idnits 2.17.1 draft-ietf-fecframe-interleaved-fec-scheme-06.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** You're using the IETF Trust Provisions' Section 6.b License Notice from 12 Sep 2009 rather than the newer Notice from 28 Dec 2009. (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 : ---------------------------------------------------------------------------- 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 seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (December 15, 2009) is 5238 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) == Missing Reference: 'RFCXXXX' is mentioned on line 827, but not defined ** Obsolete normative reference: RFC 4566 (Obsoleted by RFC 8866) == Outdated reference: A later version (-10) exists of draft-ietf-mmusic-rfc4756bis-05 ** Obsolete normative reference: RFC 4288 (Obsoleted by RFC 6838) ** Obsolete normative reference: RFC 3555 (Obsoleted by RFC 4855, RFC 4856) == Outdated reference: A later version (-04) exists of draft-ietf-fecframe-dvb-al-fec-03 -- Obsolete informational reference (is this intentional?): RFC 2733 (Obsoleted by RFC 5109) -- Obsolete informational reference (is this intentional?): RFC 3009 (Obsoleted by RFC 5109) -- Obsolete informational reference (is this intentional?): RFC 2326 (Obsoleted by RFC 7826) -- Obsolete informational reference (is this intentional?): RFC 5246 (Obsoleted by RFC 8446) Summary: 4 errors (**), 0 flaws (~~), 4 warnings (==), 6 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 FEC Framework A. Begen 3 Internet-Draft Cisco 4 Intended status: Standards Track December 15, 2009 5 Expires: June 18, 2010 7 RTP Payload Format for 1-D Interleaved Parity FEC 8 draft-ietf-fecframe-interleaved-fec-scheme-06 10 Abstract 12 This document defines a new RTP payload format for the Forward Error 13 Correction (FEC) that is generated by the 1-D interleaved parity code 14 from a source media encapsulated in RTP. The 1-D interleaved parity 15 code is a systematic code, where a number of repair symbols are 16 generated from a set of source symbols and sent in a repair flow 17 separate from the source flow that carries the source symbols. The 18 1-D interleaved parity code offers a good protection against bursty 19 packet losses at a cost of decent complexity. The new payload format 20 defined in this document is used (with some exceptions) as a part of 21 the DVB Application-layer FEC specification. 23 Status of this Memo 25 This Internet-Draft is submitted to IETF in full conformance with the 26 provisions of BCP 78 and BCP 79. 28 Internet-Drafts are working documents of the Internet Engineering 29 Task Force (IETF), its areas, and its working groups. Note that 30 other groups may also distribute working documents as Internet- 31 Drafts. 33 Internet-Drafts are draft documents valid for a maximum of six months 34 and may be updated, replaced, or obsoleted by other documents at any 35 time. It is inappropriate to use Internet-Drafts as reference 36 material or to cite them other than as "work in progress." 38 The list of current Internet-Drafts can be accessed at 39 http://www.ietf.org/ietf/1id-abstracts.txt. 41 The list of Internet-Draft Shadow Directories can be accessed at 42 http://www.ietf.org/shadow.html. 44 This Internet-Draft will expire on June 18, 2010. 46 Copyright Notice 48 Copyright (c) 2009 IETF Trust and the persons identified as the 49 document authors. All rights reserved. 51 This document is subject to BCP 78 and the IETF Trust's Legal 52 Provisions Relating to IETF Documents 53 (http://trustee.ietf.org/license-info) in effect on the date of 54 publication of this document. Please review these documents 55 carefully, as they describe your rights and restrictions with respect 56 to this document. Code Components extracted from this document must 57 include Simplified BSD License text as described in Section 4.e of 58 the Trust Legal Provisions and are provided without warranty as 59 described in the BSD License. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 64 1.1. Use Cases . . . . . . . . . . . . . . . . . . . . . . . . 6 65 1.2. Overhead Computation . . . . . . . . . . . . . . . . . . . 8 66 1.3. Relation to Existing Specifications . . . . . . . . . . . 8 67 1.3.1. RFC 2733 and RFC 3009 . . . . . . . . . . . . . . . . 8 68 1.3.2. SMPTE 2022-1 . . . . . . . . . . . . . . . . . . . . . 8 69 1.3.3. ETSI TS 102 034 . . . . . . . . . . . . . . . . . . . 9 70 2. Requirements Notation . . . . . . . . . . . . . . . . . . . . 10 71 3. Definitions, Notations and Abbreviations . . . . . . . . . . . 10 72 3.1. Definitions . . . . . . . . . . . . . . . . . . . . . . . 10 73 3.2. Notations . . . . . . . . . . . . . . . . . . . . . . . . 10 74 4. Packet Formats . . . . . . . . . . . . . . . . . . . . . . . . 11 75 4.1. Source Packets . . . . . . . . . . . . . . . . . . . . . . 11 76 4.2. Repair Packets . . . . . . . . . . . . . . . . . . . . . . 11 77 5. Payload Format Parameters . . . . . . . . . . . . . . . . . . 14 78 5.1. Media Type Registration . . . . . . . . . . . . . . . . . 15 79 5.1.1. Registration of audio/1d-interleaved-parityfec . . . . 15 80 5.1.2. Registration of video/1d-interleaved-parityfec . . . . 16 81 5.1.3. Registration of text/1d-interleaved-parityfec . . . . 17 82 5.1.4. Registration of 83 application/1d-interleaved-parityfec . . . . . . . . . 18 84 5.2. Mapping to SDP Parameters . . . . . . . . . . . . . . . . 20 85 5.2.1. Offer-Answer Model Considerations . . . . . . . . . . 20 86 5.2.2. Declarative Considerations . . . . . . . . . . . . . . 21 87 6. Protection and Recovery Procedures . . . . . . . . . . . . . . 21 88 6.1. Overview . . . . . . . . . . . . . . . . . . . . . . . . . 21 89 6.2. Repair Packet Construction . . . . . . . . . . . . . . . . 22 90 6.3. Source Packet Reconstruction . . . . . . . . . . . . . . . 24 91 6.3.1. Associating the Source and Repair Packets . . . . . . 24 92 6.3.2. Recovering the RTP Header and Payload . . . . . . . . 25 93 7. Session Description Protocol (SDP) Signaling . . . . . . . . . 26 94 8. Congestion Control Considerations . . . . . . . . . . . . . . 27 95 9. Security Considerations . . . . . . . . . . . . . . . . . . . 28 96 10. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 28 97 11. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 29 98 12. Change Log . . . . . . . . . . . . . . . . . . . . . . . . . . 29 99 12.1. draft-ietf-fecframe-interleaved-fec-scheme-06 . . . . . . 29 100 12.2. draft-ietf-fecframe-interleaved-fec-scheme-05 . . . . . . 29 101 12.3. draft-ietf-fecframe-interleaved-fec-scheme-04 . . . . . . 29 102 12.4. draft-ietf-fecframe-interleaved-fec-scheme-03 . . . . . . 29 103 12.5. draft-ietf-fecframe-interleaved-fec-scheme-02 . . . . . . 29 104 12.6. draft-ietf-fecframe-interleaved-fec-scheme-01 . . . . . . 30 105 12.7. draft-ietf-fecframe-interleaved-fec-scheme-00 . . . . . . 30 106 13. References . . . . . . . . . . . . . . . . . . . . . . . . . . 30 107 13.1. Normative References . . . . . . . . . . . . . . . . . . . 30 108 13.2. Informative References . . . . . . . . . . . . . . . . . . 31 109 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . . 31 111 1. Introduction 113 This document extends the Forward Error Correction (FEC) header 114 defined in [RFC2733] and uses this new FEC header for the FEC that is 115 generated by the 1-D interleaved parity code from a source media 116 encapsulated in RTP [RFC3550]. The resulting new RTP payload format 117 is registered by this document. 119 The type of the source media protected by the 1-D interleaved parity 120 code can be audio, video, text or application. The FEC data are 121 generated according to the media type parameters that are 122 communicated through out-of-band means. The associations/ 123 relationships between the source and repair flows are also 124 communicated through out-of-band means. 126 The 1-D interleaved parity FEC uses the exclusive OR (XOR) operation 127 to generate the repair symbols. In a nutshell, the following steps 128 take place: 130 1. The sender determines a set of source packets to be protected 131 together based on the media type parameters. 133 2. The sender applies the XOR operation on the source symbols to 134 generate the required number of repair symbols. 136 3. The sender packetizes the repair symbols and sends the repair 137 packet(s) along with the source packets to the receiver(s) (in 138 different flows). The repair packets MAY be sent proactively or 139 on-demand. 141 Note that the source and repair packets belong to different source 142 and repair flows, and the sender MUST provide a way for the receivers 143 to demultiplex them, even in the case they are sent in the same 144 transport flow (i.e., same source/destination address/port with UDP). 145 This is required to offer backward compatibility (See Section 4). At 146 the receiver side, if all of the source packets are successfully 147 received, there is no need for FEC recovery and the repair packets 148 are discarded. However, if there are missing source packets, the 149 repair packets can be used to recover the missing information. Block 150 diagrams for the systematic parity FEC encoder and decoder are 151 sketched in Figure 1 and Figure 2, respectively. 153 +------------+ 154 +--+ +--+ +--+ +--+ --> | Systematic | --> +--+ +--+ +--+ +--+ 155 +--+ +--+ +--+ +--+ | Parity FEC | +--+ +--+ +--+ +--+ 156 | Encoder | 157 | (Sender) | --> +==+ +==+ 158 +------------+ +==+ +==+ 160 Source Packet: +--+ Repair Packet: +==+ 161 +--+ +==+ 163 Figure 1: Block diagram for systematic parity FEC encoder 165 +------------+ 166 +--+ X X +--+ --> | Systematic | --> +--+ +--+ +--+ +--+ 167 +--+ +--+ | Parity FEC | +--+ +--+ +--+ +--+ 168 | Decoder | 169 +==+ +==+ --> | (Receiver) | 170 +==+ +==+ +------------+ 172 Source Packet: +--+ Repair Packet: +==+ Lost Packet: X 173 +--+ +==+ 175 Figure 2: Block diagram for systematic parity FEC decoder 177 Suppose that we have a group of D x L source packets that have 178 sequence numbers starting from 1 running to D x L. If we apply the 179 XOR operation to the group of the source packets whose sequence 180 numbers are L apart from each other as sketched in Figure 3, we 181 generate L repair packets. This process is referred to as 1-D 182 interleaved FEC protection, and the resulting L repair packets are 183 referred to as interleaved (or column) FEC packets. 185 +-------------+ +-------------+ +-------------+ +-------+ 186 | S_1 | | S_2 | | S3 | ... | S_L | 187 | S_L+1 | | S_L+2 | | S_L+3 | ... | S_2xL | 188 | . | | . | | | | | 189 | . | | . | | | | | 190 | . | | . | | | | | 191 | S_(D-1)xL+1 | | S_(D-1)xL+2 | | S_(D-1)xL+3 | ... | S_DxL | 192 +-------------+ +-------------+ +-------------+ +-------+ 193 + + + + 194 ------------- ------------- ------------- ------- 195 | XOR | | XOR | | XOR | ... | XOR | 196 ------------- ------------- ------------- ------- 197 = = = = 198 +===+ +===+ +===+ +===+ 199 |C_1| |C_2| |C_3| ... |C_L| 200 +===+ +===+ +===+ +===+ 202 Figure 3: Generating interleaved (column) FEC packets 204 In Figure 3, S_n and C_m denote the source packet with a sequence 205 number n and the interleaved (column) FEC packet with a sequence 206 number m, respectively. 208 1.1. Use Cases 210 We generate one interleaved FEC packet out of D non-consecutive 211 source packets. This repair packet can provide a full recovery of 212 the missing information if there is only one packet missing among the 213 corresponding source packets. This implies that 1-D interleaved FEC 214 protection performs well under bursty loss conditions provided that L 215 is chosen large enough, i.e., L-packet duration SHOULD NOT be shorter 216 than the duration of the burst that is intended to be repaired. 218 For example, consider the scenario depicted in Figure 4 where the 219 sender generates interleaved FEC packets and a bursty loss hits the 220 source packets. Since the number of columns is larger than the 221 number of packets lost due to the bursty loss, the repair operation 222 succeeds. 224 +---+ 225 | 1 | X X X 226 +---+ 228 +---+ +---+ +---+ +---+ 229 | 5 | | 6 | | 7 | | 8 | 230 +---+ +---+ +---+ +---+ 232 +---+ +---+ +---+ +---+ 233 | 9 | | 10| | 11| | 12| 234 +---+ +---+ +---+ +---+ 236 +===+ +===+ +===+ +===+ 237 |C_1| |C_2| |C_3| |C_4| 238 +===+ +===+ +===+ +===+ 240 Figure 4: Example scenario where 1-D interleaved FEC protection 241 succeeds error recovery 243 The sender may generate interleaved FEC packets to combat with the 244 bursty packet losses. However, two or more random packet losses may 245 hit the source and repair packets in the same column. In that case, 246 the repair operation fails. This is illustrated in Figure 5. Note 247 that it is possible that two or more bursty losses may occur in the 248 same source block, in which case interleaved FEC packets may still 249 fail to recover the lost data. 251 +---+ +---+ +---+ 252 | 1 | X | 3 | | 4 | 253 +---+ +---+ +---+ 255 +---+ +---+ +---+ 256 | 5 | X | 7 | | 8 | 257 +---+ +---+ +---+ 259 +---+ +---+ +---+ +---+ 260 | 9 | | 10| | 11| | 12| 261 +---+ +---+ +---+ +---+ 263 +===+ +===+ +===+ +===+ 264 |C_1| |C_2| |C_3| |C_4| 265 +===+ +===+ +===+ +===+ 267 Figure 5: Example scenario where 1-D interleaved FEC protection fails 268 error recovery 270 1.2. Overhead Computation 272 The overhead is defined as the ratio of the number of bytes belonging 273 to the repair packets to the number of bytes belonging to the 274 protected source packets. 276 Assuming that each repair packet carries an equal number of bytes 277 carried by a source packet, we can compute the overhead as follows: 279 Overhead = 1/D 281 where D is the number of rows in the source block. 283 1.3. Relation to Existing Specifications 285 This section discusses the relation of the current specification to 286 other existing specifications. 288 1.3.1. RFC 2733 and RFC 3009 290 The current specification extends the FEC header defined in [RFC2733] 291 and registers a new RTP payload format. This new payload format is 292 not backward compatible with the payload format that was registered 293 by [RFC3009]. 295 1.3.2. SMPTE 2022-1 297 In 2007, the Society of Motion Picture and Television Engineers 298 (SMPTE) - Technology Committee N26 on File Management and Networking 299 Technology - decided to revise the Pro-MPEG Code of Practice (CoP) #3 300 Release 2 specification, which (was initially produced by the Pro- 301 MPEG Forum in 2004) discussed the several aspects of the transmission 302 of MPEG-2 transport streams over IP networks. The new SMPTE 303 specification is referred to as [SMPTE2022-1]. 305 The Pro-MPEG CoP #3 r2 document was originally based on [RFC2733]. 306 SMPTE revised the document by extending the FEC header (by setting 307 the E bit) proposed in [RFC2733]. This extended header offers some 308 improvements. 310 For example, instead of utilizing the bitmap field used in [RFC2733], 311 [SMPTE2022-1] introduces separate fields to convey the number of rows 312 (D) and columns (L) of the source block as well as the type of the 313 repair packet (i.e., whether the repair packet is an interleaved FEC 314 packet computed over a column or a non-interleaved FEC packet 315 computed over a row). These fields plus the base sequence number 316 allow the receiver side to establish the associations between the 317 source and repair packets. Note that although the bitmap field is 318 not utilized, the FEC header of [SMPTE2022-1] inherently carries over 319 the bitmap field from [RFC2733]. 321 On the other hand, some parts of [SMPTE2022-1] are not in compliant 322 with RTP [RFC3550]. For example, [SMPTE2022-1] sets the SSRC field 323 to zero and does not use the timestamp field in the RTP headers of 324 the repair packets (Receivers ignore the timestamps of the repair 325 packets). Furthermore, [SMPTE2022-1] also sets the CC field in the 326 RTP header to zero and does not allow any Contributing Source (CSRC) 327 entry in the RTP header. 329 The current document adopts the extended FEC header of [SMPTE2022-1] 330 and registers a new RTP payload format. At the same time, this 331 document fixes the parts of [SMPTE2022-1] that are not compliant with 332 RTP [RFC3550], except the one discussed below. 334 The baseline header format first proposed in [RFC2733] does not have 335 fields to protect the P and X bits and the CC fields of the source 336 packets associated with a repair packet. Rather, the P bit, X bit 337 and CC field in the RTP header of the repair packet are used to 338 protect those bits and fields. This, however, may sometimes result 339 in failures when doing the RTP header validity checks as specified in 340 [RFC3550]. While this behavior has been fixed in [RFC5109] that 341 obsoleted [RFC2733], the RTP payload format defined in this document 342 still allows for this behavior for legacy purposes. Implementations 343 following this specification MUST be aware of this potential issue 344 when RTP header validity checks are applied. 346 1.3.3. ETSI TS 102 034 348 In 2007, the Digital Video Broadcasting (DVB) consortium published a 349 technical specification [ETSI-TS-102-034] through European 350 Telecommunications Standards Institute (ETSI). This specification 351 covers several areas related to the transmission of MPEG-2 transport 352 stream-based services over IP networks. 354 The Annex E of [ETSI-TS-102-034] defines an optional protocol for 355 Application-layer FEC (AL-FEC) protection of streaming media for 356 DVB-IP services carried over RTP [RFC3550] transport. AL-FEC 357 protocol uses two layers for protection: a base layer that is 358 produced by a packet-based interleaved parity code, and an 359 enhancement layer that is produced by a Raptor code. While the use 360 of the enhancement layer is optional, the use of the base layer is 361 mandatory wherever AL-FEC is used. The DVB AL-FEC protocol is also 362 described in [I-D.ietf-fecframe-dvb-al-fec]. 364 The interleaved parity code that is used in the base layer is a 365 subset of [SMPTE2022-1]. In particular, AL-FEC base layer uses only 366 the 1-D interleaved FEC protection from [SMPTE2022-1]. The new RTP 367 payload format that is defined and registered in this document (with 368 some exceptions listed in [I-D.ietf-fecframe-dvb-al-fec]) is used as 369 the AL-FEC base layer. 371 2. Requirements Notation 373 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 374 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 375 document are to be interpreted as described in [RFC2119]. 377 3. Definitions, Notations and Abbreviations 379 The definitions and notations commonly used in this document are 380 summarized in this section. 382 3.1. Definitions 384 This document uses the following definitions: 386 Source Flow: The packet flow(s) carrying the source data and to 387 which FEC protection is to be applied. 389 Repair Flow: The packet flow(s) carrying the repair data. 391 Symbol: A unit of data. Its size, in bytes, is referred to as the 392 symbol size. 394 Source Symbol: The smallest unit of data used during the encoding 395 process. 397 Repair Symbol: Repair symbols are generated from the source symbols. 399 Source Packet: Data packets that contain only source symbols. 401 Repair Packet: Data packets that contain only repair symbols. 403 Source Block: A block of source symbols that are considered together 404 in the encoding process. 406 3.2. Notations 408 o L: Number of columns of the source block. 410 o D: Number of rows of the source block. 412 4. Packet Formats 414 This section defines the formats of the source and repair packets. 416 4.1. Source Packets 418 The source packets MUST contain the information that identifies the 419 source block and the position within the source block occupied by the 420 packet. Since the source packets that are carried within an RTP 421 stream already contain unique sequence numbers in their RTP headers 422 [RFC3550], we can identify the source packets in a straightforward 423 manner and there is no need to append additional field(s). The 424 primary advantage of not modifying the source packets in any way is 425 that it provides backward compatibility for the receivers that do not 426 support FEC at all. In multicast scenarios, this backward 427 compatibility becomes quite useful as it allows the non-FEC-capable 428 and FEC-capable receivers to receive and interpret the same source 429 packets sent in the same multicast session. 431 4.2. Repair Packets 433 The repair packets MUST contain information that identifies the 434 source block they pertain to and the relationship between the 435 contained repair symbols and the original source block. For this 436 purpose, we use the RTP header of the repair packets as well as 437 another header within the RTP payload, which we refer to as the FEC 438 header, as shown in Figure 6. 440 +------------------------------+ 441 | IP Header | 442 +------------------------------+ 443 | Transport Header | 444 +------------------------------+ 445 | RTP Header | __ 446 +------------------------------+ | 447 | FEC Header | \ 448 +------------------------------+ > RTP Payload 449 | Repair Symbols | / 450 +------------------------------+ __| 452 Figure 6: Format of repair packets 454 The RTP header is formatted according to [RFC3550] with some further 455 clarifications listed below: 457 o Version: The version field is set to 2. 459 o Padding (P) Bit: This bit is equal to the XOR sum of the 460 corresponding P bits from the RTP headers of the source packets 461 protected by this repair packet. However, padding octets are 462 never present in a repair packet, independent of the value of the 463 P bit. 465 o Extension (X) Bit: This bit is equal to the XOR sum of the 466 corresponding X bits from the RTP headers of the source packets 467 protected by this repair packet. However, an RTP header extension 468 is never present in a repair packet, independent of the value of 469 the X bit. 471 o CSRC Count (CC): This field is equal to the XOR sum of the 472 corresponding CC values from the RTP headers of the source packets 473 protected by this repair packet. However, a CSRC list is never 474 present in a repair packet, independent of the value of the CC 475 field. 477 o Marker (M) Bit: This bit is equal to the XOR sum of the 478 corresponding M bits from the RTP headers of the source packets 479 protected by this repair packet. 481 o Payload Type: The (dynamic) payload type for the repair packets 482 is determined through out-of-band means. Note that this document 483 registers a new payload format for the repair packets (Refer to 484 Section 5 for details). According to [RFC3550], an RTP receiver 485 that cannot recognize a payload type must discard it. This 486 provides backward compatibility. The FEC mechanisms can then be 487 used in a multicast group with mixed FEC-capable and non-FEC- 488 capable receivers. If a non-FEC-capable receiver receives a 489 repair packet, it will not recognize the payload type, and hence, 490 discards the repair packet. 492 o Sequence Number (SN): The sequence number has the standard 493 definition. It MUST be one higher than the sequence number in the 494 previously transmitted repair packet. The initial value of the 495 sequence number SHOULD be random (unpredictable) [RFC3550]. 497 o Timestamp (TS): The timestamp SHALL be set to a time 498 corresponding to the repair packet's transmission time. Note that 499 the timestamp value has no use in the actual FEC protection 500 process and is usually useful for jitter calculations. 502 o Synchronization Source (SSRC): The SSRC value SHALL be randomly 503 assigned as suggested by [RFC3550]. This allows the sender to 504 multiplex the source and repair flows on the same port, or 505 multiplex multiple repair flows on a single port. The repair 506 flows SHOULD use the RTCP CNAME field to associate themselves with 507 the source flow. 509 In some networks, the RTP Source, which produces the source 510 packets and the FEC Source, which generates the repair packets 511 from the source packets may not be the same host. In such 512 scenarios, using the same CNAME for the source and repair flows 513 means that the RTP Source and the FEC Source MUST share the same 514 CNAME (for this specific source-repair flow association). A 515 common CNAME may be produced based on an algorithm that is known 516 both to the RTP and FEC Source. This usage is compliant with 517 [RFC3550]. 519 Note that due to the randomness of the SSRC assignments, there is 520 a possibility of SSRC collision. In such cases, the collisions 521 MUST be resolved as described in [RFC3550]. 523 Note that the P bit, X bit, CC field and M bit of the source packets 524 are protected by the corresponding bits/fields in the RTP header of 525 the repair packet. On the other hand, the payload of a repair packet 526 protects the concatenation of (if present) the CSRC list, RTP 527 extension, payload and padding of the source RTP packets associated 528 with this repair packet. 530 The FEC header is 16 octets. The format of the FEC header is shown 531 in Figure 7. 533 0 1 2 3 534 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 1 535 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 536 | SN base low | Length recovery | 537 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 538 |E| PT recovery | Mask | 539 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 540 | TS recovery | 541 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 542 |N|D|Type |Index| Offset | NA | SN base ext | 543 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 545 Figure 7: Format of the FEC header 547 The FEC header consists of the following fields: 549 o The SN base low field is used to indicate the lowest sequence 550 number, taking wrap around into account, of those source packets 551 protected by this repair packet. 553 o The Length recovery field is used to determine the length of any 554 recovered packets. 556 o The E bit is the extension flag introduced in [RFC2733] and used 557 to extend the [RFC2733] FEC header. 559 o The PT recovery field is used to determine the payload type of the 560 recovered packets. 562 o The Mask field is not used. 564 o The TS recovery field is used to determine the timestamp of the 565 recovered packets. 567 o The N bit is the extension flag that is reserved for future uses. 569 o The D bit is not used. 571 o The Type field indicates the type of the error-correcting code 572 used. This document defines only one error-correcting code. 574 o The Index field is not used. 576 o The Offset and NA fields are used to indicate the number of 577 columns (L) and rows (D) of the source block, respectively. 579 o The SN base ext field is not used. 581 The details on setting the fields in the FEC header are provided in 582 Section 6.2. 584 It should be noted that a mask-based approach (similar to the one 585 specified in [RFC2733]) may not be very efficient to indicate which 586 source packets in the current source block are associated with a 587 given repair packet. In particular, for the applications that would 588 like to use large source block sizes, the size of the mask that is 589 required to describe the source-repair packet associations may be 590 prohibitively large. Instead, a systematized approach is inherently 591 more efficient. 593 5. Payload Format Parameters 595 This section provides the media subtype registration for the 1-D 596 interleaved parity FEC. The parameters that are required to 597 configure the FEC encoding and decoding operations are also defined 598 in this section. 600 5.1. Media Type Registration 602 This registration is done using the template defined in [RFC4288] and 603 following the guidance provided in [RFC3555]. 605 Note to the RFC Editor: In the following sections, please replace 606 "XXXX" with the number of this document prior to publication as an 607 RFC. 609 5.1.1. Registration of audio/1d-interleaved-parityfec 611 Type name: audio 613 Subtype name: 1d-interleaved-parityfec 615 Required parameters: 617 o rate: The RTP timestamp (clock) rate. The rate SHALL be larger 618 than 1000 Hz to provide sufficient resolution to RTCP operations. 619 However, it is RECOMMENDED to select the rate that matches the 620 rate of the protected source RTP stream. 622 o L: Number of columns of the source block. L is a positive 623 integer that is less than or equal to 255. 625 o D: Number of rows of the source block. D is a positive integer 626 that is less than or equal to 255. 628 o repair-window: The time that spans the source packets and the 629 corresponding repair packets. An FEC encoder processes a block of 630 source packets and generates a number of repair packets, which are 631 then transmitted within a certain duration. At the receiver, the 632 FEC decoder tries to decode all the packets received within the 633 repair window to recover the missing packets. Assuming that there 634 is no issue of delay variation, the FEC decoder SHOULD NOT wait 635 longer than the repair window since additional waiting would not 636 help the recovery process. The size of the repair window is 637 specified in microseconds. 639 Optional parameters: None. 641 Encoding considerations: This media type is framed (See Section 4.8 642 in the template document [RFC4288]) and contains binary data. 644 Security considerations: See Section 9 of [RFCXXXX]. 646 Interoperability considerations: None. 648 Published specification: [RFCXXXX]. 650 Applications that use this media type: Multimedia applications that 651 want to improve resiliency against packet loss by sending redundant 652 data in addition to the source media. 654 Additional information: None. 656 Person & email address to contact for further information: Ali Begen 657 and IETF Audio/Video Transport Working Group. 659 Intended usage: COMMON. 661 Restriction on usage: This media type depends on RTP framing, and 662 hence, is only defined for transport via RTP [RFC3550]. 664 Author: Ali Begen . 666 Change controller: IETF Audio/Video Transport Working Group 667 delegated from the IESG. 669 5.1.2. Registration of video/1d-interleaved-parityfec 671 Type name: video 673 Subtype name: 1d-interleaved-parityfec 675 Required parameters: 677 o rate: The RTP timestamp (clock) rate. The rate SHALL be larger 678 than 1000 Hz to provide sufficient resolution to RTCP operations. 679 However, it is RECOMMENDED to select the rate that matches the 680 rate of the protected source RTP stream. 682 o L: Number of columns of the source block. L is a positive 683 integer that is less than or equal to 255. 685 o D: Number of rows of the source block. D is a positive integer 686 that is less than or equal to 255. 688 o repair-window: The time that spans the source packets and the 689 corresponding repair packets. An FEC encoder processes a block of 690 source packets and generates a number of repair packets, which are 691 then transmitted within a certain duration. At the receiver, the 692 FEC decoder tries to decode all the packets received within the 693 repair window to recover the missing packets. Assuming that there 694 is no issue of delay variation, the FEC decoder SHOULD NOT wait 695 longer than the repair window since additional waiting would not 696 help the recovery process. The size of the repair window is 697 specified in microseconds. 699 Optional parameters: None. 701 Encoding considerations: This media type is framed (See Section 4.8 702 in the template document [RFC4288]) and contains binary data. 704 Security considerations: See Section 9 of [RFCXXXX]. 706 Interoperability considerations: None. 708 Published specification: [RFCXXXX]. 710 Applications that use this media type: Multimedia applications that 711 want to improve resiliency against packet loss by sending redundant 712 data in addition to the source media. 714 Additional information: None. 716 Person & email address to contact for further information: Ali Begen 717 and IETF Audio/Video Transport Working Group. 719 Intended usage: COMMON. 721 Restriction on usage: This media type depends on RTP framing, and 722 hence, is only defined for transport via RTP [RFC3550]. 724 Author: Ali Begen . 726 Change controller: IETF Audio/Video Transport Working Group 727 delegated from the IESG. 729 5.1.3. Registration of text/1d-interleaved-parityfec 731 Type name: text 733 Subtype name: 1d-interleaved-parityfec 735 Required parameters: 737 o rate: The RTP timestamp (clock) rate. The rate SHALL be larger 738 than 1000 Hz to provide sufficient resolution to RTCP operations. 739 However, it is RECOMMENDED to select the rate that matches the 740 rate of the protected source RTP stream. 742 o L: Number of columns of the source block. L is a positive 743 integer that is less than or equal to 255. 745 o D: Number of rows of the source block. D is a positive integer 746 that is less than or equal to 255. 748 o repair-window: The time that spans the source packets and the 749 corresponding repair packets. An FEC encoder processes a block of 750 source packets and generates a number of repair packets, which are 751 then transmitted within a certain duration. At the receiver, the 752 FEC decoder tries to decode all the packets received within the 753 repair window to recover the missing packets. Assuming that there 754 is no issue of delay variation, the FEC decoder SHOULD NOT wait 755 longer than the repair window since additional waiting would not 756 help the recovery process. The size of the repair window is 757 specified in microseconds. 759 Optional parameters: None. 761 Encoding considerations: This media type is framed (See Section 4.8 762 in the template document [RFC4288]) and contains binary data. 764 Security considerations: See Section 9 of [RFCXXXX]. 766 Interoperability considerations: None. 768 Published specification: [RFCXXXX]. 770 Applications that use this media type: Multimedia applications that 771 want to improve resiliency against packet loss by sending redundant 772 data in addition to the source media. 774 Additional information: None. 776 Person & email address to contact for further information: Ali Begen 777 and IETF Audio/Video Transport Working Group. 779 Intended usage: COMMON. 781 Restriction on usage: This media type depends on RTP framing, and 782 hence, is only defined for transport via RTP [RFC3550]. 784 Author: Ali Begen . 786 Change controller: IETF Audio/Video Transport Working Group 787 delegated from the IESG. 789 5.1.4. Registration of application/1d-interleaved-parityfec 791 Type name: application 792 Subtype name: 1d-interleaved-parityfec 794 Required parameters: 796 o rate: The RTP timestamp (clock) rate. The rate SHALL be larger 797 than 1000 Hz to provide sufficient resolution to RTCP operations. 798 However, it is RECOMMENDED to select the rate that matches the 799 rate of the protected source RTP stream. 801 o L: Number of columns of the source block. L is a positive 802 integer that is less than or equal to 255. 804 o D: Number of rows of the source block. D is a positive integer 805 that is less than or equal to 255. 807 o repair-window: The time that spans the source packets and the 808 corresponding repair packets. An FEC encoder processes a block of 809 source packets and generates a number of repair packets, which are 810 then transmitted within a certain duration. At the receiver, the 811 FEC decoder tries to decode all the packets received within the 812 repair window to recover the missing packets. Assuming that there 813 is no issue of delay variation, the FEC decoder SHOULD NOT wait 814 longer than the repair window since additional waiting would not 815 help the recovery process. The size of the repair window is 816 specified in microseconds. 818 Optional parameters: None. 820 Encoding considerations: This media type is framed (See Section 4.8 821 in the template document [RFC4288]) and contains binary data. 823 Security considerations: See Section 9 of [RFCXXXX]. 825 Interoperability considerations: None. 827 Published specification: [RFCXXXX]. 829 Applications that use this media type: Multimedia applications that 830 want to improve resiliency against packet loss by sending redundant 831 data in addition to the source media. 833 Additional information: None. 835 Person & email address to contact for further information: Ali Begen 836 and IETF Audio/Video Transport Working Group. 838 Intended usage: COMMON. 840 Restriction on usage: This media type depends on RTP framing, and 841 hence, is only defined for transport via RTP [RFC3550]. 843 Author: Ali Begen . 845 Change controller: IETF Audio/Video Transport Working Group 846 delegated from the IESG. 848 5.2. Mapping to SDP Parameters 850 Applications that are using RTP transport commonly use Session 851 Description Protocol (SDP) [RFC4566] to describe their RTP sessions. 852 The information that is used to specify the media types in an RTP 853 session has specific mappings to the fields in an SDP description. 854 In this section, we provide these mappings for the media subtype 855 registered by this document ("1d-interleaved-parityfec"). Note that 856 if an application does not use SDP to describe the RTP sessions, an 857 appropriate mapping must be defined and used to specify the media 858 types and their parameters for the control/description protocol 859 employed by the application. 861 The mapping of the media type specification for "1d-interleaved- 862 parityfec" and its parameters in SDP is as follows: 864 o The media type (e.g., "application") goes into the "m=" line as 865 the media name. 867 o The media subtype ("1d-interleaved-parityfec") goes into the 868 "a=rtpmap" line as the encoding name. The RTP clock rate 869 parameter ("rate") also goes into the "a=rtpmap" line as the clock 870 rate. 872 o The remaining required payload-format-specific parameters go into 873 the "a=fmtp" line by copying them directly from the media type 874 string as a semicolon-separated list of parameter=value pairs. 876 SDP examples are provided in Section 7. 878 5.2.1. Offer-Answer Model Considerations 880 When offering 1-D interleaved parity FEC over RTP using SDP in an 881 Offer/Answer model [RFC3264], the following considerations apply: 883 o Each combination of the L and D parameters produces a different 884 FEC data and is not compatible with any other combination. A 885 sender application may desire to offer multiple offers with 886 different sets of L and D values as long as the parameter values 887 are valid. The receiver SHOULD normally choose the offer that has 888 a sufficient amount of interleaving. If multiple such offers 889 exist, the receiver may choose the offer that has the lowest 890 overhead or the one that requires the smallest amount of 891 buffering. The selection depends on the application requirements. 893 o The value for the repair-window parameter depends on the L and D 894 values and cannot be chosen arbitrarily. More specifically, L and 895 D values determine the lower limit for the repair-window size. 896 The upper limit of the repair-window size does not depend on the L 897 and D values. 899 o Although combinations with the same L and D values but with 900 different repair-window sizes produce the same FEC data, such 901 combinations are still considered different offers. The size of 902 the repair-window is related to the maximum delay between the 903 transmission of a source packet and the associated repair packet. 904 This directly impacts the buffering requirement on the receiver 905 side and the receiver must consider this when choosing an offer. 907 o There are no optional format parameters defined for this payload. 908 Any unknown option in the offer MUST be ignored and deleted from 909 the answer. If FEC is not desired by the receiver, it can be 910 deleted from the answer. 912 5.2.2. Declarative Considerations 914 In declarative usage, like SDP in the Real-time Streaming Protocol 915 (RTSP) [RFC2326] or the Session Announcement Protocol (SAP) 916 [RFC2974], the following considerations apply: 918 o The payload format configuration parameters are all declarative 919 and a participant MUST use the configuration that is provided for 920 the session. 922 o More than one configuration may be provided (if desired) by 923 declaring multiple RTP payload types. In that case, the receivers 924 should choose the repair flow that is best for them. 926 6. Protection and Recovery Procedures 928 This section provides a complete specification of the 1-D interleaved 929 parity code and its RTP payload format. 931 6.1. Overview 933 The following sections specify the steps involved in generating the 934 repair packets and reconstructing the missing source packets from the 935 repair packets. 937 6.2. Repair Packet Construction 939 The RTP header of a repair packet is formed based on the guidelines 940 given in Section 4.2. 942 The FEC header includes 16 octets. It is constructed by applying the 943 XOR operation on the bit strings that are generated from the 944 individual source packets protected by this particular repair packet. 945 The set of the source packets that are associated with a given repair 946 packet can be computed by the formula given in Section 6.3.1. 948 The bit string is formed for each source packet by concatenating the 949 following fields together in the order specified: 951 o Padding bit (1 bit) (This is the most significant bit of the bit 952 string) 954 o Extension bit (1 bit) 956 o CC field (4 bits) 958 o Marker bit (1 bit) 960 o PT field (7 bits) 962 o Timestamp (32 bits) 964 o Unsigned network-ordered 16-bit representation of the source 965 packet length in bytes minus 12 (for the fixed RTP header), i.e., 966 the sum of the lengths of all the following if present: the CSRC 967 list, header extension, RTP payload and RTP padding (16 bits) 969 o If CC is nonzero, the CSRC list (variable length) 971 o If X is 1, the header extension (variable length) 973 o Payload (variable length) 975 o Padding, if present (variable length) 977 Note that if the lengths of the source packets are not equal, each 978 shorter packet MUST be padded to the length of the longest packet by 979 adding octet 0's at the end. Due to this possible padding and 980 mandatory FEC header, a repair packet has a larger size than the 981 source packets it protects. This may cause problems if the resulting 982 repair packet size exceeds the Maximum Transmission Unit (MTU) size 983 of the path over which the repair flow is sent. 985 By applying the parity operation on the bit strings produced from the 986 source packets, we generate the FEC bit string. Some parts of the 987 RTP header and the FEC header of the repair packet are generated from 988 the FEC bit string as follows: 990 o The first (most significant) bit in the FEC bit string is written 991 into the Padding bit in the RTP header of the repair packet. 993 o The next bit in the FEC bit string is written into the Extension 994 bit in the RTP header of the repair packet. 996 o The next 4 bits of the FEC bit string are written into the CC 997 field in the RTP header of the repair packet. 999 o The next bit of the FEC bit string is written into the Marker bit 1000 in the RTP header of the repair packet. 1002 o The next 7 bits of the FEC bit string are written into the PT 1003 recovery field in the FEC header. 1005 o The next 32 bits of the FEC bit string are written into the TS 1006 recovery field in the FEC header. 1008 o The next 16 bits are written into the Length recovery field in the 1009 FEC header. This allows the FEC procedure to be applied even when 1010 the lengths of the protected source packets are not identical. 1012 o The remaining bits are set to be the payload of the repair packet. 1014 The remaining parts of the FEC header are set as follows: 1016 o The SN base low field MUST be set to the lowest sequence number, 1017 taking wrap around into account, of those source packets protected 1018 by this repair packet. 1020 o The E bit MUST be set to 1 to extend the [RFC2733] FEC header. 1022 o The Mask field SHALL be set to 0 and ignored by the receiver. 1024 o The N bit SHALL be set to 0 and ignored by the receiver. 1026 o The D bit SHALL be set to 0 and ignored by the receiver. 1028 o The Type field MUST be set to 0. 1030 o The Index field SHALL be set to 0 and ignored by the receiver. 1032 o The Offset field MUST be set to the number of columns of the 1033 source block (L). 1035 o The NA field MUST be set to the number of rows of the source block 1036 (D). 1038 o The SN base ext field SHALL be set to 0 and ignored by the 1039 receiver. 1041 6.3. Source Packet Reconstruction 1043 This section describes the recovery procedures that are required to 1044 reconstruct the missing source packets. The recovery process has two 1045 steps. In the first step, the FEC decoder determines which source 1046 and repair packets should be used in order to recover a missing 1047 packet. In the second step, the decoder recovers the missing packet, 1048 which consists of an RTP header and RTP payload. 1050 In the following, we describe the RECOMMENDED algorithms for the 1051 first and second steps. Based on the implementation, different 1052 algorithms MAY be adopted. However, the end result MUST be identical 1053 to the one produced by the algorithms described below. 1055 6.3.1. Associating the Source and Repair Packets 1057 The first step is to associate the source and repair packets. The SN 1058 base low field in the FEC header shows the lowest sequence number of 1059 the source packets that form the particular column. In addition, the 1060 information of how many source packets are available in each column 1061 and row is available from the media type parameters specified in the 1062 SDP description. This set of information uniquely identifies all of 1063 the source packets associated with a given repair packet. 1065 Mathematically, for any received repair packet, p*, we can determine 1066 the sequence numbers of the source packets that are protected by this 1067 repair packet as follows: 1069 p*_snb + i * L (modulo 65536) 1071 where p*_snb denotes the value in the SN base low field of p*'s FEC 1072 header, L is the number of columns of the source block and 1074 0 <= i < D 1076 where D is the number of rows of the source block. 1078 We denote the set of the source packets associated with repair packet 1079 p* by set T(p*). Note that in a source block whose size is L columns 1080 by D rows, set T includes D source packets. Recall that 1-D 1081 interleaved FEC protection can fully recover the missing information 1082 if there is only one source packet missing in set T. If the repair 1083 packet that protects the source packets in set T is missing, or the 1084 repair packet is available but two or more source packets are 1085 missing, then missing source packets in set T cannot be recovered by 1086 1-D interleaved FEC protection. 1088 6.3.2. Recovering the RTP Header and Payload 1090 For a given set T, the procedure for the recovery of the RTP header 1091 of the missing packet, whose sequence number is denoted by SEQNUM, is 1092 as follows: 1094 1. For each of the source packets that are successfully received in 1095 set T, compute the bit string as described in Section 6.2. 1097 2. For the repair packet associated with set T, compute the bit 1098 string in the same fashion except use the PT recovery field 1099 instead of the PT field and TS recovery field instead of the 1100 Timestamp field, and set the CSRC list, header extension and 1101 padding to null regardless of the values of the CC field, X bit 1102 and P bit. 1104 3. If any of the bit strings generated from the source packets are 1105 shorter than the bit string generated from the repair packet, 1106 pad them to be the same length as the bit string generated from 1107 the repair packet. For padding, the padding of octet 0 MUST be 1108 added at the end of the bit string. 1110 4. Calculate the recovered bit string as the XOR of the bit strings 1111 generated from all source packets in set T and the FEC bit 1112 string generated from the repair packet associated with set T. 1114 5. Create a new packet with the standard 12-byte RTP header and no 1115 payload. 1117 6. Set the version of the new packet to 2. 1119 7. Set the Padding bit in the new packet to the first bit in the 1120 recovered bit string. 1122 8. Set the Extension bit in the new packet to the next bit in the 1123 recovered bit string. 1125 9. Set the CC field to the next 4 bits in the recovered bit string. 1127 10. Set the Marker bit in the new packet to the next bit in the 1128 recovered bit string. 1130 11. Set the Payload type in the new packet to the next 7 bits in the 1131 recovered bit string. 1133 12. Set the SN field in the new packet to SEQNUM. 1135 13. Set the TS field in the new packet to the next 32 bits in the 1136 recovered bit string. 1138 14. Take the next 16 bits of the recovered bit string and set the 1139 new variable Y to whatever unsigned integer this represents 1140 (assuming network order). Convert Y to host order and then take 1141 Y bytes from the recovered bit string and append them to the new 1142 packet. Y represents the length of the new packet in bytes 1143 minus 12 (for the fixed RTP header), i.e., the sum of the 1144 lengths of all the following if present: the CSRC list, header 1145 extension, RTP payload and RTP padding. 1147 15. Set the SSRC of the new packet to the SSRC of the source RTP 1148 stream. 1150 This procedure completely recovers both the header and payload of an 1151 RTP packet. 1153 7. Session Description Protocol (SDP) Signaling 1155 This section provides an SDP [RFC4566] example. The following 1156 example uses the FEC grouping semantics [I-D.ietf-mmusic-rfc4756bis]. 1158 In this example, we have one source video stream (mid:S1) and one FEC 1159 repair stream (mid:R1). We form one FEC group with the "a=group:FEC 1160 S1 R1" line. The source and repair streams are sent to the same port 1161 on different multicast groups. The repair window is set to 200 ms. 1163 v=0 1164 o=ali 1122334455 1122334466 IN IP4 fec.example.com 1165 s=Interleaved Parity FEC Example 1166 t=0 0 1167 a=group:FEC S1 R1 1168 m=video 30000 RTP/AVP 100 1169 c=IN IP4 233.252.0.1/127 1170 a=rtpmap:100 MP2T/90000 1171 a=mid:S1 1172 m=application 30000 RTP/AVP 110 1173 c=IN IP4 233.252.0.2/127 1174 a=rtpmap:110 1d-interleaved-parityfec/90000 1175 a=fmtp:110 L:5; D:10; repair-window:200000 1176 a=mid:R1 1178 8. Congestion Control Considerations 1180 FEC is an effective approach to provide applications resiliency 1181 against packet losses. However, in networks where the congestion is 1182 a major contributor to the packet loss, the potential impacts of 1183 using FEC SHOULD be considered carefully before injecting the repair 1184 flows into the network. In particular, in bandwidth-limited 1185 networks, FEC repair flows may consume most or all of the available 1186 bandwidth and may consequently congest the network. In such cases, 1187 the applications MUST NOT arbitrarily increase the amount of FEC 1188 protection since doing so may lead to a congestion collapse. If 1189 desired, stronger FEC protection MAY be applied only after the source 1190 rate has been reduced. 1192 In a network-friendly implementation, an application SHOULD NOT send/ 1193 receive FEC repair flows if it knows that sending/receiving those FEC 1194 repair flows would not help at all in recovering the missing packets. 1195 Such a practice helps reduce the amount of wasted bandwidth. It is 1196 RECOMMENDED that the amount of FEC protection is adjusted dynamically 1197 based on the packet loss rate observed by the applications. 1199 In multicast scenarios, it may be difficult to optimize the FEC 1200 protection per receiver. If there is a large variation among the 1201 levels of FEC protection needed by different receivers, it is 1202 RECOMMENDED that the sender offers multiple repair flows with 1203 different levels of FEC protection and the receivers join the 1204 corresponding multicast sessions to receive the repair flow(s) that 1205 is best for them. 1207 9. Security Considerations 1209 RTP packets using the payload format defined in this specification 1210 are subject to the security considerations discussed in the RTP 1211 specification [RFC3550] and in any applicable RTP profile. 1213 The main security considerations for the RTP packet carrying the RTP 1214 payload format defined within this memo are confidentiality, 1215 integrity and source authenticity. Confidentiality is achieved by 1216 encrypting the RTP payload. Altering the FEC packets can have a big 1217 impact on the reconstruction operation. An attack by changing some 1218 bits in the FEC packets can have a significant effect on the 1219 calculation and the recovery of the source packets. For example, 1220 changing the length recovery field can result in the recovery of a 1221 packet that is too long. Depending on the application, it may be 1222 helpful to perform a sanity check on the received source and FEC 1223 packets before performing the recovery operation and to determine the 1224 validity of the recovered packets before using them. 1226 Integrity of the RTP packets is achieved through a suitable 1227 cryptographic integrity protection mechanism. Such a cryptographic 1228 system may also allow the authentication of the source of the 1229 payload. A suitable security mechanism for this RTP payload format 1230 should provide source authentication capable of determining if an RTP 1231 packet is from a member of the RTP session. 1233 Note that the appropriate mechanism to provide security to RTP and 1234 payloads following this memo may vary. It is dependent on the 1235 application, transport and signaling protocol employed. Therefore, a 1236 single mechanism is not sufficient, although if suitable, using the 1237 Secure Real-time Transport Protocol (SRTP) [RFC3711] is RECOMMENDED. 1238 Other mechanisms that may be used are IPsec [RFC4301] and Transport 1239 Layer Security (TLS) [RFC5246]; other alternatives may exist. 1241 If FEC protection is applied on already encrypted source packets, 1242 there is no need for additional encryption. However, if the source 1243 packets are encrypted after FEC protection is applied, the FEC 1244 packets should be cryptographically as secure as the source packets. 1245 Failure to provide an equal level of confidentiality, integrity and 1246 authentication to the FEC packets can compromise the source packets' 1247 confidentiality, integrity or authentication since the FEC packets 1248 are generated by applying XOR operation across the source packets. 1250 10. IANA Considerations 1252 New media subtypes are subject to IANA registration. For the 1253 registration of the payload format and its parameters introduced in 1254 this document, refer to Section 5. 1256 11. Acknowledgments 1258 A major part of this document is borrowed from [RFC2733], [RFC5109] 1259 and [SMPTE2022-1]. Thus, the author would like to thank the authors 1260 and editors of these earlier specifications. The author also thanks 1261 Colin Perkins for his constructive suggestions for this document. 1263 12. Change Log 1265 12.1. draft-ietf-fecframe-interleaved-fec-scheme-06 1267 The following are the major changes compared to version 05: 1269 o Comments from IETF LC have been addressed. 1271 12.2. draft-ietf-fecframe-interleaved-fec-scheme-05 1273 The following are the major changes compared to version 04: 1275 o Comments from Vincent Roca have been addressed. 1277 12.3. draft-ietf-fecframe-interleaved-fec-scheme-04 1279 The following are the major changes compared to version 03: 1281 o Further comments from AVT WG have been addressed. 1283 12.4. draft-ietf-fecframe-interleaved-fec-scheme-03 1285 The following are the major changes compared to version 02: 1287 o Comments from WGLC have been addressed. 1289 12.5. draft-ietf-fecframe-interleaved-fec-scheme-02 1291 The following are the major changes compared to version 01: 1293 o Some details were added regarding the use of CNAME field. 1295 o Offer-Answer and Declarative Considerations sections have been 1296 completed. 1298 o Security Considerations section has been completed. 1300 12.6. draft-ietf-fecframe-interleaved-fec-scheme-01 1302 The following are the major changes compared to version 00: 1304 o The timestamp field definition has changed. 1306 12.7. draft-ietf-fecframe-interleaved-fec-scheme-00 1308 This is the initial version, which is based on an earlier individual 1309 submission. The following are the major changes compared to that 1310 document: 1312 o Per the discussion in the WG, references to the FEC Framework have 1313 been removed and the document has been turned into a pure RTP 1314 payload format specification. 1316 o A new section is added for congestion control considerations. 1318 o Editorial changes to clarify a few points. 1320 13. References 1322 13.1. Normative References 1324 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 1325 Requirement Levels", BCP 14, RFC 2119, March 1997. 1327 [RFC3550] Schulzrinne, H., Casner, S., Frederick, R., and V. 1328 Jacobson, "RTP: A Transport Protocol for Real-Time 1329 Applications", STD 64, RFC 3550, July 2003. 1331 [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session 1332 Description Protocol", RFC 4566, July 2006. 1334 [I-D.ietf-mmusic-rfc4756bis] 1335 Begen, A., "Forward Error Correction Grouping Semantics in 1336 Session Description Protocol", 1337 draft-ietf-mmusic-rfc4756bis-05 (work in progress), 1338 October 2009. 1340 [RFC4288] Freed, N. and J. Klensin, "Media Type Specifications and 1341 Registration Procedures", BCP 13, RFC 4288, December 2005. 1343 [RFC3555] Casner, S. and P. Hoschka, "MIME Type Registration of RTP 1344 Payload Formats", RFC 3555, July 2003. 1346 [RFC3264] Rosenberg, J. and H. Schulzrinne, "An Offer/Answer Model 1347 with Session Description Protocol (SDP)", RFC 3264, 1348 June 2002. 1350 13.2. Informative References 1352 [I-D.ietf-fecframe-dvb-al-fec] 1353 Begen, A. and T. Stockhammer, "DVB-IPTV Application-Layer 1354 Hybrid FEC Protection", draft-ietf-fecframe-dvb-al-fec-03 1355 (work in progress), September 2009. 1357 [RFC2733] Rosenberg, J. and H. Schulzrinne, "An RTP Payload Format 1358 for Generic Forward Error Correction", RFC 2733, 1359 December 1999. 1361 [RFC3009] Rosenberg, J. and H. Schulzrinne, "Registration of 1362 parityfec MIME types", RFC 3009, November 2000. 1364 [RFC5109] Li, A., "RTP Payload Format for Generic Forward Error 1365 Correction", RFC 5109, December 2007. 1367 [ETSI-TS-102-034] 1368 ETSI TS 102 034 V1.3.1, "Transport of MPEG 2 TS Based DVB 1369 Services over IP Based Networks", October 2007. 1371 [RFC2326] Schulzrinne, H., Rao, A., and R. Lanphier, "Real Time 1372 Streaming Protocol (RTSP)", RFC 2326, April 1998. 1374 [RFC2974] Handley, M., Perkins, C., and E. Whelan, "Session 1375 Announcement Protocol", RFC 2974, October 2000. 1377 [SMPTE2022-1] 1378 SMPTE 2022-1-2007, "Forward Error Correction for Real-Time 1379 Video/Audio Transport over IP Networks", 2007. 1381 [RFC3711] Baugher, M., McGrew, D., Naslund, M., Carrara, E., and K. 1382 Norrman, "The Secure Real-time Transport Protocol (SRTP)", 1383 RFC 3711, March 2004. 1385 [RFC4301] Kent, S. and K. Seo, "Security Architecture for the 1386 Internet Protocol", RFC 4301, December 2005. 1388 [RFC5246] Dierks, T. and E. Rescorla, "The Transport Layer Security 1389 (TLS) Protocol Version 1.2", RFC 5246, August 2008. 1391 Author's Address 1393 Ali Begen 1394 Cisco 1395 170 West Tasman Drive 1396 San Jose, CA 95134 1397 USA 1399 Email: abegen@cisco.com