idnits 2.17.1 draft-ietf-xrblock-rtcp-xr-qoe-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 == Line 67 has weird spacing: '...o/video per S...' == Line 312 has weird spacing: '...o/video per S...' -- The document date (June 18, 2013) is 3965 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 976, but not defined == Unused Reference: 'RFC5234' is defined on line 736, but no explicit reference was found in the text -- Possible downref: Non-RFC (?) normative reference: ref. 'ATSC' ** Obsolete normative reference: RFC 4566 (Obsoleted by RFC 8866) ** Obsolete normative reference: RFC 5226 (Obsoleted by RFC 8126) Summary: 2 errors (**), 0 flaws (~~), 5 warnings (==), 2 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network Working Group A. Clark 3 Internet-Draft Telchemy 4 Intended status: Standards Track Q. Wu 5 Expires: December 20, 2013 Huawei 6 R. Schott 7 Deutsche Telekom 8 G. Zorn 9 Network Zen 10 June 18, 2013 12 RTP Control Protocol (RTCP) Extended Report (XR) Blocks for QoE Metric 13 Reporting 14 draft-ietf-xrblock-rtcp-xr-qoe-09 16 Abstract 18 This document defines an RTP Control Protocol (RTCP) Extended Report 19 (XR) Block including two new segment types and associated SDP 20 parameters that allow the reporting of QoE metrics for use in a range 21 of RTP applications. 23 Status of this Memo 25 This Internet-Draft is submitted 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). Note that other groups may also distribute 30 working documents as Internet-Drafts. The list of current Internet- 31 Drafts is at http://datatracker.ietf.org/drafts/current/. 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 This Internet-Draft will expire on December 20, 2013. 40 Copyright Notice 42 Copyright (c) 2013 IETF Trust and the persons identified as the 43 document authors. All rights reserved. 45 This document is subject to BCP 78 and the IETF Trust's Legal 46 Provisions Relating to IETF Documents 47 (http://trustee.ietf.org/license-info) in effect on the date of 48 publication of this document. Please review these documents 49 carefully, as they describe your rights and restrictions with respect 50 to this document. Code Components extracted from this document must 51 include Simplified BSD License text as described in Section 4.e of 52 the Trust Legal Provisions and are provided without warranty as 53 described in the Simplified BSD License. 55 Table of Contents 57 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 4 58 1.1. QoE Metrics Report Block . . . . . . . . . . . . . . . . . 4 59 1.2. RTCP and RTCP XR Reports . . . . . . . . . . . . . . . . . 4 60 1.3. Performance Metrics Framework . . . . . . . . . . . . . . 4 61 1.4. Applicability . . . . . . . . . . . . . . . . . . . . . . 4 62 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 5 63 2.1. Standards Language . . . . . . . . . . . . . . . . . . . . 5 64 3. QoE Metrics Block . . . . . . . . . . . . . . . . . . . . . . 6 65 3.1. Metric Block Structure . . . . . . . . . . . . . . . . . . 6 66 3.2. Definition of Fields in QoE Metrics Block . . . . . . . . 7 67 3.2.1. Single Stream audio/video per SSRC Segment . . . . . 8 68 3.2.2. Multi-Channel audio per SSRC Segment . . . . . . . . . 9 69 4. SDP Signaling . . . . . . . . . . . . . . . . . . . . . . . . 10 70 4.1. SDP rtcp-xr-attrib Attribute Extension . . . . . . . . . . 11 71 4.2. Offer/Answer Usage . . . . . . . . . . . . . . . . . . . . 12 72 5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 14 73 5.1. New RTCP XR Block Type value . . . . . . . . . . . . . . . 14 74 5.2. New RTCP XR SDP Parameter . . . . . . . . . . . . . . . . 14 75 5.3. The SDP calgextmap Attribute . . . . . . . . . . . . . . . 14 76 5.4. New registry of calculation algorithms . . . . . . . . . . 15 77 6. Security Considerations . . . . . . . . . . . . . . . . . . . 16 78 7. Authors . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 79 8. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 16 80 9. References . . . . . . . . . . . . . . . . . . . . . . . . . . 17 81 9.1. Normative References . . . . . . . . . . . . . . . . . . . 17 82 9.2. Informative References . . . . . . . . . . . . . . . . . . 17 83 Appendix A. Example of User Quality of Experience Evaluation 84 for video stream . . . . . . . . . . . . . . . . . . 19 85 Appendix B. Metrics represented using RFC6390 Template . . . . . 20 86 Appendix C. Change Log . . . . . . . . . . . . . . . . . . . . . 22 87 C.1. draft-ietf-xrblock-rtcp-xr-qoe-09 . . . . . . . . . . . . 22 88 C.2. draft-ietf-xrblock-rtcp-xr-qoe-08 . . . . . . . . . . . . 23 89 C.3. draft-ietf-xrblock-rtcp-xr-qoe-07 . . . . . . . . . . . . 23 90 C.4. draft-ietf-xrblock-rtcp-xr-qoe-06 . . . . . . . . . . . . 23 91 C.5. draft-ietf-xrblock-rtcp-xr-qoe-04 . . . . . . . . . . . . 23 92 C.6. draft-ietf-xrblock-rtcp-xr-qoe-03 . . . . . . . . . . . . 23 93 C.7. draft-ietf-xrblock-rtcp-xr-qoe-02 . . . . . . . . . . . . 23 94 C.8. draft-ietf-xrblock-rtcp-xr-qoe-01 . . . . . . . . . . . . 24 95 C.9. draft-ietf-xrblock-rtcp-xr-qoe-00 . . . . . . . . . . . . 24 97 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 24 99 1. Introduction 101 1.1. QoE Metrics Report Block 103 This document defines a new block type to augment those defined in 104 [RFC3611], for use in a range of RTP applications. 106 The new block type provides information on media quality using one of 107 several standard metrics (i.e.,Mean Opinion Score(MoS)). 109 The metrics belong to the class of application level metrics defined 110 in [RFC6792]. 112 1.2. RTCP and RTCP XR Reports 114 The use of RTCP for reporting is defined in [RFC3550]. [RFC3611] 115 defined an extensible structure for reporting using an RTCP Extended 116 Report (XR). This document defines a new Extended Report block for 117 use with [RFC3550] and [RFC3611]. 119 1.3. Performance Metrics Framework 121 The Performance Metrics Framework [RFC6390] provides guidance on the 122 definition and specification of performance metrics. The RTP 123 Monitoring Architectures [RFC6792] provides guidelines for reporting 124 block format using RTCP XR. The XR block type described in this 125 document are in accordance with the guidelines in [RFC6390] and 126 [RFC6792]. 128 1.4. Applicability 130 The QoE Metrics Report Block can be used in any application of RTP 131 for which QoE measurement algorithms are defined. 133 The factors that affect real-time audio/video application quality can 134 be split into two categories. The first category consists of 135 transport-specific factors such as packet loss, delay and jitter 136 (which also translates into losses in the playback buffer). The 137 factors in the second category are application-specific factors that 138 affect real time application (e.g., video) quality. These factors 139 can be but are not limited to video codec and loss recovery 140 technique, coding bit rate, packetization scheme, and content 141 characteristics. 143 Compared with application-specific factors, the transport-specific 144 factors sometimes are not sufficient to measure real time media 145 quality, since the ability to analyze the real time media in the 146 application layer provides quantifiable measurements for end user 147 Quality of Experience (QoE) that may not be captured in the 148 transmission layers or from the RTP layer down. In a typical 149 scenario, monitoring of the transmission layers can produce 150 statistics suggesting that quality is not an issue, such as the fact 151 that network jitter is not excessive. However, problems may occur in 152 the service layers leading to poor subscriber QoE. Therefore 153 monitoring using only network-level measurements may be insufficient 154 when application layer media quality is required. 156 In order to provide accurate measures of real time media quality when 157 transporting real time media across a network, the QoE Metrics is 158 highly required which can be conveyed in the RTCP XR packets 159 [RFC3611] and may have the following three benefits: 161 o Tuning the content encoder algorithm to satisfy real time data 162 quality requirements. 163 o Determining which system techniques to use in a given situation 164 and when to switch from one technique to another as system 165 parameters change. 166 o Verifying the continued correct operation of an existing system. 168 2. Terminology 170 2.1. Standards Language 172 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 173 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 174 document are to be interpreted as described in RFC 2119 [RFC2119]. 176 The terminology used is 178 Numeric formats S X:Y 180 where S indicates a two's complement signed representation, X 181 the number of bits prior to the decimal place and Y the number 182 of bits after the decimal place. 183 Hence 8:8 represents an unsigned number in the range 0.0 to 184 255.996 with a granularity of 0.0039. S7:8 would represent the 185 range -127.996 to +127.996. 0:16 represents a proper binary 186 fraction with range 187 0.0 to 1 - 1/65536 = 0.9999847 188 though note that use of flag values at the top of the numeric 189 range slightly reduces this upper limit. For example, if the 190 16- bit values 0xfffe and 0xffff are used as flags for "over- 191 range" and "unavailable" conditions, a 0:16 quantity has range 192 0.0 to 1 - 3/65536 = 0.9999542 194 3. QoE Metrics Block 196 Multimedia application QoE metric is commonly expressed as a MOS 197 ("Mean Opinion Score"), MOS is on a scale from 1 to 5, in which 5 198 represents excellent and 1 represents unacceptable. MOS scores are 199 usually obtained using subjective testing or using objective 200 algorithm. However Subjective testing to estimate the multimedia 201 quality may be not suitable for measuring the multimedia quality 202 since the results may vary from test to test. Therefore using 203 objective algorithm to calculate MOS scores is RECOMMENDED. ITU-T 204 recommendations (e.g., 205 [G.107][G.107.1][P.862][P.862.1][P.862.2][P.863][P.564][G.1082][P.120 206 1.1][P.1201.2][P.1202.1][P.1202.2]) define the methodologies for 207 assessment of the performance of multimedia stream and provides a 208 method to evaluate QoE estimation algorithms and objective model for 209 video and audio. Hence this document recommends vendors and 210 implementers to use these International Telecommunication Union 211 (ITU)-specified methodologies to measure parameters when possible. 213 This block reports the multimedia application performance or media 214 quality beyond the information carried in the standard RTCP packet 215 format. Information is recorded about QoE metric which provides a 216 measure that gives a numerical indication of the perceived quality of 217 the media received. The measurement of metrics in this block are 218 usually made at the receiving end of the RTP stream. Instances of 219 this Metrics Block refer by Synchronization source (SSRC) to the 220 separate auxiliary Measurement Information block [RFC6776] which 221 describes measurement periods in use (see RFC6776 section 4.2). 223 This Metrics Block relies on the measurement period in the 224 Measurement Information block indicating the span of the report. 225 Senders MUST send this block in the same compound RTCP packet as the 226 measurement information block. Receivers MUST verify that the 227 measurement period is received in the same compound RTCP packet as 228 this Metrics Block. If not, this Metrics Block MUST be discarded. 230 3.1. Metric Block Structure 232 The report block contents are dependent upon a series of flag bits 233 carried in the first part of the header. Not all parameters need to 234 be reported in each block. Flags indicate which are and which are 235 not reported. The fields corresponding to unreported parameters MUST 236 be present, and MUST be set to zero. The receiver MUST ignore any 237 QoE Metrics Block with a non-zero value in any field flagged as 238 unreported. The encoding of QoE metrics block payload consists of a 239 series of 32 bit units called segments that describe payload Type, 240 MoS algorithm and MoS value. 242 The QoE Metrics Block has the following format: 244 0 1 2 3 245 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 1 246 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 247 | BT=QMB | I | Reserved | Block Length | 248 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 249 | SSRC of source | 250 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 251 | Segment 1 | 252 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 253 | Segment 2 | 254 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 255 .................. 256 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 257 | Segment n | 258 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 260 3.2. Definition of Fields in QoE Metrics Block 262 Block type (BT): 8 bits 264 The QoE Metrics Block is identified by the constant . 266 Interval Metric flag (I): 2 bits 268 This field is used to indicate whether the QoE metrics are 269 Sampled, Interval or Cumulative metrics [RFC6792]: 271 I=10: Interval Duration - the reported value applies to the 272 most recent measurement interval duration between successive 273 metrics reports. 274 I=11: Cumulative Duration - the reported value applies to the 275 accumulation period characteristic of cumulative measurements. 276 I=01: Sampled Value - the reported value is a sampled 277 instantaneous value. 279 In this document, the value I=00 is reserved for future use. 280 Senders MUST NOT use the values I=00. If a block is received with 281 I=00, the receiver MUST discard the block. 283 Reserved: 6 bits 285 This field is reserved for future definition. In the absence of 286 such a definition, the bits in this field MUST be set to zero and 287 ignored by the receiver (See RFC6709 section 4.2). 289 Block Length: 16 bits 291 The length of this report block in 32-bit words, minus one. For 292 the QoE Metrics Block, the block length is variable length. 294 SSRC of source: 32 bits 296 As defined in Section 4.1 of [RFC3611]. 298 Segment i: 32 bits 300 There are two segment types defined in this document: single 301 stream Audio/Video per SSRC segment, multi-channel audio per SSRC 302 segment. Multi-channel audio per SSRC segment is used to deal 303 with the case where Multi-channel audios are carried in one RTP 304 stream while single stream Audio/Video per SSRC segment is used to 305 deal with the case where each media stream is identified by SSRC 306 and sent in separate RTP stream. The leftmost bit of the segment 307 determines its type. If the leftmost bit of the segment is zero, 308 then it is single stream segment. If the leftmost bit is one, 309 then it is multi-channel audio segment. Note that two segment 310 types can not be present in the same metric block. 312 3.2.1. Single Stream audio/video per SSRC Segment 314 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 315 |S| CAID | PT | MOS Value | 316 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 318 Segment Type (S): 1 bit 320 This field is used to identify the segment type used in this 321 report block. A zero identifies this as a single stream Audio/ 322 Video per SSRC segment. Single stream means there is only one 323 media stream carried in one RTP stream. The single stream Audio/ 324 Video per SSRC segment can be used to report the MoS value 325 associated with the media stream identified by SSRC. If there are 326 multiple media streams and they want to use the single stream 327 Audio/Video per SSRC segment to report the MOS value, they should 328 be carried in the separate RTP streams with each identified by 329 different SSRC. In this case, multiple QoE Metrics Blocks are 330 required to report the MOS value corresponding to each media 331 stream using single stream Audio/Video per SSRC segment in the 332 same RTCP XR packet. 334 Calg Algorithm ID (CAID) : 8bits 336 The 8-bit CAID is the local identifier of calculation algorithm 337 associated with this segment in the range 1-255 inclusive. 339 Payload Type (PT): 7 bits 341 QoE metrics reporting depends on the payload format in use. This 342 field identifies the format of the RTP payload. For RTP sessions 343 where multiple payload formats can be negotiated or the payload 344 format changes during the mid-session), the value of this field 345 will be used to indicate what payload format was in use for the 346 reporting interval. 348 MOS Value: 16 bits 350 The estimated mean opinion score for multimedia application 351 performance is defined as including the effects of delay,loss, 352 discard,jitter and other effects that would affect media quality. 353 It is expressed in numeric format 8:8 with the value in the range 354 0.0 to 255.996. The valid the measured value ranges from 0.0 to 355 50.0, corresponding to MoS x 10 as for MoS. If the measured value 356 is over ranged, the value 0xFFFE MUST be reported to indicate an 357 over-range measurement. If the measurement is unavailable, the 358 value 0xFFFF MUST be reported. Values other than 0xFFFE,0xFFFF 359 and the valid range defined above MUST NOT be sent and MUST be 360 ignored by the receiving system. 362 3.2.2. Multi-Channel audio per SSRC Segment 364 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 365 |S| CAID | PT |CHID | MOS Value | 366 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 368 Segment Type (S): 1 bit 370 This field is used to identify the segment type used in this 371 report block. A one identifies this as a multi-channel audio 372 segment. 374 CAlg Algorithm ID (CAID) : 8bits 376 The 8-bit ID is the local identifier of this segment in the range 377 1-255 inclusive. 379 Payload Type (PT): 7 bits 381 As defined in Section 3.2.1 of this document. 383 Channel Identifier (CHID): 3 bits 385 If multiple channels of audio are carried in one RTP stream, each 386 channel of audio will be viewed as a independent channel(e.g., 387 left channel audio, right channel audio). This field is used to 388 identify each channel carried in the same media stream. The 389 default Channel mapping follows static ordering rule described in 390 the section 4.1 of [RFC3551]. However there are some payload 391 formats that use different channel mappings, e.g., AC-3 audio over 392 RTP [RFC4184] only follow AC-3 channel order scheme defined in 393 [ATSC]. Enhanced AC-3 Audio over RTP [RFC4598] uses dynamic 394 channel transform mechanism. In order that the appropriate 395 channel mapping can be determined, QoE reports need to be tied to 396 an RTP payload format, i.e., including the payload type of the 397 reported media according to [RFC6792] and using Payload Type to 398 determine the appropriate channel mapping. 400 MOS Value: 13 bits 402 The estimated mean opinion score for multimedia application 403 performance is defined as including the effects of delay,loss, 404 discard,jitter and other effects that would affect multimedia 405 quality . It is expressed in numeric format 6:7 with the value in 406 the range 0.0 to 63.992. The valid the measured value ranges from 407 0.0 to 50.0, corresponding to MoS x 10 as for MoS. If the 408 measured value is over ranged, the value 0x1FFE MUST be reported 409 to indicate an over-range measurement. If the measurement is 410 unavailable, the value 0x1FFF MUST be reported. Values other than 411 0x1FFE,0x1FFF and the valid range defined above MUST NOT be sent 412 and MUST be ignored by the receiving system. 414 4. SDP Signaling 416 [RFC3611] defines the use of SDP (Session Description Protocol) 417 [RFC4566] for signaling the use of XR blocks. However XR blocks MAY 418 be used without prior signaling (see section 5 of RFC3611). 420 4.1. SDP rtcp-xr-attrib Attribute Extension 422 This section augments the SDP [RFC4566] attribute "rtcp-xr" defined 423 in [RFC3611] by providing an additional value of "xr-format" to 424 signal the use of the report block defined in this document. Within 425 the "xr-format", the syntax element "calgextmap" is an attribute as 426 defined in [RFC4566] and used to signal the mapping of the local 427 identifier (CAID) in the segment extension defined in section 3.2 to 428 the calculation algorithm. Specific extensionattributes are defined 429 by the specification that defines a specific extension name; there 430 might be several. 432 xr-format =/ xr-qoe-block 433 xr-qoe-block = "qoe-metrics" ["=" extmap *("," calgextmap)] 434 calgextmap = mapentry "=" extensionname [SP extentionattributes] 435 direction = "sendonly" / "recvonly" / "sendrecv" / "inactive" 436 mapentry = "calg:" 1*5 DIGIT ["/" direction] 437 extensionname = "P564";ITU-T P.564 Compliant Algorithm [P.564] 438 / "G107";ITU-T G.107 [G.107] 439 / "G107_1";ITU-T G.107.1 [G.107.1] 440 / "TS101_329";ETSI TS 101 329-5 Annex E [ ETSI] 441 /"JJ201_1 ";TTC JJ201.1 [TTC] 442 /"P1201_1";ITU-T P.1201.2 [P.1201.1] 443 /"P1201_2";ITU-T P.1201.2 [P.1201.2] 444 /"P1202_1";ITU-T P.1202.1 [P.1202.1] 445 /"P1202_2";ITU-T P.1202.2 [P.1202.2] 446 /"P.862.2";ITU-T P.862.2 [P.862.2] 447 /"P.863"; ITU-T P.863 [P.863] 448 / non-ws-string 449 extentionattributes = mosref 450 /attributes-ext 451 mosref = "mosref=" ("l"; lower resolution 452 /"m"; middle resolution 453 / "h";higher resolution 454 / non-ws-string) 455 mostype = "mostype=" ("e"; Estimated MoS [P.800.1] 456 /"s";subjective MoS [P.800.1] 457 /"o";objective MoS [P.800.1] 458 /non-ws-string) 459 attributes-ext = non-ws-string 460 SP = 461 non-ws-string = 1*(%x21-FF) 463 Each local identifier (CAID)of calculation algorithm used in the 464 segment defined in the section 3.2 is mapped to a string using an 465 attribute of the form: 467 a=calgextmap: ["/"] [] 468 where is a calculation algorithm name, as above, is 469 the local identifier (CAID)of the calculation algorithm associated 470 with the segment defined in this document and is an integer in the 471 valid range inclusive. 473 Example: 474 a=rtcp-xr:qoe-metrics=calg:1=G107,calg:2=P1202_1 476 A usable mapping MUST use IDs in the valid range, and each ID in this 477 range MUST be unique and used only once for each stream or each 478 channel in the stream. 480 The mapping MUST be provided per media stream (in the media-level 481 section(s) of SDP, i.e., after an "m=" line). 483 The syntax element "mosref" is referred to the media resolution 484 relative reference (e.g., Narrowband (3.4kHz) Speech and Standard 485 Definition (SD) Resolution Video with lower resolution, Wideband 486 (7kHz) Speech and High Definition (HD) Resolution Video with higher 487 resolution). MOS scores reported in the QoE block might vary with 488 the MoS reference; For example, MOS values for narrowband, wideband 489 codecs occupy the same range but SHOULD be reported in different 490 value. For video application, MoS scores for SD resolution, HD 491 resolution video also occupy the same ranges and SHOULD be reported 492 in different value. 494 4.2. Offer/Answer Usage 496 When SDP is used in offer-answer context, the SDP Offer/Answer usage 497 defined in [RFC3611] applies. In the offer answer context, the 498 signaling described above might be used in three ways: 500 o asymmetric behavior (segment extensions sent in only one 501 direction), 502 o the offer of mutually exclusive alternatives, or 503 o the offer of more segments than can be sent in a single session. 505 A direction attribute MAY be included in a calgextmap; without it, 506 the direction implicitly inherits, of course, from the RTCP stream 507 direction. 509 Segment extension, with their directions, MAY be signaled for an 510 "inactive" stream. It is an error to use an extension direction 511 incompatible with the stream direction (e.g., a "sendonly" attribute 512 for a "recvonly" stream). 514 If an segment extension is offered as "sendrecv", explicitly or 515 implicitly, and asymmetric behavior is desired, the SDP MAY be 516 modified to modify or add direction qualifiers for that segment 517 extension. 519 A mosref attribute and mos type attribute MAY be included in an 520 calgextmap; without it, the mosref and most type attribute implicitly 521 inherits, of course, from the name attribute (e.g., P.1201.1 522 [P.1201.1] indicates lower resolution used while P.1201.2 [P.1201.2] 523 indicates higher resolution used) or payload type carried in the 524 segment extension (e.g.,EVRC-WB [RFC5188] indicates using Wideband 525 Codec). However not all payload types or MoS algorithm names 526 indicate resolution to be used and mos type to be used. 528 If an answerer receives an offer with an mosref attribute value it 529 doesn't support (e.g.,the answerer only supports "l" and receives 530 "h"from offerer), the answer SHOULD reject the mosref attribute value 531 offered by the offerer. 533 If the answerer wishes to reject a mosref attribute offered by the 534 offerer, it sets identifiers associated with segment extensions in 535 the answer to the value in the range 4096-4351. The rejected answer 536 MUST contain 'mosref ' attribute whose value is the value of the SDP 537 offer. 539 Local identifiers in the valid range inclusive in an offer or answer 540 must not be used more than once per media section. A session update 541 MAY change the direction qualifiers of segment extensions under use. 542 A session update MAY add or remove segment extension(s). Identifiers 543 values in the valid range MUST NOT be altered (remapped). 545 If a party wishes to offer mutually exclusive alternatives, then 546 multiple segment extensions with the same identifier in the 547 (unusable) range 4096-4351 MAY be offered; the answerer SHOULD select 548 at most one of the offered extensions with the same identifier, and 549 remap it to a free identifier in the valid range, for that extension 550 to be usable. Note that two segment types defined in section 3 are 551 also two exclusive alternatives. 553 If more segment extensions are offered in the valid range, the 554 answerer SHOULD choose those that are desired, and place the offered 555 identifier value "as is" in the SDP answer. 557 Similarly, if more segment extensions are offered than can be fit in 558 the valid range, identifiers in the range 4096-4351 MAY be offered; 559 the answerer SHOULD choose those that are desired, and remap them to 560 a free identifier in the valid range. 562 Note that the range 4096-4351 for these negotiation identifiers is 563 deliberately restricted to allow expansion of the range of valid 564 identifiers in future. Segment extensions with an identifier outside 565 the valid range cannot, of course, be used. 567 Example (port numbers, RTP profiles, payload IDs and rtpmaps, etc. 568 all omitted for brevity): 570 The offer: 572 a=rtcp-xr:qoe- metrics=calg:4906=P1201_l,calg:4906=P1202_l,calg: 573 4907=G107 575 The answerer is interested in transmission P.1202.1 on lower 576 resolution application, but doesn't support P.1201.1 on lower 577 resolution application at all. It is interested in transmission 578 G.107. It therefore adjusts the declarations: 580 a=rtcp-xr:qoe-metrics=calg:1=P1202_l,calg:2=G107 582 5. IANA Considerations 584 New block types for RTCP XR are subject to IANA registration. For 585 general guidelines on IANA considerations for RTCP XR, refer to 586 [RFC3611]. 588 5.1. New RTCP XR Block Type value 590 This document assigns the block type value QMB in the IANA " RTP 591 Control Protocol Extended Reports (RTCP XR) Block Type Registry" to 592 the "QoE Metrics Block". 594 [Note to RFC Editor: please replace QMB with the IANA provided RTCP 595 XR block type for this block.] 597 5.2. New RTCP XR SDP Parameter 599 This document also registers a new parameter "qoe-metrics" in the " 600 RTP Control Protocol Extended Reports (RTCP XR) Session Description 601 Protocol (SDP) Parameters Registry". 603 5.3. The SDP calgextmap Attribute 605 This section contains the information required by [RFC4566] for an 606 SDP attribute. 607 o contact name, email address: 609 Qin Wu 610 sunseawq@huawei.com 612 o attribute name (as it will appear in SDP): calgextmap 613 o long-form attribute name in English: calculation algorithm map 614 definition 615 o type of attribute (session level, media level, or both): both 616 o whether the attribute value is subject to the charset attribute: 617 not subject to the charset attribute 618 o a one-paragraph explanation of the purpose of the attribute: This 619 attribute defines the mapping from the local identifier (CAID) in 620 the segment extension defined in section 3.2 into the calculation 621 algorithm name as documented in specifications and appropriately 622 registered. 623 o a specification of appropriate attribute values for this 624 attribute: see RFC xxxx. 626 5.4. New registry of calculation algorithms 628 This document creates a new registry to be called "RTCP XR QoE metric 629 block - multimedia application Calculation Algorithm" as a sub- 630 registry of the "RTP Control Protocol Extended Reports (RTCP XR) 631 Block Type Registry". This registry applies to the multimedia 632 session where each type of media are sent in a separate RTP stream 633 and also applies to the session where Multi-channel audios are 634 carried in one RTP stream. Policies for this new registry are as 635 follows: 637 o The information required to support this assignment is an 638 unambiguous definition of the new metric, covering the base 639 measurements and how they are processed to generate the reported 640 metric. 642 o The review process for the registry is "Specification Required" as 643 described in Section 4.1 of [RFC5226]. 645 o Entries in the registry are identified by entry name and mapped to 646 the local identifier (CAID) in the segment extension defined in 647 section 3.2. 649 o Registration Template 651 The following information must be provided with each registration: 652 * Name: A string uniquely and unambiguously identifying the 653 Calculation algorithm for use in protocols. 654 * Name Description: A valid Description of the Calculation 655 algorithm name. 656 * Reference: The reference which defines the calculation 657 algorithm corresponding to the Name and Name Description. 659 * Type: The media type to which the calculation algorithm is 660 applied 662 o Initial assignments are as follows: 664 Name Name Description Reference Type 665 ========= =================================== ========== ==== 666 P564 ITU-T P.564 Compliant Algorithm [P.564] Voice 667 G107 ITU-T G.107 [G.107] Voice 668 TS101_329 ETSI TS 101 329-5 Annex E [ETSI] Voice 669 JJ201_1 TTC JJ201.1 [TTC] Voice 670 G107_1 ITU-T G.107.1 [G.107.1] Voice 671 P862 ITU-T P.862 [P.862] Voice 672 P862_2 ITU-T P.862.2 [P.862.2] Voice 673 P863 ITU-T P.863 [P.863] Voice 674 P1201_1 ITU-T P.1201.1 [P.1201.1] Multimedia 675 P1201_2 ITU-T P.1201.2 [P.1201.2] Multimedia 676 P1202_1 ITU-T P.1202.1 [P.1202.1] Video 677 P1202_2 ITU-T P.1202.2 [P.1202.2] Video 679 6. Security Considerations 681 The new RTCP XR report blocks proposed in this document introduces no 682 new security considerations beyond those described in [RFC3611]. 684 7. Authors 686 This draft merges ideas from two drafts addressing the QoE metric 687 Reporting issue. The authors of these drafts are listed below (in 688 alphabetical order): 690 Alan Clark < alan.d.clark@telchemy.com > 691 Geoff Hunt < r.geoff.hunt@gmail.com > 692 Martin Kastner < martin.kastner@telchemy.com > 693 Kai Lee < leekai@ctbri.com.cn > 694 Roland Schott < roland.schott@telekom.de > 695 Qin Wu < sunseawq@huawei.com > 696 Glen Zorn < gwz@net-zen.net > 698 8. Acknowledgements 700 The authors gratefully acknowledge the comments and contributions 701 made by Bruce Adams, Philip Arden, Amit Arora, Bob Biskner, Kevin 702 Connor, Claus Dahm, Randy Ethier, Roni Even, Jim Frauenthal, Albert 703 Higashi, Tom Hock, Shane Holthaus, Paul Jones, Rajesh Kumar, Keith 704 Lantz, Mohamed Mostafa, Amy Pendleton, Colin Perkins, Mike Ramalho, 705 Ravi Raviraj, Albrecht Schwarz, Tom Taylor, Bill Ver Steeg, David R 706 Oran, Ali Begen and Hideaki Yamada. 708 9. References 710 9.1. Normative References 712 [ATSC] U.S. Advanced Television Systems Committee (ATSC), "ATSC 713 Standard: Digital Audio Compression (AC-3), Revision B", 714 ATSC Doc A/52B, June 2005. 716 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 717 Requirement Levels", BCP 14, RFC 2119, March 1997. 719 [RFC3550] Schulzrinne, H., "RTP: A Transport Protocol for Real-Time 720 Applications", RFC 3550, July 2003. 722 [RFC3551] Schulzrinne, H. and S. Casner, "RTP Profile for Audio and 723 Video Conferences with Minimal Control", RFC 3551, 724 July 2003. 726 [RFC3611] Friedman, T., Caceres, R., and A. Clark, "RTP Control 727 Protocol Extended Reports (RTCP XR)", RFC 3611, 728 November 2003. 730 [RFC4566] Handley, M., Jacobson, V., and C. Perkins, "SDP: Session 731 Description Protocol", RFC 4566, July 2006. 733 [RFC5226] Narten, T., "Guidelines for Writing an IANA Considerations 734 Section in RFCs", RFC 5226, May 2008. 736 [RFC5234] Crocker, D. and P. Overell, "Augmented BNF for Syntax 737 Specifications: ABNF", STD 68, RFC 5234, January 2008. 739 [RFC6776] Wu, Q., "Measurement Identity and information Reporting 740 using SDES item and XR Block", RFC 6776, October 2012. 742 9.2. Informative References 744 [ETSI] ETSI, "Quality of Service (QoS) measurement 745 methodologies", ETSI TS 101 329-5 V1.1.1, November 2000. 747 [G.107] ITU-T, "The E Model, a computational model for use in 748 transmission planning", ITU-T Recommendation G.107, 749 April 2009. 751 [G.107.1] ITU-T, "Wideband E-model", ITU-T Recommendation G.107.1, 752 December 2011. 754 [G.1082] ITU-T, "Measurement-based methods for improving the 755 robustness of IPTV performance", ITU-T 756 Recommendation G.1082, April 2009. 758 [P.1201.1] 759 ITU-T, "Parametric non-intrusive assessment of audiovisual 760 media streaming quality - lower resolution application 761 area", ITU-T Recommendation P.1201.1, October 2012. 763 [P.1201.2] 764 ITU-T, "Parametric non-intrusive assessment of audiovisual 765 media streaming quality - higher resolution application 766 area", ITU-T Recommendation P.1201.2, October 2012. 768 [P.1202.1] 769 ITU-T, "Parametric non-intrusive bitstream assessment of 770 video media streaming quality - lower resolution 771 application area", ITU-T Recommendation P.1202.1, 772 October 2012. 774 [P.1202.2] 775 ITU-T, "Parametric non-intrusive bitstream assessment of 776 video media streaming quality - higher resolution 777 application area", ITU-T Recommendation P.1202.2, 778 May 2013. 780 [P.564] ITU-T, "Conformance testing for narrowband Voice over IP 781 transmission quality assessment models", ITU-T 782 Recommendation P.564, July 2006. 784 [P.862] ITU-T, "Perceptual evaluation of speech quality (PESQ): An 785 objective method for end-to-end speech quality assessment 786 of narrow-band telephone networks and speech codecs", 787 ITU-T Recommendation P.862, Febuary 2001. 789 [P.862.1] ITU-T, "Mapping function for transforming P.862 raw result 790 scores to MOS-LQO", ITU-T Recommendation P.862.1, 791 November 2003. 793 [P.862.2] ITU-T, "Wideband extension to Recommendation P.862 for the 794 assessment of wideband telephone networks and speech 795 codecs", ITU-T Recommendation P.862.2, November 2007. 797 [P.863] ITU-T, "Perceptual objective listening quality 798 assessment", ITU-T Recommendation P.863, January 2011. 800 [RFC4184] Link, B., Hager, T., and J. Flaks, "RTP Payload Format for 801 AC-3 Audio", RFC 4184, October 2005. 803 [RFC4598] Link, B., "Real-time Transport Protocol (RTP) Payload 804 Format for Enhanced AC-3 (E-AC-3) Audio", RFC 4598, 805 July 2006. 807 [RFC5188] Desineni, H. and Q. Xie, "RTP Payload Format for the 808 Enhanced Variable Rate Wideband Codec (EVRC-WB) and the 809 Media Subtype Updates for EVRC-B Codec", RFC 5188, 810 February 2008. 812 [RFC6390] Clark, A. and B. Claise, "Framework for Performance Metric 813 Development", RFC 6390, October 2011. 815 [RFC6792] Wu, Q., "Monitoring Architectures for RTP", RFC 6792, 816 November 2012. 818 [TTC] TTC 201.01 (Japan), "A method for speech quality 819 assessment for Voice over IP". 821 Appendix A. Example of User Quality of Experience Evaluation for video 822 stream 824 To evaluate user quality of experience levels using objective test 825 data, MoS Scores provide a familiar, easily understood numeric 826 representation of video, audio, and overall audiovisual quality. 827 Unlike audio, video is even more sensitive to transport impairments , 828 and even low rates of packet loss can cause severe degradation in 829 perceived quality. However, all occurrences of packet loss do not 830 have an equal impact on perceptual quality, in part because of the 831 way video frames are structured during the encoding process - such as 832 frame properties including frame type, frame structure and 833 quantization parameter (QP), and in part due to subjective factors - 834 such as the degree to which perception is affected by the levels of 835 motion, detail in the video sequence, and decoder characteristic 836 parameters including media resolution,codec type. When a video 837 stream is sent from the media source to RTP receiving end and get 838 monitored. in order to provide accurate evaluation of video quality, 839 one possible QoE evaluation method is having network nodes that 840 implement network management tools in place. They may know frame 841 properties,perception degree, decoder characteristic parameters of 842 this video stream using some out of band means, gather transport 843 impairment information received from the RTP receiving end and use 844 them as MoS calculation input parameters to calculate MoS scores by 845 choosing appropriate MoS calculation algorithm. Such MoS Scores 846 value can be useful for troubleshooting or comparing video quality 847 across different service types. 849 Appendix B. Metrics represented using RFC6390 Template 851 RFC EDITOR NOTE: please change XXXX in [RFCXXXX] by the new RFC 852 number, when assigned. 854 a. MoS Value Metric 856 * Metric Name: MoS 858 * Metric Description: The estimated mean opinion score for 859 multimedia application performance is defined as including the 860 effects of delay,loss, discard,jitter and other effects that 861 would affect multimedia quality. 863 * Method of Measurement or Calculation: See section 3.2.1, MoS 864 value definition [RFCXXXX]. 866 * Units of Measurement: See section 3.2.1, MoS value definition 867 [RFCXXXX]. 869 * Measurement Point(s) with Potential Measurement Domain: See 870 section 3, 2nd paragraph [RFCXXXX]. 872 * Measurement Timing: See section 3, 3rd paragraph [RFCXXXX] for 873 measurement timing and section 3.1 [RFCXXXX] for Interval 874 Metric flag. 876 * Use and applications: See section 1.4 [RFCXXXX]. 878 * Reporting model: See RFC3611. 880 b. Segment Type Metric 882 * Metric Name: Segment Type 884 * Metric Description: It is used to identify the segment type 885 used in this report block. For more details, see section 886 3.2.1, Segment type definition. 888 * Method of Measurement or Calculation: See section 3.2.1, 889 Segment Type definition [RFCXXXX]. 891 * Units of Measurement: See section 3.2.1, Segment Type 892 definition [RFCXXXX]. 894 * Measurement Point(s) with Potential Measurement Domain: See 895 section 3, 2nd paragraph [RFCXXXX]. 897 * Measurement Timing: See section 3, 3rd paragraph [RFCXXXX] for 898 measurement timing and section 3.1 [RFCXXXX] for Interval 899 Metric flag. 901 * Use and applications: See section 1.4 [RFCXXXX]. 903 * Reporting model: See RFC3611. 905 c. Calg Algorithm Identifier Metric 907 * Metric Name: Calg Algorithm Identifier 909 * Metric Description: It is the local identifier of calculation 910 Algorithm associated with this segment in the range 1-255 911 inclusive. 913 * Method of Measurement or Calculation: See section 3.2.1, Calg 914 Algorithm ID definition [RFCXXXX]. 916 * Units of Measurement: See section 3.2.1, Calg Algorithm ID 917 definition[RFCXXXX]. 919 * Measurement Point(s) with Potential Measurement Domain: See 920 section 3, 2nd paragraph [RFCXXXX]. 922 * Measurement Timing: See section 3, 3rd paragraph [RFCXXXX] for 923 measurement timing and section 3.1 [RFCXXXX] for Interval 924 Metric flag. 926 * Use and applications: See section 1.4 [RFCXXXX]. 928 * Reporting model: See RFC3611. 930 d. Payload Type Metric 932 * Metric Name: Payload Type 934 * Metric Description: It is used to identify the format of the 935 RTP payload. For more details, see section 3.2.1, payload 936 type definition. 938 * Method of Measurement or Calculation: See section 3.2.1, 939 Payload type definition [RFCXXXX]. 941 * Units of Measurement: See section 3.2.1, payload type 942 definition[RFCXXXX]. 944 * Measurement Point(s) with Potential Measurement Domain: See 945 section 3, 2nd paragraph [RFCXXXX]. 947 * Measurement Timing: See section 3, 3rd paragraph [RFCXXXX] for 948 measurement timing and section 3.1 [RFCXXXX] for Interval 949 Metric flag. 951 * Use and applications: See section 1.4 [RFCXXXX]. 953 * Reporting model: See RFC3611. 955 e. Channel Identifier Metric 957 * Metric Name: Payload Type 959 * Metric Description: It is used to identify each channel 960 carried in the same media stream. For more details, see 961 section 3.2.2, channel identifier definition. 963 * Method of Measurement or Calculation: See section 3.2.2, 964 Channel Identifier definition [RFCXXXX]. 966 * Units of Measurement: See section 3.2.2, channel identifier 967 definition[RFCXXXX]. 969 * Measurement Point(s) with Potential Measurement Domain: See 970 section 3, 2nd paragraph [RFCXXXX]. 972 * Measurement Timing: See section 3, 3rd paragraph [RFCXXXX] for 973 measurement timing and section 3.1 [RFCXXXX] for Interval 974 Metric flag. 976 * Use and applications: See section 1.4 [RFCXXXX]. 978 * Reporting model: See RFC3611. 980 Appendix C. Change Log 982 C.1. draft-ietf-xrblock-rtcp-xr-qoe-09 984 The following are the major changes compared to previous version: 986 o Address comments recieved from WGLC, PM-DIR Review and SDP review. 987 o Change an existing SDP attribute 'extmap' to new SDP attribute 988 'calgextmap' and add new SDP attribute registry. 989 o Add Reference to G.107.1, P.862.1, P.862.2 and P.863 for new 990 calculation algorithms. 991 o Add MoS type attribute to distinguish different MoS type. 992 o Other Editorial changes. 994 C.2. draft-ietf-xrblock-rtcp-xr-qoe-08 996 The following are the major changes compared to previous version: 997 o Remove mostype attribute from SDP extension since it can inferred 998 from payload type. 999 o Clarify mosref attribute usage in the O/A. 1001 C.3. draft-ietf-xrblock-rtcp-xr-qoe-07 1003 The following are the major changes compared to previous version: 1004 o Some editorial changes to get in line with burst gap related 1005 draft. 1006 o Add an appendix to apply RFC6390 template. 1008 C.4. draft-ietf-xrblock-rtcp-xr-qoe-06 1010 The following are the major changes compared to previous two 1011 versions: 1012 o A few Contact information update. 1013 o A few Acknowledgement section update. 1015 C.5. draft-ietf-xrblock-rtcp-xr-qoe-04 1017 The following are the major changes compared to previous version: 1018 o Split two references P.NAMS and P.NBAMS into four references. 1019 o SDP signaling update. 1020 o Add one example to explain User QoE evaluation for video stream 1022 C.6. draft-ietf-xrblock-rtcp-xr-qoe-03 1024 The following are the major changes compared to previous version: 1025 o Add one new reference to support TTC JJ201.01. 1026 o Update two references P.NAMS and P.NBAMS. 1027 o Other Editorial changes based on comments applied to PDV and Delay 1028 drafts. 1030 C.7. draft-ietf-xrblock-rtcp-xr-qoe-02 1032 The following are the major changes compared to previous version: 1034 o Remove leftmost second bit since it is ueeless. 1035 o Change 13bits MoS value field into 14 bits to increase MoS 1036 precision. 1037 o Fix some typo and make some editorial changes. 1039 C.8. draft-ietf-xrblock-rtcp-xr-qoe-01 1041 The following are the major changes compared to previous version: 1042 o Remove layered support from the QoE metric draft. 1043 o Allocate 7 bits in the block header for payload type to indicate 1044 what type of payload format is in use and add associated 1045 definition of payload type. 1046 o Clarify using Payload Type to determine the appropriate channel 1047 mapping in the definition of Channel Identifier. 1049 C.9. draft-ietf-xrblock-rtcp-xr-qoe-00 1051 The following are the major changes compared to previous version: 1052 o Allocate one more bit in the single stream per SSC segment to get 1053 alignment with the other two segment type. 1055 Authors' Addresses 1057 Alan Clark 1058 Telchemy Incorporated 1059 2905 Premiere Parkway, Suite 280 1060 Duluth, GA 30097 1061 USA 1063 Email: alan.d.clark@telchemy.com 1065 Qin Wu 1066 Huawei 1067 101 Software Avenue, Yuhua District 1068 Nanjing, Jiangsu 210012 1069 China 1071 Email: sunseawq@huawei.com 1072 Roland Schott 1073 Deutsche Telekom 1074 Heinrich-Hertz-Strasse 3-7 1075 Darmstadt 64295 1076 Germany 1078 Email: Roland.Schott@telekom.de 1080 Glen Zorn 1081 Network Zen 1082 77/440 Soi Phoomjit, Rama IV Road 1083 Phra Khanong, Khlong Toie 1084 Bangkok 10110 1085 Thailand 1087 Phone: +66 (0) 87 502 4274 1088 Email: gwz@net-zen.net