idnits 2.17.1 draft-ietf-avtext-splicing-for-rtp-09.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year -- The document date (August 13, 2012) is 4273 days in the past. Is this intentional? Checking references for intended status: Informational ---------------------------------------------------------------------------- == Unused Reference: 'RFC3711' is defined on line 671, but no explicit reference was found in the text Summary: 0 errors (**), 0 flaws (~~), 2 warnings (==), 1 comment (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 AVTEXT Working Group J. Xia 3 Internet-Draft Huawei 4 Intended status: Informational August 13, 2012 5 Expires: February 14, 2013 7 Content Splicing for RTP Sessions 8 draft-ietf-avtext-splicing-for-rtp-09 10 Abstract 12 Content splicing is a process that replaces the content of a main 13 multimedia stream with other multimedia content, and delivers the 14 substitutive multimedia content to the receivers for a period of 15 time. Splicing is commonly used for local advertisement insertion by 16 cable operators, replacing a national advertisement content with a 17 local advertisement. 19 This memo describes some use cases for content splicing and a set of 20 requirements for splicing content delivered by RTP. It provides 21 concrete guidelines for how an RTP mixer can be used to handle 22 content splicing. 24 Status of this Memo 26 This Internet-Draft is submitted in full conformance with the 27 provisions of BCP 78 and BCP 79. 29 Internet-Drafts are working documents of the Internet Engineering 30 Task Force (IETF). Note that other groups may also distribute 31 working documents as Internet-Drafts. The list of current Internet- 32 Drafts is at http://datatracker.ietf.org/drafts/current/. 34 Internet-Drafts are draft documents valid for a maximum of six months 35 and may be updated, replaced, or obsoleted by other documents at any 36 time. It is inappropriate to use Internet-Drafts as reference 37 material or to cite them other than as "work in progress." 39 This Internet-Draft will expire on February 14, 2013. 41 Copyright Notice 43 Copyright (c) 2012 IETF Trust and the persons identified as the 44 document authors. All rights reserved. 46 This document is subject to BCP 78 and the IETF Trust's Legal 47 Provisions Relating to IETF Documents 48 (http://trustee.ietf.org/license-info) in effect on the date of 49 publication of this document. Please review these documents 50 carefully, as they describe your rights and restrictions with respect 51 to this document. Code Components extracted from this document must 52 include Simplified BSD License text as described in Section 4.e of 53 the Trust Legal Provisions and are provided without warranty as 54 described in the Simplified BSD License. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 59 2. System Model and Terminology . . . . . . . . . . . . . . . . . 3 60 3. Requirements for RTP Splicing . . . . . . . . . . . . . . . . 6 61 4. Content Splicing for RTP sessions . . . . . . . . . . . . . . 7 62 4.1. RTP Processing in RTP Mixer . . . . . . . . . . . . . . . 7 63 4.2. RTCP Processing in RTP Mixer . . . . . . . . . . . . . . . 8 64 4.3. Media Clipping Considerations . . . . . . . . . . . . . . 10 65 4.4. Congestion Control Considerations . . . . . . . . . . . . 11 66 4.5. Processing Splicing in User Invisibility Case . . . . . . 12 67 5. Implementation Considerations . . . . . . . . . . . . . . . . 13 68 6. Security Considerations . . . . . . . . . . . . . . . . . . . 13 69 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 14 70 8. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 14 71 9. 10. Appendix- Why Mixer Is Chosen . . . . . . . . . . . . . . 14 72 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 15 73 10.1. Normative References . . . . . . . . . . . . . . . . . . . 15 74 10.2. Informative References . . . . . . . . . . . . . . . . . . 15 75 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 16 77 1. Introduction 79 This document outlines how content splicing can be used in RTP 80 sessions. Splicing, in general, is a process where part of a 81 multimedia content is replaced with other multimedia content, and 82 delivered to the receivers for a period of time. The substitutive 83 content can be provided for example via another stream or via local 84 media file storage. One representative use case for splicing is 85 local advertisement insertion, allowing content providers to replace 86 the national advertising content with its own regional advertising 87 content prior to delivering the regional advertising content to the 88 receivers. Besides the advertisement insertion use case, there are 89 other use cases in which splicing technology can be applied. For 90 example, splicing a recorded video into a video conferencing session, 91 or implementing a playlist server that stitches pieces of video 92 together. 94 Content splicing is a well-defined operation in MPEG-based cable TV 95 systems. Indeed, the Society for Cable Telecommunications Engineers 96 (SCTE) has created two standards, [SCTE30] and [SCTE35], to 97 standardize MPEG2-TS splicing procedure. SCTE 30 creates a 98 standardized method for communication between advertisements server 99 and splicer, and SCTE 35 supports splicing of MPEG2 transport 100 streams. 102 When using multimedia splicing into the internet, the media may be 103 transported by RTP. In this case the original media content and 104 substitutive media content will use the same time period, but may 105 contain different numbers of RTP packets due to different media 106 codecs and entropy coding. This mismatch may require some 107 adjustments of the RTP header sequence number to maintain 108 consistency. [RFC3550] provides the tools to enabled seamless 109 content splicing in RTP session, but to date there has been no clear 110 guidelines on how to use these tools. 112 This memo outlines the requirements for content splicing in RTP 113 sessions and describes how an RTP mixer can be used to meet these 114 requirements. 116 2. System Model and Terminology 118 In this document, an intermediary network element, the Splicer 119 handles RTP splicing. The Splicer can receive main content and 120 substitutive content simultaneously, but will send one of them at one 121 point of time. 123 When RTP splicing begins, the splicer sends the substitutive content 124 to the RTP receiver instead of the main content for a period of time. 125 When RTP splicing ends, the splicer switches back sending the main 126 content to the RTP receiver. 128 A simplified RTP splicing diagram is depicted in Figure 1, in which 129 only one main content flow and one substitutive content flow are 130 given. Actually, the splicer can handle multiple splicing for 131 multiple RTP sessions simultaneously. RTP splicing may happen more 132 than once in multiple time slots during the lifetime of the main RTP 133 stream. The methods how splicer learns when to start and end the 134 splicing is out of scope for this document. 136 +---------------+ 137 | | Main Content +-----------+ 138 | Main RTP |------------->| | Output Content 139 | Content | | Splicer |---------------> 140 +---------------+ ---------->| | 141 | +-----------+ 142 | 143 | Substitutive Content 144 | 145 | 146 +-----------------------+ 147 | Substitutive RTP | 148 | Content | 149 | or | 150 | Local File Storage | 151 +-----------------------+ 153 Figure 1: RTP Splicing Architecture 155 This document uses the following terminologies. 157 Output RTP Stream 159 The RTP stream that the RTP receiver is currently receiving. The 160 content of output RTP stream can be either main content or 161 substitutive content. 163 Main Content 165 The multimedia content that are conveyed in main RTP stream. Main 166 content will be replaced by the substitutive content during 167 splicing. 169 Main RTP Stream 171 The RTP stream that the splicer is receiving. The content of main 172 RTP stream can be replaced by substitutive content for a period of 173 time. 175 Main RTP Sender 177 The sender of RTP packets carrying the main RTP stream. 179 Substitutive Content 181 The multimedia content that replaces the main content during 182 splicing. The substitutive content can for example be contained 183 in an RTP stream from a media sender or fetched from local media 184 file storage. 186 Substitutive RTP Stream 188 A RTP stream with new content that will replace the content in the 189 main RTP stream. Substitutive RTP stream and main RTP stream are 190 two separate streams. If the substitutive content is provided via 191 substitutive RTP stream, the substitutive RTP Stream must pass 192 through the splicer before the substitutive content is delivered 193 to receiver. 195 Substitutive RTP Sender 197 The sender of RTP packets carrying the substitutive RTP stream. 199 Splicing In Point 201 A virtual point in the RTP stream, suitable for substitutive 202 content entry, typically in the boundary between two independently 203 decodable frames. 205 Splicing Out Point 207 A virtual point in the RTP stream, suitable for substitutive 208 content exist, typically in the boundary between two independently 209 decodable frames. 211 Splicer 213 An intermediary node that inserts substitutive content into main 214 RTP stream. The splicer sends substitutive content to RTP 215 receiver instead of main content during splicing. It is also 216 responsible for processing RTCP traffic between the RTP sender and 217 the RTP receiver. 219 3. Requirements for RTP Splicing 221 In order to allow seamless content splicing at the RTP layer, the 222 following requirements must be met. Meeting these will also allow, 223 but not require, seamless content splicing at layers above RTP. 225 REQ-1: 227 The splicer should be agnostic about the network and transport 228 layer protocols used to deliver the RTP streams. 230 REQ-2: 232 The splicing operation at the RTP layer must allow splicing at any 233 point required by the media content, and must not constrain when 234 splicing in or splicing out operations can take place. 236 REQ-3: 238 Splicing of RTP content must be backward compatible with the RTP/ 239 RTCP protocol, associated profiles, payload formats, and 240 extensions. 242 REQ-4: 244 The splicer will modify the content of RTP packets, and break the 245 end-to-end security, e.g., breaking data integrity and source 246 authentication. If the Splicer is designated to insert 247 substitutive content, it must be trusted, i.e., be in the security 248 context(s) as the main RTP sender, the substitutive RTP sender, 249 and the receivers. If encryption is employed, the splicer must be 250 able to decrypt the inbound RTP packets and re-encrypt the 251 outbound RTP packets after splicing. 253 REQ-5: 255 The splicer should rewrite as necessary and forward RTCP messages 256 (e.g., including packet loss, jitter, etc.) sent from downstream 257 receiver to the main RTP sender or the substitutive RTP sender, 258 and thus allow the main RTP sender or substitutive RTP sender to 259 learn the performance of the downstream receiver when its content 260 is being passed to RTP receiver. In addition, the splicer should 261 rewrite RTCP messages from the main RTP sender or substitutive RTP 262 sender to the receiver. 264 REQ-6: 266 The splicer must not affect other RTP sessions running between the 267 RTP sender and the RTP receiver, and must be transparent for the 268 RTP sessions it does not splice. 270 REQ-7: 272 The splicer should be able to modify the RTP stream across a 273 splicing in or splicing out point such that the splicing point is 274 not easy to be detected in the RTP stream. For the advertisement 275 insertion use case, it is important to make it difficult for the 276 receiver to detect where an advertisement insertion is starting or 277 ending from the RTP packets. Ensuring the splicing point is not 278 visible in the media content may be easy with some codecs, but 279 extremely difficult with others; in the worst case, the splicer 280 may need to perform full media transcoding if it has to hide the 281 splicing point in the media content. This memo only focuses on 282 making the splicing invisible at the RTP layer. How (or if) the 283 splicing is made invisible in the media stream is outside the 284 scope of this memo. 286 4. Content Splicing for RTP sessions 288 The RTP specification [RFC3550] defines two types of middlebox: RTP 289 translators and RTP mixers. Splicing is best viewed as a mixing 290 operation. The splicer generates a new RTP stream that is a mix of 291 the main RTP stream and the substitutive RTP stream. An RTP mixer is 292 therefore an appropriate model for a content splicer. In next four 293 subsections (from subsection 4.1 to subsection 4.4), the document 294 analyzes how the mixer handles RTP splicing and how it satisfies the 295 general requirements listed in section 3. In subsection 4.5, the 296 document looks at REQ-7 in order to hide the fact that splicing take 297 place. 299 4.1. RTP Processing in RTP Mixer 301 A splicer could be implemented as a mixer that receives the main RTP 302 stream and the substitutive content (possibly via a substitutive RTP 303 stream), and sends a single output RTP stream to the receiver(s). 304 That output RTP stream will contain either the main content or the 305 substitutive content. The output RTP stream will come from the 306 mixer, and will have the SSRC of the mixer rather than the main RTP 307 sender or the substitutive RTP sender. 309 The mixer uses its own SSRC, sequence number space and timing model 310 when generating the output stream. Moreover, the mixer may insert 311 the SSRC of main RTP stream into CSRC list in the output media 312 stream. 314 At the splicing in point, when the substitutive content becomes 315 active, the mixer chooses the substitutive RTP stream as input stream 316 at splicing in point, and extracts the payload data (i.e., 317 substitutive content). If the substitutive content comes from local 318 media file storage, the mixer directly fetches the substitutive 319 content. After that, the mixer encapsulates substitutive content 320 instead of main content as the payload of the output media stream, 321 and then sends the output RTP media stream to receiver. The mixer 322 may insert the SSRC of substitutive RTP stream into CSRC list in the 323 output media stream. If the substitutive content comes from local 324 media file storage, the mixer should leave the CSRC list blank. 326 At the splicing out point, when the substitutive content ends, the 327 mixer retrieves the main RTP stream as input stream at splicing out 328 point, and extracts the payload data (i.e., main content). After 329 that, the mixer encapsulates main content instead of substitutive 330 content as the payload of the output media stream, and then sends the 331 output media stream to the receivers. Moreover, the mixer may insert 332 the SSRC of main RTP stream into CSRC list in the output media stream 333 as before. 335 Note that if the content is too large to fit into RTP packets sent to 336 RTP receiver, the mixer needs to transcode or perform application- 337 layer fragmentation. Usually the mixer is deployed as part of a 338 managed system and MTU will be carefully managed by this system. 339 This document does not raise any new MTU related issues compared to a 340 standard mixer described in [RFC3550]. 342 Splicing may occur more than once during the lifetime of main RTP 343 stream, this means the mixer needs to send main content and 344 substitutive content in turn with its own SSRC identifier. From 345 receiver point of view, the only source of the output stream is the 346 mixer regardless of where the content is coming from. 348 4.2. RTCP Processing in RTP Mixer 350 By monitoring available bandwidth and buffer levels and by computing 351 network metrics such as packet loss, network jitter, and delay, RTP 352 receiver can learn the network performance and communicate this to 353 the RTP sender via RTCP reception reports. 355 According to the description in section 7.3 of [RFC3550], the mixer 356 splits the RTCP flow between sender and receiver into two separate 357 RTCP loops, RTP sender has no idea about the situation on the 358 receiver. But splicing is a processing that the mixer selects one 359 media stream from multiple streams rather than mixing them, so the 360 mixer can leave the SSRC identifier in the RTCP report intact (i.e., 361 the SSRC of downstream receiver), this enables the main RTP sender or 362 the substitutive RTP sender to learn the situation on the receiver. 364 If the RTCP report corresponds to a time interval that is entirely 365 main content or entirely substitutive content, the number of output 366 RTP packets containing substitutive content is equal to the number of 367 input substitutive RTP packets (from substitutive RTP stream) during 368 splicing, in the same manner, the number of output RTP packets 369 containing main content is equal to the number of input main RTP 370 packets (from main RTP stream) during non-splicing unless the mixer 371 fragment the input RTP packets. This means that the mixer does not 372 need to modify the loss packet fields in reception report blocks in 373 RTCP reports. But if the mixer fragments the input RTP packets, it 374 may need to modify the loss packet fields to compensate for the 375 fragmentation. Whether the input RTP packets are fragmented or not, 376 the mixer still needs to change the SSRC field in report block to the 377 SSRC identifier of the main RTP sender or the substitutive RTP 378 sender, and rewrite the extended highest sequence number field to the 379 corresponding original extended highest sequence number before 380 forwarding the RTCP report to the main RTP sender or the substitutive 381 RTP sender. 383 If the RTCP report spans the splicing in point or the splicing out 384 point, it reflects the characteristics of the combination of main RTP 385 packets and substitutive RTP packets. In this case, the mixer needs 386 to divide the RTCP report into two separate RTCP reports and send 387 them to their original RTP senders respectively. For each RTCP 388 report, the mixer also needs to make the corresponding changes to the 389 packet loss fields in report block besides the SSRC field and the 390 extended highest sequence number field. 392 If the mixer receives an RTCP extended report (XR) block, it should 393 rewrite the XR report block in a similar way to the reception report 394 block in the RTCP report. 396 Besides forwarding the RTCP reports sent from RTP receiver, the mixer 397 can also generate its own RTCP reports to inform the main RTP sender 398 or the substitutive RTP sender of the reception quality of the 399 content reaches the mixer when the content is not sent to the RTP 400 receiver. These RTCP reports use the SSRC of the mixer. If the 401 substitutive content comes from local media file storage, the mixer 402 does not need to generate RTCP reports for the substitutive stream. 404 Based on above RTCP operating mechanism, the RTP sender whose content 405 is being passed to receiver will see the reception quality of its 406 stream as received by the mixer, and the reception quality of spliced 407 stream as received by the receiver. The RTP sender whose content is 408 not being passed to receiver will only see the reception quality of 409 its stream as received by the mixer. 411 The mixer must forward RTCP SDES and BYE packets from the receiver to 412 the sender, and may forward them in inverse direction as defined in 413 section 7.3 of [RFC3550]. 415 Once the mixer receives an RTP/AVPF [RFC4585] transport layer 416 feedback packet, it must handle it carefully as the feedback packet 417 may contain the information of the content that come from different 418 RTP senders. In this case the mixer needs to divide the feedback 419 packet into two separate feedback packets and process the information 420 in the feedback control information (FCI) in the two feedback 421 packets, just as the RTCP report process described above. 423 If the substitutive content comes from local media file storage 424 (i.e., the mixer can be regarded as the substitutive RTP sender), any 425 RTCP packets received from downstream relate to the substitutive 426 content must be terminated on the mixer without any further 427 processing. 429 4.3. Media Clipping Considerations 431 This section provides informative guideline about how media clipping 432 is shaped and how the mixer deal with the media clipping. 434 If the time slot for substitutive content mismatches (is shorter or 435 longer than) the duration of the main content to be replaced, then 436 media clipping may occur at the splicing point. 438 If the substitutive content has shorter duration from the main 439 content, then there will be a gap in the output RTP stream. The RTP 440 sequence number will be contiguous across this gap, but there will be 441 an unexpected jump in the RTP timestamp. This gap will cause the 442 receiver to have nothing to play. This is unavoidable, unless the 443 mixer adjusts the splice in or splice out point to compensate, 444 sending more of the main RTP stream in place of the shorter 445 substitutive stream, or unless the mixer can vary the length of the 446 substitutive content. It is the responsibility of the higher layer 447 protocols to ensure that the substitutive content is of the same 448 duration as the main content to be replaced. 450 If the insertion duration is longer than the reserved gap duration, 451 there will be an overlap between the substitutive RTP stream and the 452 main RTP stream at splicing out point. One straightforward approach 453 is that the mixer takes an ungraceful action, terminating the 454 splicing and switching back to main RTP stream even if this may cause 455 media stuttering on receiver. Alternatively, the mixer may transcode 456 the substitutive content to play at a faster rate than normal, to 457 adjust it to the length of the gap in the main content, and generate 458 a new RTP stream for the transcoded content. This is a complex 459 operation, and very specific to the content and media codec used. 461 4.4. Congestion Control Considerations 463 If the substitutive content has somewhat different characteristics 464 from the main content it replaces, or if the substitutive content is 465 encoded with a different codec or has different encoding bitrate, it 466 might overload the network and might cause network congestion on the 467 path between the mixer and the RTP receiver(s) that would not have 468 been caused by the main content. 470 To be robust to network congestion and packet loss, a mixer that is 471 performing splicing must continuously monitor the status of 472 downstream network by monitoring any of the following RTCP reports 473 that are used: 475 1. RTCP receiver reports indicate packet loss [RFC3550]. 477 2. RTCP NACKs for lost packet recovery [RFC4585]. 479 3. RTCP ECN Feedback information [I-D.ietf-avtcore-ecn-for-rtp]. 481 Once the mixer detects congestion on its downstream link, it will 482 treat these reports as follows: 484 1. If the mixer receives the RTCP receiver reports with packet loss 485 indication, it will forward the reports to the substitutive RTP 486 sender or the main RTP sender as described in section 4.2. 488 2. If mixer receives the RTCP NACK packets defined in [RFC4585] from 489 RTP receiver for packet loss recovery, it first identifies the 490 content category of lost packets to which the NACK corresponds. 491 Then, the mixer will generate new RTCP NACK for the lost packets 492 with its own SSRC, and make corresponding changes to their 493 sequence numbers to match original, pre-spliced, packets. If the 494 lost substitutive content comes from local media file storage, 495 the mixer acting as substitutive RTP sender will directly fetch 496 the lost substitutive content and retransmit it to RTP receiver. 497 The mixer may buffer the sent RTP packets and do the 498 retransmission. 500 It is somewhat complex that the lost packets requested in a 501 single RTCP NACK message not only contain the main content but 502 also the substitutive content. To address this, the mixer must 503 divide the RTCP NACK packet into two separate RTCP NACK packets: 504 one requests for the lost main content, and another requests for 505 the lost substitutive content. 507 3. If an ECN-aware mixer receives RTCP ECN feedbacks (RTCP ECN 508 feedback packets or RTCP XR summary reports) defined in 509 [I-D.ietf-avtcore-ecn-for-rtp] from the RTP receiver, it must 510 process them in a similar way to the RTP/AVPF feedback packet or 511 RTCP XR process described in section 4.2 of this memo. 513 These three methods require the mixer to run a congestion control 514 loop and bitrate adaptation between itself and RTP receiver. The 515 mixer can thin or transcode the main RTP stream or the substitutive 516 RTP stream, but such operations are very inefficient and difficult, 517 and bring undesirable delay. Fortunately in this memo, the mixer 518 acting as splicer can rewrite the RTCP packets sent from the RTP 519 receiver and forward them to the RTP sender, thus letting the RTP 520 sender knows that congestion is being experienced on the path between 521 the mixer and the RTP receiver. Then, the RTP sender applies its 522 congestion control algorithm and reduces the media bitrate to a value 523 that is in compliance with congestion control principles for the 524 slowest link. The congestion control algorithm may be a TCP-friendly 525 bitrate adaptation algorithm specified in [RFC5348], or a DCCP 526 congestion control algorithms defined in [RFC5762]. 528 If the substitutive content comes from local media file storage, the 529 mixer must directly reduce the bitrate as if it were the substitutive 530 RTP sender. 532 From above analysis, to reduce the risk of congestion and remain the 533 bandwidth consumption stable over time, the substitutive RTP stream 534 is recommended to be encoded at an appropriate bitrate to match that 535 of main RTP stream. If the substitutive RTP stream comes from the 536 substitutive RTP sender, this sender had better has some knowledge 537 about the media encoding bitrate of main content in advance. How it 538 knows that is out of scope in this draft. 540 4.5. Processing Splicing in User Invisibility Case 542 If it is desirable to prevent receivers from detecting that splicing 543 is occurring at the RTP layer, the mixer must not include a CSRC list 544 in outgoing RTP packets, and must not forward RTCP messages from the 545 main RTP sender or from the substitutive RTP sender. Due to the 546 absence of CSRC list in the output RTP stream, the RTP receiver only 547 initiates SDES, BYE and APP packets to the mixer without any 548 knowledge of the main RTP sender and the substitutive RTP sender. 550 CSRC list identifies the contributing sources, these SSRC identifiers 551 of contributing sources are kept globally unique for each RTP 552 session. The uniqueness of SSRC identifier is used to resolve 553 collisions and detecting RTP-level forwarding loops as defined in 554 section 8.2 of [RFC3550]. The absence of CSRC list in this case will 555 create a danger that loops involving those contributing sources could 556 not be detected. The Loops could occur if either the mixer is 557 misconfigured to form a loop, or a second mixer/translator is added, 558 causing packets to loop back to upstream of the original mixer. So 559 Non-RTP means must be used to detect and resolve loops if the mixer 560 does not add a CSRC list. 562 5. Implementation Considerations 564 When the mixer is used to handle RTP splicing, RTP receiver does not 565 need any RTP/RTCP extension for splicing. As a trade-off, additional 566 overhead could be induced on the mixer which uses its own sequence 567 number space and timing model. So the mixer will rewrite RTP 568 sequence number and timestamp whatever splicing is active or not, and 569 generate RTCP flows for both sides. In case the mixer serves 570 multiple main RTP streams simultaneously, this may lead to more 571 overhead on the mixer. 573 If User Invisibility Requirement is required, CSRC list is not 574 included in outgoing RTP packet, this brings a potential issue with 575 loop detection as briefly described in section 4.5. 577 6. Security Considerations 579 The splicing application is subject to the general security 580 considerations of the RTP specification [RFC3550]. 582 The mixer acting as splicer replaces some content with other content 583 in RTP packets, thus breaking any RTP level end-to-end security, such 584 as integrity protection and source authentication. Thus any RTP 585 level or outside security mechanism, such as IPsec or DTLS will use a 586 security association between the splicer and the receiver. When 587 using SRTP the splicer could be provisioned with the same security 588 association as the main RTP sender. Using a limitation in the SRTP 589 security services, the splicer can modify and re-protect the RTP 590 packets without enabling the receiver to detect if the data comes 591 from the original source or from the splicer. 593 Security goals to have source authentication all the way from the RTP 594 main sender to the receiver through the splicer is not possible with 595 splicing. The nature of this RTP service offered by a network 596 operator employing a content splicer is that the RTP layer security 597 relationship is between the receiver and the splicer, and between the 598 senders and the splicer, are not end-to-end. This appears to 599 invalidate the invisibility goal, but in the common case the receiver 600 will consider the splicer as the main media source. 602 Commonly no RTP level security mechanism is employed. Instead only 603 payload security mechanisms (e.g., ISMACryp [ISMACryp]) are used. If 604 any payload internal security mechanisms are used, only the RTP 605 sender and the RTP receiver can learn the security keying material 606 generated by such internal security mechanism, in which case, any 607 middlebox (e.g., splicer) between the RTP sender and the RTP receiver 608 can't get such keying material, and thus fail to perform splicing. 610 7. IANA Considerations 612 No IANA actions are required. 614 8. Acknowledgments 616 The following individuals have reviewed the earlier versions of this 617 specification and provided very valuable comments: Colin Perkins, 618 Magnus Westerlund, Roni Even, Tom Van Caenegem, Joerg Ott, David R 619 Oran, Cullen Jennings, Ali C Begen, Charles Eckel and Ning Zong. 621 9. 10. Appendix- Why Mixer Is Chosen 623 Translator and mixer both can realize splicing by changing a set of 624 RTP parameters. 626 Translator has no SSRC, hence it is transparent to RTP sender and 627 receiver. Therefore, RTP sender sees the full path to the receiver 628 when translator is passing its content. When translator insert the 629 substitutive content RTP sender could get a report on the path up to 630 translator itself. Additionally, if splicing does not occur yet, 631 translator does not need to rewrite RTP header, the overhead on 632 translator can be avoided. 634 If mixer is used to do splicing, it can also allow RTP sender to 635 learn the situation of its content on receiver or on mixer just like 636 translator does, which is specified in section 4.2. Compared to 637 translator, mixer's outstanding benefit is that it is pretty straight 638 forward to do with RTCP messages, for example, bit-rate adaptation to 639 handle varying network conditions. But translator needs more 640 considerations and its implementation is more complex. 642 From above analysis, both translator and mixer have their own 643 advantages: less overhead or less complexity on handling RTCP. 644 Through long and sophisticated discussion, the avtext WG members 645 prefer less complexity rather than less overhead and incline to mixer 646 to do splicing. 648 If one chooses mixer as splicer, the overhead on mixer must be taken 649 into account even if the splicing does not occur yet. 651 10. References 653 10.1. Normative References 655 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 656 Jacobson, "RTP: A Transport Protocol for Real-Time 657 Applications", STD 64, RFC 3550, July 2003. 659 [RFC4585] Ott, J., Wenger, S., Sato, N., Burmeister, C., and J. Rey, 660 "Extended RTP Profile for Real-time Transport Control 661 Protocol (RTCP)-Based Feedback (RTP/AVPF)", RFC 4585, 662 July 2006. 664 [I-D.ietf-avtcore-ecn-for-rtp] 665 Westerlund, M., "Explicit Congestion Notification (ECN) 666 for RTP over UDP", draft-ietf-avtcore-ecn-for-rtp-08 (work 667 in progress), May 2012. 669 10.2. Informative References 671 [RFC3711] Baugher, M., McGrew, D., Naslund, M., Carrara, E., and K. 672 Norrman, "The Secure Real-time Transport Protocol (SRTP)", 673 RFC 3711, March 2004. 675 [RFC5348] Floyd, S., Handley, M., Padhye, J., and J. Widmer, "TCP 676 Friendly Rate Control (TFRC): Protocol Specification", 677 RFC 5348, September 2008. 679 [RFC5762] Perkins, C., "RTP and the Datagram Congestion Control 680 Protocol (DCCP)", RFC 5762, April 2010. 682 [SCTE30] Society of Cable Telecommunications Engineers (SCTE), 683 "Digital Program Insertion Splicing API", 2009. 685 [SCTE35] Society of Cable Telecommunications Engineers (SCTE), 686 "Digital Program Insertion Cueing Message for Cable", 687 2011. 689 [ISMACryp] 690 Internet Streaming Media Alliance (ISMA), "ISMA Encryption 691 and Authentication Specification 2.0", November 2007. 693 Author's Address 695 Jinwei Xia 696 Huawei 697 Software No.101 698 Nanjing, Yuhuatai District 210012 699 China 701 Phone: +86-025-86622310 702 Email: xiajinwei@huawei.com