Internet Engineering Task Force                                   AVT WG
Internet Draft                                        Schulzrinne/Casner
draft-ietf-avt-profile-new-10.txt              Columbia U./Packet Design
March 2, 2001
Expires: August 2, 2001

    RTP Profile for Audio and Video Conferences with Minimal Control


   This document is an Internet-Draft and is in full conformance with
   all provisions of Section 10 of RFC2026.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress".

   The list of current Internet-Drafts can be accessed at

   To view the list Internet-Draft Shadow Directories, see


   This memorandum is a revision of RFC 1890 in preparation for
   advancement from Proposed Standard to Draft Standard status. Readers
   are encouraged to use the PostScript form of this draft to see where
   changes from RFC 1890 are marked by change bars.

   This document describes a profile called "RTP/AVP" for the use of the
   real-time transport protocol (RTP), version 2, and the associated
   control protocol, RTCP, within audio and video multiparticipant
   conferences with minimal control. It provides interpretations of
   generic fields within the RTP specification suitable for audio and
   video conferences. In particular, this document defines a set of
   default mappings from payload type numbers to encodings.

Schulzrinne/Casner                                            [Page 1]

Internet Draft                  Profile                    March 2, 2001

   This document also describes how audio and video data may be carried
   within RTP. It defines a set of standard encodings and their names
   when used within RTP. The descriptions provide pointers to reference
   implementations and the detailed standards. This document is meant as
   an aid for implementors of audio, video and other real-time
   multimedia applications.

   Resolution of Open Issues

   [Note to the RFC Editor: This section is to be deleted when this
   draft is published as an RFC but is shown here for reference during
   the Last Call. The first paragraph of the Abstract is also to be
   deleted.  All RFC XXXX should be filled in with the number of the RTP
   specification RFC submitted for Draft Standard status, and all RFC
   YYYY should be filled in with the number of the draft specifying MIME
   registration of RTP payload types as it is submitted for Proposed
   Standard status. These latter references are intended to be non-

   Readers are directed to Appendix 9, Changes from RFC 1890, for a
   listing of the changes that have been made in this draft.  The
   changes from RFC 1890 are marked with change bars in the PostScript
   form of this draft.

   The changes in this revision of the draft from the previous one are:

       o An paragraph further explaining the requirements for congestion
         control was added to Section 2 based on the discussion at IETF

       o Packetization of G.726 audio at rates 40, 24 and 16 kb/s is
         specified in addition to 32 kb/s.

       o The mapping of a user pass-phrase string into an encryption key
         was deleted from Section 2 because two interoperable
         implementations were not found.

       o The specification of a two-byte encapsulation for RTP over TCP
         was deleted because two interoperable implementations were not

       o The audio payload formats 1016, G723, GSM-HR and GSM-EFR were
         removed because two interoperable implementations were not

       o The video payload formats H263, BT656, MP2T, MP1S, MP2P and
         BMPEG were removed because two interoperable implementations
         were not found.

Schulzrinne/Casner                                            [Page 2]

Internet Draft                  Profile                    March 2, 2001

   This version of the draft is intended to be complete for Last Call.
   The following open issues from previous drafts have been addressed:

        o The procedure for registering RTP encoding names as MIME
          subtypes was moved to a separate RFC-to-be that may also serve
          to specify how (some of) the encodings here may be used with
          mail and other not-RTP transports. That procedure is not
          required to implement this profile, but may be used in those
          contexts where it is needed.

        o This profile follows the suggestion in the RTP spec that RTCP
          bandwidth may be specified separately from the session
          bandwidth and separately for active senders and passive

        o No specific action is taken in this document to address
          generic payload formats; it is assumed that if any generic
          payload formats are developed, they can be specified in
          separate RFCs and that the session parameters they require for
          operation can be specified in the MIME registration of those

        o The specification of the CN (comfort noise) payload format has
          been removed to a separate draft so that it may be enhanced as
          a result of additional work in ITU-T. That draft is intended
          for publication at Proposed Standard status. Static payload
          type 13 is marked reserved here for the use of that payload
          format (since CN has already been implemented from earlier
          drafts of this profile). Static payload type 19 is also
          reserved because some revisions of the draft assigned that
          number to CN to avoid an historic use of 13.

        o The requirement for congestion control in RTP is addressed in
          the RTP spec with an explanation that the behavior is context
          specific and should be defined in RTP profiles. Text has been
          added to this profile in Section 2 to describe the
          requirements only in general terms because specific algorithms
          have not been devised yet for multicast congestion control.

1 Introduction

   This profile defines aspects of RTP left unspecified in the RTP
   Version 2 protocol definition (RFC XXXX) [1].  This profile is
   intended for the use within audio and video conferences with minimal
   session control. In particular, no support for the negotiation of
   parameters or membership control is provided. The profile is expected
   to be useful in sessions where no negotiation or membership control
   are used (e.g., using the static payload types and the membership
   indications provided by RTCP), but this profile may also be useful in
   conjunction with a higher-level control protocol.

Schulzrinne/Casner                                            [Page 3]

Internet Draft                  Profile                    March 2, 2001

   Use of this profile may be implicit in the use of the appropriate
   applications; there may be no explicit indication by port number,
   protocol identifier or the like.  Applications such as session
   directories may use the name for this profile specified in Section 3.

   Other profiles may make different choices for the items specified

   This document also defines a set of encodings and payload formats for
   audio and video.

1.1 Terminology

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   document are to be interpreted as described in RFC 2119 [2] and
   indicate requirement levels for implementations compliant with this
   RTP profile.

   This draft defines the term media type as dividing encodings of audio
   and video content into three classes: audio, video and audio/video

2 RTP and RTCP Packet Forms and Protocol Behavior

   The section "RTP Profiles and Payload Format Specification" of RFC
   XXXX enumerates a number of items that can be specified or modified
   in a profile. This section addresses these items. Generally, this
   profile follows the default and/or recommended aspects of the RTP

        RTP data header: The standard format of the fixed RTP data
             header is used (one marker bit).

        Payload types: Static payload types are defined in Section 6.

        RTP data header additions: No additional fixed fields are
             appended to the RTP data header.

        RTP data header extensions: No RTP header extensions are
             defined, but applications operating under this profile MAY
             use such extensions. Thus, applications SHOULD NOT assume
             that the RTP header X bit is always zero and SHOULD be
             prepared to ignore the header extension. If a header
             extension is defined in the future, that definition MUST
             specify the contents of the first 16 bits in such a way
             that multiple different extensions can be identified.

Schulzrinne/Casner                                            [Page 4]

Internet Draft                  Profile                    March 2, 2001

        RTCP packet types: No additional RTCP packet types are defined
             by this profile specification.

        RTCP report interval: The suggested constants are to be used for
             the RTCP report interval calculation.  Sessions operating
             under this profile MAY specify a separate parameter for the
             RTCP traffic bandwidth rather than using the default
             fraction of the session bandwidth. The RTCP traffic
             bandwidth MAY be divided into two separate session
             parameters for those participants which are active data
             senders and those which are not. Following the
             recommendation in the RTP specification [1] that 1/4 of the
             RTCP bandwidth be dedicated to data senders, the
             RECOMMENDED default values for these two parameters would
             be 1.25% and 3.75%, respectively. For a particular session,
             the RTCP bandwidth for non-data-senders MAY be set to zero
             when operating on unidirectional links or for sessions that
             don't require feedback on the quality of reception. The
             RTCP bandwidth for data senders SHOULD be kept non-zero so
             that sender reports can still be sent for inter-media
             synchronization and to identify the source by CNAME. The
             means by which the one or two session parameters for RTCP
             bandwidth are specified is beyond the scope of this memo.

        SR/RR extension: No extension section is defined for the RTCP SR
             or RR packet.

        SDES use: Applications MAY use any of the SDES items described
             in the RTP specification. While CNAME information MUST be
             sent every reporting interval, other items SHOULD only be
             sent every third reporting interval, with NAME sent seven
             out of eight times within that slot and the remaining SDES
             items cyclically taking up the eighth slot, as defined in
             Section 6.2.2 of the RTP specification. In other words,
             NAME is sent in RTCP packets 1, 4, 7, 10, 13, 16, 19,
             while, say, EMAIL is used in RTCP packet 22.

        Security: The RTP default security services are also the default
             under this profile.

        String-to-key mapping: No mapping is specified by this profile.

Schulzrinne/Casner                                            [Page 5]

Internet Draft                  Profile                    March 2, 2001

        Congestion: RTP and this profile may be used in the context of
             enhanced network service, for example, through Integrated
             Services (RFC 1633) [3] or Differentiated Services (RFC
             2475) [4], or they may be used with best effort service.

             If enhanced service is being used, RTP receivers SHOULD
             monitor packet loss to ensure that the service that was
             requested is actually being delivered. If it is not, then
             they SHOULD assume that they are receiving best-effort
             service and behave accordingly.

             If best-effort service is being used, RTP receivers SHOULD
             monitor packet loss to ensure that the packet loss rate is
             within acceptable parameters. Packet loss is considered
             acceptable if a TCP flow across the same network path and
             experiencing the same network conditions would achieve an
             average throughput, measured on a reasonable timescale,
             that is not less the RTP flow is achieving. This condition
             can be satisfied by implementing congestion control
             mechanisms to adapt the transmission rate (or the number of
             layers subscribed for a layered multicast session), or by
             arranging for a receiver to leave the session if the loss
             rate is unacceptably high.

             The comparison to TCP cannot be specified exactly, but is
             intended as an "order-of-magnitude" comparison in timescale
             and throughput.  The timescale on which TCP throughput is
             measured is the round-trip time of the connection.  In
             essence, this requirement states that it is not acceptable
             to deploy an application (using RTP or any other transport
             protocol) on the best-effort Internet which consumes
             bandwidth arbitrarily and does not compete fairly with TCP
             within an order of magnitude.

        Underlying protocol: The profile specifies the use of RTP over
             unicast and multicast UDP as well as TCP.  (This does not
             preclude the use of these definitions when RTP is carried
             by other lower-layer protocols.)

        Transport mapping: The standard mapping of RTP and RTCP to
             transport-level addresses is used.

        Encapsulation: This profile leaves to applications the
             specification of RTP encapsulation in protocols other
             than UDP.

Schulzrinne/Casner                                            [Page 6]

Internet Draft                  Profile                    March 2, 2001

3 IANA Considerations

   The RTP specification establishes a registry of profile names for use
   by higher-level control protocols, such as the Session Description
   Protocol (SDP), RFC 2327 [5], to refer to transport methods. This
   profile registers the name "RTP/AVP".

3.1 Registering Additional Encodings

   This profile lists a set of encodings, each of which is comprised of
   a particular media data compression or representation plus a payload
   format for encapsulation within RTP. Some of those payload formats
   are specified here, while others are specified in separate RFCs. It
   is expected that additional encodings beyond the set listed here will
   be created in the future and specified in additional payload format

   This profile also assigns to each encoding a short name which MAY be
   used by higher-level control protocols, such as the Session
   Description Protocol (SDP), RFC 2327 [5], to identify encodings
   selected for a particular RTP session.

   In some contexts it may be useful to refer to these encodings in the
   form of a MIME content-type. To facilitate this, RFC YYYY [6]
   provides registrations for all of the encodings names listed here as
   MIME subtype names under the "audio" and "video" MIME types through
   the MIME registration procedure as specified in RFC 2048 [7].

   Any additional encodings specified for use under this profile (or
   others) may also be assigned names registered as MIME subtypes with
   the Internet Assigned Numbers Authority (IANA). This registry
   provides a means to insure that the names assigned to the additional
   encodings are kept unique. RFC YYYY specifies the information that is
   required for the registration of RTP encodings.

   In addition to assigning names to encodings, this profile also also
   assigns static RTP payload type numbers to some of them. However, the
   payload type number space is relatively small and cannot accommodate
   assignments for all existing and future encodings. During the early
   stages of RTP development, it was necessary to use statically
   assigned payload types because no other mechanism had been specified
   to bind encodings to payload types. It was anticipated that non-RTP
   means beyond the scope of this memo (such as directory services or
   invitation protocols) would be specified to establish a dynamic

Schulzrinne/Casner                                            [Page 7]

Internet Draft                  Profile                    March 2, 2001

   mapping between a payload type and an encoding. Now, mechanisms for
   defining dynamic payload type bindings have been specified in the
   Session Description Protocol (SDP) and in other protocols such as
   ITU-T recommendation H.323/H.245.  These mechanisms associate the
   registered name of the encoding/payload format, along with any
   additional required parameters such as the RTP timestamp clock rate
   and number of channels, to a payload type number.  This association
   is effective only for the duration of the RTP session in which the
   dynamic payload type binding is made. This association applies only
   to the RTP session for which it is made, thus the numbers can be re-
   used for different encodings in different sessions so the number
   space limitation is avoided.

   This profile reserves payload type numbers in the range 96-127
   exclusively for dynamic assignment. Applications SHOULD first use
   values in this range for dynamic payload types. Those applications
   which need to define more than 32 dynamic payload types MAY bind
   codes below 96, in which case it is RECOMMENDED that unassigned
   payload type numbers be used first. However, the statically assigned
   payload types are default bindings and MAY be dynamically bound to
   new encodings if needed. Redefining payload types below 96 may cause
   incorrect operation if an attempt is made to join a session without
   obtaining session description information that defines the dynamic
   payload types.

   Dynamic payload types SHOULD NOT be used without a well-defined
   mechanism to indicate the mapping. Systems that expect to
   interoperate with others operating under this profile SHOULD NOT make
   their own assignments of proprietary encodings to particular, fixed
   payload types.

   This specification establishes the policy that no additional static
   payload types will be assigned beyond the ones defined in this
   document. Establishing this policy avoids the problem of trying to
   create a set of criteria for accepting static assignments and
   encourages the implementation and deployment of the dynamic payload
   type mechanisms.

4 Audio

4.1 Encoding-Independent Rules

   For applications which send either no packets or comfort-noise
   packets during silence, the first packet of a talkspurt, that is, the
   first packet after a silence period, SHOULD be distinguished by
   setting the marker bit in the RTP data header to one. The marker bits
   in all other packets is zero. The beginning of a talkspurt MAY be
   used to adjust the playout delay to reflect changing network delays.

Schulzrinne/Casner                                            [Page 8]

Internet Draft                  Profile                    March 2, 2001

   Applications without silence suppression MUST set the marker bit to

   The RTP clock rate used for generating the RTP timestamp is
   independent of the number of channels and the encoding; it equals the
   number of sampling periods per second. For N-channel encodings, each
   sampling period (say, 1/8000 of a second) generates N samples. (This
   terminology is standard, but somewhat confusing, as the total number
   of samples generated per second is then the sampling rate times the
   channel count.)

   If multiple audio channels are used, channels are numbered left-to-
   right, starting at one. In RTP audio packets, information from
   lower-numbered channels precedes that from higher-numbered channels.
   For more than two channels, the convention followed by the AIFF-C
   audio interchange format SHOULD be followed [8], using the following
   notation, unless some other convention is specified for a particular
   encoding or payload format:

   l  left
   r  right
   c  center
   S  surround
   F  front
   R  rear

   channels  description   channel
                              1     2   3   4   5   6
   2         stereo           l     r
   3                          l     r   c
   4         quadrophonic    Fl     Fr  Rl  Rr
   4                          l     c   r   S
   5                         Fl     Fr  Fc  Sl  Sr
   6                          l     lc  c   r   rc  S

   Samples for all channels belonging to a single sampling instant MUST
   be within the same packet. The interleaving of samples from different
   channels depends on the encoding. General guidelines are given in
   Section 4.3 and 4.4.

   The sampling frequency SHOULD be drawn from the set: 8000, 11025,
   16000, 22050, 24000, 32000, 44100 and 48000 Hz.  (Older Apple
   Macintosh computers had a native sample rate of 22254.54 Hz, which
   can be converted to 22050 with acceptable quality by dropping 4

Schulzrinne/Casner                                            [Page 9]

Internet Draft                  Profile                    March 2, 2001

   samples in a 20 ms frame.)  However, most audio encodings are defined
   for a more restricted set of sampling frequencies. Receivers SHOULD
   be prepared to accept multi-channel audio, but MAY choose to only
   play a single channel.

4.2 Operating Recommendations

   The following recommendations are default operating parameters.
   Applications SHOULD be prepared to handle other values. The ranges
   given are meant to give guidance to application writers, allowing a
   set of applications conforming to these guidelines to interoperate
   without additional negotiation. These guidelines are not intended to
   restrict operating parameters for applications that can negotiate a
   set of interoperable parameters, e.g., through a conference control

   For packetized audio, the default packetization interval SHOULD have
   a duration of 20 ms or one frame, whichever is longer, unless
   otherwise noted in Table 1 (column "ms/packet").  The packetization
   interval determines the minimum end-to-end delay; longer packets
   introduce less header overhead but higher delay and make packet loss
   more noticeable. For non-interactive applications such as lectures or
   for links with severe bandwidth constraints, a higher packetization
   delay MAY be used.  A receiver SHOULD accept packets representing
   between 0 and 200 ms of audio data. (For framed audio encodings, a
   receiver SHOULD accept packets with a number of frames equal to 200
   ms divided by the frame duration, rounded up.) This restriction
   allows reasonable buffer sizing for the receiver.

4.3 Guidelines for Sample-Based Audio Encodings

   In sample-based encodings, each audio sample is represented by a
   fixed number of bits. Within the compressed audio data, codes for
   individual samples may span octet boundaries. An RTP audio packet may
   contain any number of audio samples, subject to the constraint that
   the number of bits per sample times the number of samples per packet
   yields an integral octet count. Fractional encodings produce less
   than one octet per sample.

   The duration of an audio packet is determined by the number of
   samples in the packet.

   For sample-based encodings producing one or more octets per sample,
   samples from different channels sampled at the same sampling instant
   SHOULD be packed in consecutive octets. For example, for a two-
   channel encoding, the octet sequence is (left channel, first sample),
   (right channel, first sample), (left channel, second sample), (right
   channel, second sample), .... For multi-octet encodings, octets

Schulzrinne/Casner                                           [Page 10]

Internet Draft                  Profile                    March 2, 2001

   SHOULD be transmitted in network byte order (i.e., most significant
   octet first).

   The packing of sample-based encodings producing less than one octet
   per sample is encoding-specific.

   The RTP timestamp reflects the instant at which the first sample in
   the packet was sampled, that is, the oldest information in the

4.4 Guidelines for Frame-Based Audio Encodings

   Frame-based encodings encode a fixed-length block of audio into
   another block of compressed data, typically also of fixed length. For
   frame-based encodings, the sender MAY choose to combine several such
   frames into a single RTP packet. The receiver can tell the number of
   frames contained in an RTP packet, if all the frames have the same
   length, by dividing the RTP payload length by the audio frame size
   which is defined as part of the encoding. This does not work when
   carrying frames of different sizes unless the frame sizes are
   relatively prime.  If not, the frames MUST indicate their size.

   For frame-based codecs, the channel order is defined for the whole
   block. That is, for two-channel audio, right and left samples SHOULD
   be coded independently, with the encoded frame for the left channel
   preceding that for the right channel.

   All frame-oriented audio codecs SHOULD be able to encode and decode
   several consecutive frames within a single packet. Since the frame
   size for the frame-oriented codecs is given, there is no need to use
   a separate designation for the same encoding, but with different
   number of frames per packet.

   RTP packets SHALL contain a whole number of frames, with frames
   inserted according to age within a packet, so that the oldest frame
   (to be played first) occurs immediately after the RTP packet header.
   The RTP timestamp reflects the instant at which the first sample in
   the first frame was sampled, that is, the oldest information in the

4.5 Audio Encodings

   The characteristics of the audio encodings described in this document
   are shown in Table 1; they are listed in order of their payload type
   in Table 4. While most audio codecs are only specified for a fixed
   sampling rate, some sample-based algorithms (indicated by an entry of
   "var." in the sampling rate column of Table 1) may be used with

Schulzrinne/Casner                                           [Page 11]

Internet Draft                  Profile                    March 2, 2001

    name of                              sampling              default
    encoding  sample/frame  bits/sample      rate  ms/frame  ms/packet
    DVI4      sample        4                var.                   20
    G722      sample        8              16,000                   20
    G726-40   sample        5               8,000                   20
    G726-32   sample        4               8,000                   20
    G726-24   sample        3               8,000                   20
    G726-16   sample        2               8,000                   20
    G728      frame         N/A             8,000       2.5         20
    G729      frame         N/A             8,000        10         20
    G729D     frame         N/A             8,000        10         20
    G729E     frame         N/A             8,000        10         20
    GSM       frame         N/A             8,000        20         20
    L8        sample        8                var.                   20
    L16       sample        16               var.                   20
    LPC       frame         N/A             8,000        20         20
    MPA       frame         N/A              var.      var.
    PCMA      sample        8                var.                   20
    PCMU      sample        8                var.                   20
    QCELP     frame         N/A             8,000        20         20
    VDVI      sample        var.             var.                   20

   Table 1: Properties of Audio Encodings (N/A:  not  applicable;  var.:

   different sampling rates, resulting in different coded bit rates.
   When used with a sampling rate other than that for which a static
   payload type is defined, non-RTP means beyond the scope of this memo
   MUST be used to define a dynamic payload type and MUST indicate the
   selected RTP timestamp clock rate, which is usually the same as the
   sampling rate for audio.

4.5.1 DVI4

   DVI4 is specified, with pseudo-code, in [9] as the IMA ADPCM wave

   However, the encoding defined here as DVI4 differs in three respects
   from this recommendation:

Schulzrinne/Casner                                           [Page 12]

Internet Draft                  Profile                    March 2, 2001

        o The RTP DVI4 header contains the predicted value rather than
          the first sample value contained the IMA ADPCM block header.

        o IMA ADPCM blocks contain an odd number of samples, since the
          first sample of a block is contained just in the header
          (uncompressed), followed by an even number of compressed
          samples. DVI4 has an even number of compressed samples only,
          using the `predict' word from the header to decode the first

        o For DVI4, the 4-bit samples are packed with the first sample
          in the four most significant bits and the second sample in the
          four least significant bits. In the IMA ADPCM codec, the
          samples are packed in the opposite order.

   Each packet contains a single DVI block. This profile only defines
   the 4-bit-per-sample version, while IMA also specifies a 3-bit-per-
   sample encoding.

   The "header" word for each channel has the following structure:

     int16  predict;  /* predicted value of first sample
                         from the previous block (L16 format) */
     u_int8 index;    /* current index into stepsize table */
     u_int8 reserved; /* set to zero by sender, ignored by receiver */

   Each octet following the header contains two 4-bit samples, thus the
   number of samples per packet MUST be even because there is no means
   to indicate a partially filled last octet.

   Packing of samples for multiple channels is for further study.

   The document IMA Recommended Practices for Enhancing Digital Audio
   Compatibility in Multimedia Systems (version 3.0) contains the
   algorithm description. It is available from

   Interactive Multimedia Association
   48 Maryland Avenue, Suite 202
   Annapolis, MD 21401-8011
   phone: +1 410 626-1380

Schulzrinne/Casner                                           [Page 13]

Internet Draft                  Profile                    March 2, 2001

4.5.2 G722

   G722 is specified in ITU-T Recommendation G.722, "7 kHz audio-coding
   within 64 kbit/s".  The G.722 encoder produces a stream of octets,
   each of which SHALL be octet-aligned in an RTP packet. The first bit
   transmitted in the G.722 octet, which is the most significant bit of
   the higher sub-band sample, SHALL correspond to the most significant
   bit of the octet in the RTP packet.

   Even though the actual sampling rate for G.722 audio is 16000 Hz, the
   RTP clock rate for the G722 payload format is 8000 Hz because that
   value was erroneously assigned in RFC 1890 and must remain unchanged
   for backward compatibility. The octet rate or sample-pair rate is
   8000 Hz.

4.5.3 G726-40, G726-32, G726-24, and G726-16

   ITU-T Recommendation G.726 describes, among others, the algorithm
   recommended for conversion of a single 64 kbit/s A-law or mu-law
   PCM channel encoded at 8000 samples/sec to and from a 40, 32, 24,
   or 16 kbit/s channel.  The conversion is applied to the PCM stream
   using an Adaptive Differential Pulse Code Modulation (ADPCM)
   transcoding technique.  The ADPCM representation consists of a
   series of codewords with a one-to-one correspondance to the samples
   in the PCM stream.  The G726 data rates of 40, 32, 24, and 16
   kbit/s have codewords of 5, 4, 3, and 2 bits respectively.

   The 16 and 24 kbit/s encodings do not provide toll quality speech.
   They are designed for used in overloaded Digital Circuit
   Multiplication Equipment (DCME).  ITU-T G.726 recommends that the
   16 and 24 kbit/s encodings should be alternated with higher data
   rate encodings to provide an average sample size of between 3.5 and
   3.7 bits per sample.

   The encodings of G.726 are here denoted as G726-40, G726-32,
   G726-24, and G726-16.  Prior to 1990, G721 described the 32 kbit/s
   ADPCM encoding, and G723 described the 40, 32, and 16 kbit/s
   encodings.  Thus, G726-32 designates the same algorithm as G721 in
   RFC 1890.

   A stream of G726 codewords contains no information on the encoding
   being used, therefore transitions between G726 encoding types is
   not permitted within a sequence of packed codewords.  Applications
   MUST determine the encoding type of packed codewords from the RTP
   payload identifier.

Schulzrinne/Casner                                           [Page 14]

Internet Draft                  Profile                    March 2, 2001

   No payload-specific header information SHALL be included as part
   of the audio data. A stream of G726 codewords MUST be packed into
   octets as follows: the first codeword is placed into the first
   octet such that the least significant bit of the codeword aligns
   with the least significant bit in the octet, the second codeword
   is then packed so that its least significant bit coincides with
   the least significant unoccupied bit in the octet.  When a
   complete codeword cannot be placed into an octet, the bits
   overlapping the octet boundary are placed into the least
   significant bits of the next octet.  Packing MUST end with a
   completely packed final octet.  The number of codewords packed
   will therefore be a multiple of 8, 2, 8, and 4 for G726-40,
   G726-32, G726-24, and G726-16 respectively.  An examples of the
   packing scheme for G726-32 codewords is as shown:

     0                   1
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
    |B B B B|A A A A|D D D D|C C C C| ...
    |0 1 2 3|0 1 2 3|0 1 2 3|0 1 2 3|

    An example of the packing scheme for G726-24 codewords is:

     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
    |C C|B B B|A A A|F|E E E|D D D|C|H H H|G G G|F F| ...
    |1 2|0 1 2|0 1 2|2|0 1 2|0 1 2|0|0 1 2|0 1 2|0 1|

Schulzrinne/Casner                                           [Page 15]

Internet Draft                  Profile                    March 2, 2001

4.5.4 G728

   G728 is specified in ITU-T Recommendation G.728, "Coding of speech at
   16 kbit/s using low-delay code excited linear prediction".

   A G.278 encoder translates 5 consecutive audio samples into a 10-bit
   codebook index, resulting in a bit rate of 16 kb/s for audio sampled
   at 8,000 samples per second. The group of five consecutive samples is
   called a vector. Four consecutive vectors, labeled V1 to V4 (where V1
   is to be played first by the receiver), build one G.728 frame. The
   four vectors of 40 bits are packed into 5 octets, labeled B1 through
   B5. B1 SHALL be placed first in the RTP packet.

   Referring to the figure below, the principle for bit order is
   "maintenance of bit significance". Bits from an older vector are more
   significant than bits from newer vectors. The MSB of the frame goes
   to the MSB of B1 and the LSB of the frame goes to LSB of B5.

             1         2         3        3
   0         0         0         0        9
   <---V1---><---V2---><---V3---><---V4---> vectors
   <--B1--><--B2--><--B3--><--B4--><--B5--> octets
   <------------- frame 1 ---------------->

   In particular, B1 contains the eight most significant bits of V1,
   with the MSB of V1 being the MSB of B1. B2 contains the two least
   significant bits of V1, the more significant of the two in its MSB,
   and the six most significant bits of V2. B1 SHALL be placed first in
   the RTP packet and B5 last.

4.5.5 G729

   G729 is specified in ITU-T Recommendation G.729, "Coding of speech at
   8 kbit/s using conjugate structure-algebraic code excited linear
   prediction (CS-ACELP)". A reduced-complexity version of the G.729
   algorithm is specified in Annex A to Rec. G.729. The speech coding
   algorithms in the main body of G.729 and in G.729 Annex A are fully
   interoperable with each other, so there is no need to further
   distinguish between them. The G.729 and G.729 Annex A codecs were
   optimized to represent speech with high quality, where G.729 Annex A
   trades some speech quality for an approximate 50% complexity

Schulzrinne/Casner                                           [Page 16]

Internet Draft                  Profile                    March 2, 2001

   reduction [10]. See the next Section (4.5.6) for other data rates
   added in later G.729 Annexes. For all data rates, the sampling
   frequency (and RTP timestamp clock rate) is 8000 Hz.

   A voice activity detector (VAD) and comfort noise generator (CNG)
   algorithm in Annex B of G.729 is RECOMMENDED for digital simultaneous
   voice and data applications and can be used in conjunction with G.729
   or G.729 Annex A. A G.729 or G.729 Annex A frame contains 10 octets,
   while the G.729 Annex B comfort noise frame occupies 2 octets.

   A G729 RTP packet may consist of zero or more G.729 or G.729 Annex A
   frames, followed by zero or one G.729 Annex B frames. The presence of
   a comfort noise frame can be deduced from the length of the RTP
   payload. The default packetization interval is 20 ms (two frames),
   but in some situations it may be desireable to send 10 ms packets. An
   example would be a transition from speech to comfort noise in the
   first 10 ms of the packet. For some applications, a longer
   packetization interval may be required to reduce the packet rate.

   The transmitted parameters of a G.729/G.729A 10-ms frame, consisting
   of 80 bits, are defined in Recommendation G.729, Table 8/G.729. The
   mapping of the these parameters is given below in Fig. 4.  The
   diagrams show the bit packing in "network byte order," also known as
   big-endian order. The bits of each 32-bit word are numbered 0 to 31,
   with the most significant bit on the left and numbered 0. The octets
   (bytes) of each word are transmitted most significant octet first.
   The bits of each data field are numbered in the order as produced by
   the G.729 C code reference implementation.

     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
    |L|      L1     |    L2   |    L3   |       P1      |P|    C1   |
    |0|             |         |         |               |0|         |
    | |0 1 2 3 4 5 6|0 1 2 3 4|0 1 2 3 4|0 1 2 3 4 5 6 7| |0 1 2 3 4|
    |       C1      |  S1   | GA1 |  GB1  |    P2   |      C2       |
    |          1 1 1|       |     |       |         |               |
    |5 6 7 8 9 0 1 2|0 1 2 3|0 1 2|0 1 2 3|0 1 2 3 4|0 1 2 3 4 5 6 7|
    |   C2    |  S2   | GA2 |  GB2  |
    |    1 1 1|       |     |       |
    |8 9 0 1 2|0 1 2 3|0 1 2|0 1 2 3|

Schulzrinne/Casner                                           [Page 17]

Internet Draft                  Profile                    March 2, 2001

   The packing of the G.729 Annex B comfort noise frame is as follows:

     0                   1
     0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5
    |L|  LSF1   |  LSF2 |   GAIN  |R|
    |S|         |       |         |E|
    |F|         |       |         |S|
    |0|0 1 2 3 4|0 1 2 3|0 1 2 3 4|V|    RESV = Reserved (zero)

4.5.6 G729D and G729E

   Annexes D and E to ITU-T Recommendation G.729 provide additional data
   rates. Because the data rate is not signaled in the bitstream, the
   different data rates are given distinct RTP encoding names which are
   mapped to distinct payload type numbers. G729D indicates a 6.4 kbit/s
   coding mode (G.729 Annex D, for momentary reduction in channel
   capacity), while G729E indicates an 11.8 kbit/s mode (G.729 Annex E,
   for improved performance with a wide range of narrow-band input
   signals, e.g. music and background noise). Annex E has two operating
   modes, backward adaptive and forward adaptive, which are signaled by
   the first two bits in each frame (the most significant two bits of
   the first octet).

   The voice activity detector (VAD) and comfort noise generator (CNG)
   algorithm specified in Annex B of G.729 may be used with Annex D and
   Annex E frames in addition to G.729 and G.729 Annex A frames. The
   algorithm details for the operation of Annexes D and E with the Annex
   B CNG are specified in G.729 Annexes F and G. Note that Annexes F and
   G do not introduce any new encodings.

   For G729D, an RTP packet may consist of zero or more G.729 Annex D
   frames, followed by zero or one G.729 Annex B frame. Similarly, for
   G729E, an RTP packet may consist of zero or more G.729 Annex E
   frames, followed by zero or one G.729 Annex B frame. The presence of
   a comfort noise frame can be deduced from the length of the RTP

   A single RTP packet must contain frames of only one data rate,
   optionally followed by one comfort noise frame. The data rate may be
   changed from packet to packet by changing the payload type number.
   G.729 Annexes D, E and H describe what the encoding and decoding
   algorithms must do to accommodate a change in data rate.

Schulzrinne/Casner                                           [Page 18]

Internet Draft                  Profile                    March 2, 2001

   For G729D, the bits of a G.729 Annex D frame are formatted as shown
   below in Fig. 6 (cf. Table D.1/G.729). The frame length is 64 bits.

     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
    |L|      L1     |    L2   |    L3   |        P1     |     C1    |
    |0|             |         |         |               |           |
    | |0 1 2 3 4 5 6|0 1 2 3 4|0 1 2 3 4|0 1 2 3 4 5 6 7|0 1 2 3 4 5|
    | C1  |S1 | GA1 | GB1 |  P2   |        C2       |S2 | GA2 | GB2 |
    |     |   |     |     |       |                 |   |     |     |
    |6 7 8|0 1|0 1 2|0 1 2|0 1 2 3|0 1 2 3 4 5 6 7 8|0 1|0 1 2|0 1 2|

   The net bit rate for the G.729 Annex E algorithm is 11.8 kbit/s and a
   total of 118 bits are used. Two bits are appended as "don't care"
   bits to complete an integer number of octets for the frame. For
   G729E, the bits of a data frame are formatted as shown in the next
   two diagrams (cf. Table E.1/G.729). The fields for the G729E forward
   adaptive mode are packed as follows:

     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
    |0 0|L|      L1     |    L2   |    L3   |        P1     |P| C0_1|
    |   |0|             |         |         |               |0|     |
    |   | |0 1 2 3 4 5 6|0 1 2 3 4|0 1 2 3 4|0 1 2 3 4 5 6 7| |0 1 2|
    |       |   C1_1      |     C2_1    |   C3_1      |    C4_1     |
    |       |             |             |             |             |
    |3 4 5 6|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2 3 4 5 6|
    | GA1 |  GB1  |    P2   |   C0_2      |     C1_2    |   C2_2    |
    |     |       |         |             |             |           |
    |0 1 2|0 1 2 3|0 1 2 3 4|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2 3 4 5|
    | |    C3_2     |     C4_2    | GA2 | GB2   |DC |
    | |             |             |     |       |   |
    |6|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2|0 1 2 3|0 1|

Schulzrinne/Casner                                           [Page 19]

Internet Draft                  Profile                    March 2, 2001

   The fields for the G729E backward adaptive mode are packed as shown
   in Fig. 8.

     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
    |1 1|       P1      |P|       C0_1              |     C1_1      |
    |   |               |0|                    1 1 1|               |
    |   |0 1 2 3 4 5 6 7|0|0 1 2 3 4 5 6 7 8 9 0 1 2|0 1 2 3 4 5 6 7|
    |   |  C2_1       | C3_1        | C4_1        |GA1  | GB1   |P2 |
    |   |             |             |             |     |       |   |
    |8 9|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2|0 1 2 3|0 1|
    |     |          C0_2           |       C1_2        |    C2_2   |
    |     |                    1 1 1|                   |           |
    |2 3 4|0 1 2 3 4 5 6 7 8 9 0 1 2|0 1 2 3 4 5 6 7 8 9|0 1 2 3 4 5|
    | |    C3_2     |     C4_2    | GA2 | GB2   |DC |
    | |             |             |     |       |   |
    |6|0 1 2 3 4 5 6|0 1 2 3 4 5 6|0 1 2|0 1 2 3|0 1|

Schulzrinne/Casner                                           [Page 20]

Internet Draft                  Profile                    March 2, 2001

4.5.7 GSM

   GSM (group speciale mobile) denotes the European GSM 06.10 standard
   for full-rate speech transcoding, ETS 300 961, which is based on
   RPE/LTP (residual pulse excitation/long term prediction) coding at a
   rate of 13 kb/s [15,16,17]. The text of the standard can be obtained

   ETSI (European Telecommunications Standards Institute)
   ETSI Secretariat: B.P.152
   F-06561 Valbonne Cedex
   Phone: +33 92 94 42 00
   Fax: +33 93 65 47 16

   Blocks of 160 audio samples are compressed into 33 octets, for an
   effective data rate of 13,200 b/s. General Packaging Issues

   The GSM standard (ETS 300 961) specifies the bit stream produced by
   the codec, but does not specify how these bits should be packed for
   transmission. The packetization specified here has subsequently been

   adopted in ETSI Technical Specification TS 101 318.  Some software
   implementations of the GSM codec use a different packing than that
   specified here.

   In the GSM packing used by RTP, the bits SHALL be packed beginning
   from the most significant bit. Every 160 sample GSM frame is coded
   into one 33 octet (264 bit) buffer. Every such buffer begins with a 4
   bit signature (0xD), followed by the MSB encoding of the fields of
   the frame. The first octet thus contains 1101 in the 4 most
   significant bits (0-3) and the 4 most significant bits of F1 (0-3) in
   the 4 least significant bits (4-7). The second octet contains the 2
   least significant bits of F1 in bits 0-1, and F2 in bits 2-7, and so
   on. The order of the fields in the frame is described in Table 2. GSM variable names and numbers

   In the RTP encoding we have the bit pattern described in Table 3,
   where F.i signifies the ith bit of the field F, bit 0 is the most
   significant bit, and the bits of every octet are numbered from 0 to 7
   from most to least significant.

4.5.8 L8

   L8 denotes linear audio data samples, using 8-bits of precision with
   an offset of 128, that is, the most negative signal is encoded as

Schulzrinne/Casner                                           [Page 21]

Internet Draft                  Profile                    March 2, 2001

             field  field name  bits  field  field name  bits
             1      LARc[0]     6     39     xmc[22]     3
             2      LARc[1]     6     40     xmc[23]     3
             3      LARc[2]     5     41     xmc[24]     3
             4      LARc[3]     5     42     xmc[25]     3
             5      LARc[4]     4     43     Nc[2]       7
             6      LARc[5]     4     44     bc[2]       2
             7      LARc[6]     3     45     Mc[2]       2
             8      LARc[7]     3     46     xmaxc[2]    6
             9      Nc[0]       7     47     xmc[26]     3
             10     bc[0]       2     48     xmc[27]     3
             11     Mc[0]       2     49     xmc[28]     3
             12     xmaxc[0]    6     50     xmc[29]     3
             13     xmc[0]      3     51     xmc[30]     3
             14     xmc[1]      3     52     xmc[31]     3
             15     xmc[2]      3     53     xmc[32]     3
             16     xmc[3]      3     54     xmc[33]     3
             17     xmc[4]      3     55     xmc[34]     3
             18     xmc[5]      3     56     xmc[35]     3
             19     xmc[6]      3     57     xmc[36]     3
             20     xmc[7]      3     58     xmc[37]     3
             21     xmc[8]      3     59     xmc[38]     3
             22     xmc[9]      3     60     Nc[3]       7
             23     xmc[10]     3     61     bc[3]       2
             24     xmc[11]     3     62     Mc[3]       2
             25     xmc[12]     3     63     xmaxc[3]    6
             26     Nc[1]       7     64     xmc[39]     3
             27     bc[1]       2     65     xmc[40]     3
             28     Mc[1]       2     66     xmc[41]     3
             29     xmaxc[1]    6     67     xmc[42]     3
             30     xmc[13]     3     68     xmc[43]     3
             31     xmc[14]     3     69     xmc[44]     3
             32     xmc[15]     3     70     xmc[45]     3
             33     xmc[16]     3     71     xmc[46]     3
             34     xmc[17]     3     72     xmc[47]     3
             35     xmc[18]     3     73     xmc[48]     3
             36     xmc[19]     3     74     xmc[49]     3
             37     xmc[20]     3     75     xmc[50]     3
             38     xmc[21]     3     76     xmc[51]     3

   Table 2: Ordering of GSM variables

4.5.9 L16

   L16 denotes uncompressed audio data samples, using 16-bit signed
   representation with 65535 equally divided steps between minimum and

Schulzrinne/Casner                                           [Page 22]

Internet Draft                  Profile                    March 2, 2001

   Octet   Bit 0    Bit 1    Bit 2    Bit 3    Bit 4    Bit 5    Bit 6    Bit 7
       0     1        1        0        1     LARc0.0  LARc0.1  LARc0.2  LARc0.3
       1  LARc0.4  LARc0.5  LARc1.0  LARc1.1  LARc1.2  LARc1.3  LARc1.4  LARc1.5
       2  LARc2.0  LARc2.1  LARc2.2  LARc2.3  LARc2.4  LARc3.0  LARc3.1  LARc3.2
       3  LARc3.3  LARc3.4  LARc4.0  LARc4.1  LARc4.2  LARc4.3  LARc5.0  LARc5.1
       4  LARc5.2  LARc5.3  LARc6.0  LARc6.1  LARc6.2  LARc7.0  LARc7.1  LARc7.2
       5   Nc0.0    Nc0.1    Nc0.2    Nc0.3    Nc0.4    Nc0.5    Nc0.6   bc0.0
       6   bc0.1    Mc0.0    Mc0.1   xmaxc00  xmaxc01  xmaxc02  xmaxc03  xmaxc04
       7  xmaxc05  xmc0.0   xmc0.1   xmc0.2   xmc1.0   xmc1.1   xmc1.2   xmc2.0
       8  xmc2.1   xmc2.2   xmc3.0   xmc3.1   xmc3.2   xmc4.0   xmc4.1   xmc4.2
       9  xmc5.0   xmc5.1   xmc5.2   xmc6.0   xmc6.1   xmc6.2   xmc7.0   xmc7.1
      10  xmc7.2   xmc8.0   xmc8.1   xmc8.2   xmc9.0   xmc9.1   xmc9.2   xmc10.0
      11  xmc10.1  xmc10.2  xmc11.0  xmc11.1  xmc11.2  xmc12.0  xmc12.1  xcm12.2
      12   Nc1.0    Nc1.1    Nc1.2    Nc1.3    Nc1.4    Nc1.5    Nc1.6    bc1.0
      13   bc1.1    Mc1.0    Mc1.1   xmaxc10  xmaxc11  xmaxc12  xmaxc13  xmaxc14
      14  xmax15   xmc13.0  xmc13.1  xmc13.2  xmc14.0  xmc14.1  xmc14.2  xmc15.0
      15  xmc15.1  xmc15.2  xmc16.0  xmc16.1  xmc16.2  xmc17.0  xmc17.1  xmc17.2
      16  xmc18.0  xmc18.1  xmc18.2  xmc19.0  xmc19.1  xmc19.2  xmc20.0  xmc20.1
      17  xmc20.2  xmc21.0  xmc21.1  xmc21.2  xmc22.0  xmc22.1  xmc22.2  xmc23.0
      18  xmc23.1  xmc23.2  xmc24.0  xmc24.1  xmc24.2  xmc25.0  xmc25.1  xmc25.2
      19   Nc2.0    Nc2.1    Nc2.2    Nc2.3    Nc2.4    Nc2.5    Nc2.6    bc2.0
      20   bc2.1    Mc2.0    Mc2.1   xmaxc20  xmaxc21  xmaxc22  xmaxc23  xmaxc24
      21  xmaxc25  xmc26.0  xmc26.1  xmc26.2  xmc27.0  xmc27.1  xmc27.2  xmc28.0
      22  xmc28.1  xmc28.2  xmc29.0  xmc29.1  xmc29.2  xmc30.0  xmc30.1  xmc30.2
      23  xmc31.0  xmc31.1  xmc31.2  xmc32.0  xmc32.1  xmc32.2  xmc33.0  xmc33.1
      24  xmc33.2  xmc34.0  xmc34.1  xmc34.2  xmc35.0  xmc35.1  xmc35.2  xmc36.0
      25  Xmc36.1  xmc36.2  xmc37.0  xmc37.1  xmc37.2  xmc38.0  xmc38.1  xmc38.2
      26   Nc3.0    Nc3.1    Nc3.2    Nc3.3    Nc3.4    Nc3.5    Nc3.6    bc3.0
      27   bc3.1    Mc3.0    Mc3.1   xmaxc30  xmaxc31  xmaxc32  xmaxc33  xmaxc34
      28  xmaxc35  xmc39.0  xmc39.1  xmc39.2  xmc40.0  xmc40.1  xmc40.2  xmc41.0
      29  xmc41.1  xmc41.2  xmc42.0  xmc42.1  xmc42.2  xmc43.0  xmc43.1  xmc43.2
      30  xmc44.0  xmc44.1  xmc44.2  xmc45.0  xmc45.1  xmc45.2  xmc46.0  xmc46.1
      31  xmc46.2  xmc47.0  xmc47.1  xmc47.2  xmc48.0  xmc48.1  xmc48.2  xmc49.0
      32  xmc49.1  xmc49.2  xmc50.0  xmc50.1  xmc50.2  xmc51.0  xmc51.1  xmc51.2

   Table 3: GSM payload format

   maximum signal level, ranging from -32768 to 32767. The value is
   represented in two's complement notation and transmitted in network
   byte order (most significant byte first).

4.5.10 LPC

   LPC designates an experimental linear predictive encoding contributed
   by Ron Frederick, which is based on an implementation written by Ron
   Zuckerman posted to the Usenet group comp.dsp on June 26, 1992.  The

Schulzrinne/Casner                                           [Page 23]

Internet Draft                  Profile                    March 2, 2001

   codec generates 14 octets for every frame. The framesize is set to 20
   ms, resulting in a bit rate of 5,600 b/s.

4.5.11 MPA

   MPA denotes MPEG-1 or MPEG-2 audio encapsulated as elementary
   streams.  The encoding is defined in ISO standards ISO/IEC 11172-3
   and 13818-3.  The encapsulation is specified in RFC 2250 [14].

   The encoding may be at any of three levels of complexity, called
   Layer I, II and III. The selected layer as well as the sampling rate
   and channel count are indicated in the payload. The RTP timestamp
   clock rate is always 90000, independent of the sampling rate.  MPEG-1
   audio supports sampling rates of 32, 44.1, and 48 kHz (ISO/IEC
   11172-3, section 1.1; "Scope"). MPEG-2 supports sampling rates of 16,
   22.05 and 24 kHz.  The number of samples per frame is fixed, but the
   frame size will vary with the sampling rate and bit rate.

4.5.12 PCMA and PCMU

   PCMA and PCMU are specified in ITU-T Recommendation G.711. Audio data
   is encoded as eight bits per sample, after logarithmic scaling. PCMU
   denotes mu-law scaling, PCMA A-law scaling. A detailed description is
   given by Jayant and Noll [15].  Each G.711 octet SHALL be octet-
   aligned in an RTP packet. The sign bit of each G.711 octet SHALL
   correspond to the most significant bit of the octet in the RTP packet
   (i.e., assuming the G.711 samples are handled as octets on the host
   machine, the sign bit SHALL be the most signficant bit of the octet
   as defined by the host machine format). The 56 kb/s and 48 kb/s modes
   of G.711 are not applicable to RTP, since PCMA and PCMU MUST always
   be transmitted as 8-bit samples.

4.5.13 QCELP

   The Electronic Industries Association (EIA) & Telecommunications
   Industry Association (TIA) standard IS-733, "TR45: High Rate Speech
   Service Option for Wideband Spread Spectrum Communications Systems,"
   defines the QCELP audio compression algorithm for use in wireless
   CDMA applications. The QCELP CODEC compresses each 20 milliseconds of
   8000 Hz, 16- bit sampled input speech into one of four different size
   output frames: Rate 1 (266 bits), Rate 1/2 (124 bits), Rate 1/4 (54
   bits) or Rate 1/8 (20 bits). For typical speech patterns, this
   results in an average output of 6.8 k bits/sec for normal mode and
   4.7 k bits/sec for reduced rate mode. The packetization of the QCELP
   audio codec is described in [16].

4.5.14 RED

Schulzrinne/Casner                                           [Page 24]

Internet Draft                  Profile                    March 2, 2001

   The redundant audio payload format "RED" is specified by RFC 2198
   [17]. It defines a means by which multiple redundant copies of an
   audio packet may be transmitted in a single RTP stream. Each packet
   in such a stream contains, in addition to the audio data for that
   packetization interval, a (more heavily compressed) copy of the data
   from a previous packetization interval. This allows an approximation
   of the data from lost packets to be recovered upon decoding of a
   subsequent packet, giving much improved sound quality when compared
   with silence substitution for lost packets.

4.5.15 VDVI

   VDVI is a variable-rate version of DVI4, yielding speech bit rates of
   between 10 and 25 kb/s. It is specified for single-channel operation
   only.  Samples are packed into octets starting at the most-
   significant bit.  The last octet is padded with 1 bits if the last
   sample does not fill the last octet. This padding is distinct from
   the valid codewords.  The receiver needs to detect the padding
   because there is no explicit count of samples in the packet.

   It uses the following encoding:

                      DVI4 codeword  VDVI bit pattern
                                  0  00
                                  1  010
                                  2  1100
                                  3  11100
                                  4  111100
                                  5  1111100
                                  6  11111100
                                  7  11111110
                                  8  10
                                  9  011
                                 10  1101
                                 11  11101
                                 12  111101
                                 13  1111101
                                 14  11111101
                                 15  11111111

5 Video

   The following sections describe the video encodings that are defined
   in this memo and give their abbreviated names used for
   identification.  These video encodings and their payload types are
   listed in Table 5.

Schulzrinne/Casner                                           [Page 25]

Internet Draft                  Profile                    March 2, 2001

   All of these video encodings use an RTP timestamp frequency of 90,000
   Hz, the same as the MPEG presentation time stamp frequency. This
   frequency yields exact integer timestamp increments for the typical
   24 (HDTV), 25 (PAL), and 29.97 (NTSC) and 30 Hz (HDTV) frame rates
   and 50, 59.94 and 60 Hz field rates. While 90 kHz is the RECOMMENDED
   rate for future video encodings used within this profile, other rates
   MAY be used.  However, it is not sufficient to use the video frame
   rate (typically between 15 and 30 Hz) because that does not provide
   adequate resolution for typical synchronization requirements when
   calculating the RTP timestamp corresponding to the NTP timestamp in
   an RTCP SR packet. The timestamp resolution MUST also be sufficient
   for the jitter estimate contained in the receiver reports.

   For most of these video encodings, the RTP timestamp encodes the
   sampling instant of the video image contained in the RTP data packet.
   If a video image occupies more than one packet, the timestamp is the
   same on all of those packets. Packets from different video images are
   distinguished by their different timestamps.

   Most of these video encodings also specify that the marker bit of the
   RTP header SHOULD be set to one in the last packet of a video frame
   and otherwise set to zero. Thus, it is not necessary to wait for a
   following packet with a different timestamp to detect that a new
   frame should be displayed.

5.1 CelB

   The CELL-B encoding is a proprietary encoding proposed by Sun
   Microsystems. The byte stream format is described in RFC 2029 [18].

5.2 JPEG

   The encoding is specified in ISO Standards 10918-1 and 10918-2. The
   RTP payload format is as specified in RFC 2435 [19].

5.3 H261

   The encoding is specified in ITU-T Recommendation H.261, "Video codec
   for audiovisual services at p x 64 kbit/s". The packetization and
   RTP-specific properties are described in RFC 2032 [20].

Schulzrinne/Casner                                           [Page 26]

Internet Draft                  Profile                    March 2, 2001

5.4 H263-1998

   The encoding is specified in the 1998 version of ITU-T Recommendation
   H.263, "Video coding for low bit rate communication". The
   packetization and RTP-specific properties are described in RFC 2429
   [21]. Because the 1998 version of H.263 is a superset of the 1996
   syntax, this payload format can also be used with the 1996 version of
   H.263, and is RECOMMENDED for this use by new implementations. This
   payload format does not replace RFC 2190, which continues to be used
   by existing implementations, and may be required for backward
   compatibility in new implementations. Implementations using the new
   features of the 1998 version of H.263 MUST use the payload format
   described in RFC 2429.

5.5 MPV

   MPV designates the use of MPEG-1 and MPEG-2 video encoding elementary
   streams as specified in ISO Standards ISO/IEC 11172 and 13818-2,
   respectively. The RTP payload format is as specified in RFC 2250
   [14], Section 3.

5.8 nv

   The encoding is implemented in the program `nv', version 4, developed
   at Xerox PARC by Ron Frederick. Further information is available from
   the author:

   Ron Frederick
   Entera, Inc.
   40971 Encyclopedia Circle
   Fremont, CA 94538
   United States
   electronic mail:

Schulzrinne/Casner                                           [Page 27]

Internet Draft                  Profile                    March 2, 2001

6 Payload Type Definitions

   Tables 4 and 5 define this profile's static payload type values for
   the PT field of the RTP data header.  In addition, payload type
   values in the range 96-127 MAY be defined dynamically through a
   conference control protocol, which is beyond the scope of this
   document. For example, a session directory could specify that for a
   given session, payload type 96 indicates PCMU encoding, 8,000 Hz
   sampling rate, 2 channels.  Entries in Tables 4 and 5 with payload
   type "dyn" have no static payload type assigned and are only used
   with a dynamic payload type. Payload type 13 is reserved for a
   comfort noise payload format to be specified in a separate RFC.
   Payload type 19 is also marked "reserved" because some draft versions
   of this specification assigned that number to a comfort noise payload
   format.  The payload type range 72-76 is marked "reserved" so that
   RTCP and RTP packets can be reliably distinguished (see Section
   "Summary of Protocol Constants" of the RTP protocol specification).

   The payload types currently defined in this profile are assigned to
   exactly one of three categories or media types : audio only, video
   only and those combining audio and video. The media types are marked
   in Tables 4 and 5 as "A", "V" and "AV", respectively.  Payload types
   of different media types SHALL NOT be interleaved or multiplexed
   within a single RTP session, but multiple RTP sessions MAY be used in
   parallel to send multiple media types. An RTP source MAY change
   payload types within the same media type during a session.  See the
   section "Multiplexing RTP Sessions" of RFC XXXX for additional

   Session participants agree through mechanisms beyond the scope of
   this specification on the set of payload types allowed in a given
   session.  This set MAY, for example, be defined by the capabilities
   of the applications used, negotiated by a conference control protocol
   or established by agreement between the human participants.

   Audio applications operating under this profile SHOULD, at a minimum,
   be able to send and/or receive payload types 0 (PCMU) and 5 (DVI4).
   This allows interoperability without format negotiation and ensures
   successful negotation with a conference control protocol.

Schulzrinne/Casner                                           [Page 28]

Internet Draft                  Profile                    March 2, 2001

            PT   encoding    media type  clock rate  channels
                 name                    (Hz)
            0    PCMU        A           8000        1
            1    reserved    A
            2    G726-32     A           8000        1
            3    GSM         A           8000        1
            4    reserved    A
            5    DVI4        A           8000        1
            6    DVI4        A           16000       1
            7    LPC         A           8000        1
            8    PCMA        A           8000        1
            9    G722        A           8000        1
            10   L16         A           44100       2
            11   L16         A           44100       1
            12   QCELP       A           8000        1
            13   reserved    A
            14   MPA         A           90000       (see text)
            15   G728        A           8000        1
            16   DVI4        A           11025       1
            17   DVI4        A           22050       1
            18   G729        A           8000        1
            19   reserved    A
            20   unassigned  A
            21   unassigned  A
            22   unassigned  A
            23   unassigned  A
            dyn  G726-40     A           8000        1
            dyn  G726-24     A           8000        1
            dyn  G726-16     A           8000        1
            dyn  G729D       A           8000        1
            dyn  G729E       A           8000        1
            dyn  L8          A           var.        var.
            dyn  RED         A                       (see text)
            dyn  VDVI        A           var.        1

   Table 4: Payload types (PT) for audio encodings

Schulzrinne/Casner                                           [Page 29]

Internet Draft                  Profile                    March 2, 2001

               PT      encoding    media type  clock rate
                       name                    (Hz)
               24      unassigned  V
               25      CelB        V           90000
               26      JPEG        V           90000
               27      unassigned  V
               28      nv          V           90000
               29      unassigned  V
               30      unassigned  V
               31      H261        V           90000
               32      MPV         V           90000
               33      reserved    V
               34      reserved    V
               35-71   unassigned  ?
               72-76   reserved    N/A         N/A
               77-95   unassigned  ?
               96-127  dynamic     ?
               dyn     BT656       V           90000
               dyn     H263-1998   V           90000

   Table 5: Payload types (PT) for video and combined encodings

7 RTP over TCP and Similar Byte Stream Protocols

   Under special circumstances, it may be necessary to carry RTP in
   protocols offering a byte stream abstraction, such as TCP, possibly
   multiplexed with other data.  The application MUST define its own
   method of delineating RTP and RTCP packets (RTSP [22] provides an
   example of such an encapsulation specification.)

8 Port Assignment

   As specified in the RTP protocol definition, RTP data SHOULD be
   carried on an even UDP port number and the corresponding RTCP
   packets SHOULD be carried on the next higher (odd) port number.

   Applications operating under this profile MAY use any such UDP port
   pair. For example, the port pair MAY be allocated randomly by a
   session management program. A single fixed port number pair cannot be
   required because multiple applications using this profile are likely
   to run on the same host, and there are some operating systems that do
   not allow multiple processes to use the same UDP port with different
   multicast addresses.

Schulzrinne/Casner                                           [Page 30]

Internet Draft                  Profile                    March 2, 2001

   However, port numbers 5004 and 5005 have been registered for use with
   this profile for those applications that choose to use them as the
   default pair. Applications that operate under multiple profiles MAY
   use this port pair as an indication to select this profile if they
   are not subject to the constraint of the previous paragraph.
   Applications need not have a default and MAY require that the port
   pair be explicitly specified. The particular port numbers were chosen
   to lie in the range above 5000 to accommodate port number allocation
   practice within some versions of the Unix operating system, where
   port numbers below 1024 can only be used by privileged processes and
   port numbers between 1024 and 5000 are automatically assigned by the
   operating system.

9 Changes from RFC 1890

   This RFC revises RFC 1890. It is mostly backwards-compatible with RFC
   1890 and codifies existing practice. The changes are listed below.

        o The mapping of a user pass-phrase string into an encryption
          key was deleted from Section 2 because two interoperable
          implementations were not found.

        o The payload formats for 1016 audio and MP2T video were removed
          and their static payload type assignments 1 and 33 were marked
          "reserved" because two interoperable implementations were not

        o Additional payload formats and/or expanded descriptions were
          included for G722, G726, G728, G729, GSM, QCELP, RED, VDVI,
          and H263-1998.

        o Static payload types 12, 16, 17 and 18 were added, and 13 and
          19 were reserved.

        o Requirements for congestion control were added in Section 2.

        o A new Section "IANA Considerations" was added to specify the
          regstration of the name for this profile and to establish a
          new policy that no additional registration of static payload
          types for this profile will be made beyond those included in
          Tables 4 and 5, but that additional encoding names may be
          registered as MIME subtypes for binding to dynamic payload

        o In Section 4.1, the requirement level for setting of the
          marker bit on the first packet after silence for audio was
          changed from "is" to "SHOULD be".

Schulzrinne/Casner                                           [Page 31]

Internet Draft                  Profile                    March 2, 2001

        o Similarly, text was added to specify that the marker bit
          SHOULD be set to one on the last packet of a video frame, and
          that video frames are distinguished by their timestamps.

        o This profile follows the suggestion in the RTP spec that RTCP
          bandwidth may be specified separately from the session
          bandwidth and separately for active senders and passive

        o RFC references are added for payload formats published after
          RFC 1890.

        o The security considerations and full copyright sections were

        o According to Peter Hoddie of Apple, only pre-1994 Macintosh
          used the 22254.54 rate and none the 11127.27 rate, so the
          latter was dropped from the discussion of suggested sampling

        o Table 1 was corrected to move some values from the "ms/packet"
          column to the "default ms/packet" column where they belonged.

        o A note has been added for G722 to clarify a discrepancy
          between the actual sampling rate and the RTP timestamp clock

        o Small clarifications of the text have been made in several
          places, some in response to questions from readers. In

          - A definition for "media type" is given in Section 1.1 to
            allow the explanation of multiplexing RTP sessions in
            Section 6 to be more clear regarding the multiplexing of
            multiple media.

          - The explanation of how to determine the number of audio
            frames in a packet from the length was expanded.

          - More description of the allocation of bandwidth to SDES
            items is given.

          - A note was added that the convention for the order of
            channels specified in Section 4.1 may be overridden by a
            particular encoding or payload format specification.

          - The terms MUST, SHOULD, MAY, etc. are used as defined in RFC

Schulzrinne/Casner                                           [Page 32]

Internet Draft                  Profile                    March 2, 2001

        o A second author for this document was added.

10 Security Considerations

   Implementations using the profile defined in this specification are
   subject to the security considerations discussed in the RTP
   specification [1]. This profile does not specify any different
   security services other than giving rules for mapping characters in a
   user-provided pass phrase to canonical form.  The primary function of
   this profile is to list a set of data compression encodings for audio
   and video media.

   Confidentiality of the media streams is achieved by encryption.
   Because the data compression used with the payload formats described
   in this profile is applied end-to-end, encryption may be performed
   after compression so there is no conflict between the two operations.

   A potential denial-of-service threat exists for data encodings using
   compression techniques that have non-uniform receiver-end
   computational load. The attacker can inject pathological datagrams
   into the stream which are complex to decode and cause the receiver to
   be overloaded. However, the encodings described in this profile do
   not exhibit any significant non-uniformity.

   As with any IP-based protocol, in some circumstances a receiver may
   be overloaded simply by the receipt of too many packets, either
   desired or undesired. Network-layer authentication MAY be used to
   discard packets from undesired sources, but the processing cost of
   the authentication itself may be too high. In a multicast
   environment, pruning of specific sources may be implemented in future
   versions of IGMP [23] and in multicast routing protocols to allow a
   receiver to select which sources are allowed to reach it.

11 Full Copyright Statement

   Copyright (C) The Internet Society (2000). All Rights Reserved.

   This document and translations of it may be copied and furnished to
   others, and derivative works that comment on or otherwise explain it
   or assist in its implmentation may be prepared, copied, published and
   distributed, in whole or in part, without restriction of any kind,
   provided that the above copyright notice and this paragraph are
   included on all such copies and derivative works. However, this
   document itself may not be modified in any way, such as by removing
   the copyright notice or references to the Internet Society or other
   Internet organizations, except as needed for the purpose of
   developing Internet standards in which case the procedures for
   copyrights defined in the Internet Standards process must be

Schulzrinne/Casner                                           [Page 33]

Internet Draft                  Profile                    March 2, 2001

   followed, or as required to translate it into languages other than

   The limited permissions granted above are perpetual and will not be
   revoked by the Internet Society or its successors or assigns.

   This document and the information contained herein is provided on an

12 Acknowledgements

   The comments and careful review of Simao Campos, Richard Cox and AVT
   Working Group participants are gratefully acknowledged. The GSM
   description was adopted from the IMTC Voice over IP Forum Service
   Interoperability Implementation Agreement (January 1997). Fred Burg
   and Terry Lyons helped with the G.729 description.

13 Addresses of Authors

   Henning Schulzrinne
   Dept. of Computer Science
   Columbia University
   1214 Amsterdam Avenue
   New York, NY 10027
   electronic mail:

   Stephen L. Casner
   Packet Design
   2465 Latham Street
   Mountain View, CA 94040
   United States
   electronic mail:

A Bibliography

   [1] H. Schulzrinne, S. Casner, R. Frederick, and V. Jacobson, "RTP: A
   transport protocol for real-time applications," Internet Draft,
   Internet Engineering Task Force, Feb. 1999 Work in progress, revision
   to RFC 1889.

   [2] S. Bradner, "Key words for use in RFCs to Indicate Requirement
   Levels,"  RFC 2119, Internet Engineering Task Force, Mar. 1997.

Schulzrinne/Casner                                           [Page 34]

Internet Draft                  Profile                    March 2, 2001

   [3] R. Braden, D. Clark, S. Shenker, "Integrated Services in the
   Internet Architecture: an Overview," Request for Comments
   (Informational) RFC 1633, Internet Engineering Task Force, June 1994.

   [4] S. Blake, D. Black, M. Carlson, E. Davies, Z. Wang, W. Weiss, "An
   Architecture for Differentiated Service," Request for Comments
   (Proposed Standard) RFC 2475, Internet Engineering Task Force, Dec.

   [5] M. Handley and V. Jacobson, "SDP: Session Description Protocol,"
   Request for Comments (Proposed Standard) RFC 2327, Internet
   Engineering Task Force, Apr. 1998.

   [6] P. Hoschka, "MIME Type Registration of RTP Payload Types,"
   Internet Draft, Internet Engineering Task Force, Feb. 1999 Work in

   [7] N. Freed, J. Klensin, and J. Postel, "Multipurpose Internet Mail
   Extensions (MIME) Part Four: Registration Procedures,"  RFC 2048,
   Internet Engineering Task Force, Nov. 1996.

   [8] Apple Computer, "Audio interchange file format AIFF-C," Aug.
   1991.  (also

   [9] IMA Digital Audio Focus and Technical Working Groups,
   "Recommended practices for enhancing digital audio compatibility in
   multimedia systems (version 3.00)," tech. rep., Interactive
   Multimedia Association, Annapolis, Maryland, Oct. 1992.

   [10] D. Deleam and J.-P. Petit, "Real-time implementations of the
   recent ITU-T low bit rate speech coders on the TI TMS320C54X DSP:
   results, methodology, and applications," in Proc. of International
   Conference on Signal Processing, Technology, and Applications
   (ICSPAT) , (Boston, Massachusetts), pp. 1656--1660, Oct. 1996.

   [11] M. Mouly and M.-B. Pautet, The GSM system for mobile
   communications Lassay-les-Chateaux, France: Europe Media Duplication,

   [12] J. Degener, "Digital speech compression," Dr. Dobb's Journal ,
   Dec.  1994.

   [13] S. M. Redl, M. K. Weber, and M. W. Oliphant, An Introduction to
   GSM Boston: Artech House, 1995.

   [14] D. Hoffman, G. Fernando, V. Goyal, and M. Civanlar, "RTP payload
   format for MPEG1/MPEG2 video," Request for Comments (Proposed
   Standard) RFC 2250, Internet Engineering Task Force, Jan. 1998.

Schulzrinne/Casner                                           [Page 35]

Internet Draft                  Profile                    March 2, 2001

   [15] N. S. Jayant and P. Noll, Digital Coding of Waveforms--
   Principles and Applications to Speech and Video Englewood Cliffs, New
   Jersey: Prentice-Hall, 1984.

   [16] K. McKay, "RTP Payload Format for PureVoice(tm) Audio", Request
   for Comments (Proposed Standard) RFC 2658, Internet Engineering Task
   Force, Aug. 1999.

   [17] C. Perkins, I. Kouvelas, O. Hodson, V. Hardman, M. Handley, J.C.
   Bolot, A. Vega-Garcia, and S. Fosse-Parisis, "RTP Payload for
   Redundant Audio Data," Request for Comments (Proposed Standard) RFC
   2198, Internet Engineering Task Force, Sep. 1997.

   [18] M. Speer and D. Hoffman, "RTP payload format of sun's CellB
   video encoding," Request for Comments (Proposed Standard) RFC 2029,
   Internet Engineering Task Force, Oct. 1996.

   [19] L. Berc, W. Fenner, R. Frederick, and S. McCanne, "RTP payload
   format for JPEG-compressed video," Request for Comments (Proposed
   Standard) RFC 2435, Internet Engineering Task Force, Oct. 1996.

   [20] T. Turletti and C. Huitema, "RTP payload format for H.261 video
   streams," Request for Comments (Proposed Standard) RFC 2032, Internet
   Engineering Task Force, Oct. 1996.

   [21] C. Bormann, L. Cline, G. Deisher, T. Gardos, C. Maciocco, D.
   Newell, J. Ott, G. Sullivan, S. Wenger, C. Zhu, "RTP Payload Format
   for the 1998 Version of ITU-T Rec. H.263 Video (H.263+)," Request for
   Comments (Proposed Standard) RFC 2429, Internet Engineering Task
   Force, Oct. 1998.

   [22] H. Schulzrinne, A. Rao, and R. Lanphier, "Real time streaming
   protocol (RTSP)," Request for Comments (Proposed Standard) RFC 2326,
   Internet Engineering Task Force, Apr. 1998.

   [23] S. Deering, "Host Extensions for IP Multicasting," Request for
   Comments RFC 1112, STD 5, Internet Engineering Task Force, Aug. 1989.

Schulzrinne/Casner                                           [Page 36]

Internet Draft                  Profile                    March 2, 2001

   Current Locations of Related Resources

   Note: Several sections below refer to the ITU-T Software Tool Library
   (STL). It is available from the ITU Sales Service, Place des Nations,
   CH-1211 Geneve 20, Switzerland (also check The
   ITU-T STL is covered by a license defined in ITU-T Recommendation
   G.191, "Software tools for speech and audio coding standardization".


   Information on the UCS Transformation Format 8 (UTF-8) is available



   An implementation is available from Jack Jansen at



   An implementation of the G.722 algorithm is available as part of the
   ITU-T STL, described above.


   G726 is specified in the ITU-T Recommendation G.726, "40, 32, 24,
   and 16 kb/s Adaptive Differential Pulse Code Modulation (ADPCM)". An
   implementation of the G.726 algorithm is available as part of the
   ITU-T STL, described above.


   The reference C code implementation defining the G.729 algorithm and
   its Annexes A through I are available as an integral part of
   Recommendation G.729 from the ITU Sales Service, listed above. Annex
   I contains the integrated C source code for all G.729 operating
   modes.  The G.729 algorithm and associated C code are covered by a
   specific license. The contact information for obtaining the license
   is available from the ITU-T Secretariat.

Schulzrinne/Casner                                           [Page 37]

Internet Draft                  Profile                    March 2, 2001


   A reference implementation was written by Carsten Borman and Jutta
   Degener (TU Berlin, Germany). It is available at


   Although the RPE-LTP algorithm is not an ITU-T standard, there is a C
   code implementation of the RPE-LTP algorithm available as part of the
   ITU-T STL. The STL implementation is an adaptation of the TU Berlin


   An implementation is available at



   An implementation of these algorithm is available as part of the
   ITU-T STL, described above. Code to convert between linear and mu-law
   companded data is also available in [9].

Schulzrinne/Casner                                           [Page 38]

Internet Draft                  Profile                    March 2, 2001

                           Table of Contents

   1          Introduction ........................................    3
   1.1        Terminology .........................................    4
   2          RTP and RTCP Packet Forms and Protocol Behavior .....    4
   3          IANA Considerations .................................    7
   3.1        Registering Additional Encodings ....................    7
   4          Audio ...............................................    8
   4.1        Encoding-Independent Rules ..........................    8
   4.2        Operating Recommendations ...........................   10
   4.3        Guidelines for Sample-Based Audio Encodings .........   10
   4.4        Guidelines for Frame-Based Audio Encodings ..........   11
   4.5        Audio Encodings .....................................   11
   4.5.1      DVI4 ................................................   12
   4.5.2      G722 ................................................   14
   4.5.3      G726-40, G726-32, G726-24, and G726-16...............   14
   4.5.4      G728 ................................................   16
   4.5.5      G729 ................................................   16
   4.5.6      G729D and G729E .....................................   18
   4.5.7      GSM .................................................   21    General Packaging Issues ............................   21    GSM variable names and numbers ......................   21
   4.5.8      L8 ..................................................   21
   4.5.9      L16 .................................................   22
   4.5.10     LPC .................................................   23
   4.5.11     MPA .................................................   24
   4.5.12     PCMA and PCMU .......................................   24
   4.5.13     QCELP ...............................................   24
   4.5.14     RED .................................................   24
   4.5.15     VDVI ................................................   25
   5          Video ...............................................   25
   5.1        CelB ................................................   26
   5.2        JPEG ................................................   26
   5.3        H261 ................................................   26
   5.4        H263-1998 ...........................................   27
   5.5        MPV .................................................   27
   5.8        nv ..................................................   27
   6          Payload Type Definitions ............................   28
   7          RTP over TCP and Similar Byte Stream Protocols ......   30
   8          Port Assignment .....................................   30
   9          Changes from RFC 1890 ...............................   31
   10         Security Considerations .............................   33
   11         Full Copyright Statement ............................   33
   12         Acknowledgements ....................................   34
   13         Addresses of Authors ................................   34
   A          Bibliography ........................................   34

Schulzrinne/Casner                                           [Page 39]