H., Rao, A., and R. Lanphier, "Real Time Streaming 4207 Protocol (RTSP)", RFC 2326, April 1998. 4209 [28] Handley, M., Perkins, C., and E. Whelan, "Session Announcement 4210 Protocol", RFC 2974, October 2000. 4212 [29] Westerlund, M. and S. Wenger, "RTP Topologies", RFC 5117, 4213 January 2008. 4215 [30] Wenger, S., Chandra, U., and M. Westerlund, "Codec Control 4216 Messages in the RTP Audio-Visual Profile with Feedback (AVPF)", 4217 RFC 5104, February 2008. 4219 16. Authors' Addresses 4221 Ye-Kui Wang 4222 Huawei Technologies 4223 400 Somerset Corporate Blvd 4224 Bridgewater, NJ 08807 4225 USA 4227 Phone: +1-908-393-4758 4228 EMail: yekuiwang@huawei.com 4230 Roni Even 4231 14 David Hamelech 4232 Tel Aviv 64953 4233 Israel 4235 Phone: +972-545481099 4236 Email: ron.even.tlv@gmail.com 4238 Tom Kristensen 4239 TANDBERG 4240 Philip Pedersens vei 22 4241 N-1366 Lysaker 4242 Norway 4244 Phone: +47 67125125 4245 Email: tom.kristensen@tandberg.com, tomkri@ifi.uio.no 4247 17. Backward Compatibility to RFC 3984 4249 The current document is a revision of RFC 3984 and intends to 4250 obsolete it. This section addresses the backward compatibility 4251 issues. 4253 The technical changes are listed in section 18. 4255 Items 1), 2), 3), 7), 9), 10), 12), 13) are bug-fix type of changes, 4256 and do not incur any backward compatibility issues. 4258 Item 4), addition of six new media type parameters, does not incur 4259 any backward compatibility issues for SDP Offer/Answer based 4260 applications, as legacy RFC 3984 receivers ignore these parameters, 4261 and it is fine for legacy RFC 3984 senders not to use these 4262 parameters as they are optional. However, there is a backward 4263 compatibility issue for SDP declarative usage based applications, e.g. 4264 those using RTSP and SAP, because the SDP receiver per RFC 3984 4265 cannot accept a session for which the SDP includes an unrecognized 4266 parameter. Therefore, the RTSP or SAP server may have to prepare two 4267 sets of streams, one for legacy RFC 3984 receivers and one for 4268 receivers according to this memo. 4270 Items 5), 6) and 11) are related to out-of-band transport of 4271 parameter sets. There are following backward compatibility issues. 4273 1) When a legacy sender per RFC 3984 includes parameter sets for a 4274 level different than the default level indicated by profile-level- 4275 id to sprop-parameter-sets, the parameter value of sprop- 4276 parameter-sets is invalid to the receiver per this memo and 4277 therefore the session may be rejected. 4279 2) In SDP Offer/Answer between a legacy offerer per RFC 3984 and an 4280 answerer per this memo, when the answerer includes in the answer 4281 parameter sets that are not a superset of the parameter sets 4282 included in the offer, the parameter value of sprop-parameter-sets 4283 is invalid to offerer and the session may not be initiated 4284 properly (related to change item 11)). 4286 3) When one endpoint A per this memo includes in-band-parameter-sets 4287 equal to 1, the other side B per RFC 3984 does not understand that 4288 it must transmit parameter sets in-band and B may still exclude 4289 parameter sets in the in-band stream it is sending. Consequently 4290 endpoint A cannot decode the stream it receives. 4292 Item 7), allowance of conveying sprop-parameter-sets and sprop-level- 4293 parameter-sets using the "fmtp" source attribute as specified in 4294 section 6.3 of [9], is similar as item 4). It does not incur any 4295 backward compatibility issues for SDP Offer/Answer based applications, 4296 as legacy RFC 3984 receivers ignore the "fmtp" source attribute, and 4297 it is fine for legacy RFC 3984 senders not to use the "fmtp" source 4298 attribute as it is optional. However, there is a backward 4299 compatibility issue for SDP declarative usage based applications, e.g. 4300 those using RTSP and SAP, because the SDP receiver per RFC 3984 4301 cannot accept a session for which the SDP includes an unrecognized 4302 parameter (i.e., the "fmtp" source attribute). Therefore, the RTSP 4303 or SAP server may have to prepare two sets of streams, one for legacy 4304 RFC 3984 receivers and one for receivers according to this memo. 4306 Item 14) removed that use of out-of-band transport of parameter sets 4307 is recommended. As out-of-band transport of parameter sets is still 4308 allowed, this change does not incur any backward compatibility issues. 4310 Item 15) does not incur any backward compatibility issues as the 4311 added subsection 8.5 is informative. 4313 18. Changes from RFC 3984 4315 Following is the list of technical changes (including bug fixes) from 4316 RFC 3984. Besides this list of technical changes, numerous editorial 4317 changes have been made, but not documented in this memo. 4319 1) In subsections 5.4, 5.5, 6.2, 6,3 and 6.4, removed that the 4320 packetization mode in use may be signaled by external means. 4322 2) In subsection 7.2.2, changed the sentence 4324 There are N VCL NAL units in the deinterleaving buffer. 4326 to 4328 There are N or more VCL NAL units in the de-interleaving buffer. 4330 3) In subsection 8.1, the semantics of sprop-init-buf-time, paragraph 4331 2, changed the sentence 4333 The parameter is the maximum value of (transmission time of a NAL 4334 unit - decoding time of the NAL unit), assuming reliable and 4335 instantaneous transmission, the same timeline for transmission 4336 and decoding, and that decoding starts when the first packet 4337 arrives. 4339 to 4341 The parameter is the maximum value of (decoding time of the NAL 4342 unit - transmission time of a NAL unit), assuming reliable and 4343 instantaneous transmission, the same timeline for transmission 4344 and decoding, and that decoding starts when the first packet 4345 arrives. 4347 4) Added six new media type parameters, namely max-smbps, sprop- 4348 level-parameter-sets, use-level-src-parameter-sets, in-band- 4349 parameter-sets, sar-understood and sar-supported. 4351 5) In subsection 8.1, removed the specification of parameter-add. 4352 Other descriptions of parameter-add (in subsections 8.2 and 8.4) 4353 are also removed. 4355 6) In subsection 8.1, added a constraint to sprop-parameter-sets such 4356 that it can only contain parameter sets for the same profile and 4357 level as indicated by profile-level-id. 4359 7) In subsection 8.2.1, added that sprop-parameter-sets and sprop- 4360 level-parameter-sets may be either included in the "a=fmtp" line 4361 of SDP or conveyed using the "fmtp" source attribute as specified 4362 in section 6.3 of [9]. 4364 8) In subsection 8.2.2, removed sprop-deint-buf-req from being part 4365 of the media format configuration in usage with the SDP 4366 Offer/Answer model. 4368 9) In subsection 8.2.2, made it clear that level is downgradable in 4369 the SDP Offer/Answer model, i.e. the use of the level part of 4370 "profile-level-id" does not need to be symmetric (the level 4371 included in the answer can be lower than or equal to the level 4372 included in the offer). 4374 10)In subsection 8.2.2, removed that the capability parameters may be 4375 used to declare encoding capabilities. 4377 11)In subsection 8.2.2, added rules on how to use sprop-parameter- 4378 sets and sprop-level-parameter-sets for out-of-band transport of 4379 parameter sets, with or without level downgrading. 4381 12)In subsection 8.2.2, clarified the rules of using the media type 4382 parameters with SDP Offer/Answer for multicast. 4384 13)In subsection 8.2.2, completed and corrected the list of how 4385 different media type parameters shall be interpreted in the 4386 different combinations of offer or answer and direction attribute. 4388 14)In subsection 8.4, changed the text such that both out-of-band and 4389 in-band transport of parameter sets are allowed and neither is 4390 recommended or required. 4392 15)Added subsection 8.5 (informative) providing example methods for 4393 decoder refresh to handle parameter set losses.