rfc8872xml2.original.xml   rfc8872.xml 
<?xml version="1.0" encoding="US-ASCII"?> <?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE rfc SYSTEM "rfc2629.dtd" [
<!ENTITY RFC2119 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R <!DOCTYPE rfc SYSTEM "rfc2629-xhtml.ent">
FC.2119.xml">
<!ENTITY RFC2198 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R <rfc xmlns:xi="http://www.w3.org/2001/XInclude" docName="draft-ietf-avtcore-mult
FC.2198.xml"> iplex-guidelines-12" number="8872" ipr="trust200902" submissionType="IETF" categ
<!ENTITY RFC2205 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R ory="info" consensus="true" obsoletes="" updates="" xml:lang="en" tocInclude="tr
FC.2205.xml"> ue" tocDepth="3" symRefs="true" sortRefs="true" version="3">
<!ENTITY RFC2474 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.2474.xml"> <!-- xml2rfc v2v3 conversion 2.45.3 -->
<!ENTITY RFC4588 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.4588.xml">
<!ENTITY RFC5109 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5109.xml">
<!ENTITY RFC3264 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3264.xml">
<!ENTITY RFC2974 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.2974.xml">
<!ENTITY RFC3261 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3261.xml">
<!ENTITY RFC3550 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3550.xml">
<!ENTITY RFC3389 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3389.xml">
<!ENTITY RFC3551 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3551.xml">
<!ENTITY RFC3711 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3711.xml">
<!ENTITY RFC3830 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.3830.xml">
<!ENTITY RFC4103 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.4103.xml">
<!ENTITY RFC4383 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.4383.xml">
<!ENTITY RFC4566 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.4566.xml">
<!ENTITY RFC4568 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.4568.xml">
<!ENTITY RFC4585 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.4585.xml">
<!ENTITY RFC5104 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5104.xml">
<!ENTITY RFC5389 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5389.xml">
<!ENTITY RFC5576 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5576.xml">
<!ENTITY RFC5760 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5760.xml">
<!ENTITY RFC5761 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5761.xml">
<!ENTITY RFC5764 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5764.xml">
<!ENTITY RFC5888 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.5888.xml">
<!ENTITY RFC6465 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.6465.xml">
<!ENTITY RFC7201 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.7201.xml">
<!ENTITY RFC7656 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.7656.xml">
<!ENTITY RFC7657 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.7657.xml">
<!ENTITY RFC7667 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.7667.xml">
<!ENTITY RFC7826 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.7826.xml">
<!ENTITY RFC7983 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.7983.xml">
<!ENTITY RFC8088 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.8088.xml">
<!ENTITY RFC8108 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.8108.xml">
<!ENTITY RFC8445 SYSTEM "http://xml.resource.org/public/rfc/bibxml/reference.R
FC.8445.xml">
<!ENTITY I-D.ietf-mmusic-rid SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc
/bibxml3/reference.I-D.ietf-mmusic-rid.xml">
<!ENTITY I-D.ietf-mmusic-sdp-bundle-negotiation SYSTEM "https://xml2rfc.tools.
ietf.org/public/rfc/bibxml3/reference.I-D.ietf-mmusic-sdp-bundle-negotiation.xml
">
<!ENTITY I-D.ietf-avtcore-multi-media-rtp-session SYSTEM "https://xml2rfc.tool
s.ietf.org/public/rfc/bibxml3/reference.I-D.ietf-avtcore-multi-media-rtp-session
.xml">
<!ENTITY I-D.ietf-perc-srtp-ekt-diet SYSTEM "https://xml2rfc.tools.ietf.org/pu
blic/rfc/bibxml3/reference.I-D.ietf-perc-srtp-ekt-diet.xml">
<!ENTITY I-D.ietf-avtext-rid SYSTEM "https://xml2rfc.tools.ietf.org/public/rfc
/bibxml3/reference.I-D.ietf-avtext-rid.xml">
<!ENTITY I-D.ietf-perc-private-media-framework SYSTEM "https://xml2rfc.tools.i
etf.org/public/rfc/bibxml3/reference.I-D.ietf-perc-private-media-framework.xml">
]>
<?rfc toc="yes"?>
<?rfc tocompact="yes"?>
<?rfc tocdepth="3"?>
<?rfc tocindent="yes"?>
<?rfc symrefs="yes"?>
<?rfc sortrefs="yes"?>
<?rfc comments="yes"?>
<?rfc inline="yes"?>
<?rfc compact="yes"?>
<?rfc subcompact="no"?>
<rfc
category="info"
docName="draft-ietf-avtcore-multiplex-guidelines-12"
ipr="trust200902"
submissionType="IETF">
<front> <front>
<title abbrev="Guidelines for Multiplexing in RTP">Guidelines for <title abbrev="Guidelines for Multiplexing in RTP">Guidelines for
using the Multiplexing Features of RTP to Support Multiple Media Using the Multiplexing Features of RTP to Support Multiple Media
Streams</title> Streams</title>
<author <seriesInfo name="RFC" value="8872"/>
fullname="Magnus Westerlund" <author fullname="Magnus Westerlund" initials="M." surname="Westerlund">
initials="M."
surname="Westerlund">
<organization>Ericsson</organization> <organization>Ericsson</organization>
<address> <address>
<postal> <postal>
<street>Torshamnsgatan 23</street> <street>Torshamnsgatan 23</street>
<street>SE-164 80 Kista</street> <code>164 80</code>
<street>Sweden</street> <city>Kista</city>
<country>Sweden</country>
</postal> </postal>
<phone>+46 10 714 82 87</phone>
<email>magnus.westerlund@ericsson.com</email> <email>magnus.westerlund@ericsson.com</email>
</address> </address>
</author> </author>
<author fullname="Bo Burman" initials="B." surname="Burman"> <author fullname="Bo Burman" initials="B." surname="Burman">
<organization>Ericsson</organization> <organization>Ericsson</organization>
<address> <address>
<postal> <postal>
<street>Gronlandsgatan 31</street> <street>Gronlandsgatan 31</street>
<street>SE-164 60 Kista</street> <code>164 60</code>
<street>Sweden</street> <city>Kista</city>
<country>Sweden</country>
</postal> </postal>
<email>bo.burman@ericsson.com</email> <email>bo.burman@ericsson.com</email>
</address> </address>
</author> </author>
<author fullname="Colin Perkins" initials="C." surname="Perkins"> <author fullname="Colin Perkins" initials="C." surname="Perkins">
<organization>University of Glasgow</organization> <organization>University of Glasgow</organization>
<address> <address>
<postal> <postal>
<street>School of Computing Science</street> <extaddr>School of Computing Science</extaddr>
<street>Glasgow G12 8QQ</street> <city>Glasgow</city>
<street>United Kingdom</street> <code>G12 8QQ</code>
<country>United Kingdom</country>
</postal> </postal>
<email>csp@csperkins.org</email> <email>csp@csperkins.org</email>
</address> </address>
</author> </author>
<author <author fullname="Harald Tveit Alvestrand" initials="H." surname="Alvestrand
fullname="Harald Tveit Alvestrand" ">
initials="H."
surname="Alvestrand">
<organization>Google</organization> <organization>Google</organization>
<address> <address>
<postal> <postal>
<street>Kungsbron 2</street> <street>Kungsbron 2</street>
<street>Stockholm 11122</street> <city>Stockholm</city>
<street>Sweden</street> <code>11122</code>
<country>Sweden</country>
</postal> </postal>
<email>harald@alvestrand.no</email> <email>harald@alvestrand.no</email>
</address> </address>
</author> </author>
<author fullname="Roni Even" initials="R." surname="Even"> <author fullname="Roni Even" initials="R." surname="Even">
<address> <address>
<email>ron.even.tlv@gmail.com</email> <email>ron.even.tlv@gmail.com</email>
</address> </address>
</author> </author>
<date day="16" month="June" year="2020"/> <date month="January" year="2021"/>
<keyword>Simulcast</keyword>
<abstract> <abstract>
<t>The Real-time Transport Protocol (RTP) is a flexible protocol that <t>The Real-time Transport Protocol (RTP) is a flexible protocol that
can be used in a wide range of applications, networks, and system can be used in a wide range of applications, networks, and system
topologies. That flexibility makes for wide applicability, but can topologies. That flexibility makes for wide applicability but can
complicate the application design process. One particular design complicate the application design process. One particular design
question that has received much attention is how to support multiple question that has received much attention is how to support multiple
media streams in RTP. This memo discusses the available options and media streams in RTP. This memo discusses the available options and
design trade-offs, and provides guidelines on how to use the design trade-offs, and provides guidelines on how to use the
multiplexing features of RTP to support multiple media streams.</t> multiplexing features of RTP to support multiple media streams.</t>
</abstract> </abstract>
</front> </front>
<middle> <middle>
<section anchor="section-1" title="Introduction"> <section anchor="sect-1" numbered="true" toc="default">
<name>Introduction</name>
<t>The Real-time Transport Protocol (RTP) <t>The Real-time Transport Protocol (RTP)
<xref target="RFC3550"/> <xref target="RFC3550" format="default"/>
is a commonly used protocol for real-time media transport. It is a is a commonly used protocol for real-time media transport. It is a
protocol that provides great flexibility and can support a large set protocol that provides great flexibility and can support a large set
of different applications. RTP was from the beginning designed for of different applications. From the beginning, RTP was designed for
multiple participants in a communication session. It supports many multiple participants in a communication session. It supports many
topology paradigms and usages, as defined in topology paradigms and usages, as defined in
<xref target="RFC7667"/>. RTP has several multiplexing points designed <xref target="RFC7667" format="default"/>. RTP has several multiplexing
for different purposes. These enable support of multiple RTP streams points designed
and switching between different encoding or packetization of the for different purposes; these points enable support of multiple RTP stre
ams
and switching between different encoding or packetization techniques for
the
media. By using multiple RTP sessions, sets of RTP streams can be media. By using multiple RTP sessions, sets of RTP streams can be
structured for efficient processing or identification. Thus, an structured for efficient processing or identification. Thus,
RTP application designer needs to understand how to best use the RTP to meet an application's needs, an RTP application designer needs to understand
session, the RTP stream identifier (SSRC), and the RTP payload type to how best to use the RTP
meet the application's needs.</t> session, the RTP stream identifier (synchronization source (SSRC)), and
<t>There have been increased interest in more advanced usage of RTP. the RTP payload type.</t>
<t>There has been increased interest in more-advanced usage of RTP.
For example, multiple RTP streams can be used when a single endpoint For example, multiple RTP streams can be used when a single endpoint
has multiple media sources (like multiple cameras or microphones) that has multiple media sources (like multiple cameras or microphones) from
need to be sent simultaneously. Consequently, questions are raised which streams of media need to be sent simultaneously. Consequently, que
stions are raised
regarding the most appropriate RTP usage. The limitations in some regarding the most appropriate RTP usage. The limitations in some
implementations, RTP/RTCP extensions, and signalling have also been implementations, RTP/RTCP extensions, and signaling have also been
exposed. This document aims to clarify the usefulness exposed. This document aims to clarify the usefulness
of some functionalities in RTP which will hopefully result in more compl of some functionalities in RTP that, hopefully, will result in future
ete implementations that are more complete.</t>
implementations in the future.</t>
<t>The purpose of this document is to provide clear information about <t>The purpose of this document is to provide clear information about
the possibilities of RTP when it comes to multiplexing. The RTP the possibilities of RTP when it comes to multiplexing. The RTP
application designer needs to understand the implications arising application designer needs to understand the implications arising
from a particular usage of the RTP multiplexing points. The document from a particular usage of the RTP multiplexing points. This document
will provide some guidelines and recommend against some usages as provides some guidelines and recommends against some usages as
being unsuitable, in general or for particular purposes.</t> being unsuitable, in general or for particular purposes.</t>
<t>The document starts with some definitions and then goes into the <t>This document starts with some definitions and then goes into
existing RTP functionalities around multiplexing. Both the desired existing RTP functionalities around multiplexing. Both the desired
behaviour and the implications of a particular behaviour depend on behavior and the implications of a particular behavior depend on
which topologies are used, which requires some consideration. This is which topologies are used; therefore, this topic requires some
followed by a discussion of some choices in multiplexing behaviour and consideration. We then discuss some choices regarding multiplexing
their impacts. Some designs of RTP usage are discussed. Finally, some behavior and the impacts of those choices. Some designs of RTP usage
are also discussed. Finally, some
guidelines and examples are provided.</t> guidelines and examples are provided.</t>
</section> </section>
<section anchor="section-2" title="Definitions"> <section anchor="sect-2" numbered="true" toc="default">
<section anchor="section-2.1" title="Terminology"> <name>Definitions</name>
<t>The definitions in Section 3 of <section anchor="sect-2.1" numbered="true" toc="default">
<xref target="RFC3550"/> <name>Terminology</name>
are referenced normatively.</t> <t>The definitions in <xref target="RFC3550" sectionFormat="of"
<t>The taxonomy defined in section="3"/> are referenced normatively.</t>
<xref target="RFC7656"/> <t>The taxonomy defined in <xref target="RFC7656" format="default"/>
is referenced normatively.</t> is referenced normatively.</t>
<t>The following terms and abbreviations are used in this document:</t> <t>The following terms and abbreviations are used in this document:</t>
<t> <dl newline="true" spacing="normal">
<list hangIndent="3" style="hanging"> <dt>Multi-party:</dt>
<t hangText="Multiparty:">A communication situation including multip <dd>Communication that includes multiple endpoints.
le endpoints. In this document, "multi-party" will be used to refer to scenarios whe
<vspace blankLines="0"/> re
In this document, it will be used to refer to situations where mor more than two endpoints communicate.</dd>
e <dt>Multiplexing:</dt>
than two endpoints communicate.</t> <dd>An operation that takes multiple entities as input, aggregating
<t hangText="Multiplexing:">The operation of taking multiple entitie them onto some common resource while keeping the individual entities
s as input, addressable such that they can later be fully and unambiguously
<vspace blankLines="0"/> separated (demultiplexed) again.</dd>
aggregating them onto some common resource while keeping the <dt>RTP Receiver:</dt>
individual entities addressable such that they can later be fully <dd>An endpoint or middlebox receiving RTP streams and RTCP
and messages. It uses at least one SSRC to send RTCP messages. An RTP
unambiguously separated (de-multiplexed) again.</t> receiver may also be an RTP sender.</dd>
<t hangText="RTP Receiver:">An Endpoint or Middlebox receiving RTP <dt>RTP Sender:</dt>
streams and RTCP messages. It uses at least one SSRC to send RTCP <dd>An endpoint sending one or more RTP streams but also sending
messages. An RTP Receiver may also be an RTP Sender. RTCP messages.</dd>
</t> <dt>RTP Session Group:</dt>
<t hangText="RTP Sender:">An Endpoint sending one or more RTP <dd>One or more RTP sessions that are used together to perform some
streams, but also sending RTCP messages. function. Examples include multiple RTP sessions used to carry differe
</t> nt
<t hangText="RTP Session Group:">One or more RTP sessions that are us layers of a layered encoding. In an RTP Session Group, CNAMEs are
ed together assumed to be valid across all RTP sessions and designate
<vspace blankLines="0"/> synchronization contexts that can cross RTP sessions; i.e., SSRCs
to perform some function. Examples are multiple RTP sessions used that map to a common CNAME can be assumed to have RTCP Sender Report
to (SR) timing information derived from a common clock such that they
carry different layers of a layered encoding. In an RTP Session Gr can be synchronized for playout.</dd>
oup, <dt>Signaling:</dt>
CNAMEs are assumed to be valid across all RTP sessions, and design <dd>The process of configuring endpoints to participate in one or
ate more RTP sessions.</dd>
synchronisation contexts that can cross RTP sessions; i.e. SSRCs t </dl>
hat <aside><t> Note: The above definitions of "RTP receiver" and "RTP sender
map to a common CNAME can be assumed to have RTCP Sender Report (S " are
R) timing consistent with the usage in <xref target="RFC3550" format="default"/>
information derived from a common clock such that they can be .
synchronised for playout. </t></aside>
</t>
<t hangText="Signalling:">The process of configuring endpoints to pa
rticipate in
<vspace blankLines="0"/>
one or more RTP sessions.</t>
</list>
</t>
<t> Note: The above definitions of RTP Receiver and RTP Sender are
consistent with the usage in <xref target="RFC3550"/>.
</t>
</section> </section>
<section anchor="section-2.2" title="Subjects Out of Scope"> <section anchor="sect-2.2" numbered="true" toc="default">
<name>Focus of This Document</name>
<t>This document is focused on issues that affect RTP. Thus, issues <t>This document is focused on issues that affect RTP. Thus, issues
that involve signalling protocols, such as whether SIP that involve signaling protocols -- such as whether SIP
<xref target="RFC3261"/>, Jingle <xref target="JINGLE"/> or some <xref target="RFC3261" format="default"/>, Jingle <xref target="JINGLE"
other protocol is in use for session configuration, the particular format="default"/>, or some
syntaxes used to define RTP session properties, or the constraints other protocol is in use for session configuration; the particular
imposed by particular choices in the signalling protocols, are syntaxes used to define RTP session properties; or the constraints
imposed by particular choices in the signaling protocols -- are
mentioned only as examples in order to describe the RTP issues more mentioned only as examples in order to describe the RTP issues more
precisely.</t> precisely.</t>
<t>This document assumes the applications will use RTCP. While there <t>This document assumes that the applications will use RTCP. While ther e
are applications that don't send RTCP, they do not conform to the RTP are applications that don't send RTCP, they do not conform to the RTP
specification, and thus can be regarded as reusing the RTP packet specification and thus can be regarded as reusing the RTP packet
format but not implementing the RTP protocol.</t> format but not implementing RTP.</t>
</section> </section>
</section> </section>
<section anchor="section-3" title="RTP Multiplexing Overview"> <section anchor="sect-3" numbered="true" toc="default">
<section <name>RTP Multiplexing Overview</name>
anchor="section-3.1" <section anchor="sect-3.1" numbered="true" toc="default">
title="Reasons for Multiplexing and Grouping RTP Streams"> <name>Reasons for Multiplexing and Grouping RTP Streams</name>
<t>There are several reasons why an endpoint might choose to send <t>There are several reasons why an endpoint might choose to send
multiple media streams. In the below discussion, please keep in mind multiple media streams. In the discussion below, please keep in mind
that the reasons for having multiple RTP streams vary and include but that the reasons for having multiple RTP streams vary and include, but
are not limited to the following:</t> are not limited to, the following:</t>
<t> <ul spacing="normal">
<list style="symbols"> <li>There might be multiple media sources.</li>
<t>Multiple media sources</t> <li>
<t>Multiple RTP streams might be needed to represent one media sourc <t>Multiple RTP streams might be needed to represent one media
e for instance: source, for example:
<list style="symbols"> </t>
<t>To carry different layers of an scalable encoding of a media s <ul spacing="normal">
ource</t> <li>To carry different layers of a scalable encoding of a media so
<t>Alternative encodings during simulcast, for instance using dif urce</li>
ferent codecs for the <li>Alternative encodings during simulcast, using different codecs
same audio stream</t> for the
<t>Alternative formats during simulcast, for instance multiple r same audio stream</li>
esolutions of the same <li>Alternative formats during simulcast, multiple resolutions of
video stream</t> the same
</list> video stream</li>
</t> </ul>
<t>A retransmission stream might repeat some parts of the content of </li>
another RTP stream</t> <li>A retransmission stream might repeat some parts of the content of
<t>A Forward Error Correction (FEC) stream might provide material th another RTP stream.</li>
at <li>A Forward Error Correction (FEC) stream might provide material tha
can be used to repair another RTP stream</t> t
</list> can be used to repair another RTP stream.</li>
</t> </ul>
<t>For each of these reasons, it is necessary to decide if each <t>For each of these reasons, it is necessary to decide whether each
additional RTP stream is sent within the same RTP session as the other additional RTP stream is sent within the same RTP session as the other
RTP streams, or if it is necessary to use additional RTP sessions to RTP streams or it is necessary to use additional RTP sessions to
group the RTP streams. The choice suitable for one situation, might no group the RTP streams. For a combination of reasons, the suitable choi
t ce for one situation might not
be the choice suitable in another situation or combination of reasons. be the suitable choice for another situation. The choice is easiest
The clearest understanding when multiplexing multiple media sources of the same
is associated with multiplexing multiple media sources of the same
media type. However, all reasons warrant discussion and clarification media type. However, all reasons warrant discussion and clarification
on how to deal with them. As the discussion below will show, in regarding how to deal with them. As the discussion below will show,
reality we cannot choose a single one of SSRC or RTP session a single solution does not suit all purposes.
multiplexing solutions for all purposes. To utilise RTP well and as ef To utilize RTP well and as efficiently as
ficiently as possible, both are needed.
possible, both are needed. The real issue is finding the right The real issue is knowing when to create multiple RTP sessions versus when to
guidance on when to create additional RTP sessions and when additional send multiple RTP streams in a single RTP session.</t>
RTP streams in the same RTP session is the right choice.</t>
</section> </section>
<section anchor="section-3.2" title="RTP Multiplexing Points"> <section anchor="sect-3.2" numbered="true" toc="default">
<t>This section describes the multiplexing points present in the RTP <name>RTP Multiplexing Points</name>
protocol that can be used to distinguish RTP streams and groups of RTP <t>This section describes the multiplexing points present in RTP
streams. Figure 1 outlines the process of demultiplexing incoming RTP that can be used to distinguish RTP streams and groups of RTP
streams starting already at the socket representing reception of one streams. <xref target="ref-rtp-demultiplexing-process"/> outlines
or more transport flows, e.g. based on the UDP destination port. It als the process of demultiplexing incoming RTP
o demultiplexes streams, starting with one or more sockets representing the reception
RTP/RTCP from any other protocols, such as STUN <xref target="RFC5389"/ of one
> or more transport flows, e.g., based on the UDP destination port. It a
and DTLS-SRTP <xref target="RFC5764"/> on the same transport as lso demultiplexes
described in <xref target="RFC7983"/>. The Processing and Buffering (PB RTP/RTCP from any other protocols, such as Session Traversal
) Utilities for NAT (STUN) <xref target="RFC5389" format="default"/>
step of Figure 1 terminates the RTP/RTCP protocol and prepares the and DTLS-SRTP <xref target="RFC5764" format="default"/> on the same tr
ansport as
described in <xref target="RFC7983" format="default"/>.
The Processing and Buffering (PB)
step in <xref target="ref-rtp-demultiplexing-process"/> terminates
RTP/RTCP and prepares the
RTP payload for input to the decoder.</t> RTP payload for input to the decoder.</t>
<figure <figure anchor="ref-rtp-demultiplexing-process">
anchor="ref-rtp-demultiplexing-process" <name>RTP Demultiplexing Process</name>
title="RTP Demultiplexing Process"> <artwork name="" type="" align="left" alt=""><![CDATA[
<artwork>
<![CDATA[
| | | | | |
| | | packets | | | packets
+-- v v v +-- v v v
| +------------+ | +------------+
| | Socket(s) | Transport Protocol Demultiplexing | | Socket(s) | Transport Protocol Demultiplexing
| +------------+ | +------------+
| || || | || ||
RTP | RTP/ || |+-----> DTLS (SRTP Keying, SCTP, etc) RTP | RTP/ || |+-----> DTLS (SRTP keying, SCTP, etc.)
Session | RTCP || +------> STUN (multiplexed using same port) Session | RTCP || +------> STUN (multiplexed using same port)
+-- || +-- ||
+-- || +-- ||
| ++(split by SSRC)-++---> Identify SSRC collision | ++(split by SSRC)-++---> Identify SSRC collision
| || || || || | || || || ||
| (associate with signalling by MID/RID) | (associate with signaling by MID/RID)
| vv vv vv vv | vv vv vv vv
RTP | +--+ +--+ +--+ +--+ Jitter buffer, RTP | +--+ +--+ +--+ +--+ Jitter buffer,
Streams | |PB| |PB| |PB| |PB| process RTCP, etc. Streams | |PB| |PB| |PB| |PB| process RTCP, etc.
| +--+ +--+ +--+ +--+ | +--+ +--+ +--+ +--+
+-- | | | | +-- | | | |
(select decoder based on PT) (select decoder based on payload type (PT))
+-- | / | / +-- | / | /
| +-----+ | / | +-----+ | /
| / | |/ | / | |/
Payload | v v v Payload | v v v
Formats | +---+ +---+ +---+ Formats | +---+ +---+ +---+
| |Dec| |Dec| |Dec| Decoders | |Dec| |Dec| |Dec| Decoders
| +---+ +---+ +---+ | +---+ +---+ +---+
+-- +--]]></artwork>
]]>
</artwork>
</figure> </figure>
<t/> <section anchor="sect-3.2.1" numbered="true" toc="default">
<section anchor="section-3.2.1" title="RTP Session"> <name>RTP Session</name>
<t>An RTP session is the highest semantic layer in the RTP protocol, <t>An RTP session is the highest semantic layer in RTP
and represents an association between a group of communicating and represents an association between a group of communicating
endpoints. RTP does not contain a session identifier, yet different endpoints. RTP does not contain a session identifier, yet different
RTP sessions must be possible to identify both across a set of diffe rent RTP sessions must be possible to identify both across a set of diffe rent
endpoints and from the perspective of a single endpoint.</t> endpoints and from the perspective of a single endpoint.</t>
<t>For RTP session separation across endpoints, the set of <t>For RTP session separation across endpoints, the set of
participants that form an RTP session is defined as those that share a participants that form an RTP session is defined as those that share a
single synchronisation source space single SSRC space
<xref target="RFC3550"/>. That is, if a group of participants are ea <xref target="RFC3550" format="default"/>. That is, if a group of pa
ch rticipants are each
aware of the synchronisation source identifiers belonging to the oth aware of the SSRC identifiers belonging to the other
er
participants, then those participants are in a single RTP session. A participants, then those participants are in a single RTP session. A
participant can become aware of a synchronisation source identifier participant can become aware of an SSRC identifier by
by receiving an RTP packet containing the identifier in the SSRC field
receiving an RTP packet containing it in the SSRC field or CSRC list or
, contributing source (CSRC) list,
by receiving an RTCP packet mentioning it in an SSRC field, or throu by receiving an RTCP packet listing it in an SSRC field, or through
gh signaling (e.g., the Session Description Protocol (SDP)
signalling (e.g., the Session Description Protocol (SDP) <xref target="RFC4566" format="default"/>
<xref target="RFC4566"/>
"a=ssrc:" attribute "a=ssrc:" attribute
<xref target="RFC5576"/>). Thus, the scope of an RTP session is <xref target="RFC5576" format="default"/>). Thus, the scope of an RT P session is
determined by the participants' network interconnection topology, in determined by the participants' network interconnection topology, in
combination with RTP and RTCP forwarding strategies deployed by the combination with RTP and RTCP forwarding strategies deployed by the
endpoints and any middleboxes, and by the signalling.</t> endpoints and any middleboxes, and by the signaling.</t>
<t>For RTP session separation within a single endpoint RTP relies on <t>For RTP session separation within a single endpoint, RTP relies on
the underlying transport layer, and on the signalling to identify RT the underlying transport layer and the signaling to identify RTP
P
sessions in a manner that is meaningful to the application. A single sessions in a manner that is meaningful to the application. A single
endpoint can have one or more transport flows for the same RTP endpoint can have one or more transport flows for the same RTP
session, and a single RTP session can span multiple transport session, and a single RTP session can span multiple transport-layer
layer flows even if all endpoints use a single transport layer flow flows even if all endpoints use a single transport-layer flow per endpoint
per endpoint for that RTP session. The signaling layer might give RTP sessions an
for that RTP session. The signalling layer might give RTP sessions a explicit
n explicit
identifier, or the identification might be implicit based on the identifier, or the identification might be implicit based on the
addresses and ports used. Accordingly, a single RTP session can have addresses and ports used. Accordingly, a single RTP session can have
multiple associated identifiers, explicit and implicit, belonging to multiple associated identifiers, explicit and implicit, belonging to
different contexts. For example, when running RTP on top of UDP/IP, an different contexts. For example, when running RTP on top of UDP/IP, an
endpoint can identify and delimit an RTP session from other RTP endpoint can identify and delimit an RTP session from other RTP
sessions by their UDP source and destination IP addresses and UDP po sessions by their UDP source and destination IP addresses and
rt numbers. their UDP port numbers.
A single RTP session can be using multiple IP/UDP flows for receiving A single RTP session can be using multiple IP/UDP flows for receivin
and/or g and/or
sending RTP packets to other endpoints or middleboxes, even if the sending RTP packets to other endpoints or middleboxes, even if the
endpoint does not have multiple IP addresses. Using multiple IP addre endpoint does not have multiple IP addresses. Using multiple IP addr
sses esses
only makes it more likely to require multiple IP/UDP flows. only makes it more likely that multiple IP/UDP flows will be
Another example is SDP media descriptions (the "m=" line and the required. Another example is SDP media descriptions (the "m=" line a
following associated lines) that signal the transport flow and RTP s nd the
ession subsequent associated lines) that signal the transport flow and RTP
session
configuration for the endpoint's part of the RTP session. The SDP gr ouping configuration for the endpoint's part of the RTP session. The SDP gr ouping
framework framework
<xref target="RFC5888"/> <xref target="RFC5888" format="default"/>
allows labeling of the media descriptions to be used so that allows labeling of the media descriptions to be used so that
RTP Session Groups can be created. Through use of Negotiating Media RTP Session Groups can be created. Through the use of
Multiplexing <xref target="RFC8843">"Negotiating Media Multiplexing Using the
Using the Session Description Protocol (SDP) Session Description Protocol (SDP)"</xref>,
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/>,
multiple media descriptions become part of a common RTP session wher e each multiple media descriptions become part of a common RTP session wher e each
media description represents the RTP streams sent or received for a media source.</t> media description represents the RTP streams sent or received for a media source.</t>
<t>The RTP protocol makes no normative statements about the <t>RTP makes no normative statements about the
relationship between different RTP sessions, however the application relationship between different RTP sessions; however, applications
s that use more than one RTP session need to understand how the
that use more than one RTP session will have some higher layer different RTP sessions that they create relate to one another.</t>
understanding of the relationship between the sessions they create.<
/t>
</section> </section>
<section anchor="section-3.2.2" title="Synchronisation Source (SSRC)"> <section anchor="sect-3.2.2" numbered="true" toc="default">
<t>A synchronisation source (SSRC) identifies a source of an RTP <name>Synchronization Source (SSRC)</name>
<t>An SSRC identifies a source of an RTP
stream, or an RTP receiver when sending RTCP. Every endpoint has at stream, or an RTP receiver when sending RTCP. Every endpoint has at
least one SSRC identifier, even if it does not send RTP packets. RTP least one SSRC identifier, even if it does not send RTP packets. RTP
endpoints that are only RTP receivers still send RTCP and use their endpoints that are only RTP receivers still send RTCP and use their
SSRC identifiers in the RTCP packets they send. An endpoint can have SSRC identifiers in the RTCP packets they send. An endpoint can have
multiple SSRC identifiers if it sends multiple RTP streams. Endpoint s multiple SSRC identifiers if it sends multiple RTP streams. Endpoint s
that are both RTP sender and RTP receiver use the same SSRC(s) in that function as both RTP sender and RTP receiver use the same SSRC( s) in
both roles.</t> both roles.</t>
<t>The SSRC is a 32-bit identifier. It is present in every RTP and <t>The SSRC is a 32-bit identifier. It is present in every RTP and
RTCP packet header, and in the payload of some RTCP packet types. It RTCP packet header and in the payload of some RTCP packet types. It
can also be present in SDP signalling. Unless pre-signalled, e.g. can also be present in SDP signaling. Unless presignaled, e.g.,
using the SDP "a=ssrc:" attribute using the SDP "a=ssrc:" attribute
<xref target="RFC5576"/>, the SSRC is chosen at random. It is not <xref target="RFC5576" format="default"/>, the SSRC is chosen at ran
dependent on the network address of the endpoint, and is intended to dom. It is not
be unique within an RTP session. SSRC collisions can occur, and are dependent on the network address of the endpoint and is intended to
be unique within an RTP session. SSRC collisions can occur and are
handled as specified in handled as specified in
<xref target="RFC3550"/> <xref target="RFC3550" format="default"/>
and and
<xref target="RFC5576"/>, resulting in the SSRC of the colliding RTP <xref target="RFC5576" format="default"/>, resulting in the SSRC of the colliding RTP
streams or receivers changing. An endpoint that changes streams or receivers changing. An endpoint that changes
its network transport address during a session has to choose a new its network transport address during a session has to choose a new
SSRC identifier to avoid being interpreted as looped source, unless SSRC identifier to avoid being interpreted as a looped source, unles
a mechanism providing a virtual transport (such as ICE s
<xref target="RFC8445"/>) abstracts the changes.</t> a mechanism providing a virtual transport (such as Interactive
<t>SSRC identifiers that belong to the same synchronisation context Connectivity Establishment (ICE)
(i.e., that represent RTP streams that can be synchronised using <xref target="RFC8445" format="default"/>) abstracts the changes.</t
>
<t>SSRC identifiers that belong to the same synchronization context
(i.e., that represent RTP streams that can be synchronized using
information in RTCP SR packets) use identical CNAME chunks in information in RTCP SR packets) use identical CNAME chunks in
corresponding RTCP SDES packets. SDP signalling can also be used to corresponding RTCP source description (SDES) packets. SDP signaling can also be used to
provide explicit SSRC grouping provide explicit SSRC grouping
<xref target="RFC5576"/>.</t> <xref target="RFC5576" format="default"/>.</t>
<t>In some cases, the same SSRC identifier value is used to relate <t>In some cases, the same SSRC identifier value is used to relate
streams in two different RTP sessions, such as in RTP retransmission streams in two different RTP sessions, such as in RTP retransmission
<xref target="RFC4588"/>. This is to be avoided since there is no <xref target="RFC4588" format="default"/>. This is to be avoided, si
guarantee that SSRC values are unique across RTP sessions. For the R nce there is no
TP guarantee that SSRC values are unique across RTP sessions. In the
retransmission case of RTP retransmission
<xref target="RFC4588"/> <xref target="RFC4588" format="default"/>,
case it is recommended to use explicit binding of the source RTP it is recommended to use explicit binding of the source RTP
stream and the redundancy stream, e.g. using the RepairedRtpStreamId stream and the redundancy stream, e.g., using the RepairedRtpStreamI
RTCP SDES item <xref target="I-D.ietf-avtext-rid"/>. The d
RepairedRtpStreamId is a rather recent mechanism, so one cannot expec RTCP SDES item <xref target="RFC8852" format="default"/>. The
t RepairedRtpStreamId is a rather recent mechanism, so one cannot expe
older applications to follow this recommendation. ct
</t> older applications to follow this recommendation.
<t>Note that RTP sequence number and RTP timestamp are scoped by the </t>
SSRC and thus specific per RTP stream.</t> <t>Note that the RTP sequence number and RTP timestamp are scoped by t
he
<t>Different types of entities use an SSRC to identify themselves, as SSRC and are thus specific per RTP stream.</t>
follows: <t>Different types of entities use an SSRC to identify themselves, as
</t> follows:
<t> </t>
<list hangIndent="3" style="hanging"> <ul spacing="normal">
<t hangText="A real media source:">Uses the SSRC to identify a "phy <li>A real media source uses the SSRC to identify a "physical" media
sical" source.</li>
media source.</t> <li>A conceptual media source uses the SSRC to identify the result o
<t hangText="A conceptual media source:">Uses the SSRC to identify f
the result of applying some filtering function in a network node -- for exampl
applying some filtering function in a network node, for example e, a
a
filtering function in an RTP mixer that provides the most active filtering function in an RTP mixer that provides the most active
speaker based on some criteria, or a mix representing a set of o ther speaker based on some criteria, or a mix representing a set of o ther
sources.</t> sources.</li>
<t hangText="An RTP receiver:">Uses the SSRC to identify itself as <li>An RTP receiver uses the SSRC to identify itself as the
the source of its RTCP reports.</li>
source of its RTCP reports.</t> </ul>
</list> <t>An endpoint that generates more than one media type, e.g.,
</t>
<t>An endpoint that generates more than one media type, e.g.
a conference participant sending both audio and video, need not (and , a conference participant sending both audio and video, need not (and ,
indeed, should not) use the same SSRC value across RTP sessions. RTC indeed, should not) use the same SSRC value across RTP
P compound sessions. Using RTCP compound
packets containing the CNAME SDES item is the designated method to packets containing the CNAME SDES item is the designated method for
bind an SSRC to a CNAME, effectively cross-correlating SSRCs within binding an SSRC to a CNAME, effectively cross-correlating SSRCs with
and between RTP Sessions as coming from the same endpoint. The main in
and between RTP sessions as coming from the same endpoint. The main
property attributed to SSRCs associated with the same CNAME is that property attributed to SSRCs associated with the same CNAME is that
they are from a particular synchronisation context and can be they are from a particular synchronization context and can be
synchronised at playback.</t> synchronized at playback.</t>
<t>An RTP receiver receiving a previously unseen SSRC value will <t>An RTP receiver receiving a previously unseen SSRC value will
interpret it as a new source. It might in fact be a previously interpret it as a new source. It might in fact be a previously
existing source that had to change SSRC number due to an SSRC existing source that had to change its SSRC number due to an SSRC
conflict. Use of the MID extension conflict. Using the media identification (MID) extension
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/> helps to ide <xref target="RFC8843" format="default"/> helps to identify
ntify which media source the new SSRC represents, and using the
which media source the new SSRC represents and use of the RID extensi restriction identifier (RID) extension
on <xref target="RFC8851" format="default"/> helps to identify what enc
<xref target="I-D.ietf-mmusic-rid"/> helps to identify what encoding oding
or redundancy stream it represents, even though the SSRC changed. or redundancy stream it represents, even though the SSRC changed.
However, the originator of the previous SSRC ought to have However, the originator of the previous SSRC ought to have
ended the conflicting source by sending an RTCP BYE for it prior to ended the conflicting source by sending an RTCP BYE for it prior to
starting to send with the new SSRC, making the new SSRC a new source .</t> starting to send with the new SSRC, making the new SSRC a new source .</t>
</section> </section>
<section anchor="section-3.2.3" title="Contributing Source (CSRC)"> <section anchor="sect-3.2.3" numbered="true" toc="default">
<t>The Contributing Source (CSRC) is not a separate identifier. Rather <name>Contributing Source (CSRC)</name>
<t>The CSRC is not a separate identifier. Rather,
an SSRC identifier is listed as a CSRC in the RTP header of a packet an SSRC identifier is listed as a CSRC in the RTP header of a packet
generated by an RTP mixer or video MCU/switch, if the corresponding generated by an RTP mixer or video Multipoint Control Unit (MCU) /
SSRC switch, if the corresponding SSRC
was in the header of one of the packets that contributed to the outp ut.</t> was in the header of one of the packets that contributed to the outp ut.</t>
<t>It is not possible, in general, to extract media represented by an <t>It is not possible, in general, to extract media represented by an
individual CSRC since it is typically the result of a media merge individual CSRC, since it is typically the result of a media merge
(e.g. mix) operation on the individual media streams (e.g., mix) operation on the individual media streams
corresponding to the CSRC identifiers. The exception is the case whe corresponding to the CSRC identifiers. The exception is the case whe
n re
only a single CSRC is indicated as this represent forwarding of an R only a single CSRC is indicated, as this represents the forwarding o
TP f an RTP
stream, possibly modified. The RTP header extension for stream that might have been modified. The RTP header extension (<xre
Mixer-to-Client Audio Level Indication f target="RFC6465">"A Real-time Transport Protocol (RTP)
<xref target="RFC6465"/> Header Extension for Mixer-to-Client Audio Level Indication"</xref>)
expands on the receiver's information about a packet with a CSRC lis t. expands on the receiver's information about a packet with a CSRC lis t.
Due to these restrictions, CSRC will not be considered a fully Due to these restrictions, a CSRC will not be considered a fully
qualified multiplexing point and will be disregarded in the rest of qualified multiplexing point and will be disregarded in the rest of
this document.</t> this document.</t>
</section> </section>
<section anchor="section-3.2.4" title="RTP Payload Type"> <section anchor="sect-3.2.4" numbered="true" toc="default">
<t>Each RTP stream utilises one or more RTP payload formats. An RTP <name>RTP Payload Type</name>
<t>Each RTP stream utilizes one or more RTP payload formats. An RTP
payload format describes how the output of a particular media codec is payload format describes how the output of a particular media codec is
framed and encoded into RTP packets. The payload format is framed and encoded into RTP packets. The payload format is
identified by the payload type (PT) field in the RTP packet header. identified by the payload type (PT) field in the RTP packet header.
The combination of SSRC and PT therefore identifies a specific RTP s tream The combination of SSRC and PT therefore identifies a specific RTP s tream
in a specific encoding format. The format definition can be taken fro in a specific encoding format. The format definition can be taken fr
m om
<xref target="RFC3551"/> <xref target="RFC3551" format="default"/>
for statically allocated payload types, but ought to be explicitly for statically allocated payload types but ought to be explicitly
defined in signalling, such as SDP, both for static and dynamic defined in signaling, such as SDP, for both static and dynamic
payload types. The term "format" here includes those aspects describ ed payload types. The term "format" here includes those aspects describ ed
by out-of-band signalling means; in SDP, the term "format" includes by out-of-band signaling means; in SDP, the term "format" includes
media type, RTP timestamp sampling rate, codec, codec configuration, media type, RTP timestamp sampling rate, codec, codec configuration
payload format configurations, and various robustness mechanisms such ,
as redundant encodings <xref target="RFC2198"/>.</t> payload format configurations, and various robustness mechanisms suc
h
as redundant encodings <xref target="RFC2198" format="default"/>.</t
>
<t>The RTP payload type is scoped by the sending endpoint within an <t>The RTP payload type is scoped by the sending endpoint within an
RTP session. PT has the same meaning across all RTP streams in an RT P RTP session. PT has the same meaning across all RTP streams in an RT P
session. All SSRCs sent from a single endpoint share the same payloa d session. All SSRCs sent from a single endpoint share the same payloa d
type definitions. The RTP payload type is designed such that only a type definitions. The RTP payload type is designed such that only a
single payload type is valid at any time instant in the RTP stream's single payload type is valid at any instant in time in the RTP strea
timestamp time line, effectively time-multiplexing different payload m's
timestamp timeline, effectively time-multiplexing different payload
types if any change occurs. The payload type can change on a types if any change occurs. The payload type can change on a
per-packet basis for an SSRC, for example a speech codec making use of per-packet basis for an SSRC -- for example, a speech codec making u se of
generic comfort noise generic comfort noise
<xref target="RFC3389"/>. If there is a true need to send multiple <xref target="RFC3389" format="default"/>. If there is a true need t o send multiple
payload types for the same SSRC that are valid for the same instant, payload types for the same SSRC that are valid for the same instant,
then redundant encodings then redundant encodings
<xref target="RFC2198"/> <xref target="RFC2198" format="default"/>
can be used. Several additional constraints than the ones mentioned can be used. Several additional constraints, other than those mentio
above need to be met to enable this use, one of which is that the ned
above, need to be met to enable this usage, one of which is that the
combined payload sizes of the different payload types ought not exce ed combined payload sizes of the different payload types ought not exce ed
the transport MTU.</t> the transport MTU.</t>
<t>Other aspects of RTP payload format use are described in How to <t>Other aspects of using the RTP payload format are described in
Write an RTP Payload Format <xref target="RFC8088">"How to Write an RTP Payload Format"</xref>.<
<xref target="RFC8088"/>.</t> /t>
<t>The payload type is not a multiplexing point at the RTP layer (see <t>The payload type is not a multiplexing point at the RTP layer (see
<xref target="section-a"/> <xref target="sect-a" format="default"/>
for a detailed discussion of why using the payload type as an RTP for a detailed discussion of why using the payload type as an RTP
multiplexing point does not work). The RTP payload type is, however, multiplexing point does not work). The RTP payload type is, however,
used to determine how to consume and decode an RTP stream. The RTP used to determine how to consume and decode an RTP stream. The RTP
payload type number is sometimes used to associate an RTP stream wit h payload type number is sometimes used to associate an RTP stream wit h
the signalling, which in general requires that unique RTP payload the signaling, which in general requires that unique RTP payload
type numbers are used in each context. Use of MID, e.g. when bundlin type numbers be used in each context. Using MID, e.g., when bundling
g "m=" sections "m=" sections
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/>, <xref target="RFC8843" format="default"/>,
can replace the payload type as signalling association and unique can replace the payload type as a signaling association, and unique
RTP payload types are then no longer required for that purpose.</t> RTP payload types are then no longer required for that purpose.</t>
</section> </section>
</section> </section>
<section <section anchor="sect-3.3" numbered="true" toc="default">
anchor="section-3.3" <name>Issues Related to RTP Topologies</name>
title="Issues Related to RTP Topologies">
<t>The impact of how RTP multiplexing is performed will in general <t>The impact of how RTP multiplexing is performed will in general
vary with how the RTP session participants are interconnected, vary with how the RTP session participants are interconnected,
described by RTP Topology as described in
<xref target="RFC7667"/>.</t> <xref target="RFC7667">"RTP Topologies"</xref>.</t>
<t>Even the most basic use case, denoted Topo-Point-to-Point in <t>Even the most basic use case -- "Topo-Point-to-Point" as described in
<xref target="RFC7667"/>, raises a number of considerations that are <xref target="RFC7667" format="default"/> -- raises a number of
discussed in detail in following sections. They range over such considerations, which are
aspects as:</t> discussed in detail in the following sections. They range over such
<t> aspects as the following:</t>
<list style="symbols"> <ul spacing="normal">
<t>Does my communication peer support RTP as defined with multiple <li>Does my communication peer support RTP as defined with multiple
SSRCs per RTP session?</t> SSRCs per RTP session?</li>
<t>Do I need network differentiation in form of QoS <li>Do I need network differentiation in the form of QoS
(<xref target="section-4.2.1"/>)?</t> (<xref target="sect-4.2.1" format="default"/>)?</li>
<t>Can the application more easily process and handle the media <li>Can the application more easily process and handle the media
streams if they are in different RTP sessions?</t> streams if they are in different RTP sessions?</li>
<t>Do I need to use additional RTP streams for RTP retransmission or <li>Do I need to use additional RTP streams for RTP retransmission or
FEC?</t> FEC?</li>
</list> </ul>
</t> <t>For some point-to-multipoint topologies (e.g., Topo-ASM and
<t>For some point to multi-point topologies (e.g. Topo-ASM and Topo-SSM
Topo-SSM in <xref target="RFC7667" format="default"/>), multicast is used to inter
<xref target="RFC7667"/>), multicast is used to interconnect the connect the
session participants. Special considerations (documented in session participants. Special considerations (documented in
<xref target="section-4.2.3"/>) are then needed as multicast is a <xref target="sect-4.2.3" format="default"/>) are then needed, as mult icast is a
one-to-many distribution system.</t> one-to-many distribution system.</t>
<t>Sometimes an RTP communication can end up in a situation when the <t>Sometimes, an RTP communication session can end up in a situation whe
communicating peers are not compatible for various reasons:</t> re the
<t> communicating peers are not compatible, for various reasons:</t>
<list style="symbols"> <ul spacing="normal">
<t>No common media codec for a media type thus requiring transcoding <li>No common media codec for a media type, thus requiring transcoding
.</t> .</li>
<t>Different support for multiple RTP streams and RTP sessions.</t> <li>Different support for multiple RTP streams and RTP sessions.</li>
<t>Usage of different media transport protocols, i.e., RTP or other. <li>Usage of different media transport protocols (i.e., one peer
</t> uses RTP, but the other peer uses a different transport protocol).</li
<t>Usage of different transport protocols, e.g., UDP, DCCP, or TCP.< >
/t> <li>Usage of different transport protocols, e.g., UDP, the Datagram
<t>Different security solutions, e.g., IPsec, TLS, DTLS, or SRTP wit Congestion Control Protocol (DCCP), or TCP.</li>
h <li>Different security solutions (e.g., IPsec, TLS, DTLS, or the
different keying mechanisms.</t> Secure Real-time Transport Protocol (SRTP)) with
</list> different keying mechanisms.</li>
</t> </ul>
<t>In many situations this is resolved by the inclusion of a <t>These compatibility issues can often be resolved by the inclusion of
translator between the two peers, as described by Topo-PtP-Translator a
in translator between the two peers -- the Topo-PtP-Translator, as
<xref target="RFC7667"/>. The translator's main purpose is to make the described in
peers look compatible to each other. There can also be other reasons <xref target="RFC7667" format="default"/>. The translator's main purpo
than compatibility to insert a translator in the form of a middlebox se is to make the
or gateway, for example a need to monitor the RTP streams. Beware that peers look compatible to each other. There can also be reasons other
than compatibility for inserting a translator in the form of a middleb
ox
or gateway -- for example, a need to monitor the RTP streams. Beware t
hat
changing the stream transport characteristics in the translator changing the stream transport characteristics in the translator
can require thorough understanding of aspects from congestion control can require a thorough understanding of aspects ranging from congestio
and media adaptation to application-layer semantics.</t> n control
<t>Within the uses enabled by the RTP standard, the point to point and media-level adaptations to application-layer semantics.</t>
topology can contain one or more RTP sessions <t>Within the uses enabled by the RTP standard, the point-to-point
topology can contain one or more RTP sessions
with one or more media sources per session, each having one or more with one or more media sources per session, each having one or more
RTP streams per media source.</t> RTP streams per media source.</t>
</section> </section>
<section <section anchor="sect-3.4" numbered="true" toc="default">
anchor="section-3.4" <name>Issues Related to RTP and RTCP</name>
title="Issues Related to RTP and RTCP Protocol">
<t>Using multiple RTP streams is a well-supported feature of RTP. <t>Using multiple RTP streams is a well-supported feature of RTP.
However, for most implementers or people writing RTP/RTCP applications However, for most implementers or people writing RTP/RTCP applications
or extensions attempting to apply multiple streams, it can be unclear or extensions attempting to apply multiple streams, it can be unclear
when it is most appropriate to add an additional RTP stream in an when it is most appropriate to add an additional RTP stream in an
existing RTP session and when it is better to use multiple RTP existing RTP session and when it is better to use multiple RTP
sessions. This section discusses the various considerations needed.</t sessions. This section discusses the various considerations that
> need to be taken into account.</t>
<section anchor="section-3.4.1" title="The RTP Specification"> <section anchor="sect-3.4.1" numbered="true" toc="default">
<t>RFC 3550 contains some recommendations and a bullet list with 5 <name>The RTP Specification</name>
arguments for different aspects of RTP multiplexing. Please review <t>RFC 3550 contains some
Section 5.2 of <xref target="RFC3550"/>. Five important aspects recommendations and a numbered list (<xref target="RFC3550"
are quoted below.</t> sectionFormat="of" section="5.2"/>) of five arguments regarding differ
<t><list hangIndent="3" style="hanging"> ent
<t hangText="1.">If, say, two audio streams shared the same RTP sessi aspects of RTP multiplexing. Please review <xref target="RFC3550"
on and the same sectionFormat="of" section="5.2"/>. Five important aspects are
quoted below.</t>
<ol spacing="normal" type="1">
<li><blockquote>If, say, two audio streams shared the same RTP sessi
on and the same
SSRC value, and one were to change encodings and thus acquire a SSRC value, and one were to change encodings and thus acquire a
different RTP payload type, there would be no general way of different RTP payload type, there would be no general way of
identifying which stream had changed encodings.</t></list> identifying which stream had changed encodings.</blockquote>
</t> <t>This argument advocates the use of different SSRCs for each individ
<t>The first argument is to use different SSRC for each individual RTP ual RTP
stream, which is fundamental to RTP operation.</t> stream, as this is fundamental to RTP operation.</t></li>
<t><list hangIndent="3" style="hanging"> <li><blockquote>An SSRC is defined to identify a single timing and s
<t hangText="2.">An SSRC is defined to identify a single timing and s equence number
equence number
space. Interleaving multiple payload types would require differe nt space. Interleaving multiple payload types would require differe nt
timing spaces if the media clock rates differ and would require timing spaces if the media clock rates differ and would require
different sequence number spaces to tell which payload type suff ered different sequence number spaces to tell which payload type suff ered
packet loss.</t></list> packet loss.</blockquote>
</t> <t>This argument advocates against demultiplexing RTP
<t>The second argument is advocating against demultiplexing RTP streams within a session based only on their RTP payload type number
streams within a session based only on their RTP payload type number s;
s, it still stands, as can be seen by the extensive list of issues
which still stands as can been seen by the extensive list of issues discussed in <xref target="sect-a"/>.</t></li>
found in Appendix A.</t> <!-- Note: "Section 6.4" is in RFC 3550, so no xref -->
<t><list hangIndent="3" style="hanging"> <li><blockquote>The RTCP sender and receiver reports (see Section 6.
<t hangText="3.">The RTCP sender and receiver reports (see Section 6. 4) can only
4) can only
describe one timing and sequence number space per SSRC and do no t describe one timing and sequence number space per SSRC and do no t
carry a payload type field.</t></list> carry a payload type field.</blockquote>
</t> <t>This argument is yet another argument against payload type
<t>The third argument is yet another argument against payload type multiplexing.</t></li>
multiplexing.</t> <li><blockquote>An RTP mixer would not be able to combine interleave
<t><list hangIndent="3" style="hanging"> d streams of
<t hangText="4.">An RTP mixer would not be able to combine interleave incompatible media into one stream.</blockquote>
d streams of <t>This argument advocates against multiplexing RTP packets that
incompatible media into one stream.</t></list> require different handling into the same session. In most cases,
</t> the RTP mixer must embed application logic
<t>The fourth argument is against multiplexing RTP packets that
require different handling into the same session. In most cases
the RTP mixer must embed application logic
to handle streams; the separation of streams according to to handle streams; the separation of streams according to
stream type is just another piece of application logic, which might or stream type is just another piece of application logic, which might or
might not be appropriate for a particular application. One type of might not be appropriate for a particular application. One type of
application that can mix different media sources blindly is the application that can mix different media sources blindly is the
audio-only telephone bridge, although the ability to do that comes audio-only telephone bridge, although the ability to do that comes
from the well-defined scenario that is aided by use of a single medi a from the well-defined scenario that is aided by the use of a single media
type, even though individual streams may use incompatible codec type s; type, even though individual streams may use incompatible codec type s;
most other types of applications need application-specific logic to most other types of applications need application-specific logic to
perform the mix correctly.</t> perform the mix correctly.</t></li>
<t><list hangIndent="3" style="hanging"> <li><blockquote><t>Carrying multiple media in one RTP session preclu
<t hangText="5.">Carrying multiple media in one RTP session precludes des: the use of
: the use of
different network paths or network resource allocations if different network paths or network resource allocations if
appropriate; reception of a subset of the media if desired, for appropriate; reception of a subset of the media if desired, for
example just audio if video would exceed the available bandwidth ; and example just audio if video would exceed the available bandwidth ; and
receiver implementations that use separate processes for the dif ferent receiver implementations that use separate processes for the dif ferent
media, whereas using separate RTP sessions permits either single - or media, whereas using separate RTP sessions permits either single - or
multiple-process implementations.</t></list></t> multiple-process implementations.</t></blockquote>
<t>The fifth argument discusses network aspects that are described in <t>This argument discusses network aspects that are described in
<xref target="section-4.2"/>. It also goes into aspects of <xref target="sect-4.2" format="default"/>. It also goes into aspect
implementation, like Split Component Terminal (see Section 3.10 of s of
<xref target="RFC7667"/>) endpoints where different processes or implementation, like split component terminals (see
inter-connected devices handle different aspects of the whole <xref target="RFC7667" sectionFormat="of" section="3.10"/>) -- endpo
multi-media session.</t> ints where different processes or
<t>A summary of RFC 3550's view on multiplexing is to use unique SSRC interconnected devices handle different aspects of the whole
s multimedia session.</t></li>
for anything that is its own media/packet stream, and to use </ol>
different RTP sessions for media streams that don't share a media <t>To summarize, RFC 3550's view on multiplexing is to use unique SSRC
type. This document supports the first point; it is very valid. The s
latter needs further discussion, as imposing a single solution on all for anything that is its own media/packet stream and use
usages of RTP is inappropriate. "Multiple Media Types in an RTP different RTP sessions for media streams that don't share a media
Session specification" <xref target="I-D.ietf-avtcore-multi-media-rtp type. This document supports the first point; it is very valid. Th
-session"/> e
updates RFC 3550 to allow multiple media types in a RTP session. latter needs further discussion, as imposing a single solution on al
It also provides a detailed analysis of the potential benefits l
and issues in having usages of RTP is inappropriate. <xref target="RFC8860">"Sending
multiple media types in the same RTP session. Thus, that document pr Multiple Types of Media in a Single RTP Session"</xref>
ovides updates RFC 3550 to allow multiple media types in an RTP session
a wider scope for an RTP session and considers multiple media types and provides a detailed analysis of the potential benefits
in one RTP session as a possible choice for the RTP application and issues related to having
designer.</t> multiple media types in the same RTP session. Thus, <xref target="R
</section> FC8860"/> provides
<section anchor="section-3.4.2" title="Multiple SSRCs in a Session"> a wider scope for an RTP session and considers multiple media types
in one RTP session as a possible choice for the RTP application
designer.</t>
</section>
<section anchor="sect-3.4.2" numbered="true" toc="default">
<name>Multiple SSRCs in a Session</name>
<t>Using multiple SSRCs at one endpoint in an RTP session requires <t>Using multiple SSRCs at one endpoint in an RTP session requires
resolving some unclear aspects of the RTP specification. These could that some unclear aspects of the RTP specification be resolved. These
potentially lead to some interoperability issues as well as some items could potentially lead to some interoperability issues as
potential significant inefficiencies, as further discussed in "RTP well as some potential significant inefficiencies, as further
Considerations for Endpoints Sending Multiple Media Streams" discussed in "Sending Multiple RTP Streams in a Single RTP Session"
<xref target="RFC8108"/>. An RTP application designer should conside <xref target="RFC8108" format="default"/>. An RTP
r application designer should consider these issues and the
these issues and the possible application impact from lack of application's possible impact caused by a lack of appropriate RTP hand
appropriate RTP handling or optimization in the peer endpoints.</t> ling or
optimization in the peer endpoints.</t>
<t>Using multiple RTP sessions can potentially mitigate application <t>Using multiple RTP sessions can potentially mitigate application
issues caused by multiple SSRCs in an RTP session.</t> issues caused by multiple SSRCs in an RTP session.</t>
</section> </section>
<section anchor="section-3.4.3" title="Binding Related Sources"> <section anchor="sect-3.4.3" numbered="true" toc="default">
<name>Binding Related Sources</name>
<t>A common problem in a number of various RTP extensions has been how <t>A common problem in a number of various RTP extensions has been how
to bind related RTP streams together. This issue is common to both to bind related RTP streams together. This issue is common to both
using additional SSRCs and multiple RTP sessions.</t> using additional SSRCs and multiple RTP sessions.</t>
<t>The solutions can be divided into a few groups:</t> <t>The solutions can be divided into a few groups:</t>
<t> <ul spacing="normal">
<list style="symbols"> <li>RTP/RTCP based</li>
<t>RTP/RTCP based</t> <li>Signaling based, e.g., SDP</li>
<t>Signalling based, e.g. SDP</t> <li>Grouping related RTP sessions</li>
<t>Grouping related RTP sessions</t> <li>Grouping SSRCs within an RTP session</li>
<t>Grouping SSRCs within an RTP session</t> </ul>
</list>
</t>
<t>Most solutions are explicit, but some implicit methods have also <t>Most solutions are explicit, but some implicit methods have also
been applied to the problem.</t> been applied to the problem.</t>
<t>The SDP-based signalling solutions are:</t> <t>The SDP-based signaling solutions are:</t>
<t> <dl newline="true" spacing="normal">
<list hangIndent="3" style="hanging"> <dt>SDP media description grouping:</dt>
<t hangText="SDP Media Description Grouping:">The SDP Grouping Fra <dd>The SDP grouping framework <xref target="RFC5888"
mework format="default"/> uses various semantics to group any number of
<xref target="RFC5888"/> media descriptions. SDP media description grouping has primarily
<vspace blankLines="0"/> been used to group RTP sessions,
uses various semantics to group any number of media descriptions but in combination with <xref target="RFC8843" format="default"/>,
. it can also group multiple media descriptions within a single RTP
This has primarily been grouping RTP sessions, but in combinatio session.</dd>
n with <dt>SDP media multiplexing:</dt>
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/> <dd><xref target="RFC8843">"Negotiating Media
it can also group multiple media descriptions within a single RT Multiplexing Using the Session Description Protocol (SDP)"</xref>
P session.</t> uses information taken from both SDP and RTCP to associate RTP streams to SDP me
<t hangText="SDP Media Multiplexing:">Negotiating Media Multiplexi dia
ng Using descriptions. This allows both SDP and RTCP to group RTP streams bel
the Session Description Protocol (SDP) onging to
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/> an SDP media description and group multiple SDP media
<vspace blankLines="0"/> descriptions into a single RTP session.</dd>
uses both SDP and RTCP information to associate RTP streams to S <dt>SDP SSRC grouping:</dt>
DP media <dd><xref target="RFC5576">"Source-Specific Media Attributes in
descriptions. This allows both to group RTP streams belonging to the Session Description Protocol (SDP)"</xref> includes a solution f
an SDP or grouping
media description, and to group multiple SDP media descriptions SSRCs in the same
into a way that the grouping framework groups media descriptions.</dd>
single RTP session.</t> </dl>
<t hangText="SDP SSRC grouping:">Source-Specific Media Attributes
in SDP
<xref target="RFC5576"/>
<vspace blankLines="0"/>
includes a solution for grouping SSRCs the same way as the Group
ing
framework groups Media Descriptions.</t>
</list>
</t>
<t>The above grouping constructs support many use cases. Those solutio ns have <t>The above grouping constructs support many use cases. Those solutio ns have
shortcomings in cases where the session's dynamic properties are suc h shortcomings in cases where the session's dynamic properties are suc h
that it is difficult or a drain on resources to keep the list of rel ated that it is difficult or a drain on resources to keep the list of rel ated
SSRCs up to date.</t> SSRCs up to date.</t>
<t>An RTP/RTCP-based grouping solution is to use the RTCP SDES CNAME t <t>One RTP/RTCP-based grouping solution is to use the RTCP SDES CNAME
o bind to bind
related RTP streams to an endpoint or to a synchronization context. related RTP streams to an endpoint or a synchronization context. For
For applications with a single RTP stream per type (media, source, or
applications with a single RTP stream per type (media, source or redundancy stream), the CNAME is sufficient for that purpose, indepe
redundancy stream), CNAME is sufficient for that purpose independent ndent of whether one or more RTP sessions
ly of whether one or more RTP sessions are used. However, some applications choose not to use a CNAME becau
are used. However, some applications choose not to use CNAME because se of
of
perceived complexity or a desire not to implement RTCP and instead u se perceived complexity or a desire not to implement RTCP and instead u se
the same SSRC value to bind related RTP streams across multiple RTP the same SSRC value to bind related RTP streams across multiple RTP
sessions. RTP Retransmission sessions. RTP retransmission
<xref target="RFC4588"/> <xref target="RFC4588" format="default"/>,
in multiple RTP session mode and Generic FEC when configured to use multiple RTP sessions, and generic FEC
<xref target="RFC5109"/> <xref target="RFC5109" format="default"/>
both use the CNAME method to relate the RTP streams, which may work but might have some both use the CNAME method to relate the RTP streams, which may work but might have some
downsides in RTP sessions with many participating SSRCs. It is not r ecommended to downsides in RTP sessions with many participating SSRCs. It is not r ecommended to
use identical SSRC values across RTP sessions to relate RTP streams; use identical SSRC values across RTP sessions to relate RTP streams;
When an SSRC when an SSRC
collision occurs, this will force change of that SSRC in all RTP collision occurs, this will force a change of that SSRC in all RTP
sessions and thus resynchronize all of them instead of only the sing sessions and will thus resynchronize all of the streams instead of o
le nly the single
media stream having the collision.</t> media stream experiencing the collision.</t>
<t>Another method to implicitly bind SSRCs is used by RTP <t>Another method for implicitly binding SSRCs is used by RTP
Retransmission retransmission
<xref target="RFC4588"/> <xref target="RFC4588" format="default"/>
when using the same RTP session as the source RTP stream for retrans missions. when using the same RTP session as the source RTP stream for retrans missions.
The receiver missing a packet issues an RTP retransmission A receiver that is missing a packet issues an RTP retransmission
request, and then awaits a new SSRC carrying the RTP retransmission request and then awaits a new SSRC carrying the RTP retransmission
payload and where that SSRC is from the same CNAME. This limits a payload, where that SSRC is from the same CNAME. This limits a
requester to having only one outstanding retransmission request on a ny requester to having only one outstanding retransmission request on a ny
new source SSRCs per endpoint.</t> new SSRCs per endpoint.</t>
<t>RTP Payload Format Restrictions <t><xref target="RFC8851">"RTP Payload Format Restrictions"</xref>
<xref target="I-D.ietf-mmusic-rid"/> provides an RTP/RTCP-based mechanism to unambiguously identify the R
provides an RTP/RTCP based mechanism to unambiguously identify the R TP
TP
streams within an RTP session and restrict the streams' payload form at streams within an RTP session and restrict the streams' payload form at
parameters in a codec-agnostic way beyond what is provided with the parameters in a codec-agnostic way beyond what is provided with the
regular payload types. The mapping is done by specifying an "a=rid" regular payload types. The mapping is done by specifying an "a=rid"
value in the SDP offer/answer signalling and having the correspondin value in the SDP offer/answer signaling and having the corresponding
g RtpStreamId value as an SDES item and an RTP header extension
RtpStreamId value as an SDES item and an RTP header extension. The <xref target="RFC8852"/>. The
RID solution also includes a solution for binding redundancy RTP RID solution also includes a solution for binding redundancy RTP
streams to their original source RTP streams, given that those use R streams to their original source RTP streams, given that those
ID streams use RID
identifiers.</t> identifiers. The redundancy stream uses the RepairedRtpStreamId
<t>Experience has found that an explicit binding between the RTP str SDES item and RTP header extension to declare the RtpStreamId
eams, value of the source stream to create the binding.</t>
agnostic of SSRC values, behaves well. That way, solutions using <t>Experience has shown that an explicit binding between the RTP strea
ms,
agnostic of SSRC values, behaves well. That way, solutions using
multiple RTP streams in a single RTP session and in multiple RTP ses sions multiple RTP streams in a single RTP session and in multiple RTP ses sions
will use the same type of binding.</t> will use the same type of binding.</t>
</section> </section>
<section anchor="section-3.4.4" title="Forward Error Correction"> <section anchor="sect-3.4.4" numbered="true" toc="default">
<t>There exist a number of Forward Error Correction (FEC) based <name>Forward Error Correction</name>
schemes for how to mitigate packet loss in the original streams. <t>There exist a number of FEC-based schemes designed to mitigate pack
Most of the FEC schemes protect a single source flow. The et loss in the original streams.
Most of the FEC schemes protect a single source flow. This
protection is achieved by transmitting a certain amount of redundant protection is achieved by transmitting a certain amount of redundant
information that is encoded such that it can repair one or more pack information that is encoded such that it can repair one or more
et instances of packet
losses over the set of packets the redundant information protects. loss over the set of packets the redundant information protects.
This sequence of redundant information needs to be transmitted as This sequence of redundant information needs to be transmitted as
its own media stream, or in some cases, instead of the original medi a its own media stream or, in some cases, instead of the original medi a
stream. Thus, many of these schemes create a need for binding relate d stream. Thus, many of these schemes create a need for binding relate d
flows as discussed above. Looking at the history of these schemes, flows, as discussed above. Looking at the history of these schemes,
there are schemes using multiple SSRCs and schemes using multiple RT P there are schemes using multiple SSRCs and schemes using multiple RT P
sessions, and some schemes that support both modes of operation.</t> sessions, and some schemes that support both modes of operation.</t>
<t>Using multiple RTP sessions supports the case where some set of <t>Using multiple RTP sessions supports the case where some set of
receivers might not be able to utilise the FEC information. By placi receivers might not be able to utilize the FEC information. By placi
ng ng
it in a separate RTP session and if separating RTP sessions on it in a separate RTP session and if separating RTP sessions at the
transport level, FEC can easily be ignored already on the transport transport level, FEC can easily be ignored at the transport level,
level, without considering any RTP-layer information.</t>
without considering any RTP layer information.</t> <t>In usages involving multicast, sending FEC information in a separat
<t>In usages involving multicast, having the FEC information on its e multicast group allows for similar flexibility. This is especially
own multicast group allows for similar flexibility. This is especial
ly
useful when receivers see heterogeneous packet loss rates. A receive r useful when receivers see heterogeneous packet loss rates. A receive r
can decide, based on measurement of experienced packet loss rates, can decide, based on measurement of experienced packet loss rates,
whether to join a multicast group with the suitable FEC data repair whether to join a multicast group with suitable FEC data repair
capabilities.</t> capabilities.</t>
</section> </section>
</section> </section>
</section> </section>
<section <section anchor="sect-4" numbered="true" toc="default">
anchor="section-4" <name>Considerations for RTP Multiplexing</name>
title="Considerations for RTP Multiplexing"> <section anchor="sect-4.1" numbered="true" toc="default">
<section anchor="section-4.1" title="Interworking Considerations"> <name>Interworking Considerations</name>
<t>There are several different kinds of interworking, and this section <t>There are several different kinds of interworking, and this section
discusses two; interworking directly between different applications, a discusses two: interworking directly between different applications an
nd d
interworking of applications through an RTP Translator. The discussion the interworking of applications through an RTP translator. The discus
includes sion includes
the implications of potentially different RTP multiplexing point the implications of potentially different RTP multiplexing point
choices and limitations that have to be considered when working with choices and limitations that have to be considered when working with
some legacy applications.</t> some legacy applications.</t>
<section anchor="section-4.1.1" title="Application Interworking"> <section anchor="sect-4.1.1" numbered="true" toc="default">
<name>Application Interworking</name>
<t>It is not uncommon that applications or services of similar but not <t>It is not uncommon that applications or services of similar but not
identical usage, especially the ones intended for interactive identical usage, especially those intended for interactive
communication, encounter a situation where one want to interconnect communication, encounter a situation where one wants to interconnect
two or more of these applications.</t> two or more of these applications.</t>
<t>In these cases, one ends up in a situation where one might use a <t>In these cases, one ends up in a situation where one might use a
gateway to interconnect applications. This gateway must then either gateway to interconnect applications. This gateway must then either
change the multiplexing structure or adhere to the respective change the multiplexing structure or adhere to the respective
limitations in each application.</t> limitations in each application.</t>
<t>There are two fundamental approaches to building a gateway: using <t>There are two fundamental approaches to building a gateway: using
RTP Translator interworking (RTP bridging), where the gateway acts RTP translator interworking (RTP bridging), where the gateway acts
as an RTP Translator with the two interconnected applications being as an RTP translator with the two interconnected applications being
members of the same RTP session; or using Gateway Interworking with members of the same RTP session; or using gateway interworking
(<xref target="sect-4.1.3"/>) with
RTP termination, where there are independent RTP sessions between RTP termination, where there are independent RTP sessions between
each interconnected application and the gateway.</t> each interconnected application and the gateway.</t>
<t>For interworking to be feasible, any security solution in use needs <t>For interworking to be feasible, any security solution in use needs
to be compatible and capable of exchanging keys with either the peer to be compatible and capable of exchanging keys with either the peer
or the gateway under the used trust model. Secondly, the applications or the gateway under the trust model being used. Secondly, the appli
need to use media streams in a way that makes sense in both applicati cations
ons. need to use media streams in a way that makes sense in both applicat
</t> ions.
</t>
</section> </section>
<section anchor="section-4.1.2" title="RTP Translator Interworking"> <section anchor="sect-4.1.2" numbered="true" toc="default">
<t>From an RTP perspective, the RTP Translator approach could work if <name>RTP Translator Interworking</name>
<t>From an RTP perspective, the RTP translator approach could work if
all the applications are using the same codecs with the same payload all the applications are using the same codecs with the same payload
types, have made the same multiplexing choices, and have the same types, have made the same multiplexing choices, and have the same
capabilities in number of simultaneous RTP streams combined with the capabilities regarding the number of simultaneous RTP streams combin ed with the
same set of RTP/RTCP extensions being supported. Unfortunately, this same set of RTP/RTCP extensions being supported. Unfortunately, this
might not always be true.</t> might not always be true.</t>
<t>When a gateway is implemented via an RTP Translator, an important <t>When a gateway is implemented via an RTP translator, an important
consideration is if the two applications being interconnected need t o consideration is if the two applications being interconnected need t o
use the same approach to multiplexing. If one side is using RTP use the same approach to multiplexing. If one side is using RTP
session multiplexing and the other is using SSRC multiplexing with B UNDLE session multiplexing and the other is using SSRC multiplexing with B UNDLE
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/>, it may be p ossible <xref target="RFC8843" format="default"/>, it may be possible
for the RTP translator to map the RTP streams between both for the RTP translator to map the RTP streams between both
sides using some method, e.g. based on the number and order of SDP " sides using some method, e.g., based on the number and order of SDP
m=" "m="
lines from each side. There are also challenges with lines from each side. There are also challenges related to
SSRC collision handling since, unless SSRC translation is applied on SSRC collision handling, since, unless SSRC translation is applied o
the n the
RTP translator, there may be a collision on the SSRC multiplexing RTP translator, there may be a collision on the SSRC multiplexing
side that the RTP session multiplexing side will not be aware of. side that the RTP session multiplexing side will not be aware of.
Furthermore, if one of the applications is capable of Furthermore, if one of the applications is capable of
working in several modes (such as being able to use additional RTP working in several modes (such as being able to use additional RTP
streams in one RTP session or multiple RTP sessions at will), and th e streams in one RTP session or multiple RTP sessions at will) and the
other one is not, successful interconnection depends on locking the other one is not, successful interconnection depends on locking the
more flexible application into the operating mode where more flexible application into the operating mode where
interconnection can be successful, even if none of the participants are using interconnection can be successful, even if none of the participants are using
the less flexible application when the RTP sessions are being create d.</t> the less flexible application when the RTP sessions are being create d.</t>
</section> </section>
<section anchor="section-4.1.3" title="Gateway Interworking"> <section anchor="sect-4.1.3" numbered="true" toc="default">
<name>Gateway Interworking</name>
<t>When one terminates RTP sessions at the gateway, there are certain <t>When one terminates RTP sessions at the gateway, there are certain
tasks that the gateway has to carry out:</t> tasks that the gateway has to carry out:</t>
<t> <ul spacing="normal">
<list style="symbols"> <li>Generating appropriate RTCP reports for all RTP streams (possibl
<t>Generating appropriate RTCP reports for all RTP streams (possib y
ly based on incoming RTCP reports) originating from SSRCs controlle
based on incoming RTCP reports), originating from SSRCs controll d by
ed by the gateway.</li>
the gateway.</t> <li>Handling SSRC collision resolution in each application's RTP ses
<t>Handling SSRC collision resolution in each application's RTP se sions.</li>
ssions.</t> <li>Signaling, choosing, and policing appropriate bitrates for each
<t>Signalling, choosing, and policing appropriate bit-rates for ea session.</li>
ch </ul>
session.</t>
</list>
</t>
<t>For applications that use any security mechanism, e.g., in the form <t>For applications that use any security mechanism, e.g., in the form
of SRTP, the gateway needs to be able to decrypt and verify source of SRTP, the gateway needs to be able to decrypt and verify source
integrity of the incoming packets, and re-encrypt, integrity protect, integrity of the incoming packets and then re-encrypt, integrity pro
and sign the packets as the peer in the other application's security tect,
context. and sign the packets as the peer in the other application's security
This is necessary even if all that's needed is a simple remapping of context.
SSRC This is necessary even if all that's needed is a simple remapping of
SSRC
numbers. If this is done, the gateway also needs to be a member of t he numbers. If this is done, the gateway also needs to be a member of t he
security contexts of both sides, and thus a trusted entity.</t> security contexts of both sides and thus a trusted entity.</t>
<t>The gateway might also need to apply transcoding (for <t>The gateway might also need to apply transcoding (for
incompatible codec types), media-level adaptations that cannot be incompatible codec types), media-level adaptations that cannot be
solved through media negotiation (such as rescaling for incompatible solved through media negotiation (such as rescaling for incompatible
video size requirements), suppression of content that is known not t o video size requirements), suppression of content that is known not t o
be handled in the destination application, or the addition or remova l be handled in the destination application, or the addition or remova l
of redundancy coding or scalability layers to fit the needs of the of redundancy coding or scalability layers to fit the needs of the
destination domain.</t> destination domain.</t>
<t>From the above, we can see that the gateway needs to have an <t>From the above, we can see that the gateway needs to have an
intimate knowledge of the application requirements; a gateway is by intimate knowledge of the application requirements; a gateway is by
its nature application specific, not a commodity product.</t> its nature application specific and not a commodity product.</t>
<t>These gateways might therefore potentially block <t>These gateways might therefore potentially block
application evolution by blocking RTP and RTCP extensions that the application evolution by blocking RTP and RTCP extensions that the
applications have been extended with but that are unknown to the applications have been extended with but that are unknown to the
gateway.</t> gateway.</t>
<t>If one uses security mechanism, like SRTP, the gateway and the <t>If one uses a security mechanism like SRTP, the gateway and the
necessary trust in it by the peers is an additional risk to the necessary trust in it by the peers pose an additional risk to
communication security. The gateway also incur additional complexitie communication security. The gateway also incurs additional
s in complexities in the
form of the decrypt-encrypt cycles needed for each forwarded packet. form of the decrypt-encrypt cycles needed for each forwarded packet.
SRTP, due to its keying structure, also requires that each RTP sessi on SRTP, due to its keying structure, also requires that each RTP sessi on
needs different master keys, as use of the same key in two RTP need different master keys, as the use of the same key in two RTP
sessions can for some ciphers result in a reuse of a one-time pad th sessions can, for some ciphers, result in a reuse of a one-time pad
at that
completely breaks the confidentiality of the packets.</t> completely breaks the confidentiality of the packets.</t>
</section> </section>
<section <section anchor="sect-4.1.4" numbered="true" toc="default">
anchor="section-4.1.4" <name>Legacy Considerations for Multiple SSRCs</name>
title="Multiple SSRC Legacy Considerations">
<t>Historically, the most common RTP use cases have been point-to-poin t <t>Historically, the most common RTP use cases have been point-to-poin t
Voice over IP (VoIP) or streaming applications, commonly with no Voice over IP (VoIP) or streaming applications, commonly with no
more than one media source per endpoint and media type (typically more than one media source per endpoint and media type (typically
audio or video). Even in conferencing applications, especially audio or video). Even in conferencing applications, especially
voice-only, the conference focus or bridge has provided a single str voice-only, the conference focus or bridge provides to each particip
eam ant a single stream
to each participant containing a mix of the other participants. It i containing a mix of the other participants. It is
s
also common to have individual RTP sessions between each endpoint an d also common to have individual RTP sessions between each endpoint an d
the RTP mixer, meaning that the mixer functions as an RTP-terminatin g the RTP mixer, meaning that the mixer functions as an RTP-terminatin g
gateway.</t> gateway.</t>
<t>Applications and systems that aren't updated to handle multiple str eams following <t>Applications and systems that aren't updated to handle multiple str eams following
these recommendations can have issues with participating in RTP these recommendations can have issues with participating in RTP
sessions containing multiple SSRCs within a single session, such as: </t> sessions containing multiple SSRCs within a single session, such as: </t>
<t> <ol spacing="normal" type="1">
<list style="numbers"> <li>The need to handle more than one stream simultaneously rather th
<t>Need to handle more than one stream simultaneously rather than an
replacing an already existing stream with a new one.</t> replacing an already-existing stream with a new one.</li>
<t>Be capable of decoding multiple streams simultaneously.</t> <li>Being capable of decoding multiple streams simultaneously.</li>
<t>Be capable of rendering multiple streams simultaneously.</t> <li>Being capable of rendering multiple streams simultaneously.</li>
</list> </ol>
</t>
<t>This indicates that gateways attempting to interconnect to this <t>This indicates that gateways attempting to interconnect to this
class of devices have to make sure that only one RTP stream of each class of devices have to make sure that only one RTP stream of each
media type gets delivered to the endpoint if it's expecting only one , and media type gets delivered to the endpoint if it's expecting only one and
that the multiplexing format is what the device expects. It is highl y that the multiplexing format is what the device expects. It is highl y
unlikely that RTP translator-based interworking can be made to unlikely that RTP translator-based interworking can be made to
function successfully in such a context.</t> function successfully in such a context.</t>
</section> </section>
</section> </section>
<section anchor="section-4.2" title="Network Considerations"> <section anchor="sect-4.2" numbered="true" toc="default">
<name>Network Considerations</name>
<t>The RTP implementer needs to consider that the RTP multiplexing choic e <t>The RTP implementer needs to consider that the RTP multiplexing choic e
also impacts network level mechanisms.</t> also impacts network-level mechanisms.</t>
<section anchor="section-4.2.1" title="Quality of Service"> <section anchor="sect-4.2.1" numbered="true" toc="default">
<t>Quality of Service mechanisms are either flow based or packet marki <name>Quality of Service</name>
ng <t>QoS mechanisms are either flow based or packet marking
based. RSVP based. RSVP
<xref target="RFC2205"/> <xref target="RFC2205" format="default"/>
is an example of a flow based mechanism, while Diff-Serv is an example of a flow-based mechanism, while Diffserv
<xref target="RFC2474"/> <xref target="RFC2474" format="default"/>
is an example of a packet marking based one.</t> is an example of a packet-marking-based mechanism.</t>
<t>For a flow based scheme, additional SSRC will receive the <t>For a flow-based scheme, additional SSRCs will receive the
same QoS as all other RTP streams being part of the same 5-tuple same QoS as all other RTP streams being part of the same 5-tuple
(protocol, source address, destination address, source port, (protocol, source address, destination address, source port,
destination port), which is the most common selector for flow based destination port), which is the most common selector for flow-based
QoS.</t> QoS.</t>
<t>For a packet marking based scheme, the method of multiplexing will <t>For a packet-marking-based scheme, the method of multiplexing will
not affect the possibility to use QoS. Different not affect the possibility of using QoS. Different
Differentiated Services Code Points (DSCP) can be assigned to Differentiated Services Code Points (DSCPs) can be assigned to
different packets within a transport flow (5-Tuple) as well as withi different packets within a transport flow (5-tuple) as well as withi
n an RTP stream, n an RTP stream,
assuming usage of UDP or other transport protocol that do not have is assuming the usage of UDP or other transport protocols that do not h
sues ave issues
with packet reordering within the transport flow (5-tuple). with packet reordering within the transport flow (5-tuple).
To avoid packet reording issues, packets belonging to the same RTP To avoid packet-reordering issues, packets belonging to the same RTP
flow should limits its use of DSCP to those whose corresponding flow should limit their use of DSCPs to packets whose corresponding
Per Hop Behavior (PHB) that do not enable reordering. Per-Hop Behavior (PHB) do not enable reordering. If the transport pr
If the transport protocol used assumes in order delivery of packet, otocol being used assumes in&nbhy;order
such as TCP and SCTP, then a single DSCP should be used. delivery of packets (e.g., TCP and the Stream Control Transmission
For more discussion of this see <xref target="RFC7657"/>.</t> Protocol (SCTP)),
then a single DSCP should be used.
For more discussion on this topic, see <xref target="RFC7657" format
="default"/>.</t>
<t>The method for assigning marking to packets can impact what number <t>The method for assigning marking to packets can impact what number
of RTP sessions to choose. If this marking is done using a network of RTP sessions to choose. If this marking is done using a network
ingress function, it can have issues discriminating the different RT P ingress function, it can have issues discriminating the different RT P
streams. The network API on the endpoint also needs to be capable of streams. The network API on the endpoint also needs to be capable of
setting the marking on a per-packet basis to reach the full setting the marking on a per-packet basis to reach full
functionality.</t> functionality.</t>
</section> </section>
<section anchor="section-4.2.2" title="NAT and Firewall Traversal"> <section anchor="sect-4.2.2" numbered="true" toc="default">
<t>In today's networks there exist a large number of middleboxes. The <name>NAT and Firewall Traversal</name>
ones that normally have most impact on RTP are Network Address <t>In today's networks, there exist a large number of middleboxes. Tho
Translators (NAT) and Firewalls (FW).</t> se
<t>Below we analyse and comment on the impact of requiring more that normally have the most impact on RTP are Network Address
underlying transport flows in the presence of NATs and Firewalls:</t Translators (NATs) and Firewalls (FWs).</t>
> <t>Below, we analyze and comment on the impact of requiring more
<t> underlying transport flows in the presence of NATs and FWs:</t>
<list hangIndent="3" style="hanging"> <dl newline="true" spacing="normal">
<t hangText="End-Point Port Consumption:">A given IP address only <dt>Endpoint Port Consumption:</dt>
has 65536 <dd>A given IP address only has 65536
<vspace blankLines="0"/>
available local ports per transport protocol for all consumers o f available local ports per transport protocol for all consumers o f
ports that exist on the machine. This is normally never an issue for ports that exist on the machine. This is normally never an issue for
an end-user machine. It can become an issue for servers that han an end-user machine. It can become an issue for servers that
dle handle a
large number of simultaneous streams. However, if the applicatio n uses large number of simultaneous streams. However, if the applicatio n uses
ICE to authenticate STUN requests, a server can serve multiple ICE to authenticate STUN requests, a server can serve multiple
endpoints from the same local port, and use the whole 5-tuple (s ource endpoints from the same local port and use the whole 5-tuple (so urce
and destination address, source and destination port, protocol) as and destination address, source and destination port, protocol) as
identifier of flows after having securely bound them to the remo te the identifier of flows after having securely bound them to the remote
endpoint address using the STUN request. In theory, the minimum number endpoint address using the STUN request. In theory, the minimum number
of media server ports needed are the maximum number of simultane of media server ports needed is the maximum number of simultaneo
ous us
RTP sessions a single endpoint can use. In practice, implementat RTP sessions a single endpoint can use. In practice, implementat
ion ions
will probably benefit from using more server ports to simplify will probably benefit from using more server ports to simplify
implementation or avoid performance bottlenecks.</t> implementation or avoid performance bottlenecks.</dd>
<t hangText="NAT State:">If an endpoint sits behind a NAT, each fl <dt>NAT State:</dt>
ow it generates <dd>If an endpoint sits behind a NAT, each flow it generates
<vspace blankLines="0"/>
to an external address will result in a state that has to be kep t in to an external address will result in a state that has to be kep t in
the NAT. That state is a limited resource. In home or Small the NAT. That state is a limited resource. In home or Small
Office/Home Office (SOHO) NATs, memory or processing are usually Office&wj;/Home Office (SOHO) NATs, the most limited resource is
the memory or processing. For large-scale NATs serving many internal
most limited resources. For large scale NATs serving many intern
al
endpoints, available external ports are likely the scarce resour ce. endpoints, available external ports are likely the scarce resour ce.
Port limitations is primarily a problem for larger centralised N Port limitations are primarily a problem for larger centralized
ATs NATs
where endpoint independent mapping requires each flow to use one where endpoint-independent mapping requires each flow to use one
port port
for the external IP address. This affects the maximum number of for the external IP address. This affects the maximum number of
internal users per external IP address. However, as a comparison , a internal users per external IP address. However, as a comparison , a
real-time video conference session with audio and video likely u ses real-time video conference session with audio and video likely u ses
less than 10 UDP flows, compared to certain web applications tha t can less than 10 UDP flows, compared to certain web applications tha t can
use 100+ TCP flows to various servers from a single browser inst use 100+ TCP flows to various servers from a single browser
ance.</t> instance.</dd>
<t hangText="NAT Traversal Extra Delay:">Performing the NAT/FW tra <dt>Extra Delay Added by NAT Traversal:</dt>
versal takes a <dd>Performing the NAT/FW traversal takes a
<vspace blankLines="0"/> certain amount of time for each flow. The best-case scenario for
certain amount of time for each flow. It also takes time in a ph
ase of
communication between accepting to communicate and the media pat
h
being established, which is fairly critical. The best case scena
rio for
additional NAT/FW traversal time after finding the first valid c andidate additional NAT/FW traversal time after finding the first valid c andidate
pair following the specified ICE procedures is 1.5*RTT + pair following the specified ICE procedures is 1.5*RTT +
Ta*(Additional_Flows-1), where Ta is the pacing timer. That assu mes a Ta*(Additional_Flows-1), where Ta is the pacing timer. That assu mes a
message in one direction, immediately followed by a check back. message in one direction, immediately followed by a
The reason it isn't more, is that ICE first finds one candidate return message in the opposite direction to confirm reachability.
pair It isn't more, because ICE first finds one candidate pair
that works prior to attempting to establish multiple flows. Thus that works, prior to attempting to establish multiple flows. Thu
, s,
there is no extra time until one has found a working candidate p air. there is no extra time until one has found a working candidate p air.
Based on that working pair, the extra time is needed to in paral Based on that working pair, the extra time is needed to
lel establish the additional flows (two or three, in most cases)
establish the, in most cases 2-3, additional flows. However, pac in parallel. However, packet
ket loss causes extra delays of at least 500 ms (the minimal
loss causes extra delays, at least 500 ms, which is the minimal retransmission timer for ICE).</dd>
retransmission timer for ICE.</t> <dt>NAT Traversal Failure Rate:</dt>
<t hangText="NAT Traversal Failure Rate:">Due to the need to estab <dd>Due to the need to establish more than a
lish more than a
<vspace blankLines="0"/>
single flow through the NAT, there is some risk that establishin g the single flow through the NAT, there is some risk that establishin g the
first flow succeeds but that one or more of the additional flows first flow will succeed but one or more of the additional
fail. flows will fail.
The risk that this happens is hard to quantify, but ought to be The risk of this happening is hard to quantify but should be fai
fairly rly
low as one flow from the same interfaces has just been successfu low, as one flow from the same interfaces has just been successf
lly ully
established. Thus only rare events such as NAT resource overload established. Thus, only such rare events as NAT resource overloa
, or d,
selecting particular port numbers that are filtered etc., ought selecting particular port numbers that are filtered, etc., ought
to be to be
reasons for failure.</t> reasons for failure.</dd>
<t hangText="Deep Packet Inspection and Multiple Streams:">Firewal <dt>Deep Packet Inspection and Multiple Streams:</dt>
ls differ in how <dd>FWs differ in how
<vspace blankLines="0"/> deeply they inspect packets.
deeply they inspect packets. Due to all previous issues with fir Previous experience using FWs and Session Border Gateways
ewall and (SBGs) with RTP shows that there is a significant risk that
Session Boarder Gateways (SBG) with RTP transport media e.g. in V the FWs and SBGs will reject RTP sessions that use multiple SSRC
oice over s.</dd>
IP (VoIP) systems, there exists a significant risk that deeply </dl>
inspecting firewalls will have similar legacy issues with multip
le
SSRCs as some RTP stack implementations.</t>
</list>
</t>
<t>Using additional RTP streams in the same RTP session and transport <t>Using additional RTP streams in the same RTP session and transport
flow does not introduce any additional NAT traversal complexities pe r flow does not introduce any additional NAT traversal complexities pe r
RTP stream. This can be compared with normally one or two additional RTP stream. This can be compared with (normally) one or two addition al
transport flows per RTP session when using multiple RTP sessions. transport flows per RTP session when using multiple RTP sessions.
Additional lower layer transport flows will be needed, unless an Additional lower-layer transport flows will be needed, unless an
explicit de-multiplexing layer is added between RTP and the transpor explicit demultiplexing layer is added between RTP and the transport
t protocol. At the time of this writing, no such mechanism was defined
protocol. At time of writing no such mechanism was defined.</t> .</t>
</section> </section>
<section anchor="section-4.2.3" title="Multicast"> <section anchor="sect-4.2.3" numbered="true" toc="default">
<t>Multicast groups provides a powerful tool for a number of real-time <name>Multicast</name>
applications, especially the ones that desire broadcast-like <t>Multicast groups provide a powerful tool for a number of real-time
behaviours with one endpoint transmitting to a large number of applications, especially those that desire broadcast-like
receivers, like in IPTV. There is also the RTP/RTCP extension to behaviors with one endpoint transmitting to a large number of
better support Source Specific Multicast (SSM) receivers, like in IPTV. An RTP/RTCP extension to
<xref target="RFC5760"/>. Many-to-many communication, which RTP better support Source-Specific Multicast (SSM)
<xref target="RFC3550"/> <xref target="RFC5760" format="default"/> is also available. Many-to
-many communication, which RTP
<xref target="RFC3550" format="default"/>
was originally built to support, has several limitations in common w ith was originally built to support, has several limitations in common w ith
multicast.</t> multicast.</t>
<t>One limitation is that, for any group, sender side adaptation with the <t>One limitation is that, for any group, sender-side adaptations with the
intent to suit all receivers would have to adapt to the most limited intent to suit all receivers would have to adapt to the most limited
receiver experiencing the worst conditions among the group participa nts, receiver experiencing the worst conditions among the group participa nts,
which imposes degradation for all participants. For broadcast-type which imposes degradation for all participants. For broadcast-type
applications with a large number of receivers, this is not applications with a large number of receivers, this is not
acceptable. Instead, various receiver-based solutions are employed t o acceptable. Instead, various receiver-based solutions are employed t o
ensure that the receivers achieve best possible performance. By usin g ensure that the receivers achieve the best possible performance. By using
scalable encoding and placing each scalability layer in a different scalable encoding and placing each scalability layer in a different
multicast group, the receiver can control the amount of traffic it multicast group, the receiver can control the amount of traffic it
receives. To have each scalability layer on a different multicast receives. To have each scalability layer in a different multicast
group, one RTP session per multicast group is used.</t> group, one RTP session per multicast group is used.</t>
<t>In addition, the transport flow considerations in multicast are a <t>In addition, the transport flow considerations in multicast are a
bit different from unicast; NATs with port translation are not usefu l bit different from unicast; NATs with port translation are not usefu l
in the multicast environment, meaning that the entire port range of in the multicast environment, meaning that the entire port range of
each multicast address is available for distinguishing between RTP each multicast address is available for distinguishing between RTP
sessions.</t> sessions.</t>
<t>Thus, when using broadcast applications it appears easiest and most <t>Thus, when using broadcast applications it appears easiest and most
straightforward to use multiple RTP sessions for sending different straightforward to use multiple RTP sessions for sending different
media flows used for adapting to network conditions. It is also comm on media flows used for adapting to network conditions. It is also comm on
that streams improving transport robustness are sent in their own that streams improving transport robustness are sent in their own
multicast group to allow for interworking with legacy or to support multicast group to allow for interworking with legacy applications o r to support
different levels of protection.</t> different levels of protection.</t>
<t>Many-to-many applications have different needs and the most <t>Many-to-many applications have different needs, and the most
appropriate multiplexing choice will depend on how the actual applic ation is appropriate multiplexing choice will depend on how the actual applic ation is
realized. Multicast applications that are capable of using sender si de realized. Multicast applications that are capable of using sender-si de
congestion control can avoid the use of multiple multicast sessions and RTP congestion control can avoid the use of multiple multicast sessions and RTP
sessions that result from use of receiver side congestion control.</ sessions that result from the use of receiver-side congestion contro
t> l.</t>
<t>The properties of a broadcast application using RTP multicast:</t> <t>The properties of a broadcast application using RTP multicast are
<t> as follows:</t>
<list style="numbers"> <ol spacing="normal" type="1">
<t>Uses a group of RTP sessions, not just one. Each endpoint will <li>The application uses a group of RTP sessions -- not just one. Ea
need to ch endpoint will need to
be a member of a number of RTP sessions in order to perform well be a member of a number of RTP sessions in order to perform well
.</t> .</li>
<t>Within each RTP session, the number of RTP receivers is likely <li>Within each RTP session, the number of RTP receivers is likely t
to o
be much larger than the number of RTP senders.</t> be much larger than the number of RTP senders.</li>
<t>The applications need signalling functions to identify the <li>The application needs signaling functions to identify the
relationships between RTP sessions.</t> relationships between RTP sessions.</li>
<t>The applications need signalling or RTP/RTCP functions to ident <li>The application needs signaling or RTP/RTCP functions to identif
ify y
the relationships between SSRCs in different RTP sessions when n the relationships between SSRCs in different RTP sessions when
eeds more complex relations than those that can be expressed by the CNAME exist.</li>
beyond CNAME exist.</t> </ol>
</list>
</t>
<t>Both broadcast and many-to-many multicast applications share a <t>Both broadcast and many-to-many multicast applications share a
signalling requirement; all of the participants need the signaling requirement; all of the participants need the
same RTP and payload type configuration. Otherwise, A could for same RTP and payload type configuration. Otherwise, A could, for
example be using payload type 97 as the video codec H.264 while B example, be using payload type 97 as the video codec H.264 while B
thinks it is MPEG-2. SDP offer/answer thinks it is MPEG-2. SDP offer/answer
<xref target="RFC3264"/> <xref target="RFC3264" format="default"/>
is not appropriate for ensuring this property in broadcast/multicast is not appropriate for ensuring this property in a broadcast/multica
context. The signalling aspects of broadcast/multicast are not st
context. The signaling aspects of broadcast/multicast are not
explored further in this memo.</t> explored further in this memo.</t>
<t>Security solutions for this type of group communication are also <t>Security solutions for this type of group communication are also
challenging. First, the key-management and the security protocol nee d challenging. First, the key-management mechanism and the security pr otocol need
to support group communication. Second, source authentication requir es to support group communication. Second, source authentication requir es
special solutions. For more discussion on this please review Options special solutions. For more discussion on this topic, please review
for Securing RTP Sessions <xref target="RFC7201">"Options for Securing RTP Sessions"</xref>.</t>
<xref target="RFC7201"/>.</t>
</section> </section>
</section> </section>
<section <section anchor="sect-4.3" numbered="true" toc="default">
anchor="section-4.3" <name>Security and Key-Management Considerations</name>
title="Security and Key Management Considerations"> <t>When dealing with point-to-point two-member RTP sessions only, there
<t>When dealing with point-to-point, 2-member RTP sessions only, there
are few security issues that are relevant to the choice of having one are few security issues that are relevant to the choice of having one
RTP session or multiple RTP sessions. However, there are a few aspects RTP session or multiple RTP sessions. However, there are a few aspects
of multiparty sessions that might warrant consideration. For general of multi-party sessions that might warrant consideration. For general
information of possible methods of securing RTP, please review RTP information regarding possible methods of securing RTP, please review
Security Options
<xref target="RFC7201"/>.</t> <xref target="RFC7201"/>.</t>
<section anchor="section-4.3.1" title="Security Context Scope"> <section anchor="sect-4.3.1" numbered="true" toc="default">
<name>Security Context Scope</name>
<t>When using SRTP <t>When using SRTP
<xref target="RFC3711"/>, <xref target="RFC3711" format="default"/>,
the security context scope is important and can be a necessary the security context scope is important and can be a necessary
differentiation in some applications. As SRTP's crypto suites are (s o differentiation in some applications. As SRTP's crypto suites are (s o
far) built around symmetric keys, the receiver will need to have the far) built around symmetric keys, the receiver will need to have the
same key as the sender. This results in that no one in a multi-party same key as the sender. As a result, no one in a multi-party
session can be certain that a received packet really was sent by the session can be certain that a received packet was really sent by the
claimed sender and not by another party having access to the key. Th e claimed sender and not by another party having access to the key. Th e
single SRTP algorithm not having this propery is the TESLA source single SRTP algorithm not having this property is Timed
authentication <xref target="RFC4383"/>. However, TESLA adds delay Efficient Stream Loss-Tolerant Authentication (TESLA) source
to achieve source authentication. In most cases, symmetric ciphers authentication <xref target="RFC4383" format="default"/>. However, T
provide sufficient security properties but create issues in a few cas ESLA adds delay
es.</t> to achieve source authentication. In most cases, symmetric ciphers
provide sufficient security properties, but in a few cases they can
create issues.</t>
<t>The first case is when someone leaves a multi-party session and one <t>The first case is when someone leaves a multi-party session and one
wants to ensure that the party that left can no longer access the RT P wants to ensure that the party that left can no longer access the RT P
streams. This requires that everyone re-keys without disclosing the streams. This requires that everyone rekey without disclosing the
new keys to the excluded party.</t> new keys to the excluded party.</t>
<t>A second case is when using security as an enforcing mechanism for <t>A second case is when security is used as an enforcing mechanism fo
stream access differentiation between different receivers. Take for r
example a scalable layer or a high quality simulcast version that on stream access differentiation between different receivers. Take, for
ly example, a scalable layer or a high-quality simulcast version that o
nly
users paying a premium are allowed to access. The mechanism preventi ng a receiver users paying a premium are allowed to access. The mechanism preventi ng a receiver
from getting the high quality stream can be based on the stream bein from getting the high-quality stream can be based on the stream bein
g g
encrypted with a key that user can't access without paying premium, encrypted with a key that users can't access without paying a premiu
using the key-management to limit access to the key.</t> m,
<t>SRTP <xref target="RFC3711"/> as specified uses per SSRC unique k using the key-management mechanism to limit access to the key.</t>
eys, <t>As specified in <xref target="RFC3711" format="default"/>, SRTP use
however the original assumption was a single session master key from s
which SSRC specific RTP and RTCP keys where derived. However, that unique keys per SSRC;
assumption was proven incorrect, as the application usage and however, the original assumption was a single-session master key fro
the developed key-mamangement mechanisms have chosen many different m
methods for ensuring SSRC unique keys. The key-management functions h which SSRC-specific RTP and RTCP keys were derived. However, that
ave different assumption was proven incorrect, as the application usage and
capabilities to establish different sets of keys, normally on a the developed key-management mechanisms have chosen many different
methods for ensuring unique keys per SSRC. The key-management functi
ons have different
abilities to establish different sets of keys, normally on a
per-endpoint basis. For example, DTLS-SRTP per-endpoint basis. For example, DTLS-SRTP
<xref target="RFC5764"/> <xref target="RFC5764" format="default"/>
and Security Descriptions and Security Descriptions
<xref target="RFC4568"/> <xref target="RFC4568" format="default"/>
establish different keys for outgoing and incoming traffic from an establish different keys for outgoing and incoming traffic from an
endpoint. This key usage has to be written into the cryptographic endpoint. This key usage has to be written into the cryptographic
context, possibly associated with different SSRCs. Thus, limitations context, possibly associated with different SSRCs. Thus, limitations
do exist depending on chosen key-management method and due to integra do exist, depending on the chosen key-management method and due to
tion the integration
of particular implementations of the key-management and SRTP.</t> of particular implementations of the key-management method and SRTP.
</t>
</section> </section>
<section <section anchor="sect-4.3.2" numbered="true" toc="default">
anchor="section-4.3.2" <name>Key Management for Multi-party Sessions</name>
title="Key Management for Multi-party Sessions"> <t>The capabilities of the key-management method combined with the RTP
<t>The capabilities of the key-management combined with the RTP multip multiplexing
lexing choices affect the resulting security properties, control over the
choices affects the resulting security properties, control over the secured media, and who has access to it.</t>
secured media, and who have access to it.</t>
<t>Multi-party sessions contain at least one RTP stream from each acti ve <t>Multi-party sessions contain at least one RTP stream from each acti ve
participant. Depending on the multi-party topology participant. Depending on the multi-party topology
<xref target="RFC7667"/>, <xref target="RFC7667" format="default"/>,
each participant can both send and receive multiple RTP streams. each participant can both send and receive multiple RTP streams.
Transport translator-based sessions (Topo-Trn-Translator) and multic ast Transport translator-based sessions (Topo-Trn-Translator) and multic ast
sessions (Topo-ASM), can neither use Security Description sessions (Topo-ASM) can use neither Security Descriptions
<xref target="RFC4568"/> <xref target="RFC4568" format="default"/>
nor DTLS-SRTP nor DTLS-SRTP
<xref target="RFC5764"/> <xref target="RFC5764" format="default"/>
without an extension as each endpoint provides its set of keys. In without an extension, because each endpoint provides its own set of
centralised conferences, the signalling counterpart is a conference keys. In
server, and the transport translator is the media plane unicast centralized conferences, the signaling counterpart is a conference
counterpart (to which DTLS messages would be sent). Thus, an extensio server, and the transport translator is the media-plane unicast
n counterpart (to which DTLS messages would be sent). Thus, an extensi
like Encrypted Key Transport <xref target="I-D.ietf-perc-srtp-ekt-die on
t"/> like Encrypted Key Transport <xref target="RFC8870" format="default"
or a MIKEY <xref target="RFC3830"/> based solution that allows for />
keying all session participants with the same master key is needed.</ or a solution based on Multimedia Internet KEYing (MIKEY) <xref targ
t> et="RFC3830" format="default"/> that allows for
<t>Privacy Enchanced RTP Conferencing (PERC) also enables a different keying all session participants with the same master key is needed.<
trust model with semi-trusted media switching RTP middleboxes /t>
<xref target="I-D.ietf-perc-private-media-framework"/>.</t> <t>Privacy-Enhanced RTP Conferencing (PERC) also enables a different
trust model with semi-trusted media-switching RTP middleboxes
<xref target="RFC8871" format="default"/>.</t>
</section> </section>
<section anchor="section-4.3.3" title="Complexity Implications"> <section anchor="sect-4.3.3" numbered="true" toc="default">
<t>The usage of security functions can surface complexity implications <name>Complexity Implications</name>
from the choice of multiplexing and topology. This becomes especiall <t>There can be complex interactions between the choice of
y multiplexing and topology and the security functions. This becomes esp
ecially
evident in RTP topologies having any type of middlebox that processe s evident in RTP topologies having any type of middlebox that processe s
or modifies RTP/RTCP packets. While there is very small overhead for or modifies RTP/RTCP packets. While the overhead of
an RTP translator or mixer to rewrite an SSRC value in the RTP packe an RTP translator or mixer rewriting an SSRC value in the RTP packet
t of an unencrypted session is low, the cost is higher when using cryp
of an unencrypted session, the cost is higher when using cryptograph tographic
ic
security functions. For example, if using SRTP security functions. For example, if using SRTP
<xref target="RFC3711"/>, the actual security context and exact cryp <xref target="RFC3711" format="default"/>, the actual security conte
to xt and exact crypto
key are determined by the SSRC field value. If one changes SSRC, the key are determined by the SSRC field value. If one changes the
SSRC value, the
encryption and authentication must use another key. Thus, changing t he encryption and authentication must use another key. Thus, changing t he
SSRC value implies a decryption using the old SSRC and its security SSRC value implies a decryption using the old SSRC and its security
context, followed by an encryption using the new one.</t> context, followed by an encryption using the new one.</t>
</section> </section>
</section> </section>
</section> </section>
<section anchor="section-5" title="RTP Multiplexing Design Choices"> <section anchor="sect-5" numbered="true" toc="default">
<name>RTP Multiplexing Design Choices</name>
<t>This section discusses how some RTP multiplexing design choices can <t>This section discusses how some RTP multiplexing design choices can
be used in applications to achieve certain goals, and a summary of the be used in applications to achieve certain goals and summarizes the
implications of such choices. For each design there is discussion of implications of such choices. The benefits and downsides of each
benefits and downsides.</t> design are also discussed.</t>
<section <section anchor="sect-5.1" numbered="true" toc="default">
anchor="section-5.1" <name>Multiple Media Types in One Session</name>
title="Multiple Media Types in One Session">
<t>This design uses a single RTP session for multiple different media <t>This design uses a single RTP session for multiple different media
types, like audio and video, and possibly also transport robustness types, like audio and video, and possibly also transport robustness
mechanisms like FEC or retransmission. An endpoint can send zero, one mechanisms like FEC or retransmission. An endpoint can send zero,
or more media sources per media type, resulting in a number of RTP one, or multiple media sources per media type, resulting in a number o
f RTP
streams of various media types for both source and redundancy streams. </t> streams of various media types for both source and redundancy streams. </t>
<t>The Advantages:</t> <t>Advantages:</t>
<t> <ol spacing="normal" type="1">
<list style="numbers"> <li>
<t>Only a single RTP session is used, which implies:<list style="sym <t>Only a single RTP session is used, which implies:</t>
bols"> <ul spacing="normal">
<t>Minimal need to keep NAT/FW state.</t> <li>Minimal need to keep NAT/FW state.</li>
<t>Minimal NAT/FW-traversal cost.</t> <li>Minimal NAT/FW traversal cost.</li>
<t>Fate-sharing for all media flows.</t> <li>Fate-sharing for all media flows.</li>
<t>Minimal overhead for security association establishment.</t> <li>Minimal overhead for security association establishment.</li>
</list> </ul>
</t> </li>
<t>Dynamic allocation of RTP streams can be handled almost entirely <li>Dynamic allocation of RTP streams can be handled almost entirely
at RTP level. at the RTP level.
How localized this can be kept to RTP level depends on the applica The extent to which this allocation can be kept at the RTP level depen
tion's needs ds on the application's needs
for explicit indication of the stream usage and how timely that ca for an explicit indication of stream usage and in how timely a
n be signalled.</t> fashion that information can be signaled.</li>
</list> </ol>
</t> <t>Disadvantages:</t>
<t>The Disadvantages:</t> <ol spacing="normal" type="1">
<t> <li>It is less suitable for interworking with other applications that
<list style="letters"> use
<t>It is less suitable for interworking with other applications that
use
individual RTP sessions per media type or multiple sessions for a individual RTP sessions per media type or multiple sessions for a
single media type, due to the risk of SSRC collision and thus pote single media type, due to the risk of SSRC collisions and thus a p
ntial otential
need for SSRC translation.</t> need for SSRC translation.</li>
<t>Negotiation of individual bandwidths for the different media type <li>Negotiation of individual bandwidths for the different media types
s is is
currently only possible in SDP when using RID currently only possible in SDP when using RID
<xref target="I-D.ietf-mmusic-rid"/>.</t> <xref target="RFC8851" format="default"/>.</li>
<t>It is not suitable for Split Component Terminal (see Section 3.10 <li>It is not suitable for split component terminals (see
of <xref target="RFC7667" sectionFormat="of" section="3.10"/>).</li>
<xref target="RFC7667"/>).</t> <li>Flow-based QoS cannot be used to provide separate treatment of RTP
<t>Flow-based QoS cannot be used to provide separate treatment of RT streams compared to others in the single RTP session.</li>
P <li>If there is significant asymmetry between the RTP streams' RTCP
streams compared to others in the single RTP session.</t> reporting needs, there are some challenges related to configuratio
<t>If there is significant asymmetry between the RTP streams' RTCP n and usage
reporting needs, there are some challenges in configuration and us
age
to avoid wasting RTCP reporting on the RTP stream that does not ne ed to avoid wasting RTCP reporting on the RTP stream that does not ne ed
that frequent reporting.</t> such frequent reporting.</li>
<t>It is not suitable for applications where some receivers like to <li>It is not suitable for applications where some receivers like to r
receive eceive
only a subset of the RTP streams, especially if multicast or trans only a subset of the RTP streams, especially if multicast or a tra
port nsport
translator is being used.</t> translator is being used.</li>
<t>There is some additional concern with legacy implementations that <li>There are some additional concerns regarding legacy implementation
do s that do
not support the RTP specification fully when it comes to handling multiple not support the RTP specification fully when it comes to handling multiple
SSRC per endpoint, as multiple simultaneous media types are sent a SSRCs per endpoint, as multiple simultaneous media types are sent
s as
separate SSRC in the same RTP session.</t> separate SSRCs in the same RTP session.</li>
<t>If the applications need finer control over which session <li>If the applications need finer control over which session
participants are included in different sets of security participants are included in different sets of security
associations, most key-management mechanisms will have difficultie s establishing associations, most key-management mechanisms will have difficultie s establishing
such a session.</t> such a session.</li>
</list> </ol>
</t>
</section> </section>
<section <section anchor="sect-5.2" numbered="true" toc="default">
anchor="section-5.2" <name>Multiple SSRCs of the Same Media Type</name>
title="Multiple SSRCs of the Same Media Type">
<t>In this design, each RTP session serves only a single media type. <t>In this design, each RTP session serves only a single media type.
The RTP session can contain multiple RTP streams, either from a single The RTP session can contain multiple RTP streams, from either a single
endpoint or from multiple endpoints. This commonly creates a low endpoint or multiple endpoints. This commonly creates a low
number of RTP sessions, typically only one for audio and one for number of RTP sessions, typically only one for audio and one for
video, with a corresponding need for two listening ports when using video, with a corresponding need for two listening ports when using
RTP/RTCP multiplexing RTP/RTCP multiplexing
<xref target="RFC5761"/>.</t> <xref target="RFC5761" format="default"/>.</t>
<t>The Advantages</t> <t>Advantages:</t>
<t> <ol spacing="normal" type="1">
<list style="numbers"> <li>It works well with split component terminals (see <xref
<t>It works well with Split Component Terminal (see Section 3.10 of target="RFC7667" sectionFormat="of" section="3.10"/>) where the
<xref target="RFC7667"/>) where the split is per media type.</t> split is per media type.</li>
<t>It enables flow-based QoS with different prioritisation between m <li>It enables flow-based QoS with different prioritization levels bet
edia ween media
types.</t> types.</li>
<t>For applications with dynamic usage of RTP streams, i.e. frequent <li>For applications with dynamic usage of RTP streams (i.e.,
ly streams are frequently
added and removed, having much of the state associated with the RT added and removed), having much of the state associated with the R
P TP
session rather than per individual SSRC can avoid the need for session rather than per individual SSRC can avoid the need for
in-session signalling of meta-information about each SSRC. In the in-session signaling of meta-information about each SSRC. In simpl
simple e
cases this allows for unsignalled RTP streams where ses cases, this allows for unsignaled RTP streams where se
sion level ssion-level
information and RTCP SDES item (e.g. CNAME) are suffien information and an RTCP SDES item (e.g., CNAME) are
t. In the more sufficient. In the more complex cases where more sourc
complex cases where more source-specific metadata needs e-specific metadata needs to be
to be signaled, the SSRC can be associated with an intermedi
signalled the SSRC can be associated with an intermedia ate identifier,
te identifier, e.g., the MID conveyed as an SDES item as defined in
e.g. the MID conveyed as an SDES item as defined in Sec <xref target="RFC8843" sectionFormat="of" section="15"
tion 15 of />.</li>
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/> <li>The overhead of security association establishment is low.</li>
.</t> </ol>
<t>There is low overhead for security association establishment.</t> <t>Disadvantages:</t>
</list> <ol spacing="normal" type="1">
</t> <li>
<t>The Disadvantages</t> <t>A slightly higher number of RTP sessions are needed, compared
<t> to multiple media types in one session
<list style="letters"> (<xref target="sect-5.1" format="default"/>). This implies the fol
<t>There are a slightly higher number of RTP sessions needed compare lowing:
d
to Multiple Media Types in one Session
<xref target="section-5.1"/>. This implies:
<list style="symbols">
<t>More NAT/FW state is needed.</t>
<t>There is increased NAT/FW-traversal cost in both processing a
nd delay.</t>
</list>
</t> </t>
<t>There is some potential for concern with legacy implementations t <ul spacing="normal">
hat don't <li>More NAT/FW state is needed.</li>
<li>The cost of NAT/FW traversal is increased in terms of both pro
cessing and delay.</li>
</ul>
</li>
<li>There is some potential for concern regarding legacy implementatio
ns that don't
support the RTP specification fully when it comes to handling mult iple support the RTP specification fully when it comes to handling mult iple
SSRC per endpoint.</t> SSRCs per endpoint.</li>
<t>It is not possible to control security association for sets of RT <li>It is not possible to control security associations for sets of RT
P P
streams within the same media type with today's key-management streams within the same media type with today's key-management
mechanisms, unless these are split into different RTP sessions mechanisms, unless these are split into different RTP sessions
(<xref target="section-5.3"/>).</t> (<xref target="sect-5.3" format="default"/>).</li>
</list> </ol>
</t>
<t>For RTP applications where all RTP streams of the same media type <t>For RTP applications where all RTP streams of the same media type
share same usage, this structure provides efficiency gains in amount share the same usage, this structure provides efficiency gains in
of network state used and provides more fate sharing with other media the amount
flows of the same type. At the same time, it is still maintaining of network state used and provides more fate-sharing with other media
flows of the same type. At the same time, it still maintains
almost all functionalities for the negotiation signaling of properties per almost all functionalities for the negotiation signaling of properties per
individual media type, and also individual media type and also
enables flow based QoS prioritisation between media types. It handles enables flow-based QoS prioritization between media types. It handles
multi-party sessions well, independently of multicast or centralised multi-party sessions well, independently of multicast or centralized
transport distribution, as additional sources can dynamically enter transport distribution, as additional sources can dynamically enter
and leave the session.</t> and leave the session.</t>
</section> </section>
<section <section anchor="sect-5.3" numbered="true" toc="default">
anchor="section-5.3" <name>Multiple Sessions for One Media Type</name>
title="Multiple Sessions for One Media Type"> <t>This design goes one step further than the design discussed in <xref
<t>This design goes one step further than above (<xref target="section-5 target="sect-5.2" format="default"/>
.2"/>) by also using multiple RTP sessions for a single media type. The main
by using multiple RTP sessions also for a single media type. The main
reason for going in this direction is that the RTP application needs reason for going in this direction is that the RTP application needs
separation of the RTP streams due to their usage, such as e.g. scalabi separation of the RTP streams according to their usage, such as, for e
lity xample, scalability
over multicast, simulcast, need for extended QoS prioritisation, or th over multicast, simulcast, the need for extended QoS prioritization, o
e need r the need
for fine-grained signalling using RTP session-focused signalling tools for fine-grained signaling using RTP session-focused signaling tools.<
.</t> /t>
<t>The Advantages:</t> <t>Advantages:</t>
<t> <ol spacing="normal" type="1">
<list style="numbers"> <li>This design is more suitable for multicast usage where receivers c
<t>This is more suitable for multicast usage where receivers can ind an individually
ividually select which RTP sessions they want to participate in, assuming
select which RTP sessions they want to participate in, assuming ea that each
ch RTP session has its own multicast group.</li>
RTP session has its own multicast group.</t> <li>When multiple different usages exist, the application can
<t>The application can indicate its usage of the RTP streams on RTP indicate its usage of the RTP streams at the RTP
session level, when multiple different usages exist.</t> session level.</li>
<t>There is less need for SSRC-specific explicit signalling for each <li>There is less need for SSRC-specific explicit signaling for each m
media edia
stream and thus reduced need for explicit and timely signalling wh stream and thus a reduced need for explicit and timely signaling w
en hen
RTP streams are added or removed.</t> RTP streams are added or removed.</li>
<t>It enables detailed QoS prioritisation for flow-based mechanisms. <li>It enables detailed QoS prioritization for flow-based mechanisms.<
</t> /li>
<t>It works well with Split Component Terminal (see Section 3.10 of <li>It works well with split component terminals (see
<xref target="RFC7667"/>).</t> <xref target="RFC7667" sectionFormat="of" section="3.10"/>).</li>
<t>The scope for who is included in a security association can be <li>The scope for who is included in a security association can be
structured around the different RTP sessions, thus enabling such structured around the different RTP sessions, thus enabling such
functionality with existing key-management.</t> functionality with existing key-management mechanisms.</li>
</list> </ol>
</t> <t>Disadvantages:</t>
<t>The Disadvantages:</t> <ol spacing="normal" type="1">
<t> <li>There is an increased amount of session configuration state compar
<list style="letters"> ed
<t>There is an increased amount of session configuration state compa to multiple SSRCs of the same media type (<xref target="sect-5.2"/
red >), due to the increased amount
to Multiple SSRCs of the Same Media Type, due to the increased amo of RTP sessions.</li>
unt <li>For RTP streams that are part of scalability, simulcast, or
of RTP sessions.</t> transport robustness, a method for binding sources across multiple
<t>For RTP streams that are part of scalability, simulcast or RTP
transport robustness, a method to bind sources across multiple RTP sessions is needed.</li>
sessions is needed.</t> <li>There is some potential for concern regarding legacy implementatio
<t>There is some potential for concern with legacy implementations t ns that
hat
don't support the RTP specification fully when it comes to handlin g don't support the RTP specification fully when it comes to handlin g
multiple SSRC per endpoint.</t> multiple SSRCs per endpoint.</li>
<t>There is higher overhead for security association establishment, <li>The overhead of security association establishment is higher, due
due to the increased number of RTP sessions.</li>
to the increased number of RTP sessions.</t> <li>If the applications need finer control over which participants
<t>If the applications need more fine-grained control than per RTP s in a given RTP session are included in different sets of
ession security associations, most of today's key-management mechanisms
over which participants that are included in different sets of sec will have difficulties establishing such a session.</li>
urity </ol>
associations, most of today's key-management will have difficultie <t>For more-complex RTP applications that have several different
s usages for RTP streams of the same media type or that use scalability
establishing such a session.</t> or
</list> simulcast, this solution can enable those functions, at the cost of
</t>
<t>For more complex RTP applications that have several different
usages for RTP streams of the same media type, or uses scalability or
simulcast, this solution can enable those functions at the cost of
increased overhead associated with the additional sessions. This type increased overhead associated with the additional sessions. This type
of structure is suitable for more advanced applications as well as of structure is suitable for more-advanced applications as well as
multicast-based applications requiring differentiation to different multicast-based applications requiring differentiation to different
participants.</t> participants.</t>
</section> </section>
<section anchor="section-5.4" title="Single SSRC per Endpoint"> <section anchor="sect-5.4" numbered="true" toc="default">
<t>In this design each endpoint in a point-to-point session has only a <name>Single SSRC per Endpoint</name>
single SSRC, thus the RTP session contains only two SSRCs, one local <t>In this design, each endpoint in a point-to-point session has only a
and one remote. This session can be used both unidirectional, i.e. single SSRC; thus, the RTP session contains only two SSRCs -- one loca
only a single RTP stream, or bi-directional, i.e. both endpoints have l
one RTP stream each. If the application needs additional media flows and one remote. This session can be used either unidirectionally
(i.e., one SSRC sends an RTP stream that is received by the other
SSRC) or bidirectionally (i.e., the two SSRCs both send an RTP
stream and receive the RTP stream sent by the other endpoint).
If the application needs additional media flows
between the endpoints, it will have to establish additional RTP between the endpoints, it will have to establish additional RTP
sessions.</t> sessions.</t>
<t>The Advantages:</t> <t>Advantages:</t>
<t> <ol spacing="normal" type="1">
<list style="numbers"> <li>This design has great potential for interoperability with legacy
<t>This design has great legacy interoperability potential as it wil applications, as it will
l not tax any RTP stack implementations.</li>
not tax any RTP stack implementations.</t> <li>The signaling system makes it possible to negotiate and describe
<t>The signalling has good possibilities to negotiate and describe t the exact formats and bitrates for each RTP stream, especially
he using today's tools in SDP.</li>
exact formats and bitrates for each RTP stream, especially using <li>It is possible to control security associations per RTP stream wit
today's tools in SDP.</t> h
<t>It is possible to control security association per RTP stream wit current key-management functions, since each RTP stream is directl
h y related to
current key-management, since each RTP stream is directly related an RTP session and the most commonly used keying mechanisms operat
to e on a
an RTP session, and the most used keying mechanisms operates on a per-session basis.</li>
per-session basis.</t> </ol>
</list> <t>Disadvantages:</t>
</t> <ol spacing="normal" type="1">
<t>The Disadvantages:</t> <li>The amount of NAT/FW state grows linearly with the number
<t> of RTP streams.</li>
<list style="letters"> <li>NAT/FW traversal increases delay and resource consumption.</li>
<t>There is a linear growth of the amount of NAT/FW state with numbe <li>There are likely more signaling message and signaling processing
r requirements due to the increased amount of session-related inform
of RTP streams.</t> ation.</li>
<t>There is increased delay and resource consumption from <li>There is higher potential for a single RTP stream to fail during
NAT/FW-traversal.</t> transport between the endpoints, due to the need for a separate
<t>There are likely larger signalling message and signalling process NAT/FW traversal for every RTP stream, since there is only one str
ing eam per session.</li>
requirements due to the increased amount of session-related inform <li>The amount of explicit state for relating RTP streams grows, depen
ation.</t> ding
<t>There is higher potential for a single RTP stream to fail during on how the application relates RTP streams.</li>
transport between the endpoints, due to the need for separate NAT/ <li>Port consumption might become a problem for centralized
FW-
traversal for every RTP stream since there is only one stream per
session.</t>
<t>The amount of explicit state for relating RTP streams grows, depe
nding
on how the application relates RTP streams.</t>
<t>The port consumption might become a problem for centralised
services, where the central node's port or 5-tuple filter consumpt ion services, where the central node's port or 5-tuple filter consumpt ion
grows rapidly with the number of sessions.</t> grows rapidly with the number of sessions.</li>
<t>For applications where the RTP stream usage is highly dynamic, i. <li>For applications where RTP stream usage is highly dynamic,
e. i.e., entities frequently enter and leave sessions, the amount of sign
entering and leaving, the amount of signalling can become high. Is aling can become high. Issues
sues
can also arise from the need for timely establishment of additiona l RTP can also arise from the need for timely establishment of additiona l RTP
sessions.</t> sessions.</li>
<t>If, against the recommendation, the same SSRC value is reused in <li>If, against the recommendation in <xref target="RFC3550"/>, the sa
me SSRC value is reused in
multiple RTP sessions rather than being randomly chosen, interwork ing multiple RTP sessions rather than being randomly chosen, interwork ing
with applications that use a different multiplexing structure will with applications that use a different multiplexing structure will
require SSRC translation.</t> require SSRC translation.</li>
</list> </ol>
</t>
<t>RTP applications with a strong need to interwork with legacy RTP <t>RTP applications with a strong need to interwork with legacy RTP
applications can potentially benefit from this structure. However, a applications can potentially benefit from this structure. However, a
large number of media descriptions in SDP can also run into issues large number of media descriptions in SDP can also run into issues
with existing implementations. For any application needing a larger with existing implementations. For any application needing a larger
number of media flows, the overhead can become very significant. This number of media flows, the overhead can become very significant. This
structure is also not suitable for non-mixed multi-party sessions, as any given structure is also not suitable for non-mixed multi-party sessions, as any given
RTP stream from each participant, although having same usage in the RTP stream from each participant, although having the same usage in th e
application, needs its own RTP session. In addition, the dynamic application, needs its own RTP session. In addition, the dynamic
behaviour that can arise in multi-party applications can tax the behavior that can arise in multi-party applications can tax the
signalling system and make timely media establishment more difficult.< signaling system and make timely media establishment more difficult.</
/t> t>
</section> </section>
<section anchor="section-5.5" title="Summary"> <section anchor="sect-5.5" numbered="true" toc="default">
<t>Both the <name>Summary</name>
"Single SSRC per Endpoint" and the "Multiple Media Types in One <t>Both the "single SSRC per endpoint" (<xref
Session" are cases that require full explicit signalling of the media target="sect-5.4"/>) and "multiple media types in one
stream relations. However, they operate on two different levels where session" (<xref target="sect-5.1"/>) cases require full explicit signa
the first primarily enables session level binding, and the second ling of the media
needs SSRC level binding. From another perspective, the two solutions stream relationships. However, they operate on two different levels, w
are the two extreme points when it comes to number of RTP sessions here
the first primarily enables session-level binding and the second
needs SSRC-level binding. From another perspective, the two solutions
are the two extremes when it comes to the number of RTP sessions
needed.</t> needed.</t>
<t>The two other designs, "Multiple SSRCs of the Same Media Type" and <t>The two other designs -- multiple SSRCs of the same media type
"Multiple Sessions for One Media Type", are two examples that primaril (<xref target="sect-5.2"/>) and
y multiple sessions for one media type (<xref target="sect-5.3"/>) -- ar
allows for some implicit mapping of the role or usage of the RTP e two examples that primarily
streams based on which RTP session they appear in. It thus potentially allow for some implicit mapping of the role or usage of the RTP
allows for less signalling and in particular reduces the need for streams based on which RTP session they appear in. Thus, they potentia
real-time signalling in sessions with dynamically changing number lly
allow for less signaling and, in particular, reduce the need for
real-time signaling in sessions with a dynamically changing number
of RTP streams. They also represent points of RTP streams. They also represent points
in-between the first two designs when it comes to amount of RTP between the first two designs when it comes to the amount of RTP
sessions established, i.e. representing an attempt to balance the sessions established, i.e., they represent an attempt to balance the
amount of RTP sessions with the functionality the communication amount of RTP sessions with the functionality the communication
session provides both on network level and on signalling level.</t> session provides at both the network level and the signaling level.</t >
</section> </section>
</section> </section>
<section anchor="section-6" title="Guidelines"> <section anchor="sect-6" numbered="true" toc="default">
<name>Guidelines</name>
<t>This section contains a number of multi-stream guidelines for <t>This section contains a number of multi-stream guidelines for
implementers, system designers, or specification writers.</t> implementers, system designers, and specification writers.</t>
<t> <dl newline="true" spacing="normal">
<list hangIndent="3" style="hanging"> <dt>Do not require the use of the same SSRC value across RTP sessions:</
<t hangText="Do not require use of the same SSRC value across RTP sess dt>
ions:"> <dd>
<vspace blankLines="0"/> As discussed in <xref target="sect-3.4.3" format="default"/>,
As discussed in <xref target="section-3.4.3"/> there are downsides to using the same SSRC in multiple RTP sessions
there exist drawbacks in using the same SSRC in multiple RTP session
s
as a mechanism to bind related RTP streams together. It is instead as a mechanism to bind related RTP streams together. It is instead
recommended to use a mechanism to explicitly signal the relation, recommended to use a mechanism to explicitly signal the relationship
either in RTP/RTCP or in the signalling mechanism used to establish ,
the RTP session(s).</t> in either RTP&wj;/RTCP or the signaling mechanism used to establish
<t hangText="Use additional RTP streams for additional media sources:" the RTP session(s).</dd>
>In <dt>Use additional RTP streams for additional media sources:</dt>
the cases where an RTP endpoint needs to transmit additional RTP <dd>In
the cases where an RTP endpoint needs to transmit additional RTP
streams of the same media type in the application, with the same streams of the same media type in the application, with the same
processing requirements at the network and RTP layers, it is suggest ed processing requirements at the network and RTP layers, it is suggest ed
to send them in the same RTP session. For example a telepresence roo to send them in the same RTP session. For example, in the case of a
m telepresence room
where there are three cameras, and each camera captures 2 persons where there are three cameras and each camera captures two persons
sitting at the table, sending each camera as its own RTP stream with sitting at the table, we suggest that each camera send its own RTP s
in tream within
a single RTP session is suggested.</t> a single RTP session.</dd>
<t hangText="Use additional RTP sessions for streams with different re <dt>Use additional RTP sessions for streams with different requirements:
quirements:"> </dt>
When RTP streams have different processing requirements from the netw <dd>
ork or When RTP streams have different processing requirements from the net
work or
the RTP layer at the endpoints, it is suggested that the different the RTP layer at the endpoints, it is suggested that the different
types of streams are put in different RTP sessions. This includes th e types of streams be put in different RTP sessions. This includes the
case where different participants want different subsets of the set of case where different participants want different subsets of the set of
RTP streams.</t> RTP streams.</dd>
<t hangText="When using multiple RTP sessions, use grouping:"> When <dt>Use grouping when using multiple RTP sessions:</dt>
<dd> When
using multiple RTP session solutions, it is suggested to explicitly using multiple RTP session solutions, it is suggested to explicitly
group the involved RTP sessions when needed using a signalling group the involved RTP sessions when needed using a signaling
mechanism, for example The Session Description Protocol (SDP) Groupi mechanism -- for example, see <xref target="RFC5888">"The Session
ng Description Protocol (SDP) Grouping Framework"</xref> -- using some
Framework appropriate grouping semantics.</dd>
<xref target="RFC5888"/>, using some appropriate grouping semantics. <dt>Ensure that RTP/RTCP extensions support multiple RTP streams as well
</t> as multiple RTP sessions:</dt>
<t <dd>When
hangText="RTP/RTCP Extensions Support Multiple RTP Streams as Well a
s Multiple RTP Sessions:">When
defining an RTP or RTCP extension, the creator needs to consider if defining an RTP or RTCP extension, the creator needs to consider if
this extension is applicable to use with additional SSRCs and multip le this extension is applicable for use with additional SSRCs and multi ple
RTP sessions. Any extension intended to be generic must support both . RTP sessions. Any extension intended to be generic must support both .
Extensions that are not as generally applicable will have to conside r Extensions that are not as generally applicable will have to conside r
if interoperability is better served by defining a single solution o whether interoperability is better served by defining a single solut
r ion or
providing both options.</t> providing both options.</dd>
<t hangText="Extensions for Transport Support:">When defining new RTP/ <dt>Provide adequate extensions for transport support:</dt>
RTCP <dd>When defining new RTP/RTCP
extensions intended for transport support, like the retransmission o r extensions intended for transport support, like the retransmission o r
FEC mechanisms, they must include support for both multiple RTP FEC mechanisms, they must include support for both multiple RTP
streams in the same RTP session and multiple RTP sessions, such that streams in the same RTP session and multiple RTP sessions, such that
application developers can choose freely from the set of mechanisms application developers can choose freely from the set of mechanisms
without concerning themselves with which of the multiplexing choices a without concerning themselves with which of the multiplexing choices a
particular solution supports.</t> particular solution supports.</dd>
</list> </dl>
</t>
</section> </section>
<section anchor="section-8" title="IANA Considerations"> <section anchor="sect-8" numbered="true" toc="default">
<t>This document makes no request of IANA.</t> <name>IANA Considerations</name>
<t>Note to RFC Editor: this section can be removed on publication as <t>This document has no IANA actions.</t>
an RFC.</t>
</section> </section>
<section anchor="section-9" title="Security Considerations"> <section anchor="sect-9" numbered="true" toc="default">
<t>The security considerations of the RTP specification <name>Security Considerations</name>
<xref target="RFC3550"/>, <t>The security considerations discussed in the RTP specification
<xref target="RFC3550" format="default"/>;
any applicable RTP profile any applicable RTP profile
<xref target="RFC3551"/>,<xref target="RFC4585"/>,<xref target="RFC3711" <xref target="RFC3551" format="default"/> <xref target="RFC4585"
/>, format="default"/> <xref target="RFC3711" format="default"/>;
and the extensions for sending multiple media types in a single RTP and the extensions for sending multiple media types in a single RTP
session session
<xref target="I-D.ietf-avtcore-multi-media-rtp-session"/>, RID <xref target="RFC8860" format="default"/>, RID
<xref target="I-D.ietf-mmusic-rid"/>, BUNDLE <xref target="RFC8851" format="default"/>, BUNDLE
<xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/>, <xref target="RFC8843" format="default"/>,
<xref target="RFC5760"/>, <xref target="RFC5760" format="default"/>, and
<xref target="RFC5761"/>, apply if selected and thus need to be consider <xref target="RFC5761" format="default"/> apply if selected and thus nee
ed in the evaluation.</t> d to be considered in the evaluation.</t>
<t><xref target="sect-4.3" format="default"/> discusses the security impli
<t>There is discussion of the security implications of choosing cations of choosing
multiple SSRC vs multiple RTP sessions in multiple SSRCs vs. multiple RTP sessions.</t>
<xref target="section-4.3"/>.</t>
</section>
<section title="Contributors">
<t>Hui Zheng (Marvin) contributed to WG draft versions -04
and -05 of the document.
</t>
</section>
<section title="Acknowledgments">
<t>The Authors like to acknowledge and thank Cullen Jennings, Dale R Worle
y, Huang Yihong (Rachel), Benjamin Kaduk, Mirja Kuehlewind, and Vijay Gurbani
for review and comments.
</t>
</section> </section>
</middle> </middle>
<back> <back>
<references title="Normative References"> <references>
&RFC3550; &RFC3551; &RFC3711; &RFC4585; &RFC5576; <name>References</name>
&RFC5760; &RFC5761; &RFC7656; &RFC7667; <references>
&I-D.ietf-avtcore-multi-media-rtp-session; &I-D.ietf-mmusic-rid; <name>Normative References</name>
&I-D.ietf-mmusic-sdp-bundle-negotiation; <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3550.
&I-D.ietf-perc-srtp-ekt-diet; xml"/>
</references> <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3551.
<references title="Informative References"> xml"/>
&RFC2198; &RFC2205; &RFC2474; &RFC2974; &RFC3261; &RFC3264; &RFC3389; &RFC <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3711.
3830; xml"/>
&RFC4103; &RFC4383; &RFC4566; &RFC4568; &RFC4588; &RFC5104; &RFC5109; <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4585.
&RFC5389; &RFC5764; &RFC5888; &RFC6465; &RFC7201; &RFC7657; &RFC7826; xml"/>
&RFC7983; &RFC8088; &RFC8108; &RFC8445; <xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5576.
&I-D.ietf-avtext-rid; &I-D.ietf-perc-private-media-framework; xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5760.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5761.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7656.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7667.
xml"/>
<reference anchor="JINGLE"> <!-- draft-ietf-avtcore-multi-media-rtp-session (RFC 8860) -->
<front> <reference anchor="RFC8860" target="https://www.rfc-editor.org/info/rfc8860">
<title>XEP-0166: Jingle</title> <front>
<author initials="S." surname="Ludwig"> <title>Sending Multiple Types of Media in a Single RTP Session</title>
<author initials="M." surname="Westerlund" fullname="Magnus Westerlund">
<organization/>
</author>
<author initials="C." surname="Perkins" fullname="Colin Perkins">
<organization/>
</author>
<author initials="J." surname="Lennox" fullname="Jonathan Lennox">
<organization/>
</author>
<date month="January" year="2021"/>
</front>
<seriesInfo name="RFC" value="8860"/>
<seriesInfo name="DOI" value="10.17487/RFC8860"/>
</reference>
<!-- draft-ietf-mmusic-rid (RFC 8851) -->
<reference anchor='RFC8851' target="https://www.rfc-editor.org/info/rfc8851">
<front>
<title>RTP Payload Format Restrictions</title>
<author initials='A.B.' surname='Roach' fullname='Adam Roach' role="editor">
<organization />
</author>
<date month='January' year='2021' />
</front>
<seriesInfo name="RFC" value="8851"/>
<seriesInfo name="DOI" value="10.17487/RFC8851"/>
</reference>
<!-- draft-ietf-mmusic-sdp-bundle-negotiation (RFC 8843) -->
<reference anchor="RFC8843" target="https://www.rfc-editor.org/info/rfc8843"
>
<front>
<title>Negotiating Media Multiplexing Using the Session Description Prot
ocol (SDP)</title>
<author initials="C" surname="Holmberg" fullname="Christer Holmberg">
<organization/>
</author>
<author initials="H" surname="Alvestrand" fullname="Harald Alvestrand">
<organization/>
</author>
<author initials="C" surname="Jennings" fullname="Cullen Jennings">
<organization/>
</author>
<date month="January" year="2021"/>
</front>
<seriesInfo name="RFC" value="8843"/>
<seriesInfo name="DOI" value="10.17487/RFC8843"/>
</reference>
<!-- draft-ietf-avtext-rid (RFC 8852) -->
<reference anchor='RFC8852' target="https://www.rfc-editor.org/info/rfc8852">
<front>
<title>RTP Stream Identifier Source Description (SDES)</title>
<author initials='A.B.' surname='Roach' fullname='Adam Roach'>
<organization />
</author>
<author initials='S' surname='Nandakumar' fullname='Suhas Nandakumar'>
<organization />
</author>
<author initials='P' surname='Thatcher' fullname='Peter Thatcher'>
<organization />
</author>
<date month='January' year='2021' />
</front>
<seriesInfo name="RFC" value="8852"/>
<seriesInfo name="DOI" value="10.17487/RFC8852"/>
</reference>
<!-- draft-ietf-perc-srtp-ekt-diet (RFC 8870) -->
<reference anchor="RFC8870" target="https://www.rfc-editor.org/info/rfc8870">
<front>
<title>Encrypted Key Transport for DTLS and Secure RTP</title>
<author initials="C" surname="Jennings" fullname="Cullen Jennings">
<organization>company</organization>
</author>
<author initials="J" surname="Mattsson" fullname="John Mattsson">
<organization>company</organization>
</author>
<author initials="D" surname="McGrew" fullname="David A. McGrew">
<organization>company</organization>
</author>
<author initials="D" surname="Wing" fullname="Dan Wing">
<organization>company</organization>
</author>
<author initials="F" surname="Andreasen" fullname="Flemming Andreasen">
<organization>company</organization>
</author>
<date month="January" year="2021"/>
</front>
<seriesInfo name="RFC" value="8870"/>
<seriesInfo name="DOI" value="10.17487/RFC8870"/>
</reference>
</references>
<references>
<name>Informative References</name>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2198.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2205.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2474.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2974.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3261.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3264.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3389.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3830.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4103.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4383.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4566.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4568.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4588.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5104.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5109.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5389.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5764.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5888.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6465.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7201.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7657.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7826.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7983.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8088.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8108.
xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8445.
xml"/>
<!-- draft-ietf-perc-private-media-framework (RFC 8871) -->
<reference anchor='RFC8871' target="https://www.rfc-editor.org/info/rfc8871">
<front>
<title>A Solution Framework for Private Media in Privacy-Enhanced RTP Conferenci
ng (PERC)</title>
<author initials='P' surname='Jones' fullname='Paul Jones'>
<organization />
</author>
<author initials='D' surname='Benham' fullname='David Benham'>
<organization />
</author>
<author initials='C' surname='Groves' fullname='Christian Groves'>
<organization />
</author>
<date month='January' year='2021'/>
</front>
<seriesInfo name="RFC" value="8871"/>
<seriesInfo name="DOI" value="10.17487/RFC8871"/>
</reference>
<reference anchor="JINGLE" target="https://xmpp.org/extensions/xep-0166.
html">
<front>
<title>XEP-0166: Jingle</title>
<author initials="S." surname="Ludwig">
</author> </author>
<author initials="J." surname="Beda"> <author initials="J." surname="Beda">
</author> </author>
<author initials="P." surname="Saint-Andre"> <author initials="P." surname="Saint-Andre">
</author> </author>
<author initials="R." surname="McQueen"> <author initials="R." surname="McQueen">
</author> </author>
<author initials="S." surname="Egan"> <author initials="S." surname="Egan">
</author> </author>
<author initials="J." surname="Hildebrand"> <author initials="J." surname="Hildebrand">
</author> </author>
<date month="September" year="2018"/> <date month="September" year="2018"/>
</front> </front>
<seriesInfo </reference>
name="XMPP.org" </references>
value="https://xmpp.org/extensions/xep-0166.html"/>
</reference>
</references> </references>
<section <section anchor="sect-a" numbered="true" toc="default">
anchor="section-a" <name>Dismissing Payload Type Multiplexing</name>
title="Dismissing Payload Type Multiplexing">
<t>This section documents a number of reasons why using the payload <t>This section documents a number of reasons why using the payload
type as a multiplexing point is unsuitable for most issues related to type as a multiplexing point is unsuitable for most issues related to
multiple RTP streams. Attempting to use Payload type multiplexing multiple RTP streams. Attempting to use payload type multiplexing
beyond its defined usage has well known negative effects on RTP beyond its defined usage has well-known negative effects on RTP, as
discussed below. discussed below.
To use payload type as the single discriminator for multiple streams To use the payload type as the single discriminator for multiple streams
implies that all the different RTP streams are being sent with the implies that all the different RTP streams are being sent with the
same SSRC, thus using the same timestamp and sequence number space. same SSRC, thus using the same timestamp and sequence number space.
This has many effects:</t> The many effects of using payload type multiplexing are as follows:</t>
<t> <ol spacing="normal" type="1">
<list style="numbers"> <li>Constraints are placed on the RTP timestamp rate for the multiplexed
<t>Putting constraints on RTP timestamp rate for the multiplexed media media.
.
For example, RTP streams that use different RTP timestamp rates cann ot For example, RTP streams that use different RTP timestamp rates cann ot
be combined, as the timestamp values need to be consistent across al l be combined, as the timestamp values need to be consistent across al l
multiplexed media frames. Thus streams are forced to use the same RT P multiplexed media frames. Thus, streams are forced to use the same R TP
timestamp rate. When this is not possible, payload type multiplexing timestamp rate. When this is not possible, payload type multiplexing
cannot be used.</t> cannot be used.</li>
<t>Many RTP payload formats can fragment a media object over multiple <li>Many RTP payload formats can fragment a media object over multiple
RTP packets, like parts of a video frame. These payload formats need RTP packets, like parts of a video frame. These payload formats need
to determine the order of the fragments to correctly decode them. to determine the order of the fragments to correctly decode them.
Thus, it is important to ensure that all fragments related to a fram e Thus, it is important to ensure that all fragments related to a fram e
or a similar media object are transmitted in sequence and without or a similar media object are transmitted in sequence and without
interruptions within the object. This can relatively simple be solve d interruptions within the object. This can be done relatively easily
on the sender side by ensuring that the fragments of each RTP stream on the sender side by ensuring that the fragments of each RTP stream
are sent in sequence.</t> are sent in sequence.</li>
<t>Some media formats require uninterrupted sequence number space <li>Some media formats require uninterrupted sequence number space
between media parts. These are media formats where any missing RTP between media parts. These are media formats where any missing RTP
sequence number will result in decoding failure or invoking a repair sequence number will result in decoding failure or invoking a repair
mechanism within a single media context. The text/ T140 payload form mechanism within a single media context. The text&wj;/t140 payload f
at ormat
<xref target="RFC4103"/> <xref target="RFC4103" format="default"/>
is an example of such a format. These formats will need a sequence is an example of such a format. These formats will need a sequence
numbering abstraction function between RTP and the individual RTP numbering abstraction function between RTP and the individual RTP
stream before being used with payload type multiplexing.</t> stream before being used with payload type multiplexing.</li>
<t>Sending multiple media streams in the same sequence number space ma <li>Sending multiple media streams in the same sequence number space
kes it makes it
impossible to determine which media stream lost a packet. This as th impossible to determine which media stream lost a packet.
e Such a scenario causes difficulties, since the receiver cannot deter
payload type that is used for demultiplex the media stream is not rec mine to which stream it should
eived. apply packet-loss concealment or other stream-specific
Thus, causing the receiver difficulties in determining which stream loss-mitigation mechanisms.</li>
to <li>If RTP retransmission
apply packet loss concealment or other stream-specific loss mitigatio <xref target="RFC4588" format="default"/>
n is used and packet loss occurs, it is possible to ask for the missin
mechanisms.</t> g
<t>If RTP Retransmission packet(s) by SSRC and sequence number -- not by payload type. If onl
<xref target="RFC4588"/> y
is used and there is a loss, it is possible to ask for the missing
packet(s) by SSRC and sequence number, not by payload type. If only
some of the payload type multiplexed streams are of interest, there is some of the payload type multiplexed streams are of interest, there is
no way of telling which missing packet(s) belong to the interesting no way to tell which missing packet or packets belong to the
stream(s) and all lost packets need be requested, wasting bandwidth. stream or streams of interest, and all lost packets need to be reque
</t> sted, wasting bandwidth.</li>
<t>The current RTCP feedback mechanisms are built around providing <li>The current RTCP feedback mechanisms are built around providing
feedback on RTP streams based on stream ID (SSRC), packet (sequence feedback on RTP streams based on stream ID (SSRC), packet (sequence
numbers) and time interval (RTP timestamps). There is almost never a numbers), and time interval (RTP timestamps). There is almost never a
field to indicate which payload type is reported, so sending feedbac k field to indicate which payload type is reported, so sending feedbac k
for a specific RTP payload type is difficult without extending for a specific RTP payload type is difficult without extending
existing RTCP reporting.</t> existing RTCP reporting.</li>
<t>The current RTCP media control messages <li>The current RTCP media control messages specification
<xref target="RFC5104"/> <xref target="RFC5104" format="default"/>
specification is oriented around controlling particular media flows, is oriented around controlling particular media flows,
i.e. requests are done addressing a particular SSRC. Such mechanisms i.e., requests are done by addressing a particular SSRC. Such mechan
would need to be redefined to support payload type multiplexing.</t> isms
<t>The number of payload types are inherently limited. Accordingly, would need to be redefined to support payload type multiplexing.</li
>
<li>The number of payload types is inherently limited. Accordingly,
using payload type multiplexing limits the number of streams that ca n using payload type multiplexing limits the number of streams that ca n
be multiplexed and does not scale. This limitation is exacerbated if be multiplexed and does not scale. This limitation is exacerbated if
one uses solutions like RTP and RTCP multiplexing one uses solutions like RTP and RTCP multiplexing
<xref target="RFC5761"/> <xref target="RFC5761" format="default"/>
where a number of payload types are blocked due to the overlap betwe en where a number of payload types are blocked due to the overlap betwe en
RTP and RTCP.</t> RTP and RTCP.</li>
<t>At times, there is a need to group multiplexed streams and this is <li>At times, there is a need to group multiplexed streams. This is
currently possible for RTP sessions and for SSRC, but there is no currently possible for RTP sessions and SSRCs, but there is no
defined way to group payload types.</t> defined way to group payload types.</li>
<t>It is currently not possible to signal bandwidth requirements per <li>It is currently not possible to signal bandwidth requirements per
RTP stream when using payload type multiplexing.</t> RTP stream when using payload type multiplexing.</li>
<t>Most existing SDP media level attributes cannot be applied on a per <li>Most existing SDP media-level attributes cannot be applied on a
payload type level and would require re-definition in that context.< per-payload-type basis and would require redefinition in that contex
/t> t.</li>
<t>A legacy endpoint that does not understand the indication that <li>A legacy endpoint that does not understand the indication that
different RTP payload types are different RTP streams might be different RTP payload types are different RTP streams might be
slightly confused by the large amount of possibly overlapping or slightly confused by the large amount of possibly overlapping or
identically defined RTP payload types.</t> identically defined RTP payload types.</li>
</list> </ol>
</t>
</section> </section>
<section anchor="section-b" title="Signalling Considerations"> <section anchor="sect-b" numbered="true" toc="default">
<t>Signalling is not an architectural consideration for RTP itself, so <name>Signaling Considerations</name>
<t>Signaling is not an architectural consideration for RTP itself, so
this discussion has been moved to an appendix. However, it is extremely this discussion has been moved to an appendix. However, it is extremely
important for anyone building complete applications, so it is important for anyone building complete applications, so it is
deserving of discussion.</t> deserving of discussion.</t>
<t>We document salient issues here that need to be addressed by the WGs <t>We document some issues here that need to be addressed when using some
that use some form of signaling to establish RTP sessions. These form of signaling to establish RTP sessions. These
issues cannot simply be addressed by tweaking, extending, or profilin issues cannot be addressed by simply tweaking, extending, or profilin
g g
RTP, but require a dedicated and indepth look at the signaling RTP; rather, they require a dedicated and in-depth look at the signal
ing
primitives that set up the RTP sessions.</t> primitives that set up the RTP sessions.</t>
<t>There exist various signalling solutions for establishing RTP <t>There exist various signaling solutions for establishing RTP
sessions. Many are SDP sessions. Many are based on SDP
<xref target="RFC4566"/> <xref target="RFC4566" format="default"/>;
based, however SDP functionality is also dependent on the signalling however, SDP functionality is also dependent on the signaling
protocols carrying the SDP. RTSP protocols carrying the SDP. The Real-Time Streaming Protocol (RTSP)
<xref target="RFC7826"/> <xref target="RFC7826" format="default"/>
and SAP and the Session Announcement Protocol (SAP)
<xref target="RFC2974"/> <xref target="RFC2974" format="default"/>
both use SDP in a declarative fashion, while SIP both use SDP in a declarative fashion, while SIP
<xref target="RFC3261"/> <xref target="RFC3261" format="default"/>
uses SDP with the additional definition of Offer/Answer uses SDP with the additional definition of offer/answer
<xref target="RFC3264"/>. The impact on signalling and especially SDP <xref target="RFC3264" format="default"/>. The impact on signaling,
needs to be considered as it can greatly affect how to deploy a and especially on SDP,
needs to be considered, as it can greatly affect how to deploy a
certain multiplexing point choice.</t> certain multiplexing point choice.</t>
<section anchor="section-b.1" title="Session Oriented Properties"> <section anchor="sect-b.1" numbered="true" toc="default">
<t>One aspect of the existing signalling is that it is focused on <name>Session-Oriented Properties</name>
RTP sessions, or in the case of SDP, the media description concept. <t>One aspect of existing signaling protocols is that they are focused o
There are a number of things that are signalled on media description n
level but those are not necessarily strictly bound to an RTP session RTP sessions or, in the case of SDP, the concept of media
and could be of interest to signal specifically for a particular RTP descriptions. A number of things are signaled at the media
stream (SSRC) within the session. The following properties have been description level, but those are not necessarily strictly bound to
identified as being potentially useful to signal not only on RTP an RTP session and could be of interest for signaling, especially
session level:</t> for a particular RTP stream (SSRC) within the session.
<t> The following properties have been
<list style="symbols"> identified as being potentially useful for signaling, and not only
<t>Bitrate/Bandwidth exist today only at aggregate or as a common "a at the RTP session level:</t>
ny <ul spacing="normal">
RTP stream" limit, unless either codec-specific bandwidth limiting <li>Bitrate and/or bandwidth can be specified today only as an
or aggregate limit, or as a common "any RTP stream" limit, unless
RTCP signalling using TMMBR <xref target="RFC5104"/> is used.</t> either codec-specific bandwidth limiting or
<t>Which SSRC that will use which RTP payload type (this will be RTCP signaling using Temporary Maximum Media Stream Bit Rate
visible from the first media packet, but is sometimes useful to kn Request (TMMBR) messages <xref target="RFC5104"
ow format="default"/> is used.
before packet arrival).</t> </li>
</list> <li>Which SSRC will use which RTP payload type (this information will
</t> be
visible in the first media packet but is sometimes useful to have
before the packet arrives).</li>
</ul>
<t>Some of these issues are clearly SDP's problem rather than RTP <t>Some of these issues are clearly SDP's problem rather than RTP
limitations. However, if the aim is to deploy an solution using limitations. However, if the aim is to deploy a solution that uses
additional SSRCs that contains several sets of RTP streams with several SSRCs and contains several sets of RTP streams with
different properties (encoding/packetization parameter, bit-rate, different properties (encoding/packetization parameters, bitrate,
etc.), putting each set in a different RTP session would directly etc.), putting each set in a different RTP session would directly
enable negotiation of the parameters for each set. If insisting on enable negotiation of the parameters for each set. If insisting on
additional SSRC only, a number of signalling extensions are needed to additional SSRCs only, a number of signaling extensions are needed to
clarify that there are multiple sets of RTP streams with different clarify that there are multiple sets of RTP streams with different
properties and that they need in fact be kept different, since a properties and that they in fact need to be kept different, since a
single set will not satisfy the application's requirements.</t> single set will not satisfy the application's requirements.</t>
<t>For some parameters, such as RTP payload type, resolution and <t>For some parameters, such as RTP payload type, resolution, and
framerate, a SSRC-linked mechanism has been proposed in frame rate, an SSRC-linked mechanism has been proposed in
<xref target="I-D.ietf-mmusic-rid"/></t> <xref target="RFC8851" format="default"/>.</t>
</section> </section>
<section <section anchor="sect-b.2" numbered="true" toc="default">
anchor="section-b.2" <name>SDP Prevents Multiple Media Types</name>
title="SDP Prevents Multiple Media Types"> <t>SDP uses the "m=" line to both delineate an RTP session and specify
<t>SDP chose to use the m= line both to delineate an RTP session and the top-level media type: audio, video, text, image, application.
to specify the top level of the MIME media type; audio, video, text, This media type is used as the top-level media type for identifying
image, application. This media type is used as the top-level media the actual payload format and is bound to a particular payload type
type for identifying the actual payload format and is bound to a using the "a=rtpmap:" attribute. This binding has to be loosened in
particular payload type using the rtpmap attribute. This binding has order to use SDP to describe RTP sessions containing multiple
to be loosened in order to use SDP to describe RTP sessions containing top-level media types.</t>
multiple MIME top level types.</t> <t><xref target="RFC8843" format="default"/>
<t><xref target="I-D.ietf-mmusic-sdp-bundle-negotiation"/>
describes how to let multiple SDP media descriptions use a single describes how to let multiple SDP media descriptions use a single
underlying transport in SDP, which allows to define one RTP session underlying transport in SDP, which allows the definition of one RTP se
with media types having different MIME top level types.</t> ssion
with different top-level media types.</t>
</section> </section>
<section anchor="section-b.3" title="Signalling RTP Stream Usage"> <section anchor="sect-b.3" numbered="true" toc="default">
<t>RTP streams being transported in RTP have some particular usage in <name>Signaling RTP Stream Usage</name>
an RTP application. This usage of the RTP stream is in many <t>RTP streams being transported in RTP have a particular usage in
applications so far implicitly signalled. For example, an application an RTP application. In many applications to date, this usage of the RT
might choose to take all incoming audio RTP streams, mix them and play P
them out. However, in more advanced applications that use multiple RTP stream is implicitly signaled. For example, an application
streams there will be more than a single usage or purpose among the might choose to take all incoming audio RTP streams, mix them, and pla
y
them out. However, in more-advanced applications that use multiple RTP
streams, there will be more than a single usage or purpose among the
set of RTP streams being sent or received. RTP applications will need set of RTP streams being sent or received. RTP applications will need
to signal this usage somehow. The signalling used will have to to somehow signal this usage. The signaling that is used will have to
identify the RTP streams affected by their RTP- level identifiers, identify the RTP streams affected by their RTP-level identifiers,
which means that they have to be identified either by their session or which means that they have to be identified by either their session or
by their SSRC + session.</t> their SSRC + session.</t>
<t>In some applications, the receiver cannot utilise the RTP stream at <t>In some applications, the receiver cannot utilize the RTP stream at
all before it has received the signalling message describing the RTP all before it has received the signaling message describing the RTP
stream and its usage. In other applications, there exists a default stream and its usage. In other applications, there exists a default
handling that is appropriate.</t> handling method that is appropriate.</t>
<t>If all RTP streams in an RTP session are to be treated in the same <t>If all RTP streams in an RTP session are to be treated in the same
way, identifying the session is enough. If SSRCs in a session are to way, identifying the session is enough. If SSRCs in a session are to
be treated differently, signalling needs to identify both the session be treated differently, signaling needs to identify both the session
and the SSRC.</t> and the SSRC.</t>
<t>If this signalling affects how any RTP central node, like an RTP <t>If this signaling affects how any RTP central node, like an RTP
mixer or translator that selects, mixes or processes streams, treats mixer or translator that selects, mixes, or processes streams, treats
the streams, the node will also need to receive the same signalling to the streams, the node will also need to receive the same signaling to
know how to treat RTP streams with different usage in the right know how to treat RTP streams with different usages in the right
fashion.</t> fashion.</t>
</section> </section>
</section> </section>
<section numbered="false" toc="default">
<name>Acknowledgments</name>
<t>The authors would like to acknowledge and thank <contact fullname="Cull
en
Jennings"/>, <contact fullname="Dale R. Worley"/>, <contact
fullname="Huang Yihong (Rachel)"/>, <contact fullname="Benjamin
Kaduk"/>, <contact fullname="Mirja Kühlewind"/>, and <contact
fullname="Vijay Gurbani"/> for review and comments.</t>
</section>
<section numbered="false" toc="default">
<name>Contributors</name>
<t><contact fullname="Hui Zheng (Marvin)"/> contributed to WG draft versio
ns -04
and -05 of the document.
</t>
</section>
</back> </back>
</rfc> </rfc>
 End of changes. 268 change blocks. 
1502 lines changed or deleted 1579 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/