Network Working Group R. Kumar
Request for Comments: 3108 M. Mostafa
Category: Standards Track Cisco Systems
May 2001
Conventions for the use of the Session Description Protocol (SDP)
for ATM Bearer Connections
Status of this Memo
This document specifies an Internet standards track protocol for the
Internet community, and requests discussion and suggestions for
improvements. Please refer to the current edition of the "Internet
Official Protocol Standards" (STD 1) for the standardization state
and status of this protocol. Distribution of this memo is unlimited.
Copyright Notice
Copyright (C) The Internet Society (2001). All Rights Reserved.
Abstract
This document describes conventions for using the Session Description
Protocol (SDP) described in RFC 2327 for controlling ATM Bearer
Connections, and any associated ATM Adaptation Layer (AAL). The AALs
addressed are Type 1, Type 2 and Type 5. This list of conventions is
meant to be exhaustive. Individual applications can use subsets of
these conventions. Further, these conventions are meant to comply
strictly with the SDP syntax as defined in RFC 2327.
Table of Contents
1. Introduction................................................... 3
1.1 Key words to indicate Requirement Levels..................... 5
2. Representation of Certain Fields within SDP description lines.. 5
2.1 Representation of Extension Attributes....................... 5
2.2 Representation of Parameter Values........................... 5
2.3 Directionality Convention.................................... 6
2.4 Case convention............................................... 7
2.5 Use of special characters in SDP parameter values............. 8
3. Capabilities Provided by SDP conventions....................... 8
4. Format of the ATM Session Description.......................... 9
5. Structure of the Session Description Lines.................... 11
5.1 The Origin Line.............................................. 11
5.2 The Session Name Line........................................ 12
5.3 The Connection Information Line.............................. 13
5.4 The Timestamp Line........................................... 15
Kumar & Mostafa Standards Track [Page 1]
RFC 3108 ATM SDP May 2001
5.5 Media Information Line for ATM connections................... 16
5.5.1 The Virtual Connection ID.................................. 16
5.5.2 The Transport Parameter.................................... 19
5.5.3 The Format List for AAL1 and AAL5 applications............. 21
5.5.4 The Format List for AAL2 applications...................... 21
5.5.5 Media information line construction........................ 22
5.6 The Media Attribute Lines.................................... 27
5.6.1 ATM bearer connection attributes........................... 28
5.6.1.1 The 'eecid' attribute.................................... 30
5.6.1.2 The 'aalType' attribute.................................. 31
5.6.1.3 The 'capability' attribute............................... 32
5.6.1.4 The 'qosClass' attribute................................. 33
5.6.1.5 The 'bcob' attribute..................................... 34
5.6.1.6 The 'stc' attribute...................................... 34
5.6.1.7 The 'upcc' attribute..................................... 35
5.6.1.8 The 'atmQOSparms' attribute.............................. 35
5.6.1.9 The 'atmTrfcDesc' attribute............................. 37
5.6.1.10 The 'abrParms' attribute................................. 39
5.6.1.11 The 'abrSetup' attribute................................. 40
5.6.1.12 The 'bearerType' attribute............................... 41
5.6.1.13 The 'lij' attribute...................................... 42
5.6.1.14 The 'anycast' attribute.................................. 43
5.6.1.15 The 'cache' attribute.................................... 43
5.6.1.16 The 'bearerSigIE' attribute.............................. 44
5.6.2 ATM Adaptation Layer (AAL) attributes...................... 45
5.6.2.1 The 'aalApp' attribute................................... 46
5.6.2.2 The 'cbrRate' attribute.................................. 48
5.6.2.3 The 'sbc' attribute...................................... 49
5.6.2.4 The 'clkrec' attribute................................... 51
5.6.2.5 The 'fec' attribute...................................... 51
5.6.2.6 The 'prtfl' attribute.................................... 51
5.6.2.7 The 'structure' attribute................................ 52
5.6.2.8 The 'cpsSDUsize' attribute............................... 53
5.6.2.9 The 'aal2CPS' attribute.................................. 53
5.6.2.10 The 'aal2CPSSDUrate' attribute........................... 54
5.6.2.11 The 'aal2sscs3661unassured' attribute.................... 54
5.6.2.12 The 'aal2sscs3661assured' attribute...................... 55
5.6.2.13 The 'aal2sscs3662' attribute............................. 56
5.6.2.14 The 'aal5sscop' attribute................................ 58
5.6.3 Service attributes......................................... 58
5.6.3.1 The 'atmmap' attribute................................... 60
5.6.3.2 The 'silenceSupp' attribute.............................. 63
5.6.3.3 The 'ecan' attribute..................................... 65
5.6.3.4 The 'gc' attributes...................................... 66
5.6.3.5 The 'profileDesc' attribute.............................. 67
5.6.3.6 The 'vsel' attribute..................................... 68
5.6.3.7 The 'dsel' attribute..................................... 70
5.6.3.8 The 'fsel' attribute..................................... 72
Kumar & Mostafa Standards Track [Page 2]
RFC 3108 ATM SDP May 2001
5.6.3.9 The 'onewaySel' attribute................................ 73
5.6.3.10 The 'codecconfig' attribute.............................. 75
5.6.3.11 The 'isup_usi' attribute................................. 76
5.6.3.12 The 'uiLayer1_Prot' attribute............................ 76
5.6.4 Miscellaneous media attributes............................. 77
5.6.4.1 The 'chain' attribute..................................... 77
5.6.5 Use of the second media-level part in H.323 Annex C
applications............................................... 78
5.6.6 Use of the eecid media attribute in call establishment
procedures................................................. 78
6. List of Parameters with Representations....................... 83
7. Examples of ATM session descriptions using SDP................. 93
8. Security Considerations........................................ 94
8.1 Bearer Security.............................................. 94
8.2 Security of the SDP description.............................. 95
9. ATM SDP Grammar................................................ 95
References........................................................104
Acknowledgements..................................................109
Authors' Addresses................................................109
Full Copyright Statement..........................................110
1. Introduction
SDP will be used in conjunction with a connection handling /device
control protocol such as Megaco (H.248) [26], SIP [18] or MGCP [25]
to communicate the information needed to set up ATM and AAL2 bearer
connections. These connections include voice connections, voiceband
data connections, clear channel circuit emulation connections, video
connections and baseband data connections (such as fax relay, modem
relay, SSCOP, frame relay etc.).
These conventions use standard SDP syntax as defined in RFC 2327 [1]
to describe the ATM-level and AAL-level connections, addresses and
other parameters. In general, parameters associated with layers
higher than the ATM adaptation layer are included only if they are
tightly coupled to the ATM or AAL layers. Since the syntax conforms
to RFC 2327, standard SDP parsers should react in a well-defined and
safe manner on receiving session descriptions based on the SDP
conventions in this document. This is done by extending the values
of fields defined in RFC 2327 rather than by defining new fields.
This is true for all SDP lines except the of the media attribute
lines, in which case new attributes are defined. The SDP protocol
allows the definition of new attributes in the media attribute lines
which are free-form. For the remaining lines, the fact that the
field in an SDP descriptor is set to "ATM" should
preclude the misinterpretation of extended parameter values by RFC
2327-compliant SDP parsers.
Kumar & Mostafa Standards Track [Page 3]
RFC 3108 ATM SDP May 2001
These conventions are meant to address the following ATM
applications:
1. Applications in which a new SVC is set-up for each service
connection. These SVCs could be AAL1 or AAL5 SVCs or single-
CID AAL2 SVCs.
2. Applications in which existing path resources are assigned to
service connections. These resources could be:
* AAL1/AAL5 PVCs, SPVCs or cached SVCs,
* AAL2 single-CID PVCs, SPVCs or cached SVCs,
* CIDs within AAL2 SVCs/PVCs/SPVCs that multiplex multiple
CIDs.
* Subchannels (identified by CIDs) within AAL1 [8] or AAL2
[11] SVCs/PVCs/SPVCs.
Note that the difference between PVCs and SPVCs is in the way the
bearer virtual circuit connection is set up. SPVCs are a class of
PVCs that use bearer signaling, as opposed to node-by-node
provisioning, for connection establishment.
This document is limited to the case when the network type is ATM.
This includes raw RTP encapsulation [45] or voice sample
encapsulation [46] over AAL5 with no intervening IP layer. It does
not address SDP usage for IP, with or without ATM as a lower layer.
In some cases, IP connection set-up is independent of lower layers,
which are configured prior to it. For example, AAL5 PVCs that
connect IP routers can be used for VoIP calls. In other cases, VoIP
call set-up is closely tied to ATM-level connection set-up. This
might require a chaining of IP and ATM descriptors, as described in
section 5.6.4.1.
This document makes no assumptions on who constructs the session
descriptions (media gateway, intermediate ATM/AAL2 switch, media
gateway controller etc.). This will be different in different
applications. Further, it allows the use of one session description
for both directions of a connection (as in SIP and MGCP applications)
or the use of separate session descriptions for different directions.
It also addresses the ATM multicast and anycast capabilities.
This document makes no assumptions about how the SDP description will
be coded. Although the descriptions shown here are encoded as text,
alternate codings are possible:
- Binary encoding such as ASN.1. This is an option (in addition to
text encoding) in the Megaco context.
Kumar & Mostafa Standards Track [Page 4]
RFC 3108 ATM SDP May 2001
- Use of extended ISUP parameters [36] to encode the information in
SDP descriptors, with conversion to/from binary/text-based SDP
encoding when needed.
1.1 Key words to indicate Requirement Levels
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119 [62].
2. Representation of Certain Fields within SDP description lines
This document conforms to the syntactic conventions of standard SDP
as defined in RFC 2327 [1].
2.1 Representation of Extension Attributes
The SDP protocol [1] requires that non-standard attributes and codec
names use an "X-" prefix.
In this internet document, the "X-" prefix is used consistently for
codec names (Table 2) that have not been registered with the IANA.
The IANA-registered codec names listed in [31] do not use this
prefix, regardless of whether they are statically or dynamically
assigned payload types.
However, this prefix is not used for the extension SDP attributes
defined in this document. This has been done to enhance legibility.
This document suggests that parsers be flexible in the use of the
"X-" prefix convention. They should accept codec names and attribute
names with or without the "X-" prefix.
2.2 Representation of Parameter Values
Depending on the format of their representation in SDP, the
parameters defined in this document fall into the following classes:
(1) Parameters always represented in a decimal format.
(2) Parameters always represented in a hexadecimal format.
(3) Parameters always represented as character strings.
(4) Parameters that can be represented in either decimal or
hexadecimal format.
No prefixes are needed for classes 1 - 3, since the format is fixed.
For class 4, a "0x" prefix shall always be used to differentiate the
hexadecimal from the decimal format.
Kumar & Mostafa Standards Track [Page 5]
RFC 3108 ATM SDP May 2001
For both decimal and hex representations, if the underlying bit field
is smaller or larger than the binary equivalent of the SDP
representation, then leading 0 bits should be added or removed as
needed. Thus, 3 and 0x3 translate into the following five-bit
pattern: 0 0011. The SDP representations 0x12 and 18 translate into
the following five-bit pattern: 1 0010.
Leading 0 digits shall not be used in decimal representations.
Generally, these are also not used in hexadecimal representations.
Exceptions are when an exact number of hex digits is expected, as in
the case of NSAP addresses. Parsers shall not reject leading zeros
in hex values.
Both single-character and multi-character string values are enclosed
in double quotes (i.e., "). By contrast, single quotes (i.e., ') are
used for emphasizing keywords rather than to refer to characters or
strings.
In the text representation of decimal and hex numbers, digits to the
left are more significant than digits to the right.
2.3 Directionality Convention
This section defined the meaning of the terms 'forward' and
'backward' as used in this document. This is specially applicable to
parameters that have a specific direction associated with them.
In this document, 'forward' refers to the direction away from the ATM
node under consideration, while 'backward' refers to the direction
towards the ATM node. This convention must be used in all SDP-based
session descriptions regardless of whether underlying bearer is an
SVC, a dynamically allocated PVC/SPVC or a dynamically allocated CID.
This is regardless of which side originates the service connection.
If ATM SVC or AAL2 Q.2630.1 signaling is used, the directionality
convention is independent of which side originates the SVC or AAL2
connection.
This provides a simple way of identifying the direction in which a
parameter is applicable, in a manner that is independent of the
underlying ATM or AAL2 bearer. This simplicity comes at a price,
described below.
The convention used by all ATM/AAL2 signaling specifications (e.g.,
Q.2931 Section 1.3.3 and Q.2630.1) mandates that forward direction is
from the end initiating setup/establishment via bearer signaling
towards the end receiving the setup/establishment request. The
backward direction is in the opposite direction. In some cases, the
'forward' and 'backward' directions of the ATM signaling convention
Kumar & Mostafa Standards Track [Page 6]
RFC 3108 ATM SDP May 2001
might be the exact opposite of the SDP convention described above,
requiring the media gateway to perform the necessary translation. An
example case in which this is needed is described below.
Consider an SDP description sent by a media gateway controller to the
gateway originating a service-level call. In the backward SVC call
set-up model, this gateway terminates (rather than originates) an SVC
call. The media gateway refers to the traffic descriptor (and hence
the PCR) in the direction away from this gateway as the forward
traffic descriptor and forward PCR. Clearly, this is at odds with
ATM SVC signaling which refers to this very PCR as the backward PCR.
The gateway needs to be able to perform the required swap of
directions. In this example, the media gateway terminating the
service level call (and hence originating the SVC call) does not need
to perform this swap.
Certain parameters within attributes are defined exclusively for the
forward or backward directions. Examples for the forward direction
are the subparameter within the 'aal2sscs3661unassured'
media attribute line, the , and
subparameters within the 'aal2sscs3661assured' media attribute line,
the and subparameters within the 'aal5sscop'
media attribute line, and the parameter within the
'aal2sscs3662' media attribute line. Examples for the backward
direction are the subparameter within the
'aal2sscs3661unassured' media attribute line, the ,
and subparameters within the
'aal2sscs3661assured' media attribute line, the and
subparameters within the 'aal5sscop' media attribute line,
and the parameter within the 'aal2sscs3662' media
attribute line.
2.4 Case convention
As defined in RFC 2327 [1], SDP syntax is case-sensitive. Since
these ATM conventions conform strictly with SDP syntax, they are
case-sensitive. SDP line types (e.g., "c", "m", "o", "a") and fields
in the SDP lines should be built according to the case conventions in
[1] and in this document. It is suggested, but not required, that
SDP parsers for ATM applications be case-tolerant where ignoring case
does not result in ambiguity. Encoding names, which are defined
outside the SDP protocol, are case-insensitive.
Kumar & Mostafa Standards Track [Page 7]
RFC 3108 ATM SDP May 2001
2.5 Use of special characters in SDP parameter values
In general, RFC 2327-conformant string values of SDP parameters [1]
do not include special characters that are neither alphabets nor
digits. An exception is the "/" character used in the value
"RTP/AVP" of transport sub-field of the 'm' line.
String values used in SDP descriptions of ATM connections retain this
convention, while allowing the use of the special character "/" in a
manner commensurate with [1]. In addition, the special characters
"$" and "-" are used in the following manner. A "$" value is a
wildcard that allows the recipient of the SDP description to select
any permitted value of the parameter. A "-" value indicates that it
is not necessary to specify the value of the parameter in the SDP
description because this parameter is irrelevant for this
application, or because its value can be known from another source
such as provisioning, defaults, another protocol, another SDP
descriptor or another part of the same SDP descriptor. If the use of
these special characters is construed as a violation of RFC 2327 [1]
syntax, then reserved string values can be used. The string "CHOOSE"
can be used in lieu of "$". The string "OMIT" can be used in lieu of
"-" for an omitted parameter.
3. Capabilities Provided by SDP conventions
To support the applications listed in section 1, the SDP conventions
in this document provide the following session control capabilities:
* Identification of the underlying bearer network type as ATM.
* Identification by an ATM network element of its own address, in
one of several possible formats. A connection peer can
initiate SVC set-up to this address. A call agent or
connection peer can select an pre-established bearer path to
this address.
* Identification of the ATM bearer connection that is to be bound
to the service-level connection. Depending on the application,
this is either a VCC or a subchannel (identified by a CID)
within a VCC.
* Identification of media type: audio, video, data.
* In AAL1/AAL5 applications, declaration of a set of payload
types that can be bound to the ATM bearer connection. The
encoding names and payload types defined for use in the RTP
context [31] are re-used for AAL1 and AAL5, if applicable.
Kumar & Mostafa Standards Track [Page 8]
RFC 3108 ATM SDP May 2001
* In AAL2 applications, declaration of a set of profiles that can
be bound to the ATM bearer connection. A mechanism for
dynamically defining custom profiles within the SDP session
description is included. This allows the use of custom
profiles for connections that span multi-network interfaces.
* A means of correlating service-level connections with
underlying ATM bearer connections. The backbone network
connection identifier or bnc-id specified in ITU Q.1901 [36]
standardization work is used for this purpose. In order to
provide a common SDP base for applications based on Q.1901 and
SIP/SIP+, the neutral term 'eecid' is used in lieu of 'bnc-id'
in the SDP session descriptor.
* A means of mapping codec types and packetization periods into
service types (voice, voiceband data and facsimile). This is
useful in determining the encoding to use when the connection
is upspeeded in response to modem or facsimile tones.
* A means of describing the adaptation type, QoS class, ATM
transfer capability/service category, broadband bearer class,
traffic parameters, CPS parameters and SSCS parameters related
the underlying bearer connection.
* Means for enabling or describing special functions such as
leaf- initiated-join, anycast and SVC caching.
* For H.323 Annex C applications, a means of specifying the IP
address and port number on which the node will receive RTCP
messages.
* A means of chaining consecutive SDP descriptors so that they
refer to different layers of the same connection.
4. Format of the ATM Session Description
The sequence of lines in the session descriptions in this document
conforms to RFC 2327 [1]. In general, a session description consists
of a session-level part followed by zero or more media-level parts.
ATM session descriptions consist of a session-level part followed by
one or two media-level parts. The only two media applicable are the
ATM bearer medium and RTCP control (where applicable).
The session level part consists of the following lines:
v= (protocol version, zero or one line)
o= (origin, zero or one line)
s= (session name, zero or one line)
Kumar & Mostafa Standards Track [Page 9]
RFC 3108 ATM SDP May 2001
c= (connection information, one line)
b= (bandwidth, zero or more lines)
t= (timestamp, zero or one line)
k= (encryption key, zero or one line)
In ATM session descriptions, there are no media attribute lines in
the session level part. These are present in the media-level parts.
The media-level part for the ATM bearer consists of the following
lines:
m= (media information and transport address, one line)
b= (bandwidth, zero or more lines)
k= (encryption key, zero or more lines)
a= (media attribute, zero or more lines)
The media-level part for RTCP control consists of the following
lines:
m= (media information and transport address, one line)
c= (connection information for control only, one line)
In general, the 'v', 'o', 's', and 't' lines are mandatory. However,
in the Megaco [26] context, these lines have been made optional. The
'o', 's', and 't' lines are omitted in most MGCP [25] applications.
Note that SDP session descriptors for ATM can contain bandwidth (b=)
and encryption key (k=) lines. Like all other lines, these lines
should strictly conform to the SDP standard [1].
The bandwidth (b=) line is not necessarily redundant in the ATM
context since, in some applications, it can be used to convey
application-level information which does not map directly into the
atmTrfcDesc media attribute line. For instance, the 'b' line can be
used in SDP descriptors in RTSP commands to describe content
bandwidth.
The encryption key line (k=) can be used to indicate an encryption
key for the bearer, and a method to obtain the key. At present, the
encryption of ATM and AAL2 bearers has not been conventionalized,
unlike the encryption of RTP payloads. Nor has the authentication or
encryption of ATM or AAL2 bearer signaling. In the ATM and AAL2
contexts, the term 'bearer' can include 'bearer signaling' as well as
'bearer payloads'.
The order of lines in an ATM session description is exactly in the
RFC 2327-conformant order depicted above. However, there is no order
of the media attribute ('a') lines with respect to other 'a' lines.
Kumar & Mostafa Standards Track [Page 10]
RFC 3108 ATM SDP May 2001
The SDP protocol version for session descriptions using these
conventions is 0. In conformance with standard SDP, it is strongly
recommended that the 'v' line be included at the beginning of each
SDP session description. In some contexts such as Megaco, the
'v' line is optional and may be omitted unless several session
descriptions are provided in sequence, in which case the 'v' line
serves as a delimiter. Depending on the application, sequences of
session descriptions might refer to:
- Different connections or sessions.
- Alternate ways of realizing the same connection or session.
- Different layers of the same session (section 5.6.4.1).
The 'o', 's' and 't' lines are included for strict conformance with
RFC 2327. It is possible that these lines might not carry useful
information in some ATM-based applications. Therefore, some
applications might omit these lines, although it is recommended that
they not do so. For maximum interoperability, it is preferable that
SDP parsers not reject session descriptions that do not contain these
lines.
5. Structure of the Session Description Lines
5.1 The Origin Line
The origin line for an ATM-based session is structured as follows:
o=
The is set to "-".
The can be set to one of the following:
* an NTP timestamp referring to the moment when the SDP session
descriptor was created.
* a Call ID, connection ID or context ID that uniquely identifies
the session within the scope of the ATM node. Since calls can
comprise multiple connections (sessions), call IDs are
generally not suitable for this purpose.
NTP time stamps can be represented as decimal or hex integers. The
part of the NTP timestamp that refers to an integer number of seconds
is sufficient. This is a 32-bit field
On the other hand, call IDs, connection IDs and context IDs can be
can be 32 hex digits long.
Kumar & Mostafa Standards Track [Page 11]
RFC 3108 ATM SDP May 2001
The field is represented as a decimal or hex number of up
to 32 digits. A "0x" prefix is used before the hex representation.
The refers to the version of the SDP session descriptor
(not that of the SDP protocol). This is can be set to one of the
following:
* 0.
* an NTP timestamp referring to the moment when the SDP session
descriptor was modified. If the SDP session descriptor has not
been modified by an intermediate entity (such as an MGC), then
the timestamp will be the same as the
timestamp, if any. As with the , only the integer
part of the NTP timestamp is used.
When equated to the integer part of an NTP timestamp, the
field is 10 digits wide. This is more restricted than [1], which
allows unlimited size. As in [1], the most significant digit is
non-zero when an NTP timestamp is used.
The in SDP session descriptions for ATM applications
should be assigned the string value "ATM" or wildcarded to a "$" or
"-".
The and parameters are identical to those
for the connection information ('c') line (Section 5.3). Each of
these parameters can be wildcarded per the conventions described for
the 'c' line in Section 5.3. These parameters should not me omitted
since this would violate SDP syntax [1].
As with the 'c' line, SDP parsers are not expected to check the
consistency of with , pairs.
The and need to be consistent with each
other.
5.2 The Session Name Line
In general, the session name line is structured as follows:
s=
For ATM-based sessions, the parameter is set to a "-".
The resulting line is:
s=-
Kumar & Mostafa Standards Track [Page 12]
RFC 3108 ATM SDP May 2001
5.3 The Connection Information Line
In general, the connection information line [1] is structured as
follows:
c=
For ATM networks, additional values of ,
and are defined, over and above those listed in [1]. The
ABNF syntax (Section 9) for ATM SDP does not limit the ways in which
can be combined with , pairs.
However, some combinations will not be valid in certain applications,
while others will never be valid. Invalid combinations should be
rejected by application-specific functions, and not by generic
parsers. The ABNF syntax does limit the ways in which
and can be paired.
For ATM networks, the value of should be set to "ATM".
Further, this may be wildcarded to "$" or "-". If this is done, an
node using ATM as the basic transport mechanism will select a value
of "ATM". A node that interfaces with multiple network types ("IN",
"ATM" etc.) that include ATM can also choose a value of "ATM".
When the SDP description is built by a node such as a media gateway,
the refers to the address of the node building the SDP
description. When this description is forwarded to another node, it
still contains the original node's address. When the media gateway
controller builds part or all of the SDP description, the local
descriptor contains the address of the local node, while the remote
descriptor contains the address of the remote node. If the
and/or are irrelevant or are known by other means, they
can be set to a "$" or a "-", as described below.
Additionally, in all contexts, the 'm' line can have an ATM address
in the subparameter which, if present, is the
remote address if the 'c' line address is local, and vice versa.
For ATM networks, the can be NSAP, E164 or GWID
(ALIAS). For ATM networks, the syntax depends on the
syntax of the . SDP parsers should check the
consistency of with .
NSAP: If the addressType is NSAP, the address is expressed in the
standard dotted hex form. This is a string of 40 hex digits, with
dots after the 2nd, 6th, 10th, 14th, 18th, 22nd, 26th, 30th, 34th and
38th digits. The last octet of the NSAP address is the 'selector'
field that is available for non-standard use. An example of a line
with an NSAP address is:
Kumar & Mostafa Standards Track [Page 13]
RFC 3108 ATM SDP May 2001
c=ATM NSAP 47.0091.8100.0000.0060.3e64.fd01.0060.3e64.fd01.00
A "0x" prefix shall not be used in this case since this is always in
hexadecimal format.
E164: If the addressType is E164, the address is expressed as a
decimal number with up to 15 digits. For example:
c=ATM E164 9738294382
The use of E.164 numbers in the B-ISDN context is defined in ITU
E.191. There is a disparity between the ATM forum and the ITU in the
use of E.164 numbers for ATM addressing. The ATM forum (e.g., UNI
Signaling 4.0) allows only International Format E.164 numbers, while
the ITU (e.g., Q.2931) allows private numbering plans. Since the
goal of this SDP specification is to interoperate with all bearer
signaling protocols, it allows the use of numbers that do not conform
to the E.164 International Format. However, to maximize overall
consistency, network administrators can restrict the provisioning of
numbers to the E.164 International Format.
GWID (ALIAS): If the addressType is GWID, it means that the address
is a Gateway Identifier or Node Alias. This may or may not be
globally unique. In this format, the address is expressed as an
alphanumeric string ("A"-"Z", "a"-"z", "0" - "9",".","-","_"). For
example:
c=ATM GWID officeABCmgx101vism12
Since these SDP conventions can be used for more than gateways, the
string "ALIAS" can be used instead of "GWID" in the 'c' line. Thus,
the example above is equivalent to:
c=ATM ALIAS officeABCmgx101vism12
An example of a GWID (ALIAS)is the CLLI code used for telecom
equipment. For all practical purposes, it should be adequate for the
GWID (ALIAS) to be a variable length string with a maximum size of 32
characters.
The connection information line is always present in an SDP session
descriptor. However, each of the parameters on this line can be
wildcarded to a "$" or a "-", independently of whether other
parameters on this line are wildcarded or not. Not all syntactically
legal wildcard combinations are meaningful in a particular
application.
Kumar & Mostafa Standards Track [Page 14]
RFC 3108 ATM SDP May 2001
Examples of meaningful wildcard combinations in the ATM context are:
c=- - -
c=$ $ $
c=ATM - -
c=ATM $ $
c=ATM -
c=ATM $
Specifying the ATM address type without specifying the ATM address is
useful when the recipient is asked to select an ATM address of a
certain type (NSAP, E.164 etc.).
Examples of syntactically legal wildcard combinations of dubious
utility are:
c=- $ -
c=- $ $
c=- -
c=$ $
c=-
c=$
Note that and/or should not omitted without
being set to a "-" or "$" since this would violate basic SDP syntax
[1].
5.4 The Timestamp Line
The timestamp line for an SDP session descriptor is structured as
follows:
t=
Per Ref. [49], NTP time stamps use a 32 bit unsigned representation
of seconds, and a 32 bit unsigned representation of fractional
seconds. For ATM-based sessions, the parameter can be
made equal to the NTP timestamp referring to the moment when the SDP
session descriptor was created. It can also be set to 0 indicating
its irrelevance. If it made equal to the NTP timestamp in seconds,
the fractional part of the NTP timestamp is omitted. When equated to
the integer part of an NTP timestamp, the field is 10
digits wide. This is more restricted than [1], which allows
unlimited size. As in [1], the most significant digit is non-zero
when an NTP timestamp is used.
The parameter is set to 0 for ATM-based SDP descriptors.
Kumar & Mostafa Standards Track [Page 15]
RFC 3108 ATM SDP May 2001
5.5 Media Information Line for ATM connections
The general format of the media information line adapted for AAL1 and
AAL5 applications is:
m=
The general format of the media information line adapted for AAL2
applications is:
m=
...
Note that is equivalent to in [1].
The subparameter can take on all the values defined in [1].
These are: "audio", "video", "application", "data" and "control".
When the parameter has more than one value in the 'm'
line, the pairs can be arranged in
preferential order.
5.5.1 The Virtual Connection ID
In applications in which the media-level part of a session descriptor
is bound to an ATM virtual circuit, the can be
in one of the following formats:
*
* -/
* /
* /
* //
* //
* /
* -//
* //
In applications in which the media-level part of a session descriptor
is bound to a subchannel within an ATM virtual circuit, the
can be in one of the following formats:
* /
* -//
* //
* //
* ///
* ///
Kumar & Mostafa Standards Track [Page 16]
RFC 3108 ATM SDP May 2001
* //
* -///
* ///
Here,
= VCCI-
= VPCI-
= BCG-
= PORT-
= VPI-
= VCI-
= CID-
The , , , and are decimal numbers or
hexadecimal numbers. An "0x" prefix is used before their values when
they are in the hex format.
The is always a hexadecimal number. An "0x" prefix is not
used with it.
The and are identical to their definitions
above for the connection information line with the difference that
this address refers to the remote peer in the media information line.
Since the , as defined here, is meant for use in
ATM networks, the values of and in the
are limited to ATM-specific values.
The , and are the Virtual Path Identifier, Virtual
Circuit Identifier and Channel Identifier respectively. The is
an 8 or 12 bit field. The is a 16-bit field. The is an
8-bit field ([8] and [11]). For AAL1 applications, it corresponds to
the channel number defined in Annex C of [8].
The is a 16-bit field defined in Section 4.5.16 of ITU Q.2931
[Ref. 15]. The is similar to the , except for its width
and the fact that it retains its value across VP crossconnects. In
some applications, the size of the is the same as the size of
the (8 or 12 bits). In this case, the most significant 8 or 4
bits are ignored.
The is a 16-bit field defined in ITU Recommendation Q.2941.2
[32]. The is similar to the , except for the fact that
it retains its value across VC crossconnects.
In general, and values are unique between a pair of
nodes. When they are unique between a pair of nodes but not unique
within a network, they need to be qualified, at any node, by the ATM
Kumar & Mostafa Standards Track [Page 17]
RFC 3108 ATM SDP May 2001
address of the remote node. These parameters can be pre-provisioned
or signaled. When signaled, the is encapsulated in the
connection identifier information element of SVC signaling messages.
The is encapsulated in the Generic Information Transport (GIT)
information element of SVC signaling messages. In an ATM node pair,
either node can assign values and signal it to the other end
via SVC signaling. A glare avoidance scheme is defined in [32] and
[44]. This mechanism works in SVC applications. A different glare
avoidance technique is needed when a pool of existing PVCs/SPVCs is
dynamically assigned to calls. One such scheme for glare reduction
is the assignment of values from different ends of the
range, using the lowest or highest available value as applicable.
When and values are pre-provisioned, administrations
have the option of provisioning them uniquely in a network. In this
case, the ATM address of the far end is not needed to qualify these
parameters.
In the AAL2 context, the definition of a VCC implies that there is no
CID-level switching between its ends. If either end can assign
values, then a glare reduction mechanism is needed. One such scheme
for glare reduction is the assignment of values from different
ends of the range, using the lowest or highest available value
as applicable.
The parameter is used to identify the physical trunk port on
an ATM module. It can be represented as a hexadecimal number of up
to 32 hex digits.
In some applications, it is meaningful to bundle a set of connections
between a pair of ATM nodes into a bearer connection group. The
subparameter is an eight bit field that allows the bundling of
up to 255 VPCs or VCCs.
In some applications, it is necessary to wildcard the
parameter, or some elements of this parameter.
The "$" wildcard character can be substituted for the entire
parameter, or some of its terms. In the latter
case, the constant strings that qualify the terms in the
are retained. The concatenation
- can be wildcarded in the following ways:
* The entire concatenation, -, is replaced
with a "$".
* is replaced with a "$", but is not.
Kumar & Mostafa Standards Track [Page 18]
RFC 3108 ATM SDP May 2001
Examples of wildcarding the in the AAL1 and
AAL5 contexts are: $, VCCI-$, BCG-100/VPI-20/VCI-$. Examples of
wildcarding the in the AAL2 context are: $,
VCCI-40/CID-$, BCG-100/VPI-20/VCI-120/CID-$, NSAP-$/VCCI-$/CID-$,
$/VCCI-$/CID-$.
It is also permissible to set the entire
parameter to a "-" indicating its irrelevance.
5.5.2 The Transport Parameter
The parameter indicates the method used to encapsulate
the service payload. These methods are not defined in this document,
which refers to existing ATMF and ITU-T standards, which, in turn,
might refer to other standards. For ATM applications, the following
values are defined:
Kumar & Mostafa Standards Track [Page 19]
RFC 3108 ATM SDP May 2001
Table 1: List of Transport Parameter values used in SDP in the ATM
context
+---------------------------------------------------------------------+
| | Controlling Document for |
| Transport | Encapsulation of Service Payload |
+------------------------+--------------------------------------------+
| AAL1/ATMF | af-vtoa-0078.000 [7] |
+------------------------+--------------------------------------------+
| AAL1/ITU | ITU-T H.222.1 [51] |
+------------------------+--------------------------------------------+
| AAL5/ATMF | af-vtoa-0083.000 [46] |
+------------------------+--------------------------------------------+
| AAL5/ITU | ITU-T H.222.1 [51] |
+------------------------+--------------------------------------------+
| AAL2/ATMF | af-vtoa-0113.000 [44] and |
| | af-vmoa-0145.000 [52] |
+------------------------+--------------------------------------------+
| AAL2/ITU | ITU-T I.366.2 [13] |
+------------------------+--------------------------------------------+
| AAL1/custom | Corporate document or |
| AAL2/custom | application-specific interoperability |
| AAL5/custom | statement. |
+------------------------+--------------------------------------------+
| AAL1/ | |
| AAL2/ | |
| AAL5/ | |
| AAL1/IEEE: | Corporate document |
| AAL2/IEEE: | |
| AAL5/IEEE: | |
+------------------------+--------------------------------------------+
| RTP/AVP | Annex C of H.323 [45] |
+------------------------+--------------------------------------------+
In H.323 Annex C applications [45], the parameter has a
value of "RTP/AVP". This is because these applications use the RTP
protocol [2] and audio/video profile [3]. The fact that RTP is
carried directly over AAL5 per [45] can be indicated explicitly via
the aalApp media attribute.
A value of "AAL1/custom", "AAL2/custom" or "AAL5/custom" for the
parameter can indicate non-standard or semi-standard
encapsulation schemes defined by a corporation or a multi-vendor
agreement. Since there is no standard administration of this
convention, care should be taken to preclude inconsistencies within
the scope of a deployment.
Kumar & Mostafa Standards Track [Page 20]
RFC 3108 ATM SDP May 2001
The use of values "AAL1/",
"AAL2/", "AAL5/", "AAL1/IEEE:",
"AAL2/IEEE:" and "AAL5/IEEE:" is similar. These indicate
non-standard transport mechanisms or AAL2 profiles which should be
used consistently within the scope of an application or deployment.
The parameter is the registered, globally unique name
of a corporation (e.g., Cisco, Telcordia etc.). The parameter
is the hex representation of a three-octet field identical to the OUI
maintained by the IEEE. Since this is always represented in hex, the
"0x" prefix shall not be used. Leading zeros can be omitted. For
example, "IEEE:00000C" and "IEEE:C" both refer to Cisco Systems, Inc.
5.5.3 The Format List for AAL1 and AAL5 applications
In the AAL1 and AAL5 contexts, the is a list of payload
types:
...
In most AAL1 and AAL5 applications, the ordering of payload types
implies a preference (preferred payload types before less favored
ones). The payload type can be statically assigned or dynamically
mapped. Although the transport is not the same, SDP in the ATM
context leverages the encoding names and payload types registered
with IANA [31] for RTP. Encoding names not listed in [31] use a "X-"
prefix. Encodings that are not statically mapped to payload types in
[31] are to be dynamically mapped at the time of connection
establishment to payload types in the decimal range 96-127. The SDP
'atmmap' attribute (similar to 'rtpmap') is used for this purpose.
In addition to listing the IANA-registered encoding names and payload
types found in [31], Table 2 defines a few non-standard encoding
names(with "X-" prefixes).
5.5.4 The Format List for AAL2 applications
In the AAL2 context, the is a list of AAL2 profile
types:
...
In most applications, the ordering of profiles implies a preference
(preferred profiles before less favored ones). The
parameter is expressed as a decimal number in the range 1-255.
Kumar & Mostafa Standards Track [Page 21]
RFC 3108 ATM SDP May 2001
5.5.5 Media information line construction
Using the parameter definitions above, the 'm' for AAL1-based audio
media can be constructed as follows:
m=audio AAL1/ATMF
...
Note that only those payload types, whether statically mapped or
dynamically assigned, that are consistent with af-vtoa-78 [7] can be
used in this construction.
Backwards compatibility note: The transport value "AAL1/AVP" used in
previous versions of this document should be considered equivalent to
the value "AAL1/ATMF" defined above. "AAL1/AVP" is unsuitable
because the AVP profile is closely tied to RTP.
An example 'm' line use for audio media over AAL1 is:
m=audio VCCI-27 AAL1/ATMF 0
This indicates the use of an AAL1 VCC with VCCI=24 to carry PCMU
audio that is encapsulated according to ATMF's af-vtoa-78 [7].
Another example of the use of the 'm' line use for audio media over
AAL1 is:
m=audio $ AAL1/ATMF 0 8
This indicates that any AAL1 VCC may be used. If it exists already,
then its selection is subject to glare rules. The audio media on
this VCC is encapsulated according to ATMF's af-vtoa-78 [7]. The
encodings to be used are either PCMU or PCMA, in preferential order.
The 'm' for AAL5-based audio media can be constructed as follows:
m=audio AAL5/ATMF
...
An example 'm' line use for audio media over AAL5 is:
m=audio PORT-2/VPI-6/$ AAL5/ITU 9 15
implies that any VCI on VPI= 6 of trunk port #2 may be used. The
identities of the terms in the virtual connection ID are implicit in
the application context. The audio media on this VCC is encapsulated
according to ITU-T H.222.1 [51]. The encodings to be used are either
ITU-T G.722 or ITU-T G.728 (LD-CELP), in preferential order.
Kumar & Mostafa Standards Track [Page 22]
RFC 3108 ATM SDP May 2001
The 'm' for AAL5-based H.323 Annex C audio [45] can be constructed as
follows:
m=audio