draft-ietf-avt-evrc-smv-01.txt   draft-ietf-avt-evrc-smv-02.txt 
Internet Draft Adam H. Li Internet Draft Adam H. Li
draft-ietf-avt-evrc-smv-01.txt UCLA draft-ietf-avt-evrc-smv-02.txt UCLA
May 16, 2002 Editor June 7, 2002 Editor
Expires: November 16, 2002 Expires: December 7, 2002
RTP Payload Format for EVRC and SMV Vocoders RTP Payload Format for Enhanced Variable Rate Codecs (EVRC) and
Selectable Mode Vocoders (SMV)
STATUS OF THIS MEMO STATUS OF THIS MEMO
This document is an Internet-Draft and is in full conformance with This document is an Internet-Draft and is in full conformance with
all provisions of Section 10 of RFC 2026. all provisions of Section 10 of RFC 2026.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that other Task Force (IETF), its areas, and its working groups. Note that other
groups may also distribute working documents as Internet-Drafts. groups may also distribute working documents as Internet-Drafts.
skipping to change at page 2, line 5 skipping to change at page 2, line 6
Table of Contents Table of Contents
1. Introduction ................................................... 2 1. Introduction ................................................... 2
2. Background ..................................................... 2 2. Background ..................................................... 2
3. The Codecs Supported ........................................... 3 3. The Codecs Supported ........................................... 3
3.1. EVRC ......................................................... 3 3.1. EVRC ......................................................... 3
3.2. SMV .......................................................... 3 3.2. SMV .......................................................... 3
3.3. Other Frame-Based Vocoders ................................... 4 3.3. Other Frame-Based Vocoders ................................... 4
4. RTP/Vocoder Packet Format ...................................... 4 4. RTP/Vocoder Packet Format ...................................... 4
4.1. Type 1 Interleaved/Bundled Packet Format ..................... 4 4.1. Interleaved/Bundled Packet Format ............................ 4
4.2. Type 2 Header-Free Packet Format ............................. 6 4.2. Header-Free Packet Format .................................... 6
4.3. Determining the Format of Packets ............................ 6 4.3. Determining the Format of Packets ............................ 6
5. Packet Table of Contents Entries and Codec Data Frame Format ... 7 5. Packet Table of Contents Entries and Codec Data Frame Format ... 7
5.1. Packet Table of Contents entries ............................. 7 5.1. Packet Table of Contents entries ............................. 7
5.2. Codec Data Frames ............................................ 8 5.2. Codec Data Frames ............................................ 7
6. Interleaving Codec Data Frames in Type 1 Packets ............... 9 6. Interleaving Codec Data Frames ................................. 8
6.1. Finding Interleave Group Boundaries ......................... 11 7. Bundling Codec Data Frames .................................... 11
6.2. Additional Receiver Responsibilities ........................ 11 8. Handling Missing Codec Data Frames ............................ 11
7. Bundling Codec Data Frames in Type 1 Packets .................. 11 9. Implementation Issues ......................................... 11
8. Handling Missing Codec Data Frames ............................ 12 9.1. Interleaving Length ......................................... 11
9. Implementation Issues ......................................... 12
9.1. Interleaving Length ......................................... 12
9.2. Validation of Received Packets .............................. 12 9.2. Validation of Received Packets .............................. 12
10. Mode Request ................................................. 13 9.3. Processing the Late Packets ................................. 12
11. Storage Mode ................................................. 13 10. Mode Request ................................................. 12
11. Storage Format ............................................... 13
12. IANA Considerations .......................................... 14 12. IANA Considerations .......................................... 14
12.1. Registration of Media Type EVRC ............................ 14 12.1. Registration of Media Type EVRC ............................ 14
12.2. Registration of Media Type EVRC0 ........................... 15 12.2. Registration of Media Type EVRC0 ........................... 15
12.3. Registration of Media Type SMV ............................. 16 12.3. Registration of Media Type SMV ............................. 16
12.4. Registration of Media Type SMV0 ............................ 17 12.4. Registration of Media Type SMV0 ............................ 17
13. Mapping to SDP Parameters .................................... 17 13. Mapping to SDP Parameters .................................... 18
14. Security Considerations ...................................... 18 14. Security Considerations ...................................... 18
15. Adding Support of Other Frame-Based Vocoders ................. 19 15. Adding Support of Other Frame-Based Vocoders ................. 19
16. Acknowledgements ............................................. 19 16. Acknowledgements ............................................. 19
17. References ................................................... 20 17. References ................................................... 20
18. Authors' Address ............................................. 20 18. Authors' Address ............................................. 20
1. Introduction 1. Introduction
This document describes how speech compressed with EVRC [1] or SMV This document describes how speech compressed with EVRC [1] or SMV
[2] may be formatted for use as an RTP payload type. The format is [2] may be formatted for use as an RTP payload type. The format is
skipping to change at page 4, line 7 skipping to change at page 4, line 7
The SMV codec can operate in four modes. Each mode may produce frames The SMV codec can operate in four modes. Each mode may produce frames
of any of the rates (full rate to 1/8 rate) for varying percentages of any of the rates (full rate to 1/8 rate) for varying percentages
of time, based on the characteristics of the speech samples and the of time, based on the characteristics of the speech samples and the
selected mode. The SMV mode can change on a frame-by-frame basis. The selected mode. The SMV mode can change on a frame-by-frame basis. The
SMV codec does not need additional information other than the codec SMV codec does not need additional information other than the codec
data frames to correctly decode the data of various modes; therefore, data frames to correctly decode the data of various modes; therefore,
the mode of the encoder does not need to be transmitted with the the mode of the encoder does not need to be transmitted with the
encoded frames. encoded frames.
The percentage of different frame rates for the four SMV modes are The SMV codec chooses the output frame rate based on analysis of the
shown in the table below. input speech and the current operating mode. For typical speech
patterns, this results in an average output of 4.2 kilobits/second
Mode 0 Mode 1 Mode 2 Mode 3 for Mode 0 in two way conversation (approximately 50% active speech
------------------------------------------------------------- time and 50% in eighth rate while listening) and lower for other
Rate 1 68.90% 38.14% 15.43% 07.49% reduced rate modes. SMV is more bandwidth efficient than EVRC. EVRC
Rate 1/2 06.03% 15.82% 38.34% 46.28% is equivalent in performance to SMV mode 1.
Rate 1/4 00.00% 17.37% 16.38% 16.38%
Rate 1/8 25.07% 28.67% 29.85% 29.85%
The SMV codec chooses the output frame rate based on an analysis of
the input speech and the current operating mode. For typical speech
patterns, this results in an average output of 4.2kilobits/second for
Mode 0 in two way conversation (assuming 50% active speech time and
50% in eighth rate while listening) and lower for other reduced rate
modes.
SMV is more bandwidth efficient than EVRC. EVRC is equivalent in
performance to SMV mode 1.
3.3. Other Frame-Based Vocoders 3.3. Other Frame-Based Vocoders
Other frame-based vocoders can be carried in the packet format Other frame-based vocoders can be carried in the packet format
defined in this document, as long as they possess the following defined in this document, as long as they possess the following
properties: properties:
o The codec is frame-based; o The codec is frame-based;
o blank and erasure frames are supported; o blank and erasure frames are supported;
o the total number of rates is less than 17; o the total number of rates is less than 17;
o the maximum full rate frame can be transported in a single RTP o the maximum full rate frame can be transported in a single RTP
packet using this specific format. packet using this specific format.
Vocoders with the characteristics listed above can be transported Vocoders with the characteristics listed above can be transported
using the packet format specified in this document with some using the packet format specified in this document with some
additional specification work; the pieces that must be defined are additional specification work; the pieces that must be defined are
listed in Section 15. listed in Section 15.
4. RTP/Vocoder Packet Format 4. RTP/Vocoder Packet Format
In the packet format diagrams shown in this document, bit 0 is the The vocoder speech data may be transmitted in either of the two RTP
most significant bit. The vocoder speech data MUST be transmitted in packet formats specified in the following two subsections, as
RTP packets of one of the following two types. appropriate for the application scenario. In the packet format
diagrams shown in this document, bit 0 is the most significant bit.
4.1. Type 1 Interleaved/Bundled Packet Format 4.1. Interleaved/Bundled Packet Format
This format is used to send one or more vocoder frames per packet. This format is used to send one or more vocoder frames per packet.
Interleaving or bundling MAY be used. The RTP packet for this format Interleaving or bundling MAY be used. The RTP packet for this format
is as follows: is as follows:
0 1 2 3 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 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 [4] | | RTP Header [4] |
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
|R|R| LLL | NNN | FFF | Count | TOC | ... | TOC |padding| |R|R| LLL | NNN | MMM | Count | TOC | ... | TOC |padding|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| one or more codec data frames, one per TOC entry | | one or more codec data frames, one per TOC entry |
| .... | | .... |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
The RTP header has the expected values as described in the RTP The RTP header has the expected values as described in the RTP
specification [4]. The RTP timestamp is in 1/8000 of a second units specification [4]. The RTP timestamp is in 1/8000 of a second units
for EVRC and SMV. For any other vocoders that use this packet format, for EVRC and SMV. For any other vocoders that use this packet format,
the timestamp unit needs to be defined explicitly. The M bit should the timestamp unit needs to be defined explicitly. The M bit should
be set as specified in the applicable RTP profile, for example, RFC be set as specified in the applicable RTP profile, for example, RFC
1890 [5]. Note that RFC 1890 [5] specifies that if the sender does 1890 [5]. Note that RFC 1890 [5] specifies that if the sender does
not suppress silence, the M bit will always be zero. When multiple not suppress silence, the M bit will always be zero. When multiple
codec data frames are present in a single RTP packet, the timestamp codec data frames are present in a single RTP packet, the timestamp
is that of the oldest data represented in the RTP packet. The is that of the oldest data represented in the RTP packet. The
assignment of an RTP payload type for this new packet format is assignment of an RTP payload type for this packet format is outside
outside the scope of this document; it is specified by the RTP the scope of this document; it is specified by the RTP profile under
profile under which this payload format is used. which this payload format is used.
The first octet of a Type 1 Interleaved/Bundled format packet is the
The first octet of a Interleaved/Bundled format packet is the
Interleave Octet. The second octet contains the Mode Request and Interleave Octet. The second octet contains the Mode Request and
Frame Count fields. The Table of Contents (ToC) field then follows. Frame Count fields. The Table of Contents (ToC) field then follows.
The fields are specified as follows: The fields are specified as follows:
Reserved (RR): 2 bits Reserved (RR): 2 bits
Reserved bits. MUST be set to zero by sender, SHOULD be ignored Reserved bits. MUST be set to zero by sender, SHOULD be ignored
by receiver. by receiver.
Interleave Length (LLL): 3 bits Interleave Length (LLL): 3 bits
Indicates the length of interleave; a value of 0 indicates Indicates the length of interleave; a value of 0 indicates
bundling, a special case of interleaving. See Section 6 and bundling, a special case of interleaving. See Section 6 and
Section 7 for more detailed discussion. Section 7 for more detailed discussion.
Interleave Index (NNN): 3 bits Interleave Index (NNN): 3 bits
Indicates the index within an interleave group. MUST have a value Indicates the index within an interleave group. MUST have a value
less than or equal to the value of LLL. Values of NNN greater less than or equal to the value of LLL. Values of NNN greater
than the value of LLL are invalid. Packet with invalid NNN values than the value of LLL are invalid. Packet with invalid NNN values
SHOULD be ignored by the receiver. SHOULD be ignored by the receiver.
Mode Request (FFF): 3 bits Mode Request (MMM): 3 bits
The Mode Request field is used to signal Mode Request The Mode Request field is used to signal Mode Request
information. See Section 10 for details. information. See Section 10 for details.
Frame Count (Count): 5 bits Frame Count (Count): 5 bits
The number of ToC fields (and vocoder frames) present in the The number of ToC fields (and vocoder frames) present in the
packet is the value of the frame count field plus one. A value of packet is the value of the frame count field plus one. A value of
zero indicates that the packet contains one ToC field, while a zero indicates that the packet contains one ToC field, while a
value of 31 indicates that the packet contains 32 ToC fields. value of 31 indicates that the packet contains 32 ToC fields.
Padding (padding): 0 or 4 bits Padding (padding): 0 or 4 bits
skipping to change at page 6, line 18 skipping to change at page 6, line 10
of padding following the last TOC. When the frame count is even, of padding following the last TOC. When the frame count is even,
the sender MUST NOT add padding bits. If padding is present, the the sender MUST NOT add padding bits. If padding is present, the
padding bits MUST be set to zero by sender, and SHOULD be ignored padding bits MUST be set to zero by sender, and SHOULD be ignored
by receiver. by receiver.
The Table of Contents field (ToC) provides information on the codec The Table of Contents field (ToC) provides information on the codec
data frame(s) in the packet. There is one ToC entry for each codec data frame(s) in the packet. There is one ToC entry for each codec
data frame. The detailed formats of the ToC field and codec data data frame. The detailed formats of the ToC field and codec data
frames are specified in Section 5. frames are specified in Section 5.
Multiple data frames may be included within a Type 1 Multiple data frames may be included within a Interleaved/Bundled
Interleaved/Bundled packet using interleaving or bundling as packet using interleaving or bundling as described in Section 6 and
described in Section 6 and Section 7. Section 7.
4.2. Type 2 Header-Free Packet Format 4.2. Header-Free Packet Format
The Type 2 Header-Free Packet Format is designed for maximum The Header-Free Packet Format is designed for maximum bandwidth
bandwidth efficiency and low latency. Only one codec data frame can efficiency and low latency. Only one codec data frame can be sent in
be sent in each Type 2 Header-Free format packet. None of the payload each Header-Free format packet. None of the payload header fields
header fields (LLL, NNN, FFF, Count) nor ToC entries are present. The (LLL, NNN, MMM, Count) nor ToC entries are present. The codec rate
codec rate for the data frame can be determined from the length of for the data frame can be determined from the length of the codec
the codec data frame, since there is only one codec data frame in data frame, since there is only one codec data frame in each Header-
each Type 2 Header-Free packet. Free packet.
Use of the RTP header fields for Type 2 Header-Free RTP/Vocoder Use of the RTP header fields for Header-Free RTP/Vocoder Packet
Packet Format is the same as described in Section 4.1 for Type 1 Format is the same as described in Section 4.1 for
Interleaved/Bundled RTP/Vocoder Packet Format. The detailed format of Interleaved/Bundled RTP/Vocoder Packet Format. The detailed format of
the codec data frame is specified in Section 5. the codec data frame is specified in Section 5.
0 1 2 3 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 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 [4] | | RTP Header [4] |
+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+ +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
| | | |
+ ONLY one codec data frame +-+-+-+-+-+-+-+-+ + ONLY one codec data frame +-+-+-+-+-+-+-+-+
| | | |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
4.3. Determining the Format of Packets 4.3. Determining the Format of Packets
All receivers SHOULD be able to process both types of packets. The All receivers SHOULD be able to process both packet formats. The
sender MAY choose to use one or both types of packets. sender MAY choose to use one or both packet formats.
A receiver MUST have prior knowledge of the packet type to correctly
decode the RTP packets. The packet types used in an RTP session MUST
be specified by the sender, and signaled through out-of-band means,
for example by SDP during the setup of a session.
A receiver MUST have prior knowledge of the packet format to
correctly decode the RTP packets.
When packets of both formats are used within the same session, When packets of both formats are used within the same session,
different RTP payload type values MUST be used for each format to different RTP payload type values MUST be used for each format to
distinguish the packet formats. The association of payload type distinguish the packet formats. The association of payload type
number with the packet format is done out-of-band, for example by SDP number with the packet format is done out-of-band, for example by SDP
during the setup of a session. during the setup of a session.
5. Packet Table of Contents Entries and Codec Data Frame Format 5. Packet Table of Contents Entries and Codec Data Frame Format
5.1. Packet Table of Contents entries 5.1. Packet Table of Contents entries
Each codec data frame in a Type 1 Interleaved/Bundled packet has a Each codec data frame in a Interleaved/Bundled packet has a
corresponding Table of Contents (ToC) entry. The ToC entry indicates corresponding Table of Contents (ToC) entry. The ToC entry indicates
the rate of the codec frame. (Type 2 Header-Free packets MUST NOT the rate of the codec frame. (Header-Free packets MUST NOT have a ToC
have a ToC field, and there is always only one codec data frame in field.)
each Type 2 Header-Free packet.)
Each ToC entry is occupies four bits. The format of the bits is Each ToC entry is occupies four bits. The format of the bits is
indicated below: indicated below:
0 1 2 3 0 1 2 3
+-+-+-+-+ +-+-+-+-+
|fr type| |fr type|
+-+-+-+-+ +-+-+-+-+
Frame Type: 4 bits Frame Type: 4 bits
skipping to change at page 9, line 4 skipping to change at page 8, line 45
|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0| |0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|0|
|0|0|0|0|0|0|0|0|0|1|1|1|1|1|1|1|1|1|1|2|2|2|2|2|2|2|2|2|2|3|3|3| |0|0|0|0|0|0|0|0|0|1|1|1|1|1|1|1|1|1|1|2|2|2|2|2|2|2|2|2|2|3|3|3|
|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|2| |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|2|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
: : : :
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1| | | | | | |1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1|1| | | | | |
|4|4|4|4|4|5|5|5|5|5|5|5|5|5|5|6|6|6|6|6|6|6|6|6|6|7|7|Z|Z|Z|Z|Z| |4|4|4|4|4|5|5|5|5|5|5|5|5|5|5|6|6|6|6|6|6|6|6|6|6|7|7|Z|Z|Z|Z|Z|
|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| | | | | | |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| | | | | |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
6. Interleaving Codec Data Frames in Type 1 Packets
6. Interleaving Codec Data Frames
As indicated in Section 4.1, more than one codec data frame MAY be As indicated in Section 4.1, more than one codec data frame MAY be
included in a single Type 1 Interleaved/Bundled packet by a sender. included in a single Interleaved/Bundled packet by a sender. This is
This is accomplished by interleaving or bundling. accomplished by interleaving or bundling.
Bundling is used to spread the transmission overhead of the RTP and Bundling is used to spread the transmission overhead of the RTP and
payload header over multiple vocoder frames. Interleaving payload header over multiple vocoder frames. Interleaving
additionally reduces the listener's perception of data loss by additionally reduces the listener's perception of data loss by
spreading such loss over non-consecutive vocoder frames. EVRC, SMV, spreading such loss over non-consecutive vocoder frames. EVRC, SMV,
and similar vocoders are able to compensate for an occasional lost and similar vocoders are able to compensate for an occasional lost
frame, but speech quality degrades exponentially with consecutive frame, but speech quality degrades exponentially with consecutive
frame loss. frame loss.
Bundling is signaled by setting the LLL field to zero and the Count Bundling is signaled by setting the LLL field to zero and the Count
field to greater than zero. Interleaving is indicated by setting the field to greater than zero. Interleaving is indicated by setting the
LLL field to a value greater than zero. LLL field to a value greater than zero.
The discussions on general interleaving apply to the bundling (which The discussions on general interleaving apply to the bundling (which
can be viewed as a reduced case of interleaving) with reduced can be viewed as a reduced case of interleaving) with reduced
complexity. The bundling case is discussed in detail in Section 7. complexity. The bundling case is discussed in detail in Section 7.
Senders MAY support interleaving and/or bundling. All receivers MUST Senders MAY support interleaving and/or bundling. All receivers that
support interleaving and bundling. support Interleave/Bundling packet format MUST support both
interleaving and bundling.
Given a time-ordered sequence of output frames from the codec Given a time-ordered sequence of output frames from the codec
numbered 0..n, a bundling value B (the value in the Count field plus numbered 0..n, a bundling value B (the value in the Count field plus
one), and an interleave length L where n = B * (L+1) - 1, the output one), and an interleave length L where n = B * (L+1) - 1, the output
frames are placed into RTP packets as follows (the values of the frames are placed into RTP packets as follows (the values of the
fields LLL and NNN are indicated for each RTP packet): fields LLL and NNN are indicated for each RTP packet):
First RTP Packet in Interleave group: First RTP Packet in Interleave group:
LLL=L, NNN=0 LLL=L, NNN=0
Frame 0, Frame L+1, Frame 2(L+1), Frame 3(L+1), ... for a total of Frame 0, Frame L+1, Frame 2(L+1), Frame 3(L+1), ... for a total of
skipping to change at page 10, line 44 skipping to change at page 10, line 34
than will fit in the MTU of the underlying network. than will fit in the MTU of the underlying network.
o Once beginning a session with a given maximum interleaving value o Once beginning a session with a given maximum interleaving value
set by maxinterleave in Section 12, MUST NOT increase the set by maxinterleave in Section 12, MUST NOT increase the
interleaving value (LLL) to exceed the maximum interleaving value interleaving value (LLL) to exceed the maximum interleaving value
that is signaled. that is signaled.
o MAY change the interleaving value, but MUST do so only between o MAY change the interleaving value, but MUST do so only between
interleave groups. interleave groups.
o Silence suppression MAY only be used between interleave groups. A o Silence suppression MUST only be used between interleave groups. A
ToC with Frame Type 0 (Blank Frame, Section 5.1) MUST be used ToC with Frame Type 0 (Blank Frame, Section 5.1) MUST be used
within interleaving groups if the codec outputs a blank frame. within interleaving groups if the codec outputs a blank frame.
The M bits in the RTP header is not set for these blank frames, The M bit in the RTP header is not set for these blank frames, as
as the stream is continuous in time. Because there is only one the stream is continuous in time. Because there is only one time
time stamp for each RTP packet, silence suppression used within stamp for each RTP packet, silence suppression used within an
an interleave group would cause ambiguities when reconstructing interleave group would cause ambiguities when reconstructing the
the speech at the receiver side, and thus is prohibited. speech at the receiver side, and thus is prohibited.
6.1. Finding Interleave Group Boundaries
Given an RTP packet with sequence number S, interleave length (field Given an RTP packet with sequence number S, interleave length (field
LLL) L, interleave index value (field NNN) N, and bundling value B, LLL) L, interleave index value (field NNN) N, and bundling value B,
the interleave group consists of this RTP packet and other RTP the interleave group consists of this RTP packet and other RTP
packets with sequence numbers from S-N mod 65536 to S-N+L mod 65536 packets with sequence numbers from S-N mod 65536 to S-N+L mod 65536
inclusive. In other words, the interleave group always consists of inclusive. In other words, the interleave group always consists of
L+1 RTP packets with sequential sequence numbers. The bundling value L+1 RTP packets with sequential sequence numbers. The bundling value
for all RTP packets in an interleave group MUST be the same. for all RTP packets in an interleave group MUST be the same.
The receiver determines the expected bundling value for all RTP The receiver determines the expected bundling value for all RTP
packets in an interleave group by the number of codec data frames packets in an interleave group by the number of codec data frames
bundled in the first RTP packet of the interleave group received. bundled in the first RTP packet of the interleave group received.
Note that this may not be the first RTP packet of the interleave Note that this may not be the first RTP packet of the interleave
group if packets are delivered out of order by the underlying group if packets are delivered out of order by the underlying
transport. transport.
6.2. Additional Receiver Responsibilities 7. Bundling Codec Data Frames
Assume that the receiver has begun playing frames from an interleave
group. The time has come to play frame x from packet n of the
interleave group. Further assume that packet n of the interleave
group has not been received. As described in Section 8, an erasure
frame will be sent to the receiving vocoder.
Now, assume that packet n of the interleave group arrives before
frame x+1 of that packet is needed. Receivers SHOULD use frame x+1 of
the newly received packet n rather than substituting an erasure
frame. In other words, just because packet n was not available the
first time it was needed to reconstruct the interleaved speech, the
receiver SHOULD NOT assume it is not available when it is
subsequently needed for interleaved speech reconstruction.
7. Bundling Codec Data Frames in Type 1 Packets
As discussed in Section 6, the bundling of codec data frames is a As discussed in Section 6, the bundling of codec data frames is a
special reduced case of interleaving with LLL value in the Interleave special reduced case of interleaving with LLL value in the Interleave
Octet set to 0. Octet set to 0.
Bundling codec data frames indicates multiple data frames are Bundling codec data frames indicates that multiple data frames are
included consecutively in a packet, because the interleaving length included consecutively in a packet, because the interleaving length
(LLL) is 0. The interleaving group is thus reduced to a single RTP (LLL) is 0. The interleaving group is thus reduced to a single RTP
packet, and the reconstruction of the code data frames from RTP packet, and the reconstruction of the codec data frames from RTP
packets becomes a much simpler process. packets becomes a much simpler process.
Furthermore, the additional restrictions on senders are reduced to: Furthermore, the additional restrictions on senders are reduced to:
o MUST NOT bundle more codec data frames in a single RTP packet than o MUST NOT bundle more codec data frames in a single RTP packet than
indicated by maxptime (see Section 12) if it is signaled. indicated by maxptime (see Section 12) if it is signaled.
o SHOULD NOT bundle more codec data frames in a single RTP packet o SHOULD NOT bundle more codec data frames in a single RTP packet
than will fit in the MTU of the underlying network. than will fit in the MTU of the underlying network.
8. Handling Missing Codec Data Frames 8. Handling Missing Codec Data Frames
The vocoders covered by this payload format support erasure frame as The vocoders covered by this payload format support erasure frames as
an indication when frames are not available. The erasure frames are an indication when frames are not available. The erasure frames are
normally used internally by a receiver to advance the state of the normally used internally by a receiver to advance the state of the
voice decoder by exactly one frame time for each missing frame. Using voice decoder by exactly one frame time for each missing frame. Using
the information from packet sequence number, time stamp, and the M the information from packet sequence number, time stamp, and the M
bit, the receiver can detect missing codec data frames from RTP bit, the receiver can detect missing codec data frames from RTP
packet loss and/or silence suppression, and generate corresponding packet loss and/or silence suppression, and generate corresponding
erasure frames. Erasure frames MUST also be used in storage mode to erasure frames. Erasure frames MUST also be used in storage format to
record missing frames. record missing frames.
9. Implementation Issues 9. Implementation Issues
9.1. Interleaving Length 9.1. Interleaving Length
The vocoder interpolates the missing speech content when given an The vocoder interpolates the missing speech content when given an
erasure frame. However, the best quality is perceived by the listener erasure frame. However, the best quality is perceived by the listener
when erasure frames are not consecutive. This makes interleaving when erasure frames are not consecutive. This makes interleaving
desirable as it increases speech quality when packet loss occurs. desirable as it increases speech quality when packet loss occurs.
On the other hand, interleaving can greatly increase the end-to-end On the other hand, interleaving can greatly increase the end-to-end
delay. Where an interactive session is desired, either Type 1 delay. Where an interactive session is desired, either
Interleaved/Bundled with interleaving length (field LLL) 0 or Type 2 Interleaved/Bundled packet format with interleaving length (field
Header-Free RTP payload types are RECOMMENDED. LLL) 0 or Header-Free packet format is RECOMMENDED.
When end-to-end delay is not a primary concern, an interleaving When end-to-end delay is not a primary concern, an interleaving
length (field LLL) of 4 or 5 is RECOMMENDED as it offers a reasonable length (field LLL) of 4 or 5 is RECOMMENDED as it offers a reasonable
compromise between robustness and latency. compromise between robustness and latency.
9.2. Validation of Received Packets 9.2. Validation of Received Packets
When receiving an RTP packet, the receiver SHOULD check the validity When receiving an RTP packet, the receiver SHOULD check the validity
of the ToC fields and match the length of the packet with what is of the ToC fields and match the length of the packet with what is
indicated by the ToC fields. If any invalidity or mismatch is indicated by the ToC fields. If any invalidity or mismatch is
skipping to change at page 13, line 5 skipping to change at page 12, line 26
fields, the RTP packet SHOULD be treated as lost by the receiver for fields, the RTP packet SHOULD be treated as lost by the receiver for
the purpose of generating erasure frames as described in Section 8. the purpose of generating erasure frames as described in Section 8.
On receipt of an RTP packet in an interleave group with other than On receipt of an RTP packet in an interleave group with other than
the expected frame count value, the receiver MAY discard codec data the expected frame count value, the receiver MAY discard codec data
frames off the end of the RTP packet or add erasure codec data frames frames off the end of the RTP packet or add erasure codec data frames
to the end of the packet in order to manufacture a substitute packet to the end of the packet in order to manufacture a substitute packet
with the expected bundling value. The receiver MAY instead choose to with the expected bundling value. The receiver MAY instead choose to
discard the whole interleave group. discard the whole interleave group.
9.3. Processing the Late Packets
Assume that the receiver has begun playing frames from an interleave
group. The time has come to play frame x from packet n of the
interleave group. Further assume that packet n of the interleave
group has not been received. As described in Section 8, an erasure
frame will be sent to the receiving vocoder.
Now, assume that packet n of the interleave group arrives before
frame x+1 of that packet is needed. Receivers should use frame x+1 of
the newly received packet n rather than substituting an erasure
frame. In other words, just because packet n was not available the
first time it was needed to reconstruct the interleaved speech, the
receiver should not assume it is not available when it is
subsequently needed for interleaved speech reconstruction.
10. Mode Request 10. Mode Request
The Mode Request signal requests a particular encoding mode for the The Mode Request signal requests a particular encoding mode for the
speech encoding in the reverse direction. All implementations are speech encoding in the reverse direction. All implementations are
RECOMMENDED to honor the Mode Request signal. The Mode Request signal RECOMMENDED to honor the Mode Request signal. The Mode Request signal
SHOULD only be used in one-to-one sessions. In multiparty sessions, SHOULD only be used in one-to-one sessions. In multiparty sessions,
any received Mode Request signals SHOULD be ignored. any received Mode Request signals SHOULD be ignored.
In addition, the Mode Request signal MAY also be sent through non-RTP In addition, the Mode Request signal MAY also be sent through non-RTP
means, which is out of the scope of this specification. means, which is out of the scope of this specification.
skipping to change at page 13, line 47 skipping to change at page 13, line 32
If an unknown value above '100' (4) is received, it MUST be handled If an unknown value above '100' (4) is received, it MUST be handled
as if '100' (4) were received, for interoperability with potential as if '100' (4) were received, for interoperability with potential
future revisions. future revisions.
For SMV codec, the Mode Request field MUST be interpreted according For SMV codec, the Mode Request field MUST be interpreted according
to Table 2.2-2 of the SMV codec specifications [2]. Values above to Table 2.2-2 of the SMV codec specifications [2]. Values above
'101' (5) are currently reserved. If an unknown value above '101' (5) '101' (5) are currently reserved. If an unknown value above '101' (5)
is received, it MUST be handled as if '101' (5) were received, also is received, it MUST be handled as if '101' (5) were received, also
for interoperability with potential future revisions. for interoperability with potential future revisions.
11. Storage Mode 11. Storage Format
The storage mode is used for storing speech frames, e.g., as a file The storage format is used for storing speech frames, e.g., as a file
or e-mail attachment. or e-mail attachment.
The file begins with a magic number to identify the vocoder that is The file begins with a magic number to identify the vocoder that is
used. The magic number for EVRC corresponds to the ASCII character used. The magic number for EVRC corresponds to the ASCII character
string "#!EVRC\n", i.e., "0x23 0x21 0x45 0x56 0x52 0x43 0x0A" in string "#!EVRC\n", i.e., "0x23 0x21 0x45 0x56 0x52 0x43 0x0A". The
network byte order. The magic number for SMV corresponds to the ASCII magic number for SMV corresponds to the ASCII character string
character string "#!SMV\n", i.e., "0x23 0x21 0x53 0x4d 0x56 0x0a" in "#!SMV\n", i.e., "0x23 0x21 0x53 0x4d 0x56 0x0a".
network byte order.
The codec data frames are stored in consecutive order, with a single The codec data frames are stored in consecutive order, with a single
TOC entry field, extended to one octet, prefixing each codec data TOC entry field, extended to one octet, prefixing each codec data
frame. The ToC field is extended to one octet by setting the four frame. The ToC field is extended to one octet by setting the four
most significant bits of the octet to zero. For example, a ToC value most significant bits of the octet to zero. For example, a ToC value
of 4 (a full-rate frame) is stored as 0x04. of 4 (a full-rate frame) is stored as 0x04.
Speech frames lost in transmission and non-received frames MUST be Speech frames lost in transmission and non-received frames MUST be
stored as erasure frames (frame type 5, see definition in Section stored as erasure frames (frame type 5, see definition in Section
5.1) to maintain synchronization with the original media. 5.1) to maintain synchronization with the original media.
12. IANA Considerations 12. IANA Considerations
Two new MIME sub-types as described in this section are to be Four new MIME sub-types as described in this section are to be
registered. registered.
The MIME-names for the EVRC and SMV codec are allocated from the IETF The MIME-names for the EVRC and SMV codec are allocated from the IETF
tree since all the vocoders covered are expected to be widely used tree since all the vocoders covered are expected to be widely used
for Voice-over-IP applications. for Voice-over-IP applications.
12.1. Registration of Media Type EVRC 12.1. Registration of Media Type EVRC
Media Type Name: audio Media Type Name: audio
Media Subtype Name: EVRC Media Subtype Name: EVRC
Type 1 Interleaved/Bundled packet format for EVRC
Required Parameter: none Required Parameter: none
Optional parameters: Optional parameters:
The following parameter applies to RTP mode only. The following parameters apply to RTP transfer only.
ptime: Defined as usual for RTP audio [6]. ptime: Defined as usual for RTP audio RFC 2327.
maxptime: The maximum amount of media which can be encapsulated maxptime: The maximum amount of media which can be encapsulated
in each packet, expressed as time in milliseconds. The time in each packet, expressed as time in milliseconds. The time
SHALL be calculated as the sum of the time the media present SHALL be calculated as the sum of the time the media present
in the packet represents. The time SHOULD be a multiple of the in the packet represents. The time SHOULD be a multiple of the
duration of a single codec data frame (20 msec). If not duration of a single codec data frame (20 msec). If not
signaled, the default maxptime value SHALL be 200 signaled, the default maxptime value SHALL be 200
milliseconds. milliseconds.
maxinterleave: Maximum number for interleaving length (field LLL maxinterleave: Maximum number for interleaving length (field LLL
in the Interleaving Octet). The interleaving lengths used in in the Interleaving Octet). The interleaving lengths used in
the entire session MUST NOT exceed this maximum value. If not the entire session MUST NOT exceed this maximum value. If not
signaled, the maxinterleave length SHALL be 5. signaled, the maxinterleave length SHALL be 5.
Encoding considerations: Encoding considerations:
For RTP mode, see Section 6 and Section 7 of RFC xxxx. This type is defined for transfer of EVRC-encoded data via RTP
For storage mode, see Section 11 of RFC xxxx. using the Interleaved/Bundled packet format specified in Sections
4.1, 6, and 7 of RFC xxxx. It is also defined for other transfer
methods using the storage format specified in Section 11 of RFC
xxxx.
Security considerations: Security considerations:
See Section 14 "Security Considerations" of RFC xxxx. See Section 14 "Security Considerations" of RFC xxxx.
Public specification: Public specification:
RFC xxxx. The EVRC vocoder is specified in 3GPP2 C.S0014.
Transfer methods are specified in RFC xxxx.
Additional information: Additional information:
The following information applies for storage mode only. The following information applies for storage format only.
Magic number: #!EVRC\n Magic number: #!EVRC\n (see Section 11 of RFC xxxx)
File extensions: evc, EVC File extensions: evc, EVC
Macintosh file type code: none Macintosh file type code: none
Object identifier or OID: none Object identifier or OID: none
Intended usage: Intended usage:
COMMON. It is expected that many VoIP applications (as well as COMMON. It is expected that many VoIP applications (as well as
mobile applications) will use this type. mobile applications) will use this type.
Person & email address to contact for further information: Person & email address to contact for further information:
Adam Li Adam Li
skipping to change at page 15, line 37 skipping to change at page 15, line 31
Author/Change controller: Author/Change controller:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
IETF Audio/Video Transport Working Group IETF Audio/Video Transport Working Group
12.2. Registration of Media Type EVRC0 12.2. Registration of Media Type EVRC0
Media Type Name: audio Media Type Name: audio
Media Subtype Name: EVRC0 Media Subtype Name: EVRC0
Type 2 Header-Free packet format for EVRC
Required Parameter: none Required Parameters: none
Optional parameters: none Optional parameters: none
Encoding considerations: none Encoding considerations: none
This type is only defined for transfer of EVRC-encoded data via
RTP using the Header-Free packet format specified in Section 4.2
of RFC xxxx.
Security considerations: Security considerations:
See Section 14 "Security Considerations" of RFC xxxx. See Section 14 "Security Considerations" of RFC xxxx.
Public specification: Public specification:
RFC xxxx. The EVRC vocoder is specified in 3GPP2 C.S0014.
Transfer methods are specified in RFC xxxx.
Additional information: none Additional information: none
Intended usage: Intended usage:
COMMON. It is expected that many VoIP applications (as well as COMMON. It is expected that many VoIP applications (as well as
mobile applications) will use this type. mobile applications) will use this type.
Person & email address to contact for further information: Person & email address to contact for further information:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
skipping to change at page 16, line 19 skipping to change at page 16, line 14
Author/Change controller: Author/Change controller:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
IETF Audio/Video Transport Working Group IETF Audio/Video Transport Working Group
12.3. Registration of Media Type SMV 12.3. Registration of Media Type SMV
Media Type Name: audio Media Type Name: audio
Media Subtype Name: SMV Media Subtype Name: SMV
Type 1 Interleaved/Bundled packet format for SMV
Required Parameter: none Required Parameter: none
Optional parameters: Optional parameters:
The following parameter applies to RTP mode only. The following parameters apply to RTP transfer only.
ptime: Defined as usual for RTP audio [6]. ptime: Defined as usual for RTP audio 2327.
maxptime: The maximum amount of media which can be encapsulated maxptime: The maximum amount of media which can be encapsulated
in each packet, expressed as time in milliseconds. The time in each packet, expressed as time in milliseconds. The time
SHALL be calculated as the sum of the time the media present SHALL be calculated as the sum of the time the media present
in the packet represents. The time SHOULD be a multiple of the in the packet represents. The time SHOULD be a multiple of the
duration of a single codec data frame (20 msec). If not duration of a single codec data frame (20 msec). If not
signaled, the default maxptime value SHALL be 200 signaled, the default maxptime value SHALL be 200
milliseconds. milliseconds.
maxinterleave: Maximum number for interleaving length (field LLL maxinterleave: Maximum number for interleaving length (field LLL
in the Interleaving Octet). The interleaving lengths used in in the Interleaving Octet). The interleaving lengths used in
the entire session MUST NOT exceed this maximum value. If not the entire session MUST NOT exceed this maximum value. If not
signaled, the maxinterleave length SHALL be 5. signaled, the maxinterleave length SHALL be 5.
Encoding considerations: Encoding considerations:
For RTP mode, see Section 6 and Section 7 of RFC xxxx. This type is defined for transfer of SMV-encoded data via RTP
For storage mode, see Section 11 of RFC xxxx. using the Interleaved/Bundled packet format specified in Section
4.1, 6, and 7 of RFC xxxx. It is also defined for other transfer
methods using the storage format specified in Section 11 of RFC
xxxx.
Security considerations: Security considerations:
See Section 14 "Security Considerations" of RFC xxxx. See Section 14 "Security Considerations" of RFC xxxx.
Public specification: Public specification:
RFC xxxx. The SMV vocoder is specified in 3GPP2 C.S0030-0 v2.0.
Transfer methods are specified in RFC xxxx.
Additional information: Additional information:
The following information applies to storage mode only. The following information applies to storage format only.
Magic number: #!SMV\n Magic number: #!SMV\n (see Section 11 of RFC xxxx)
File extensions: smv, SMV File extensions: smv, SMV
Macintosh file type code: none Macintosh file type code: none
Object identifier or OID: none Object identifier or OID: none
Intended usage: Intended usage:
COMMON. It is expected that many VoIP applications (as well as COMMON. It is expected that many VoIP applications (as well as
mobile applications) will use this type. mobile applications) will use this type.
Person & email address to contact for further information: Person & email address to contact for further information:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
skipping to change at page 17, line 22 skipping to change at page 17, line 22
Author/Change controller: Author/Change controller:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
IETF Audio/Video Transport Working Group IETF Audio/Video Transport Working Group
12.4. Registration of Media Type SMV0 12.4. Registration of Media Type SMV0
Media Type Name: audio Media Type Name: audio
Media Subtype Name: SMV0 Media Subtype Name: SMV0
Type 2 Header-Free packet format for SMV
Required Parameter: none Required Parameter: none
Optional parameters: none Optional parameters: none
Encoding considerations: none Encoding considerations: none
This type is only defined for transfer of SMV-encoded data via
RTP using the Header-Free packet format specified in Section 4.2
of RFC xxxx.
Security considerations: Security considerations:
See Section 14 "Security Considerations" of RFC xxxx. See Section 14 "Security Considerations" of RFC xxxx.
Public specification: Public specification:
RFC xxxx. The SMV vocoder is specified in 3GPP2 C.S0030-0 v2.0.
Transfer methods are specified in RFC xxxx.
Additional information: none Additional information: none
Intended usage: Intended usage:
COMMON. It is expected that many VoIP applications (as well as COMMON. It is expected that many VoIP applications (as well as
mobile applications) will use this type. mobile applications) will use this type.
Person & email address to contact for further information: Person & email address to contact for further information:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
skipping to change at page 17, line 50 skipping to change at page 18, line 4
mobile applications) will use this type. mobile applications) will use this type.
Person & email address to contact for further information: Person & email address to contact for further information:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
Author/Change controller: Author/Change controller:
Adam Li Adam Li
adamli@icsl.ucla.edu adamli@icsl.ucla.edu
IETF Audio/Video Transport Working Group IETF Audio/Video Transport Working Group
13. Mapping to SDP Parameters 13. Mapping to SDP Parameters
Please note that this section applies to the RTP mode only. Please note that this section applies to the RTP transfer only.
The information carried in the MIME media type specification has a The information carried in the MIME media type specification has a
specific mapping to fields in the Session Description Protocol (SDP) specific mapping to fields in the Session Description Protocol (SDP)
[6], which is commonly used to describe RTP sessions. When SDP is [6], which is commonly used to describe RTP sessions. When SDP is
used to specify sessions employing the EVRC or EMV codec, the mapping used to specify sessions employing the EVRC or EMV codec, the mapping
is as follows: is as follows:
o The MIME type ("audio") goes in SDP "m=" as the media name. o The MIME type ("audio") goes in SDP "m=" as the media name.
o The MIME subtype (payload format name) goes in SDP "a=rtpmap" o The MIME subtype (payload format name) goes in SDP "a=rtpmap"
as the encoding name. as the encoding name.
o The parameters "ptime" and "maxptime" go in the SDP "a=ptime" o The parameters "ptime" and "maxptime" go in the SDP "a=ptime"
and "a=maxptime" attributes, respectively. and "a=maxptime" attributes, respectively.
o Any remaining parameters go in the SDP "a=fmtp" attribute by o The parameter Ÿmaxinterleave÷ goes in the SDP "a=fmtp"
copying them directly from the MIME media type string as a attribute by copying it directly from the MIME media type string
semicolon separated list of parameter=value pairs. as ÷maxinterleave=value÷.
Some examples of SDP session descriptions for EVRC and SMV encodings Some examples of SDP session descriptions for EVRC and SMV encodings
follow below. follow below.
Example of usage of EVRC: Example of usage of EVRC:
m = audio 49120 RTP/AVP 97 m = audio 49120 RTP/AVP 97
a = rtpmap:97 EVRC a = rtpmap:97 EVRC
a = fmtp:97 maxinterleave=2 a = fmtp:97 maxinterleave=2
a = maxptime:80 a = maxptime:80
skipping to change at page 19, line 18 skipping to change at page 19, line 23
As with any IP-based protocol, in some circumstances, a receiver may As with any IP-based protocol, in some circumstances, a receiver may
be overloaded simply by the receipt of too many packets, either be overloaded simply by the receipt of too many packets, either
desired or undesired. Network-layer authentication may be used to desired or undesired. Network-layer authentication may be used to
discard packets from undesired sources, but the processing cost of discard packets from undesired sources, but the processing cost of
the authentication itself may be too high. In a multicast the authentication itself may be too high. In a multicast
environment, pruning of specific sources may be implemented in environment, pruning of specific sources may be implemented in
future versions of IGMP [7] and in multicast routing protocols to future versions of IGMP [7] and in multicast routing protocols to
allow a receiver to select which sources are allowed to reach it. allow a receiver to select which sources are allowed to reach it.
Interleaving MAY affect encryption. Depending on the used encryption Interleaving may affect encryption. Depending on the used encryption
scheme there MAY be restrictions on for example the time when keys scheme there may be restrictions on for example the time when keys
can be changed. Specifically, the key change may need to occur at the can be changed. Specifically, the key change may need to occur at the
boundary between interleave groups. boundary between interleave groups.
15. Adding Support of Other Frame-Based Vocoders 15. Adding Support of Other Frame-Based Vocoders
As described above, the RTP packet format defined in this document is As described above, the RTP packet format defined in this document is
very flexible and designed to be usable by other frame-based very flexible and designed to be usable by other frame-based
vocoders. vocoders.
Additional vocoders using this format MUST have properties as Additional vocoders using this format MUST have properties as
described in Section 3.3. described in Section 3.3.
For an eligible vocoder to use the payload format mechanisms defined For an eligible vocoder to use the payload format mechanisms defined
in this document, a new RTP payload format document needs to be in this document, a new RTP payload format document needs to be
published as an RFC. That document can simply refer to this document published as a standards track RFC. That document can simply refer to
and then specify the following parameters: this document and then specify the following parameters:
o Define the unit used for RTP time stamp; o Define the unit used for RTP time stamp;
o Define the meaning of the Mode Request bits; o Define the meaning of the Mode Request bits;
o Define corresponding codec data frame type values for ToC; o Define corresponding codec data frame type values for ToC;
o Define the conversion procedure for vocoders output data frame; o Define the conversion procedure for vocoders output data frame;
o Define a magic number for storage mode, and complete the o Define a magic number for storage format, and complete the
corresponding MIME registration. corresponding MIME registration.
16. Acknowledgements 16. Acknowledgements
The following authors have made significant contributions to this The following authors have made significant contributions to this
document: Adam H. Li, John D. Villasenor, Dong-Seek Park, Jeong-Hoon document: Adam H. Li, John D. Villasenor, Dong-Seek Park, Jeong-Hoon
Park, Keith Miller, S. Craig Greer, David Leon, Nikolai Leung, Park, Keith Miller, S. Craig Greer, David Leon, Nikolai Leung,
Marcello Lioy, Kyle J. McKay, Magdalena L. Espelien, Randall Gellens, Marcello Lioy, Kyle J. McKay, Magdalena L. Espelien, Randall Gellens,
Tom Hiller, Peter J. McCann, Stinson S. Mathai, Michael D. Turner, Tom Hiller, Peter J. McCann, Stinson S. Mathai, Michael D. Turner,
Ajay Rajkumar, Dan Gal, Magnus Westerlund, Lars-Erik Jonsson, Greg Ajay Rajkumar, Dan Gal, Magnus Westerlund, Lars-Erik Jonsson, Greg
Sherwood, and Thomas Zeng. Sherwood, and Thomas Zeng.
17. References 17. References
[1] 3GPP2 C.S0014, "Enhanced Variable Rate Codec, Speech Service [1] 3GPP2 C.S0014, "Enhanced Variable Rate Codec, Speech Service
Option 3 for Wideband Spread Spectrum Digital Systems", January Option 3 for Wideband Spread Spectrum Digital Systems", January
1997. 1997.
[2] C.S0030-0 v2.0, "Selectable Mode Vocoder, Service Option for [2] 3GPP2 C.S0030-0 v2.0, "Selectable Mode Vocoder, Service Option
Wideband Spread Spectrum Communication Systems", May 2002. for Wideband Spread Spectrum Communication Systems", May 2002.
[3] Bradner, S., "Key words for use in RFCs to Indicate Requirement [3] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", BCP 14, RFC 2119, March 1997. Levels", BCP 14, RFC 2119, March 1997.
[4] Schulzrinne, H., Casner, S., Frederick, R. and V. Jacobson, [4] Schulzrinne, H., Casner, S., Frederick, R. and V. Jacobson,
"RTP: A Transport Protocol for Real-Time Applications", RFC "RTP: A Transport Protocol for Real-Time Applications", RFC
1889, January 1996. 1889, January 1996.
[5] Schulzrinne, H., "RTP Profile for Audio and Video Conferences [5] Schulzrinne, H., "RTP Profile for Audio and Video Conferences
with Minimal Control", RFC 1890, January 1996. with Minimal Control", RFC 1890, January 1996.
 End of changes. 

This html diff was produced by rfcdiff 1.23, available from http://www.levkowetz.com/ietf/tools/rfcdiff/