RTP payload format for mU-law EMbedded Codec for Low-delay IP communication (UEMCLIP) speech codec

This document specifies the payload format for sending "mU-law EMbedded Coder for Low-delay IP communication" (UEMCLIP) encoded speech using the Real-time Transport Protocol (RTP) . UEMCLIP is a proprietary codec which enhances u-law ITU-T G.711 , and designed to help the market for smooth transition towards the forthcoming wideband communication environment while achieving a very little media transcoding load with the existing terminals, in which the implementation of G.711 is mandatory. It should be noted that, generally speaking, codecs are negotiated and changed using an SDP exchange. Also, defines general RTP mixer and translator models, where media transcoding may not take place at the node. For those cases, the design concept of the embedded structure is not useful. However, there are other cases when costly transcoding is unavoidable in commonly deployed types of Multi-point Control Units (MCUs) which terminates media and RTCP packets , and when narrowband and wideband terminals co-exist. This embedded bitstream structure can reduce the media transcoding to a simple bitstream truncation. The background and the basic idea of the media format is described in . The details of the payload format are given in . The transcoding issues with G.711 are discussed in , and the considerations for congestion control are in . In , the payload format parameters for a media type registration for UEMCLIP RTP payload format and SDP mappings are provided. The security considerations and IANA considerations are dealt in and , respectively.

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in .

UEMCLIP is an enhanced version of u-law ITU-T G.711, otherwise known as PCMU . It is targeted at Voice over Internet Protocol (VoIP) applications, and its main goal is to provide a wideband communication platform that is highly interoperable with existing terminals equipped with G.711, and to stimulate the market to gradually shift to using wideband communication. In widely deployed multi-point conferencing systems, the packets usually go through RTCP-terminating MCUs, "Topo-RTCP-terminating-MCU" as defined in . Because the G.711 bitstream is embedded in the bitstream, costly media transcoding can be avoided in this case. This document does not discuss the implementation details of the encoder and decoder, but only describes the bitstream format. Because of its scalable nature, there are a number of sub-bitstreams (sub-layer) in a UEMCLIP bitstream. By choosing appropriate sub-layers, the codec can adapt to the following requirements: Sampling frequency, Number of channels, Speech quality, and Bit-rate. The UEMCLIP codec operates at 20-ms frame, and includes three sub-coders as shown in . The core layer is u-law G.711 at 64 kbit/s, and other two are quality and bandwidth enhancement layers with bit-rate of 16 kbit/s each. Layer Description Bit-rate Coding algorithm aG.711 core64u-law PCM bLower-band enhancement16Time domain block quantization cHigher-band16MDCT block quantization Based on these sub-layers, UEMCLIP codec operates in four modes as shown in . Here, "Ch" is the number of channels and "Fs" is the sampling frequency in kHz. It should be noted that the current version only supports single channel operation and there might be future extensions with multi-channel capabilities. The absent Modes 2 and 5 are reserved for possible future extension to 32 kHz sampling modes. As the mode definition is expected to grow, any other modes not defined in this table MUST NOT be used for compatibility and interoperability reasons. Mode Ch Fs Layer a Layer b Layer c Bit-rate w/o headers [kbps] Total bit-rate [kbps] 01 8x--6467.2 1116x-x8084.0 2- ------ 31 8xx-8084.0 4116xxx96100.8 5- ------ UEMCLIP bitstream contains internal headers and other side-information apart from the layer data. This results in total bit-rate larger than the sum of the layers shown in the above table. The detail of the internal headers and auxiliary information are described in . Defining the sampling frequency and the number of channels does not result in a singular mode, i.e., there can be multiple modes for the same sampling frequency or number of channels. The supported modes would differ between implementations, thus the sender and the receiver must negotiate what mode to use for transmission.

As an RTP payload, UEMCLIP bitstream can contain one or more frames as shown in .

0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | RTP Header | +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ | | | one or more frames of UEMCLIP | | | +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ UEMCLIP bitstream has a scalable structure, thus it is possible to reconstruct the signal by decoding a part of it. A UEMCLIP frame is composed of a main header (MH) followed by one or more (up to three) sub-layers (SL) as shown in .

+--+-------+//-+ |MH| SL #1 |...| +--+-------+//-+ As a sub-layer, the core layer, i.e., "Layer a", MUST always be included. It should be noted that the location of the core layer may not immediately follow MH field. The decoder MUST always refer to the layer indices for proper decoding. The UEMCLIP bitstream does not explicitly include the following information: mode and sampling frequency (Fs). As described before, this information MUST be exchanged while establishing a connection, for example, by means of SDP.

Each RTP packet starts with a fixed RTP header, as explained in . The following fields of the RTP fixed header used specifically for UEMCLIP streams are emphasized: The assignment of an RTP payload type for this packet format is outside the scope of this document, however, it is expected that a payload type in the dynamic range shall be assigned. This encodes the sampling instant of the first speech signal sample in the RTP data packet. For UEMCLIP streams, the RTP timestamp MUST advance based on a clock either at 8000 or 16000 (Hz). In cases where the audio sampling rate can change during a session, the RTP timestamp rate MUST equal to the maximum rate (in Hz) given in the mode range (see ). This implies that the RTP timestamp rate for UEMCLIP payload type MUST NOT change during a session. For example, for a UEMCLIP stream with 8-kHz audio sampling, where a transition to a 16-kHz audio sampling mode is allowed, the RTP time stamp must always advance using 16-kHz clock rate. For a fixed audio sampling mode, the RTP timestamp rate should be either 8 or 16 kHz, depending on the sampling rate. If the codec is used for applications with discontinuous transmission (DTX, or silence compression), the first packet after a silence period during which packets have not been transmitted contiguously SHOULD have the marker bit in the RTP data header set to one. The marker bit in all other packets MUST be zero. Applications without DTX MUST set the marker bit to zero.

More than one UEMCLIP frame may be included in a single RTP packet by a sender. However, senders have the following additional restrictions: A single RTP packet SHOULD NOT include more UEMCLIP frames than will fit in the path MTU. All frames contained in a single RTP packet MUST be of the same mode. Frames MUST NOT be split between RTP packets. It is RECOMMENDED that the number of frames contained within an RTP packet be consistent with the application. Since UEMCLIP is designed for telephony application where delay has a great impact on the quality, then fewer frames per packet for lower delay, is preferable.

In a UEMCLIP bitstream, all numbers are encoded in a network byte-order.

The main header (MH) is placed at the top of a frame and has size of 6 bytes. The content of the main header is shown in .

0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | MX | PC | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | PC(cont'd) | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 8 bits Mixing information field. This field is only relevant when Topo-RTCP-terminating-MCUs are utilized to interpret these fields. See for details of the fields. 40 bits Packet-loss concealment (PLC) information field. See .

0 1 2 3 4 5 6 7 +-+-+-+-+-+-+-+-+ |C|R|V| PW1 | |1|1|1| | +-+-+-+-+-+-+-+-+ 1 bit Validity flag of V1 and PW1. This bit being "1" indicates that both parameters are valid, and "0" indicates that the parameters should be ignored. If any of these parameters is invalid, this bit should be set to "0". This flag is mainly intended for a UEMCLIP-conscious Topo-RTCP-terminating-MCU. This flag should be set to "0" in case of upward transcoding from G.711 (See ). 1 bit This bit should be ignored. The default of this bit is 0. 1 bit Voice activity detection flag of the current frame, designed to be used for MCU operations. This flag being "1" indicates that the frame is an active (voice) segment, and "0" indicates that it is an inactive (non-voice) or a silent segment. This flag is specifically designed for mixing information. DTX judgment based this flag is not recommended. 5 bits Signal power code of the current frame. The code is obtained by calculating a root mean square (RMS) of "Layer a" and encoding this RMS using G.711 u-law . Denoting the encoded RMS as R, then PW1 is obtained by PW1 = ((~R)>>2) & 0x1F, where "~", ">>", "&" are one's complement arithmetic, right SHIFT, and bitwise AND operators, respectively.

0 1 2 3 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 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |C|R2 |V| K |U| P1 |U| P2 | PW2 | |2| |2| |1| |2| | | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | R3 | | | +-+-+-+-+-+-+-+-+ 1 bit Validity flag of V2, K, U1, P1, U2, P2, and PW2. If the flag is "1", it means that all these parameters are valid, and "0" means that the parameters should be ignored. If any of these parameters is invalid, this bit should be set to "0". Similarly to C1, this flag should be set to "0" in case of upward transcoding from G.711 (See ). 2 bits These bits should be ignored. The default of these bits are 0. 1 bit Voice activity detection flag of the current frame, designed to be used for packet-loss concealment. This might not be as same as V1 in the mixing information, and might not be synchronous to the marker bit in the RTP header. DTX judgment based this flag is not recommended. 4 bits This value indicates the frame offset of U2, P2 and PW2. Since it is a better idea to carry the speech feature parameters as PLC information in a different frame to maintain the speech quality, this frame offset value gives which frame the parameters are to be associated with. The value ranges between "0" and "15". If the current frame number is N, for example, the value K indicates that U2, P2 and PW2 are associated with frame of N-K. Frame indicator is equal to the difference in the RTP sequence number when one UEMCLIP frame is contained in a single RTP packet. 1 bit Voiced/Unvoiced signal indicator of the current frame. This flag being "0" indicates that the frame is a voiced signal segment, and "1" indicates that it is an unvoiced signal segment. 7 bits Pitch code of the current frame. The actual pitch lag is calculated as P1+20 samples in 8-kHz sampling rate. Pitch lag must be 20 <= pitch length <= 120. Codes ranging between "0x65" and "0x7F" are not used. To obtain the pitch lag, any pitch estimation method can be used, such as the one used in G.711 Appendix I . 1 bit Voiced/Unvoiced signal indicator of the offset frame. This flag being "0" indicates that the frame is a voiced signal segment, and "1" indicates that it is an unvoiced signal segment. The offset value is defined as K. 7 bits Pitch code of the offset frame. The offset value is defined as K. The calculation method is identical to "P1", except that it is based on the signal of offset frame. 8 bits Signal power code of the offset frame. The offset value is defined as K. 8 bits These bits should be ignored. The default of all bits are "0".

Sub-layer (SL) is a sub-header followed by layer bitstreams, as shown in . The sub-header indicates the layer location and the number of bytes.

0 1 2 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 . . . +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+//-+-+-+ |CI |FI |QI |R4 | SB | LD ... | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+//-+-+-+ 2 bits Indicates the channel number. For all modes given in , this should be "0". The detail is given in . 2 bits Indicates the frequency number. "0" means that the layer is in the base frequency band, higher number means that the layer is in respective frequency band. The detail is given in . 2 bits Indicates the quality layer number. "0" means that the layer is in the base layer, and higher number means that the layer is in respective quality layer. The detail is given in . 2 bits Not used (reserved). The default value is "0". 8 bits Indicates the byte size of the following sub-layer data. SB*8 bits The actual sub-layer data. For all the layers shown in , the layer indices are shown in . Layer CI FI QI a000 b001 c010

As given in , u-law encoded G.711 bitstream (Layer a) is the core layer of a UEMCLIP bitstream, and is always embedded. This means that media transcoding from UEMCLIP bitstream to G.711 does not have to undergo decoding and re-encoding procedures, but simple extraction would suffice. However, this does not apply for the reverse procedure, i.e., transcoding from G.711 to UEMCLIP, because the auxiliary information in the main header (MH) must be assigned separately. It should be noted that this media transcoding is useful for a Media Translator (Topo-Media-Translator) or a Point to Multipoint Using RTCP Terminating MCU (Topo-RTCP-terminating-MCU) in and all the requirements apply. This means that a transcoding device of this sort MUST rewrite RTCP packets, together with the RTP media packets. The transcoding from UEMCLIP to u-law G.711 can be done easily by finding an appropriate sub-layer. Within a frame, the transcoder should look for a sub-layer with layer index "0x00", and subsequent LD which has size of SB*8 bits (UEMCLIP has a 20-ms frame thus, SB=160) are the actual G.711 bitstream data. It should be noted that transcoder should not always expect the core layer to be located right after the main header. On the other hand, the transcoding from G.711 to UEMCLIP is not entirely straight-forward. Since there are no means to generate enhancement sub-layers, a G.711 bitstream can only be converted to UEMCLIP Mode 0 bitstream. If the original G.711 bitstream is encoded in A-law, it should first be converted to u-law to become the core layer. Because a UEMCLIP frame size is 20 ms, u-law encoded G.711 bitstream MUST be a 160-sample chunk to become a core layer. For the main header contents, when the UEMCLIP encoder is not available, it should follow the following guidelines. The check bits for mixing and PLC (C1 and C2) are set to 0. The reserved bits (R1 to R3) in MH are set to respective default values. For the core layer (i.e., u-law G.711 bitstream), it should have the following sub-layer header: All CI, FI, QI, and R4 MUST be 0. Sub-layer size (SB) MUST be 160 for 20 ms frame.

The general congestion control considerations for transporting RTP data also apply to UEMCLIP over RTP as well as any applicable RTP profile like AVP . The bandwidth of a UEMCLIP bitstream can be reduced by changing to lower-bit-rate modes. The embedded layer structure of UEMCLIP may help to control congestion, when dynamic mode changing (see ) is available, and the range of modes is obtained by offer-answer negotiation as given in . It should be noted that this involves proper RTCP handling when the bit-rate is modified in an RTP translator or a mixer . Packing more frames in each RTP payload can reduce the number of packets sent, and hence the overhead from IP/UDP/RTP headers, at the expense of increased delay and reduced error robustness against packet losses. It should be treated with care because increased delay means reduced quality.

This registration is done using the template defined in and following . audio UEMCLIP Defines the sampling rate, and MUST be either 8000 or 16000. See "Mode specification" of RFC xxxx (this RFC) for details. See RFC 4566 . See RFC 4566 . Indicates the range of dynamically changeable modes during a session. Possible values are comma separated list of modes from the supported mode set: 0, 1, 3, and 4. If only one mode is specified, it means that the mode must not be changed during the session. When not specified, the mode transmission defaults to a singular mode as specified in . See "Mode specification" of RFC xxxx (this RFC) for details. This media type is framed and contains binary data. See Section 4.8 of RFC 4288. See "Security Considerations" of RFC xxxx (this RFC). This media may be readily transcoded to u-law encoded ITU-T G.711. See "Transcoding between UEMCLIP and G.711" of RFC xxxx (this RFC). RFC xxxx (This RFC) Audio and video streaming and conferencing tools. None COMMON This media type depends on RTP framing, and hence is only defined for transfer via RTP. Yusuke Hiwasaki <hiwasaki.yusuke@lab.ntt.co.jp> Yusuke Hiwasaki IETF Audio/Video Transport Working Group delegated from the IESG

The media types audio/UEMCLIP are mapped to fields in the Session Description Protocol (SDP) as follows: The "m=" line of SDP MUST be audio. Registered media subtype name should be used for the "a=rtpmap" line. Depending on the mode, clock rate (sampling frequency) specified in "a=rtpmap" MUST be selected from the ones defined in . See for details. Since this is an audio stream, the encoding parameters indicate the number of audio channels, and this SHOULD default to "1", as selected from the ones defined in . This is OPTIONAL. A frame length of any UEMCLIP is 20 ms, thus the argument of "a=ptime" SHOULD be a multiple of "20". When not listed in SDP, it should also default to the minimum size: "20". Any description specific to UEMCLIP are defined in the Format Specification Parameters ("a=fmtp"). Each parameter MUST be separated with ";", and if any attributes (value) exists, it MUST be defined with "=". For compatibility reasons, any application/terminal MUST ignore any parameters that it does not understand. This is to ensure the upper-compatibility with parameters added in future enhancements. The mode specification should be made here (see ).

Since UEMCLIP codec can operate in number of modes (bit-rates), it is desirable to specify the range of modes that an encoder or a decoder can operate at. When exchanging SDP messages, an offerer should specify all possible combination of mode numbers as arguments to "mode=" in "a=fmtp" line, delimited by commas ",". In case of specifying multiple modes, those SHOULD appear in the descending priority order. Although UEMCLIP decoders SHOULD accept bitstreams in any modes, an implementation may fail to adopt to the dynamic mode changes during a session. For this reason, an application may choose to operate either with one fixed mode or with multiple modes that can be dynamically changed. If the mode is to be fixed and changes are not allowed, this can be indicated by specifying a single mode per payload type. The mode numbers that can be specified in a payload type as arguments to "mode" are restricted by a combination of a clock rate and a number of audio channels. This is because SDP binds a payload type to a combination of a sampling frequency and a number of audio channels. gives selectable mode numbers that attributed with clock rates. When mode specifications are not given at all, a payload type MUST default to a single mode using the default value specified in this table. Clock rate Channels Selectable modes Default mode 800010,30 1600010,1,3,41 It should be noted that a mode attributed with a larger sampling frequency (Fs) is not used in conjunction with smaller clock rates specified in "a=rtpmap". This means that Modes 0 and 3 can be specified in a payload type having clock rate of both 8000 and 16000 in "a=rtpmap", but Modes 1 and 4 cannot be specified with one having clock rate of 8000.

The procedures related to exchanging SDP messages MUST follow . The following is a detalied list on the semantics of using the UEMCLIP payload format in an offer-answer exchange. An offerer SHOULD offer every possible combination of UEMCLIP payload type it can handle, i.e., sampling frequency, channel number, and fmtp parameters, in a preferred order. When the transmission bandwidth is restricted, it MUST be offered in accordance to the restriction. When multiple UEMCLIP payload types are offered, it is RECOMMENDED that the answerer selects a single UEMCLIP payload type and answers it back. In a UEMCLIP payload type, an answerer MUST answer back suitable mode number(s) as a subset of what has been offered. This means that there is a symmetry assumption on sent and received streams, and offerer MUST NOT send in modes that it does not offer. In an offering/answering SDP, any fmtp parameters which are not known MUST be ignored. If any unknown/undefined parameters should be offered, an answerer MUST delete the entry from the answer message. A receiver of an SDP message MUST only use specified payload types and modes. When a mode specification is missing, i.e., a mode is not specified at all, the session MUST default to one single mode without mode changes during a session. For this case, the default mode values, as shown in , MUST be used based on the sampling frequency and number of channels. This table must be looked up only when there are no mode specifications, thus the offerer/answerer MUST NOT assume that the default modes are always available when it is not in the specified list of modes. When an offered condition does not fit an answerer's capabilities, it naturally MUST NOT answer any of the conditions, and session MAY proceed to re-INVITE, if possible. If a condition (mode) is decided upon, an offerer and an answerer MUST transmit on this condition.

When an offerer indicates that he/she wishes to dynamically switch between modes (0,1,3, and 4) during a session, an example of an offered SDP can be:

v=0 o=john 51050101 51050101 IN IP4 offhost.example.com s=- c=IN IP4 offhost.example.com t=0 0 m=audio 5004 RTP/AVP 96 a=rtpmap:96 UEMCLIP/16000/1 a=fmtp:96 mode=4,1,3,0 It should be noted that the listed modes appears in the offerer's preference. When an answerer can only operate in Modes 1 and 0 but can dynamically switch between those modes during a session, an answerer MUST delete the entries of Mode 3 and 4, and answer back as:

v=0 o=lena 549947322 549947322 IN IP4 anshost.example.org s=- c=IN IP4 anshost.example.org t=0 0 m=audio 5004 RTP/AVP 96 a=rtpmap:96 UEMCLIP/16000/1 a=fmtp:96 mode=1,0 As a result, both would start communicating in either Mode 1 or 0, and can dynamically switch between those modes during the session. On the other hand, when the answerer is capable of communicating either in Modes 1 or 0, and cannot switch between modes during a session, an example of such answer is as follows:

v=0 o=lena 549947322 549947322 IN IP4 anshost.example.org s=- c=IN IP4 anshost.example.org t=0 0 m=audio 5004 RTP/AVP 96 a=rtpmap:96 UEMCLIP/16000/1 a=fmtp:96 mode=1 As a result, both will start communicating in Mode 1. It should be noted that mode change during this session is not allowed because the answerer responded with a single mode, and answerer selected Mode 1 above Mode 0 according to the offered order. If an offerer does not want a mode change during a session but is capable of receiving either Modes 4 or 1 bitstreams, the SDP should somewhat look like:

v=0 o=john 51050101 51050101 IN IP4 offhost.example.com s=- c=IN IP4 offhost.example.com t=0 0 m=audio 5004 RTP/AVP 96 97 a=rtpmap:96 UEMCLIP/16000/1 a=fmtp:96 mode=4 a=rtpmap:97 UEMCLIP/16000/1 a=fmtp:97 mode=1 and if the answerer prefers to communicate in Mode 1, an answer would be:

v=0 o=lena 549947322 549947322 IN IP4 anshost.example.org s=- c=IN IP4 anshost.example.org t=0 0 m=audio 5004 RTP/AVP 97 a=rtpmap:97 UEMCLIP/16000/1 a=fmtp:97 mode=1 Please note that it is RECOMMENDED to select a single UEMCLIP payload type for answers. The "ptime" attribute is used to denote the desired packetization interval. When not specified, it SHOULD default to 20. Since UEMCLIP uses 20 msec frames, ptime values of multiples of 20 imply multiple frames per packet. In the example below, the ptime is set to 60, and this means that offerer wants to receive 3 frames in each packet.

v=0 o=kosuke 2890844730 2890844730 IN IP4 anotherhost.example.com s=- c=IN IP4 anotherhost.example.com t=0 0 m=audio 5004 RTP/AVP 96 a=ptime:60 a=rtpmap:96 UEMCLIP/16000/1 When mode specification is not present, it should default to a fixed mode, and in this case, Mode 1 (see ).

RTP packets using the payload format defined in this specification are subject to the security considerations discussed in the RTP specification and any appropriate profiles. This implies that confidentiality of the media streams is achieved by encryption unless the applicable profile specifies other means. A potential denial-of-service threat exists for data encoding using compression techniques that have non-uniform receiver-end computational load. The attacker can inject pathological datagrams into the stream that are complex to decode and cause the receiver output to become overloaded. However, UEMCLIP covered in this document do not exhibit any significant non-uniformity. Another potential threats are memory attacks by illegal layer indices or byte numbers. The implementor of the decoder should always be aware that the indicated numbers may be corrupted and does not point to the right sub-layer and may force reading beyond the bitstream boundaries. It is advised that a decoder implementation rejects layers such indices.

It is requested that one new media subtype (audio/UEMCLIP) is registered by IANA. For details, see .