idnits 2.17.1 draft-ietf-avt-rtp-mpeg4-02.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** Looks like you're using RFC 2026 boilerplate. This must be updated to follow RFC 3978/3979, as updated by RFC 4748. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- ** The document seems to lack a 1id_guidelines paragraph about 6 months document validity -- however, there's a paragraph with a matching beginning. Boilerplate error? == No 'Intended status' indicated for this document; assuming Proposed Standard == The page length should not exceed 58 lines per page, but there was 2 longer pages, the longest (page 10) being 97 lines == It seems as if not all pages are separated by form feeds - found 0 form feeds but 10 pages Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The document seems to lack an IANA Considerations section. (See Section 2.2 of https://www.ietf.org/id-info/checklist for how to handle the case when there are no actions for IANA.) ** The document seems to lack separate sections for Informative/Normative References. All references will be assumed normative when checking for downward references. ** There are 7 instances of too long lines in the document, the longest one being 4 characters in excess of 72. ** There is 1 instance of lines with control characters in the document. Miscellaneous warnings: ---------------------------------------------------------------------------- == Line 154 has weird spacing: '... media unawar...' == Line 429 has weird spacing: '...cessing to ca...' -- 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 (October 22, 1999) is 8951 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) -- Possible downref: Non-RFC (?) normative reference: ref. '1' -- Possible downref: Non-RFC (?) normative reference: ref. '2' -- Possible downref: Non-RFC (?) normative reference: ref. '3' -- Possible downref: Non-RFC (?) normative reference: ref. '4' ** Obsolete normative reference: RFC 1889 (ref. '5') (Obsoleted by RFC 3550) -- Possible downref: Normative reference to a draft: ref. '7' Summary: 7 errors (**), 0 flaws (~~), 5 warnings (==), 7 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 1 Internet Engineering Task Force Civanlar-AT&T/ Basso-AT&T 2 INTERNET DRAFT Casner-Cisco 3 File: draft-ietf-avt-rtp-mpeg4-02.txt Herpel-Thomson/Perkins-UCL 4 October 22, 1999 5 Expires: April 22, 2000 7 RTP Payload Format for MPEG-4 Streams 9 STATUS OF THIS MEMO 11 This document is an Internet-Draft and is in full conformance with all 12 provisions of Section 10 of RFC2026. 14 Internet-Drafts are working documents of the Internet Engineering Task 15 Force (IETF), its areas, and its working groups. Note that other 16 groups may also distribute working documents as Internet-Drafts. 18 Internet-Drafts are draft documents valid for a maximum of six months 19 and may be updated, replaced, or obsoleted by other documents at any 20 time. It is inappropriate to use Internet- Drafts as reference 21 material or to cite them other than as "work in progress." 23 The list of current Internet-Drafts can be accessed at 24 http://www.ietf.org/ietf/1id-abstracts.txt 26 The list of Internet-Draft Shadow Directories can be accessed at 27 http://www.ietf.org/shadow.html. 29 Abstract 31 This document describes a payload format for transporting MPEG-4 32 encoded data using RTP. MPEG-4 is a recent standard from ISO/IEC for 33 the coding of natural and synthetic audio-visual data. Several 34 services provided by RTP are beneficial for MPEG-4 encoded data 35 transport over the Internet. Additionally, the use of RTP makes it 36 possible to synchronize MPEG-4 data with other real-time data types. 38 This specification is a product of the Audio/Video Transport working 39 group within the Internet Engineering Task Force and ISO/IEC MPEG-4 ad 40 hoc group on MPEG-4 over Internet. Comments are solicited and should 41 be addressed to the working group's mailing list at rem-conf@es.net 42 and/or the authors. 44 1. Introduction 46 MPEG-4 is a recent standard from ISO/IEC for the coding of natural and 47 synthetic audio-visual data in the form of audiovisual objects that 48 are arranged into an audiovisual scene by means of a scene description 49 [1][2][3][4]. This draft specifies an RTP [5] payload format for 50 transporting MPEG-4 encoded data streams. 52 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 53 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 54 document are to be interpreted as described in RFC 2119 [6]. 56 The benefits of using RTP for MPEG-4 data stream transport include: 58 i. Ability to synchronize MPEG-4 streams with other RTP payloads 60 ii. Monitoring MPEG-4 delivery performance through RTCP 62 iii. Combining MPEG-4 and other real-time data streams received 63 from multiple end-systems into a set of consolidated streams 64 through RTP mixers 66 iv. Converting data types, etc. through the use of RTP translators. 68 1.1 Overview of MPEG-4 End-System Architecture 70 Fig. 1 below shows the general layered architecture of MPEG-4 71 terminals. The Compression Layer processes individual audio-visual 72 media streams. The MPEG-4 compression schemes are defined in the 73 ISO/IEC specifications 14496-2 [2] and 14496-3 [3]. The compression 74 schemes in MPEG-4 achieve efficient encoding over a bandwidth ranging 75 from several Kbps to many Mbps. The audio-visual content compressed by 76 this layer is organized into Elementary Streams (ESs). The MPEG-4 77 standard specifies MPEG-4 compliant streams. Within the constraint of 78 this compliance the compression layer is unaware of a specific delivery 79 technology, but it can be made to react to the characteristics of a 80 particular delivery layer such as the path-MTU or loss characteristics. 81 Also, some compressors can be designed to be delivery specific for 82 implementation efficiency. In such cases the compressor may work in a 83 non-optimal fashion with delivery technologies that are different than 84 the one it is specifically designed to operate with. 86 The hierarchical relations, location and properties of ESs in a 87 presentation are described by a dynamic set of Object Descriptors 88 (ODs). Each OD groups one or more ES Descriptors referring to a single 89 content item (audio-visual object). Hence, multiple alternative or 90 hierarchical representations of each content item are possible. 92 ODs are themselves conveyed through one or more ESs. A complete set of 93 ODs can be seen as an MPEG-4 resource or session description at a 94 stream level. The resource description may itself be hierarchical, i.e. 95 an ES conveying an OD may describe other ESs conveying other ODs. 97 The session description is accompanied by a dynamic scene description, 98 Binary Format for Scene (BIFS), again conveyed through one or more ESs. 99 At this level, content is identified in terms of audio-visual objects. 100 The spatiotemporal location of each object is defined by BIFS. The 101 audio-visual content of those objects that are synthetic and static are 102 described by BIFS also. Natural and animated synthetic objects may 103 refer to an OD that points to one or more ESs that carry the coded 104 representation of the object or its animation data. 106 By conveying the session (or resource) description as well as the scene 107 (or content composition) description through their own ESs, it is made 108 possible to change portions of the content composition and the number 109 and properties of media streams that carry the audio-visual content 110 separately and dynamically at well known instants in time. 112 One or more initial Scene Description streams and the corresponding OD 113 streams has to be pointed to by an initial object descriptor (IOD). The 114 IOD needs to be made available to the receivers through some out-of- 115 band means which are not defined in this document. 117 A homogeneous encapsulation of ESs carrying media or control (ODs, 118 BIFS) data is defined by the Sync Layer (SL) that primarily provides 119 the synchronization between streams. The Compression Layer organizes 120 the ESs in Access Units (AU), the smallest elements that can be 121 attributed individual timestamps. Integer or fractional AUs are then 122 encapsulated in SL packets. All consecutive data from one stream is 123 called an SL-packetized stream at this layer. The interface between the 124 compression layer and the SL is called the Elementary Stream Interface 125 (ESI). The ESI is informative. 127 The Delivery Layer in MPEG-4 consists of the Delivery Multimedia 128 Integration Framework defined in ISO/IEC 14496-6 [4]. This layer is 129 media unaware but delivery technology aware. It provides transparent 130 access to and delivery of content irrespective of the technologies 131 used. The interface between the SL and DMIF is called the DMIF 132 Application Interface (DAI). It offers content location independent 133 procedures for establishing MPEG-4 sessions and access to transport 134 channels. The specification of this payload format is considered as a 135 part of the MPEG-4 Delivery Layer. 137 media aware +-----------------------------------------+ 138 delivery unaware | COMPRESSION LAYER | 139 14496-2 Visual |streams from as low as Kbps to multi-Mbps| 140 14496-3 Audio +-----------------------------------------+ Elementary 141 Stream 142 ================================================================Interface 143 (ESI) 144 +-------------------------------------------+ 145 media and | SYNC LAYER | 146 delivery unaware | manages elementary streams, their synch- | 147 14496-1 Systems | ronization and hierarchical relations | 148 +-------------------------------------------+ DMIF 149 Application 150 ================================================================Interface 151 (DAI) 152 +-------------------------------------------+ 153 delivery aware | DELIVERY LAYER | 154 media unaware |provides transparent access to and delivery| 155 14496-6 DMIF | of content irrespective of delivery | 156 | technology | 157 +-------------------------------------------+ 159 Figure 1: General MPEG-4 terminal architecture 161 1.2 MPEG-4 Elementary Stream Data Packetization 163 The ESs from the encoders are fed into the SL with indications of AU 164 boundaries, random access points, desired composition time and the 165 current time. 167 The Sync Layer fragments the ESs into SL packets, each containing a 168 header which encodes information conveyed through the ESI. If the AU is 169 larger than an SL packet, subsequent packets containing remaining parts 170 of the AU are generated with subset headers until the complete AU is 171 packetized. 173 The syntax of the Sync Layer is not fixed and can be adapted to the 174 needs of the stream to be transported. This includes the possibility to 175 select the presence or absence of individual syntax elements as well as 176 configuration of their length in bits. The configuration for each 177 individual stream is conveyed in an SLConfigDescriptor, which is an 178 integral part of the ES Descriptor for this stream. 180 2. Analysis of the alternatives for carrying MPEG-4 over IP 182 2.1 MPEG-4 over UDP 184 Considering that the MPEG-4 SL defines several transport related 185 functions such as timing, sequence numbering, etc., this seems to be 186 the most straightforward alternative for carrying MPEG-4 data over IP. 187 One group of problems with this approach, however, stems from the 188 monolithic architecture of MPEG-4. No other multimedia data stream 189 (including those carried with RTP) can be synchronized with MPEG-4 data 190 carried directly over UDP. Furthermore, the dynamic scene and session 191 control concepts can't be extended to non-MPEG-4 data. 193 Even if the coordination with non-MPEG-4 data is overlooked, carrying 194 MPEG-4 data over UDP has the following additional shortcomings: 196 i. Mechanisms need to be defined to protect sensitive parts of 197 MPEG-4 data. Some of these (like FEC) are already defined for 198 RTP. 200 ii. There is no defined technique for synchronizing MPEG-4 201 streams from different servers in the variable delay environment 202 of the Internet. 204 iii. MPEG-4 streams originating from two servers may collide (their 205 sources may become unresolvable at the destination) in a multicast 206 session. 208 iv. An MPEG-4 backchannel needs to be defined for quality 209 feedback similar to that provided by RTCP. 211 v. RTP mixers and translators can't be used. 213 The backchannel problem may be alleviated by developing a reception 214 reporting protocol like RTCP. Such an effort may benefit from RTCP 215 design knowledge, but needs extensions. 217 2.2 RTP header followed by full MPEG-4 headers 219 This alternative may be implemented by using the send time or the 220 composition time coming from the reference clock as the RTP timestamp. 221 This way no new feedback protocol needs to be defined for MPEG-4's 222 backchannel, but RTCP may not be sufficient for MPEG-4's feedback 223 requirements which are still in the definition stage. Additionally, due 224 to the duplication of header information, such as the sequence numbers 225 and time stamps, this alternative causes unnecessary increases in the 226 overhead. Scene description or dynamic session control can't be extended 227 to non-MPEG-4 streams also. 229 2.3 MPEG-4 ESs over RTP with individual payload types 231 This is the most suitable alternative for coordination with the existing 232 Internet multimedia transport techniques and does not use MPEG-4 systems 233 at all. Complete implementation of it requires definition of potentially 234 many payload types and might lead to constructing new session and scene 235 description mechanisms. Considering the size of the work involved which 236 essentially reconstructs MPEG-4 systems, this may only be a long term 237 alternative if no other solution can be found. 239 2.4 RTP header followed by a reduced SL header 241 The inefficiency of the approach described in 2.2 can be fixed by using 242 a reduced SL header that does not carry duplicate information following 243 the RTP header. 245 2.5 Recommendation 247 Based on the above analysis, the best compromise is to map the MPEG-4 SL 248 packets onto RTP packets, such that the common pieces of the headers 249 reside in the RTP header that is followed by an optional reduced SL 250 header providing the MPEG-4 specific information. The details of this 251 payload format are described in the next section. 253 3. Payload Format 255 The RTP Payload consists of a single SL packet, including an SL packet 256 header without the sequenceNumber and compositionTimeStamp fields. Use 257 of all other fields in the SL packet headers that the RTP header does 258 not duplicate (including the decodingTimeStamp) is OPTIONAL. Packets 259 SHOULD be sent in the decoding order. 261 If the resulting, smaller, SL packet header consumes a non-integer 262 number of bytes, zero padding bits MUST be inserted at the end of the SL 263 header to byte-align the SL packet payload. 265 The size of the SL packets SHOULD be adjusted such that the resulting 266 RTP packet is not larger than the path-MTU. To handle larger packets, 267 this payload format relies on lower layers for fragmentation which may 268 not be desirable. 270 0 1 2 3 271 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 1 272 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 273 |V=2|P|X| CC |M| PT | sequence number | RTP 274 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 275 | timestamp | Header 276 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 277 | synchronization source (SSRC) identifier | 278 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 279 : contributing source (CSRC) identifiers : 280 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 281 |SL Packet Header (variable # of bytes) | | 282 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | RTP 283 | | 284 | SL Packet Payload (byte aligned) | Payload 285 | | 286 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 287 | :...OPTIONAL RTP padding | 288 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 289 Figure 2 - An RTP packet for MPEG-4 291 3.1 RTP Header Fields Usage: 293 Payload Type (PT): The assignment of an RTP payload type for this new 294 packet format is outside the scope of this document, and will not be 295 specified here. It is expected that the RTP profile for a particular 296 class of applications will assign a payload type for this encoding, or 297 if that is not done then a payload type in the dynamic range shall be 298 chosen. 300 Marker (M) bit: Set to one to mark the last fragment (or only fragment) 301 of an AU. 303 Extension (X) bit: Defined by the RTP profile used. 305 Sequence Number: Derived from the sequenceNumber field of the SL packet 306 by adding a constant random offset. If the sequenceNumber has less than 307 16-bit length, the MSBs MUST initially be filled with a random value 308 that is incremented by one each time the sequenceNumber value of the SL 309 packet returns to zero. If the value sequenceNumber=0 is encountered in 310 multiple consecutive SL packets, indicating a deliberate duplication of 311 the SL packet, the sequence number SHOULD be incremented by one for each 312 of these packets after the first one. 314 In implementations where full SL packets are generated first and then 315 packetised in RTP, the sequenceNumber MUST be removed from the SL packet 316 header by bit-shifting the subsequent header elements towards the 317 beginning of the header. When unpacking the RTP packet this process can 318 be reversed with the knowledge of the SLConfigDescriptor. For using this 319 payload format, MPEG-4 implementations that do not produce the full SL 320 packet in the first place, but rather produce the RTP header and 321 stripped down (perhaps null) SL header directly are preferable. 323 However, the choice between generating SL packets and converting, or 324 generating RTP directly is an implementation detail, and does not affect 325 what goes on the wire. Both forms will interwork. 327 If no sequenceNumber field is configured for this stream (no 328 sequenceNumber field present in the SL packet header), then the RTP 329 packetizer MUST generate its own sequence numbers. 331 Timestamp: Set to the value in the compositionTimeStamp field of the SL 332 packet, if present. If compositionTimeStamp has less than 32 bits 333 length, the MSBs of timestamp MUST be set to zero. 335 Although it is available from the SL configuration data, the resolution 336 of the timestamp may need to be conveyed explicitly through some out- 337 of-band means to be used by network elements which are not MPEG-4 aware. 339 If compositionTimeStamp has more than 32 bits length, this payload 340 format cannot be used. 342 In case compositionTimeStamp is not present in the current SL packet, 343 but has been present in a previous SL packet, this same value MUST be 344 taken again as the as the compositionTimeStamp of the current SL packet. 346 If compositionTimeStamp is never present in SL packets for this stream, 347 the RTP packetizer SHOULD convey a reading of a local clock at the time 348 the RTP packet is created. 350 Similar to handling of the sequence numbers in implementations that 351 generate full SL packets, the compositionTimeStamp, if present, MUST 352 then be removed from the SL packet header by bit-shifting the subsequent 353 header elements towards the beginning of the SL packet header. When 354 unpacking the RTP packet this process can be reversed with the knowledge 355 of the SLConfigDescriptor and by evaluating the 356 compositionTimeStampFlag. 358 Timestamps are recommended to start at a random value for security 359 reasons [5, Section 5.1]. 361 SSRC: set as described in RFC1889 [5]. A mapping between the ES 362 identifiers (ESIDs) and SSRCs should be provided through out-of-band 363 means. 365 CC and CSRC fields are used as described in RFC 1889 [5]. 367 RTCP SHOULD be used as defined in RFC 1889 [5]. 369 RTP timestamps in RTCP SR packets: according to the RTP timing model, 370 the RTP timestamp that is carried into an RTCP SR packet is the same 371 as the CTS that would be applied to an RTP packet for data that was sampled 372 at the instant the SR packet is being generated and sent. The RTP 373 timestamp value is calculated from the NTP timestamp for the current 374 time which also goes in the RTCP SR packet. To perform that calculation, 375 an implementation needs to periodically establish a correspondence between 376 the CTS value of a data packet and the NTP time at which that data was 377 sampled. 379 4. Multiplexing 381 Since a typical MPEG-4 session may involve a large number of objects, 382 that may be as many as a few hundred, transporting each ES as an 383 individual RTP session may not always be practical. Allocating and 384 controlling hundreds of destination addresses for each MPEG-4 session 385 may pose insurmountable session administration problems. The 386 input/output processing overhead at the end-points will be extremely 387 high also. Additionally, low delay transmission of low bitrate data 388 streams, e.g. facial animation parameters, results in extremely high 389 header overheads. 391 To solve these problems, MPEG-4 data transport requires a multiplexing 392 scheme that allows selective bundling of several ESs. This is beyond the 393 scope of the payload format defined here. MPEG-4's Flexmux multiplexing 394 scheme may be used for this purpose by defining an additional RTP 395 payload format for "multiplexed MPEG-4 streams." On the other hand, 396 considering that many other payload types may have similar needs, a 397 better approach may be to develop a generic RTP multiplexing scheme 398 usable for MPEG-4 data. The generic multiplexing scheme reported in [7] 399 is a candidate for this approach. 401 For MPEG-4 applications, the multiplexing technique needs to address the 402 following requirements: 404 i. The ESs multiplexed in one stream can change frequently during 405 a session. Consequently, the coding type, individual packet size 406 and temporal relationships between the multiplexed data units must 407 be handled dynamically. 409 ii. The multiplexing scheme should have a mechanism to determine 410 the ES identifier (ES_ID) for each of the multiplexed packets. 411 ES_ID is not a part of the SL header. 413 iii. In general, an SL packet does not contain information about its 414 size. The multiplexing scheme should be able to delineate the 415 multiplexed packets whose lengths may vary from a few bytes to 416 close to the path-MTU. 418 5. Security Considerations 420 RTP packets using the payload format defined in this specification are 421 subject to the security considerations discussed in the RTP 422 specification [5]. This implies that confidentiality of the media 423 streams is achieved by encryption. Because the data compression used 424 with this payload format is applied end-to-end, encryption may be 425 performed on the compressed data so there is no conflict between the two 426 operations. 428 This payload type does not exhibit any significant non-uniformity in the 429 receiver side computational complexity for packet processing to cause a 430 potential denial-of-service threat. 432 6. References 434 [1] ISO/IEC 14496-1 FDIS MPEG-4 Systems November 1998 436 [2] ISO/IEC 14496-2 FDIS MPEG-4 Visual November 1998 438 [3] ISO/IEC 14496-3 FDIS MPEG-4 Audio November 1998 440 [4] ISO/IEC 14496-6 FDIS Delivery Multimedia Integration 441 Framework, November 1998. 443 [5] Schulzrinne, Casner, Frederick, Jacobson RTP: A 444 Transport Protocol for Real Time Applications RFC 1889, 445 Internet Engineering Task Force, January 1996. 447 [6] S. Bradner, Key words for use in RFCs to Indicate 448 Requirement Levels, RFC 2119, March 1997. 450 [7] M. Handley, "GeRM: Generic RTP Multiplexing," work 451 in progress, draft-ietf-avt-germ-00.txt, November 1998. 453 7. Authors' Addresses 455 M. Reha Civanlar 456 AT&T Labs - Research 457 100 Schultz Drive 458 Red Bank, NJ 07701 459 USA 460 e-mail: civanlar@research.att.com 462 Andrea Basso 463 AT&T Labs - Research 464 100 Schultz Drive 465 Red Bank, NJ 07701 466 USA 467 e-mail: basso@research.att.com 469 Stephen L. Casner 470 Cisco Systems, Inc. 471 170 West Tasman Drive 472 San Jose, CA 95134 473 USA 474 e-mail: casner@cisco.com 476 Carsten Herpel 477 THOMSON multimedia 478 Karl-Wiechert-Allee 74 479 30625 Hannover 480 Germany 481 e-mail: herpelc@thmulti.com 483 Colin Perkins 484 Department of Computer Science 485 University College London 486 Gower Street 487 London WC1E 6BT 488 United Kingdom 489 e-mail: C.Perkins@cs.ucl.ac.uk