idnits 2.17.1 draft-ietf-avt-rapid-rtp-sync-03.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** The document seems to lack a License Notice according IETF Trust Provisions of 28 Dec 2009, Section 6.b.i or Provisions of 12 Sep 2009 Section 6.b -- however, there's a paragraph with a matching beginning. Boilerplate error? (You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Feb 2009 rather than one of the newer Notices. See https://trustee.ietf.org/license-info/.) Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- == The 'Updates: ' line in the draft header should list only the _numbers_ of the RFCs which will be updated by this document (if approved); it should not include the word 'RFC' in the list. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (June 30, 2009) is 5407 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(See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) == Outdated reference: A later version (-19) exists of draft-ietf-avt-rtcpssm-18 ** Obsolete normative reference: RFC 5285 (ref. '6') (Obsoleted by RFC 8285) ** Obsolete normative reference: RFC 1305 (ref. '7') (Obsoleted by RFC 5905) == Outdated reference: A later version (-27) exists of draft-ietf-avt-rtp-svc-18 == Outdated reference: A later version (-07) exists of draft-ietf-avt-dtls-srtp-05 == Outdated reference: A later version (-22) exists of draft-zimmermann-avt-zrtp-13 -- Obsolete informational reference (is this intentional?): RFC 5117 (ref. '15') (Obsoleted by RFC 7667) Summary: 3 errors (**), 0 flaws (~~), 6 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group C. Perkins 3 Internet-Draft University of Glasgow 4 Updates: RFC3550 T. Schierl 5 (if approved) Fraunhofer HHI 6 Intended status: Standards Track June 30, 2009 7 Expires: January 1, 2010 9 Rapid Synchronisation of RTP Flows 10 draft-ietf-avt-rapid-rtp-sync-03.txt 12 Status of this Memo 14 This Internet-Draft is submitted to IETF in full conformance with the 15 provisions of BCP 78 and BCP 79. 17 Internet-Drafts are working documents of the Internet Engineering 18 Task Force (IETF), its areas, and its working groups. Note that 19 other groups may also distribute working documents as Internet- 20 Drafts. 22 Internet-Drafts are draft documents valid for a maximum of six months 23 and may be updated, replaced, or obsoleted by other documents at any 24 time. It is inappropriate to use Internet-Drafts as reference 25 material or to cite them other than as "work in progress." 27 The list of current Internet-Drafts can be accessed at 28 http://www.ietf.org/ietf/1id-abstracts.txt. 30 The list of Internet-Draft Shadow Directories can be accessed at 31 http://www.ietf.org/shadow.html. 33 This Internet-Draft will expire on January 1, 2010. 35 Copyright Notice 37 Copyright (c) 2009 IETF Trust and the persons identified as the 38 document authors. All rights reserved. 40 This document is subject to BCP 78 and the IETF Trust's Legal 41 Provisions Relating to IETF Documents in effect on the date of 42 publication of this document (http://trustee.ietf.org/license-info). 43 Please review these documents carefully, as they describe your rights 44 and restrictions with respect to this document. 46 Abstract 48 This memo outlines how RTP sessions are synchronised, and discusses 49 how rapidly such synchronisation can occur. We show that most RTP 50 sessions can be synchronised immediately, but that the use of video 51 switching multipoint conference units (MCUs) or large source specific 52 multicast (SSM) groups can greatly increase the synchronisation 53 delay. This increase in delay can be unacceptable to some 54 applications that use layered and/or multi-description codecs. 56 This memo introduces three mechanisms to reduce the synchronisation 57 delay for such sessions. First, it updates the RTP Control Protocol 58 (RTCP) timing rules to reduce the initial synchronisation delay for 59 SSM sessions. Second, a new feedback packet is defined for use with 60 the Extended RTP Profile for RTCP-based Feedback (RTP/AVPF), allowing 61 video switching MCUs to rapidly request resynchronisation. Finally, 62 new RTP header extensions are defined to allow rapid synchronisation 63 of late joiners, and guarantee correct timestamp based decoding order 64 recovery for layered codecs in the presence of clock skew. 66 Table of Contents 68 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 69 2. Synchronisation of RTP Flows . . . . . . . . . . . . . . . . . 5 70 2.1. Initial Synchronisation Delay . . . . . . . . . . . . . . 6 71 2.1.1. Unicast Sessions . . . . . . . . . . . . . . . . . . . 6 72 2.1.2. Source Specific Multicast (SSM) Sessions . . . . . . . 7 73 2.1.3. Any Source Multicast (ASM) Sessions . . . . . . . . . 8 74 2.1.4. Discussion . . . . . . . . . . . . . . . . . . . . . . 9 75 2.2. Synchronisation for Late Joiners . . . . . . . . . . . . . 10 76 3. Reducing RTP Synchronisation Delays . . . . . . . . . . . . . 10 77 3.1. Reduced Initial RTCP Interval for SSM Senders . . . . . . 11 78 3.2. Rapid Resynchronisation Request . . . . . . . . . . . . . 11 79 3.3. In-band Delivery of Synchronisation Metadata . . . . . . . 12 80 4. Application to Decoding Order Recovery in Layered Codecs . . . 14 81 4.1. Problem description . . . . . . . . . . . . . . . . . . . 14 82 4.2. In-band Synchronisation for Decoding Order Recovery . . . 15 83 4.3. Timestamp based decoding order recovery . . . . . . . . . 16 84 5. Security Considerations . . . . . . . . . . . . . . . . . . . 19 85 6. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 20 86 7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 20 87 8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 20 88 8.1. Normative References . . . . . . . . . . . . . . . . . . . 20 89 8.2. Informative References . . . . . . . . . . . . . . . . . . 21 90 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 22 92 1. Introduction 94 When using RTP to deliver multimedia content it's often necessary to 95 synchronise playout of audio and video components of a presentation. 96 This is achieved using information contained in RTP Control Protocol 97 (RTCP) Sender Report (SR) packets [1]. These are sent periodically, 98 and the components of a multimedia session cannot be synchronised 99 until sufficient RTCP SR packets have been received for each RTP flow 100 to allow the receiver to establish mappings between the media clock 101 used for each RTP flow, and the common (NTP-format) reference clock 102 used to establish synchronisation. 104 Recently, concern has been expressed that this synchronisation delay 105 is problematic for some applications, for example those using layered 106 or multi-description video coding. This memo reviews the operations 107 of RTP synchronisation, and describes the synchronisation delay that 108 can be expected. Three backwards compatible extensions to the basic 109 RTP synchronisation mechanism are proposed: 111 o The RTCP transmission timing rules are relaxed for SSM senders, to 112 reduce the initial synchronisation latency for large SSM groups. 113 See Section 3.1. 115 o An enhancement to the Extended RTP Profile for RTCP-based Feedback 116 (RTP/AVPF) [2] is defined to allow receivers to request additional 117 RTCP SR packets, providing the metadata needed to synchronise RTP 118 flows. This can reduce the synchronisation delay when joining 119 sessions with large RTCP reporting intervals, in the presence of 120 packet loss, or when video switching MCUs are employed. See 121 Section 3.2. 123 o Two RTP header extensions are defined, to deliver synchronisation 124 metadata in-band with RTP data packets. These extensions provide 125 synchronisation metadata that is aligned with RTP data packets, 126 and so eliminate the need to estimate clock-skew between flows 127 before synchronisation. They can also reduce the need to receive 128 RTCP SR packets before flows can be synchronising, although it 129 does not eliminate the need for RTCP. See Section 3.3. 131 The immediate use-case for these extensions is to reduce the delay 132 due to synchronisation when joining a layered video session (e.g. an 133 H.264/SVC session in NI-T mode [9]). The extensions are not specific 134 to layered coding, however, and can be used in any environment when 135 synchronisation latency is an issue. 137 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 138 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 139 document are to be interpreted as described in RFC 2119 [3]. 141 2. Synchronisation of RTP Flows 143 RTP flows are synchronised by receivers based on information that is 144 contained in RTCP SR packets generated by senders (specifically, the 145 NTP-format timestamp and the RTP timestamp). Synchronisation 146 requires that a common reference clock MUST be used to generate the 147 NTP-format timestamps in a set of flows that are to be synchronised. 148 Furthermore, to achieve more rapid and accurate synchronisation, it 149 is RECOMMENDED that senders and receivers use a common reference 150 clock where possible (recognising that this is often not possible 151 when RTP is used outside of controlled environments); the means by 152 which that common reference clock is distributed are outside the 153 scope of this memo. 155 For multimedia sessions, each type of media (e.g. audio or video) is 156 sent in a separate RTP session, and the receiver associates RTP flows 157 to be synchronised by means of the canonical end-point identifier 158 (CNAME) item included in the RTCP Source Description (SDES) packets 159 generated by the sender or signalled out of band [10]. For layered 160 media, different layers can be sent in different RTP sessions, or 161 using different SSRC values within a single RTP session; in both 162 cases, the CNAME is used to identify flows to be synchronised. To 163 ensure synchronisation, an RTP sender MUST therefore send periodic 164 compound RTCP packets following Section 6 of RFC 3550 [1]. 166 The timing of these periodic compound RTCP packets will depend on the 167 number of members in each RTP session, the fraction of those that are 168 sending data, the session bandwidth, the configured RTCP bandwidth 169 fraction, and whether the session is multicast or unicast (see RFC 170 3550 Section 6.2 for details). In summary, RTCP control traffic is 171 allocated a small fraction, generally 5%, of the session bandwidth, 172 and of that fraction, one quarter is allocated to active RTP senders, 173 while receivers use the remaining three quarters (these fractions can 174 be configured via SDP [11]). Each member of an RTP session derives 175 an RTCP reporting interval based on these fractions, whether the 176 session is multicast or unicast, the number of members it has 177 observed, and whether it is actively sending data or not. It then 178 sends a compound RTCP packet on average once per reporting interval 179 (the actual packet transmission time is randomised in the range [0.5 180 ... 1.5] times the reporting interval to avoid synchronisation of 181 reports). 183 A minimum reporting interval of 5 seconds is RECOMMENDED, except that 184 the delay before sending the initial report "MAY be set to half the 185 minimum interval to allow quicker notification that the new 186 participant is present" [1]. Also, for unicast sessions, "the delay 187 before sending the initial compound RTCP packet MAY be zero" [1]. In 188 addition, for unicast sessions, and for active senders in a multicast 189 session, the fixed minimum reporting interval MAY be scaled to "360 190 divided by the session bandwidth in kilobits/second. This minimum is 191 smaller than 5 seconds for bandwidths greater than 72 kb/s." [1] 193 2.1. Initial Synchronisation Delay 195 A multimedia session comprises a set of concurrent RTP sessions among 196 a common group of participants, using one RTP session for each media 197 type. For example, a videoconference (which is a multimedia session) 198 might contain an audio RTP session and a video RTP session. To allow 199 a receiver to synchronise the components of a multimedia session, a 200 compound RTCP packet containing an RTCP SR packet and an RTCP SDES 201 packet with a CNAME item MUST be sent to each of the RTP sessions in 202 the multimedia session. A receiver cannot synchronise playout across 203 the multimedia session until such RTCP packets have been received on 204 all of the component RTP sessions. If there is no packet loss, this 205 gives an expected initial synchronisation delay equal to the average 206 time taken to receive the first RTCP packet in the RTP session with 207 the longest RTCP reporting interval. This will vary between unicast 208 and multicast RTP sessions. 210 The initial synchronisation delay for layered sessions is similar to 211 that for multimedia sessions. The layers cannot be synchronised 212 until the RTCP SR and CNAME information has been received for each 213 layer in the session. 215 2.1.1. Unicast Sessions 217 For unicast multimedia or layered sessions, senders SHOULD transmit 218 an initial compound RTCP packet (containing an RTCP SR packet and an 219 RTCP SDES packet with a CNAME item) immediately on joining each RTP 220 session in the multimedia session. The individual RTP sessions are 221 considered to be joined once any in-band signalling for NAT traversal 222 (e.g. [12]) and/or security keying (e.g. [13],[14]) has concluded, 223 and the media path is open. This implies that the initial RTCP 224 packet is sent in parallel with the first data packet following the 225 guidance in RFC 3550 that "the delay before sending the initial 226 compound RTCP packet MAY be zero" and, in the absence of any packet 227 loss, flows can be synchronised immediately. 229 Note that NAT pinholes, firewall holes, quality-of-service, and media 230 security keys should have been negotiated as part of the signalling, 231 whether in-band or out-of-band, before the first RTCP packet is sent. 232 This should ensure that any middleboxes are ready to accept traffic, 233 and reduce the likelihood that the initial RTCP packet will be lost. 235 2.1.2. Source Specific Multicast (SSM) Sessions 237 For multicast sessions, the delay before sending the initial RTCP 238 packet, and hence the synchronisation delay, varies with the session 239 bandwidth and the number of members in the session. For a multicast 240 multimedia or layered session, the average synchronisation delay will 241 depend on the slowest of the component RTP sessions; this will 242 generally be the session with the lowest bandwidth (assuming all the 243 RTP sessions have the same number of members). 245 When sending to a multicast group, the reduced minimum RTCP reporting 246 interval of 360 seconds divided by the session bandwidth in kilobits 247 per second [1] should be used when synchronisation latency is likely 248 to be an issue. Also, as usual, the reporting interval is halved for 249 the first RTCP packet. Depending on the session bandwidth and the 250 number of members, this gives the average synchronisation delays 251 shown in Figure 1. 253 Session| Number of receivers: 254 Bandwidth| 2 3 4 5 10 100 1000 10000 255 --+------------------------------------------------ 256 8 kbps| 2.73 4.10 5.47 5.47 5.47 5.47 5.47 5.47 257 16 kbps| 2.50 2.50 2.73 2.73 2.73 2.73 2.73 2.73 258 32 kbps| 2.50 2.50 2.50 2.50 2.50 2.50 2.50 2.50 259 64 kbps| 2.50 2.50 2.50 2.50 2.50 2.50 2.50 2.50 260 128 kbps| 1.41 1.41 1.41 1.41 1.41 1.41 1.41 1.41 261 256 kbps| 0.70 0.70 0.70 0.70 0.70 0.70 0.70 0.70 262 512 kbps| 0.35 0.35 0.35 0.35 0.35 0.35 0.35 0.35 263 1 Mbps| 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 264 2 Mbps| 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.09 265 4 Mbps| 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 267 Figure 1: Average RTCP reporting interval in seconds for an RTP 268 Session with 1 sender. 270 These numbers assume a source specific multicast channel with a 271 single active sender, which the rules in RFC 3550 section 6.3 give a 272 fixed fraction of the RTCP bandwidth irrespective of the number of 273 receivers. It can be seen that they are sufficient for lip- 274 synchronisation without excessive delay, but might be viewed as 275 having too much latency for synchronising parts of a layered video 276 stream. 278 The RTCP interval is randomised in the usual manner, so the minimum 279 synchronisation delay will be half these intervals, and the maximum 280 delay will be 1.5 times these intervals. Note also that these RTCP 281 intervals are calculated assuming perfect knowledge of the number of 282 members in the session. 284 2.1.3. Any Source Multicast (ASM) Sessions 286 For ASM sessions, the fraction of members that are senders plays an 287 important role, and causes more variation in average RTCP reporting 288 interval. This is illustrated in Figure 2 and Figure 3, which show 289 the RTCP reporting interval for the same session bandwidths and 290 receiver populations as the SSM session described in Figure 1, but 291 for sessions with 2 and 10 senders respectively. It can be seen that 292 the initial synchronisation delay scales with the number of senders 293 (this is to ensure that the total RTCP traffic from all group members 294 does not grow without bound) and can be significantly larger than for 295 single source groups. Despite this, the initial synchronisation time 296 remains acceptable for lip-synchronisation in typical small-to-medium 297 sized group conferencing scenarios. 299 Session| Number of receivers: 300 Bandwidth| 2 3 4 5 10 100 1000 10000 301 --+------------------------------------------------ 302 8 kbps| 2.73 4.10 5.47 6.84 10.94 10.94 10.94 10.94 303 16 kbps| 2.50 2.50 2.73 3.42 5.47 5.47 5.47 5.47 304 32 kbps| 2.50 2.50 2.50 2.50 2.73 2.73 2.73 2.73 305 64 kbps| 2.50 2.50 2.50 2.50 2.50 2.50 2.50 2.50 306 128 kbps| 1.41 1.41 1.41 1.41 1.41 1.41 1.41 1.41 307 256 kbps| 0.70 0.70 0.70 0.70 0.70 0.70 0.70 0.70 308 512 kbps| 0.35 0.35 0.35 0.35 0.35 0.35 0.35 0.35 309 1 Mbps| 0.18 0.18 0.18 0.18 0.18 0.18 0.18 0.18 310 2 Mbps| 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.09 311 4 Mbps| 0.04 0.04 0.04 0.04 0.04 0.04 0.04 0.04 313 Figure 2: Average RTCP reporting interval in seconds for an RTP 314 Session with 2 senders. 316 Session| Number of receivers: 317 Bandwidth| 2 3 4 5 10 100 1000 10000 318 --+------------------------------------------------ 319 8 kbps| 2.73 4.10 5.47 6.84 13.67 54.69 54.69 54.69 320 16 kbps| 2.50 2.50 2.73 3.42 6.84 27.34 27.34 27.34 321 32 kbps| 2.50 2.50 2.50 2.50 3.42 13.67 13.67 13.67 322 64 kbps| 2.50 2.50 2.50 2.50 2.50 6.84 6.84 6.84 323 128 kbps| 1.41 1.41 1.41 1.41 1.41 3.42 3.42 3.42 324 256 kbps| 0.70 0.70 0.70 0.70 0.70 1.71 1.71 1.71 325 512 kbps| 0.35 0.35 0.35 0.35 0.35 0.85 0.85 0.85 326 1 Mbps| 0.18 0.18 0.18 0.18 0.18 0.43 0.43 0.43 327 2 Mbps| 0.09 0.09 0.09 0.09 0.09 0.21 0.21 0.21 328 4 Mbps| 0.04 0.04 0.04 0.04 0.04 0.11 0.11 0.11 330 Figure 3: Average RTCP reporting interval in seconds for an RTP 331 Session with 10 senders. 333 Note that multi-sender groups implemented using multi-unicast with a 334 central RTP translator (Topo-Translator in the terminology of [15]) 335 or mixer (Topo-Mixer), or some forms of video switching MCU (Topo- 336 Video-switch-MCU) distribute RTCP packets to all members of the 337 group, and so scale in the same way as an ASM group with regards to 338 initial synchronisation latency. 340 2.1.4. Discussion 342 For unicast sessions, the existing RTCP SR-based mechanism allows for 343 immediate synchronisation, provided the initial RTCP packet is not 344 lost. 346 For SSM sessions, the initial synchronisation delay is sufficient for 347 lip-synchronisation, but may be larger than desired for some layered 348 codecs. The rationale for not sending immediate RTCP packets for 349 multicast groups is to avoid implosion of requests when large numbers 350 of members simultaneously join the group ("flash crowd"). This is 351 not an issue for SSM senders, since there can be at most one sender, 352 so it is desirable to allow SSM senders to send an immediate RTCP SR 353 on joining a session (as is currently allowed for unicast sessions, 354 which also don't suffer from the implosion problem). SSM receivers 355 using unicast feedback would not be allowed to send immediate RTCP. 356 For ASM sessions, implosion of responses is a concern, so no change 357 is proposed to the RTCP timing rules. 359 In all cases, it is possible that the initial RTCP SR packet is lost. 360 In this case, the receiver will not be able to synchronise the media 361 until the reporting interval has passed, and the next RTCP SR packet 362 is sent. This is undesirable. Section 3.2 defines a new RTP/AVPF 363 transport layer feedback message to request an RTCP SR be generated, 364 allowing rapid resynchronisation in the case of packet loss. 366 2.2. Synchronisation for Late Joiners 368 Synchronisation between RTP sessions is potentially slower for late 369 joiners than for participants present at the start of the session. 370 The reasons for this are two-fold: 372 1. Many of the optimisations that allow rapid transmission of RTCP 373 SR packets apply only at the start of a session. This implies 374 that a new participant may have to wait a complete RTCP reporting 375 interval for each session before receiving the necessary data to 376 synchronise media streams. This might potentially take several 377 seconds, depending on the configured session bandwidth and the 378 number of participants. 380 2. Additional synchronisation delay comes from the nature of the 381 RTCP timing rules. Packets are generated on average once per 382 reporting interval, but with the exact transmission times being 383 randomised +/- 50% to avoid synchronisation of reports. This is 384 important to avoid network congestion in multicast sessions, but 385 does mean that the timing of RTCP SR reports for different RTP 386 sessions isn't synchronised. Accordingly, a receiver must 387 estimate the skew on the NTP-format clock in order to align RTP 388 timestamps across sessions. This estimation is an essential part 389 of an RTP synchronisation implementation, and can be done with 390 high accuracy given sufficient reports. Collecting sufficient 391 RTCP SR data to perform this estimation, however, may require 392 reception of several RTCP reports, further increasing the 393 synchronisation delay. 395 3. Many media codecs have the notion of periodic access points, such 396 that a newly joined receiver often cannot start decoding a media 397 stream until the packets corresponding to the access point have 398 been received. These access points may be sent less often than 399 RTCP SR packets, and so may be the limiting factor in starting 400 synchronised media playout for late joiners. 402 These delays are likely an issue for tuning in to an ongoing 403 multicast RTP session, or for video switching MCUs. 405 3. Reducing RTP Synchronisation Delays 407 Three backwards compatible RTP extensions are defined to reduce the 408 possible synchronisation delay: a reduced initial RTCP interval for 409 SSM senders, a rapid resynchronisation request message, and RTP 410 header extensions that can convey synchronisation metadata in-band. 412 3.1. Reduced Initial RTCP Interval for SSM Senders 414 In SSM sessions where the initial synchronisation delay is important, 415 the RTP sender MAY set the delay before sending the initial compound 416 RTCP packet to zero, and send its first RTCP packet immediately upon 417 joining the SSM session. RTP receivers in an SSM session, sending 418 unicast RTCP feedback, MUST NOT send RTCP packets with zero initial 419 delay; the timing rules defined in [4] apply unchanged to receivers. 421 3.2. Rapid Resynchronisation Request 423 The general format of an RTP/AVPF transport layer feedback message is 424 shown in Figure 4 (see [2] for details). 426 0 1 2 3 427 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 1 428 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 429 |V=2|P| FMT | PT=RTPFB=205 | length | 430 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 431 | SSRC of packet sender | 432 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 433 | SSRC of media source | 434 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 435 : Feedback Control Information (FCI) : 436 : : 438 Figure 4: RTP/AVP Transport Layer Feedback Message 440 A new feedback message type, RTCP-SR-REQ, is defined with FMT = 5. 441 The Feedback Control Information (FCI) part of the feedback message 442 MUST be empty. The SSRC of packet sender indicates the member that 443 is unable to synchronise media streams, while the SSRC of media 444 source indicates the sender of the media it is unable to synchronise. 445 The length MUST equal 2. 447 This feedback message MAY be sent by a receiver to indicate that it's 448 unable to synchronise some media streams, and desires that the media 449 source transmit an RTCP SR packet as soon as possible (within the 450 constraints of the RTCP timing rules for early feedback). When it 451 receives such an indication, the media source SHOULD generate an RTCP 452 SR packet as soon as possible within the RTCP early feedback rules. 453 If the use of non-compound RTCP [5] was previously negotiated, both 454 the feedback request and the RTCP SR response may be sent as non- 455 compound RTCP packets. The RTCP-SR-REQ packet MAY be repeated once 456 per RTCP reporting interval if no RTCP SR packet is forthcoming. 458 When using SSM sessions with unicast feedback, is possible that the 459 feedback target and media source are not co-located. If a feedback 460 target receives an RTCP-SR-REQ feedback message in such a case, the 461 request should be forwarded to the media source. The mechanism to be 462 used for forwarding such requests is not defined here. 464 3.3. In-band Delivery of Synchronisation Metadata 466 The RTP header extension mechanism defined in [6] can be adopted to 467 carry an OPTIONAL NTP format timestamp in RTP data packets. If such 468 a timestamp is included, it MUST correspond to the same time instant 469 as the RTP timestamp in the packet's header, and MUST be derived from 470 the same clock used to generate the NTP format timestamps included in 471 RTCP SR packets. Provided it has knowledge of the SSRC to CNAME 472 mapping, either from prior receipt of an RTCP CNAME packet or via 473 out-of-band signalling [10], the receiver can use the information 474 provided as input to the synchronisation algorithm, in exactly the 475 same way as if an additional RTCP SR packet was been received for the 476 flow. 478 Two variants are defined for this header extension. The first 479 variant extends the RTP header with a 64 bit NTP timestamp format 480 timestamp as defined in [7]. The second variant carries the lower 24 481 bit part of the Seconds of a NTP timestamp format timestamp and the 482 32 bit of the Fraction of a NTP timestamp format timestamp. The 483 formats of the two variants are shown below. 485 Variant A/64-bit NTP RTP header extension (length: 16 bytes): 487 0 1 2 3 488 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 1 489 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 490 |V=2|P|1| CC |M| PT | sequence number | 491 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+R 492 | timestamp |T 493 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+P 494 | synchronization source (SSRC) identifier | 495 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 496 | 0xBE | 0xDE | length=3 | 497 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+E 498 | ID-A | L=7 | NTP timestamp format - Seconds (bit 0-23) |x 499 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+t 500 |NTP Sec.(24-31)| NTP timestamp format - Fraction(bit 0-23) |n 501 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 502 |NTP Frc.(24-31)| 0 (pad) | 0 (pad) | 0 (pad) | 503 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 504 | payload data | 505 | .... | 506 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 508 Variant B/56-bit NTP RTP header extension (length: 12 bytes): 510 0 1 2 3 511 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 1 512 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 513 |V=2|P|1| CC |M| PT | sequence number | 514 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+R 515 | timestamp |T 516 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+P 517 | synchronization source (SSRC) identifier | 518 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 519 | 0xBE | 0xDE | length=2 | 520 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+E 521 | ID-B | L=6 | NTP timestamp format - Seconds (bit 8-31) |x 522 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+t 523 | NTP timestamp format - Fraction (bit 0-31) |n 524 +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ 525 | payload data | 526 | .... | 527 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 529 An NTP timestamp format timestamp MAY be included on any RTP packets 530 the sender chooses, but it is RECOMMENDED when performing timestamp 531 based decoding order recovery for layered codecs transported in 532 multiple RTP flows, as further specified in Section 4.2. This header 533 extension SHOULD be also sent on the RTP packets corresponding to a 534 video random access point, and on the associated audio packets, to 535 allow rapid synchronisation for late joiners in multimedia sessions, 536 and in video switching scenarios. 538 Note: The inclusion of an RTP header extension will reduce the 539 efficiency of RTP header compression, if it is used. Furthermore, 540 middle boxes which do not understand the header extensions may remove 541 them or may not update the content according to this memo. 543 In all cases, irrespective of whether in-band NTP timestamp format 544 timestamps are included or not, regular RTCP SR packets MUST be sent 545 to provide backwards compatibility with receivers that synchronize 546 RTP flows according to [1], and robustness in the face of middleboxes 547 (RTP translators) that might strip RTP header extensions. The sender 548 reports are also required to receive the upper 8 bit of the Seconds 549 of the NTP timestamp format timestamp not included in the Variant 550 B/56-bit NTP RTP header extension (although this may generally be 551 inferred from context). 553 When the SDP is used, the use of the RTP header extensions defined 554 above MUST be indicated as specified in [6]. Therefore the following 555 URIs MUST be used: 557 o The URI used for signaling the use of Variant A/64-bit NTP RTP 558 header extension in SDP is "urn:ietf:params:rtp-hdrext:ntp-64". 560 o The URI used for signaling the use of Variant B/56-bit NTP RTP 561 header extension in SDP is "urn:ietf:params:rtp-hdrext:ntp-56". 563 4. Application to Decoding Order Recovery in Layered Codecs 565 Based on the timestamp contained in each RTP data packet, and the 566 mapping to an NTP format timestamp, a decoding order recovery process 567 may be applied if a media as result of a layered coding process is 568 transported in multiple RTP flows. This recovers the decoding order 569 of media frames or samples at the receiver. Especially when 570 transporting layered video, the decoding order recovery process is 571 not straight forward. In this section, we provide guidance on how to 572 use RTP/NTP timing information for decoding order recovery. 574 4.1. Problem description 576 One option for decoding order recovery in layered codecs is to use 577 the NTP/sample presentation timestamps to reorder data of the same 578 layered media transported in multiple RTP flows. For a timestamp- 579 based decoding order recovery process, it is crucial to allow exact 580 alignment of media frames respectively samples using the NTP timing 581 information. 583 In the presence of clock skew in NTP-format clock, it may not be 584 possible to derive exact matching NTP timestamps using the NTP format 585 clock in each RTP flow's RTCP sender reports. This is due to the 586 fact that RTCP sender reports are not sent at the same point of time 587 in the multiple RTP flows transporting data of the same layered 588 media, while having a skew between those samples in the RTP flows 589 RTCP sender reports. If the RTCP SR packets are not send 590 synchronously in the multiple RTP flows, they therefore do not 591 contain the same NTP-format timestamp. If there is a skew present in 592 the clock used for NTP-format timestamp generation, using different 593 NTP-format timestamps for the same sampling instance in the RTP flow 594 inevitably leads to non-matching NTP timestamps generated from RTP 595 timestamps and NTP-format timestamps in the multiple RTP flows. In 596 order to allow a common and straight forward timestamp-based decoding 597 order recovery process, it is important to guarantee exact matching 598 of NTP timestamps. Thus in the presence of non-perfect clocks, which 599 should be the normal case, an additional mechanism SHALL be used. An 600 exact inter-flow alignment of NTP timestamps can be guaranteed, if an 601 RTP header extension containing an NTP timestamp is always inserted 602 at the same timing position in all the RTP flows in question, and if 603 those NTP header extensions are used to update the NTP-RTP relation 604 in all RTP flows at the same point of time. This is called 605 synchronous insertion of RTP header extensions in the following. 607 4.2. In-band Synchronisation for Decoding Order Recovery 609 The RTP header extension to convey an NTP timestamp SHOULD be used 610 with a layered, multi-description, or multi-view codec, to provide 611 exact matching of NTP timestamps between layers, descriptions, or 612 views transported in different RTP flows to allow timestamp-based 613 decoding order recovery. If this header extension is inserted for 614 RTP flows transporting samples or parts of samples of the same 615 layered media, it SHALL be included at least once in each of the RTP 616 flows of the same media for the sampling time instance of an 617 insertion of an RTP header extension. Such synchronously inserted 618 RTP header extensions SHALL contain the same NTP format timestamp. 619 The frequency of inserting the header extensions in the RTP flows is 620 up to the sender, but it should be noticed that higher insertion 621 frequencies obviously lead to higher synchronization frequencies. 622 For use cases where the same clock source has been used to generate 623 the RTP timestamps in the multiple RTP flows, an application MAY rely 624 on the RTP timestamps only for decoding order recovery starting from 625 the point of synchronous insertion of the RTP header extensions 626 containing NTP timestamps. 628 Note: If the decoding order of RTP flows is given by any means (as 629 e.g., for RTP session by mechanism defined in [8]), the NTP timestamp 630 provided by the header extension allows to collect data of the same 631 sample from the RTP flows, forming the sample decoding order. There 632 may be future mechanism to allow indication of dependencies of RTP 633 flows transported as RTP streams using SSRC multiplexing 635 It is RECOMMENDED that the receiver uses for timestamp-based decoding 636 order recovery the NTP timestamps provided in the RTP header 637 extensions only, if such extensions are present for the RTP flows. 638 Section 4.3 gives further details about the timestamp-based decoding 639 order recovery. 641 Note: Using the RTP header extensions described above allows the 642 receiver to find the corresponding sample of the layered media, or 643 parts thereof, in all RTP flows at the instant the RTP header 644 extension is inserted into the flows. This guarantees that any clock 645 skew present in the NTP timestamp generation process based on RTCP 646 sender reports is avoided, and so allows direct comparison of NTP 647 timestamps across multiple RTP flows. Furthermore, this approach 648 solves the possible problem of clock skews identified for the NI-T 649 mode as defined in [9]. To ensure the absence of clock skew, a 650 header extension containing the NTP timestamp MUST be inserted into 651 the RTP flows comprising a layered media stream at the same instant 652 in each RTP flow. This may require the insertion of extra packets in 653 some of the RTP flows, since in layered video codecs not all sampling 654 instances may be present in all the flows. If such a header 655 extension is included in all flows at a sampling time instance, the 656 NTP timestamps for samples following in decoding order the RTP header 657 insertion point can be constructed using the RTP timestamps and 658 identical reference NTP timestamps in the NTP header extension in all 659 RTP flows. It should be noted that the frequency of inserting the 660 RTP header extension containing the NTP timestamp is crucial in 661 presence of clock skew, since the points of insertion may be the only 662 points for a receiver to start the decoding order recovery. 664 4.3. Timestamp based decoding order recovery 666 If parts or complete samples as result of a layered coding process 667 are transported as different RTP flows, i.e. as different RTP 668 streams, and/or as different RTP sessions, a decoding order recovery 669 process is required to reorder the samples or parts of samples 670 received. Such mechanism may be based on the NTP presentation 671 timestamp which can be derived from the RTP timestamp using the NTP- 672 format timestamp provided in the RTCP sender report packets. 674 In order to guarantee the exact alignment of those derived NTP 675 presentation timestamps, the RTP header extension as defined in this 676 memo in Section 3.3 allows the receiver to start the decoding order 677 recovery before the reception of a RTCP sender report if the RTP 678 header extension is earlier provided in the RTP flow. Using the RTP 679 header extensions may be the only way to allow correct decoding order 680 recovery based on exact matching of NTP timestamps in the presence of 681 clock skew in the clock used for generating the NTP format clock. 683 Furthermore, some use cases may allow to use synchronously inserted 684 RTP header extensions containing NTP timestamps to align the RTP 685 timestamps of the multiple RTP flows, i.e. use cases where the RTP 686 timestamps of the multiple RTP flows are generated from the same 687 clock source. In such use cases, starting from a synchronous 688 insertion of the RTP header extensions, the application may use the 689 detected difference of RTP random offset values in the multiple 690 sessions to align the media samples of parts of samples. 692 Since typically for layered video codecs as, e.g. SVC [9], the 693 decoding order is not equal to the presentation order of the media 694 samples, media samples or parts of media samples cannot be simply 695 ordered according to the presentation timestamp order. For this 696 reason, if transporting media samples or parts of media samples of a 697 layered, multi-view or multi description codec in different RTP 698 flows, the following rules SHOULD be kept for sending such flows: 700 Note: The following rules are typically kept for layered audio 701 codecs, which allows using the same algorithm for decoding order 702 recovery of audio samples. 704 Terminology: Following the decoding order of RTP flows as described 705 above, an RTP flow containing sample data which is required to be 706 accessed and/or decoded before decoding a second sample data of 707 another RTP flow is called a lower RTP flow with respect to the 708 second RTP flow. A second RTP flow, which requires for the decoding 709 process accessing and/or decoding the sample data of the lower RTP 710 flow is called the higher RTP flow. The lowest RTP flow is the flow, 711 which does not require the presence of any other data. 713 o The decoding order of media samples or part of the media samples 714 transported in different RTP flows SHOULD be derivable by any 715 means. This can be accomplished, e.g. by using the mechanisms 716 defined in [8] if the sample data or parts of the sample data are 717 transported in different RTP sessions or by any other means. 719 o For each two RTP flows the following rules SHOULD be true in order 720 to allow decoding order recovery based on matching NTP timestamps 721 present in the different RTP flows: 723 1. The order of the RTP samples within an RTP flow is equal to 724 the decoding order. 726 2. A higher RTP flow contains all the same sampling instances of 727 the lower RTP flow. A higher RTP flow may contain additional 728 sampling instances. 730 Note: In some cases, it may be required to add packets in higher RTP 731 flows in order to satisfy the second rule above. This may be 732 achieved by placing empty RTP packets (containing padding data only) 733 or by other payload means as, e.g. the Empty NAL unit packet as 734 defined in [9]. 736 If a packet must be inserted for satisfying the above rule, the NTP 737 timestamp of such an inserted packet MUST be set equal to the NTP 738 timestamp of a packet of the same sample present in any lower RTP 739 flow and the lowest RTP flow. This is easy to accomplish if the 740 packet can be inserted at the time of the RTP flow generation, since 741 the NTP timestamp must be the same for the inserted packet and the 742 packet of the corresponding sample. 744 The above rules allow the receiver to process the data of the RTP 745 flows as follows: 747 o Go through all received RTP flows starting with the highest RTP 748 flow and aggregate the sample data or parts of the sample data 749 with the same NTP timestamp in the order of RTP flows, starting 750 from the lowest RTP flow up to the highest RTP flow received, to 751 the sample with the NTP timestamp present in the highest RTP flow. 752 The NTP timestamps MAY be derived using RTCP sender reports or MAY 753 be directly taken from the NTP timestamp provided in an RTP header 754 extension. The order of RTP flows may e.g. be indicated by 755 mechanisms as defined in [8] or any other implicit or explicit 756 means. Repeat the aforementioned process for each different NTP 757 timestamp present in the highest RTP flow. 759 Informative example: The example shown in Figure 3 refers to three 760 RTP flows A, B and C containing a layered, a multi-view or a multi- 761 description media stream. In the example, the dependency signalling 762 as defined in [8] indicates that flow A is the lowest RTP flow, B is 763 the first higher RTP flow and depends on A, and C is the second 764 higher RTP flow corresponding to flow A and depends on A and B. A 765 media coding structure is used that results in samples present in 766 higher flows but not present in all lower flows. Flow A has the 767 lowest frame rate and Flow B and C have the same but higher frame 768 rate. The figure shows the full video samples with their 769 corresponding RTP timestamps "(x)". The video samples are already 770 re-ordered according to their RTP sequence number order. The figure 771 indicates for the received sample in decoding order within each RTP 772 flow, as well as the associated NTP media timestamps ("TS[..]"). 773 These timestamps may be derived using the NTP format timestamp 774 provided in the RTCP sender reports or as shown in the figure 775 directly from the NTP timestamp contained in the RTP header 776 extensions as indicate by the timestamp in "". Note that the 777 timestamps are not in increasing order since, in this example, the 778 decoding order is different from the output/presentation order. 780 The process first proceeds to the sample parts associated with the 781 first available synchronous insertion of NTP timestamp into RTP 782 header extensions at NTP media timestamp TS=[8] and starts in the 783 highest RTP flow C and removes/ignores all preceding sample parts (in 784 decoding order) to sample parts with TS=[8] in each of the de- 785 jittering buffers of RTP flows A, B, and C. Then, starting from flow 786 C, the first media timestamp available in decoding order (TS=[8]) is 787 selected and sample parts starting from RTP flow A, and flow B and C 788 are placed in order of the RTP flow dependency as indicated by 789 mechanisms defined in [8] (in the example for TS[8]: first flow B and 790 then flow C into the video sample VS(TS[8]) associated with NTP media 791 timestamp TS=[8]. Then the next media timestamp TS=[6] (RTP 792 timestamp=(4)) in order of appearance in the highest RTP flow C is 793 processed and the process described above is repeated. Note that 794 there may be video samples with no sample parts present, e.g., in the 795 lowest RTP flow A (see, e.g., TS=[5]). The decoding order recovery 796 process could be also started after receiving all RTP sender reports 797 RTP timestamp to NTP-format timestamp mapping (indicated as 798 timestamps "(x){y}") assuming that there is no clock skew in the 799 source used for the NTP-format timestamp generation. 801 C:-(0)----(2)----(7)<8>--(5)----(4)----(6)-----(11)----(9){10}- 802 | | | | | | | | 803 B:-(3)----(5)---(10)<8>--(8)----(7)----(9){7}--(14)----(12)---- 804 | | | | 805 A:---------------(3)<8>--(1)-------------------(7){12}-(5)----- 807 ---------------------------------------decoding/transmission order-> 808 TS:[1] [3] [8]=<8> [6] [5] [7] [12] [10] 810 Key: 811 A, B, C - RTP flows 812 Integer values in "()"- video sample with its RTP timestamp as 813 indicated in its RTP packet. 814 "|" - indicates corresponding samples / parts of 815 sample of the same video sample VS(TS[..]) 816 in the RTP flows. 817 Integer values in "[]"- NTP media timestamp TS, sampling time 818 as derived from the NTP timestamp associated 819 with the video sample AU(TS[..]), consisting 820 of sample parts in the flows above. 821 Integer values in "<>"- NTP media timestamp TS as directly 822 taken from the NTP RTP header extensions. 823 Integer values in "{}"- NTP media timestamp TS as provided in the 824 RTCP sender reports. 826 5. Security Considerations 828 The security considerations of the RTP specification [1], the 829 Extended RTP profile for RTCP-Based Feedback [2], and the General 830 Mechanism for RTP Header Extensions [6] apply. The extensions we 831 define in this memo are not believed to introduce any additional 832 security considerations. 834 6. IANA Considerations 836 NOTE TO RFC EDITOR: Please replace "RFC XXXX" in the following with 837 the RFC number assigned to this memo, and delete this note. 839 The IANA is requested to register one new value in the table of FMT 840 Values for RTPFB Payload Types [2] as follows: 842 Name: RTCP-SR-REQ 843 Long name: RTCP Rapid Resynchronisation Request 844 Value: 5 845 Reference: RFC XXXX 847 The IANA is also requested to register two new RTP Compact Header 848 Extensions [6], according to the following: 850 Extension URI: urn:ietf:params:rtp-hdrext:ntp-64 851 Description: Synchronisation metadata: 64-bit timestamp format 852 Contact: Thomas Schierl 853 IETF Audio/Video Transport Working Group 854 Reference: RFC XXXX 856 Extension URI: urn:ietf:params:rtp-hdrext:ntp-56 857 Description: Synchronisation metadata: 56-bit timestamp format 858 Contact: Thomas Schierl 859 IETF Audio/Video Transport Working Group 860 Reference: RFC XXXX 862 7. Acknowledgements 864 This memo has benefitted from discussions with numerous members of 865 the IETF AVT working group, including Jonathan Lennox, Magnus 866 Westerlund, Randell Jesup, Gerard Babonneau, Ingemar Johansson, Ali 867 C. Begen, Ye-Kui Wang, Roni Even, Michael Dolan, and Art Allison. 868 The header extension format of Variant A in Section 3.3 was suggested 869 by Dave Singer, matching a similar mechanism specified by ISMA. 871 8. References 873 8.1. Normative References 875 [1] Schulzrinne, H., Casner, S., Frederick, R., and V. Jacobson, 876 "RTP: A Transport Protocol for Real-Time Applications", STD 64, 877 RFC 3550, July 2003. 879 [2] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey, 880 "Extended RTP Profile for Real-time Transport Control Protocol 881 (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585, July 2006. 883 [3] Bradner, S., "Key words for use in RFCs to Indicate Requirement 884 Levels", BCP 14, RFC 2119, March 1997. 886 [4] Schooler, E., Ott, J., and J. Chesterfield, "RTCP Extensions 887 for Single-Source Multicast Sessions with Unicast Feedback", 888 draft-ietf-avt-rtcpssm-18 (work in progress), March 2009. 890 [5] Johansson, I. and M. Westerlund, "Support for Reduced-Size 891 Real-Time Transport Control Protocol (RTCP): Opportunities and 892 Consequences", RFC 5506, April 2009. 894 [6] Singer, D. and H. Desineni, "A General Mechanism for RTP Header 895 Extensions", RFC 5285, July 2008. 897 [7] Mills, D., "Network Time Protocol (Version 3) Specification, 898 Implementation", RFC 1305, March 1992. 900 [8] Schierl, T. and S. Wenger, "Signaling media decoding dependency 901 in Session Description Protocol (SDP)", 902 draft-ietf-mmusic-decoding-dependency-08 (work in progress), 903 April 2009. 905 8.2. Informative References 907 [9] Wenger, S., Wang, Y., Schierl, T., and A. Eleftheriadis, "RTP 908 Payload Format for SVC Video", draft-ietf-avt-rtp-svc-18 (work 909 in progress), March 2009. 911 [10] Lennox, J., Ott, J., and T. Schierl, "Source-Specific Media 912 Attributes in the Session Description Protocol (SDP)", 913 draft-ietf-mmusic-sdp-source-attributes-02 (work in progress), 914 October 2008. 916 [11] Casner, S., "Session Description Protocol (SDP) Bandwidth 917 Modifiers for RTP Control Protocol (RTCP) Bandwidth", RFC 3556, 918 July 2003. 920 [12] Rosenberg, J., "Interactive Connectivity Establishment (ICE): A 921 Protocol for Network Address Translator (NAT) Traversal for 922 Offer/Answer Protocols", draft-ietf-mmusic-ice-19 (work in 923 progress), October 2007. 925 [13] McGrew, D. and E. Rescorla, "Datagram Transport Layer Security 926 (DTLS) Extension to Establish Keys for Secure Real-time 927 Transport Protocol (SRTP)", draft-ietf-avt-dtls-srtp-05 (work 928 in progress), September 2008. 930 [14] Zimmermann, P., Johnston, A., and J. Callas, "ZRTP: Media Path 931 Key Agreement for Secure RTP", draft-zimmermann-avt-zrtp-13 932 (work in progress), January 2009. 934 [15] Westerlund, M. and S. Wenger, "RTP Topologies", RFC 5117, 935 January 2008. 937 Authors' Addresses 939 Colin Perkins 940 University of Glasgow 941 Department of Computing Science 942 Sir Alwyn Williams Building 943 Lilybank Gardens 944 Glasgow G12 8QQ 945 UK 947 Email: csp@csperkins.org 949 Thomas Schierl 950 Fraunhofer HHI 951 Einsteinufer 37 952 D-10587 Berlin 953 Germany 955 Phone: +49-30-31002-227 956 Email: Thomas.Schierl@hhi.fraunhofer.de