idnits 2.17.1 draft-ietf-avt-rtp-rfc3984bis-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 17. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 4253. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 4230. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 4237. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 4243. 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 2 instances of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust 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. -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (December 16, 2008) is 5609 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 4094, 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 2327 (ref. '6') (Obsoleted by RFC 4566) ** 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: 3 errors (**), 0 flaws (~~), 4 warnings (==), 14 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 Nokia 3 Intended status: Standards track R. Even 4 Expires: June 2009 Self-employed 5 T. Kristensen 6 Tandberg 7 December 16, 2008 9 RTP Payload Format for H.264 Video 10 draft-ietf-avt-rtp-rfc3984bis-02.txt 12 Status of this Memo 14 By submitting this Internet-Draft, each author represents that any 15 applicable patent or other IPR claims of which he or she is aware 16 have been or will be disclosed, and any of which he or she becomes 17 aware will be disclosed, in accordance with Section 6 of BCP 79. 19 Internet-Drafts are working documents of the Internet Engineering 20 Task Force (IETF), its areas, and its working groups. Note that 21 other groups may also distribute working documents as Internet- 22 Drafts. 24 Internet-Drafts are draft documents valid for a maximum of six months 25 and may be updated, replaced, or obsoleted by other documents at any 26 time. It is inappropriate to use Internet-Drafts as reference 27 material or to cite them other than as "work in progress." 29 The list of current Internet-Drafts can be accessed at 30 http://www.ietf.org/ietf/1id-abstracts.txt 32 The list of Internet-Draft Shadow Directories can be accessed at 33 http://www.ietf.org/shadow.html 35 This Internet-Draft will expire on June 16, 2009. 37 Copyright Notice 39 Copyright (C) The IETF Trust (2008). 41 Abstract 43 This memo describes an RTP Payload format for the ITU-T 44 Recommendation H.264 video codec and the technically identical 45 ISO/IEC International Standard 14496-10 video codec, excluding the 46 Scalable Video Coding (SVC) extension and the Multivew Video Coding 47 extension, for which the RTP payload formats are defined elsewhere. 48 The RTP payload format allows for packetization of one or more 49 Network Abstraction Layer Units (NALUs), produced by an H.264 video 50 encoder, in each RTP payload. The payload format has wide 51 applicability, as it supports applications from simple low bit-rate 52 conversational usage, to Internet video streaming with interleaved 53 transmission, to high bit-rate video-on-demand. 55 This memo intends to obsolete RFC 3984. Changes from RFC 3984 are 56 summarized in section 17. Issues on backward compatibility to RFC 57 3984 are discussed in section 16. 59 Table of Contents 61 1. Introduction...................................................3 62 1.1. The H.264 Codec...........................................3 63 1.2. Parameter Set Concept.....................................3 64 1.3. Network Abstraction Layer Unit Types......................3 65 2. Conventions....................................................3 66 3. Scope..........................................................3 67 4. Definitions and Abbreviations..................................3 68 4.1. Definitions...............................................3 69 4.2. Abbreviations.............................................3 70 5. RTP Payload Format.............................................3 71 5.1. RTP Header Usage..........................................3 72 5.2. Payload Structures........................................3 73 5.3. NAL Unit Header Usage.....................................3 74 5.4. Packetization Modes.......................................3 75 5.5. Decoding Order Number (DON)...............................3 76 5.6. Single NAL Unit Packet....................................3 77 5.7. Aggregation Packets.......................................3 78 5.7.1. Single-Time Aggregation Packet.......................3 79 5.7.2. Multi-Time Aggregation Packets (MTAPs)...............3 80 5.7.3. Fragmentation Units (FUs)............................3 81 6. Packetization Rules............................................3 82 6.1. Common Packetization Rules................................3 83 6.2. Single NAL Unit Mode......................................3 84 6.3. Non-Interleaved Mode......................................3 85 6.4. Interleaved Mode..........................................3 86 7. De-Packetization Process.......................................3 87 7.1. Single NAL Unit and Non-Interleaved Mode..................3 88 7.2. Interleaved Mode..........................................3 89 7.2.1. Size of the De-interleaving Buffer...................3 90 7.2.2. De-interleaving Process..............................3 91 7.3. Additional De-Packetization Guidelines....................3 92 8. Payload Format Parameters......................................3 93 8.1. Media Type Registration...................................3 94 8.2. SDP Parameters............................................3 95 8.2.1. Mapping of Payload Type Parameters to SDP............3 96 8.2.2. Usage with the SDP Offer/Answer Model................3 97 8.2.3. Usage in Declarative Session Descriptions............3 98 8.3. Examples..................................................3 99 8.4. Parameter Set Considerations..............................3 100 8.5. Decoder Refresh Point Procedure using In-Band Transport of 101 Parameter Sets (Informative)...................................3 102 8.5.1. IDR Procedure to Respond to a Request for a Decoder 103 Refresh Point...............................................3 104 8.5.2. Gradual Recovery Procedure to Respond to a Request for a 105 Decoder Refresh Point.......................................3 106 9. Security Considerations........................................3 107 10. Congestion Control............................................3 108 11. IANA Consideration............................................3 109 12. Informative Appendix: Application Examples....................3 110 12.1. Video Telephony according to ITU-T Recommendation H.241 111 Annex A........................................................3 112 12.2. Video Telephony, No Slice Data Partitioning, No NAL Unit 113 Aggregation....................................................3 114 12.3. Video Telephony, Interleaved Packetization Using NAL Unit 115 Aggregation....................................................3 116 12.4. Video Telephony with Data Partitioning...................3 117 12.5. Video Telephony or Streaming with FUs and Forward Error 118 Correction.....................................................3 119 12.6. Low Bit-Rate Streaming...................................3 120 12.7. Robust Packet Scheduling in Video Streaming..............3 121 13. Informative Appendix: Rationale for Decoding Order Number.....3 122 13.1. Introduction.............................................3 123 13.2. Example of Multi-Picture Slice Interleaving..............3 124 13.3. Example of Robust Packet Scheduling......................3 125 13.4. Robust Transmission Scheduling of Redundant Coded Slices.3 126 13.5. Remarks on Other Design Possibilities....................3 127 14. Acknowledgements..............................................3 128 15. References....................................................3 129 15.1. Normative References.....................................3 130 15.2. Informative References...................................3 131 Authors' Addresses................................................3 132 Intellectual Property Statement...................................3 133 Disclaimer of Validity............................................3 134 Acknowledgement...................................................3 135 16. Backward Compatibility to RFC 3984............................3 136 17. Changes from RFC 3984.........................................3 138 1. Introduction 140 This memo intends to obsolete RFC 3984. Changes from RFC 3984 are 141 summarized in section 17. Issues on backward compatibility to RFC 142 3984 are discussed in section 16. 144 1.1. The H.264 Codec 146 This memo specifies an RTP payload specification for the video coding 147 standard known as ITU-T Recommendation H.264 [1] and ISO/IEC 148 International Standard 14496 Part 10 [2] (both also known as Advanced 149 Video Coding, or AVC). In this memo the H.264 acronym is used for 150 the codec and the standard, but the memo is equally applicable to the 151 ISO/IEC counterpart of the coding standard. 153 The H.264 video codec has a very broad application range that covers 154 all forms of digital compressed video from, low bit-rate Internet 155 streaming applications to HDTV broadcast and Digital Cinema 156 applications with nearly lossless coding. Compared to the current 157 state of technology, the overall performance of H.264 is such that 158 bit rate savings of 50% or more are reported. Digital Satellite TV 159 quality, for example, was reported to be achievable at 1.5 Mbit/s, 160 compared to the current operation point of MPEG 2 video at around 3.5 161 Mbit/s [10]. 163 The codec specification [1] itself distinguishes conceptually between 164 a video coding layer (VCL) and a network abstraction layer (NAL). 165 The VCL contains the signal processing functionality of the codec; 166 mechanisms such as transform, quantization, and motion compensated 167 prediction; and a loop filter. It follows the general concept of 168 most of today's video codecs, a macroblock-based coder that uses 169 inter picture prediction with motion compensation and transform 170 coding of the residual signal. The VCL encoder outputs slices: a bit 171 string that contains the macroblock data of an integer number of 172 macroblocks, and the information of the slice header (containing the 173 spatial address of the first macroblock in the slice, the initial 174 quantization parameter, and similar information). Macroblocks in 175 slices are arranged in scan order unless a different macroblock 176 allocation is specified, by using the so-called Flexible Macroblock 177 Ordering syntax. In-picture prediction is used only within a slice. 178 More information is provided in [10]. 180 The Network Abstraction Layer (NAL) encoder encapsulates the slice 181 output of the VCL encoder into Network Abstraction Layer Units (NAL 182 units), which are suitable for transmission over packet networks or 183 use in packet oriented multiplex environments. Annex B of H.264 184 defines an encapsulation process to transmit such NAL units over 185 byte-stream oriented networks. In the scope of this memo, Annex B is 186 not relevant. 188 Internally, the NAL uses NAL units. A NAL unit consists of a one- 189 byte header and the payload byte string. The header indicates the 190 type of the NAL unit, the (potential) presence of bit errors or 191 syntax violations in the NAL unit payload, and information regarding 192 the relative importance of the NAL unit for the decoding process. 193 This RTP payload specification is designed to be unaware of the bit 194 string in the NAL unit payload. 196 One of the main properties of H.264 is the complete decoupling of the 197 transmission time, the decoding time, and the sampling or 198 presentation time of slices and pictures. The decoding process 199 specified in H.264 is unaware of time, and the H.264 syntax does not 200 carry information such as the number of skipped frames (as is common 201 in the form of the Temporal Reference in earlier video compression 202 standards). Also, there are NAL units that affect many pictures and 203 that are, therefore, inherently timeless. For this reason, the 204 handling of the RTP timestamp requires some special considerations 205 for NAL units for which the sampling or presentation time is not 206 defined or, at transmission time, unknown. 208 1.2. Parameter Set Concept 210 One very fundamental design concept of H.264 is to generate self- 211 contained packets, to make mechanisms such as the header duplication 212 of RFC 2429 [11] or MPEG-4's Header Extension Code (HEC) [12] 213 unnecessary. This was achieved by decoupling information relevant to 214 more than one slice from the media stream. This higher layer meta 215 information should be sent reliably, asynchronously, and in advance 216 from the RTP packet stream that contains the slice packets. 217 (Provisions for sending this information in-band are also available 218 for applications that do not have an out-of-band transport channel 219 appropriate for the purpose.) The combination of the higher-level 220 parameters is called a parameter set. The H.264 specification 221 includes two types of parameter sets: sequence parameter set and 222 picture parameter set. An active sequence parameter set remains 223 unchanged throughout a coded video sequence, and an active picture 224 parameter set remains unchanged within a coded picture. The sequence 225 and picture parameter set structures contain information such as 226 picture size, optional coding modes employed, and macroblock to slice 227 group map. 229 To be able to change picture parameters (such as the picture size) 230 without having to transmit parameter set updates synchronously to the 231 slice packet stream, the encoder and decoder can maintain a list of 232 more than one sequence and picture parameter set. Each slice header 233 contains a codeword that indicates the sequence and picture parameter 234 set to be used. 236 This mechanism allows the decoupling of the transmission of parameter 237 sets from the packet stream, and the transmission of them by external 238 means (e.g., as a side effect of the capability exchange), or through 239 a (reliable or unreliable) control protocol. It may even be possible 240 that they are never transmitted but are fixed by an application 241 design specification. 243 1.3. Network Abstraction Layer Unit Types 245 Tutorial information on the NAL design can be found in [13], [14], 246 and [15]. 248 All NAL units consist of a single NAL unit type octet, which also co- 249 serves as the payload header of this RTP payload format. The payload 250 of a NAL unit follows immediately. 252 The syntax and semantics of the NAL unit type octet are specified in 253 [1], but the essential properties of the NAL unit type octet are 254 summarized below. The NAL unit type octet has the following format: 256 +---------------+ 257 |0|1|2|3|4|5|6|7| 258 +-+-+-+-+-+-+-+-+ 259 |F|NRI| Type | 260 +---------------+ 262 The semantics of the components of the NAL unit type octet, as 263 specified in the H.264 specification, are described briefly below. 265 F: 1 bit 266 forbidden_zero_bit. The H.264 specification declares a value of 267 1 as a syntax violation. 269 NRI: 2 bits 270 nal_ref_idc. A value of 00 indicates that the content of the NAL 271 unit is not used to reconstruct reference pictures for inter 272 picture prediction. Such NAL units can be discarded without 273 risking the integrity of the reference pictures. Values greater 274 than 00 indicate that the decoding of the NAL unit is required to 275 maintain the integrity of the reference pictures. 277 Type: 5 bits 278 nal_unit_type. This component specifies the NAL unit payload 279 type as defined in Table 7-1 of [1], and later within this memo. 280 For a reference of all currently defined NAL unit types and their 281 semantics, please refer to section 7.4.1 in [1]. 283 This memo introduces new NAL unit types, which are presented in 284 section 5.2. The NAL unit types defined in this memo are marked as 285 unspecified in [1]. Moreover, this specification extends the 286 semantics of F and NRI as described in section 5.3. 288 2. Conventions 290 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 291 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 292 document are to be interpreted as described in RFC-2119 [3]. 294 This specification uses the notion of setting and clearing a bit when 295 bit fields are handled. Setting a bit is the same as assigning that 296 bit the value of 1 (On). Clearing a bit is the same as assigning 297 that bit the value of 0 (Off). 299 3. Scope 301 This payload specification can only be used to carry the "naked" 302 H.264 NAL unit stream over RTP, and not the bitstream format 303 discussed in Annex B of H.264. Likely, the first applications of 304 this specification will be in the conversational multimedia field, 305 video telephony or video conferencing, but the payload format also 306 covers other applications, such as Internet streaming and TV over IP. 308 4. Definitions and Abbreviations 310 4.1. Definitions 312 This document uses the definitions of [1]. The following terms, 313 defined in [1], are summed up for convenience: 315 access unit: A set of NAL units always containing a primary coded 316 picture. In addition to the primary coded picture, an access 317 unit may also contain one or more redundant coded pictures or 318 other NAL units not containing slices or slice data partitions of 319 a coded picture. The decoding of an access unit always results 320 in a decoded picture. 322 coded video sequence: A sequence of access units that consists, 323 in decoding order, of an instantaneous decoding refresh (IDR) 324 access unit followed by zero or more non-IDR access units 325 including all subsequent access units up to but not including any 326 subsequent IDR access unit. 328 IDR access unit: An access unit in which the primary coded 329 picture is an IDR picture. 331 IDR picture: A coded picture containing only slices with I or SI 332 slice types that causes a "reset" in the decoding process. After 333 the decoding of an IDR picture, all following coded pictures in 334 decoding order can be decoded without inter prediction from any 335 picture decoded prior to the IDR picture. 337 primary coded picture: The coded representation of a picture to 338 be used by the decoding process for a bitstream conforming to 339 H.264. The primary coded picture contains all macroblocks of the 340 picture. 342 redundant coded picture: A coded representation of a picture or a 343 part of a picture. The content of a redundant coded picture 344 shall not be used by the decoding process for a bitstream 345 conforming to H.264. The content of a redundant coded picture 346 may be used by the decoding process for a bitstream that contains 347 errors or losses. 349 VCL NAL unit: A collective term used to refer to coded slice and 350 coded data partition NAL units. 352 In addition, the following definitions apply: 354 decoding order number (DON): A field in the payload structure or 355 a derived variable indicating NAL unit decoding order. Values of 356 DON are in the range of 0 to 65535, inclusive. After reaching 357 the maximum value, the value of DON wraps around to 0. 359 NAL unit decoding order: A NAL unit order that conforms to the 360 constraints on NAL unit order given in section 7.4.1.2 in [1]. 362 NALU-time: The value that the RTP timestamp would have if the NAL 363 unit would be transported in its own RTP packet. 365 transmission order: The order of packets in ascending RTP 366 sequence number order (in modulo arithmetic). Within an 367 aggregation packet, the NAL unit transmission order is the same 368 as the order of appearance of NAL units in the packet. 370 media aware network element (MANE): A network element, such as a 371 middlebox or application layer gateway that is capable of parsing 372 certain aspects of the RTP payload headers or the RTP payload and 373 reacting to the contents. 375 Informative note: The concept of a MANE goes beyond normal 376 routers or gateways in that a MANE has to be aware of the 377 signaling (e.g., to learn about the payload type mappings of 378 the media streams), and in that it has to be trusted when 379 working with SRTP. The advantage of using MANEs is that they 380 allow packets to be dropped according to the needs of the 381 media coding. For example, if a MANE has to drop packets due 382 to congestion on a certain link, it can identify and remove 383 those packets whose elimination produces the least adverse 384 effect on the user experience. 386 static macroblock: A certain amount of macroblocks in the video 387 stream can be defined as static, as defined in section 8.3.2.8 in 388 [3]. Static macroblocks free up additional processing cycles for 389 the handling of non-static macroblocks. Based on a given amount 390 of video processing resources and a given resolution, a higher 391 number of static macroblocks enables a correspondingly higher 392 frame rate. 394 default sub-profile: The subset of coding tools, which may be all 395 coding tools of one profile or the common subset of coding tools 396 of more than one profile, indicated by the profile-level-id 397 parameter. In SDP Offer/Answer, the default sub-profile must be 398 used in a symmetric manner, i.e. the answer must either use the 399 same sub-profile as the offer or reject the offer. 401 default level: The level indicated by the profile-level-id 402 parameter. In SDP Offer/Answer, level is downgradable, i.e., the 403 answer may either use the default level or a lower level. 405 4.2. Abbreviations 407 DON: Decoding Order Number 408 DONB: Decoding Order Number Base 409 DOND: Decoding Order Number Difference 410 FEC: Forward Error Correction 411 FU: Fragmentation Unit 412 IDR: Instantaneous Decoding Refresh 413 IEC: International Electrotechnical Commission 414 ISO: International Organization for Standardization 415 ITU-T: International Telecommunication Union, 416 Telecommunication Standardization Sector 417 MANE: Media Aware Network Element 418 MTAP: Multi-Time Aggregation Packet 419 MTAP16: MTAP with 16-bit timestamp offset 420 MTAP24: MTAP with 24-bit timestamp offset 421 NAL: Network Abstraction Layer 422 NALU: NAL Unit 423 SAR: Sample Aspect Ratio 424 SEI: Supplemental Enhancement Information 425 STAP: Single-Time Aggregation Packet 426 STAP-A: STAP type A 427 STAP-B: STAP type B 428 TS: Timestamp 429 VCL: Video Coding Layer 430 VUI: Video Usability Information 432 5. RTP Payload Format 434 5.1. RTP Header Usage 436 The format of the RTP header is specified in RFC 3550 [5] and 437 reprinted in Figure 1 for convenience. This payload format uses the 438 fields of the header in a manner consistent with that specification. 440 When one NAL unit is encapsulated per RTP packet, the RECOMMENDED RTP 441 payload format is specified in section 5.6. The RTP payload (and the 442 settings for some RTP header bits) for aggregation packets and 443 fragmentation units are specified in sections 5.7 and 5.8, 444 respectively. 446 0 1 2 3 447 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 1 448 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 449 |V=2|P|X| CC |M| PT | sequence number | 450 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 451 | timestamp | 452 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 453 | synchronization source (SSRC) identifier | 454 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 455 | contributing source (CSRC) identifiers | 456 | .... | 457 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 459 Figure 1 RTP header according to RFC 3550 461 The RTP header information to be set according to this RTP payload 462 format is set as follows: 464 Marker bit (M): 1 bit 465 Set for the very last packet of the access unit indicated by the 466 RTP timestamp, in line with the normal use of the M bit in video 467 formats, to allow an efficient playout buffer handling. For 468 aggregation packets (STAP and MTAP), the marker bit in the RTP 469 header MUST be set to the value that the marker bit of the last 470 NAL unit of the aggregation packet would have been if it were 471 transported in its own RTP packet. Decoders MAY use this bit as 472 an early indication of the last packet of an access unit, but 473 MUST NOT rely on this property. 475 Informative note: Only one M bit is associated with an 476 aggregation packet carrying multiple NAL units. Thus, if a 477 gateway has re-packetized an aggregation packet into several 478 packets, it cannot reliably set the M bit of those packets. 480 Payload type (PT): 7 bits 481 The assignment of an RTP payload type for this new packet format 482 is outside the scope of this document and will not be specified 483 here. The assignment of a payload type has to be performed 484 either through the profile used or in a dynamic way. 486 Sequence number (SN): 16 bits 487 Set and used in accordance with RFC 3550. For the single NALU 488 and non-interleaved packetization mode, the sequence number is 489 used to determine decoding order for the NALU. 491 Timestamp: 32 bits 492 The RTP timestamp is set to the sampling timestamp of the 493 content. A 90 kHz clock rate MUST be used. 495 If the NAL unit has no timing properties of its own (e.g., 496 parameter set and SEI NAL units), the RTP timestamp is set to the 497 RTP timestamp of the primary coded picture of the access unit in 498 which the NAL unit is included, according to section 7.4.1.2 of 499 [1]. 501 The setting of the RTP Timestamp for MTAPs is defined in section 502 5.7.2. 504 Receivers SHOULD ignore any picture timing SEI messages included 505 in access units that have only one display timestamp. Instead, 506 receivers SHOULD use the RTP timestamp for synchronizing the 507 display process. 509 RTP senders SHOULD NOT transmit picture timing SEI messages for 510 pictures that are not supposed to be displayed as multiple 511 fields. 513 If one access unit has more than one display timestamp carried in 514 a picture timing SEI message, then the information in the SEI 515 message SHOULD be treated as relative to the RTP timestamp, with 516 the earliest event occurring at the time given by the RTP 517 timestamp, and subsequent events later, as given by the 518 difference in SEI message picture timing values. Let tSEI1, 519 tSEI2, ..., tSEIn be the display timestamps carried in the SEI 520 message of an access unit, where tSEI1 is the earliest of all 521 such timestamps. Let tmadjst() be a function that adjusts the 522 SEI messages time scale to a 90-kHz time scale. Let TS be the 523 RTP timestamp. Then, the display time for the event associated 524 with tSEI1 is TS. The display time for the event with tSEIx, 525 where x is [2..n] is TS + tmadjst (tSEIx - tSEI1). 527 Informative note: Displaying coded frames as fields is needed 528 commonly in an operation known as 3:2 pulldown, in which film 529 content that consists of coded frames is displayed on a 530 display using interlaced scanning. The picture timing SEI 531 message enables carriage of multiple timestamps for the same 532 coded picture, and therefore the 3:2 pulldown process is 533 perfectly controlled. The picture timing SEI message 534 mechanism is necessary because only one timestamp per coded 535 frame can be conveyed in the RTP timestamp. 537 Informative note: Because H.264 allows the decoding order to 538 be different from the display order, values of RTP timestamps 539 may not be monotonically non-decreasing as a function of RTP 540 sequence numbers. Furthermore, the value for inter-arrival 541 jitter reported in the RTCP reports may not be a trustworthy 542 indication of the network performance, as the calculation 543 rules for inter-arrival jitter (section 6.4.1 of RFC 3550) 544 assume that the RTP timestamp of a packet is directly 545 proportional to its transmission time. 547 5.2. Payload Structures 549 The payload format defines three different basic payload structures. 550 A receiver can identify the payload structure by the first byte of 551 the RTP packet payload, which co-serves as the RTP payload header 552 and, in some cases, as the first byte of the payload. This byte is 553 always structured as a NAL unit header. The NAL unit type field 554 indicates which structure is present. The possible structures are as 555 follows: 557 Single NAL Unit Packet: Contains only a single NAL unit in the 558 payload. The NAL header type field will be equal to the original NAL 559 unit type; i.e., in the range of 1 to 23, inclusive. Specified in 560 section 5.6. 562 Aggregation Packet: Packet type used to aggregate multiple NAL units 563 into a single RTP payload. This packet exists in four versions, the 564 Single-Time Aggregation Packet type A (STAP-A), the Single-Time 565 Aggregation Packet type B (STAP-B), Multi-Time Aggregation Packet 566 (MTAP) with 16-bit offset (MTAP16), and Multi-Time Aggregation Packet 567 (MTAP) with 24-bit offset (MTAP24). The NAL unit type numbers 568 assigned for STAP-A, STAP-B, MTAP16, and MTAP24 are 24, 25, 26, and 569 27, respectively. Specified in section 5.7. 571 Fragmentation Unit: Used to fragment a single NAL unit over multiple 572 RTP packets. Exists with two versions, FU-A and FU-B, identified 573 with the NAL unit type numbers 28 and 29, respectively. Specified in 574 section 5.8. 576 Informative note: This specification does not limit the size of 577 NAL units encapsulated in single NAL unit packets and 578 fragmentation units. The maximum size of a NAL unit encapsulated 579 in any aggregation packet is 65535 bytes. 581 Table 1 summarizes NAL unit types and the corresponding RTP packet 582 types when each of these NAL units is directly used a packet payload, 583 and where the types are described in this memo. 585 Table 1. Summary of NAL unit types and the corresponding packet 586 types 588 NAL Unit Packet Packet Type Name Section 589 Type Type 590 --------------------------------------------------------- 591 0 reserved - 592 1-23 NAL unit Single NAL unit packet 5.6 593 24 STAP-A Single-time aggregation packet 5.7.1 594 25 STAP-B Single-time aggregation packet 5.7.1 595 26 MTAP16 Multi-time aggregation packet 5.7.2 596 27 MTAP24 Multi-time aggregation packet 5.7.2 597 28 FU-A Fragmentation unit 5.8 598 29 FU-B Fragmentation unit 5.8 599 30-31 reserved - 601 5.3. NAL Unit Header Usage 603 The structure and semantics of the NAL unit header were introduced in 604 section 1.3. For convenience, the format of the NAL unit header is 605 reprinted below: 607 +---------------+ 608 |0|1|2|3|4|5|6|7| 609 +-+-+-+-+-+-+-+-+ 610 |F|NRI| Type | 611 +---------------+ 613 This section specifies the semantics of F and NRI according to this 614 specification. 616 F: 1 bit 617 forbidden_zero_bit. A value of 0 indicates that the NAL unit 618 type octet and payload should not contain bit errors or other 619 syntax violations. A value of 1 indicates that the NAL unit type 620 octet and payload may contain bit errors or other syntax 621 violations. 623 MANEs SHOULD set the F bit to indicate detected bit errors in the 624 NAL unit. The H.264 specification requires that the F bit is 625 equal to 0. When the F bit is set, the decoder is advised that 626 bit errors or any other syntax violations may be present in the 627 payload or in the NAL unit type octet. The simplest decoder 628 reaction to a NAL unit in which the F bit is equal to 1 is to 629 discard such a NAL unit and to conceal the lost data in the 630 discarded NAL unit. 632 NRI: 2 bits 633 nal_ref_idc. The semantics of value 00 and a non-zero value 634 remain unchanged from the H.264 specification. In other words, a 635 value of 00 indicates that the content of the NAL unit is not 636 used to reconstruct reference pictures for inter picture 637 prediction. Such NAL units can be discarded without risking the 638 integrity of the reference pictures. Values greater than 00 639 indicate that the decoding of the NAL unit is required to 640 maintain the integrity of the reference pictures. 642 In addition to the specification above, according to this RTP 643 payload specification, values of NRI indicate the relative 644 transport priority, as determined by the encoder. MANEs can use 645 this information to protect more important NAL units better than 646 they do less important NAL units. The highest transport priority 647 is 11, followed by 10, and then by 01; finally, 00 is the lowest. 649 Informative note: Any non-zero value of NRI is handled 650 identically in H.264 decoders. Therefore, receivers need not 651 manipulate the value of NRI when passing NAL units to the 652 decoder. 654 An H.264 encoder MUST set the value of NRI according to the H.264 655 specification (subclause 7.4.1) when the value of nal_unit_type 656 is in the range of 1 to 12, inclusive. In particular, the H.264 657 specification requires that the value of NRI SHALL be equal to 0 658 for all NAL units having nal_unit_type equal to 6, 9, 10, 11, or 659 12. 661 For NAL units having nal_unit_type equal to 7 or 8 (indicating a 662 sequence parameter set or a picture parameter set, respectively), 663 an H.264 encoder SHOULD set the value of NRI to 11 (in binary 664 format). For coded slice NAL units of a primary coded picture 665 having nal_unit_type equal to 5 (indicating a coded slice 666 belonging to an IDR picture), an H.264 encoder SHOULD set the 667 value of NRI to 11 (in binary format). 669 For a mapping of the remaining nal_unit_types to NRI values, the 670 following example MAY be used and has been shown to be efficient 671 in a certain environment [14]. Other mappings MAY also be 672 desirable, depending on the application and the H.264/AVC Annex A 673 profile in use. 675 Informative note: Data Partitioning is not available in 676 certain profiles; e.g., in the Main or Baseline profiles. 677 Consequently, the NAL unit types 2, 3, and 4 can occur only if 678 the video bitstream conforms to a profile in which data 679 partitioning is allowed and not in streams that conform to the 680 Main or Baseline profiles. 682 Table 2. Example of NRI values for coded slices and coded slice data 683 partitions of primary coded reference pictures 685 NAL Unit Type Content of NAL unit NRI (binary) 686 ---------------------------------------------------------------- 687 1 non-IDR coded slice 10 688 2 Coded slice data partition A 10 689 3 Coded slice data partition B 01 690 4 Coded slice data partition C 01 692 Informative note: As mentioned before, the NRI value of non- 693 reference pictures is 00 as mandated by H.264/AVC. 695 An H.264 encoder SHOULD set the value of NRI for coded slice and 696 coded slice data partition NAL units of redundant coded reference 697 pictures equal to 01 (in binary format). 699 Definitions of the values for NRI for NAL unit types 24 to 29, 700 inclusive, are given in sections 5.7 and 5.8 of this memo. 702 No recommendation for the value of NRI is given for NAL units 703 having nal_unit_type in the range of 13 to 23, inclusive, because 704 these values are reserved for ITU-T and ISO/IEC. No 705 recommendation for the value of NRI is given for NAL units having 706 nal_unit_type equal to 0 or in the range of 30 to 31, inclusive, 707 as the semantics of these values are not specified in this memo. 709 5.4. Packetization Modes 711 This memo specifies three cases of packetization modes: 713 o Single NAL unit mode 715 o Non-interleaved mode 717 o Interleaved mode 719 The single NAL unit mode is targeted for conversational systems that 720 comply with ITU-T Recommendation H.241 [3] (see section 12.1). The 721 non-interleaved mode is targeted for conversational systems that may 722 not comply with ITU-T Recommendation H.241. In the non-interleaved 723 mode, NAL units are transmitted in NAL unit decoding order. The 724 interleaved mode is targeted for systems that do not require very low 725 end-to-end latency. The interleaved mode allows transmission of NAL 726 units out of NAL unit decoding order. 728 The packetization mode in use MAY be signaled by the value of the 729 OPTIONAL packetization-mode media type parameter. The used 730 packetization mode governs which NAL unit types are allowed in RTP 731 payloads. Table 3 summarizes the allowed packet payload types for 732 each packetization mode. Packetization modes are explained in more 733 detail in section 6. 735 Table 3. Summary of allowed NAL unit types for each packetization 736 mode (yes = allowed, no = disallowed, ig = ignore) 738 Payload Packet Single NAL Non-Interleaved Interleaved 739 Type Type Unit Mode Mode Mode 740 ------------------------------------------------------------- 741 0 reserved ig ig ig 742 1-23 NAL unit yes yes no 743 24 STAP-A no yes no 744 25 STAP-B no no yes 745 26 MTAP16 no no yes 746 27 MTAP24 no no yes 747 28 FU-A no yes yes 748 29 FU-B no no yes 749 30-31 reserved ig ig ig 751 Some NAL unit or payload type values (indicated as reserved in 752 Table 3) are reserved for future extensions. NAL units of those 753 types SHOULD NOT be sent by a sender (direct as packet payloads, or 754 as aggregation units in aggregation packets, or as fragmented units 755 in FU packets) and MUST be ignored by a receiver. For example, the 756 payload types 1-23, with the associated packet type "NAL unit", are 757 allowed in "Single NAL Unit Mode" and in "Non-Interleaved Mode", but 758 disallowed in "Interleaved Mode". However, NAL units of NAL unit 759 types 1-23 can be used in "Interleaved Mode" as aggregation units in 760 STAP-B, MTAP16 and MTAP14 packets as well as fragmented units in FU-A 761 and FU-B packets. Similarly, NAL units of NAL unit types 1-23 can 762 also be used in the "Non-Interleaved Mode" as aggregation units in 763 STAP-A packets or fragmented units in FU-A packets, in addition to 764 being directly used as packet payloads. 766 5.5. Decoding Order Number (DON) 768 In the interleaved packetization mode, the transmission order of NAL 769 units is allowed to differ from the decoding order of the NAL units. 770 Decoding order number (DON) is a field in the payload structure or a 771 derived variable that indicates the NAL unit decoding order. 773 Rationale and examples of use cases for transmission out of decoding 774 order and for the use of DON are given in section 13. 776 The coupling of transmission and decoding order is controlled by the 777 OPTIONAL sprop-interleaving-depth media type parameter as follows. 778 When the value of the OPTIONAL sprop-interleaving-depth media type 779 parameter is equal to 0 (explicitly or per default), the transmission 780 order of NAL units MUST conform to the NAL unit decoding order. When 781 the value of the OPTIONAL sprop-interleaving-depth media type 782 parameter is greater than 0, 784 o the order of NAL units in an MTAP16 and an MTAP24 is NOT REQUIRED 785 to be the NAL unit decoding order, and 787 o the order of NAL units generated by de-packetizing STAP-Bs, MTAPs, 788 and FUs in two consecutive packets is NOT REQUIRED to be the NAL 789 unit decoding order. 791 The RTP payload structures for a single NAL unit packet, an STAP-A, 792 and an FU-A do not include DON. STAP-B and FU-B structures include 793 DON, and the structure of MTAPs enables derivation of DON as 794 specified in section 5.7.2. 796 Informative note: When an FU-A occurs in interleaved mode, it 797 always follows an FU-B, which sets its DON. 799 Informative note: If a transmitter wants to encapsulate a single 800 NAL unit per packet and transmit packets out of their decoding 801 order, STAP-B packet type can be used. 803 In the single NAL unit packetization mode, the transmission order of 804 NAL units, determined by the RTP sequence number, MUST be the same as 805 their NAL unit decoding order. In the non-interleaved packetization 806 mode, the transmission order of NAL units in single NAL unit packets, 807 STAP-As, and FU-As MUST be the same as their NAL unit decoding order. 808 The NAL units within an STAP MUST appear in the NAL unit decoding 809 order. Thus, the decoding order is first provided through the 810 implicit order within a STAP, and second provided through the RTP 811 sequence number for the order between STAPs, FUs, and single NAL unit 812 packets. 814 Signaling of the value of DON for NAL units carried in STAP-B, MTAP, 815 and a series of fragmentation units starting with an FU-B is 816 specified in sections 5.7.1, 5.7.2, and 5.8, respectively. The DON 817 value of the first NAL unit in transmission order MAY be set to any 818 value. Values of DON are in the range of 0 to 65535, inclusive. 819 After reaching the maximum value, the value of DON wraps around to 0. 821 The decoding order of two NAL units contained in any STAP-B, MTAP, or 822 a series of fragmentation units starting with an FU-B is determined 823 as follows. Let DON(i) be the decoding order number of the NAL unit 824 having index i in the transmission order. Function don_diff(m,n) is 825 specified as follows: 827 If DON(m) == DON(n), don_diff(m,n) = 0 829 If (DON(m) < DON(n) and DON(n) - DON(m) < 32768), 830 don_diff(m,n) = DON(n) - DON(m) 832 If (DON(m) > DON(n) and DON(m) - DON(n) >= 32768), 833 don_diff(m,n) = 65536 - DON(m) + DON(n) 835 If (DON(m) < DON(n) and DON(n) - DON(m) >= 32768), 836 don_diff(m,n) = - (DON(m) + 65536 - DON(n)) 838 If (DON(m) > DON(n) and DON(m) - DON(n) < 32768), 839 don_diff(m,n) = - (DON(m) - DON(n)) 841 A positive value of don_diff(m,n) indicates that the NAL unit having 842 transmission order index n follows, in decoding order, the NAL unit 843 having transmission order index m. When don_diff(m,n) is equal to 0, 844 then the NAL unit decoding order of the two NAL units can be in 845 either order. A negative value of don_diff(m,n) indicates that the 846 NAL unit having transmission order index n precedes, in decoding 847 order, the NAL unit having transmission order index m. 849 Values of DON related fields (DON, DONB, and DOND; see section 5.7) 850 MUST be such that the decoding order determined by the values of DON, 851 as specified above, conforms to the NAL unit decoding order. If the 852 order of two NAL units in NAL unit decoding order is switched and the 853 new order does not conform to the NAL unit decoding order, the NAL 854 units MUST NOT have the same value of DON. If the order of two 855 consecutive NAL units in the NAL unit stream is switched and the new 856 order still conforms to the NAL unit decoding order, the NAL units 857 MAY have the same value of DON. For example, when arbitrary slice 858 order is allowed by the video coding profile in use, all the coded 859 slice NAL units of a coded picture are allowed to have the same value 860 of DON. Consequently, NAL units having the same value of DON can be 861 decoded in any order, and two NAL units having a different value of 862 DON should be passed to the decoder in the order specified above. 863 When two consecutive NAL units in the NAL unit decoding order have a 864 different value of DON, the value of DON for the second NAL unit in 865 decoding order SHOULD be the value of DON for the first, incremented 866 by one. 868 An example of the de-packetization process to recover the NAL unit 869 decoding order is given in section 7. 871 Informative note: Receivers should not expect that the absolute 872 difference of values of DON for two consecutive NAL units in the 873 NAL unit decoding order will be equal to one, even in error-free 874 transmission. An increment by one is not required, as at the 875 time of associating values of DON to NAL units, it may not be 876 known whether all NAL units are delivered to the receiver. For 877 example, a gateway may not forward coded slice NAL units of non- 878 reference pictures or SEI NAL units when there is a shortage of 879 bit rate in the network to which the packets are forwarded. In 880 another example, a live broadcast is interrupted by pre-encoded 881 content, such as commercials, from time to time. The first intra 882 picture of a pre-encoded clip is transmitted in advance to ensure 883 that it is readily available in the receiver. When transmitting 884 the first intra picture, the originator does not exactly know how 885 many NAL units will be encoded before the first intra picture of 886 the pre-encoded clip follows in decoding order. Thus, the values 887 of DON for the NAL units of the first intra picture of the pre- 888 encoded clip have to be estimated when they are transmitted, and 889 gaps in values of DON may occur. 891 5.6. Single NAL Unit Packet 893 The single NAL unit packet defined here MUST contain only one NAL 894 unit, of the types defined in [1]. This means that neither an 895 aggregation packet nor a fragmentation unit can be used within a 896 single NAL unit packet. A NAL unit stream composed by de-packetizing 897 single NAL unit packets in RTP sequence number order MUST conform to 898 the NAL unit decoding order. The structure of the single NAL unit 899 packet is shown in Figure 2. 901 Informative note: The first byte of a NAL unit co-serves as the 902 RTP payload header. 904 0 1 2 3 905 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 1 906 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 907 |F|NRI| Type | | 908 +-+-+-+-+-+-+-+-+ | 909 | | 910 | Bytes 2..n of a Single NAL unit | 911 | | 912 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 913 | :...OPTIONAL RTP padding | 914 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 916 Figure 2 RTP payload format for single NAL unit packet 918 5.7. Aggregation Packets 920 Aggregation packets are the NAL unit aggregation scheme of this 921 payload specification. The scheme is introduced to reflect the 922 dramatically different MTU sizes of two key target networks: wireline 923 IP networks (with an MTU size that is often limited by the Ethernet 924 MTU size; roughly 1500 bytes), and IP or non-IP (e.g., ITU-T H.324/M) 925 based wireless communication systems with preferred transmission unit 926 sizes of 254 bytes or less. To prevent media transcoding between the 927 two worlds, and to avoid undesirable packetization overhead, a NAL 928 unit aggregation scheme is introduced. 930 Two types of aggregation packets are defined by this specification: 932 o Single-time aggregation packet (STAP): aggregates NAL units with 933 identical NALU-time. Two types of STAPs are defined, one without 934 DON (STAP-A) and another including DON (STAP-B). 936 o Multi-time aggregation packet (MTAP): aggregates NAL units with 937 potentially differing NALU-time. Two different MTAPs are defined, 938 differing in the length of the NAL unit timestamp offset. 940 Each NAL unit to be carried in an aggregation packet is encapsulated 941 in an aggregation unit. Please see below for the four different 942 aggregation units and their characteristics. 944 The structure of the RTP payload format for aggregation packets is 945 presented in Figure 3. 947 0 1 2 3 948 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 1 949 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 950 |F|NRI| Type | | 951 +-+-+-+-+-+-+-+-+ | 952 | | 953 | one or more aggregation units | 954 | | 955 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 956 | :...OPTIONAL RTP padding | 957 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 959 Figure 3 RTP payload format for aggregation packets 961 MTAPs and STAPs share the following packetization rules: The RTP 962 timestamp MUST be set to the earliest of the NALU-times of all the 963 NAL units to be aggregated. The type field of the NAL unit type 964 octet MUST be set to the appropriate value, as indicated in Table 4. 965 The F bit MUST be cleared if all F bits of the aggregated NAL units 966 are zero; otherwise, it MUST be set. The value of NRI MUST be the 967 maximum of all the NAL units carried in the aggregation packet. 969 Table 4. Type field for STAPs and MTAPs 971 Type Packet Timestamp offset DON related fields 972 field length (DON, DONB, DOND) 973 (in bits) present 974 -------------------------------------------------------- 975 24 STAP-A 0 no 976 25 STAP-B 0 yes 977 26 MTAP16 16 yes 978 27 MTAP24 24 yes 980 The marker bit in the RTP header is set to the value that the marker 981 bit of the last NAL unit of the aggregated packet would have if it 982 were transported in its own RTP packet. 984 The payload of an aggregation packet consists of one or more 985 aggregation units. See sections 5.7.1 and 5.7.2 for the four 986 different types of aggregation units. An aggregation packet can 987 carry as many aggregation units as necessary; however, the total 988 amount of data in an aggregation packet obviously MUST fit into an IP 989 packet, and the size SHOULD be chosen so that the resulting IP packet 990 is smaller than the MTU size. An aggregation packet MUST NOT contain 991 fragmentation units specified in section 5.8. Aggregation packets 992 MUST NOT be nested; i.e., an aggregation packet MUST NOT contain 993 another aggregation packet. 995 5.7.1. Single-Time Aggregation Packet 997 Single-time aggregation packet (STAP) SHOULD be used whenever NAL 998 units are aggregated that all share the same NALU-time. The payload 999 of an STAP-A does not include DON and consists of at least one 1000 single-time aggregation unit, as presented in Figure 4. The payload 1001 of an STAP-B consists of a 16-bit unsigned decoding order number 1002 (DON) (in network byte order) followed by at least one single-time 1003 aggregation unit, as presented in Figure 5. 1005 0 1 2 3 1006 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 1007 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1008 : | 1009 +-+-+-+-+-+-+-+-+ | 1010 | | 1011 | single-time aggregation units | 1012 | | 1013 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1014 | : 1015 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1017 Figure 4 Payload format for STAP-A 1019 0 1 2 3 1020 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 1021 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1022 : decoding order number (DON) | | 1023 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1024 | | 1025 | single-time aggregation units | 1026 | | 1027 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1028 | : 1029 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1031 Figure 5 Payload format for STAP-B 1033 The DON field specifies the value of DON for the first NAL unit in an 1034 STAP-B in transmission order. For each successive NAL unit in 1035 appearance order in an STAP-B, the value of DON is equal to (the 1036 value of DON of the previous NAL unit in the STAP-B + 1) % 65536, in 1037 which '%' stands for the modulo operation. 1039 A single-time aggregation unit consists of 16-bit unsigned size 1040 information (in network byte order) that indicates the size of the 1041 following NAL unit in bytes (excluding these two octets, but 1042 including the NAL unit type octet of the NAL unit), followed by the 1043 NAL unit itself, including its NAL unit type byte. A single-time 1044 aggregation unit is byte aligned within the RTP payload, but it may 1045 not be aligned on a 32-bit word boundary. Figure 6 presents the 1046 structure of the single-time aggregation unit. 1048 0 1 2 3 1049 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 1050 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1051 : NAL unit size | | 1052 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1053 | | 1054 | NAL unit | 1055 | | 1056 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1057 | : 1058 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1060 Figure 6 Structure for single-time aggregation unit 1062 Figure 7 presents an example of an RTP packet that contains an STAP- 1063 A. The STAP contains two single-time aggregation units, labeled as 1 1064 and 2 in the figure. 1066 0 1 2 3 1067 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 1068 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1069 | RTP Header | 1070 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1071 |STAP-A NAL HDR | NALU 1 Size | NALU 1 HDR | 1072 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1073 | NALU 1 Data | 1074 : : 1075 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1076 | | NALU 2 Size | NALU 2 HDR | 1077 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1078 | NALU 2 Data | 1079 : : 1080 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1081 | :...OPTIONAL RTP padding | 1082 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1084 Figure 7 An example of an RTP packet including an STAP-A containing 1085 two single-time aggregation units 1087 Figure 8 presents an example of an RTP packet that contains an STAP- 1088 B. The STAP contains two single-time aggregation units, labeled as 1 1089 and 2 in the figure. 1091 0 1 2 3 1092 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 1093 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1094 | RTP Header | 1095 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1096 |STAP-B NAL HDR | DON | NALU 1 Size | 1097 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1098 | NALU 1 Size | NALU 1 HDR | NALU 1 Data | 1099 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + 1100 : : 1101 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1102 | | NALU 2 Size | NALU 2 HDR | 1103 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1104 | NALU 2 Data | 1105 : : 1106 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1107 | :...OPTIONAL RTP padding | 1108 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1110 Figure 8 An example of an RTP packet including an STAP-B containing 1111 two single-time aggregation units 1113 5.7.2. Multi-Time Aggregation Packets (MTAPs) 1115 The NAL unit payload of MTAPs consists of a 16-bit unsigned decoding 1116 order number base (DONB) (in network byte order) and one or more 1117 multi-time aggregation units, as presented in Figure 9. DONB MUST 1118 contain the value of DON for the first NAL unit in the NAL unit 1119 decoding order among the NAL units of the MTAP. 1121 Informative note: The first NAL unit in the NAL unit decoding 1122 order is not necessarily the first NAL unit in the order in which 1123 the NAL units are encapsulated in an MTAP. 1125 0 1 2 3 1126 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 1127 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1128 : decoding order number base | | 1129 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1130 | | 1131 | multi-time aggregation units | 1132 | | 1133 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1134 | : 1135 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1137 Figure 9 NAL unit payload format for MTAPs 1139 Two different multi-time aggregation units are defined in this 1140 specification. Both of them consist of 16 bits unsigned size 1141 information of the following NAL unit (in network byte order), an 8- 1142 bit unsigned decoding order number difference (DOND), and n bits (in 1143 network byte order) of timestamp offset (TS offset) for this NAL 1144 unit, whereby n can be 16 or 24. The choice between the different 1145 MTAP types (MTAP16 and MTAP24) is application dependent: the larger 1146 the timestamp offset is, the higher the flexibility of the MTAP, but 1147 the overhead is also higher. 1149 The structure of the multi-time aggregation units for MTAP16 and 1150 MTAP24 are presented in Figures 10 and 11, respectively. The 1151 starting or ending position of an aggregation unit within a packet is 1152 NOT REQUIRED to be on a 32-bit word boundary. The DON of the NAL 1153 unit contained in a multi-time aggregation unit is equal to (DONB + 1154 DOND) % 65536, in which % denotes the modulo operation. This memo 1155 does not specify how the NAL units within an MTAP are ordered, but, 1156 in most cases, NAL unit decoding order SHOULD be used. 1158 The timestamp offset field MUST be set to a value equal to the value 1159 of the following formula: If the NALU-time is larger than or equal to 1160 the RTP timestamp of the packet, then the timestamp offset equals 1161 (the NALU-time of the NAL unit - the RTP timestamp of the packet). 1162 If the NALU-time is smaller than the RTP timestamp of the packet, 1163 then the timestamp offset is equal to the NALU-time + (2^32 - the RTP 1164 timestamp of the packet). 1166 0 1 2 3 1167 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 1168 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1169 : NAL unit size | DOND | TS offset | 1170 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1171 | TS offset | | 1172 +-+-+-+-+-+-+-+-+ NAL unit | 1173 | | 1174 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1175 | : 1176 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1178 Figure 10 Multi-time aggregation unit for MTAP16 1180 0 1 2 3 1181 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 1182 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1183 : NAL unit size | DOND | TS offset | 1184 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1185 | TS offset | | 1186 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1187 | NAL unit | 1188 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1189 | : 1190 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1192 Figure 11 Multi-time aggregation unit for MTAP24 1194 For the "earliest" multi-time aggregation unit in an MTAP the 1195 timestamp offset MUST be zero. Hence, the RTP timestamp of the MTAP 1196 itself is identical to the earliest NALU-time. 1198 Informative note: The "earliest" multi-time aggregation unit is 1199 the one that would have the smallest extended RTP timestamp among 1200 all the aggregation units of an MTAP if the NAL units contained 1201 in the aggregation units were encapsulated in single NAL unit 1202 packets. An extended timestamp is a timestamp that has more than 1203 32 bits and is capable of counting the wraparound of the 1204 timestamp field, thus enabling one to determine the smallest 1205 value if the timestamp wraps. Such an "earliest" aggregation 1206 unit may not be the first one in the order in which the 1207 aggregation units are encapsulated in an MTAP. The "earliest" 1208 NAL unit need not be the same as the first NAL unit in the NAL 1209 unit decoding order either. 1211 Figure 12 presents an example of an RTP packet that contains a multi- 1212 time aggregation packet of type MTAP16 that contains two multi-time 1213 aggregation units, labeled as 1 and 2 in the figure. 1215 0 1 2 3 1216 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 1217 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1218 | RTP Header | 1219 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1220 |MTAP16 NAL HDR | decoding order number base | NALU 1 Size | 1221 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1222 | NALU 1 Size | NALU 1 DOND | NALU 1 TS offset | 1223 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1224 | NALU 1 HDR | NALU 1 DATA | 1225 +-+-+-+-+-+-+-+-+ + 1226 : : 1227 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1228 | | NALU 2 SIZE | NALU 2 DOND | 1229 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1230 | NALU 2 TS offset | NALU 2 HDR | NALU 2 DATA | 1231 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1232 : : 1233 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1234 | :...OPTIONAL RTP padding | 1235 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1237 Figure 12 An RTP packet including a multi-time aggregation packet of 1238 type MTAP16 containing two multi-time aggregation units 1240 Figure 13 presents an example of an RTP packet that contains a multi- 1241 time aggregation packet of type MTAP24 that contains two multi-time 1242 aggregation units, labeled as 1 and 2 in the figure. 1244 0 1 2 3 1245 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 1246 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1247 | RTP Header | 1248 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1249 |MTAP24 NAL HDR | decoding order number base | NALU 1 Size | 1250 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1251 | NALU 1 Size | NALU 1 DOND | NALU 1 TS offs | 1252 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1253 |NALU 1 TS offs | NALU 1 HDR | NALU 1 DATA | 1254 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ + 1255 : : 1256 + +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1257 | | NALU 2 SIZE | NALU 2 DOND | 1258 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1259 | NALU 2 TS offset | NALU 2 HDR | 1260 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1261 | NALU 2 DATA | 1262 : : 1263 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1264 | :...OPTIONAL RTP padding | 1265 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1267 Figure 13 An RTP packet including a multi-time aggregation packet of 1268 type MTAP24 containing two multi-time aggregation units 1270 5.7.3. Fragmentation Units (FUs) 1272 This payload type allows fragmenting a NAL unit into several RTP 1273 packets. Doing so on the application layer instead of relying on 1274 lower layer fragmentation (e.g., by IP) has the following advantages: 1276 o The payload format is capable of transporting NAL units bigger 1277 than 64 kbytes over an IPv4 network that may be present in pre- 1278 recorded video, particularly in High Definition formats (there is 1279 a limit of the number of slices per picture, which results in a 1280 limit of NAL units per picture, which may result in big NAL 1281 units). 1283 o The fragmentation mechanism allows fragmenting a single NAL unit 1284 and applying generic forward error correction as described in 1285 section 12.5. 1287 Fragmentation is defined only for a single NAL unit and not for any 1288 aggregation packets. A fragment of a NAL unit consists of an integer 1289 number of consecutive octets of that NAL unit. Each octet of the NAL 1290 unit MUST be part of exactly one fragment of that NAL unit. 1292 Fragments of the same NAL unit MUST be sent in consecutive order with 1293 ascending RTP sequence numbers (with no other RTP packets within the 1294 same RTP packet stream being sent between the first and last 1295 fragment). Similarly, a NAL unit MUST be reassembled in RTP sequence 1296 number order. 1298 When a NAL unit is fragmented and conveyed within fragmentation units 1299 (FUs), it is referred to as a fragmented NAL unit. STAPs and MTAPs 1300 MUST NOT be fragmented. FUs MUST NOT be nested; i.e., an FU MUST NOT 1301 contain another FU. 1303 The RTP timestamp of an RTP packet carrying an FU is set to the NALU- 1304 time of the fragmented NAL unit. 1306 Figure 14 presents the RTP payload format for FU-As. An FU-A 1307 consists of a fragmentation unit indicator of one octet, a 1308 fragmentation unit header of one octet, and a fragmentation unit 1309 payload. 1311 0 1 2 3 1312 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 1313 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1314 | FU indicator | FU header | | 1315 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | 1316 | | 1317 | FU payload | 1318 | | 1319 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1320 | :...OPTIONAL RTP padding | 1321 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1323 Figure 14 RTP payload format for FU-A 1325 Figure 15 presents the RTP payload format for FU-Bs. An FU-B 1326 consists of a fragmentation unit indicator of one octet, a 1327 fragmentation unit header of one octet, a decoding order number (DON) 1328 (in network byte order), and a fragmentation unit payload. In other 1329 words, the structure of FU-B is the same as the structure of FU-A, 1330 except for the additional DON field. 1332 0 1 2 3 1333 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 1334 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1335 | FU indicator | FU header | DON | 1336 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-| 1337 | | 1338 | FU payload | 1339 | | 1340 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1341 | :...OPTIONAL RTP padding | 1342 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 1344 Figure 15 RTP payload format for FU-B 1346 NAL unit type FU-B MUST be used in the interleaved packetization mode 1347 for the first fragmentation unit of a fragmented NAL unit. NAL unit 1348 type FU-B MUST NOT be used in any other case. In other words, in the 1349 interleaved packetization mode, each NALU that is fragmented has an 1350 FU-B as the first fragment, followed by one or more FU-A fragments. 1352 The FU indicator octet has the following format: 1354 +---------------+ 1355 |0|1|2|3|4|5|6|7| 1356 +-+-+-+-+-+-+-+-+ 1357 |F|NRI| Type | 1358 +---------------+ 1360 Values equal to 28 and 29 in the Type field of the FU indicator octet 1361 identify an FU-A and an FU-B, respectively. The use of the F bit is 1362 described in section 5.3. The value of the NRI field MUST be set 1363 according to the value of the NRI field in the fragmented NAL unit. 1365 The FU header has the following format: 1367 +---------------+ 1368 |0|1|2|3|4|5|6|7| 1369 +-+-+-+-+-+-+-+-+ 1370 |S|E|R| Type | 1371 +---------------+ 1373 S: 1 bit 1374 When set to one, the Start bit indicates the start of a 1375 fragmented NAL unit. When the following FU payload is not the 1376 start of a fragmented NAL unit payload, the Start bit is set to 1377 zero. 1379 E: 1 bit 1380 When set to one, the End bit indicates the end of a fragmented 1381 NAL unit, i.e., the last byte of the payload is also the last 1382 byte of the fragmented NAL unit. When the following FU payload 1383 is not the last fragment of a fragmented NAL unit, the End bit is 1384 set to zero. 1386 R: 1 bit 1387 The Reserved bit MUST be equal to 0 and MUST be ignored by the 1388 receiver. 1390 Type: 5 bits 1391 The NAL unit payload type as defined in Table 7-1 of [1]. 1393 The value of DON in FU-Bs is selected as described in section 5.5. 1395 Informative note: The DON field in FU-Bs allows gateways to 1396 fragment NAL units to FU-Bs without organizing the incoming NAL 1397 units to the NAL unit decoding order. 1399 A fragmented NAL unit MUST NOT be transmitted in one FU; i.e., the 1400 Start bit and End bit MUST NOT both be set to one in the same FU 1401 header. 1403 The FU payload consists of fragments of the payload of the fragmented 1404 NAL unit so that if the fragmentation unit payloads of consecutive 1405 FUs are sequentially concatenated, the payload of the fragmented NAL 1406 unit can be reconstructed. The NAL unit type octet of the fragmented 1407 NAL unit is not included as such in the fragmentation unit payload, 1408 but rather the information of the NAL unit type octet of the 1409 fragmented NAL unit is conveyed in F and NRI fields of the FU 1410 indicator octet of the fragmentation unit and in the type field of 1411 the FU header. An FU payload MAY have any number of octets and MAY 1412 be empty. 1414 Informative note: Empty FUs are allowed to reduce the latency of 1415 a certain class of senders in nearly lossless environments. 1416 These senders can be characterized in that they packetize NALU 1417 fragments before the NALU is completely generated and, hence, 1418 before the NALU size is known. If zero-length NALU fragments 1419 were not allowed, the sender would have to generate at least one 1420 bit of data of the following fragment before the current fragment 1421 could be sent. Due to the characteristics of H.264, where 1422 sometimes several macroblocks occupy zero bits, this is 1423 undesirable and can add delay. However, the (potential) use of 1424 zero-length NALU fragments should be carefully weighed against 1425 the increased risk of the loss of at least a part of the NALU 1426 because of the additional packets employed for its transmission. 1428 If a fragmentation unit is lost, the receiver SHOULD discard all 1429 following fragmentation units in transmission order corresponding to 1430 the same fragmented NAL unit. 1432 A receiver in an endpoint or in a MANE MAY aggregate the first n-1 1433 fragments of a NAL unit to an (incomplete) NAL unit, even if fragment 1434 n of that NAL unit is not received. In this case, the 1435 forbidden_zero_bit of the NAL unit MUST be set to one to indicate a 1436 syntax violation. 1438 6. Packetization Rules 1440 The packetization modes are introduced in section 5.2. The 1441 packetization rules common to more than one of the packetization 1442 modes are specified in section 6.1. The packetization rules for the 1443 single NAL unit mode, the non-interleaved mode, and the interleaved 1444 mode are specified in sections 6.2, 6.3, and 6.4, respectively. 1446 6.1. Common Packetization Rules 1448 All senders MUST enforce the following packetization rules regardless 1449 of the packetization mode in use: 1451 o Coded slice NAL units or coded slice data partition NAL units 1452 belonging to the same coded picture (and thus sharing the same RTP 1453 timestamp value) MAY be sent in any order; however, for delay- 1454 critical systems, they SHOULD be sent in their original decoding 1455 order to minimize the delay. Note that the decoding order is the 1456 order of the NAL units in the bitstream. 1458 o Parameter sets are handled in accordance with the rules and 1459 recommendations given in section 8.4. 1461 o MANEs MUST NOT duplicate any NAL unit except for sequence or 1462 picture parameter set NAL units, as neither this memo nor the 1463 H.264 specification provides means to identify duplicated NAL 1464 units. Sequence and picture parameter set NAL units MAY be 1465 duplicated to make their correct reception more probable, but any 1466 such duplication MUST NOT affect the contents of any active 1467 sequence or picture parameter set. Duplication SHOULD be 1468 performed on the application layer and not by duplicating RTP 1469 packets (with identical sequence numbers). 1471 Senders using the non-interleaved mode and the interleaved mode MUST 1472 enforce the following packetization rule: 1474 o MANEs MAY convert single NAL unit packets into one aggregation 1475 packet, convert an aggregation packet into several single NAL unit 1476 packets, or mix both concepts, in an RTP translator. The RTP 1477 translator SHOULD take into account at least the following 1478 parameters: path MTU size, unequal protection mechanisms (e.g., 1479 through packet-based FEC according to RFC 2733 [18], especially 1480 for sequence and picture parameter set NAL units and coded slice 1481 data partition A NAL units), bearable latency of the system, and 1482 buffering capabilities of the receiver. 1484 Informative note: An RTP translator is required to handle RTCP 1485 as per RFC 3550. 1487 6.2. Single NAL Unit Mode 1489 This mode is in use when the value of the OPTIONAL packetization-mode 1490 media type parameter is equal to 0 or the packetization-mode is not 1491 present. All receivers MUST support this mode. It is primarily 1492 intended for low-delay applications that are compatible with systems 1493 using ITU-T Recommendation H.241 [3] (see section 12.1). Only single 1494 NAL unit packets MAY be used in this mode. STAPs, MTAPs, and FUs 1495 MUST NOT be used. The transmission order of single NAL unit packets 1496 MUST comply with the NAL unit decoding order. 1498 6.3. Non-Interleaved Mode 1500 This mode is in use when the value of the OPTIONAL packetization-mode 1501 media type parameter is equal to 1. This mode SHOULD be supported. 1502 It is primarily intended for low-delay applications. Only single NAL 1503 unit packets, STAP-As, and FU-As MAY be used in this mode. STAP-Bs, 1504 MTAPs, and FU-Bs MUST NOT be used. The transmission order of NAL 1505 units MUST comply with the NAL unit decoding order. 1507 6.4. Interleaved Mode 1509 This mode is in use when the value of the OPTIONAL packetization-mode 1510 media type parameter is equal to 2. Some receivers MAY support this 1511 mode. STAP-Bs, MTAPs, FU-As, and FU-Bs MAY be used. STAP-As and 1512 single NAL unit packets MUST NOT be used. The transmission order of 1513 packets and NAL units is constrained as specified in section 5.5. 1515 7. De-Packetization Process 1517 The de-packetization process is implementation dependent. Therefore, 1518 the following description should be seen as an example of a suitable 1519 implementation. Other schemes may be used as well as long as the 1520 output for the same input is the same as the process described below. 1521 The output is the same meaning that the number of NAL units and their 1522 order are both the identical. Optimizations relative to the 1523 described algorithms are likely possible. Section 7.1 presents the 1524 de-packetization process for the single NAL unit and non-interleaved 1525 packetization modes, whereas section 7.2 describes the process for 1526 the interleaved mode. Section 7.3 includes additional de- 1527 packetization guidelines for intelligent receivers. 1529 All normal RTP mechanisms related to buffer management apply. In 1530 particular, duplicated or outdated RTP packets (as indicated by the 1531 RTP sequences number and the RTP timestamp) are removed. To 1532 determine the exact time for decoding, factors such as a possible 1533 intentional delay to allow for proper inter-stream synchronization 1534 must be factored in. 1536 7.1. Single NAL Unit and Non-Interleaved Mode 1538 The receiver includes a receiver buffer to compensate for 1539 transmission delay jitter. The receiver stores incoming packets in 1540 reception order into the receiver buffer. Packets are de-packetized 1541 in RTP sequence number order. If a de-packetized packet is a single 1542 NAL unit packet, the NAL unit contained in the packet is passed 1543 directly to the decoder. If a de-packetized packet is an STAP-A, the 1544 NAL units contained in the packet are passed to the decoder in the 1545 order in which they are encapsulated in the packet. For all the FU-A 1546 packets containing fragments of a single NAL unit, the de-packetized 1547 fragments are concatenated in their sending order to recover the NAL 1548 unit, which is then passed to the decoder. 1550 Informative note: If the decoder supports Arbitrary Slice Order, 1551 coded slices of a picture can be passed to the decoder in any 1552 order regardless of their reception and transmission order. 1554 7.2. Interleaved Mode 1556 The general concept behind these de-packetization rules is to reorder 1557 NAL units from transmission order to the NAL unit decoding order. 1559 The receiver includes a receiver buffer, which is used to compensate 1560 for transmission delay jitter and to reorder NAL units from 1561 transmission order to the NAL unit decoding order. In this section, 1562 the receiver operation is described under the assumption that there 1563 is no transmission delay jitter. To make a difference from a 1564 practical receiver buffer that is also used for compensation of 1565 transmission delay jitter, the receiver buffer is here after called 1566 the de-interleaving buffer in this section. Receivers SHOULD also 1567 prepare for transmission delay jitter; i.e., either reserve separate 1568 buffers for transmission delay jitter buffering and de-interleaving 1569 buffering or use a receiver buffer for both transmission delay jitter 1570 and de-interleaving. Moreover, receivers SHOULD take transmission 1571 delay jitter into account in the buffering operation; e.g., by 1572 additional initial buffering before starting of decoding and 1573 playback. 1575 This section is organized as follows: subsection 7.2.1 presents how o 1576 calculate the size of the de-interleaving buffer. Subsection 7.2.2 1577 specifies the receiver process how to organize received NAL units to 1578 the NAL unit decoding order. 1580 7.2.1. Size of the De-interleaving Buffer 1582 When the SDP Offer/Answer model or any other capability exchange 1583 procedure is used in session setup, the properties of the received 1584 stream SHOULD be such that the receiver capabilities are not 1585 exceeded. In the SDP Offer/Answer model, the receiver can indicate 1586 its capabilities to allocate a de-interleaving buffer with the deint- 1587 buf-cap media type parameter. The sender indicates the requirement 1588 for the de-interleaving buffer size with the sprop-deint-buf-req 1589 media type parameter. 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. 1592 See section 8.1 for further information on deint-buf-cap and sprop- 1593 deint-buf-req media type parameters and section 8.2.2 for further 1594 information on their use in the SDP Offer/Answer model. 1596 When a declarative session description is used in session setup, the 1597 sprop-deint-buf-req media type parameter signals the requirement for 1598 the de-interleaving buffer size. It is therefore RECOMMENDED to set 1599 the de-interleaving buffer size, in terms of number of bytes, equal 1600 to or greater than the value of sprop-deint-buf-req media type 1601 parameter. 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 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 1673 decoder, the value of PDON is set to the value of DON for the last 1674 NAL unit 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 found, a 1683 gateway may decide not to send the corresponding coded slice data 1684 partitions B and C, as their information is meaningless for H.264 1685 decoders. In this way a MANE can reduce network load by 1686 discarding useless packets without parsing a complex bitstream. 1688 o Intelligent RTP receivers (e.g., in gateways) may identify lost 1689 FUs. If a lost FU is found, a gateway may decide not to send the 1690 following FUs of the same fragmented NAL unit, as their 1691 information is meaningless for H.264 decoders. In this way a MANE 1692 can reduce network load by discarding useless packets without 1693 parsing a complex bitstream. 1695 o Intelligent receivers having to discard packets or NALUs should 1696 first discard all packets/NALUs in which the value of the NRI 1697 field of the NAL unit type octet is equal to 0. This will 1698 minimize the impact on user experience and keep the reference 1699 pictures intact. If more packets have to be discarded, then 1700 packets with a numerically lower NRI value should be discarded 1701 before packets with a numerically higher NRI value. However, 1702 discarding any packets with an NRI bigger than 0 very likely leads 1703 to decoder drift and SHOULD be avoided. 1705 8. Payload Format Parameters 1707 This section specifies the parameters that MAY be used to select 1708 optional features of the payload format and certain features of the 1709 bitstream. The parameters are specified here as part of the media 1710 subtype registration for the ITU-T H.264 | ISO/IEC 14496-10 codec. A 1711 mapping of the parameters into the Session Description Protocol (SDP) 1712 [6] is also provided for applications that use SDP. Equivalent 1713 parameters could be defined elsewhere for use with control protocols 1714 that do not use SDP. 1716 Some parameters provide a receiver with the properties of the stream 1717 that will be sent. The names of all these parameters start with 1718 "sprop" for stream properties. Some of these "sprop" parameters are 1719 limited by other payload or codec configuration parameters. For 1720 example, the sprop-parameter-sets parameter is constrained by the 1721 profile-level-id parameter. The media sender selects all "sprop" 1722 parameters rather than the receiver. This uncommon characteristic of 1723 the "sprop" parameters may not be compatible with some signaling 1724 protocol concepts, in which case the use of these parameters SHOULD 1725 be avoided. 1727 8.1. Media Type Registration 1729 The media subtype for the ITU-T H.264 | ISO/IEC 14496-10 codec is 1730 allocated from the IETF tree. 1732 The receiver MUST ignore any unspecified parameter. 1734 Media Type name: video 1736 Media subtype name: H264 1738 Required parameters: none 1740 OPTIONAL parameters: 1742 profile-level-id: 1743 A base16 [7] (hexadecimal) representation of the following 1744 three bytes in the sequence parameter set NAL unit specified 1745 in [1]: 1) profile_idc, 2) a byte herein referred to as 1746 profile-iop, composed of the values of constraint_set0_flag, 1747 constraint_set1_flag,constraint_set2_flag, 1748 constraint_set3_flag, and reserved_zero_4bits in bit- 1749 significance order, starting from the most significant bit, 1750 and 3) level_idc. Note that reserved_zero_4bits is required 1751 to be equal to 0 in [1], but other values for it may be 1752 specified in the future by ITU-T or ISO/IEC. 1754 The profile-level-id parameter indicates the default sub- 1755 profile, i.e. the subset of coding tools that may have been 1756 used to generate the stream or the receiver supports, and the 1757 default level of the stream or the receiver supports. 1759 The default sub-profile is indicated collectively by the 1760 profile_idc byte and some fields in the profile-iop byte. 1761 Depending on the values of the fields in the profile-iop byte, 1762 the default sub-profile may be the same set of coding tools 1763 supported by one profile, or a common subset of coding tools 1764 of multiple profiles, as specified in subsection 7.4.2.1.1 of 1765 [1]. The default level is indicated by the level_idc byte, 1766 and, when profile_idc is equal to 66, 77 or 88 (the Baseline, 1767 Main, or Extended profile) and level_idc is equal to 11, 1768 additionally by bit 4 (constraint_set3_flag) of the profile- 1769 iop byte. When profile_idc is equal to 66, 77 or 88 (the 1770 Baseline, Main, or Extended profile) and level_idc is equal to 1771 11, and bit 4 (constraint_set3_flag) of the profile-iop byte 1772 is equal to 1, the default level is level 1b. 1774 Table 5 lists all profiles defined in Annex A of [1] and, for 1775 each of the profiles, the possible combinations of profile_idc 1776 and profile-iop that represent the same sub-profile. 1778 Table 5. Combinations of profile_idc and profile-iop 1779 representing the same sub-profile corresponding to the full 1780 set of coding tools supported by one profile. In the 1781 following, x may be either 0 or 1, while the profile names 1782 are indicated as follows. CB: Constrained Baseline profile, 1783 B: Baseline profile, M: Main profile, E: Extended profile, 1784 H: High profile, H10: High 10 profile, H42: High 4:2:2 1785 profile, H44: High 4:4:4 Predictive profile, H10I: High 10 1786 Intra profile, H42I: High 4:2:2 Intra profile, H44I: High 1787 4:4:4 Intra profile, and C44I: CAVLC 4:4:4 Intra profile. 1789 Profile profile_idc profile-iop 1790 (hexadecimal) (binary) 1792 CB 42 (B) x1xx0000 1793 same as: 4D (M) 1xxx0000 1794 same as: 58 (E) 11xx0000 1795 same as: 64 (H), 6E (H10), 1xx00000 1796 7A (H42), or F4 (H44) 1797 B 42 (B) x0xx0000 1798 same as: 58 (E) 10xx0000 1799 M 4D (M) 0x0x0000 1800 same as: 64 (H), 6E (H10), 01000000 1801 7A (H42), or F4 (H44) 1802 E 58 00xx0000 1803 H 64 00000000 1804 H10 6E 00000000 1805 H42 7A 00000000 1806 H44 F4 00000000 1807 H10I 64 00010000 1808 H42I 7A 00010000 1809 H44I F4 00010000 1810 C44I 2C 00010000 1812 For example, in the table above, profile_idc equal to 58 1813 (Extended) with profile-iop equal to 11xx0000 indicates the 1814 same sub-profile corresponding to profile_idc equal to 42 1815 (Baseline) with profile-iop equal to x1xx0000. Note that 1816 other combinations of profile_idc and profile-iop (note listed 1817 in Table 5) may represent a sub-profile equivalent to the 1818 common subset of coding tools for more than one profile. Note 1819 also that a decoder conforming to a certain profile may be 1820 able to decode bitstreams conforming to other profiles. For 1821 example, a decoder conforming to the High 4:4:4 profile at 1822 certain level must be able to decode bitstreams confirming to 1823 the Constrained Baseline, Main, High, High 10 or High 4:2:2 1824 profile at the same or a lower level. 1826 If the profile-level-id parameter is used to indicate 1827 properties of a NAL unit stream, it indicates that, to decode 1828 the stream, the minimum subset of coding tools a decoder has 1829 to support is the default sub-profile, and the lowest level 1830 the decoder has to support is the default level. 1832 If the profile-level-id parameter is used for capability 1833 exchange or session setup procedure, it indicates the subset 1834 of coding tools, which is equal to the default sub-profile, 1835 and the highest level, which is equal to the default level, 1836 that the codec supports. All levels lower than the default 1837 level are also supported by the codec. 1839 Informative note: Capability exchange and session setup 1840 procedures should provide means to list the capabilities 1841 for each supported sub-profile separately. For example, 1842 the one-of-N codec selection procedure of the SDP 1843 Offer/Answer model can be used (section 10.2 of [8]). The 1844 one-of-N codec selection procedure may also be used to 1845 provide different combinations of profile_idc and profile- 1846 iop that represent the same sub-profile. When there are 1847 many different combinations of profile_idc and profile-iop 1848 that represent the same sub-profile, using the one-of-N 1849 codec selection procedure may result into a fairly large 1850 SDP message. Therefore, a receiver should understand the 1851 different equivalent combinations of profile_idc and 1852 profile-iop that represent the same sub-profile, and be 1853 ready to accept an offer using any of the equivalent 1854 combinations. 1856 If no profile-level-id is present, the Baseline Profile 1857 without additional constraints at Level 1 MUST be implied. 1859 max-mbps, max-smbps, max-fs, max-cpb, max-dpb, and max-br: 1860 These parameters MAY be used to signal the capabilities of a 1861 receiver implementation. These parameters MUST NOT be used for 1862 any other purpose. The profile-level-id parameter MUST be 1863 present in the same receiver capability description that 1864 contains any of these parameters. The level conveyed in the 1865 value of the profile-level-id parameter MUST be such that the 1866 receiver is fully capable of supporting. max-mbps, max-smbps, 1867 max-fs, max-cpb, max-dpb, and max-br MAY be used to indicate 1868 capabilities of the receiver that extend the required 1869 capabilities of the signaled level, as specified below. 1871 When more than one parameter from the set (max-mbps, max-smbps 1872 , max-fs, max-cpb, max-dpb, max-br) is present, the receiver 1873 MUST support all signaled capabilities simultaneously. For 1874 example, if both max-mbps and max-br are present, the signaled 1875 level with the extension of both the frame rate and bit rate 1876 is supported. That is, the receiver is able to decode NAL 1877 unit streams in which the macroblock processing rate is up to 1878 max-mbps (inclusive), the bit rate is up to max-br 1879 (inclusive), the coded picture buffer size is derived as 1880 specified in the semantics of the max-br parameter below, and 1881 other properties comply with the level specified in the value 1882 of the profile-level-id parameter. 1884 If a receiver can support all the properties of level A, the 1885 level specified in the value of the profile-level-id MUST be 1886 level A (i.e. MUST NOT be lower than level A). In other 1887 words, a sender or receiver MUST NOT signal values of max- 1888 mbps, max-fs, max-cpb, max-dpb, and max-br that meet the 1889 requirements of a higher level compared to the level specified 1890 in the value of the profile-level-id parameter. 1892 Informative note: When the OPTIONAL media type parameters 1893 are used to signal the properties of a NAL unit stream, 1894 max-mbps, max-smbps, max-fs, max-cpb, max-dpb, and max-br 1895 are not present, and the value of profile-level-id must 1896 always be such that the NAL unit stream complies fully with 1897 the specified profile and level. 1899 max-mbps: The value of max-mbps is an integer indicating the 1900 maximum macroblock processing rate in units of macroblocks per 1901 second. The max-mbps parameter signals that the receiver is 1902 capable of decoding video at a higher rate than is required by 1903 the signaled level conveyed in the value of the profile-level- 1904 id parameter. When max-mbps is signaled, the receiver MUST be 1905 able to decode NAL unit streams that conform to the signaled 1906 level, with the exception that the MaxMBPS value in Table A-1 1907 of [1] for the signaled level is replaced with the value of 1908 max-mbps. The value of max-mbps MUST be greater than or equal 1909 to the value of MaxMBPS for the level given in Table A-1 of 1910 [1]. Senders MAY use this knowledge to send pictures of a 1911 given size at a higher picture rate than is indicated in the 1912 signaled level. 1914 max-smbps: The value of max-smbps is an integer indicating the 1915 maximum static macroblock processing rate in units of static 1916 macroblocks per second, under the hypothetical assumption that 1917 all macroblocks are static macroblocks. When max-smbps is 1918 signalled the MaxMBPS value in Table A-1 of [1] should be 1919 replaced with the result of the following computation: 1921 o If the parameter max-mbps is signalled, set a variable 1922 MaxMacroblocksPerSecond to the value of max-mbps. 1923 Otherwise, set MaxMacroblocksPerSecond equal to the value 1924 of MaxMBPS for the level in Table A-1 [1]. 1926 o Set a variable P_non-static to the proportion of non-static 1927 macroblocks in picture n. 1929 o Set a variable P_static to the proportion of static 1930 macroblocks in picture n. 1932 o The value of MaxMBPS in Table A-1 of [1] should be 1933 considered by the encoder to be equal to: 1935 MaxMacroblocksPerSecond * max-smbps / ( P_non-static * max- 1936 smbps + P_static * MaxMacroblocksPerSecond) 1938 The encoder should recompute this value for each picture. The 1939 value of max-smbps MUST be greater than the value of MaxMBPS 1940 for the level given in Table A-1 of [1]. Senders MAY use this 1941 knowledge to send pictures of a given size at a higher picture 1942 rate than is indicated in the signalled level. 1944 max-fs: The value of max-fs is an integer indicating the maximum 1945 frame size in units of macroblocks. The max-fs parameter 1946 signals that the receiver is capable of decoding larger 1947 picture sizes than are required by the signaled level conveyed 1948 in the value of the profile-level-id parameter. When max-fs 1949 is signaled, the receiver MUST be able to decode NAL unit 1950 streams that conform to the signaled level, with the exception 1951 that the MaxFS value in Table A-1 of [1] for the signaled 1952 level is replaced with the value of max-fs. The value of max- 1953 fs MUST be greater than or equal to the value of MaxFS for the 1954 level given in Table A-1 of [1]. Senders MAY use this 1955 knowledge to send larger pictures at a proportionally lower 1956 frame rate than is indicated in the signaled level. 1958 max-cpb: The value of max-cpb is an integer indicating the 1959 maximum coded picture buffer size in units of 1000 bits for 1960 the VCL HRD parameters (see A.3.1 item i of [1]) and in units 1961 of 1200 bits for the NAL HRD parameters (see A.3.1 item j of 1962 [1]). The max-cpb parameter signals that the receiver has 1963 more memory than the minimum amount of coded picture buffer 1964 memory required by the signaled level conveyed in the value of 1965 the profile-level-id parameter. When max-cpb is signaled, the 1966 receiver MUST be able to decode NAL unit streams that conform 1967 to the signaled level, with the exception that the MaxCPB 1968 value in Table A-1 of [1] for the signaled level is replaced 1969 with the value of max-cpb. The value of max-cpb MUST be 1970 greater than or equal to the value of MaxCPB for the level 1971 given in Table A-1 of [1]. Senders MAY use this knowledge to 1972 construct coded video streams with greater variation of bit 1973 rate than can be achieved with the MaxCPB value in Table A-1 1974 of [1]. 1976 Informative note: The coded picture buffer is used in the 1977 hypothetical reference decoder (Annex C) of H.264. The use 1978 of the hypothetical reference decoder is recommended in 1979 H.264 encoders to verify that the produced bitstream 1980 conforms to the standard and to control the output bitrate. 1981 Thus, the coded picture buffer is conceptually independent 1982 of any other potential buffers in the receiver, including 1983 de-interleaving and de-jitter buffers. The coded picture 1984 buffer need not be implemented in decoders as specified in 1985 Annex C of H.264, but rather standard-compliant decoders 1986 can have any buffering arrangements provided that they can 1987 decode standard-compliant bitstreams. Thus, in practice, 1988 the input buffer for video decoder can be integrated with 1989 de-interleaving and de-jitter buffers of the receiver. 1991 max-dpb: The value of max-dpb is an integer indicating the 1992 maximum decoded picture buffer size in units of 1024 bytes. 1993 The max-dpb parameter signals that the receiver has more 1994 memory than the minimum amount of decoded picture buffer 1995 memory required by the signaled level conveyed in the value of 1996 the profile-level-id parameter. When max-dpb is signaled, the 1997 receiver MUST be able to decode NAL unit streams that conform 1998 to the signaled level, with the exception that the MaxDPB 1999 value in Table A-1 of [1] for the signaled level is replaced 2000 with the value of max-dpb. Consequently, a receiver that 2001 signals max-dpb MUST be capable of storing the following 2002 number of decoded frames, complementary field pairs, and non- 2003 paired fields in its decoded picture buffer: 2005 Min(1024 * max-dpb / ( PicWidthInMbs * FrameHeightInMbs * 2006 256 * ChromaFormatFactor ), 16) 2008 PicWidthInMbs, FrameHeightInMbs, and ChromaFormatFactor are 2009 defined in [1]. 2011 The value of max-dpb MUST be greater than or equal to the 2012 value of MaxDPB for the level given in Table A-1 of [1]. 2013 Senders MAY use this knowledge to construct coded video 2014 streams with improved compression. 2016 Informative note: This parameter was added primarily to 2017 complement a similar codepoint in the ITU-T Recommendation 2018 H.245, so as to facilitate signaling gateway designs. The 2019 decoded picture buffer stores reconstructed samples. There 2020 is no relationship between the size of the decoded picture 2021 buffer and the buffers used in RTP, especially de- 2022 interleaving and de-jitter buffers. 2024 max-br: The value of max-br is an integer indicating the maximum 2025 video bit rate in units of 1000 bits per second for the VCL 2026 HRD parameters (see A.3.1 item i of [1]) and in units of 1200 2027 bits per second for the NAL HRD parameters (see A.3.1 item j 2028 of [1]). 2030 The max-br parameter signals that the video decoder of the 2031 receiver is capable of decoding video at a higher bit rate 2032 than is required by the signaled level conveyed in the value 2033 of the profile-level-id parameter. 2035 When max-br is signaled, the video codec of the receiver MUST 2036 be able to decode NAL unit streams that conform to the 2037 signaled level, conveyed in the profile-level-id parameter, 2038 with the following exceptions in the limits specified by the 2039 level: 2041 o The value of max-br replaces the MaxBR value of the signaled 2042 level (in Table A-1 of [1]). 2044 o When the max-cpb parameter is not present, the result of the 2045 following formula replaces the value of MaxCPB in Table A-1 2046 of [1]: (MaxCPB of the signaled level) * max-br / (MaxBR of 2047 the signaled level). 2049 For example, if a receiver signals capability for Level 1.2 2050 with max-br equal to 1550, this indicates a maximum video 2051 bitrate of 1550 kbits/sec for VCL HRD parameters, a maximum 2052 video bitrate of 1860 kbits/sec for NAL HRD parameters, and a 2053 CPB size of 4036458 bits (1550000 / 384000 * 1000 * 1000). 2055 The value of max-br MUST be greater than or equal to the value 2056 MaxBR for the signaled level given in Table A-1 of [1]. 2058 Senders MAY use this knowledge to send higher bitrate video as 2059 allowed in the level definition of Annex A of H.264, to 2060 achieve improved video quality. 2062 Informative note: This parameter was added primarily to 2063 complement a similar codepoint in the ITU-T Recommendation 2064 H.245, so as to facilitate signaling gateway designs. No 2065 assumption can be made from the value of this parameter 2066 that the network is capable of handling such bit rates at 2067 any given time. In particular, no conclusion can be drawn 2068 that the signaled bit rate is possible under congestion 2069 control constraints. 2071 redundant-pic-cap: 2072 This parameter signals the capabilities of a receiver 2073 implementation. When equal to 0, the parameter indicates that 2074 the receiver makes no attempt to use redundant coded pictures 2075 to correct incorrectly decoded primary coded pictures. When 2076 equal to 0, the receiver is not capable of using redundant 2077 slices; therefore, a sender SHOULD avoid sending redundant 2078 slices to save bandwidth. When equal to 1, the receiver is 2079 capable of decoding any such redundant slice that covers a 2080 corrupted area in a primary decoded picture (at least partly), 2081 and therefore a sender MAY send redundant slices. When the 2082 parameter is not present, then a value of 0 MUST be used for 2083 redundant-pic-cap. When present, the value of redundant-pic- 2084 cap MUST be either 0 or 1. 2086 When the profile-level-id parameter is present in the same 2087 signaling as the redundant-pic-cap parameter, and the profile 2088 indicated in profile-level-id is such that it disallows the 2089 use of redundant coded pictures (e.g., Main Profile), the 2090 value of redundant-pic-cap MUST be equal to 0. When a 2091 receiver indicates redundant-pic-cap equal to 0, the received 2092 stream SHOULD NOT contain redundant coded pictures. 2094 Informative note: Even if redundant-pic-cap is equal to 0, 2095 the decoder is able to ignore redundant codec pictures 2096 provided that the decoder supports such a profile 2097 (Baseline, Extended) in which redundant coded pictures are 2098 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 packetization-mode: 2204 This parameter signals the properties of an RTP payload type 2205 or the capabilities of a receiver implementation. Only a 2206 single configuration point can be indicated; thus, when 2207 capabilities to support more than one packetization-mode are 2208 declared, multiple configuration points (RTP payload types) 2209 must be used. 2211 When the value of packetization-mode is equal to 0 or 2212 packetization-mode is not present, the single NAL mode, as 2213 defined in section 6.2 of RFC 3984, MUST be used. This mode 2214 is in use in standards using ITU-T Recommendation H.241 [3] 2215 (see section 12.1). When the value of packetization-mode is 2216 equal to 1, the non-interleaved mode, as defined in section 2217 6.3 of RFC 3984, MUST be used. When the value of 2218 packetization-mode is equal to 2, the interleaved mode, as 2219 defined in section 6.4 of RFC 3984, MUST be used. The value 2220 of packetization-mode MUST be an integer in the range of 0 to 2221 2, inclusive. 2223 sprop-interleaving-depth: 2224 This parameter MUST NOT be present when packetization-mode is 2225 not present or the value of packetization-mode is equal to 0 2226 or 1. This parameter MUST be present when the value of 2227 packetization-mode is equal to 2. 2229 This parameter signals the properties of an RTP packet stream. 2230 It specifies the maximum number of VCL NAL units that precede 2231 any VCL NAL unit in the RTP packet stream in transmission 2232 order and follow the VCL NAL unit in decoding order. 2233 Consequently, it is guaranteed that receivers can reconstruct 2234 NAL unit decoding order when the buffer size for NAL unit 2235 decoding order recovery is at least the value of sprop- 2236 interleaving-depth + 1 in terms of VCL NAL units. 2238 The value of sprop-interleaving-depth MUST be an integer in 2239 the range of 0 to 32767, inclusive. 2241 sprop-deint-buf-req: 2242 This parameter MUST NOT be present when packetization-mode is 2243 not present or the value of packetization-mode is equal to 0 2244 or 1. It MUST be present when the value of packetization-mode 2245 is equal to 2. 2247 sprop-deint-buf-req signals the required size of the de- 2248 interleaving buffer for the RTP packet stream. The value of 2249 the parameter MUST be greater than or equal to the maximum 2250 buffer occupancy (in units of bytes) required in such a de- 2251 interleaving buffer that is specified in section 7.2 of RFC 2252 3984. It is guaranteed that receivers can perform the de- 2253 interleaving of interleaved NAL units into NAL unit decoding 2254 order, when the de-interleaving buffer size is at least the 2255 value of sprop-deint-buf-req in terms of bytes. 2257 The value of sprop-deint-buf-req MUST be an integer in the 2258 range of 0 to 4294967295, inclusive. 2260 Informative note: sprop-deint-buf-req indicates the 2261 required size of the de-interleaving buffer only. When 2262 network jitter can occur, an appropriately sized jitter 2263 buffer has to be provisioned for as well. 2265 deint-buf-cap: 2266 This parameter signals the capabilities of a receiver 2267 implementation and indicates the amount of de-interleaving 2268 buffer space in units of bytes that the receiver has available 2269 for reconstructing the NAL unit decoding order. A receiver is 2270 able to handle any stream for which the value of the sprop- 2271 deint-buf-req parameter is smaller than or equal to this 2272 parameter. 2274 If the parameter is not present, then a value of 0 MUST be 2275 used for deint-buf-cap. The value of deint-buf-cap MUST be an 2276 integer in the range of 0 to 4294967295, inclusive. 2278 Informative note: deint-buf-cap indicates the maximum 2279 possible size of the de-interleaving buffer of the receiver 2280 only. When network jitter can occur, an appropriately 2281 sized jitter buffer has to be provisioned for as well. 2283 sprop-init-buf-time: 2284 This parameter MAY be used to signal the properties of an RTP 2285 packet stream. The parameter MUST NOT be present, if the 2286 value of packetization-mode is equal to 0 or 1. 2288 The parameter signals the initial buffering time that a 2289 receiver MUST wait before starting decoding to recover the NAL 2290 unit decoding order from the transmission order. The 2291 parameter is the maximum value of (decoding time of the NAL 2292 unit - transmission time of a NAL unit), assuming reliable and 2293 instantaneous transmission, the same timeline for transmission 2294 and decoding, and that decoding starts when the first packet 2295 arrives. 2297 An example of specifying the value of sprop-init-buf-time 2298 follows. A NAL unit stream is sent in the following 2299 interleaved order, in which the value corresponds to the 2300 decoding time and the transmission order is from left to 2301 right: 2303 0 2 1 3 5 4 6 8 7 ... 2305 Assuming a steady transmission rate of NAL units, the 2306 transmission times are: 2308 0 1 2 3 4 5 6 7 8 ... 2310 Subtracting the decoding time from the transmission time 2311 column-wise results in the following series: 2313 0 -1 1 0 -1 1 0 -1 1 ... 2315 Thus, in terms of intervals of NAL unit transmission times, 2316 the value of sprop-init-buf-time in this example is 1. The 2317 parameter is coded as a non-negative base10 integer 2318 representation in clock ticks of a 90-kHz clock. If the 2319 parameter is not present, then no initial buffering time value 2320 is defined. Otherwise the value of sprop-init-buf-time MUST 2321 be an integer in the range of 0 to 4294967295, inclusive. 2323 In addition to the signaled sprop-init-buf-time, receivers 2324 SHOULD take into account the transmission delay jitter 2325 buffering, including buffering for the delay jitter caused by 2326 mixers, translators, gateways, proxies, traffic-shapers, and 2327 other network elements. 2329 sprop-max-don-diff: 2330 This parameter MAY be used to signal the properties of an RTP 2331 packet stream. It MUST NOT be used to signal transmitter or 2332 receiver or codec capabilities. The parameter MUST NOT be 2333 present if the value of packetization-mode is equal to 0 or 1. 2334 sprop-max-don-diff is an integer in the range of 0 to 32767, 2335 inclusive. If sprop-max-don-diff is not present, the value of 2336 the parameter is unspecified. sprop-max-don-diff is 2337 calculated as follows: 2339 sprop-max-don-diff = max{AbsDON(i) - AbsDON(j)}, 2340 for any i and any j>i, 2342 where i and j indicate the index of the NAL unit in the 2343 transmission order and AbsDON denotes a decoding order number 2344 of the NAL unit that does not wrap around to 0 after 65535. 2345 In other words, AbsDON is calculated as follows: Let m and n 2346 be consecutive NAL units in transmission order. For the very 2347 first NAL unit in transmission order (whose index is 0), 2348 AbsDON(0) = DON(0). For other NAL units, AbsDON is calculated 2349 as follows: 2351 If DON(m) == DON(n), AbsDON(n) = AbsDON(m) 2353 If (DON(m) < DON(n) and DON(n) - DON(m) < 32768), 2354 AbsDON(n) = AbsDON(m) + DON(n) - DON(m) 2356 If (DON(m) > DON(n) and DON(m) - DON(n) >= 32768), 2357 AbsDON(n) = AbsDON(m) + 65536 - DON(m) + DON(n) 2359 If (DON(m) < DON(n) and DON(n) - DON(m) >= 32768), 2360 AbsDON(n) = AbsDON(m) - (DON(m) + 65536 - DON(n)) 2362 If (DON(m) > DON(n) and DON(m) - DON(n) < 32768), 2363 AbsDON(n) = AbsDON(m) - (DON(m) - DON(n)) 2365 where DON(i) is the decoding order number of the NAL unit 2366 having index i in the transmission order. The decoding order 2367 number is specified in section 5.5 of RFC 3984. 2369 Informative note: Receivers may use sprop-max-don-diff to 2370 trigger which NAL units in the receiver buffer can be 2371 passed to the decoder. 2373 max-rcmd-nalu-size: 2374 This parameter MAY be used to signal the capabilities of a 2375 receiver. The parameter MUST NOT be used for any other 2376 purposes. The value of the parameter indicates the largest 2377 NALU size in bytes that the receiver can handle efficiently. 2378 The parameter value is a recommendation, not a strict upper 2379 boundary. The sender MAY create larger NALUs but must be 2380 aware that the handling of these may come at a higher cost 2381 than NALUs conforming to the limitation. 2383 The value of max-rcmd-nalu-size MUST be an integer in the 2384 range of 0 to 4294967295, inclusive. If this parameter is not 2385 specified, no known limitation to the NALU size exists. 2386 Senders still have to consider the MTU size available between 2387 the sender and the receiver and SHOULD run MTU discovery for 2388 this purpose. 2390 This parameter is motivated by, for example, an IP to H.223 2391 video telephony gateway, where NALUs smaller than the H.223 2392 transport data unit will be more efficient. A gateway may 2393 terminate IP; thus, MTU discovery will normally not work 2394 beyond the gateway. 2396 Informative note: Setting this parameter to a lower than 2397 necessary value may have a negative impact. 2399 sar-understood: 2400 This parameter MAY be used to indicate a receiver capability 2401 and not anything else. The parameter indicates the maximum 2402 value of aspect_ratio_idc (specified in [1]) smaller than 255 2403 that the receiver understands. Table E-1 of [1] specifies 2404 aspect_ratio_idc equal to 0 as "unspecified", 1 to 16, 2405 inclusive, as specific Sample Aspect Ratios (SARs), 17 to 254, 2406 inclusive, as "reserved", and 255 as the Extended SAR, for 2407 which SAR width and SAR height are explicitly signaled. 2408 Therefore, a receiver with a decoder according to [1] 2409 understands aspect_ratio_idc in the range of 1 to 16, 2410 inclusive and aspect_ratio_idc equal to 255, in the sense that 2411 the receiver knows what exactly the SAR is. For such a 2412 receiver, the value of sar-understood is 16. If in the future 2413 Table E-1 of [1] is extended, e.g., such that the SAR for 2414 aspect_ratio_idc equal to 17 is specified, then for a receiver 2415 with a decoder that understands the extension, the value of 2416 sar-understood is 17. For a receiver with a decoder according 2417 to the 2003 version of [1], the value of sar-understood is 13, 2418 as the minimum reserved aspect_ratio_idc therein is 14. 2420 When sar-understood is not present, the value MUST be inferred 2421 to be equal to 13. 2423 sar-supported: 2424 This parameter MAY be used to indicate a receiver capability 2425 and not anything else. The value of this parameter is an 2426 integer in the range of 1 to sar-understood, inclusive, equal 2427 to 255. The value of sar-supported equal to N smaller than 2428 255 indicates that the reciever supports all the SARs 2429 corresponding to H.264 aspect_ratio_idc values (see Table E-1 2430 of [1]) in the range from 1 to N, inclusive, without geometric 2431 distortion. The value of sar-supported equal to 255 indicates 2432 that the receiver supports all sample aspect ratios which are 2433 expressible using two 16-bit integer values as the numerator 2434 and denominator, i.e., those that are expressible using the 2435 H.264 aspect_ratio_idc value of 255 (Extended_SAR, see Table 2436 E-1 of [1]), without geometric distortion. 2438 H.264 compliant encoders SHOULD NOT send an aspect_ratio_idc 2439 equal to 0, or an aspect_ratio_idc larger than sar-understood 2440 and smaller than 255. H.264 compliant encoders SHOULD send an 2441 aspect_ratio_idc that the receiver is able to display without 2442 geometrical distortion. However, H.264 compliant encoders MAY 2443 choose to send pictures using any SAR. 2445 Note that the actual sample aspect ratio or extended sample 2446 aspect ratio, when present, of the stream is conveyed in the 2447 Video Usability Information (VUI) part of the sequence 2448 parameter set. 2450 Encoding considerations: 2451 This type is only defined for transfer via RTP (RFC 3550). 2453 Security considerations: 2454 See section 9 of RFC xxxx. 2456 Public specification: 2457 Please refer to RFC xxxx and its section 15. 2459 Additional information: 2460 None 2462 File extensions: none 2464 Macintosh file type code: none 2466 Object identifier or OID: none 2468 Person & email address to contact for further information: 2469 Ye-Kui Wang, ye-kui.wang@nokia.com 2471 Intended usage: COMMON 2473 Author: 2474 Ye-Kui Wang, ye-kui.wang@nokia.com 2476 Change controller: 2477 IETF Audio/Video Transport working group delegated from the 2478 IESG. 2480 8.2. SDP Parameters 2482 8.2.1. Mapping of Payload Type Parameters to SDP 2484 The media type video/H264 string is mapped to fields in the Session 2485 Description Protocol (SDP) [6] as follows: 2487 o The media name in the "m=" line of SDP MUST be video. 2489 o The encoding name in the "a=rtpmap" line of SDP MUST be H264 (the 2490 media subtype). 2492 o The clock rate in the "a=rtpmap" line MUST be 90000. 2494 o The OPTIONAL parameters "profile-level-id", "max-mbps", "max- 2495 smbps", "max-fs", "max-cpb", "max-dpb", "max-br", "redundant-pic- 2496 cap", "use-level-src-parameter-sets", "packetization-mode", 2497 "sprop-interleaving-depth", "sprop-deint-buf-req", "deint-buf- 2498 cap", "sprop-init-buf-time", "sprop-max-don-diff", "max-rcmd-nalu- 2499 size", "sar-understood", and "sar-supported", when present, MUST 2500 be included in the "a=fmtp" line of SDP. These parameters are 2501 expressed as a media type string, in the form of a semicolon 2502 separated list of parameter=value pairs. 2504 o The OPTIONAL parameters "sprop-parameter-sets" and "sprop-level- 2505 parameter-sets", when present, MUST be included in the "a=fmtp" 2506 line of SDP or conveyed using the "fmtp" source attribute as 2507 specified in section 6.3 of [9]. For a particular media format 2508 (i.e., RTP payload type), a "sprop-parameter-sets" or "sprop- 2509 level-parameter-sets" MUST NOT be both included in the "a=fmtp" 2510 line of SDP and conveyed using the "fmtp" source attribute. When 2511 included in the "a=fmtp" line of SDP, these parameters are 2512 expressed as a media type string, in the form of a semicolon 2513 separated list of parameter=value pairs. When conveyed using the 2514 "fmtp" source attribute, these parameters are only associated with 2515 the given source and payload type as parts of the "fmtp" source 2516 attribute. 2518 Informative note: Conveyance of "sprop-parameter-sets" and 2519 "sprop-level-parameter-sets" using the "fmtp" source attribute 2520 allows for out-of-band transport of parameter sets in 2521 topologies like Topo-Video-switch-MCU [29]. 2523 An example of media representation in SDP is as follows (Baseline 2524 Profile, Level 3.0, some of the constraints of the Main profile may 2525 not be obeyed): 2527 m=video 49170 RTP/AVP 98 2528 a=rtpmap:98 H264/90000 2529 a=fmtp:98 profile-level-id=42A01E; 2530 packetization-mode=1; 2531 sprop-parameter-sets= 2533 8.2.2. Usage with the SDP Offer/Answer Model 2535 When H.264 is offered over RTP using SDP in an Offer/Answer model [8] 2536 for negotiation for unicast usage, the following limitations and 2537 rules apply: 2539 o The parameters identifying a media format configuration for H.264 2540 are "profile-level-id" and "packetization-mode", when present. 2541 These media format configuration parameters (except for the level 2542 part of "profile-level-id") MUST be used symmetrically; i.e., the 2543 answerer MUST either maintain all configuration parameters or 2544 remove the media format (payload type) completely, if one or more 2545 of the parameter values are not supported. Note that the level 2546 part of "profile-level-id" includes level_idc, and, for indication 2547 of level 1b when profile_idc is equal to 66, 77 or 88, bit 4 2548 (constraint_set3_flag) of profile-iop. The level part of 2549 "profile-level-id" is downgradable, i.e. the answerer MUST 2550 maintain the same or a lower level or remove the media format 2551 (payload type) completely. 2553 Informative note: The requirement for symmetric use applies 2554 only for the above media format configuration parameters 2555 excluding the level part of "profile-level-id", and not for 2556 the other stream properties and capability parameters. 2558 Informative note: In H.264 [1], all the levels except for 2559 level 1b are equal to the value of level_idc divided by 10. 2560 Level 1b is a level higher than level 1.0 but lower than level 2561 1.1, and is signaled in an ad-hoc manner, due to that the 2562 level was specified after level 1.0 and level 1.1. For the 2563 Baseline, Main and Extended profiles (with profile_idc equal 2564 to 66, 77 and 88, respectively), level 1b is indicated by 2565 level_idc equal to 11 (i.e. same as level 1.1) and 2566 constraint_set3_flag equal to 1. For other profiles, level 1b 2567 is indicated by level_idc equal to 9 (but note that level 1b 2568 for these profiles are still higher than level 1, which has 2569 level_idc equal to 10, and lower than level 1.1). In SDP 2570 Offer/Answer, an answer to an offer may indicate a level equal 2571 to or lower than the level indicated in the offer. Due to the 2572 ad-hoc indication of level 1b, offerers and answerers must 2573 check the value of bit 4 (constraint_set3_flag) of the middle 2574 octet of the parameter "profile-level-id", when profile_idc is 2575 equal to 66, 77 or 88 and level_idc is equal to 11. 2577 To simplify handling and matching of these configurations, the 2578 same RTP payload type number used in the offer SHOULD also be 2579 used in the answer, as specified in [8]. An answer MUST NOT 2580 contain a payload type number used in the offer unless the 2581 configuration is exactly the same as in the offer or the 2582 configuration in the answer only differs from that in the offer 2583 with a level lower than the default level offered. 2585 Informative note: When an offerer receives an answer, it has 2586 to compare payload types not declared in the offer based on 2587 the media type (i.e., video/H264) and the above media 2588 configuration parameters with any payload types it has already 2589 declared. This will enable it to determine whether the 2590 configuration in question is new or if it is equivalent to 2591 configuration already offered, since a different payload type 2592 number may be used in the answer. 2594 o The parameters "sprop-deint-buf-req", "sprop-interleaving-depth", 2595 "sprop-max-don-diff", and "sprop-init-buf-time" describe the 2596 properties of the RTP packet stream that the offerer or answerer 2597 is sending for the media format configuration. This differs from 2598 the normal usage of the Offer/Answer parameters: normally such 2599 parameters declare the properties of the stream that the offerer 2600 or the answerer is able to receive. When dealing with H.264, the 2601 offerer assumes that the answerer will be able to receive media 2602 encoded using the configuration being offered. 2604 Informative note: The above parameters apply for any stream 2605 sent by the declaring entity with the same configuration; 2606 i.e., they are dependent on their source. Rather than being 2607 bound to the payload type, the values may have to be applied 2608 to another payload type when being sent, as they apply for the 2609 configuration. 2611 o The capability parameters ("max-mbps", "max-smbps", "max-fs", 2612 "max-cpb", "max-dpb", "max-br", ,"redundant-pic-cap", "max-rcmd- 2613 nalu-size", "sar-understood", "sar-supported") MAY be used to 2614 declare further capabilities of the offerer or answerer for 2615 receiving. These parameters can only be present when the 2616 direction attribute is sendrecv or recvonly, and the parameters 2617 describe the limitations of what the offerer or answerer accepts 2618 for receiving streams. 2620 o An offerer has to include the size of the de-interleaving buffer, 2621 "sprop-deint-buf-req", in the offer for an interleaved H.264 2622 stream. To enable the offerer and answerer to inform each other 2623 about their capabilities for de-interleaving buffering in 2624 receiving streams, both parties are RECOMMENDED to include "deint- 2625 buf-cap". For interleaved streams, it is also RECOMMENDED to 2626 consider offering multiple payload types with different buffering 2627 requirements when the capabilities of the receiver are unknown. 2629 o The "sprop-parameter-sets" or "sprop-level-parameter-sets" 2630 parameter, when present (included in the "a=fmtp" line of SDP or 2631 conveyed using the "fmtp" source attribute as specified in section 2632 6.3 of [9]), is used for out-of-band transport of parameter sets. 2633 However, when out-of-band transport of parameter sets is used, 2634 parameter sets MAY still be additionally transported in-band. If 2635 neither "sprop-parameter-sets" nor "sprop-level-parameter-sets" is 2636 present, then only in-band transport of parameter sets is used. 2638 An offer MAY include either or both of "sprop-parameter-sets" and 2639 "sprop-level-parameter-sets". An answer MAY include "sprop- 2640 parameter-sets", and MUST NOT include "sprop-level-parameter- 2641 sets". 2643 When an offered payload type is accepted without level downgrade, 2644 i.e. the default level is accepted, the following applies. 2646 o When there is a "sprop-parameter-sets" included in the 2647 "a=fmtp" line of SDP, the answerer MUST be prepared to use 2648 the parameter sets included in "sprop-parameter-sets" for 2649 decoding the incoming NAL unit stream. 2651 o When there is a "sprop-parameter-sets" conveyed using the 2652 "fmtp" source attribute as specified in section 6.3 of [9], 2653 and the answerer understands the "fmtp" source attribute, it 2654 MUST be prepared to use the parameter sets included in 2655 "sprop-parameter-sets" for decoding the incoming NAL unit 2656 stream, and it MUST include either "use-level-src-parameter- 2657 sets" equal to 1 or the "fmtp" source attribute in the 2658 answer. 2660 o When there is a "sprop-parameter-sets" conveyed using the 2661 "fmtp" source attribute as specified in section 6.3 of [9], 2662 and the answerer does not understand the "fmtp" source 2663 attribute, in-band transport of parameter sets MUST be used, 2664 and the answerer MUST NOT include "use-level-src-parameter- 2665 sets" equal to 1 or the "fmtp" source attribute in the 2666 answer. 2668 o When "sprop-parameter-sets" is not present, in-band 2669 transport of parameter sets MUST be used, and the answer 2670 MUST NOT include "use-level-src-parameter-sets" equal to 1. 2672 o The answerer MUST ignore "sprop-level-parameter-sets", when 2673 present (either included in the "a=fmtp" line of SDP or 2674 conveyed using the "fmtp" source attribute). 2676 When level downgrade is in use, i.e., a level lower than the 2677 default level offered is accepted, the following applies. 2679 o The answerer MUST ignore "sprop-parameter-sets", when 2680 present (either included in the "a=fmtp" line of SDP or 2681 conveyed using the "fmtp" source attribute). 2683 o If "use-level-src-parameter-sets" equal to 1 the "fmtp" 2684 source attribute are not present in the answer for the 2685 accepted payload type, the answerer MUST ignore "sprop- 2686 level-parameter-sets", when present. 2688 o Otherwise ("use-level-src-parameter-sets" equal to 1 or the 2689 "fmtp" source attribute is present in the answer for the 2690 accepted payload type), the answerer MUST be prepared to use 2691 the parameter sets that are included in "sprop-level- 2692 parameter-sets" for the accepted level, when present, for 2693 decoding the incoming NAL unit stream, and ignore all other 2694 parameter sets included in "sprop-level-parameter-sets". 2696 o When no parameter sets for the accepted level are present in 2697 the "sprop-level-parameter-sets", in-band transport of 2698 parameter sets MUST be used. 2700 The answerer MAY or MAY not include "sprop-parameter-sets", i.e., 2701 the answerer MAY use either out-of-band or in-band transport of 2702 parameter sets for the stream it is sending, regardless of 2703 whether out-of-band parameter sets transport has been used in the 2704 offerer-to-answerer direction. All parameter sets included in 2705 the "sprop-parameter-sets", when present, for the accepted 2706 payload type in an answer MUST be associated with the accepted 2707 level, as indicated by the profile-level-id in the answer for the 2708 accepted payload type. 2710 Parameter sets included in "sprop-parameter-sets" in an answer 2711 are independent of those parameter sets included in the offer, as 2712 they are used for decoding two different video streams, one from 2713 the answerer to the offerer, and the other in the opposite 2714 direction. The offerer MUST be prepared to use the parameter 2715 sets included in the answer's "sprop-parameter-sets", when 2716 present, for decoding the incoming NAL unit stream. 2718 When "sprop-parameter-sets" or "sprop-level-parameter-sets" is 2719 conveyed using the "fmtp" source attribute in as specified in 2720 section 6.3 of [9], the receiver of the parameters MUST store the 2721 parameter sets included in the "sprop-parameter-sets" or "sprop- 2722 level-parameter-sets" for the accepted level and associate them 2723 to the source given as a part of the "fmtp" source attribute. 2724 Parameter sets associated with one source MUST only be used to 2725 decode NAL units conveyed in RTP packets from the same source. 2726 When this mechanism is in use, SSRC collision detection and 2727 resolution MUST be performed as specified in [9]. 2729 Informative note: Conveyance of "sprop-parameter-sets" and 2730 "sprop-level-parameter-sets" using the "fmtp" source attribute 2731 may be used in topologies like Topo-Video-switch-MCU [29] to 2732 enable out-of-band transport of parameter sets. 2734 For streams being delivered over multicast, the following rules 2735 apply: 2737 o The media format configuration is identified by the same 2738 parameters as above for unicast (i.e. "profile-level-id" and 2739 "packetization-mode", when present). These media format 2740 configuration parameters (including the level part of "profile- 2741 level-id") MUST be used symmetrically; i.e., the answerer MUST 2742 either maintain all configuration parameters or remove the media 2743 format (payload type) completely. Note that this implies that the 2744 level part of "profile-level-id" for Offer/Answer in multicast is 2745 not downgradable. 2747 To simplify handling and matching of these configurations, the 2748 same RTP payload type number used in the offer SHOULD also be 2749 used in the answer, as specified in [8]. An answer MUST NOT 2750 contain a payload type number used in the offer unless the 2751 configuration is the same as in the offer. 2753 o Parameter sets received MUST be associated with the originating 2754 source, and MUST be only used in decoding the incoming NAL unit 2755 stream from the same source. 2757 o The rules for other parameters are the same as above for unicast. 2759 Table 6 lists the interpretation of all the 20 media type parameters 2760 that MUST be used for the different direction attributes. 2762 Table 6. Interpretation of parameters for different direction 2763 attributes. 2765 sendonly --+ 2766 recvonly --+ | 2767 sendrecv --+ | | 2768 | | | 2769 profile-level-id C C P 2770 packetization-mode C C P 2771 sprop-deint-buf-req P - P 2772 sprop-interleaving-depth P - P 2773 sprop-max-don-diff P - P 2774 sprop-init-buf-time P - P 2775 max-mbps R R - 2776 max-smbps R R - 2777 max-fs R R - 2778 max-cpb R R - 2779 max-dpb R R - 2780 max-br R R - 2781 redundant-pic-cap R R - 2782 deint-buf-cap R R - 2783 max-rcmd-nalu-size R R - 2784 sar-understood R R - 2785 sar-supported R R - 2786 use-level-src-parameter-sets R R - 2787 sprop-parameter-sets S - S 2788 sprop-level-parameter-sets S - S 2790 Legend: 2792 C: configuration for sending and receiving streams 2793 P: properties of the stream to be sent 2794 R: receiver capabilities 2795 S: out-of-band parameter sets 2796 -: not usable, when present SHOULD be ignored 2798 Parameters used for declaring receiver capabilities are in general 2799 downgradable; i.e., they express the upper limit for a sender's 2800 possible behavior. Thus a sender MAY select to set its encoder using 2801 only lower/less or equal values of these parameters. 2803 Parameters declaring a configuration point are not downgradable, with 2804 the exception of the level part of the "profile-level-id" parameter 2805 for unicast usage. This expresses values a receiver expects to be 2806 used and must be used verbatim on the sender side. 2808 When a sender's capabilities are declared, and non-downgradable 2809 parameters are used in this declaration, then these parameters 2810 express a configuration that is acceptable for the sender to receive 2811 streams. In order to achieve high interoperability levels, it is 2812 often advisable to offer multiple alternative configurations; e.g., 2813 for the packetization mode. It is impossible to offer multiple 2814 configurations in a single payload type. Thus, when multiple 2815 configuration offers are made, each offer requires its own RTP 2816 payload type associated with the offer. 2818 A receiver SHOULD understand all media type parameters, even if it 2819 only supports a subset of the payload format's functionality. This 2820 ensures that a receiver is capable of understanding when an offer to 2821 receive media can be downgraded to what is supported by the receiver 2822 of the offer. 2824 An answerer MAY extend the offer with additional media format 2825 configurations. However, to enable their usage, in most cases a 2826 second offer is required from the offerer to provide the stream 2827 property parameters that the media sender will use. This also has 2828 the effect that the offerer has to be able to receive this media 2829 format configuration, not only to send it. 2831 If an offerer wishes to have non-symmetric capabilities between 2832 sending and receiving, the offerer should offer different RTP 2833 sessions; i.e., different media lines declared as "recvonly" and 2834 "sendonly", respectively. This may have further implications on the 2835 system. 2837 8.2.3. Usage in Declarative Session Descriptions 2839 When H.264 over RTP is offered with SDP in a declarative style, as in 2840 RTSP [27] or SAP [28], the following considerations are necessary. 2842 o All parameters capable of indicating both stream properties and 2843 receiver capabilities are used to indicate only stream properties. 2844 For example, in this case, the parameter "profile-level-id" 2845 declares only the values used by the stream, not the capabilities 2846 for receiving streams. This results in that the following 2847 interpretation of the parameters MUST be used: 2849 Declaring actual configuration or stream properties: 2851 - profile-level-id 2852 - packetization-mode 2853 - sprop-interleaving-depth 2854 - sprop-deint-buf-req 2855 - sprop-max-don-diff 2856 - sprop-init-buf-time 2858 Out-of-band transporting of parameter sets: 2860 - sprop-parameter-sets 2861 - sprop-level-parameter-sets 2863 Not usable(when present, they SHOULD be ignored): 2865 - max-mbps 2866 - max-smbps 2867 - max-fs 2868 - max-cpb 2869 - max-dpb 2870 - max-br 2871 - redundant-pic-cap 2872 - max-rcmd-nalu-size 2873 - deint-buf-cap 2874 - sar-understood 2875 - sar-supported 2876 - use-level-src-parameter-sets 2878 o A receiver of the SDP is required to support all parameters and 2879 values of the parameters provided; otherwise, the receiver MUST 2880 reject (RTSP) or not participate in (SAP) the session. It falls 2881 on the creator of the session to use values that are expected to 2882 be supported by the receiving application. 2884 8.3. Examples 2886 An SDP Offer/Answer exchange wherein both parties are expected to 2887 both send and receive could look like the following. Only the media 2888 codec specific parts of the SDP are shown. Some lines are wrapped 2889 due to text constraints. 2891 Offerer -> Answerer SDP message: 2893 m=video 49170 RTP/AVP 100 99 98 2894 a=rtpmap:98 H264/90000 2895 a=fmtp:98 profile-level-id=42A01E; packetization-mode=0; 2896 sprop-parameter-sets= 2897 a=rtpmap:99 H264/90000 2898 a=fmtp:99 profile-level-id=42A01E; packetization-mode=1; 2899 sprop-parameter-sets= 2900 a=rtpmap:100 H264/90000 2901 a=fmtp:100 profile-level-id=42A01E; packetization-mode=2; 2902 sprop-parameter-sets=; 2903 sprop-interleaving-depth=45; sprop-deint-buf-req=64000; 2904 sprop-init-buf-time=102478; deint-buf-cap=128000 2906 The above offer presents the same codec configuration in three 2907 different packetization formats. PT 98 represents single NALU mode, 2908 PT 99 represents non-interleaved mode, and PT 100 indicates the 2909 interleaved mode. In the interleaved mode case, the interleaving 2910 parameters that the offerer would use if the answer indicates support 2911 for PT 100 are also included. In all three cases the parameter 2912 "sprop-parameter-sets" conveys the initial parameter sets that are 2913 required by the answerer when receiving a stream from the offerer 2914 when this configuration is accepted. Note that the value for "sprop- 2915 parameter-sets" could be different for each payload type. 2917 Answerer -> Offerer SDP message: 2919 m=video 49170 RTP/AVP 100 99 97 2920 a=rtpmap:97 H264/90000 2921 a=fmtp:97 profile-level-id=42A01E; packetization-mode=0; 2922 sprop-parameter-sets= 2923 a=rtpmap:99 H264/90000 2924 a=fmtp:99 profile-level-id=42A01E; packetization-mode=1; 2925 sprop-parameter-sets=; 2926 max-rcmd-nalu-size=3980 2927 a=rtpmap:100 H264/90000 2928 a=fmtp:100 profile-level-id=42A01E; packetization-mode=2; 2929 sprop-parameter-sets=; 2930 sprop-interleaving-depth=60; 2931 sprop-deint-buf-req=86000; sprop-init-buf-time=156320; 2932 deint-buf-cap=128000; max-rcmd-nalu-size=3980 2934 As the Offer/Answer negotiation covers both sending and receiving 2935 streams, an offer indicates the exact parameters for what the offerer 2936 is willing to receive, whereas the answer indicates the same for what 2937 the answerer accepts to receive. In this case the offerer declared 2938 that it is willing to receive payload type 98. The answerer accepts 2939 this by declaring an equivalent payload type 97; i.e., it has 2940 identical values for the two parameters "profile-level-id" and 2941 "packetization-mode" (since "packetization-mode" is equal to 0, 2942 "sprop-deint-buf-req" is not present). As the offered payload type 2943 98 is accepted, the answerer needs to store parameter sets included 2944 in sprop-parameter-sets= in case the offer 2945 finally decides to use this configuration. In the answer, the 2946 answerer includes the parameter sets in sprop-parameter- 2947 sets= that the answerer would use in the 2948 stream sent from the answerer if this configuration is finally used. 2950 The answerer also accepts the reception of the two configurations 2951 that payload types 99 and 100 represent. Again, the answerer needs 2952 to store parameter sets included in sprop-parameter-sets= and sprop-parameter-sets= in case 2954 the offer finally decides to use either of these two configurations. 2955 The answerer provides the initial parameter sets for the answerer-to- 2956 offerer direction, i.e. the parameter sets in sprop-parameter- 2957 sets= and sprop-parameter-sets=, for payload types 99 and 100, respectively, that it will use 2959 to send the payload types. The answerer also provides the offerer 2960 with its memory limit for de-interleaving operations by providing a 2961 "deint-buf-cap" parameter. This is only useful if the offerer 2962 decides on making a second offer, where it can take the new value 2963 into account. The "max-rcmd-nalu-size" indicates that the answerer 2964 can efficiently process NALUs up to the size of 3980 bytes. However, 2965 there is no guarantee that the network supports this size. 2967 In the following example, the offer is accepted without level 2968 downgrading (i.e. the default level, 3.0, is accepted), and both 2969 "sprop-parameter-sets" and "sprop-level-parameter-sets" are present 2970 in the offer. The answerer must ignore sprop-level-parameter- 2971 sets= and store parameter sets in sprop- 2972 parameter-sets= for decoding the incoming NAL 2973 unit stream. The offerer must store the parameter sets in sprop- 2974 parameter-sets= in the answer for decoding the 2975 incoming NAL unit stream. Note that in this example, parameter sets 2976 in sprop-parameter-sets= must be associated 2977 with level 3.0. 2979 Offer SDP: 2981 m=video 49170 RTP/AVP 98 2982 a=rtpmap:98 H264/90000 2983 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 2984 packetization-mode=1; 2985 sprop-parameter-sets=; 2986 sprop-level-parameter-sets= 2988 Answer SDP: 2990 m=video 49170 RTP/AVP 98 2991 a=rtpmap:98 H264/90000 2992 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 2993 packetization-mode=1; 2994 sprop-parameter-sets= 2996 In the following example, the offer (Baseline profile, level 1.1) is 2997 accepted with level downgrading (the accepted level is 1b), and both 2998 "sprop-parameter-sets" and "sprop-level-parameter-sets" are present 2999 in the offer. The answerer must ignore sprop-parameter- 3000 sets= and all parameter sets not for the 3001 accepted level (level 1b) in sprop-level-parameter-sets=, and must store parameter sets for the accepted level 3003 (level 1b) in sprop-level-parameter-sets= for 3004 decoding the incoming NAL unit stream. The offerer must store the 3005 parameter sets in sprop-parameter-sets= in the 3006 answer for decoding the incoming NAL unit stream. Note that in this 3007 example, parameter sets in sprop-parameter-sets= must be associated with level 1b. 3010 Offer SDP: 3012 m=video 49170 RTP/AVP 98 3013 a=rtpmap:98 H264/90000 3014 a=fmtp:98 profile-level-id=42A00B; //Baseline profile, Level 1.1 3015 packetization-mode=1; 3016 sprop-parameter-sets=; 3017 sprop-level-parameter-sets= 3019 Answer SDP: 3021 m=video 49170 RTP/AVP 98 3022 a=rtpmap:98 H264/90000 3023 a=fmtp:98 profile-level-id=42B00B; //Baseline profile, Level 1b 3024 packetization-mode=1; 3025 sprop-parameter-sets=; 3026 use-level-src-parameter-sets=1 3028 In the following example, the offer (Baseline profile, level 1.1) is 3029 accepted with level downgrading (the accepted level is 1b), and both 3030 "sprop-parameter-sets" and "sprop-level-parameter-sets" are present 3031 in the offer. However, the answerer is a legacy RFC 3984 3032 implementation and does not understand "sprop-level-parameter-sets", 3033 hence it does not include "use-level-src-parameter-sets" (which the 3034 answerer does not understand, either) in the answer. Therefore, the 3035 answerer must ignore both sprop-parameter-sets= and sprop-level-parameter-sets=, and 3037 the offerer must transport parameter sets in-band. 3039 Offer SDP: 3041 m=video 49170 RTP/AVP 98 3042 a=rtpmap:98 H264/90000 3043 a=fmtp:98 profile-level-id=42A00B; //Baseline profile, Level 1.1 3044 packetization-mode=1; 3045 sprop-parameter-sets=; 3046 sprop-level-parameter-sets= 3048 Answer SDP: 3050 m=video 49170 RTP/AVP 98 3051 a=rtpmap:98 H264/90000 3052 a=fmtp:98 profile-level-id=42B00B; //Baseline profile, Level 1b 3053 packetization-mode=1 3055 In the following example, the offer is accepted without level 3056 downgrading, and "sprop-parameter-sets" is present in the offer. 3057 Parameter sets in sprop-parameter-sets= must 3058 be stored and used used by the encoder of the offerer and the decoder 3059 of the answerer, and parameter sets in sprop-parameter- 3060 sets=must be used by the encoder of the 3061 answerer and the decoder of the offerer. Note that sprop-parameter- 3062 sets= is basically independent of sprop- 3063 parameter-sets=. 3065 Offer SDP: 3067 m=video 49170 RTP/AVP 98 3068 a=rtpmap:98 H264/90000 3069 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3070 packetization-mode=1; 3071 sprop-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=42A01E; //Baseline profile, Level 3.0 3078 packetization-mode=1; 3079 sprop-parameter-sets= 3081 In the following example, the offer is accepted without level 3082 downgrading, and neither "sprop-parameter-sets" nor "sprop-level- 3083 parameter-sets" is present in the offer, meaning that there is no 3084 out-of-band transmission of parameter sets, which then have to be 3085 transported in-band. 3087 Offer SDP: 3089 m=video 49170 RTP/AVP 98 3090 a=rtpmap:98 H264/90000 3091 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3092 packetization-mode=1 3094 Answer SDP: 3096 m=video 49170 RTP/AVP 98 3097 a=rtpmap:98 H264/90000 3098 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3099 packetization-mode=1 3101 In the following example, the offer is accepted with level 3102 downgrading and "sprop-parameter-sets" is present in the offer. As 3103 sprop-parameter-sets= contains level_idc 3104 indicating Level 3.0, therefore cannot be used as the answerer wants 3105 Level 2.0 and must be ignored by the answerer, and in-band parameter 3106 sets must be used. 3108 Offer SDP: 3110 m=video 49170 RTP/AVP 98 3111 a=rtpmap:98 H264/90000 3112 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3113 packetization-mode=1; 3114 sprop-parameter-sets= 3116 Answer SDP: 3118 m=video 49170 RTP/AVP 98 3119 a=rtpmap:98 H264/90000 3120 a=fmtp:98 profile-level-id=42A014; //Baseline profile, Level 2.0 3121 packetization-mode=1 3123 In the following example, the offer is also accepted with level 3124 downgrading, and neither "sprop-parameter-sets" nor "sprop-level- 3125 parameter-sets" is present in the offer, meaning that there is no 3126 out-of-band transmission of parameter sets, which then have to be 3127 transported in-band. 3129 Offer SDP: 3131 m=video 49170 RTP/AVP 98 3132 a=rtpmap:98 H264/90000 3133 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3134 packetization-mode=1 3136 Answer SDP: 3138 m=video 49170 RTP/AVP 98 3139 a=rtpmap:98 H264/90000 3140 a=fmtp:98 profile-level-id=42A014; //Baseline profile, Level 2.0 3141 packetization-mode=1 3143 In the following example, the offerer is a Multipoint Control Unit 3144 (MCU) in a Topo-Video-switch-MCU like topology [29], offering 3145 parameter sets received (using out-of-band transport) from three 3146 other participants B, C, and D, and receiving parameter sets from the 3147 participant A, which is the answerer. The participants are 3148 identified by their values of CNAME, which are mapped to different 3149 SSRC values. The same codec configuration is used by all the four 3150 participants. The participant A stores and associates the parameter 3151 sets included in , , 3152 and to participants B, C, and D, 3153 respectively, and uses for decoding NAL units 3154 carried in RTP packets originated from participant B only, uses 3155 for decoding NAL units carried in RTP packets 3156 originated from participant C only, and uses 3157 for decoding NAL units carried in RTP packets originated from 3158 participant D only. 3160 Offer SDP: 3162 m=video 49170 RTP/AVP 98 3163 a=ssrc:SSRC-B cname:CNAME-B 3164 a=ssrc:SSRC-C cname:CNAME-C 3165 a=ssrc:SSRC-D cname:CNAME-D 3166 a=ssrc:SSRC-B fmtp:98 3167 sprop-parameter-sets= 3168 a=ssrc:SSRC-C fmtp:98 3169 sprop-parameter-sets= 3170 a=ssrc:SSRC-D fmtp:98 3171 sprop-parameter-sets= 3172 a=rtpmap:98 H264/90000 3173 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3174 packetization-mode=1 3176 Answer SDP: 3178 m=video 49170 RTP/AVP 98 3179 a=ssrc:SSRC-A cname:CNAME-A 3180 a=ssrc:SSRC-A fmtp:98 3181 sprop-parameter-sets= 3182 a=rtpmap:98 H264/90000 3183 a=fmtp:98 profile-level-id=42A01E; //Baseline profile, Level 3.0 3184 packetization-mode=1 3186 8.4. Parameter Set Considerations 3188 The H.264 parameter sets are a fundamental part of the video codec 3189 and vital to its operation; see section 1.2. Due to their 3190 characteristics and their importance for the decoding process, lost 3191 or erroneously transmitted parameter sets can hardly be concealed 3192 locally at the receiver. A reference to a corrupt parameter set has 3193 normally fatal results to the decoding process. Corruption could 3194 occur, for example, due to the erroneous transmission or loss of a 3195 parameter set NAL unit, but also due to the untimely transmission of 3196 a parameter set update. A parameter set update refers to a change of 3197 at least one parameter in a picture parameter set or sequence 3198 parameter set for which the picture parameter set or sequence 3199 parameter set identifier remains unchanged. Therefore, the following 3200 recommendations are provided as a guideline for the implementer of 3201 the RTP sender. 3203 Parameter set NALUs can be transported using three different 3204 principles: 3206 A. Using a session control protocol (out-of-band) prior to the actual 3207 RTP session. 3209 B. Using a session control protocol (out-of-band) during an ongoing 3210 RTP session. 3212 C. Within the RTP packet stream in the payload (in-band) during an 3213 ongoing RTP session. 3215 It is recommended to implement principles A and B within a session 3216 control protocol. SIP and SDP can be used as described in the SDP 3217 Offer/Answer model and in the previous sections of this memo. This 3218 section contains guidelines on how principles A and B should be 3219 implemented within session control protocols. It is independent of 3220 the particular protocol used. Principle C is supported by the RTP 3221 payload format defined in this specification. There are topologies 3222 like Topo-Video-switch-MCU [29] for which the use of principle C may 3223 be desirable. 3225 If in-band signaling of parameter sets is used, the picture and 3226 sequence parameter set NALUs SHOULD be transmitted in the RTP payload 3227 using a reliable method of delivering of RTP (see below), as a loss 3228 of a parameter set of either type will likely prevent decoding of a 3229 considerable portion of the corresponding RTP packet stream. 3231 If in-band signaling of parameter sets is used, the sender SHOULD 3232 take the error characteristics into account and use mechanisms to 3233 provide a high probability for delivering the parameter sets 3234 correctly. Mechanisms that increase the probability for a correct 3235 reception include packet repetition, FEC, and retransmission. The 3236 use of an unreliable, out-of-band control protocol has similar 3237 disadvantages as the in-band signaling (possible loss) and, in 3238 addition, may also lead to difficulties in the synchronization (see 3239 below). Therefore, it is NOT RECOMMENDED. 3241 Parameter sets MAY be added or updated during the lifetime of a 3242 session using principles B and C. It is required that parameter sets 3243 are present at the decoder prior to the NAL units that refer to them. 3244 Updating or adding of parameter sets can result in further problems, 3245 and therefore the following recommendations should be considered. 3247 - When parameter sets are added or updated, care SHOULD be taken to 3248 ensure that any parameter set is delivered prior to its usage. 3249 When new parameter sets are added, previously unused parameter set 3250 identifiers are used. It is common that no synchronization is 3251 present between out-of-band signaling and in-band traffic. If 3252 out-of-band signaling is used, it is RECOMMENDED that a sender 3253 does not start sending NALUs requiring the added or updated 3254 parameter sets prior to acknowledgement of delivery from the 3255 signaling protocol. 3257 - When parameter sets are updated, the following synchronization 3258 issue should be taken into account. When overwriting a parameter 3259 set at the receiver, the sender has to ensure that the parameter 3260 set in question is not needed by any NALU present in the network 3261 or receiver buffers. Otherwise, decoding with a wrong parameter 3262 set may occur. To lessen this problem, it is RECOMMENDED either 3263 to overwrite only those parameter sets that have not been used for 3264 a sufficiently long time (to ensure that all related NALUs have 3265 been consumed), or to add a new parameter set instead (which may 3266 have negative consequences for the efficiency of the video 3267 coding). 3269 Informative note: In some topologies like Topo-Video-switch- 3270 MCU [29] the origin of the whole set of parameter sets may 3271 come from multiple sources that may use non-unique parameter 3272 sets identifiers. In this case an offer may overwrite an 3273 existing parameter set if no other mechanism that enables 3274 uniqueness of the parameter sets in the out-of-band channel 3275 exists. 3277 - In a multiparty session, one participant MUST associate parameter 3278 sets coming from different sources with the source identification 3279 whenever possible, e.g. by conveying out-of-band transported 3280 parameter sets, as different sources typically use independent 3281 parameter set identifier value spaces. 3283 - Adding or modifying parameter sets by using both principles B and 3284 C in the same RTP session may lead to inconsistencies of the 3285 parameter sets because of the lack of synchronization between the 3286 control and the RTP channel. Therefore, principles B and C MUST 3287 NOT both be used in the same session unless sufficient 3288 synchronization can be provided. 3290 In some scenarios (e.g., when only the subset of this payload format 3291 specification corresponding to H.241 is used) or topologies, it is 3292 not possible to employ out-of-band parameter set transmission. In 3293 this case, parameter sets have to be transmitted in-band. Here, the 3294 synchronization with the non-parameter-set-data in the bitstream is 3295 implicit, but the possibility of a loss has to be taken into account. 3296 The loss probability should be reduced using the mechanisms discussed 3297 above. In case a loss of a parameter set is detected, recovery may 3298 be achieved by using a Decoder Refresh Point procedure, for example, 3299 using RTCP feedback Full Intra Request (FIR) [30]. Two example 3300 Decoder Refresh Point procedures are provided in the informative 3301 Section 8.5. 3303 - When parameter sets are initially provided using principle A and 3304 then later added or updated in-band (principle C), there is a risk 3305 associated with updating the parameter sets delivered out-of-band. 3306 If receivers miss some in-band updates (for example, because of a 3307 loss or a late tune-in), those receivers attempt to decode the 3308 bitstream using out-dated parameters. It is therefore RECOMMENDED 3309 that parameter set IDs be partitioned between the out-of-band and 3310 in-band parameter sets. 3312 8.5. Decoder Refresh Point Procedure using In-Band Transport of 3313 Parameter Sets (Informative) 3315 When a sender with a video encoder according to [1] receives a 3316 request for a decoder refresh point, the encoder shall enter the fast 3317 update mode by using one of the procedures specified in Section 8.5.1 3318 or 8.5.2 below. The procedure in 8.5.1 is the preferred response in 3319 a lossless transmission environment. Both procedures satisfy the 3320 requirement to enter the fast update mode for H.264 video encoding. 3322 8.5.1. IDR Procedure to Respond to a Request for a Decoder Refresh Point 3324 This section gives one possible way to respond to a request for a 3325 decoder refresh point. 3327 The encoder shall, in the order presented here: 3329 1) Immediately prepare to send an IDR picture. 3331 2) Send a sequence parameter set to be used by the IDR picture to be 3332 sent. The encoder may optionally also send other sequence 3333 parameter sets. 3335 3) Send a picture parameter set to be used by the IDR picture to be 3336 sent. The encoder may optionally also send other picture parameter 3337 sets. 3339 4) Send the IDR picture. 3341 5) From this point forward in time, send any other sequence or 3342 picture parameter sets that have not yet been sent in this 3343 procedure, prior to their reference by any NAL unit, regardless of 3344 whether such parameter sets were previously sent prior to 3345 receiving the request for a decoder refresh point. As needed, 3346 such parameter sets may be sent in a batch, one at a time, or in 3347 any combination of these two methods. Parameter sets may be re- 3348 sent at any time for redundancy. Caution should be taken when 3349 parameter set updates are present, as described above in Section 3350 8.4. 3352 8.5.2. Gradual Recovery Procedure to Respond to a Request for a Decoder 3353 Refresh Point 3355 This section gives another possible way to respond to a request for a 3356 decoder refresh point. 3358 The encoder shall, in the order presented here: 3360 1) Send a recovery point SEI message (see Sections D.1.7 and D.2.7 of 3361 [1]). 3363 2) Repeat any sequence and picture parameter sets that were sent 3364 before the recovery point SEI message, prior to their reference by 3365 a NAL unit. 3367 The encoder shall ensure that the decoder has access to all reference 3368 pictures for inter prediction of pictures at or after the recovery 3369 point, which is indicated by the recovery point SEI message, in 3370 output order, assuming that the transmission from now on is error- 3371 free. 3373 The value of the recovery_frame_cnt syntax element in the recovery 3374 point SEI message should be small enough to ensure a fast recovery. 3376 As needed, such parameter sets may be re-sent in a batch, one at a 3377 time, or in any combination of these two methods. Parameter sets may 3378 be re-sent at any time for redundancy. Caution should be taken when 3379 parameter set updates are present, as described above in Section 8.4. 3381 9. Security Considerations 3383 RTP packets using the payload format defined in this specification 3384 are subject to the security considerations discussed in the RTP 3385 specification [5], and in any appropriate RTP profile (for example, 3386 [16]). This implies that confidentiality of the media streams is 3387 achieved by encryption; for example, through the application of SRTP 3388 [26]. Because the data compression used with this payload format is 3389 applied end-to-end, any encryption needs to be performed after 3390 compression. A potential denial-of-service threat exists for data 3391 encodings using compression techniques that have non-uniform 3392 receiver-end computational load. The attacker can inject 3393 pathological datagrams into the stream that are complex to decode and 3394 that cause the receiver to be overloaded. H.264 is particularly 3395 vulnerable to such attacks, as it is extremely simple to generate 3396 datagrams containing NAL units that affect the decoding process of 3397 many future NAL units. Therefore, the usage of data origin 3398 authentication and data integrity protection of at least the RTP 3399 packet is RECOMMENDED; for example, with SRTP [26]. 3401 Note that the appropriate mechanism to ensure confidentiality and 3402 integrity of RTP packets and their payloads is very dependent on the 3403 application and on the transport and signaling protocols employed. 3404 Thus, although SRTP is given as an example above, other possible 3405 choices exist. 3407 Decoders MUST exercise caution with respect to the handling of user 3408 data SEI messages, particularly if they contain active elements, and 3409 MUST restrict their domain of applicability to the presentation 3410 containing the stream. 3412 End-to-End security with either authentication, integrity or 3413 confidentiality protection will prevent a MANE from performing media- 3414 aware operations other than discarding complete packets. And in the 3415 case of confidentiality protection it will even be prevented from 3416 performing discarding of packets in a media aware way. To allow any 3417 MANE to perform its operations, it will be required to be a trusted 3418 entity which is included in the security context establishment. 3420 10. Congestion Control 3422 Congestion control for RTP SHALL be used in accordance with RFC 3550 3423 [5], and with any applicable RTP profile; e.g., RFC 3551 [16]. An 3424 additional requirement if best-effort service is being used is: users 3425 of this payload format MUST monitor packet loss to ensure that the 3426 packet loss rate is within acceptable parameters. Packet loss is 3427 considered acceptable if a TCP flow across the same network path, and 3428 experiencing the same network conditions, would achieve an average 3429 throughput, measured on a reasonable timescale that is not less than 3430 the RTP flow is achieving. This condition can be satisfied by 3431 implementing congestion control mechanisms to adapt the transmission 3432 rate (or the number of layers subscribed for a layered multicast 3433 session), or by arranging for a receiver to leave the session if the 3434 loss rate is unacceptably high. 3436 The bit rate adaptation necessary for obeying the congestion control 3437 principle is easily achievable when real-time encoding is used. 3438 However, when pre-encoded content is being transmitted, bandwidth 3439 adaptation requires the availability of more than one coded 3440 representation of the same content, at different bit rates, or the 3441 existence of non-reference pictures or sub-sequences [22] in the 3442 bitstream. The switching between the different representations can 3443 normally be performed in the same RTP session; e.g., by employing a 3444 concept known as SI/SP slices of the Extended Profile, or by 3445 switching streams at IDR picture boundaries. Only when non- 3446 downgradable parameters (such as the profile part of the 3447 profile/level ID) are required to be changed does it become necessary 3448 to terminate and re-start the media stream. This may be accomplished 3449 by using a different RTP payload type. 3451 MANEs MAY follow the suggestions outlined in section 7.3 and remove 3452 certain unusable packets from the packet stream when that stream was 3453 damaged due to previous packet losses. This can help reduce the 3454 network load in certain special cases. 3456 11. IANA Consideration 3458 IANA has registered one new media type; see section 8.1. 3460 12. Informative Appendix: Application Examples 3462 This payload specification is very flexible in its use, in order to 3463 cover the extremely wide application space anticipated for H.264. 3464 However, this great flexibility also makes it difficult for an 3465 implementer to decide on a reasonable packetization scheme. Some 3466 information on how to apply this specification to real-world 3467 scenarios is likely to appear in the form of academic publications 3468 and a test model software and description in the near future. 3469 However, some preliminary usage scenarios are described here as well. 3471 12.1. Video Telephony according to ITU-T Recommendation H.241 Annex A 3473 H.323-based video telephony systems that use H.264 as an optional 3474 video compression scheme are required to support H.241 Annex A [3] as 3475 a packetization scheme. The packetization mechanism defined in this 3476 Annex is technically identical with a small subset of this 3477 specification. 3479 When a system operates according to H.241 Annex A, parameter set NAL 3480 units are sent in-band. Only Single NAL unit packets are used. Many 3481 such systems are not sending IDR pictures regularly, but only when 3482 required by user interaction or by control protocol means; e.g., when 3483 switching between video channels in a Multipoint Control Unit or for 3484 error recovery requested by feedback. 3486 12.2. Video Telephony, No Slice Data Partitioning, No NAL Unit 3487 Aggregation 3489 The RTP part of this scheme is implemented and tested (though not the 3490 control-protocol part; see below). 3492 In most real-world video telephony applications, picture parameters 3493 such as picture size or optional modes never change during the 3494 lifetime of a connection. Therefore, all necessary parameter sets 3495 (usually only one) are sent as a side effect of the capability 3496 exchange/announcement process, e.g., according to the SDP syntax 3497 specified in section 8.2 of this document. As all necessary 3498 parameter set information is established before the RTP session 3499 starts, there is no need for sending any parameter set NAL units. 3500 Slice data partitioning is not used, either. Thus, the RTP packet 3501 stream basically consists of NAL units that carry single coded 3502 slices. 3504 The encoder chooses the size of coded slice NAL units so that they 3505 offer the best performance. Often, this is done by adapting the 3506 coded slice size to the MTU size of the IP network. For small 3507 picture sizes, this may result in a one-picture-per-one-packet 3508 strategy. Intra refresh algorithms clean up the loss of packets and 3509 the resulting drift-related artifacts. 3511 12.3. Video Telephony, Interleaved Packetization Using NAL Unit 3512 Aggregation 3514 This scheme allows better error concealment and is used in H.263 3515 based designs using RFC 2429 packetization [11]. It has been 3516 implemented, and good results were reported [13]. 3518 The VCL encoder codes the source picture so that all macroblocks 3519 (MBs) of one MB line are assigned to one slice. All slices with even 3520 MB row addresses are combined into one STAP, and all slices with odd 3521 MB row addresses into another. Those STAPs are transmitted as RTP 3522 packets. The establishment of the parameter sets is performed as 3523 discussed above. 3525 Note that the use of STAPs is essential here, as the high number of 3526 individual slices (18 for a CIF picture) would lead to unacceptably 3527 high IP/UDP/RTP header overhead (unless the source coding tool FMO is 3528 used, which is not assumed in this scenario). Furthermore, some 3529 wireless video transmission systems, such as H.324M and the IP-based 3530 video telephony specified in 3GPP, are likely to use relatively small 3531 transport packet size. For example, a typical MTU size of H.223 AL3 3532 SDU is around 100 bytes [17]. Coding individual slices according to 3533 this packetization scheme provides further advantage in communication 3534 between wired and wireless networks, as individual slices are likely 3535 to be smaller than the preferred maximum packet size of wireless 3536 systems. Consequently, a gateway can convert the STAPs used in a 3537 wired network into several RTP packets with only one NAL unit, which 3538 are preferred in a wireless network, and vice versa. 3540 12.4. Video Telephony with Data Partitioning 3542 This scheme has been implemented and has been shown to offer good 3543 performance, especially at higher packet loss rates [13]. 3545 Data Partitioning is known to be useful only when some form of 3546 unequal error protection is available. Normally, in single-session 3547 RTP environments, even error characteristics are assumed; i.e., the 3548 packet loss probability of all packets of the session is the same 3549 statistically. However, there are means to reduce the packet loss 3550 probability of individual packets in an RTP session. A FEC packet 3551 according to RFC 2733 [18], for example, specifies which media 3552 packets are associated with the FEC packet. 3554 In all cases, the incurred overhead is substantial but is in the same 3555 order of magnitude as the number of bits that have otherwise been 3556 spent for intra information. However, this mechanism does not add 3557 any delay to the system. 3559 Again, the complete parameter set establishment is performed through 3560 control protocol means. 3562 12.5. Video Telephony or Streaming with FUs and Forward Error Correction 3564 This scheme has been implemented and has been shown to provide good 3565 performance, especially at higher packet loss rates [19]. 3567 The most efficient means to combat packet losses for scenarios where 3568 retransmissions are not applicable is forward error correction (FEC). 3569 Although application layer, end-to-end use of FEC is often less 3570 efficient than an FEC-based protection of individual links 3571 (especially when links of different characteristics are in the 3572 transmission path), application layer, end-to-end FEC is unavoidable 3573 in some scenarios. RFC 5109 [18] provides means to use generic, 3574 application layer, end-to-end FEC in packet-loss environments. A 3575 binary forward error correcting code is generated by applying the XOR 3576 operation to the bits at the same bit position in different packets. 3577 The binary code can be specified by the parameters (n,k) in which k 3578 is the number of information packets used in the connection and n is 3579 the total number of packets generated for k information packets; 3580 i.e., n-k parity packets are generated for k information packets. 3582 When a code is used with parameters (n,k) within the RFC 5109 3583 framework, the following properties are well known: 3585 a) If applied over one RTP packet, RFC 5109 provides only packet 3586 repetition. 3588 b) RFC 5109 is most bit rate efficient if XOR-connected packets have 3589 equal length. 3591 c) At the same packet loss probability p and for a fixed k, the 3592 greater the value of n is, the smaller the residual error 3593 probability becomes. For example, for a packet loss probability 3594 of 10%, k=1, and n=2, the residual error probability is about 1%, 3595 whereas for n=3, the residual error probability is about 0.1%. 3597 d) At the same packet loss probability p and for a fixed code rate 3598 k/n, the greater the value of n is, the smaller the residual error 3599 probability becomes. For example, at a packet loss probability of 3600 p=10%, k=1 and n=2, the residual error rate is about 1%, whereas 3601 for an extended Golay code with k=12 and n=24, the residual error 3602 rate is about 0.01%. 3604 For applying RFC 5109 in combination with H.264 baseline coded video 3605 without using FUs, several options might be considered: 3607 1) The video encoder produces NAL units for which each video frame is 3608 coded in a single slice. Applying FEC, one could use a simple 3609 code; e.g., (n=2, k=1). That is, each NAL unit would basically 3610 just be repeated. The disadvantage is obviously the bad code 3611 performance according to d), above, and the low flexibility, as 3612 only (n, k=1) codes can be used. 3614 2) The video encoder produces NAL units for which each video frame is 3615 encoded in one or more consecutive slices. Applying FEC, one 3616 could use a better code, e.g., (n=24, k=12), over a sequence of 3617 NAL units. Depending on the number of RTP packets per frame, a 3618 loss may introduce a significant delay, which is reduced when more 3619 RTP packets are used per frame. Packets of completely different 3620 length might also be connected, which decreases bit rate 3621 efficiency according to b), above. However, with some care and 3622 for slices of 1kb or larger, similar length (100-200 bytes 3623 difference) may be produced, which will not lower the bit 3624 efficiency catastrophically. 3626 3) The video encoder produces NAL units, for which a certain frame 3627 contains k slices of possibly almost equal length. Then, applying 3628 FEC, a better code, e.g., (n=24, k=12), can be used over the 3629 sequence of NAL units for each frame. The delay compared to that 3630 of 2), above, may be reduced, but several disadvantages are 3631 obvious. First, the coding efficiency of the encoded video is 3632 lowered significantly, as slice-structured coding reduces intra- 3633 frame prediction and additional slice overhead is necessary. 3634 Second, pre-encoded content or, when operating over a gateway, the 3635 video is usually not appropriately coded with k slices such that 3636 FEC can be applied. Finally, the encoding of video producing k 3637 slices of equal length is not straightforward and might require 3638 more than one encoding pass. 3640 Many of the mentioned disadvantages can be avoided by applying FUs in 3641 combination with FEC. Each NAL unit can be split into any number of 3642 FUs of basically equal length; therefore, FEC with a reasonable k and 3643 n can be applied, even if the encoder made no effort to produce 3644 slices of equal length. For example, a coded slice NAL unit 3645 containing an entire frame can be split to k FUs, and a parity check 3646 code (n=k+1, k) can be applied. However, this has the disadvantage 3647 that unless all created fragments can be recovered, the whole slice 3648 will be lost. Thus a larger section is lost than would be if the 3649 frame had been split into several slices. 3651 The presented technique makes it possible to achieve good 3652 transmission error tolerance, even if no additional source coding 3653 layer redundancy (such as periodic intra frames) is present. 3654 Consequently, the same coded video sequence can be used to achieve 3655 the maximum compression efficiency and quality over error-free 3656 transmission and for transmission over error-prone networks. 3657 Furthermore, the technique allows the application of FEC to pre- 3658 encoded sequences without adding delay. In this case, pre-encoded 3659 sequences that are not encoded for error-prone networks can still be 3660 transmitted almost reliably without adding extensive delays. In 3661 addition, FUs of equal length result in a bit rate efficient use of 3662 RFC 5109. 3664 If the error probability depends on the length of the transmitted 3665 packet (e.g., in case of mobile transmission [15]), the benefits of 3666 applying FUs with FEC are even more obvious. Basically, the 3667 flexibility of the size of FUs allows appropriate FEC to be applied 3668 for each NAL unit and unequal error protection of NAL units. 3670 When FUs and FEC are used, the incurred overhead is substantial but 3671 is in the same order of magnitude as the number of bits that have to 3672 be spent for intra-coded macroblocks if no FEC is applied. In [19], 3673 it was shown that the overall performance of the FEC-based approach 3674 enhanced quality when using the same error rate and same overall bit 3675 rate, including the overhead. 3677 12.6. Low Bit-Rate Streaming 3679 This scheme has been implemented with H.263 and non-standard RTP 3680 packetization and has given good results [20]. There is no technical 3681 reason why similarly good results could not be achievable with H.264. 3683 In today's Internet streaming, some of the offered bit rates are 3684 relatively low in order to allow terminals with dial-up modems to 3685 access the content. In wired IP networks, relatively large packets, 3686 say 500 - 1500 bytes, are preferred to smaller and more frequently 3687 occurring packets in order to reduce network congestion. Moreover, 3688 use of large packets decreases the amount of RTP/UDP/IP header 3689 overhead. For low bit-rate video, the use of large packets means 3690 that sometimes up to few pictures should be encapsulated in one 3691 packet. 3693 However, loss of a packet including many coded pictures would have 3694 drastic consequences for visual quality, as there is practically no 3695 other way to conceal a loss of an entire picture than to repeat the 3696 previous one. One way to construct relatively large packets and 3697 maintain possibilities for successful loss concealment is to 3698 construct MTAPs that contain interleaved slices from several 3699 pictures. An MTAP should not contain spatially adjacent slices from 3700 the same picture or spatially overlapping slices from any picture. 3701 If a packet is lost, it is likely that a lost slice is surrounded by 3702 spatially adjacent slices of the same picture and spatially 3703 corresponding slices of the temporally previous and succeeding 3704 pictures. Consequently, concealment of the lost slice is likely to 3705 be relatively successful. 3707 12.7. Robust Packet Scheduling in Video Streaming 3709 Robust packet scheduling has been implemented with MPEG-4 Part 2 and 3710 simulated in a wireless streaming environment [21]. There is no 3711 technical reason why similar or better results could not be 3712 achievable with H.264. 3714 Streaming clients typically have a receiver buffer that is capable of 3715 storing a relatively large amount of data. Initially, when a 3716 streaming session is established, a client does not start playing the 3717 stream back immediately. Rather, it typically buffers the incoming 3718 data for a few seconds. This buffering helps maintain continuous 3719 playback, as, in case of occasional increased transmission delays or 3720 network throughput drops, the client can decode and play buffered 3721 data. Otherwise, without initial buffering, the client has to freeze 3722 the display, stop decoding, and wait for incoming data. The 3723 buffering is also necessary for either automatic or selective 3724 retransmission in any protocol level. If any part of a picture is 3725 lost, a retransmission mechanism may be used to resend the lost data. 3726 If the retransmitted data is received before its scheduled decoding 3727 or playback time, the loss is recovered perfectly. Coded pictures 3728 can be ranked according to their importance in the subjective quality 3729 of the decoded sequence. For example, non-reference pictures, such 3730 as conventional B pictures, are subjectively least important, as 3731 their absence does not affect decoding of any other pictures. In 3732 addition to non-reference pictures, the ITU-T H.264 | ISO/IEC 14496- 3733 10 standard includes a temporal scalability method called sub- 3734 sequences [22]. Subjective ranking can also be made on coded slice 3735 data partition or slice group basis. Coded slices and coded slice 3736 data partitions that are subjectively the most important can be sent 3737 earlier than their decoding order indicates, whereas coded slices and 3738 coded slice data partitions that are subjectively the least important 3739 can be sent later than their natural coding order indicates. 3740 Consequently, any retransmitted parts of the most important slices 3741 and coded slice data partitions are more likely to be received before 3742 their scheduled decoding or playback time compared to the least 3743 important slices and slice data partitions. 3745 13. Informative Appendix: Rationale for Decoding Order Number 3747 13.1. Introduction 3749 The Decoding Order Number (DON) concept was introduced mainly to 3750 enable efficient multi-picture slice interleaving (see section 12.6) 3751 and robust packet scheduling (see section 12.7). In both of these 3752 applications, NAL units are transmitted out of decoding order. DON 3753 indicates the decoding order of NAL units and should be used in the 3754 receiver to recover the decoding order. Example use cases for 3755 efficient multi-picture slice interleaving and for robust packet 3756 scheduling are given in sections 13.2 and 13.3, respectively. 3757 Section 13.4 describes the benefits of the DON concept in error 3758 resiliency achieved by redundant coded pictures. Section 13.5 3759 summarizes considered alternatives to DON and justifies why DON was 3760 chosen to this RTP payload specification. 3762 13.2. Example of Multi-Picture Slice Interleaving 3764 An example of multi-picture slice interleaving follows. A subset of 3765 a coded video sequence is depicted below in output order. R denotes 3766 a reference picture, N denotes a non-reference picture, and the 3767 number indicates a relative output time. 3769 ... R1 N2 R3 N4 R5 ... 3771 The decoding order of these pictures from left to right is as 3772 follows: 3774 ... R1 R3 N2 R5 N4 ... 3776 The NAL units of pictures R1, R3, N2, R5, and N4 are marked with a 3777 DON equal to 1, 2, 3, 4, and 5, respectively. 3779 Each reference picture consists of three slice groups that are 3780 scattered as follows (a number denotes the slice group number for 3781 each macroblock in a QCIF frame): 3783 0 1 2 0 1 2 0 1 2 0 1 3784 2 0 1 2 0 1 2 0 1 2 0 3785 1 2 0 1 2 0 1 2 0 1 2 3786 0 1 2 0 1 2 0 1 2 0 1 3787 2 0 1 2 0 1 2 0 1 2 0 3788 1 2 0 1 2 0 1 2 0 1 2 3789 0 1 2 0 1 2 0 1 2 0 1 3790 2 0 1 2 0 1 2 0 1 2 0 3791 1 2 0 1 2 0 1 2 0 1 2 3793 For the sake of simplicity, we assume that all the macroblocks of a 3794 slice group are included in one slice. Three MTAPs are constructed 3795 from three consecutive reference pictures so that each MTAP contains 3796 three aggregation units, each of which contains all the macroblocks 3797 from one slice group. The first MTAP contains slice group 0 of 3798 picture R1, slice group 1 of picture R3, and slice group 2 of picture 3799 R5. The second MTAP contains slice group 1 of picture R1, slice 3800 group 2 of picture R3, and slice group 0 of picture R5. The third 3801 MTAP contains slice group 2 of picture R1, slice group 0 of picture 3802 R3, and slice group 1 of picture R5. Each non-reference picture is 3803 encapsulated into an STAP-B. 3805 Consequently, the transmission order of NAL units is the following: 3807 R1, slice group 0, DON 1, carried in MTAP,RTP SN: N 3808 R3, slice group 1, DON 2, carried in MTAP,RTP SN: N 3809 R5, slice group 2, DON 4, carried in MTAP,RTP SN: N 3810 R1, slice group 1, DON 1, carried in MTAP,RTP SN: N+1 3811 R3, slice group 2, DON 2, carried in MTAP,RTP SN: N+1 3812 R5, slice group 0, DON 4, carried in MTAP,RTP SN: N+1 3813 R1, slice group 2, DON 1, carried in MTAP,RTP SN: N+2 3814 R3, slice group 1, DON 2, carried in MTAP,RTP SN: N+2 3815 R5, slice group 0, DON 4, carried in MTAP,RTP SN: N+2 3816 N2, DON 3, carried in STAP-B, RTP SN: N+3 3817 N4, DON 5, carried in STAP-B, RTP SN: N+4 3819 The receiver is able to organize the NAL units back in decoding order 3820 based on the value of DON associated with each NAL unit. 3822 If one of the MTAPs is lost, the spatially adjacent and temporally 3823 co-located macroblocks are received and can be used to conceal the 3824 loss efficiently. If one of the STAPs is lost, the effect of the 3825 loss does not propagate temporally. 3827 13.3. Example of Robust Packet Scheduling 3829 An example of robust packet scheduling follows. The communication 3830 system used in the example consists of the following components in 3831 the order that the video is processed from source to sink: 3833 o camera and capturing 3834 o pre-encoding buffer 3835 o encoder 3836 o encoded picture buffer 3837 o transmitter 3838 o transmission channel 3839 o receiver 3840 o receiver buffer 3841 o decoder 3842 o decoded picture buffer 3843 o display 3845 The video communication system used in the example operates as 3846 follows. Note that processing of the video stream happens gradually 3847 and at the same time in all components of the system. The source 3848 video sequence is shot and captured to a pre-encoding buffer. The 3849 pre-encoding buffer can be used to order pictures from sampling order 3850 to encoding order or to analyze multiple uncompressed frames for bit 3851 rate control purposes, for example. In some cases, the pre-encoding 3852 buffer may not exist; instead, the sampled pictures are encoded right 3853 away. The encoder encodes pictures from the pre-encoding buffer and 3854 stores the output; i.e., coded pictures, to the encoded picture 3855 buffer. The transmitter encapsulates the coded pictures from the 3856 encoded picture buffer to transmission packets and sends them to a 3857 receiver through a transmission channel. The receiver stores the 3858 received packets to the receiver buffer. The receiver buffering 3859 process typically includes buffering for transmission delay jitter. 3860 The receiver buffer can also be used to recover correct decoding 3861 order of coded data. The decoder reads coded data from the receiver 3862 buffer and produces decoded pictures as output into the decoded 3863 picture buffer. The decoded picture buffer is used to recover the 3864 output (or display) order of pictures. Finally, pictures are 3865 displayed. 3867 In the following example figures, I denotes an IDR picture, R denotes 3868 a reference picture, N denotes a non-reference picture, and the 3869 number after I, R, or N indicates the sampling time relative to the 3870 previous IDR picture in decoding order. Values below the sequence of 3871 pictures indicate scaled system clock timestamps. The system clock 3872 is initialized arbitrarily in this example, and time runs from left 3873 to right. Each I, R, and N picture is mapped into the same timeline 3874 compared to the previous processing step, if any, assuming that 3875 encoding, transmission, and decoding take no time. Thus, events 3876 happening at the same time are located in the same column throughout 3877 all example figures. 3879 A subset of a sequence of coded pictures is depicted below in 3880 sampling order. 3882 ... N58 N59 I00 N01 N02 R03 N04 N05 R06 ... N58 N59 I00 N01 ... 3883 ... --|---|---|---|---|---|---|---|---|- ... -|---|---|---|- ... 3884 ... 58 59 60 61 62 63 64 65 66 ... 128 129 130 131 ... 3886 Figure 16 Sequence of pictures in sampling order 3888 The sampled pictures are buffered in the pre-encoding buffer to 3889 arrange them in encoding order. In this example, we assume that the 3890 non-reference pictures are predicted from both the previous and the 3891 next reference picture in output order, except for the non-reference 3892 pictures immediately preceding an IDR picture, which are predicted 3893 only from the previous reference picture in output order. Thus, the 3894 pre-encoding buffer has to contain at least two pictures, and the 3895 buffering causes a delay of two picture intervals. The output of the 3896 pre-encoding buffering process and the encoding (and decoding) order 3897 of the pictures are as follows: 3899 ... N58 N59 I00 R03 N01 N02 R06 N04 N05 ... 3900 ... -|---|---|---|---|---|---|---|---|- ... 3901 ... 60 61 62 63 64 65 66 67 68 ... 3903 Figure 17 Re-ordered pictures in the pre-encoding buffer 3905 The encoder or the transmitter can set the value of DON for each 3906 picture to a value of DON for the previous picture in decoding order 3907 plus one. 3909 For the sake of simplicity, let us assume that: 3911 o the frame rate of the sequence is constant, 3912 o each picture consists of only one slice, 3913 o each slice is encapsulated in a single NAL unit packet, 3914 o there is no transmission delay, and 3915 o pictures are transmitted at constant intervals (that is, 1 / 3916 (frame rate)). 3918 When pictures are transmitted in decoding order, they are received as 3919 follows: 3921 ... N58 N59 I00 R03 N01 N02 R06 N04 N05 ... 3922 ... -|---|---|---|---|---|---|---|---|- ... 3923 ... 60 61 62 63 64 65 66 67 68 ... 3925 Figure 18 Received pictures in decoding order 3927 The OPTIONAL sprop-interleaving-depth media type parameter is set to 3928 0, as the transmission (or reception) order is identical to the 3929 decoding order. 3931 The decoder has to buffer for one picture interval initially in its 3932 decoded picture buffer to organize pictures from decoding order to 3933 output order as depicted below: 3935 ... N58 N59 I00 N01 N02 R03 N04 N05 R06 ... 3936 ... -|---|---|---|---|---|---|---|---|- ... 3937 ... 61 62 63 64 65 66 67 68 69 ... 3939 Figure 19 Output order 3941 The amount of required initial buffering in the decoded picture 3942 buffer can be signaled in the buffering period SEI message or with 3943 the num_reorder_frames syntax element of H.264 video usability 3944 information. num_reorder_frames indicates the maximum number of 3945 frames, complementary field pairs, or non-paired fields that precede 3946 any frame, complementary field pair, or non-paired field in the 3947 sequence in decoding order and that follow it in output order. For 3948 the sake of simplicity, we assume that num_reorder_frames is used to 3949 indicate the initial buffer in the decoded picture buffer. In this 3950 example, num_reorder_frames is equal to 1. 3952 It can be observed that if the IDR picture I00 is lost during 3953 transmission and a retransmission request is issued when the value of 3954 the system clock is 62, there is one picture interval of time (until 3955 the system clock reaches timestamp 63) to receive the retransmitted 3956 IDR picture I00. 3958 Let us then assume that IDR pictures are transmitted two frame 3959 intervals earlier than their decoding position; i.e., the pictures 3960 are transmitted as follows: 3962 ... I00 N58 N59 R03 N01 N02 R06 N04 N05 ... 3963 ... --|---|---|---|---|---|---|---|---|- ... 3964 ... 62 63 64 65 66 67 68 69 70 ... 3966 Figure 20 Interleaving: Early IDR pictures in sending order 3968 The OPTIONAL sprop-interleaving-depth media type parameter is set 3969 equal to 1 according to its definition. (The value of sprop- 3970 interleaving-depth in this example can be derived as follows: Picture 3971 I00 is the only picture preceding picture N58 or N59 in transmission 3972 order and following it in decoding order. Except for pictures I00, 3973 N58, and N59, the transmission order is the same as the decoding 3974 order of pictures. As a coded picture is encapsulated into exactly 3975 one NAL unit, the value of sprop-interleaving-depth is equal to the 3976 maximum number of pictures preceding any picture in transmission 3977 order and following the picture in decoding order.) 3979 The receiver buffering process contains two pictures at a time 3980 according to the value of the sprop-interleaving-depth parameter and 3981 orders pictures from the reception order to the correct decoding 3982 order based on the value of DON associated with each picture. The 3983 output of the receiver buffering process is as follows: 3985 ... N58 N59 I00 R03 N01 N02 R06 N04 N05 ... 3986 ... -|---|---|---|---|---|---|---|---|- ... 3987 ... 63 64 65 66 67 68 69 70 71 ... 3989 Figure 21 Interleaving: Receiver buffer 3991 Again, an initial buffering delay of one picture interval is needed 3992 to organize pictures from decoding order to output order, as depicted 3993 below: 3995 ... N58 N59 I00 N01 N02 R03 N04 N05 ... 3996 ... -|---|---|---|---|---|---|---|- ... 3997 ... 64 65 66 67 68 69 70 71 ... 3999 Figure 22 Interleaving: Receiver buffer after reordering 4001 Note that the maximum delay that IDR pictures can undergo during 4002 transmission, including possible application, transport, or link 4003 layer retransmission, is equal to three picture intervals. Thus, the 4004 loss resiliency of IDR pictures is improved in systems supporting 4005 retransmission compared to the case in which pictures were 4006 transmitted in their decoding order. 4008 13.4. Robust Transmission Scheduling of Redundant Coded Slices 4010 A redundant coded picture is a coded representation of a picture or a 4011 part of a picture that is not used in the decoding process if the 4012 corresponding primary coded picture is correctly decoded. There 4013 should be no noticeable difference between any area of the decoded 4014 primary picture and a corresponding area that would result from 4015 application of the H.264 decoding process for any redundant picture 4016 in the same access unit. A redundant coded slice is a coded slice 4017 that is a part of a redundant coded picture. 4019 Redundant coded pictures can be used to provide unequal error 4020 protection in error-prone video transmission. If a primary coded 4021 representation of a picture is decoded incorrectly, a corresponding 4022 redundant coded picture can be decoded. Examples of applications and 4023 coding techniques using the redundant codec picture feature include 4024 the video redundancy coding [23] and the protection of "key pictures" 4025 in multicast streaming [24]. 4027 One property of many error-prone video communications systems is that 4028 transmission errors are often bursty. Therefore, they may affect 4029 more than one consecutive transmission packets in transmission order. 4030 In low bit-rate video communication, it is relatively common that an 4031 entire coded picture can be encapsulated into one transmission 4032 packet. Consequently, a primary coded picture and the corresponding 4033 redundant coded pictures may be transmitted in consecutive packets in 4034 transmission order. To make the transmission scheme more tolerant of 4035 bursty transmission errors, it is beneficial to transmit the primary 4036 coded picture and redundant coded picture separated by more than a 4037 single packet. The DON concept enables this. 4039 13.5. Remarks on Other Design Possibilities 4041 The slice header syntax structure of the H.264 coding standard 4042 contains the frame_num syntax element that can indicate the decoding 4043 order of coded frames. However, the usage of the frame_num syntax 4044 element is not feasible or desirable to recover the decoding order, 4045 due to the following reasons: 4047 o The receiver is required to parse at least one slice header per 4048 coded picture (before passing the coded data to the decoder). 4050 o Coded slices from multiple coded video sequences cannot be 4051 interleaved, as the frame number syntax element is reset to 0 in 4052 each IDR picture. 4054 o The coded fields of a complementary field pair share the same 4055 value of the frame_num syntax element. Thus, the decoding order 4056 of the coded fields of a complementary field pair cannot be 4057 recovered based on the frame_num syntax element or any other 4058 syntax element of the H.264 coding syntax. 4060 The RTP payload format for transport of MPEG-4 elementary streams 4061 [25] enables interleaving of access units and transmission of 4062 multiple access units in the same RTP packet. An access unit is 4063 specified in the H.264 coding standard to comprise all NAL units 4064 associated with a primary coded picture according to subclause 4065 7.4.1.2 of [1]. Consequently, slices of different pictures cannot be 4066 interleaved, and the multi-picture slice interleaving technique (see 4067 section 12.6) for improved error resilience cannot be used. 4069 14. Acknowledgements 4071 Stephan Wenger, Miska Hannuksela, Thomas Stockhammer, Magnus 4072 Westerlund, and David Singer are thanked as the authors of RFC 3984. 4073 Dave Lindbergh, Philippe Gentric, Gonzalo Camarillo, Gary Sullivan, 4074 Joerg Ott, and Colin Perkins are thanked for careful review during 4075 the development of RFC 3984. Randell Jesup, Stephen Botzko, Magnus 4076 Westerlund, Alex Eleftheriadis, and Thomas Schierl are thanked for 4077 their valuable comments and inputs during the development of this 4078 memo. 4080 This document was prepared using 2-Word-v2.0.template.dot. 4082 15. References 4084 15.1. Normative References 4086 [1] ITU-T Recommendation H.264, "Advanced video coding for generic 4087 audiovisual services", November 2007. 4089 [2] ISO/IEC International Standard 14496-10:2008. 4091 [3] ITU-T Recommendation H.241, "Extended video procedures and 4092 control signals for H.300 series terminals", May 2006. 4094 [4] Bradner, S., "Key words for use in RFCs to Indicate Requirement 4095 Levels", BCP 14, RFC 2119, March 1997. 4097 [5] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson, 4098 "RTP: A Transport Protocol for Real-Time Applications", STD 64, 4099 RFC 3550, July 2003. 4101 [6] Handley, M. and V. Jacobson, "SDP: Session Description 4102 Protocol", RFC 2327, April 1998. 4104 [7] Josefsson, S., "The Base16, Base32, and Base64 Data Encodings", 4105 RFC 3548, July 2003. 4107 [8] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model with 4108 Session Description Protocol (SDP)", RFC 3264, June 2002. 4110 [9] Lennox, J., Ott, J., and Schierl, T., "Source-Specific Media 4111 Attributes in the Session Description Protocol", draft-ietf- 4112 mmusic-sdp-source-attributes-02 (work in progress), October 4113 2008. 4115 15.2. Informative References 4117 [10] Luthra, A., Sullivan, G.J., and T. Wiegand (eds.), Special 4118 Issue on H.264/AVC. IEEE Transactions on Circuits and Systems 4119 on Video Technology, July 2003. 4121 [11] Bormann, C., Cline, L., Deisher, G., Gardos, T., Maciocco, C., 4122 Newell, D., Ott, J., Sullivan, G., Wenger, S., and C. Zhu, "RTP 4123 Payload Format for the 1998 Version of ITU-T Rec. H.263 Video 4124 (H.263+)", RFC 2429, October 1998. 4126 [12] ISO/IEC IS 14496-2. 4128 [13] Wenger, S., "H.26L over IP", IEEE Transaction on Circuits and 4129 Systems for Video technology, Vol. 13, No. 7, July 2003. 4131 [14] Wenger, S., "H.26L over IP: The IP Network Adaptation Layer", 4132 Proceedings Packet Video Workshop 02, April 2002. 4134 [15] Stockhammer, T., Hannuksela, M.M., and S. Wenger, "H.26L/JVT 4135 Coding Network Abstraction Layer and IP-based Transport" in 4136 Proc. ICIP 2002, Rochester, NY, September 2002. 4138 [16] Schulzrinne, H. and S. Casner, "RTP Profile for Audio and Video 4139 Conferences with Minimal Control", STD 65, RFC 3551, July 2003. 4141 [17] ITU-T Recommendation H.223, "Multiplexing protocol for low bit 4142 rate multimedia communication", July 2001. 4144 [18] Li, A., "RTP Payload Format for Generic Forward Error 4145 Correction", RFC 5109, December 2007. 4147 [19] Stockhammer, T., Wiegand, T., Oelbaum, T., and F. Obermeier, 4148 "Video Coding and Transport Layer Techniques for H.264/AVC- 4149 Based Transmission over Packet-Lossy Networks", IEEE 4150 International Conference on Image Processing (ICIP 2003), 4151 Barcelona, Spain, September 2003. 4153 [20] Varsa, V. and M. Karczewicz, "Slice interleaving in compressed 4154 video packetization", Packet Video Workshop 2000. 4156 [21] Kang, S.H. and A. Zakhor, "Packet scheduling algorithm for 4157 wireless video streaming," International Packet Video Workshop 4158 2002. 4160 [22] Hannuksela, M.M., "Enhanced concept of GOP", JVT-B042, 4161 available http://ftp3.itu.int/av-arch/video-site/0201_Gen/JVT- 4162 B042.doc, anuary 2002. 4164 [23] Wenger, S., "Video Redundancy Coding in H.263+", 1997 4165 International Workshop on Audio-Visual Services over Packet 4166 Networks, September 1997. 4168 [24] Wang, Y.-K., Hannuksela, M.M., and M. Gabbouj, "Error Resilient 4169 Video Coding Using Unequally Protected Key Pictures", in Proc. 4170 International Workshop VLBV03, September 2003. 4172 [25] van der Meer, J., Mackie, D., Swaminathan, V., Singer, D., and 4173 P. Gentric, "RTP Payload Format for Transport of MPEG-4 4174 Elementary Streams", RFC 3640, November 2003. 4176 [26] Baugher, M., McGrew, D., Naslund, M., Carrara, E., and K. 4177 Norrman, "The Secure Real-time Transport Protocol (SRTP)", RFC 4178 3711, March 2004. 4180 [27] Schulzrinne, H., Rao, A., and R. Lanphier, "Real Time Streaming 4181 Protocol (RTSP)", RFC 2326, April 1998. 4183 [28] Handley, M., Perkins, C., and E. Whelan, "Session Announcement 4184 Protocol", RFC 2974, October 2000. 4186 [29] Westerlund, M. and Wenger, S., "RTP Topologies", RFC 5117, 4187 January 2008. 4189 [30] Wenger, S., Chandra, U., and Westerlund, M., "Codec Control 4190 Messages in the RTP Audio-Visual Profile with Feedback (AVPF)", 4191 RFC 5104, February 2008. 4193 Authors' Addresses 4195 Ye-Kui Wang 4196 Nokia Research Center 4197 P.O. Box 1000 4198 33721 Tampere 4199 Finland 4201 Phone: +358-50-466-7004 4202 EMail: ye-kui.wang@nokia.com 4204 Roni Even 4205 14 David Hamelech 4206 Tel Aviv 64953 4207 Israel 4209 Phone: +972-545481099 4210 Email:ron.even.tlv@gmail.com 4212 Tom Kristensen 4213 TANDBERG 4214 Philip Pedersens vei 22 4215 N-1366 Lysaker 4216 Norway 4218 Phone: +47 67125125 4219 Email: tom.kristensen@tandberg.com, tomkri@ifi.uio.no 4221 Intellectual Property Statement 4223 The IETF takes no position regarding the validity or scope of any 4224 Intellectual Property Rights or other rights that might be claimed to 4225 pertain to the implementation or use of the technology described in 4226 this document or the extent to which any license under such rights 4227 might or might not be available; nor does it represent that it has 4228 made any independent effort to identify any such rights. Information 4229 on the procedures with respect to rights in RFC documents can be 4230 found in BCP 78 and BCP 79. 4232 Copies of IPR disclosures made to the IETF Secretariat and any 4233 assurances of licenses to be made available, or the result of an 4234 attempt made to obtain a general license or permission for the use of 4235 such proprietary rights by implementers or users of this 4236 specification can be obtained from the IETF on-line IPR repository at 4237 http://www.ietf.org/ipr. 4239 The IETF invites any interested party to bring to its attention any 4240 copyrights, patents or patent applications, or other proprietary 4241 rights that may cover technology that may be required to implement 4242 this standard. Please address the information to the IETF at 4243 ietf-ipr@ietf.org. 4245 Disclaimer of Validity 4247 This document and the information contained herein are provided on an 4248 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 4249 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 4250 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 4251 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 4252 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 4253 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 4255 Copyright Statement 4257 Copyright (C) The IETF Trust (2008). 4259 This document is subject to the rights, licenses and restrictions 4260 contained in BCP 78, and except as set forth therein, the authors 4261 retain all their rights. 4263 Acknowledgement 4265 Funding for the RFC Editor function is currently provided by the 4266 Internet Society. 4268 16. Backward Compatibility to RFC 3984 4270 The current document is a revision of RFC 3984 and intends to 4271 obsolete it. This section addresses the backward compatibility 4272 issues. 4274 The technical changes are listed in section 17. 4276 Items 1), 2), 3), 7), 9), 10), 12), 13) are bug-fix type of changes, 4277 and do not incur any backward compatibility issues. 4279 Item 4), addition of five new media type parameters, does not incur 4280 any backward compatibility issues for SDP Offer/Answer based 4281 applications, as legacy RFC 3984 receivers ignore these parameters, 4282 and it is fine for legacy RFC 3984 senders not to use these 4283 parameters as they are optional. However, there is a backward 4284 compatibility issue for SDP declarative usage based applications, 4285 e.g. those using RTSP and SAP, because the SDP receiver per RFC 3984 4286 cannot accept a session for which the SDP includes an unrecognized 4287 parameter. Therefore, the RTSP or SAP server may have to prepare two 4288 sets of streams, one for legacy RFC 3984 receivers and one for 4289 receivers according to this memo. 4291 Items 5), 6) and 11) are related to out-of-band transport of 4292 parameter sets. When a sender according to this memo is 4293 communicating with a legacy receiver according to RFC 3984, there is 4294 no backward compatibility issue. When the legacy receiver sees an SDP 4295 message with no parameter-add the value of parameter-add is inferred 4296 to be equal to 1 by the legacy receiver (related to change item 5)). 4297 As RFC 3984 allows inclusion of any parameter sets in sprop- 4298 parameter-sets, it is fine to the legacy receiver to include 4299 parameter sets only for the default level in sprop-parameter-sets 4300 (related to change item 6)). When there are new parameters e.g. 4301 sprop-level-parameter-sets present, the legacy receiver simply 4302 ignores them (related to change item 11)). When a legacy sender 4303 according to RFC 3984 is communicating with a receiver according to 4304 this memo, there is one backward compatibility issue. When the 4305 legacy sender includes parameter sets for a level different than the 4306 default level indicated by profile-level-id to sprop-parameter-sets, 4307 the parameter value of sprop-parameter-sets is invalid to the 4308 receiver and therefore the session may be rejected. In SDP 4309 Offer/Answer between a legacy offerer according to RFC 3984 and an 4310 answerer according to this memo, when the answerer includes in the 4311 answer parameter sets that are not a superset of the parameter sets 4312 included in the offer, the parameter value of sprop-parameter-sets is 4313 invalid to offerer and the session may not be initiated properly 4314 (related to change item 11)). 4316 Item 7), allowance of conveying sprop-parameter-sets and sprop-level- 4317 parameter-sets using the "fmtp" source attribute as specified in 4318 section 6.3 of [9], is similar as item 4). It does not incur any 4319 backward compatibility issues for SDP Offer/Answer based 4320 applications, as legacy RFC 3984 receivers ignore the "fmtp" source 4321 attribute, and it is fine for legacy RFC 3984 senders not to use the 4322 "fmtp" source attribute as it is optional. However, there is a 4323 backward compatibility issue for SDP declarative usage based 4324 applications, e.g. those using RTSP and SAP, because the SDP receiver 4325 per RFC 3984 cannot accept a session for which the SDP includes an 4326 unrecognized parameter (i.e., the "fmtp" source attribute). 4327 Therefore, the RTSP or SAP server may have to prepare two sets of 4328 streams, one for legacy RFC 3984 receivers and one for receivers 4329 according to this memo. 4331 Item 14) removed that use of out-of-band transport of parameter sets 4332 is recommended. As out-of-band transport of parameter sets is still 4333 allowed, this change does not incur any backward compatibility 4334 issues. 4336 Item 15) does not incur any backward compatibility issues as the 4337 added subsection 8.5 is informative. 4339 17. Changes from RFC 3984 4341 Following is the list of technical changes (including bug fixes) from 4342 RFC 3984. Besides this list of technical changes, numerous editorial 4343 changes have been made, but not documented in this memo. 4345 1) In subsections 5.4, 5.5, 6.2, 6,3 and 6.4, removed that the 4346 packetization mode in use may be signaled by external means. 4348 2) In subsection 7.2.2, changed the sentence 4350 There are N VCL NAL units in the deinterleaving buffer. 4352 to 4354 There are N or more VCL NAL units in the de-interleaving buffer. 4356 3) In subsection 8.1, the semantics of sprop-init-buf-time, paragraph 4357 2, changed the sentence 4359 The parameter is the maximum value of (transmission time of a NAL 4360 unit - decoding time of the NAL unit), assuming reliable and 4361 instantaneous transmission, the same timeline for transmission 4362 and decoding, and that decoding starts when the first packet 4363 arrives. 4365 to 4367 The parameter is the maximum value of (decoding time of the NAL 4368 unit - transmission time of a NAL unit), assuming reliable and 4369 instantaneous transmission, the same timeline for transmission 4370 and decoding, and that decoding starts when the first packet 4371 arrives. 4373 4) Added five new media type parameters, namely max-smbps, sprop- 4374 level-parameter-sets, use-level-src-parameter-sets, sar-understood 4375 and sar-supported. 4377 5) In subsection 8.1, removed the specification of parameter-add. 4378 Other descriptions of parameter-add (in subsections 8.2 and 8.4) 4379 are also removed. 4381 6) In subsection 8.1, added a constraint to sprop-parameter-sets such 4382 that it can only contain parameter sets for the same profile and 4383 level as indicated by profile-level-id. 4385 7) In subsection 8.2.1, added that sprop-parameter-sets and sprop- 4386 level-parameter-sets may be either included in the "a=fmtp" line 4387 of SDP or conveyed using the "fmtp" source attribute as specified 4388 in section 6.3 of [9]. 4390 8) In subsection 8.2.2, removed sprop-deint-buf-req from being part 4391 of the media format configuration in usage with the SDP 4392 Offer/Answer model. 4394 9) In subsection 8.2.2, made it clear that level is downgradable in 4395 the SDP Offer/Answer model, i.e. the use of the level part of 4396 "profile-level-id" does not need to be symmetric (the level 4397 included in the answer can be lower than or equal to the level 4398 included in the offer). 4400 10)In subsection 8.2.2, removed that the capability parameters may be 4401 used to declare encoding capabilities. 4403 11)In subsection 8.2.2, added rules on how to use sprop-parameter- 4404 sets and sprop-level-parameter-sets for out-of-band transport of 4405 parameter sets, with or without level downgrading. 4407 12)In subsection 8.2.2, clarified the rules of using the media type 4408 parameters with SDP Offer/Answer for multicast. 4410 13)In subsection 8.2.2, completed and corrected the list of how 4411 different media type parameters shall be interpreted in the 4412 different combinations of offer or answer and direction attribute. 4414 14)In subsection 8.4, changed the text such that both out-of-band and 4415 in-band transport of parameter sets are allowed and neither is 4416 recommended or required. 4418 15)Added subsection 8.5 (informative) providing example methods for 4419 decoder refresh to handle parameter set losses.