hash
stringlengths
32
32
doc_id
stringlengths
7
13
section
stringlengths
3
121
content
stringlengths
0
3.82M
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.3 Announcement procedures
The PES application server shall have the capability of requesting announcements during any phase of a session, according to the procedures described in TS 183 028 [6]. The PES application server shall use the ANNC URL syntax defined in [16] when requesting a PES announcement server to play an announcement. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 18 A PES application server acting as a B2BUA must appropriately handle received Call-Info, Alert-Info, and Error-Info headers that contain requests to play announcements, as defined in TS 183 028 [6]. Depending on the application, this may mean relaying the header toward the originating PES endpoint or PES access point, connecting the originating UE to an MRF, or ignoring the announcement request and providing some other call processing (such as with a simultaneous ringing application that ignores failure announcements from unsuccessful call legs).
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.4 Dial Tone Management
The PES application server shall act as a profile delivery server and manage Dial Tone Management documents as described in annex A. Changes to these documents shall be notified by reporting the "ua-profile" event defined in [15]. How the PES server keeps track of changes to these documents is outside the scope of the present document.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.5 Transport of ISUP information
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.5.1 General
Clause 5.3.3.5 describes the general behaviour of the PES application server in the case the PES application server is capable of exchanging ISUP information, as defined in ITU-T Recommendation Q.763 [30], carried within Narrowband Signalling Syntax (NSS) [32] message bodies with its SIP signalling peers to perform service related signalling that is capable of interworking with PSTN services. Clause C.1.4 lists examples of services some of which may require the use of encapsulated ISUP information. The sending or receiving of ISUP information in NSS message bodies is applicable to any of the following roles of the Application Server: • PES application server acting as terminating UA or redirect server. • PES application server acting as originating UA. • PES application server performing 3rd party call control. The procedures defined in the following clauses allow a PES application server to discover whether or not its peer SIP signalling entity (i.e. the next UA along the signalling path, which may be a B2BUA, and therefore excluding any intermediate proxies) is capable of supporting NSS message bodies within SIP messages, and to successfully support the additional exchange of ISUP information needed for those services supported in common by endpoints supporting potentially different sets of services. Each IMS PES must assure that NSS message bodies are not forwarded to UEs and are not forwarded to untrusted SIP networks according to local policy. Clause 5.3.3.5.2.4 describes the role of the PES application server in maintaining NSS security. NOTE: Services that require manipulation of the ISUP information within NSS message bodies cannot be implemented on application servers acting as a SIP proxy. The PES application server shall support parameter translations and defaults defined in ES 283 027 [33] for the ISUP information needed for supported services. The PES application server shall also send each piece of ISUP information in the appropriate SIP message so as to allow the valid interworking with the corresponding PSTN services according to ES 283 027 [33]. The PES application server shall not include in NSS message bodies any ISUP information that is already represented by SIP signalling. The handling of forked requests with NSS message bodies is for further study.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.5.2 Sending NSS message bodies to a peer SIP signalling entity
5.3.3.5.2.1 General The PES application server shall send an NSS message body with encapsulated ISUP information in the appropriate SIP message to a peer SIP signalling entity according to the clauses 5.3.3.5.2.2 through 5.3.3.5.2.7, when all the following conditions are satisfied. • The PES application server is capable of performing service related signalling using ISUP information for one or more of the services it supports. • An event associated with one of the supported services occurs that requires that ISUP information be sent to the peer SIP signalling entity. This event may be the receipt of signalling information from another SIP interface to the PES application server, or a service event internal to the PES application server. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 19 • Either the encapsulating message is the initial INVITE request, or the peer SIP signalling entity has previously indicated support for NSS message bodies within the associated dialog (see clause 5.3.3.5.2.2). • The SIP signalling in the encapsulating message cannot represent the information, and the information is different from the default values defined in ES 283 027 [33]. 5.3.3.5.2.2 Determining support for NSS message bodies When constructing an initial INVITE request, an originating PES application server that supports any services that require encapsulated ISUP information shall include an indication of support for NSS by including an Accept header in the initial INVITE request that indicates support for NSS ("application/nss") according to clause 5.3.3.5.2.3. Until the originating PES application server receives an indication of support for NSS message bodies from its peer SIP signalling entity (by receipt of a SIP message that either includes an Accept header indicating support for NSS or an NSS message body), the originating PES application server shall not send any further NSS message bodies within the dialog. If a terminating PES application server receives an initial INVITE request that does not include an Accept header indicating support for NSS, the terminating PES application server shall not send NSS message bodies within the dialog. If a terminating PES application server supporting any services that require the signalling of ISUP information receives an initial INVITE request that includes an Accept header indicating support for NSS, the terminating PES application server shall indicate support of NSS in the first SIP message to its SIP signalling peer. The terminating PES application server indicates support of NSS using the Accept header if allowed in the SIP message. Otherwise, the terminating PES application server does this by including an NSS message body in the SIP message. In the later case, if there is no need to send ISUP information to the SIP signalling peer in the first SIP message, the terminating PES application server shall send a Generic Parameter List (GPL) NSS message with no parameters. A terminating PES application server not supporting NSS will follow procedures in clause 5.3.3.5.3.1. 5.3.3.5.2.3 NSS message bodies The PES application server shall format the NSS message body according to ITU-T Recommendation Q.1980.1 [32]. The Content-Type header field associated with the NSS message body shall be included as follows: • Content-Type: application/nss. The Content-Disposition header field associated with the NSS message body shall be set in one of the following two ways (see clause 5.3.3.5.2.7): • Content-Disposition: signal; handling = required; or • Content-Disposition: signal; handling = optional. 5.3.3.5.2.4 ISUP information security If network entities in the IMS PES cannot be relied on to provide ISUP information confidentiality and integrity, then the PES application server shall not send NSS message bodies or process received NSS message bodies (other than ignoring or rejecting any NSS message bodies). A PES application server acting as routeing B2BUA shall remove NSS message bodies and indications of NSS support (e.g. an NSS entry in the Accept header) from SIP messages being forwarded towards each UE. If a PES application server is assigned to remove the NSS message bodies from SIP messages being forwarded to a UE, then the PES application server shall be configured within the terminating filter criteria for its UE. 5.3.3.5.2.5 Determining in which message to encapsulate ISUP information If a service logic event coincides with one of the SIP basic call control messages, i.e. INVITE request, re-INVITE request, BYE request, UPDATE request, or responses to these messages, any required ISUP information shall be encapsulated in the corresponding SIP message. If a service logic event requiring the sending of ISUP information occurs that does not coincide with one of the SIP basic call control messages, the PES application server shall send the ISUP information encapsulated in an INFO request or a 183 (Session Progress) response, as described below. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 20 An originating PES application server cannot send an INFO request until it receives a reliable provisional response or final response. A terminating PES application server cannot send an INFO request until a reliable provisional response or final response has been sent by the PES application server and the response has been acknowledged. If service logic requires an NSS message body marked with optional handling (see clause 5.3.3.5.2.7) to be sent before an INFO request can be sent, the terminating PES application server may send the NSS message body in a 183 (Session Progress) reliable response. If the service logic requires an NSS message body marked with required handling to be sent before an INFO request can be sent, the terminating PES application server shall first send a 183 (Session Progress) reliable response without an NSS message body and wait for acknowledgment of the response before sending the NSS message body in the INFO request. 5.3.3.5.2.6 Determining the NSS message identifier code If the ISUP information included in the NSS message body uniquely identifies the ISUP message needed to interwork the ISUP information to an ISUP interface, or if the interworking with ISUP is unambiguously identified by the SIP signalling and/or the encapsulated ISUP information, the PES application server shall include the ISUP information in an NSS GPL message. Otherwise, the PES application server shall include an explicit ISUP message name in the NSS message body. 5.3.3.5.2.7 Determining the content disposition handling 5.3.3.5.2.7.1 Content disposition for the initial INVITE request A PES application server sending an INVITE request with an NSS message body shall mark it for required handling (see clause 5.3.3.5.2.3) if the service logic requires the peer SIP signalling entity to understand the ISUP information. Otherwise, the PES application server shall mark the NSS message body in the initial INVITE request for optional handling. If the peer SIP signalling entity is unable to process an NSS message body marked for required handling in an initial INVITE request, it will reject the INVITE request with a failure response, allowing the originating PES application server, or perhaps a proxy on the path, to optionally retry the request to an alternate destination that may be capable of handling the NSS message body. 5.3.3.5.2.7.2 Content disposition for the INFO request A PES application server sending an INFO request with an NSS message body shall mark it for required handling if the service logic requires the peer SIP signalling entity to understand the ISUP information. Otherwise, the PES application server shall mark the NSS message body in the INFO request for optional handling. If the peer SIP signalling entity rejects an NSS message body in an INFO request by returning a failure response, the PES application server performs the service procedure that applies when unable to signal ISUP information, if such a procedure exists, and continues the call associated with the parent dialog. The PES application server shall release the call if the INFO request fails and there is any ISUP information in the INFO request that requires call release if it cannot be processed or forwarded. 5.3.3.5.2.7.3 Content disposition for other SIP messages If the previous clauses do not apply, the PES application server shall mark the NSS message body in the SIP message for optional handling. NOTE: A PES application server sending an NSS message body in any SIP response message will mark it for optional handling, since the peer SIP signalling entity cannot reject the message. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 21
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.5.3 Receiving an NSS message body from a peer SIP signalling entity
5.3.3.5.3.1 General On receipt of a SIP message containing an NSS message body, a PES application server supporting services that require encapsulated ISUP information shall de-encapsulate the ISUP information from the NSS message body, perform the processing described in the clause 5.3.3.5.3.2, and trigger the relevant service logic. If the PES application server does not receive ISUP information required by the service logic, the service logic defines the PES application server behaviour, e.g. by assuming default values defined by the service logic. 5.3.3.5.3.2 ISUP compatibility procedures A PES application server shall reject with a SIP 603 (Decline) response a SIP request that includes an NSS message body that is marked for required handling, that includes ISUP information that the PES application server does not support, and that either includes no compatibility parameter for the unsupported information, or includes a compatibility parameter for the unsupported information that requires release when the parameter is unsupported. Otherwise, the PES application server shall ignore any unsupported ISUP information. The PES application server may ignore any unsupported ISUP information it receives in an NSS message body marked for optional handling or perform any other behaviour determined by the service logic.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.3.6 Handling of charging information
The PES application server shall be able to process charging information contained in SIP messages received from downstream. The PES application server shall be able to insert charging information in SIP messages sent in the upstream direction, using the mechanism described in TS 183 047 [7].
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.4 PES Announcement Server
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.4.1 General
In addition to the procedures specified in the rest of clause 5.3.4, the PES announcement server shall support the procedures specified in ES 283 003 [4] appropriate to an MRFC.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.4.2 Announcement procedures
The PES announcement server shall support the ANNC URL syntax defined in [16] when providing announcements, prompt & collect functions, and multi-party conferencing.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5 PES Interworking Application
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.1 General
In addition to the procedures specified in the rest of clause 5.3.5, the PES interworking application shall support the procedures specified in ES 283 003 [4] and TS 183 028 [6] appropriate to an MGCF.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.2 Routing procedures
In case of incoming calls from legacy networks, the PES interworking application determines the next hop in IP routing depending on received signalling information, based on configuration data and/or data base look up. The next hop may be an I-CSCF, a BGCF or an IBCF. Based on the destination and local policy rules, the PES interworking application may include ISUP information in the messages sent to the next entity, as described in clause 5.3.5.4. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 22
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.3 Handling of charging information
The PES interworking application shall be able to insert charging information in SIP messages sent in the upstream direction, based on charging information received from the PSTN.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.4 Transport of ISUP information
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.4.1 General
The procedures in the clause 5.3.5.4 allow the PES interworking application to discover whether or not its peer SIP signalling entity is capable of supporting the encapsulation of ISUP information in Narrowband Signalling Syntax (NSS) [32] message bodies within SIP messages, and to successfully support the additional exchange of ISUP information needed for those services supported in common by endpoints supporting potentially different sets of services. These procedures are based on ES 283 027 [33] with the extended capabilities described in the present document. When signalling ISUP information within a SIP dialog, the PES interworking application shall act as either a Type A or Type B exchange depending on the role (e.g. gateway between operators, transit) the PES interworking application is performing for that particular call and the ability to exchange ISUP information during the call. Clause C.1.4 lists examples of services some of them may require the use of encapsulated ISUP information. Each IMS PES must assure that ISUP information is not shared with UEs and untrusted SIP networks according to local policy. Clause 5.3.5.4.2.4 describes the PES interworking application's role in maintaining NSS security. The PES interworking application shall support parameter translations and defaults defined in ES 283 027 [33] for the ISUP information needed for supported services. The PES interworking application shall also send each piece of ISUP information in the appropriate SIP message so as to allow valid interworking with the corresponding PSTN services according to ES 283 027 [33]. The PES interworking application shall not include in NSS message bodies any ISUP information that is already represented by SIP signalling. The handling of forked requests with NSS message bodies is for further study.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.4.2 Sending ISUP information to a peer SIP signalling entity
5.3.5.4.2.1 General The PES interworking application shall send an NSS message body with encapsulated ISUP information in the appropriate SIP message to a peer SIP signalling entity according to the clauses 5.3.5.4.2.2 through 5.3.5.4.2.7, when all the following conditions are satisfied. • Either the encapsulating message is the initial INVITE request, or the peer SIP signalling entity has previously indicated support for NSS message bodies within the associated dialog (see clause 5.3.5.4.2.2). • An event occurs requiring signalling to the peer SIP signalling entity. This event may be the receipt of an ISUP message from the PSTN, or an event internal to the MGCF. • According to local configuration, selected ISUP information shall be encapsulated and sent towards the peer SIP signalling entity. • The SIP signalling in the encapsulating message cannot represent the information, and the information is different from the default values defined in ES 283 027 [33]. 5.3.5.4.2.2 Determining support for NSS message bodies When constructing an initial INVITE request, the O-MGCF supporting NSS message bodies shall include an indication of support for NSS by including an Accept header in the initial INVITE request that indicates support for NSS ("application/nss") according to clause 5.3.5.4.2.3. Until the O-MGCF receives an indication of support for NSS message bodies from its peer SIP signalling entity (by receipt of a SIP message that either includes an Accept header indicating support for NSS or an NSS message body), the O-MGCF shall not send any further NSS message bodies within the dialog. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 23 If an I-MGCF receives an initial INVITE request that does not include an Accept header indicating support for NSS, the I-MGCF shall not send NSS message bodies within the dialog. If an I-MGCF supporting NSS receives an initial INVITE request that includes an Accept header indicating support for NSS, the I-MGCF shall indicate support of NSS in the first SIP message to its SIP signalling peer. The I-MGCF indicates support of NSS using the Accept header if allowed in the SIP message. Otherwise, the I-MGCF does this by including an NSS message body in the SIP message. In the later case, if there is no need to send ISUP information to the SIP signalling peer in the first SIP message, the I-MGCF shall send a generic parameter list (GPL) NSS message with no parameters. 5.3.5.4.2.3 NSS message bodies The PES interworking application shall format the NSS message body according to ITU-T Recommendation Q.1980.1 [32]. The Content-Type header field associated with the NSS message body shall be included as follows: • Content-Type: application/nss. The Content-Disposition header field associated with the NSS message body shall be set in one of the following two ways (see clause 5.3.5.4.2.7): • Content-Disposition: signal; handling = required; or • Content-Disposition: signal; handling = optional. 5.3.5.4.2.4 ISUP information security If network entities in the IMS PES cannot be relied on to provide NSS confidentiality and integrity, then the PES interworking application shall not send NSS message bodies or process received NSS message bodies (other than ignoring or rejecting any NSS message bodies). 5.3.5.4.2.5 Determining in which SIP message to encapsulate ISUP information If an event requiring the sending of ISUP information coincides with one of the SIP basic call control messages, i.e. INVITE request, re-INVITE request, BYE request, UPDATE request, or responses to these messages, any required ISUP information shall be encapsulated in the corresponding SIP message. If an event requiring the sending of ISUP information occurs that does not coincide with one of the SIP basic call control messages, the AS shall send the ISUP information encapsulated in an INFO request or a 183 (Session Progress) response, as described below. Clause 5.3.5.4.4 describes events unrelated to SIP basic call control messages that may require the sending of ISUP information. An O-MGCF cannot send an INFO request until it receives a reliable provisional response or final response. An I-MGCF cannot send an INFO request until a reliable provisional response or final response has been sent by the I-MGCF and the response has been acknowledged. If an event requires an NSS message body marked with optional handling (see clause 5.3.5.4.2.7) to be sent before an INFO request can be sent, the I-MGCF may send the NSS message body in a 183 (Session Progress) reliable response. If an event requires an NSS message body marked with required handling to be sent before an INFO request can be sent, the I-MGCF shall first send a 183 (Session Progress) reliable response without an NSS message body and wait for acknowledgment of the response before sending the NSS message body in the INFO request. 5.3.5.4.2.6 Determining the NSS message identifier code If the ISUP information included in the NSS message body uniquely identify the ISUP message needed to interwork the ISUP information to an ISUP interface, or if the interworking with ISUP is unambiguously identified by the SIP signalling and/or the encapsulated ISUP information, the PES interworking application shall include the ISUP information in an NSS GPL message. Otherwise, the PES interworking application shall include an explicit ISUP message name in the NSS message body. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 24 5.3.5.4.2.7 Determining the content disposition handling 5.3.5.4.2.7.1 Content disposition for the initial INVITE request An O-MGCF sending an initial INVITE request with an NSS message body shall mark it for required handling (see clause 5.3.5.4.2.3) in any of the following cases: • The ISUP preference indicator received from the PSTN is set to "ISUP required all the way". • The parameter compatibility parameter associated with any ISUP parameter in the NSS message body indicates "release call" or "discard message" when pass on is not possible. • As a matter of local policy. Otherwise, the O-MGCF shall mark the NSS message body in the initial INVITE request for optional handling. If the peer SIP signalling entity is unable to process an NSS message body marked for required handling in an initial INVITE request, it will reject the INVITE request with a failure response, allowing the O-MGCF, or perhaps a proxy on the path, to optionally retry the request to an alternate destination that may be capable of handling the NSS message body. 5.3.5.4.2.7.2 Content disposition for the INFO request If an event occurs requiring the sending of an NSS message body, the event does not coincide with one of the SIP basic call control messages (see table 1 in clause 5.3.5.4.4.1), and any failure procedures are defined when unable to pass on ISUP information in the NSS message body to the peer SIP signalling entity, the PES interworking application shall mark the NSS for required handling. For example, the "Interactions with other networks" clause of each relevant supplementary service specification (see table C.1) specifies the PES interworking application procedures when unable to pass on ISUP information. Otherwise, the PES interworking application shall mark the NSS message body for optional handling. Otherwise, the O-MGCF shall mark the NSS message body in the INFO request for optional handling. If the peer SIP signalling entity rejects an NSS message body in an INFO request by returning a failure response, the PES interworking application performs the service procedure for failing to pass on the ISUP information, if such a procedure exists, and continues the call associated with the parent dialog. The PES interworking application shall release the call if the INFO request fails and there is any ISUP information in the INFO request that requires call release if it cannot be processed or forwarded. 5.3.5.4.2.7.3 Content disposition for other SIP messages A PES interworking application sending an NSS message body in any SIP message other than the INVITE request and the INFO request shall mark it for optional handling. NOTE: A PES interworking application sending an NSS message body in any SIP response message will mark it for optional handling, since the peer SIP signalling entity cannot reject the message.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.4.3 Receiving an NSS message body from a peer SIP signalling entity
5.3.5.4.3.1 General On receipt of a SIP message containing an NSS message body, the PES interworking application supporting NSS shall de-encapsulate the ISUP information from the NSS message body, perform the processing described in clauses 5.3.5.4.3.2 and 5.3.5.4.3.3, and pass the ISUP information to the relevant ISUP procedures. 5.3.5.4.3.2 ISUP compatibility procedures (local policy options) A PES interworking application shall reject with a SIP 603 (Decline) response a SIP request that includes an NSS message body that is marked for required handling, that includes ISUP information that the PES interworking application does not support, and that requires release according to ISUP procedures when unsupported. Otherwise, the PES interworking application shall ignore any unsupported ISUP information. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 25 The PES interworking application may ignore any unsupported ISUP information it receives in an NSS message body marked for optional handling or perform any other behaviour determined by the ISUP procedures. 5.3.5.4.3.3 Alignment of SIP signalling and NSS message body contents On receipt of a SIP message containing an NSS message body, the PES interworking application shall use the encapsulated ISUP information in preference to any value determined by interworking procedures and default values defined in ES 283 027 [33].
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.4.4 ISUP messages for special consideration
5.3.5.4.4.1 General Table 1 lists the PES interworking application behaviour upon receipt of ISUP messages that have no counterparts in basic SIP call control messages. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 26 Table 1: ISUP messages for special consideration ISUP message Reference Subsequent address message 5.3.5.4.4.6 Reset circuit 5.3.5.4.4.2 Call progress 5.3.5.4.4.3 (see note 1) Circuit group blocking 5.3.5.4.4.2 Circuit group blocking acknowledgement 5.3.5.4.4.2 Circuit group query (national use) 5.3.5.4.4.2 Circuit group query response (national use) 5.3.5.4.4.2 Group reset 5.3.5.4.4.2 Circuit group reset acknowledgement 5.3.5.4.4.2 Confusion 5.3.5.4.4.2 or 5.3.5.4.4.3 (see note 2) Facility reject 5.3.5.4.4.2 or 5.3.5.4.4.3 (see note 2) User-to-user information 5.3.5.4.4.3 Forward transfer 5.3.5.4.4.3 Subsequent directory number (national use) 5.3.5.4.4.3 Suspend 5.3.5.4.4.3 or 5.3.5.4.4.5 Resume 5.3.5.4.4.3 or 5.3.5.4.4.5 Blocking 5.3.5.4.4.2 Blocking acknowledgement 5.3.5.4.4.2 Continuity check request 5.3.5.4.4.2 Continuity 5.3.5.4.4.2 Unblocking 5.3.5.4.4.2 Unblocking acknowledgement 5.3.5.4.4.2 Unequipped CIC (national use) 5.3.5.4.4.2 Circuit group unblocking 5.3.5.4.4.2 Circuit group unblocking acknowledgement 5.3.5.4.4.2 Charging information (national use) 5.3.5.4.4.3 Facility accepted 5.3.5.4.4.3 Facility request 5.3.5.4.4.3 User part test 5.3.5.4.4.2 User part available 5.3.5.4.4.2 Facility 5.3.5.4.4.3 Network resource management 5.3.5.4.4.3 Identification request 5.3.5.4.4.3 Identification response 5.3.5.4.4.3 Information (national use) 5.3.5.4.4.3 Information request (national use) 5.3.5.4.4.3 Segmentation 5.3.5.4.4.4 Loop back acknowledgment (national use) 5.3.5.4.4.3 Loop prevention 5.3.5.4.4.3 Overload (national use) 5.3.5.4.4.2 Pass-along (national use) 5.3.5.4.4.3 Application transport 5.3.5.4.4.2 or 5.3.5.4.4.3 (see note 2) Pre-release information 5.3.5.4.4.3 Release complete 5.3.5.4.4.2 NOTE 1: This is the default handling of the ISUP information associated with the Call Progress (CPG) message when other clauses in the present document do not apply. The ISUP information associated with a CPG message may be encapsulated in an 18X response, an UPDATE request, a reINVITE request, or an INFO request. NOTE 2: The ISUP information in the these messages is either locally terminated or sent transparently depending on whether it is destined for the PES interworking application or for another exchange. 5.3.5.4.4.2 ISUP side procedures only ISUP information from these messages is not encapsulated within SIP messages since they relate to procedures that are relevant only for the ISUP interface. Typically these messages are related to maintenance of ISUP circuits. If ISUP information associated with these ISUP messages is received within an NSS message body, the ISUP information shall be discarded. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 27 5.3.5.4.4.3 Transparent messages If the PES interworking application is configured to forward ISUP information meant for end-to-end service transparency, the PES interworking application shall send the ISUP information through the SIP network as described in clause 5.3.5.4.2.5. 5.3.5.4.4.4 ISUP segmentation ISUP information from the Segmentation message itself is not encapsulated within SIP. Instead the PES interworking application will reassemble the original message received from the PSTN with its segmented part and encapsulate any relevant information in accordance with other procedures in the present document. 5.3.5.4.4.5 Receipt of network initiated SUS and RES If the I-MGCF is not the controlling exchange for network initiated Suspend procedures and NSS message bodies are supported by the peer SIP signalling entity, then the I-MGCF shall forward ISUP information from received SUS and RES in NSS message bodies. If the I-MGCF is the controlling exchange for network initiated Suspend procedures or is unable to forward SUS information to the peer SIP signalling entity, the I-MGCF shall invoke the procedures described in ITU-T Recommendation Q.764 clause 2.4.1c [31] on the ISUP interface and shall not forward ISUP information from either SUS or RES. 5.3.5.4.4.6 Subsequent address message When receiving overlap signalling on the ISUP interface, the PES interworking application shall include the current values of all encapsulated ISUP information in each INVITE request.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.5.5 Optional SIP/ISUP interworking procedures for PSTN Bridging
If the PES interworking application can determine that a call is for PSTN bridging (i.e. that the call will end up in an ISUP network) then the PES interworking application may, as a network option, use the interworking procedures defined in EN 383 001 [20] for Profile C. The way the PES interworking application detects that a call is for PSTN bridging is outside the scope of the present document (local tables or data base lookups may be used). For PSTN bridging, the PES interworking application may route the SIP INVITE request according to the IMS transit mechanism, or any other mechanism, based on local policy.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.6 PES interconnection application
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.6.1 General
In addition to the procedures specified in the rest of clause 5.3.6, the PES interconnection application shall support the procedures specified in ES 283 003 [4] appropriate to an IBCF.
f35b69ccd2199c2dc10834594190c0f0
183 043
5.3.6.2 Procedures related to NSS message bodies
Depending on local policy, the PES interconnection application in an IMS PES supporting NSS, acting as a B2BUA, may perform any reasonable combination of the following functions related to an NSS message body within a received SIP message, where the message is to be forwarded either into or outside of the IMS PES: • removal of an Accept header list entry for NSS ("application/nss"); • removal of an NSS message body marked for optional handling; • filtering of NSS message body parameters (e.g. removal of ISUP information associated with a service not supported by the network); and ETSI ETSI TS 183 043 V1.2.1 (2009-02) 28 • rejection of a SIP request that includes an NSS message body marked for required handling by returning a SIP 415 (Unsupported Media Type) response. A PES interconnection application shall only forward NSS message bodies to or from trusted networks supporting NSS.
f35b69ccd2199c2dc10834594190c0f0
183 043
6 Protocol using SIP/SDP for PES
f35b69ccd2199c2dc10834594190c0f0
183 043
6.1 Introduction
This clause identifies the functional entities of the IMS-based PES architecture [3] that play a specific role in the provision of PES services with regards to SDP processing in the context of SIP signalling.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.2 Functional Entities
f35b69ccd2199c2dc10834594190c0f0
183 043
6.2.1 User Equipment (UE)
Conventional SIP UEs do not exist in PES (see clause 5.2.1). For the purpose of the PES, the VGW shall implement the role of a PES endpoint as described in clause 6.3.1.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.2.2 Access Gateway Control Function (AGCF)
For the purpose of the PES, the AGCF shall implement the role of the PES access point as described in clause 6.3.2. The AGCF entity encompasses the functionality of a Media Gateway Controller (MGC) and of a SIP User Agent.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.2.3 Application Server (AS)
For the purpose of the PES, the AS shall implement the role of a PES application server, as described in clause 6.3.3.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.2.4 Media Resource Function Controller (MRFC)
For the purpose of the PES, an MRFC shall implement the role of the PES announcement server as described in clause 6.3.4.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3 Roles
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.1 PES Endpoint
The PES endpoint shall support the procedures specified in ES 283 003 [4] appropriate to an IMS UE.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.2 PES Access Point
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.2.1 General
In addition to the procedures specified in the rest of clause 6.3.2, the PES access point shall support the procedures specified in ES 283 003 [4] appropriate to an AGCF.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.2.2 Originating calls
The usage of SDP by a PES access point is the same as its usage by a PES endpoint with the following exception: • Only one media description shall be included (i.e. one m= line). ETSI ETSI TS 183 043 V1.2.1 (2009-02) 29 NOTE: Support of T.38 is outside the scope of TISPAN NGN Release 1. When sending an SDP, the PES access point shall not include the "i=", "u=", "e=", "p=", "r=", and "z=" lines in the SDP, and it shall ignore them when received in the SDP. When generating an INVITE request, the PES access point shall: • populate the SDP with the codecs supported by the associated H.248 media gateway; and • in order to support DTMF, populate the SDP with MIME subtype "telephone-event" as described in RFC 2833 [17], unless ITU-T Recommendation G.711 [23] is the only proposed codec.
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.2.3 Terminating calls
The usage of SDP by the PES access point is the same as its usage by a PES endpoint as defined in ES 283 003 [4], with the following exceptions: • When the PES access point sends a 183 (Session Progress) response with SDP payload, it shall only request confirmation for the result of the resource reservation at the originating endpoint if there are any remaining unfulfilled preconditions. • Any media description whose media type is different from audio shall be ignored. When sending an SDP, the PES access point shall not include the "i=", "u=", "e=", "p=", "r=", and "z=" lines in the SDP, and it shall ignore them when received in the SDP. When receiving an initial INVITE request, the PES access point shall: • Check for a codec that matches the requested SDP, which may include the MIME subtype "telephone-event" as described in RFC 2833 [17]. When the PES access point generates and sends a 183 (Session Progress) response to an initial INVITE request, the PES access point shall: • Set SDP indicating the selected codec, which may include the MIME subtype "telephone-event" as described in RFC 2833 [17].
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.2.4 Modifying SDP within existing dialogues
Once a SIP dialog is confirmed (e.g. after a 200 OK and ACK have been used to acknowledge an INVITE request), the PES access point must be prepared to accept a re-INVITE request that contains a new SDP offer. While a SIP dialog remains unconfirmed (e.g. a 200 OK and ACK have not been used to acknowledge an INVITE request), the PES access point must be prepared to accept an UPDATE request that contains a new SDP offer, following the rules defined in RFC 3311 [29].
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.3 PES Application Server
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.3.1 General
In addition to the procedures specified in the rest of clause 6.3.3, the PES application server shall support the procedures specified in ES 283 003 [4] appropriate to an application server. When acting as a B2BUA, the PES application server must appropriately handle SDP received in responses received from a forking Proxy and must appropriately interwork SDP received in provisional and final INVITE responses, as well as other SIP requests, between the originating UA and terminating UA functions employed on a single call. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 30
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.4 PES Announcement Server
f35b69ccd2199c2dc10834594190c0f0
183 043
6.3.4.1 General
In addition to the procedures specified in clause 6.3.4, the PES announcement server shall support the procedures specified in ES 283 003 [4] appropriate to an MRFC.
f35b69ccd2199c2dc10834594190c0f0
183 043
7 Protocol using H.248 for PES
f35b69ccd2199c2dc10834594190c0f0
183 043
7.1 Introduction
This clause identifies the functional entities of the IMS-based PES architecture [3] that play a specific role in the provision of PES services with regards to H.248 processing.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.2 Functional Entities
f35b69ccd2199c2dc10834594190c0f0
183 043
7.2.1 Access Gateway Control Function (AGCF)
For the purpose of the PES, the AGCF shall implement the role of the PES access point as described in clause 7.3.1. The AGCF entity encompasses the functionality of an H.248 Media Gateway Controller (MGC) and of a SIP User Agent. Within the AGCF, the MGC and SIP UA components are coordinated by a Feature Manager entity whose logical behaviour is described in annex B.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.2.2 Media Gateway Function (MGF)
For the purpose of the PES, the MGF shall implement the role of the PES media gateway as described in clause 7.3.2.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3 Roles
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1 PES Access Point
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.1 General
In addition to the procedures specified in the rest of clause 7.3.1, the PES access point shall support the procedures specified in ES 283 002 [5] appropriate to an MGC.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.2 Registration
Registration is supported using the H.248 Service Change procedure. Registration can be performed on a global basis (all lines connected to the same PES media gateway), on a group basis or on a per line basis. This event is passed to the Feature Manager. When the global basis registration mechanism applies, on receipt of a ServiceChange command with "Root" TerminationID from the PES media gateway reporting the MG state changing event and after processing the event according to normal service change procedures has finished, the MGC in the PES access point shall send a Service-Change primitive to inform the Feature Manager if the ServiceChangeMethod is set to "Restart" with ServiceChangeReasons 901 ("Cold Boot") or 902 ("Warm Boot") or "Forced" or "Graceful". Receipt of a ServiceChange command from the PES media gateway reporting a Terminations state changing event shall not cause the MGC in the PES access point to notify the Feature Manager. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 31 When the per line basis registration mechanism applies, on receipt of a ServiceChange command from the PES media gateway, and after processing the event according to normal service change procedures has finished, the MGC in the PES access point shall send a Service-Change primitive to inform the Feature Manager if the ServiceChangeMethod is set to "Restart" with ServiceChangeReasons 900 ("Service Restored") or 902 ("Warm Boot") or "Forced" or "Graceful" When the MGC in the PES access point detects that the PES media gateway lost communication with the MGC in the PES access point, but it was subsequently restored, since MG state may have changed, the MGC in the PES access point may use the H.248 Audit mechanism to resynchronise its state with the PES media gateways. If the MGC in the PES access point detects that the audited state of the specified Terminations is different from the recorded state, the MGC in the PES access point shall use a Service-Change primitive with proper parameters to inform the Feature Manager about the event. When the MGC in the PES access point detects that the PES media gateway lost the communication with the PES media gateway, the MGC in the PES access point shall use Service-Change primitive with proper parameters to inform the Feature Manager about this event.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.3 Basic Session control procedures for analog lines
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.3.1 Originating side procedures
7.3.1.3.1.1 Call Initiation Step 0 On receipt of a notify command from the PES media gateway reporting the off-hook (al/of) event the PES access point shall: • Derive the user identity associated with the termination reporting the event. • Report the occurrence of the off-hook event to the Feature Manager using a Session-Attempt primitive. • Interact with the Feature Manager to determine which dial tone pattern shall be applied (see clause 5.3.2.2). • Request the PES media gateway to apply the signal that correspond to the selected dial tone (cg/dt or srvt/mwt or xcg/spec) according to the value of the dial-tone-pattern element of the Dial Tone Management document described in annex A. • Request the PES media gateway to detect and report the following events: - the xdd/xce event, according to the digit map in force; - the al/on event. NOTE 1: The PES media gateway terminations may also be preconfigured by the PES access point, using signals and events descriptors embedded in the al/of event descriptor, in such a way that upon detection of the off-hook event, the dial tone is automatically delivered and the digit collection is automatically started. The PES access point replaces the event descriptors in case the dial tone is modified. • Proceed with step 1. Step 1 On receipt of the al/on event, the PES access point shall: • Initiate call clearing procedures. • Report the occurrence of the on-hook event to the feature manager using a Session-Release primitive. On receipt of the notification of the xdd/xce event, the PES access point shall: • If the received digits do not correspond to a valid number but receipt of additional digits may lead to a valid number, request the PES media gateway to detect and report the following events: - the xdd/xce event, according to a subsequent digit map; ETSI ETSI TS 183 043 V1.2.1 (2009-02) 32 - the al/on event. This step may be repeated several times, until a valid number has been collected or the user has abandoned the call attempt by going back on-hook. • If the received digits do not correspond to a valid number and receipt of additional digits cannot lead to a valid number, request the PES media gateway to: - play an appropriate announcement using the an/apf signal; - report the g/sc signal completion event for the an/apf signal; - initiate call clearing. • If the received digits correspond to a complete number, the PES access point shall: - request the PES media gateway to create an H.248 context with the caller's analogue termination and an ephemeral termination; - send a Setup-Request primitive to the Feature Manager. SDP information shall be set according to the information received in the response to the Add command; - request the PES media gateway to detect and report the following events: 1) the al/fl event; 2) the al/on event; - wait for either of these events and for events from the Feature Manager. Proceed to step 2 on the occurrence of any of these events. Step 2 On receipt of the notification of the al/on event, the PES access point shall initiate call clearing procedures. On receipt of the notification of the al/fl event, the PES access point shall notify its Feature Manager using a Feature Request primitive. On receipt of a Session Progress primitive from the Feature Manager without alerting indication, the PES access point shall: • If the primitive contains an early media indication, request the PES media gateway to modify the ephemeral termination properties accordingly so that the calling party can perceive in band information. NOTE 2: The termination mode may be set to send-receive or receive-only depending on the network operator's policy. • If the primitive does not contain an early media indication, ignore the event. • Remain in the same state. On receipt of a Session Progress primitive from the Feature Manager indicating alerting, the PES access point shall: • If the primitive contains an early media indication, request the PES media gateway to modify the ephemeral termination properties accordingly so that the calling party can perceive in band information. NOTE 3: The termination mode may be set to send-receive or receive-only depending on the network operator's policy. • If the primitive does not contain an early media indication, request the PES media gateway to apply the cg/rt signal to the physical termination. • Proceed with step 3. On receipt of an internal Setup-Response primitive reporting a busy condition, the PES access point shall: • Request the PES media gateway to apply the cg/bt signal to the physical termination. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 33 • Initiate call clearing after expiry of an operator's defined local timer. On receipt of an internal Setup-Response primitive reporting that no response has been received from the destination, the PES access point shall: • Initiate Call Clearing. On receipt of an internal Setup-Response primitive reporting answer from the destination, the PES access point shall • Request the PES media gateway to set the mode of the termination to send-receive. • Modify the ephemeral termination's remote descriptor properties in accordance with information received from the remote side. • As an operator's option request the PES media gateway to apply the xal/las signal. • Proceed with step 4. Step 3 On receipt of the al/on event, the PES access point shall: • Initiate call clearing procedures. • Report the occurrence of the on-hook event to the feature manager using a Session-Release primitive. On receipt of the notification of the al/fl event, the PES access point shall notify the feature manager using a Feature Request primitive. On receipt of an internal Setup-Response primitive reporting that no response has been received from the destination, the PES access point shall: • Request the PES media gateway to stop the ring back (cg/rt) signal if application of this signal had been requested before. • Initiate Call Clearing procedures. On receipt of an internal Setup-Response primitive reporting answer from the destination, the PES access point shall • Request the PES media gateway to set the mode of the ephemeral termination to send-receive if not performed previously. • Modify the ephemeral termination's remote descriptor properties in accordance with information received from the remote side. • As an operator's option request the PES media gateway to apply the xal/las signal. • Proceed with step 4. Step 4 On receipt of an internal Session-Release primitive, the PES access point shall initiate call clearing procedures. On receipt of an internal Charging-Indication primitive, the PES access point shall request the PES media gateway to apply one of the signals for sending metering pulses (amet/em or amet/mpb). On receipt of an internal Session-Update primitive, the PES access point shall modify the ephemeral termination properties accordingly. On receipt of the al/on event, the PES access point shall: • initiate call clearing procedures; • report the occurrence of the on-hook event to the feature manager. On receipt of the notification of the al/fl event, the PES access point shall notify the feature manager. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 34 7.3.1.3.1.2 Call Clearing The PES access point shall: • If the xal/las signal was applied on the termination when entering the active phase, request the PES media gateway to apply the xal/nt signal. • Send a Session-Release primitive to the internal SIP UA (via the feature manager), if the call clearing phase has been entered on receipt of an al/on event. • Request the PES media gateway to subtract the ephemeral termination from the current context. • If the calling party is still off-hook, request the PES media gateway to apply the off-hook warning tone (xcg/roh) to the physical termination. When the calling party becomes on-hook, the PES access point shall: - Request the PES media gateway to move the termination back to the Null context (i.e. subtract command). - Report the occurrence of the on-hook event to the feature manager.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.3.2 Terminating side procedures
7.3.1.3.2.1 Call Initiation Step 0 On receipt of a Setup-Request primitive from the internal SIP UA (via the Feature Manager), the PES access point shall: • Check that the required bearer service is compatible with analogue lines, otherwise return a Setup-Response to the Feature Manager, with a reject indication. • Determine which physical termination is associated with the called party. If no termination is associated with the called part, the PES access point shall send a Setup-Response primitive with a failure indication to the internal SIP UA via the Feature Manager. • Check the administrative status of this termination. If the termination is out of service, with the called part, the PES access point shall send a setup-response primitive with a failure indication to the internal SIP UA via the Feature Manager. • Create an H.248 context with an ephemeral termination, based on the session description information received in the Setup-Request. • Check whether it is already engaged in a call. NOTE: As an operator's option, an AuditValue command may be sent to the Media Gateway to verify the status of the termination. If the termination is not engaged in a call, the PES access point shall: • Request the PES media gateway to apply ringing to the physical termination. The signal that is used (alert/ring or andisp/dwa) depends on whether or not information need to be displayed to the terminal. Annex D specifies the mapping between SIP headers and the Display Data Block parameter of the andisp/dwa signal. As directed by the AS (with a SIP Alert-Info header in the INVITE request), the PES access point may direct the PES media gateway to apply a distinctive ringing tone. • Send a Session-Progress (Alerting) primitive to the Feature Manager. • Start a no-answer timer. • Request the PES media gateway to detect the off-hook event (al/of). • Proceed to step 1. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 35 If the termination is already engaged in a call, the PES access point shall either (as an operator's option): • Option 1: Send a Setup-Response (busy) primitive to the Feature Manager; or • Option 2: Initiate a call waiting procedure: - Request the PES media gateway to apply the call waiting tone to the physical termination. The signal that is used (cg/cw or andisp/dwa) depends on whether or not information need to be displayed to the terminal. Annex D specifies the mapping between SIP headers and the Display Data Block parameter of the andisp/dwa signal. As directed by the AS (with a SIP Alert-Info header in the INVITE request), the PES access point may direct the PES media gateway to apply a distinctive call waiting tone. - Send a Session-Progress (Alerting) primitive to the Feature Manager. - Start a call-waiting timer. - Proceed to step 1. If a Session-Release primitive is received from the internal UA, via the feature manager, the PES access point shall initiate call clearing procedures. Step 1 If an off-hook (al/of) event is notified, the PES access point shall: • Request the PES media gateway to detect the following events: - the al/fl event; - the al/on event. • Report the occurrence of the off-hook event to the Feature Manager. • Move the physical termination representing the called party's line to the context created during the previous step. • Proceed to step 2. If the no-answer timer (line free) or call-waiting-timer (line busy) expires, the PES access point shall: • Send a Setup-Response (no answer) primitive to the internal UA via the Feature Manager. • Request the PES media gateway to stop the signal sent to the called party (ringing or call waiting signal). • Request the PES media gateway to stop the signal sent to the calling party (ring back tone or the announcement). • Initiate call clearing procedures. Step 2 If the al/fl event is received from the PES media gateway, notify the Feature Manager using a Feature Request primitive. On receipt of an internal Session-Update primitive, the PES access point shall modify the ephemeral termination properties accordingly. If the al/on event is received from the PES media gateway, the PES access point shall: • Initiate call clearing procedures. • Report the occurrence of the on-hook event to the Feature Manager. If a Session-Release primitive is received via the Feature Manager from the internal UA, the PES access point shall initiate call clearing procedures. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 36 7.3.1.3.2.2 Call Clearing If the call clearing phase has been entered on receipt of an al/on event, the PES access point shall: • send a Session-Release primitive to the internal SIP UA; • request the PES media gateway to subtract the ephemeral termination; • request the PES media gateway to move the physical termination back to the Null context. If the call clearing phase has been entered on receipt of an internal Session-Release primitive, the PES access point shall: • request the PES media gateway to subtract the ephemeral termination; • request the PES media gateway to apply the off-hook warning tone (xcg/roh) to the physical termination; • on receipt of the al/on event: - Request the PES media gateway to move the physical termination back to the Null context. - Report the occurrence of the on-hook event to the Feature Manager using a Session-Release primitive.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.4 Procedures for fax/modems calls over analog access
In order to be able to properly handle fax and modem calls, the PES access point requests the media gateway to monitor the ctyp/dtone event. On receipt of a notification of the ctyp/dtone event, the AGCF notifies the Feature Manager if the recognized tone corresponds to a supported fax/modem. Otherwise, the event is discarded.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.1.5 Message Waiting Indication
On receipt of a Service Notification internal primitive reporting the "message-summary" event, the PES access point requests the media gateway to apply the Generic Data Signalling signal of the andisp package defined in ITU-T Recommendation H.248.23 [13]. The value of the TAS parameter is provisioned in the media gateway or in the PES access point, per media gateway or per line. Annex D specifies the populating rules for the Data Block parameter of the Generic Data Signalling signal.
f35b69ccd2199c2dc10834594190c0f0
183 043
7.3.2 PES Media Gateway
The PES media gateway shall support the procedures specified in ES 283 002 [5] appropriate to an MG. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 37 Annex A (normative): XML document structure for Dial Tone Management Dial Tone Management documents are sub-trees or the simservs XML document defined in TS 183 023 [34]. The following schema shall be used to describe XML documents that specify the dial tone pattern applicable to an endpoint. <?xml version="1.0" encoding="UTF-8"?> <xs:schema xmlns:ss="urn:org:etsi:ngn:params:xml:ns:simservs" xmlns:xs="http://www.w3.org/2001/XMLSchema" targetNamespace="urn:org:etsi:ngn:params:xml:ns:simservs" elementFormDefault="qualified" attributeFormDefault="unqualified"> <xs:element name="dial-tone-management" substitutionGroup="ss:absService"> <xs:annotation> <xs:documentation>Dial Tone Management </xs:documentation> </xs:annotation> <xs:complexType> <xs:complexContent> <xs:extension base="ss:simservType"> <xs:sequence> <xs:element name="dial-tone-pattern" default="standard-dial-tone" minOccurs="0"> <xs:simpleType> <xs:restriction base="xs:string"> <xs:enumeration value="standard-dial-tone"/> <xs:enumeration value="special-condition-tone"/> <xs:enumeration value="message-waiting-tone"/> </xs:restriction> </xs:simpleType> </xs:element> </xs:sequence> </xs:extension> </xs:complexContent> </xs:complexType> </xs:element> </xs:schema> ETSI ETSI TS 183 043 V1.2.1 (2009-02) 38 Annex B (normative): AGCF internal communication B.1 General The AGCF entity encompasses the functionality of a Media Gateway Controller (MGC) and of a SIP User Agent. The MGC functionality is involved in registration and session processing. The SIP UA functionality provides the interface to the other components of the IMS-based architecture and the NASS and the RACS, including procedures provided by a P-CSCF on behalf of endpoints. It is involved in registration and session processing as well as in event subscription/notification procedures between the AGCF and application servers. Within the AGCF, the MGC and SIP UA components are coordinated by a Feature Manager entity (see figure B.1). Session Processing Registration Processing Event Subscription /Notification CLF NASS AGCF SIP stack Gq' R-MGF /A-MGF SPDF RACS P1 Mw e2 Feature Manager IMS Agent MGC IMS Agent IMS Agent MGC H.248 Stack Line-based Configuration data I/S-CSCF Figure B.1: AGCF functional model B.2 Internal communication principles Session and registration processing in the AGCF involves two halves: H.248-based MGC processing and SIP User Agent (UA) processing (see figure B.2). MGC processing focuses on the interactions with the media gateway functions, while SIP UA processing focuses on the interactions with the IMS components. The Feature Manager (FM) coordinates the two processing activities. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 39 MGC SIP UA MGF MGF AGCF AGCF RTP FM SIP UA FM MGC S-CSCF S-CSCF AS AS Figure B.2: AGCF session processing model B.3 Internal primitives The communication between the Feature Manager, MGC and SIP User Agent components of the AGCF is modelled using primitives. This modelling is used as an ease to describe the AGCF behaviour and is not intended to constrain implementations. The following primitives are defined: • Service-Change: This primitive is used by the MGC in the PES access point for reporting the ServiceChange event to the Feature Manager. • Register-Request: This primitive is used by the Feature Manager for requesting the SIP User Agent to initiate appropriate SIP registration procedures on behalf of emulation users. • Deregister-Request: This primitive is used by the Feature Manager for requesting the SIP User Agent to initiate appropriate SIP deregistration procedures on behalf of emulation users. • Session-Attempt: This primitive is used to notify the Feature Manager of an outgoing call attempt. • Setup-Request: This primitive is used to request establishment of a session. • Setup-Response: This primitive is used to confirm the establishment of a session. • Session-Update: This primitive is used to update a session description. • Session-Progress: This primitive is used to report intermediate events during session establishment. • Session-Release: This primitive is used to request the release of a session. • Feature-Request: This primitive is used to report the occurrence of a service feature activation request from the user. • Charging-Indication: This primitive is used to report the occurrence of a charging event. • Service-Notification: This primitive is used to report the occurrence of a service notification. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 40 B.4 Feature manager behaviour B.4.1 Registration procedures Based on the information received from the MGC in the PES access point side and local configuration data (mapping between line ids and IMS identities, authentication parameters, etc.) the Feature Manager requests the SIP UA component to initiate appropriate SIP registration procedures (per line registration, "en bloc" registration, etc.). B.4.1.1 Global registration procedures On receipt of a ServiceChange internal primitive indicating "restart", the Feature Manager sends one or more Register-Request primitives to the SIP UA. On receipt of a ServiceChange internal primitive indicating "forced" or "graceful", the Feature Manager sends one or more Deregister-Request primitives to the SIP UA. The number of primitives sent to the SIP UA depends on the AGCF configuration with regards to identity management. The following two cases are identified and lead to different procedures: • One private user identity is assigned to the AGCF (or each MGF) - Each termination connected to the MG is represented by one public user identity. • A pair of private and public user identity is associated with each termination connected to the MGs controlled by the AGCF. In the first case, the Feature Manager sends a single primitive with the private user identity associated with the AGCF (or the MGF) and one of the public user identities. All other public user identities will be implicitly registered or deregistered using implicit registration procedures defined in ES 283 003 [4]. In the second case, the Feature Manager sends one primitive for each termination connected to the MGF that caused the service change event. B.4.1.2 Per line registration procedures There may a lot of terminations to register and deregister at the same time from the MGC side. In order to avoid the impact of the flood registration and deregistration on the IMS core network, the Feature Manager may use some mechanism to control the number of registration/deregistration during a specified period of time. B.4.1.2.1 User-initiated registration On receipt of a Service-Change primitive from the MGC component with ServiceChangeMethod parameter set to "Restart" indicating that service will be restored on the specified Terminations, the Feature Manager shall: • According to the MGF address and the TerminationIDs, lookup the configured data for the public user identities and the private user identities of the related users. • If the related users have not been registered, use a Register-Request primitive to request the SIP UA to initiate appropriate SIP registration procedures. B.4.1.2.2 User-initiated deregistration On receipt of a Service-Change primitive from the MGC component with ServiceChangeMethod parameter set to "Graceful" or "Forced" indicating that the specified Terminations will be taken out of service, the Feature Manager shall: • According to the MGF address and the TerminationIDs, lookup the configured data for the public user identities and the private user identities of the related users. • If the related users have been registered, use a Deregister-Request primitive to request SIP UA to initiate appropriate SIP deregistration procedures. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 41 B.4.1.2.3 Exception procedures On receipt of a Service-Change primitive from the MGC component indicating that the communication between the AGCF and the MGF has been lost, the Feature Manager shall use the Deregister-Request primitive to request the SIP UA to initiate appropriate SIP deregistration procedures for all the registered users belong to the MGF. B.4.2 Call Processing B.4.2.1 General procedures The Feature Manager processes internal events received from the MGC side and request the SIP UA to generate the appropriate SIP messages based on the mapping described in table B.1. Table B.1: Mapping from MGC side to SIP UA side Internal primitive SIP message Setup Request INVITE Session Progress (alerting) 180 Ringing Setup Response (no Answer) 480 (Temporary Not Available) Setup Response (answer) 200 OK Setup Response (busy) 486 Busy Here Setup Response (reject) 606 Not Acceptable Feature Request re-INVITE Release BYE The Feature Manager processes SIP messages received from the SIP UA side and transmit appropriate internal primitives to the MGC side, based on the mapping described in table B.2. Table B.2: Mapping from SIP UA side to MGC side SIP message Internal primitive INVITE Setup Request 183 Session Progress Session Progress 180 Ringing Session Progress(alerting) 200 OK Setup Response (answer) 603 (Decline), 408 (Request Timeout), 480 (Temporary Not Available) Setup Response (no answer) Other 4xx, 5xx, 6xx Setup Response (failure) 486 (Busy Here) , 600 (Busy Everywhere) Setup Response (busy) re-INVITE with SDP, UPDATE Session Update (SDP) REFER Session Update (refer) INFO (charging) or NOTIFY (charging) Charging Indication NOTIFY (other) Service Notification BYE Release B.4.2.2 Flash-Hook Management B.4.2.2.1 General rules Processing flash-hook event notifications depends on the call configuration: Call Configuration: 1-party In this configuration, an unconfirmed INVITE dialog exists between the SIP UA and another endpoint. On receipt of a flash-hook notification from the MGC component, the Feature Manager shall request the MGC component to interact with the media gateway in order to play a dial tone and collect digits. If the digit collection succeeds, the Feature Manager shall request the SIP UA to send an INVITE request to the PES application server, with the collected digits as Request-URI. Otherwise, the events shall be discarded. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 42 Call Configuration : Stable 2-party In this configuration, one confirmed INVITE dialog exists between the SIP UA and another endpoint. On receipt of a flash-hook notification from the MGC component, the Feature Manager shall: • Request the SIP UA to send a re-INVITE request towards the connected party (B party). The re-INVITE request is built as follows: - The Request URI is set to the B party's identity. - The SDP description for the active media stream is set to a=sendonly. NOTE 1: This information may be used by an AS to request the sending of an announcement to the B Party. • Request the MGC component to initiate an outgoing call process as if an off-hook event had been received. Call Configuration : Stable 2-party call with additional held/waiting party In this configuration, a confirmed INVITE dialog exists between the SIP UA and some other endpoint (the "active party", and either an unconfirmed dialog exists with a third endpoint (the "waiting party") or another confirmed dialog exists with a third party that is on hold ("held party"). On receipt of a flash-hook notification from the MGC component, the Feature Manager shall request the MGC component to interact with the media gateway in order to: • Set the stream mode of the ephemeral termination to "inactive". • Play a dial tone. • Collect a feature code. The number of digits is provisioned in the AGCF. The AGCF shall also send a re-INVITE request on the initial dialogue to hold the associated media stream, as described in TS 183 010 [8]. NOTE 2: This information may be used by an AS to request the sending of an announcement to the B Party. Processing of the feature code depends on whether loose or tight coupling procedures are applied between the AGCF and the AS. The decision to use tight coupling or loose coupling is left to the network operator. However, it is important that both the AGCF and the PES application server are configured the same way. B.4.2.2.2 Loose coupling procedures With loose coupling, the AGCF provides call processing logic to manipulate call legs, much like a simulation endpoint would operate. The Feature Manager analyses the feature code according to a local mapping table. For each feature code, this table indicates the user expected feature: • The served user wishes to be connected to a particular held/waiting party and keep the other party held/waiting. • The served user wishes to be connected to a particular held/waiting party and release the other party. • The served user wishes to establish a 3-party conference with both of the other parties. If the feature code received from the internal MGC component does not match any known feature, the FM ignores the feature code and optionally requests the MGC component to interact with the media gateway in order to play an error tone or an announcement to the served user. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 43 If the feature code received from the internal MGC component indicates that the served user wishes to be connected to a particular held/waiting party, the Feature Manager shall: • Request the SIP UA to send either: - A 200 OK response to the INVITE request received from the waiting party if the dialogue is not confirmed yet; or - A re-INVITE request if the dialogue with the held party is already confirmed. The Re-INVITE request is built as follows: The Request URI is set to the held party's identity. The SDP description for the active media stream is set to a=sendrecv. • Request the MGC component to interact with the media gateway in order to: - Modify the Remote Descriptor of the ephemeral termination according to the SDP information received from the held/waiting party. - Set the Stream Mode of the ephemeral termination to sendrecv. - Monitor the flash-hook event on the physical termination. If the feature code received from the internal MGC component indicates that the served user wishes to release the call with the other held/waiting party, the Feature Manager shall: • Request the SIP UA to send a 603 response or a BYE request to the waiting/held party depending on the dialogue state. • Request the MGC component to interact with the media gateway in order to monitor the flash-hook event on the physical termination. If the feature code received from the internal MGC component indicates that the served user wishes to establish a 3-party conference, the Feature Manager shall: • Request the MGC component to interact with the media gateway in order to add a termination to the current context based on SDP information associated with the initial held party (i.e. the party that was already held before the flash-hook has been detected). • Request the SIP UA to send a re-INVITE request towards each of the held/waiting parties. The re-INVITE request is built as follows: - The Request URI is set to the held/waiting party's identity. - The SDP description for the active media stream is set to a=sendrecv. The address and port are set according to the contents of the local descriptor of the termination representing this party on the media gateway. B.4.2.2.3 Tight coupling procedures With tight coupling procedures, all collected digits are reported to the PES application server and the AS determines the appropriate call processing actions to take. Manipulation of call legs for call waiting and 3-party calls is managed by the AS, so feature logic for these services is not required in the AGCF. Unlike loose coupling, where a flash-hook notification must be followed by dialled digits in order for the AGCF to generate an INVITE request, tight coupling procedures support the reporting of a flash-hook event even when no additional digits are dialled by the user. The Feature Manager shall request the SIP UA to create a new dialogue and send an INVITE request to the Application Server, built as follows: • The Request URI is structured as follows: - A user part containing a provisioned prefix followed by collected digits. If no digits were collected following the flash-hook notification, the request URI sent to the AS shall contain a unique user part that signifies that a flash-hook without additional digits was detected (such as flash@pes-scc.operator.com). ETSI ETSI TS 183 043 V1.2.1 (2009-02) 44 - A domain name which together with the user part provides sufficient information to the S-CSCF to forward the INVITE request to the appropriate AS, based on Initial Filter Criteria stored in the user profile (see annex C). • A P-Asserted-Identity containing the public identity of the subscriber requesting the service. • An SDP offer for a voice call. The Feature Manager shall request the MGC component to interact with the media gateway in order to modify the H.248 Remote Descriptor and stream mode according to the SDP information received in re-INVITE messages sent by the Application Server. Release of the dialogue opened for the purpose of sending the feature code is the responsibility of the application server. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 45 Annex C (informative): Implementation of Supplementary Services C.1 General principles C.1.1 Introduction This annex describes guiding principles for implementing commonly deployed PSTN supplementary services using the IMS-based PES architecture. The list of services is taken from EG 201 973-2 [11]. The actual service logic resides in the Application Server and is outside the scope of standardization. This annex focuses on the interactions between the AGCF and PES application servers. Only the part of the service logic which has an impact on signalling to/from the AGCF is described in this annex. Similar procedures may also be used in case of analog lines connected to a VGW acting as a UE with regard to a P-CSCF in the PES. AGCF involvement in the execution of these services is limited to the generic capabilities described in the main body of the present document for supporting interworking between SIP and H.248 protocols and in annex B for processing flash-hook events. C.1.2 Supplementary service control This annex assumes that subscribers can control their supplementary services using service code commands and switching order commands as defined in ETS 300 738 [12]. More than one command can be dialled on a single call. For instance, a caller may dial the "inhibit call waiting" command, followed by the "calling line identity restriction" command, followed by the called party number. C.1.2.1 Service code commands Service codes commands are user requests to perform an action that does not result in a call to another party. Actions such as feature activation, feature deactivation, and feature status inquiries are triggered by service code commands. C.1.2.1.1 Command syntax The format of service code commands as defined in ETS 300 738 [12] is reproduced below: "START PX SC (SR SI) SX" or "PX SC (SR SI) SX FINISH" Where: • START is the start command, e.g. "Off-hook", an alternative to the finish command; • PX is a mandatory service prefix; • SC is a mandatory service code; • SR is one or more separator/s, as required; • SI is one or more units of supplementary information, as required; • SX is a service suffix as required; • FINISH is a finish command, e.g. SEND, an alternative to the start command. NOTE: Digit maps used by the media gateways to collect digits will include appropriate alternatives to cope with the syntax of service code commands. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 46 C.1.2.1.2 Generic procedure at the AGCF side On receipt of a service code command from a media gateway, the AGCF sends an INVITE request to the S-CSCF with the following information: • A Request-URI structured as follows: - A user part containing the service code command, excluding the START and FINISH fields. - A domain name which together with the user part provides sufficient information to the S-CSCF to forward the INVITE request to the appropriate AS, based on Initial Filter Criteria stored in the user profile, e.g. "PX SC (SR SI) SX"@pes-scc.operator.com NOTE 1: If the service code command includes a square "#" symbol, the userinfo portion of the Request-URI will be in the form of a telephone-subscriber. The series of digits that form the service code command will be encoded as a local-number. The phone-context attribute will be set to a domain name of the PES operator, e.g. phonecontext=pes-scc.homedomain.com that is specific enough to enable the application server to interpret the commandecode. Setting the phone-context attribute is required for conformance purposes with RFC 3966 [19]. PES network entities (e.g. CSCF) ignore this attribute. • A P-Asserted-Identity header containing the public user identity of the subscriber issuing the service code command. • An SDP offer for a voice call. NOTE 2: The SDP offer may be used by the Application Server in case an announcement has to be delivered. C.1.2.1.3 Generic procedure at the AS side On receipt of the INVITE request that includes a service code command as Request-URI, the AS sends back a 183 (Session Progress) message. NOTE: An SDP Answer may be included in the 183 (Session Progress) response in case the user is expected to key-in additional digits before the procedure can be completed. In such cases, the AS requests an MRFC/MRFP to play an appropriate prompting announcement and collect digits. The procedure for supporting user interactions during the call establishment phase is described within TS 183 028 [6]. If the procedure identified by the service code command is successfully performed, the AS modifies the subscriber profile according to the service code received, requests an MRFC/MRFP to play a positive acknowledgment tone or announcement and sends a 200 OK response towards the AGCF when the MRFC is connected. If the service code command corresponds to an interrogation procedure, the AS selects the appropriate announcement to notify the requested information (e.g. current supplementary service status) to the calling user. Release of the dialogue occurs when a BYE request is received from the MRFC or if a BYE/CANCEL request is received from the AGCF. If the procedure identified by the service code command cannot be performed successfully, the AS requests an MRFC/MRFP to play a negative acknowledgment tone or announcement and releases the call by sending a suitable 4xx or 5xx response to the INVITE request or by sending a BYE request if the dialogue is already established. The procedure for sending a tone or an announcement during the call establishment phase is described within TS 183 028 [6]. C.1.2.2 Switching order commands Switching order commands are typically used to invoke a service or modify the characteristics of a call, such as a request to invoke the automatic call return service, or a request to inhibit call waiting on the call that is about to be placed. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 47 The format of switching order commands (SOC) is reproduced below: "START SO (SR SI)" or "SO (SR SI) FINISH" Where: • START is a start command, e.g. "Register Recall (R)", an alternative to the finish command, as required; • SO is a mandatory switching order; • SR is one or more separators, as required; • SI is one or more units of supplementary information, as required; • FINISH is a finish command, e.g. "send", an alternative to the start command, as required. Processing of switching order commands where the start command is a Register Recall (also known as Flash-Hook event) is described in annex B. Processing of the Register Recall at the AGCF depends on the call configuration and usually involves requesting the media gateway to deliver a dial tone and collect digits. C.1.3 Setting of initial filter criteria Emulation of PSTN services requires that Initial Filter Criteria stored in the user profile be set on the following methods: • Originating INVITE methods. • Terminating INVITE methods. • Terminating MESSAGE methods. The user profile is selected based on the contents of the P-Asserted-Identity (originating method) or Request-URI (terminating method) headers. How many Initial Filter Criteria are actually required depend on how many application servers are involved in the provision of these services. When several application servers are involved, each implementing a particular service set or feature set, other Service Point Triggers (SPT) may be added to Initial Filter Criteria so as to route the SIP messages to the appropriate server, in the right order. In case of outgoing calls (including special calls for service control purposes), the Request-URI may typically be part of an Initial Filter Criteria. The content tag will typically contain an Extended Regular Expressions (ERE) as defined in clause 9 in IEEE 1003.1 such that any Request-URI that includes a particular pattern (e.g. a domain name) matches the criteria. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 48 The following example illustrates the case of an IFC used to trigger an application server dedicated to the processing of service code commands, assuming that the AGCF appends the pes-scc.homedomain.com domain name to the service code commands. <InitialFilterCriteria> <Priority>0</Priority> <TriggerPoint> <ConditionTypeCNF>0</ConditionTypeCNF> <SPT> <ConditionNegated>0</ConditionNegated> <Group>0</Group> <Method>INVITE</Method> </SPT> <SPT> <ConditionNegated>0</ConditionNegated> <Group>0</Group> <RequestURI>@pes-scc.homedomain.com$</RequestURI> </SPT> </TriggerPoint> <ApplicationServer> <ServerName>sip:PES-AS1@homedomain.com</ServerName> <DefaultHandling>0</DefaultHandling> </ApplicationServer> </InitialFilterCriteria> C.1.4 Supplementary services using ISUP information Full support of supplementary services may be realized by exchanging service information between peer SIP signalling entities via SIP signalling and/or encapsulated ISUP information. The ISUP information necessary to support each individual service is specified by the corresponding ETSI or ITU-T supplementary service specification; see table C.1. Table C.1: Supplementary Service References Supplementary Service ETSI/ITU-T Reference Calling Line Identification Presentation (CLIP) EN 300 356-3 [35] Calling Line Identification Restriction (CLIR) EN 300 356-4 [35] Connected Line Identification Presentation (COLP) EN 300 356-5 [35] Connected Line Identification Restriction (COLR) EN 300 356-6 [35] Terminal Portability (TP) EN 300 356-7 [35] User-to-User Signalling (UUS) EN 300 356-8 [35] Closed User Group (CUG) EN 300 356-9 [35] Subaddressing (SUB) EN 300 356-10 [35] Malicious Call Identification (MCID) EN 300 356-11 [35] Conference Call (CONF) EN 300 356-12 [35] Explicit Call Transfer (ECT) EN 300 356-14 [35] Call Forwarding Busy (CFB) EN 300 356-15 [35]] Call Forwarding No Reply (CFNR) EN 300 356-15 [35] Call Forwarding Unconditional (CFU) EN 300 356-15 [35] Call Deflection (CD) EN 300 356-15 [35] Call Hold (HOLD) EN 300 356-16 [35] Call Waiting (CW) EN 300 356-17 [35] Completion of Calls to Busy Subscriber (CCBS) EN 300 356-18 [35] Three-Party (3PTY) EN 300 356-19 [35] Completion of Calls on No Reply (CCNR) EN 300 356-20 [35] Anonymous Communication Rejection (ACR) EN 301 798 [40]. Multi-Level Precedence and Pre-emption (MLPP) ITU-T Recommendation Q.735.3 [36] Global Virtual Network Service (GVNS) ITU-T Recommendation Q.735.6 [37] Reverse charging (REV) ITU-T Recommendation Q.736.3 [38] ETSI ETSI TS 183 043 V1.2.1 (2009-02) 49 C.2 Advice of Charge C.2.1 Actions at the Originating AGCF The AGCF converts charging pulses received from the Application Server (via the S-CSCF) into appropriate signals of the H.248 amet package. Generation of the Advice Of Charge message (see ES 200 659-3 [10]) by the AGCF is described in annex D. C.2.2 Actions at the Originating AS Implementation of this service assumes that the AS providing the service is involved in all outgoing calls for which advice of charge information is expected (e.g. all outgoing calls, all outgoing international calls, etc.). It is the responsibility of the network operator to configure the appropriate Initial Filter Criteria in the user profile. Charging pulses are transmitted from the Application Server to the AGCF using the method described in TS 183 047 [7]. It is the responsibility of the AS to ensure that the information contained in these messages is in the form of a number of units. C.2.3 Actions at the Terminating AGCF Not applicable. C.2.4 Actions at the Terminating AS Not applicable. C.3 Anonymous Call Rejection C.3.1 Actions at the Originating AGCF This service does not require the originating AGCF to perform any specific action. C.3.2 Actions at the Originating AS This service does not require the originating AS to perform any specific action. C.3.3 Actions at the Terminating AS If the service is activated, the terminating AS rejects the call if the INVITE request includes the P-Asserted-Identity header field and includes the Privacy header field indicating "id", "header", "user" or "critical" as specified in RFC 3325 [25]. In all other cases the communication proceeds normally. NOTE: If the P-Asserted-Identity header field is not present, the call is not rejected. When the AS rejects a communication, the AS sends an indication to the calling user by sending a 433 (Anonymity Disallowed) response. Additionally, before terminating the communication an announcement can be provided to the originating user. The procedure for invoking an announcement in the call establishment phase is described within TS 183 028 [6]. As a service option the ACR service may forward the communication to a voice message service instead of rejecting the communication with a 433 (Anonymity Disallowed) final response. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 50 C.3.4 Actions at the Terminating AGCF The terminating AGCF is not involved in the execution of this service. C.4 Automatic Call Return C.4.1 Actions at the AGCF at the invoker side The Automatic Call Return service is invoked using a service code command. On receipt of this service code command, the AGCF builds an INVITE request as described in clause C.1. C.4.2 Actions at the AS at the invoker side Implementation of this service assumes that the AS providing the service is involved in all incoming calls to the served user. It is the responsibility of the network operator to configure the appropriate Initial Filter Criteria in the user profile. The AS keeps track of the most recent incoming call to each served user it manages. When receiving the service code command that identifies the Automatic Call Return service, the AS acts as a B2BUA and establishes a new call leg to the most recent caller (i.e. it sends an INVITE messages towards this most recent call party as if the number had been dialled by the served user). C.5 Calling Line Identity Presentation / Restriction C.5.1 Actions at the Originating AGCF A service code command may be received by the AGCF in case the calling user wishes to override the default setting for a particular call, in which case the called party number is embedded in the command code as part of the supplementary information. The AGCF generates an INVITE request according to the rules described in clause C.1. Otherwise the originating AGCF is not involved in the provision of this service. C.5.2 Actions at the Originating AS The AS determines whether calling identification is presented or restricted based on user profile data (permanent mode) or based on the contents of the Request-URI (per-call mode). Any service prefix is removed from the Request-URI before forwarding the INVITE request toward the called party. If calling identification restriction is in force, the AS will apply the following changes before forwarding the received INVITE request towards the called party via the S-CSCF: • Override the contents of the From header field with an "anonymous" indication. The convention for configuring an anonymous From header field described in RFC 3323 [24] and RFC 3325 [25] should be followed; i.e. From: "Anonymous" <sip:anonymous@anonymous.invalid>;tag= xxxxxxx. • Include a Privacy header field set to "header" in accordance with RFC 3323 [24] and RFC 3325 [25]. On receipt of an INVITE request with a service code command requesting that the calling line identity be presented to the called party, the AS removes the service prefix from the service code command and forwards the INVITE request unchanged towards the called party, via the S-CSCF. Otherwise the originating AS is not involved in the provision of this service. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 51 C.5.3 Actions at the Terminating AS If the service is subscription dependent and the called user has not subscribed to the service, the terminating AS removes any P-Asserted-Identity and Privacy header fields included in the request. Additionally, the Application Server may as a network option set the contents of the From header to a default non significant value. If the request includes the Privacy header field set to "header" the AS set the contents of all headers containing private information in accordance with RFC 3323 [24] and RFC 3325 [25]. If the request includes the Privacy header field set to "user" the AS removes or set the contents of all "user configurable" headers in accordance with RFC 3323 [24] and RFC 3325 [25]. In the latter case, the AS may need to act as transparent back-to-back user agent as described in RFC 3323 [24]. NOTE: If the request includes the Privacy header field set to "id", the P-Asserted-Identity header is removed by the S-CSCF. C.5.4 Actions at the Terminating AGCF User Identification information received in an INVITE request ("P-Asserted-Id" and "From" headers) is used by the AGCF to generate the appropriate Call Setup message (see ES 200 659-3 [10]) to be delivered to the called terminal, using the H.248 andisp package. Generation of the Call Setup message is further described in annex D. C.6 Calling Name Delivery C.6.1 Actions at the Originating AGCF No specific action is performed by the AGCF. C.6.2 Actions at the Originating Application Server If the calling user has subscribed to the service and no presentation restriction is invoked, the originating AS will insert a pre-configured calling name in the Display field of the From header and possibly the same or different name in the Display field of the P-Asserted-Identity header. C.6.3 Actions at the Terminating Application Server If the service is subscription dependent and the called user has not subscribed to the service, the terminating AS removes or the From header or set its contents in accordance with RFC 3323 [24]. If the request includes the Privacy header field set to "header" the AS sets the contents of all headers containing private information in accordance with RFC 3323 [24] and RFC 3325 [25]. If the request includes the Privacy header field set to "user" the AS removes or sets the contents of all "user configurable" headers in accordance with RFC 3323 [24] and RFC 3325 [25]. In the latter case, the AS may need to act as transparent back-to-back user agent as described in RFC 3323 [24]. C.6.4 Actions at the Terminating AGCF User Name information received in an INVITE request ("From" header /or "P-Asserted-Identity") is used by the AGCF to generate the appropriate Call Setup message (see ES 200 659-3 [10]) to be delivered to the called terminal, using the H.248 andisp package. Generation of the Call Setup message is further described in annex D. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 52 C.7 Call Forwarding C.7.1 Activation/Deactivation/Interrogation C.7.1.1 Actions at the AGCF Registration, Erasure, Activation, Deactivation and Interrogation of call forwarding services is performed using service code commands as described in ETS 300 738 [12]. On receipt of a service code command that corresponds to one of these procedures, the AGCF sends an INVITE request as described in clause C.1. C.7.1.2 Actions at the AS On receipt of the INVITE request that includes a service code command as Request-URI, the AS proceeds as described in clause C.1. In case of unconditional call forwarding the AS may also send a NOTIFY request to the AGCF to update the Dial Tone Management XML document. C.7.2 Invocation C.7.2.1 Actions at the Originating AGCF No specific action is performed by the AGCF. C.7.2.2 Actions at the Originating AS No specific action is performed by the AS. C.7.2.3 Actions at the Forwarding AS The AS implements procedures identical or similar to those described in TS 183 004 [9]. C.7.2.4 Actions at the Forwarding AGCF The AGCF at the forwarding side is not involved in the processing of forwarded calls. C.7.2.5 Actions at the Terminating AS No specific action is performed by the AS. C.7.2.6 Actions at the Terminating AGCF Call forwarding information received in an INVITE request ("history-info" header) is used by the AGCF to generate the appropriate Call Setup message (see ES 200 659-3 [10]) to be delivered to the called terminal, using the H.248 andisp package. Generation of the Call Setup message is further described in annex D. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 53 C.8 Distinctive Ringing C.8.1 Actions at the Originating AGCF This service does not require the originating AGCF to perform any specific action. C.8.2 Actions at the Originating Application Server This service does not require the originating AS to perform any specific action. C.8.3 Actions at the Terminating Application Server The terminating AS inserts an "alert-info" header in the INVITE request with a value selected based on the value of the P-Asserted-Identity header received in the incoming INVITE request and on the called user profile. C.8.4 Actions at the Terminating AGCF The terminating AGCF uses the value of the "alert-info" header to select the appropriate ring pattern. The pattern identifier is sent to the media gateway as a parameter of the H.248 andisp/dwa signal. C.9 Call Waiting C.9.1 General C.9.1.1 Actions at the AGCF at the terminating side On receipt of an INVITE request for a busy subscriber, the AGCF performs the following actions: • Request the media gateway to apply call waiting tone using the H.248 andisp/dwa signal. • Request the media gateway to monitor flash-hook events. • Send a 180 (Alerting) towards the AS. • Start a no answer timer. If the no answer timer expires the AGCF sends a 486 (Busy Here) response towards the Application Server. If a flash-hook event is reported by the media gateway, the AGCF stops the no-answer timer and requests the media gateway to perform the following actions: • Set the stream mode of the ephemeral termination to inactive. • Apply a dial tone. • Collect a switching order command. NOTE 1: In some networks, applying a dial tone and collecting an explicit switching order command is not required as the register recall is interpreted as a request to accept the waiting call. Subsequent AGCF procedures are executed as if an equivalent switching order command had been received from the user. The number of digits is provisioned in the AGCF. The AGCF also sends a re-INVITE request on the initial dialogue to hold the associated media stream, as described in TS 183 010 [8]. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 54 Processing of the switching order command depends on whether loose or tight coupling procedures are applied between the AGCF and the AS. Figure C.1 illustrates the loose coupling case while figure C.2 illustrates the tight-coupling case. NOTE 2: These figures do not take into account all possible service code commands that may be received from the user (e.g. service code command requesting that a waiting call be accepted and the active call be released). C.9.1.2 Actions at the AS at the terminating side Implementation of this service assumes that the AS providing the service is involved in all calls to/from the subscriber. The AS determines whether the called user is busy based on the procedures described in TS 183 028 [6]. If the called user is busy and has not subscribed to the call waiting service, the AS sends a 486 (Busy Here) response towards the calling party. If the called user is busy and has subscribed to the call waiting service, the AS builds an INVITE request using normal rules as for any incoming call. On receipt of a re-INVITE request with a SDP "sendonly" attribute, the AS interacts with an MRFC is order to play an announcement to the held party, in accordance with TS 183 028 [6]. C.9.2 Option 1 (Loose coupling) C.9.2.1 Actions at the AGCF at the terminating side The AGCF evaluates the Switch Order Command based on a provisioned mapping table. If the SOC indicates that the served user wishes the be connected to the waiting party, the AGCF performs the following actions: • Send a 200 OK response to the INVITE request received from the waiting party. • Request the media gateway to: - Modify the Remote Descriptor of the ephemeral termination according to the SDP information received from the waiting party. - Monitor the flash-hook event. If SOC indicates that the served user wishes to reject the waiting call, the AGCF performs the following actions: • Send a provisioned error response code (e.g. 603) to the INVITE request received from the waiting party. • Request the media gateway to: - Set the stream mode to send-receive. - Monitor the flash-hook event. • Send a re-INVITE request towards the held party (i.e. the party that has been held for the purpose of collecting the switching order command). The re-INVITE request is built as follows: - The Request-URI is set to the held party's identity. - The SDP description for the active media stream is set to a=sendrecv. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 55 Once the communication is established with the waiting party, the user may decide to switch back to the initial party, using a Register Recall followed by new switching order command. If the value of the switching order command indicates that the initial party is to be switched back, the AGCF performs the following actions: • Send a re-INVITE request towards the held party. The re-INVITE request is built as follows: - The Request-URI is set to the held party's identity. - The SDP description for the active media stream is set to a=sendrecv. • Send a re-INVITE request towards the active party. The re-INVITE request is built as follows: - The Request URI is set to the active party's identity. - The SDP description for the active media stream is set to a=sendonly. • Request the media gateway to: - Modify the Remote Descriptor of the ephemeral termination according to the SDP information associated with the held party. - Monitor the flash-hook event. The user may also be willing to establish a 3-party conference. On receipt of a Switch Order Command indicating that a conference is to be established, the AGCF performs the following actions: • Request the media gateway to add a termination to the current context based on SDP information associated with the held party. • Send a re-INVITE request towards the first held party (i.e. the party that was already held when the flash-hook event has been detected). The re-INVITE request is built as follows: - The Request URI is set to the held party's identity. - The SDP description is set to a=sendrecv. The address and port are set according to the contents of the local descriptor of the new termination. • Send a re-INVITE request towards the second held party (i.e. the party that has been held for the purpose of collecting the Switch Order Command). The re-INVITE request is built as follows: - The Request URI is set to the held party's identity. - The SDP description is set to a=sendrecv. If the feature code received from the internal MGC component does not match any known feature, the AGCF ignores the feature code and optionally request the media gateway to play an error tone or an announcement to the served user. C.9.2.2 Actions at the AS at the terminating side On receipt of a re-INVITE request with a SDP attribute "sendonly", the AS interacts with an MRFC is order to play an announcement to the held party, in accordance with TS 183 028 [6]. On receipt of a re-INVITE request with the SDP attribute "sendrecv", the AS interacts with the MRFC to stop any ongoing announcement. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 56 MGF AGCF AS TE (1) INVITE (D2, SDP Offer from C) (2) H.248 interaction to apply call waiting tone and monitor flash-hook events (7) NOTIFY (4) Call Waiting Tone (5) Register Recall (11) H.248 interaction to apply dial tone and collect digits (12) Dial Tone (13) Digits (15)Digit(s) detected (16) NOTIFY (17) Evaluate switch order code (19.a) H.248 Remote Descriptor set to SDP-C (6) Flash-hook detected (3) 180 Alerting (D2) (18b) 603 (D2) (18.a) 200 OK (D2) (8) RE-INVITE (D1, SDP a=sendonly) (10) Interaction with held party (B) (e.g. announcement) (9) H.248 Stream Mode set to inactive (20b) H.248 Stream Mode set to send&receive (21b) Stop announcement (19b) RE-INVITE (D1, SDP a=sendrecv) Figure C.1: Call Waiting with loose AGCF/AS coupling C.9.3 Option 2 (Tight coupling) C.9.3.1 Actions at the AGCF at the terminating side The AGCF sends a re-INVITE request to the Application Server, built as follows: • The Request URI is structured as follows: - A user part containing a provisioned prefix followed by the switching order command without the start and finish fields. NOTE: In networks where no explicit SOC is collected, a preconfigured SOC is used to populate the user part. - A domain name that provides sufficient information to the S-CSCF to forward the INVITE request to the appropriate AS, based on Initial Filter Criteria stored in the user profile, e.g. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 57 SOC- "SO (SR SI)"@pes.operator.com - A P-Asserted-Identity containing the public identity of the subscriber issuing the switching control command. - An SDP offer for a voice call. The AGCF modifies the H.248 Remote Descriptor and stream mode according to the SDP information received in re-INVITE messages sent by the Application Server. C.9.3.2 Actions at the AS at the terminating side The AS evaluates the Switch Order Command based on the current call configuration and issues appropriate re-INVITE messages. A CANCEL request is sent on the dialogue associated with the waiting call. The AS interacts with an MRFC in order to play and stop announcements to the held party, in accordance with TS 183 028 [6]. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 58 MG AGCF AS TE (1) INVITE (D2, SDP Offer from C) (2) H.248 interaction to apply call waiting tone and monitor flash-hook events (7) NOTIFY (4) Call Waiting Tone (5) Register Recall (11) H.248 interaction to apply dial tone and collect digits (12) Dial Tone (13) Digits (14)Digit(s) detected (15) NOTIFY (21a) H.248 Remote Descriptor set to SDP-C (6) Flash-hook detected (3) 180 Alerting (D2) (19b) RE-INVITE (D1, a=sendrecv) (16) INVITE (D3, digits@pes) (8) RE-INVITE (D1, SDP a=sendonly) (10) Interaction with held party (B) (e.g. announcement) (9) H.248 Stream Mode set to inactive (20b) H.248 Stream Mode set to send&receive (21b) stop announcement (17) 200 OK (D3) (18) Evaluate switch order code (19a) RE-INVITE (D1, SDP-C) (20a) CANCEL (D2) (22b) reject call Figure C.2: Call Waiting with tight AGCF/AS coupling ETSI ETSI TS 183 043 V1.2.1 (2009-02) 59 C.10 Incoming Call Barring C.10.1 Activation/Deactivation/Interrogation C.10.1.1 Actions at the AGCF Activation, deactivation and interrogation of incoming call barring is performed using service code commands as described in ETS 300 738 [12]. On receipt of a service code command, the AGCF sends an INVITE request as described in clause C.1. C.10.1.2 Actions at the AS On receipt of the INVITE request that includes a service code command as Request-URI, the AS proceeds as described in clause C.1. C.10.2 Invocation C.10.2.1 Actions at the Originating AGCF The originating AGCF is not involved in the invocation of the service. C.10.2.2 Actions at the Originating AS The originating AGCF is not involved in the invocation of the service. C.10.2.3 Actions at the Terminating AS On receipt of an INVITE request from a calling user, if incoming call barring is activated, the AS checks the P-Asserted-Identity against the list of allowed/forbidden calling parties. The rules and information used by the AS to evaluate whether the call is to be accepted are operator dependent. They may be compatible subset of the rules described in TS 183 011 [27]. If the call is rejected, the AS notifies the calling user by sending a 603 (Decline) response Additionally, before terminating the communication an announcement can be provided to the originating user. The procedure for invoking an announcement in the call establishment phase is described within TS 183 028 [6]. C.10.2.4 Actions at the Terminating AGCF The terminating AGCF is not involved in the provision of the service. C.11 Malicious Call Identification C.11.1 Actions at the Originating AGCF The AGCF at the originating side is not involved in the provision of this service. C.11.2 Actions at the Originating AS The AS at the originating side is not involved in the provision of this service. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 60 C.11.3 Actions at the Terminating AS Support of the Malicious Call Identification (MWI) service requires that the AS providing the service is involved in all incoming calls to the served user. On receipt of a service code command requesting invocation of the MCID service, the AS registers the details of the last incoming call in a special record. Processing of this record is outside the scope of standardization. C.11.4 Actions at the Terminating AGCF The MCID service is invoked using a special service code command dialled after the malicious call is released. On receipt of this service code command, the AGCF sends an INVITE request as described in clause C.1. C.12 Message Waiting Indicator C.12.1 Actions at the AGCF On receipt of a NOTIFY request reporting the "message-summary" event, the AGCF requests the following actions from the media gateway: • Modify the default dial tone. • Send a Message Waiting Indicator message (see ES 200 659-3 [10]) using the H.248 andisp/data signal. C.12.2 Actions at the AS The AS implements the procedures described in TS 183 006 [26]. Generation of the Message Waiting Indicator message (see ES 200 659-3 [10]) by the AGCF is described in annex D. C.13 Outgoing Call Barring C.13.1 Activation/Deactivation/Interrogation C.13.1.1 Actions at the AGCF Activation and deactivation of outgoing call barring is performed using service code commands as described in ETS 300 738 [12]. On receipt of a service code command, the AGCF sends an INVITE request as described in clause C.1. C.13.1.2 Actions at the AS On receipt of the INVITE request that includes a service code command as Request-URI, the AS proceeds as described in clause C.1. C.13.2 Invocation C.13.2.1 Actions at the Originating AGCF The originating AGCF is not involved in the invocation of the service. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 61 C.13.2.2 Actions at the Originating AS On receipt of an INVITE request from a calling user, if outgoing call barring is activated, the AS checks the Request-URI against the list of allowed/forbidden called destinations. The rules and information used by the AS to evaluate whether the call is to be accepted are operator dependent. They may be compatible subset of the rules described in TS 183 011 [27]. If the call is rejected, the AS notifies the calling user by sending a 603 (Decline) response Additionally, before terminating the communication an announcement can be provided to the originating user. The procedure for invoking an announcement in the call establishment phase is described within TS 183 028 [6]. C.13.2.3 Actions at the Terminating AS The terminating AS is not involved in the provision of the service. C.13.2.4 Actions at the Terminating AGCF The terminating AGCF is not involved in the provision of the service. C.14 Three party service C.14.1 General C.14.1.1 Actions at the AGCF at the service invocation side The service is invoked during a stable 2 party call (without any waiting/held call) using the Register Recall. Figure C.3 illustrates the message sequence between the AGCF and the AS. On receipt of a NOTIFY request reporting a flash-hook event, the AGCF performs the following actions: • Request the media gateway to play a dial tone and collect digits. • Send a re-INVITE request to place the current call on hold. On receipt of the dialled digits, the AGCF opens a new dialogue by sending an INVITE request with the following elements: • The dialled digits used as a Request-URI. • An SDP Offer for a voice call. And wait for a 183 (Session Progress) with an SDP Answer. On receipt of the 183 (Session Progress) response, the AGCF performs the following actions: • Request the media gateway to modify the Remote Descriptor of the ephemeral termination associated with the physical termination representing the analog line. • Request the media gateway to play a ringback tone, unless a P-Early-Media header is included in the SIP message, as described in TS 183 028 [6]. • Request the media gateway to monitor the flash-hook event. On receipt of a re-INVITE request requesting a call to be placed on hold, the AS applies the procedures described in TS 183 010 [8]. Subsequent AGCF actions are identical to those described for the call waiting service. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 62 C.14.1.2 Actions at the AS at the service invocation side Support of this service requires that all outgoing and incoming calls to the served user be handled by the same AS. On receipt of a service code command requesting a three party conference, the AS may check that this request is compatible with subscription data held in the user profile. If the user is not entitled to use this service, the AS interacts with an MRFC is order to play an announcement to the invoker and sends a negative response to the re-INVITE request. On receipt of a re-INVITE request with an SDP "a=sendonly" attribute, the AS may interact with an MRFC in order to play an announcement to the held party, in accordance with TS 183 028 [6]. MG AGCF AS TE (1) H.248 interaction to monitor flash-hook events (4) NOTIFY (2) Register Recall (8) H.248 interaction to apply dial tone and collect digits (9) Dial Tone (10) Digits (11)Digit(s) detected (12) NOTIFY (17) H.248 Remote Descriptor set to SDP-C (3) Flash-hook detected (13) INVITE (D2, digits@pes) (5) RE-INVITE (D1, SDP a=send-only) (7) Interaction with held party (e.g. announcement) (6) H.248 Stream Mode Set to send-only (14) 183 Session Progress (D2) (15) 180 Alerting (D2) (16) 200 OK (D2, SDP-C) (18) H.248 interaction for monitoring flash-hook events Figure C.3: Initiating a 3-Party call ETSI ETSI TS 183 043 V1.2.1 (2009-02) 63 C.14.2 Option 1 (Loose coupling) C.14.2.1 Actions at the AGCF at the invoking side The AGCF evaluates the Switch Order Command based on a provisioned mapping table. If the feature code received from the internal MGC component does not match any known feature, the AGCF ignores the feature code and optionally request the media gateway to play an error tone or an announcement to the served user. If the SOC indicates that the user wishes to establish a 3-party conference with the held parties. the AGCF performs the following actions: • Request the media gateway to: - Add a termination to the current context based on SDP information associated with the held party. - Monitor the flash-hook event. • Send a re-INVITE request towards the first held party (i.e. the party that was already held when the flash-hook event was detected). The re-INVITE request is built as follows: - The Request URI is set to the held party's identity. - The SDP description is set to a=sendrecv. The address and port are set according to the contents of the local descriptor of the new termination. • Send a re-INVITE request towards the second held party (i.e. the party that has been held for the purpose of collecting the Switch Order Command). The re-INVITE request is built as follows: - The Request URI is set to the held party's identity. - The SDP description is set to a=sendrecv. Once the 3-party call is established with the waiting party, processing of the flash-hook event is similar to the call waiting service. If SOC indicates that the served user wishes to reject one of the parties, the AGCF performs the following actions: • Send a BYE request towards the held party. • Request the media gateway to: - Remove the corresponding ephemeral termination. - Monitor the flash-hook event. C.14.2.2 Actions at the Originating AS at the invoking side On receipt of a re-INVITE request with an SDP attribute "sendonly", the AS may interact with an MRFC in order to play an announcement to the held party, in accordance with TS 183 028 [6]. On receipt of a re-INVITE request with the SDP attribute "sendrecv", the AS interacts with the MRFC to stop any ongoing announcement. On receipt of a BYE request, the AS releases the dialogue with the associated party. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 64 C.14.3 Option 2 (Tight coupling) C.14.3.1 Actions at the AGCF at the originating side The AGCF opens a new dialogue and sends an INVITE request to the Application Server, built as follows: • The Request URI is structured as follows: - A user part containing a provisioned prefix followed by the switching order command without the start and finish fields. NOTE: In networks where no explicit SOC is collected, a preconfigured SOC is used to populate the user part, such as "flash@pes.operator.com". - A domain name that together with the user part provides sufficient information to the S-CSCF to forward the INVITE request to the appropriate AS, based on Initial Filter Criteria stored in the user profile, e.g. SOC- "SO (SR SI)"@pes.operator.com • A P-Asserted-Identity containing the public identity of the subscriber issuing the switching control command. • An SDP offer for a voice call. The AGCF modifies the H.248 Remote Descriptor and stream mode according to the SDP information received in re-INVITE messages sent by the Application Server. C.14.3.2 Actions at the Originating AS at the originating side The AS evaluates the Switch Order Command based on the current call configuration and issues appropriate re-INVITE messages. The AS interacts with an MRFC is order to play and stop announcements to held parties, in accordance with TS 183 028 [6]. In order to provide a network-based 3 party conference, the AS may use the ANNC URL syntax defined in [16] to form a conference using an MRFC/MRFP. In this case, the AS will send re-INVITE requests to all 3 parties in order to connect all three parties together at the MRFC/MRFP. C.15 Repeat Last Call C.15.1 AGCF at the served user side The Repeat Last Call service is invoked using a service code command. On receipt of this service code command, the AGCF builds an INVITE request as described in clause C.1. C.15.2 AS at the served user side Implementation of this service assumes that the AS providing the service is involved in all outgoing calls to the served user. It is the responsibility of the network operator to configure the appropriate Initial Filter Criteria in the user profile. The AS keeps track of the last incoming call to each served user it manages. When receiving the service code command that identifies the Repeat Last Call service, the AS acts as a B2BUA and establishes a new call leg to the last called party (i.e. it sends an INVITE messages towards this last called party as if the number had been dialled again by the served user). ETSI ETSI TS 183 043 V1.2.1 (2009-02) 65 Annex D (normative): Mapping between SIP and the subscriber line protocol. D.1 Introduction This annex describes the mapping between SIP messages received by an AGCF or a VGW acting as a UE and the messages of the subscriber line protocol defined in ES 200 659-3 [10]. D.2 Call Setup message This message is used to send information related to an incoming call, e.g. Calling Line Identification Presentation (CLIP) and related services. This message is built by the AGCF or the VGW on receipt of an initial INVITE request. The AGCF requests the media gateways to send this message over the analog line using the Display Data Block parameter of the Display With Alerting signal of the andisp package defined in ITU-T Recommendation H.248.23 [13]. The AGCF or VGW populates the message parameters as described in table D.1, based on the contents of the INVITE message and local configuration data. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 66 Table D.1: Call set-up message parameters Parameter type O/M Populating rules Date and Time O Set from local clock (see note 1) Calling Line Identity M Set according to the contents of the "From" header or "P-Asserted-Identity" header dependent on national operator requirements Or Or Reason for absence of Calling Line Identity Set from "Privacy" header (see note 2) Called Line Identity O "P-Called-Party-Id" header Calling Party Name O Set according to the "Display-Name" in the "From" header or "P-Asserted-Identity" header dependent on national operator requirements Or or Reason for absence of Calling Party Name "Privacy" header (see note 2) Complementary Calling Line Identity O Setting of this parameter is based on operator specific rules. Call type O Setting of this parameter is based on operator specific rules. First Called Line Identity O Set according to the "hi-targeted-to-uri " in the first entry in the "History-Info" header Absent if the "Privacy" header set to "history" Number of Messages O Setting of this parameter is based on operator specific rules. Type of Forwarded call O Set according to the "Cause" parameter associated with the "hi-targeted-to-uri" in the last entry of the "History-Indo" header Absent if the "Privacy" header is set to "history" Type of Calling User O Set from the cpc parameter of the P-Asserted-Identity header Absent if the "Privacy" header set to "header" Redirecting Number O Set according to the "hi-targeted-to-uri" in the last entry in the "History-Info" Absent if the "Privacy" header set to "history" Network Provider Identity O Setting of this parameter is based on operator specific rules Carrier Identity O Setting of this parameter is based on operator specific rules Selection of Terminal Function O Setting of this parameter is based on operator specific rules Display Information O May be set from the contents of the message body Service Information O May be set from the contents of the message body Extension for network operator use O Setting of this parameter is based on operator specific rules NOTE 1: The AGCF and the VGW shall support an appropriate clock synchronization mechanism. NOTE 2: If the "Privacy" header is included and set to "id" or "user" or "header", the Reason for Absence is set to "Private" (0101 0000). If the "P-Asserted-Identity" header and the "From" header are absent, the Reason for Absence is set to "unavailable" (0100 1111). D.3 Message Waiting Indicator message This message type is used to handle information related to messages in a message system. This message is built by the AGCF or the VGW on receipt of a NOTIFY request reporting the "message-summary" event. The AGCF requests the media gateways to send this message over the analog line using the Data Block parameter of the Generic Data Signalling signal of the andisp package defined in ITU-T Recommendation H.248.23 [13]. The value of the TAS parameter is provisioned in the AGCF or MGF, per media gateway or per line. The AGCF or the VGW populates the message parameters as described in table D.2, using the information contained in NOTIFY requests reporting the "message-summary" event and local configuration data. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 67 Table D.2: Message Waiting Indicator message parameters Parameter type O/M Populating rules Date and Time O Set from local clock (see note) Calling Line Identity Set according to the contents of the "P-Asserted-Id" header Or Or Reason for absence of Calling Line Identity O Privacy header Calling Party Name Set according to the "P-Asserted-Id" header Or or Reason for absence of Calling Party Name O "Privacy" header Visual Indicator M Set to "FF"H if the "Messages-Waiting" header is set to "yes" Message Identification O Set according to the "Message-ID" header Last Message CLI O "From" header associated with the last message Complementary Date and Time O "Date" header associated with the last message Complementary Calling Line Identity O Setting of this parameter is based on operator specific rules Number of Messages O Set according to the "Voice-Message" header Type of Calling User O Setting of this parameter is based on operator specific rules Network Provider Identity O Setting of this parameter is based on operator specific rules Selection of Terminal Function O Setting of this parameter is based on operator specific rules Display Information O May be set from the contents of the message body Extension for network operator use O Setting of this parameter is based on operator specific rules NOTE: The AGCF and the UE shall support an appropriate clock synchronization mechanism. D.4 Advice of Charge message This message is used to send information related to the charge of a call. This message is built by the AGCF or the VGW on receipt of a SIP message that contain information defined inTS 183 087 [7]. The AGCF requests the media gateways to send this message over the analog line using the Burst Pulses Count parameter of the Metering Pulse Burst signal of the amet package defined in ITU-T Recommendation H.248.26 [14]. The AGCF or the VGW populates the message parameters as described in table D.3, based on information defined in TS 183 087 [7] and local configuration data. Table D.3: Advice Of Charge message parameters Parameter type O/M Populated From Date and Time O Set from local clock (see note) Calling Line Identity O Or Setting of this parameter is based on operator specific rules. Reason for absence of Calling Line Identity Called line identity O Setting of this parameter is based on operator specific rules. Complementary Calling Line Identity O Setting of this parameter is based on operator specific rules. Charge M Set from information defined in TS 183 087 [7] Additional Charge O Setting of this parameter is based on operator specific rules. Duration of the call O Setting of this parameter is based on operator specific rules. Network Provider Identity O Setting of this parameter is based on operator specific rules. Carrier Identity O Setting of this parameter is based on operator specific rules. Selection of Terminal Function O Setting of this parameter is based on operator specific rules. Display information O Set from information defined in TS 183 087 [7] Extension for network operator use O Setting of this parameter is based on operator specific rules. NOTE: The AGCF and the VGW shall support an appropriate clock synchronization mechanism. ETSI ETSI TS 183 043 V1.2.1 (2009-02) 68 Annex E (informative): Bibliography ETSI ES 282 007: "Telecommunications and Internet converged Services and Protocols for Advanced Networking (TISPAN); IP Multimedia Subsystem (IMS); Functional architecture". IEEE 1003.1: "Standard for IEEE Information Technology - Portable Operating System Interface (POSIX®)". ETSI ETSI TS 183 043 V1.2.1 (2009-02) 69 Annex F (informative): Change history Date WG Doc. CR Rev CA T Title / Comment Current Version New Version 01-12-06 12bTD197 001 F Correction of AGCF’s actions for CW service 1.1.1 1.1.2 30-05-08 17bTD311 002 F ACR 1.1.1 1.1.2 30-05-08 17bTD310 003 F Session Progress 1.1.1 1.1.2 18-06-08 Addition of the Change History annex and update of WI reference by ETSI Secretariat 1.1.2 1.1.3 05-11-08 19WTD203r1 004 F Correction of the Name of P-Asserted_identity Header 1.1.3 1.1.4 Publication 1.1.4 1.2.1 ETSI ETSI TS 183 043 V1.2.1 (2009-02) 70 History Document history V1.1.1 May 2006 Publication V1.2.1 February 2009 Publication
86a8cb2417653fdbaee677b0a1930040
183 020
1 Scope
The present document describes the protocol specifications and profiles for the interface between the UAAF in the visited NGN network (UAAF-proxy) and the UAAF in the home NGN network (UAAF-server). The specifications of this interface will be common for both xDSL and WLAN access networks, including possible other access network types as well. Specific differences, if any, will be called out.
86a8cb2417653fdbaee677b0a1930040
183 020
2 References
The following documents contain provisions which, through reference in this text, constitute provisions of the present document. • References are either specific (identified by date of publication and/or edition number or version number) or non-specific. • For a specific reference, subsequent revisions do not apply. • For a non-specific reference, the latest version applies. Referenced documents which are not found to be publicly available in the expected location might be found at http://docbox.etsi.org/Reference. [1] ETSI ES 282 001: "Telecommunications and Internet converged Services and Protocols for Advanced Networking (TISPAN); NGN Functional Architecture Release 1". [2] ETSI ES 282 004: "Telecommunications and Internet converged Services and Protocols for Advanced Networking (TISPAN); NGN Functional Architecture; Network Attachment Sub- System (NASS)". [3] ETSI TS 129 234: "Universal Mobile Telecommunications System (UMTS); 3GPP system to Wireless Local Area Network (WLAN) interworking; Stage 3 (3GPP TS 29.234 Release 6)". [4] Wi-Fi Alliance: "WPA™ Deployment Guidelines for Public Access Wi-Fi® Networks". NOTE: http://www.wi-fi.org/OpenSection/pdf/WPA_for_Public_Access_Final.pdf [5] IETF RFC 3748: "Extensible Authentication Protocol (EAP)". [6] IETF RFC 2486bis: "The Network Access Identifier". [7] ETSI TS 183 019: "Telecommunications and Internet converged Services and Protocols for Advanced Networking (TISPAN); Network Attachment; Network Access xDSL and WLAN Access Networks; Interface Protocol Definitions". [8] IETF RFC 2865: "Remote Authentication Dial In User Service (RADIUS)". [9] ETSI ES 283 034: "TISPAN; Network Attachment Sub-System (NASS); e4 interface based on the DIAMETER protocol". [10] ETSI TS 183 017: "Telecommunications and Internet Converged Services and Protocols for Advanced Networking (TISPAN); Resource and Admission Control: DIAMETER protocol for session based policy set-up information exchange between the Application Function (AF) and the Service Policy Decision Function (SPDF); Protocol specification". [11] ETSI ES 283 035: "TISPAN; Network Attachment Sub-System (NASS); e2 interface based on the DIAMETER protocol". [12] IETF RFC 3539: "Authentication, Authorization and Accounting (AAA) Transport Profile". [13] IETF RFC 4005: "Diameter Network Access Server Application". [14] IETF RFC 4072: "Diameter Extensible Authentication Protocol (EAP) Application". ETSI ETSI TS 183 020 V1.1.1 (2006-03) 6 [15] IETF RFC 3588: "Diameter Base Protocol". [16] IETF RFC 4372: "Chargeable User Identity". [17] IETF RFC 2866: "RADIUS Accounting". [18] IETF RFC 3580: "IEEE 802.1X Remote Authentication Dial In User Service (RADIUS) Usage Guidelines". [19] IETF RFC 2548: "Microsoft Vendor-specific RADIUS Attributes".
86a8cb2417653fdbaee677b0a1930040
183 020
3 Definitions and abbreviations
86a8cb2417653fdbaee677b0a1930040
183 020
3.1 Definitions
For the purposes of the present document, the following terms and definitions apply: access network: collection of network entities and interfaces that provide the underlying IP transport connectivity between end user devices and NGN entities Attribute-Value Pair: see RFC 3588 [15], it corresponds to an Information Element in a Diameter message functional entity: entity that comprises a specific set of functions at a given location. Functional entities are logical concepts, grouping of functional entities are used to describe practical physical realizations Core Network: portion of the delivery system composed of networks, systems equipment and infrastructures, connecting the service providers to the access network user equipment: one or more devices allowing a user to access services delivered by TISPAN NGN networks NOTE: This includes devices under user control commonly referred to as CPE, IAD, ATA, RGW, TE, etc., but not network controlled entities such as access gateways. visited NGN network: NGN network through which the User Equipment gains network connectivity NOTE: The NGN Network includes both the Access Network and the Core Network. The User Equipment does not have a service relationship with the business entity that operates this network. home NGN network: NGN network through which the User Equipment gains network connectivity NOTE: The NGN Network includes both the Access Network and the Core Network. The User Equipment has a service relationship with the business entity that operates this network.
86a8cb2417653fdbaee677b0a1930040
183 020
3.2 Abbreviations
For the purposes of the present document, the following abbreviations apply: 3GPP Third Generation Project Partnership AAA Authentication, Authorization and Accounting AAA-H AAA-Home AAA-V AAA-Visited AMF Access Management Function AP Access Point ATA Analogue Terminal Adaptor AVP Attribute Value Pair CNG Customer Network Gateway EAP Extensible Authentication Protocol EAPOL EAP Over Lan GSMA Global System for Mobile communications Association IAD Integrated Access Device IETF Internet Engineering Task Force ETSI ETSI TS 183 020 V1.1.1 (2006-03) 7 IP Internet Protocol NAI Network Access Identifier NASS Network Attachment SubSystem NGN Next Generation Network PEAP Protected EAP RGW Residential GateWay STA Station TE Terminal Equipment TLS Transport Layer Security TTLS Tunnelled TLS UAAF User Access Authorization Function UE User Equipment WLAN Wireless Local Area Network WPA Wi-Fi Protected Access xDSL Digital Subscriber Line
86a8cb2417653fdbaee677b0a1930040
183 020
4 NGN General Architecture
ES 282 001 [1] provides a description of the general network architecture of the NGN. The model is depicted in figure 1. UE Visited NGN Network Home NGN Network e1 e5 Figure 1: General NGN Network Model Interface e1 is an access-network-specific interface, and is dependant on the access technology being used (xDSL, WLAN, and so on). Interface e5 is a roaming interface, and is independent of the access technology. Interface e5 is used to provide a consistent method for the visited NGN network to communicate with the home NGN network. ETSI ETSI TS 183 020 V1.1.1 (2006-03) 8 Figure 2 depicts the functional composition of the access network and the NGN core for the roaming scenario, where a UE obtains network access via a visited NGN network and authenticates back with the home NGN network. Details of this model may be found in ES 282 004 [2]. Home NGN Access Network UAAF Visited NGN Access Network PDBF CPE (CNG/TE) e5 ARF a4 Service control subsystems a3 a1 a2 e1 e2 e4 e3 Resource and Admission Control Subsystem CLF NACF CNGCF AMF UAAF Home NGN Access Network UAAF Visited NGN Access Network PDBF UE e5 ARF Visited NGN network Home NGN network Access Transport network e1 Figure 2: NASS mapped onto functional network roles - roaming scenario
86a8cb2417653fdbaee677b0a1930040
183 020
4.1 Overview of Interface e5
The present document details the protocols and profiles for interface e5. This interface is intended to be used between a UAAF-proxy and a UAAF-server, which may be in different distractive domains. This interface allows the UAAF-proxy to request the UAAF-server for user authentication and authorization, based on user profiles. It also allows the UAAF-proxy to forward accounting data for the particular user session to the UAAF-server. The UAAF-proxy will forward access and authorization requests, as well as accounting messages, received over interface a3 from the AMF, to the UAAF-server over interface e5. Responses received back in return from the UAAF-server over interface e5 will be forwarded to the AMF over interface a3. A bilateral trust relationship will need to be setup between the UAAF-proxy and the UAAF-server in order to facilitate this exchange. The specifications for interface e5 will be similar to that of interface a3, between the AMF and the UAAF in the visited network. This interface therefore supports AAA message exchange between the UAAF-proxy and the UAAF-server. RADIUS and Diameter are two possible options for carrier protocols on this interface, and detailed requirements are listed in the present document. Interface e5 supports both authentication/authorization and accounting message exchange. The present document will be common for both xDSL and WLAN access networks, as well as for potential other types of access networks. Specific differences, if any, will be called out in the document. ETSI ETSI TS 183 020 V1.1.1 (2006-03) 9
86a8cb2417653fdbaee677b0a1930040
183 020
5 Protocols and profiles for Interface e5
86a8cb2417653fdbaee677b0a1930040
183 020
5.1 802.1X-based Authentication
Figure 4 depicts a typical protocol stack for 802.1X-based authentication. Further details may be found in Wi-Fi Alliance [4]. The EAP messages are carried over EAPOL (EAP over LAN) frames between the UE and the AP and then re-encapsulated in RADIUS or Diameter messages when sent from the AP to the home AAA Server (via zero or more AAA proxies). In figure 4, the UE (mobile station) acts as the 802.1X supplicant, the AP acts as the authenticator, and the RADIUS AAA server acts as the authentication server. For security reasons, RADIUS is sometimes also carried over IPsec (RFC 3162 (see Bibliograhy) describes use of RADIUS over IPv6-IPsec, and RFC 3580 [18] also recommends use of IPsec to protect RADIUS). Diameter may also be used instead of or in addition to RADIUS. Figure 3: 802.1X/EAP Authentication Protocol Stack ETSI ETSI TS 183 020 V1.1.1 (2006-03) 10 Figure 5 depicts a typical 802.1X-based authentication scenario. The UE attempts to associate with an AP and is challenged to authenticate. At this point, the UE needs to indicate its user identity. There are usually two parts to this identity: the user name and the realm. Typically, these are combined into a Network Access Identifier (NAI) of the form user@realm. The realm part of the NAI is used to establish a connection with the appropriate AAA-H for that user. This presumes that the visited network recognizes that realm name. If this is not the case, then the visited network will signal an authentication failure back to the UE. The UE can then either try a different account (with a different realm) or can try to establish a new account on the visited network. If those alternatives also fail, the UE will be denied access or will be granted only limited guest access. Figure 4: 802.1X-based authentication with RADIUS as AAA protocol To avoid revealing the true user identity to an entity other than the home service provider, especially across the WLAN radio link, the UE can use a generic user name like "anonymous" or "user" in the NAI given in the initial identity exchange. The realm part of the NAI is the only information the visited network needs to know at this point. If PEAP or TTLS are used to establish a secure tunnel between the STA and the AAA-H, then the protected identity exchange will not be visible to the visited network or to any eavesdroppers. The visited network will eventually need to obtain some identity value for charging and billing purposes if the authentication is successful. The home network can provide the identity that identifies the account for charging. This account is used between the visited network and the home network. This account need not be the same as that used by the home network to bill the subscriber. Furthermore, this identity can be an alias specified by the home provider rather than information that might compromise the true identity of the UE user. The identity used for charging can be shared only with the AAA infrastructure and never needs to be sent unprotected across the WLAN radio link. ETSI ETSI TS 183 020 V1.1.1 (2006-03) 11
86a8cb2417653fdbaee677b0a1930040
183 020
5.2 Intermediaries
An intermediary may be defined as an entity that manages and facilitates AAA transactions in real time between 2 service provider network entities. One or more intermediaries may exist between the visited and the home NGN networks, as shown in figure 5. The existence of these intermediaries should be transparent to the AAA message flow, and each intermediary shall support the requirements as noted in the following clauses. Figure 5: Inclusion of intermediaries
86a8cb2417653fdbaee677b0a1930040
183 020
5.3 Requirements of the visited NGN network
One endpoint of interface e5 is the visited NGN network. Specifically, the access network within the visited NGN network should support the following requirements, and interface e5 for AAA logically extends all the way from the access network to the UAAF in the home NGN network. 1) The visited NGN network shall comply with the relevant requirement in TS 129 234 [3], clauses 4 and 5. In TS 129 234 [3], all references to "3GPP AAA Proxy" implies the requirements shall be met by the visited NGN network and all references to "3GPP AAA Server" implies the requirements are to be met by the home NGN network. 2) The visited NGN network shall support either RADIUS or Diameter as the AAA transport. 3) If Diameter is supported, the visited NGN access network shall support the following IETF standards: a) RFC 3539 [12], Authentication, Authorization and Accounting (AAA) Transport Profile. b) RFC 4005 [13], Diameter Network Access Application. c) RFC 4072 [14], Diameter Extensible Authentication Protocol (EAP) Application. 4) If RADIUS is supported, the visited NGN access network shall support the following IETF standards: a) Interpretation of Idle-Timeout attribute in RFC 2865 [8]: If a wireless station is logged on to the network but has not generated outbound IP data traffic (the definition of what constitutes this traffic is being defined in the RADEXT group in the IETF) for a specific interval of time, and if the home NGN network has specified this interval of time via the Idle-Timeout attribute, the visited access network shall proactively disconnect the wireless station and generate an Accounting-Stop message. The actual Idle- Timeout value may be set locally on the access network or via the Idle-Timeout attribute provided by the Home NGN network. If this attribute is present in the AAA message, it shall override any local timeout value set on the access network. b) Interpretation of Session-Timeout attribute in RFC 2865 [8]: The timeout value is typically set according to the expiration time of the prepaid account. If the wireless station remains connected until the timeout expires, the access network either automatically disconnects that wireless station or requests for a re- authentication, depending on the value of the Termination-Action attribute. c) The access network shall be able to appropriately process receipt of at least 5 Class attributes (defined in RFC 2865 [8]). d) RFC 2866 [17], RADIUS Accounting: I) The Accounting-Start message shall be sent when the wireless station successfully authenticates to the network and is authorized for services. ETSI ETSI TS 183 020 V1.1.1 (2006-03) 12 II) The access network shall detect session termination, either via user disconnect/disassociate, inactivity timer expiry, or session-timeout. When this happens, an Accounting-Stop message shall be sent. III) The access network shall send Accounting-Interim records at the interval specified by the home network. IV) The access network shall report the wireless station IP address in the Framed-IP-Address RADIUS attribute in the Accounting-Interim and Accounting-Stop messages. Inclusion of the wireless station IP address in the Framed-IP-Address RADIUS attribute in the Accounting-start message is optional. V) If multiple accounting messages are generated by different physical entities within the access network for the same session, all related accounting messages shall contain the same Session-Id. VI) The NAS shall send the appropriate Accounting On and Off messages to the home network when it undergoes a reset, to enable session state to be cleared on the home AAA server. VII) The access network may attempt to deliver accounting records until an acknowledgement is received. e) RFC 3580 [18], RADIUS Usage Guidelines. f) RFC 3748 [5], Extensible Authentication Protocol (EAP). g) The visited NGN network shall be able to accept WPA keying information contained in Microsoft vendor-specific RADIUS attributes as defined in the following sections of RFC 2548 [19]: 1) Section 2.4.2: MS-MPPE-Send-Key. 2) Section 2.4.3: MS-MPPE-Recv-Key. h) The visited NGN network shall be able to proxy RADIUS Authentication and Accounting messages as defined in RFC 2865 [8] and RFC 2866 [17]. i) The visited NGN network shall be able to send back the Chargeable-User-Identity attribute (RFC 4372 [16]), defined in the RADEXT group in the IETF, in Accounting messages, if it was send in an authorization message by the authenticating AAA server. 5) The visited NGN access network should support IPSec for protecting AAA message flows.
86a8cb2417653fdbaee677b0a1930040
183 020
5.4 Requirements of the home NGN network
1) The home NGN network shall comply with the relevant requirements in TS 129 234 [3], clauses 4 and 5. In TS 129 234 [3], all references to "3GPP AAA Proxy" implies the requirements shall be met by the visited NGN network (specifically the access network), and all references to "3GPP AAA Server" implies the requirements are to be met by the home NGN network. 2) The home NGN network shall support WPA/802.1X: a) The home network shall support EAP methods defined in [7], clause 8.2. b) The home network shall support NAIs as specified by RFC 2486bis [6]. 3) The home NGN network may support WPA2. 4) The visited NGN network shall support either RADIUS or Diameter as the AAA transport. 5) The home NGN network shall support the following IETF standards: a) RFC 3748 [5], Extensible Authentication Protocol (EAP) b) RFC 2486 bis [6], Network Access Identifier 6) If Diameter is supported, the home NGN network shall support the following IETF standards: ETSI ETSI TS 183 020 V1.1.1 (2006-03) 13 a) RFC 3539 [12], Authentication, Authorization and Accounting (AAA) Transport Profile. b) RFC 4005 [13], Diameter Network Access Application. c) RFC 4072 [14], Diameter Extensible Authentication Protocol (EAP) Application. 7) If RADIUS is supported, the home NGN network shall support the following IETF standards: a) RFC 2865 [8], RADIUS Standard. b) RFC 2866 [17] , RADIUS Accounting. c) RFC 3580 [18], RADIUS Usage Guidelines. d) The home NGN network shall support the transmission of WPA keying information to the access network via Microsoft vendor-specific RADIUS attributes as defined in the following sections in RFC 2548 [19]: 1) Section 2.4.2: MS-MPPE-Send-Key. 2) Section 2.4.3.: MS-MPPE-Recv-Key. 8) The home network should support IPSec for protecting AAA message flows.
86a8cb2417653fdbaee677b0a1930040
183 020
5.5 Subscriber Profile Transfer
The protocol used on the e5 interface shall support the transport of subscriber profile information as defined in ES 282 004 [2]. Table 1 provides the list of information elements to be carried over the interface for that purpose and indicates the list of Diameter attributes that shall be supported to achieve it. Table 1: Diameter attributes for Subscriber Profile Transfer Information Element Description DIAMETER attribute Defined in SubscriberID The identity of the subscriber requesting IP connectivity. User-Name RFC 3588 [15] GloballyUniqueAddress This information element contains: - The IP address of the user equipment used by the subscriber for which profile information is being pushed. - The addressing domain in which the IP address is significant. Globally-Unique- Address ES 283 034 [9] InitialGateSetting This information element contains: - The list of default destination IP addresses and ports to which traffic can be sent. - The maximum amount of bandwidth that can be used without explicit authorisation in the uplink direction. - The maximum amount of bandwidth that can be used without explicit authorisation in the downlink direction. Initiate-Gate-Setting ES 283 034 [9] ETSI ETSI TS 183 020 V1.1.1 (2006-03) 14 Information Element Description DIAMETER attribute Defined in QoSProfile (NOTE 1) For each subscribed transport service class and application class, this information element contains: - The maximum amount of bandwidth subscribed by the attached user in the uplink direction. - The maximum amount of bandwidth subscribed by the attached user in the downlink direction. - The maximum priority allowed for any reservation request. QoS-Profile ES 283 034 [9] PrivacyIndicator (see note) This information element provides policy rules for disclosure of subscriber profile elements to applications. Privacy-Indicator The present document (clause 5.5.1). NOTE: This information element may be repeated. There is currently no standard solution defined for transferring subscriber profile data across networks where RADIUS is used instead of Diameter. This may be defined in Release 2.
86a8cb2417653fdbaee677b0a1930040
183 020
5.5.1 Privacy-Indicator AVP
The Privacy-Indicator AVP (AVP code 440 13019) is of type Grouped and provides policy rules for disclosure of subscriber profile elements to applications. AVP Format: Privacy-Indicator ::= < AVP Header: 440 13019 > * {Requested-Information} * [AF-Application-Identifier] Each Requested-Information AVP identifies a profile element whose disclosure is restricted to the list of applications identified by the AF-Application-Identifier AVPs. The Requested-Information AVP is defined in ES 283 035 [11] while the AF-Application-Identifier AVP is defined in TS 183 017 [10]. ETSI ETSI TS 183 020 V1.1.1 (2006-03) 15 Annex A (informative): Tracking of Standards-related Work A.1 Items to be tracked These are items that are on the standards-track in the standards bodies. Their progress should be monitored, and the relevant requirements reflected appropriately. 1) Carrying Location Objects in RADIUS a) The GEOPRIV task group in the IETF is working on this, and draft-ietf-geopriv-radius-lo (see Bibliography) should be tracked. 2) End to end capabilities support for RADIUS b) The RADEXT task group in the IETF is working on this, and draft-lior-radext-end-to-end-caps (see Bibliography) should be tracked. 3) RADIUS Extensions for IEEE 802 (see Bibliography) c) The RADEXT group in the IETF is working on this, and draft-congdon-radext-ieee802-03X3 (see Bibliography) should be tracked. 4) Network bandwidth parameters for RADIUS d) The RADEXT group in the IETF is working on this, and draft-lior-radius-bandwidth-capability (see Bibliography) should be tracked. ETSI ETSI TS 183 020 V1.1.1 (2006-03) 16 Annex B (informative): Bibliography IETF RFC 4284: "Identity selection hints for Extensible Authentication Protocol (EAP)". Draft-ietf-geopriv-radius-lo-: "Carrying Location Objects in RADIUS". Draft-lior-radext-end-to-end-caps-: "End-to-End Capabilities Support for Remote Authentication Dial In User Service (RADIUS)". VLAN, Prority, and Filtering Attributes NOTE: http://www.ietf.org/internet-drafts/draft-ietf-radext-ieee802-01.txt Draft-lior-radius-bandwidth-capability: "Network Bandwidth Parameters for Remote Authentication Dial In User Service (RADIUS)". ETSI TS 122 234: "Requirements on 3GPP system to Wireless Local Area Network (WLAN) Interworking; Release 6". ETSI TS 123 234: "3GPP system to Wireless Local Area Network (WLAN) Interworking; System descriptions; Release 6". International Roaming Access Protocols (IRAP): "Public WLAN Roaming Interface Specification". NOTE: ftp://download.intel.com/technology/comms/roaming/download/Roaming_Interfaces_v0.95.pdf ETSI TS 124 234: "3GPP system to Wireless Local Area Network (WLAN) Interworking; User Equipment to Network protocols; Stage 3; Release 6". GSMA IR 61: "WLAN Roaming Guidelines". IETF RFC 3162: "RADIUS and IPv6". ETSI ETSI TS 183 020 V1.1.1 (2006-03) 17 History Document history V1.1.1 March 2006 Publication
f92d6483a77f6b4220575ae1402a53f3
183 042
1 Scope
The present document specifies the, stage three Protocol Description of the Completion of Communications to Busy Subscriber (CCBS) service and the Completion of Communication on no Reply (CCNR) service
f92d6483a77f6b4220575ae1402a53f3
183 042
2 References
References are either specific (identified by date of publication and/or edition number or version number) or non-specific. • For a specific reference, subsequent revisions do not apply. • Non-specific reference may be made only to a complete document or a part thereof and only in the following cases: - if it is accepted that it will be possible to use all future changes of the referenced document for the purposes of the referring document; - for informative references. Referenced documents which are not found to be publicly available in the expected location might be found at http://docbox.etsi.org/Reference. NOTE: While any hyperlinks included in this clause were valid at the time of publication ETSI cannot guarantee their long term validity.
f92d6483a77f6b4220575ae1402a53f3
183 042
2.1 Normative references
The following referenced documents are indispensable for the application of the present document. For dated references, only the edition cited applies. For non-specific references, the latest edition of the referenced document (including any amendments) applies. [1] ETSI TS 124 642: "Digital cellular telecommunications system (Phase 2+); Universal Mobile Telecommunications System (UMTS); LTE; Completion of Communications to Busy Subscriber (CCBS) and Completion of Communications by No Reply (CCNR) using IP Multimedia (IM) Core Network (CN) subsystem; Protocol Specification (Release 8)".
f92d6483a77f6b4220575ae1402a53f3
183 042
2.2 Informative references
The following referenced documents are not essential to the use of the present document but they assist the user with regard to a particular subject area. For non-specific references, the latest version of the referenced document (including any amendments) applies. Not applicable.
f92d6483a77f6b4220575ae1402a53f3
183 042
3 Definitions and abbreviations
f92d6483a77f6b4220575ae1402a53f3
183 042
3.1 Definitions
For the purposes of the present document, the terms, definitions given in TS 124 642 [1] apply.
f92d6483a77f6b4220575ae1402a53f3
183 042
3.2 Abbreviations
For the purposes of the present document, the abbreviations given in TS 124 642 [1] apply. ETSI ETSI TS 183 042 V2.1.1 (2009-01) 6 4 Completion of Communications to Busy Subscriber (CCBS) The provisions of the present document are contained in TS 124 642 [1]. ETSI ETSI TS 183 042 V2.1.1 (2009-01) 7 Annex A (informative): Signalling flows Information is contained in TS 124 642 [1] ETSI ETSI TS 183 042 V2.1.1 (2009-01) 8 Annex B (informative): Example of filter criteria Information is contained in TS 124 642 [1] ETSI ETSI TS 183 042 V2.1.1 (2009-01) 9 Annex C (informative): Busy state monitoring procedures Information is contained in TS 124 642 [1] ETSI ETSI TS 183 042 V2.1.1 (2009-01) 10 Annex D (informative): Signalling requirements under discussion Information is contained in TS 124 642 [1] ETSI ETSI TS 183 042 V2.1.1 (2009-01) 11 History Document history V2.1.1 January 2009 Publication
e142e055458ad4dd142e50ecfb0407b8
183 019
1 Scope
The present document describes the protocol specifications and profiles for the interface between the CPE and the access network in the NGN network. The present document will provide specifications for the interface as it pertains to xDSL and WLAN access networks. Other access network types are not precluded, but will not be covered in this version of the present document.
e142e055458ad4dd142e50ecfb0407b8
183 019
2 References
The following documents contain provisions which, through reference in this text, constitute provisions of the present document. • References are either specific (identified by date of publication and/or edition number or version number) or non-specific. • For a specific reference, subsequent revisions do not apply. • For a non-specific reference, the latest version applies. Referenced documents which are not found to be publicly available in the expected location might be found at http://docbox.etsi.org/Reference. [1] ETSI ES 282 001: " Telecommunications and Internet converged Services and Protocols for Advanced Networking (TISPAN); NGN Functional Architecture Release 1". [2] ETSI TS 124 234: "Universal Mobile Telecommunications System (UMTS); 3GPP system to Wireless Local Area Network (WLAN) interworking; User Equipment (UE) to network protocols; Stage 3 (3GPP TS 24.234 Release 6)". [3] Wi-Fi Alliance: "WPA™ Deployment Guidelines for Public Access Wi-Fi® Networks". NOTE: See: http://www.wi-fi.org/OpenSection/pdf/WPA_for_Public_Access_Final.pdf [4] draft-adrangi-eap-network-discovery-12 (March 2005): "Identity selection hints for Extensible Authentication Protocol (EAP)". [5] IETF RFC 3748: "Extensible Authentication Protocol (EAP)". [6] draft-ietf-radext-rfc2486bis-06 (July 2005): "The Network Access Identifier". [7] IEEE 802.1X-2004: "Port Based Network Access Control". [8] IETF RFC 2865: "Remote Authentication Dial In User Service (RADIUS)". [9] DSL Forum TR69. [10] IETF RFC 2131: "Dynamic Host Configuration Protocol". [11] IETF RFC 2132: "DHCP Options and BOOTP Vendor Extensions". [12] IETF RFC 3004: "The User Class Option for DHCP". [13] IETF RFC 3825: "Dynamic Host Configuration Protocol Option for Coordinate-based Location Configuration Information". [14] IETF RFC 3046: "DHCP Relay Agent Information Option". [15] IETF RFC 3993: "Subscriber-ID Suboption for the Dynamic Host Configuration Protocol (DHCP) Relay Agent Option". [16] IETF RFC 3361: "Dynamic Host Configuration Protocol (DHCP-for-IPv4) Option for Session Initiation Protocol (SIP) Servers". ETSI ETSI TS 183 019 V1.1.1 (2005-12) 6 [17] IETF RFC 3315: "Dynamic Host Configuration Protocol for IPv6 (DHCPv6)". [18] IETF RFC 3319: "Dynamic Host Configuration Protocol (DHCPv6) Options for Session Initiation Protocol (SIP) Servers". [19] IETF RFC 3646: "DNS Configuration options for Dynamic Host Configuration Protocol for IPv6 (DHCPv6)". [20] draft-ietf-dhc-dhcpv6-remoteid-00.txt (April 2005): "DHCPv6 Relay Agent Remote ID Option". [21] draft-ietf-dhc-dhcpv6-subid-00.txt (April 2005): "DHCPv6 Relay Agent Subscriber-ID Option". [22] IETF RFC 2516: "A method for transmitting PPP over Ethernet (PPPoE)". [23] IETF RFC 3162: "RADIUS and IPv6". [24] IETF RFC 3580: "IEEE 802.1X Remote Authentication Dial In User Service (RADIUS Usage Guidelines".
e142e055458ad4dd142e50ecfb0407b8
183 019
3 Definitions and abbreviations
e142e055458ad4dd142e50ecfb0407b8
183 019
3.1 Definitions
For the purposes of the present document, the following terms and definitions apply: Access Network (AN): collection of network entities and interfaces that provide the IP transport connectivity between end user devices and NGN entities Functional Entity (FE): entity that comprises a specific set of functions at a given location NOTE: Functional entities are logical concepts, grouping of functional entities are used to describe practical physical realizations. Core Network (CN): portion of the delivery system composed of networks, systems equipment and infrastructures, connecting the service providers to the access network User Equipment (UE): one or more devices allowing a user to access services delivered by TISPAN NGN networks NOTE: This includes devices under user control commonly referred to as CPE, IAD, ATA, RGW, TE, etc., but not network controlled entities such as access gateways. visited NGN network: NGN network through which the User Equipment gains network connectivity NOTE: The NGN Network includes both the Access Network and the Core Network. The User Equipment does not have a service relationship with the business entity that operates this network. home NGN network: NGN network through which the User Equipment gains network connectivity NOTE: The NGN network includes both the Access Network and the Core Network. The User Equipment has a service relationship with the business entity that operates this network.
e142e055458ad4dd142e50ecfb0407b8
183 019
3.2 Abbreviations
For the purposes of the present document, the following abbreviations apply: 3GPP Third Generation Project Partnership AMF Access Management Function AP Access Point ATA Analog Terminal Adaptor CLF Connectivity session Location and repository Function CNG Customer Network Gateway CPECF CPE Configuration Function ETSI ETSI TS 183 019 V1.1.1 (2005-12) 7 DHCP Dynamic Host Configuration Protocol EAP Extensible Authentication Protocol GSMA Global System for Mobile communications Association IAD Integrated Access Device IETF Internet Engineering Task Force IP Internet Protocol NACF Network Access Configuration Function NAI Network Access Identifier NASS Network Attachment SubSystem NGN Next Generation Network PDBF Profile Data Base Function PEAP Protected EAP RGW Residential GateWay SIM Subscriber Identity Module TE Terminal Equipment TLS Transport Layer Security UAAF User Access Authorization Function UAM Universal Access Method UE User Equipment WLAN Wireless Local Area Network xDSL Digital Subscriber Line
e142e055458ad4dd142e50ecfb0407b8
183 019
4 NGN general architecture
ES 282 001 [1] provides a description of the general network architecture of the NGN. The model is depicted in figure 1. UE Visited NGN network Home NGN network e1 e5 Figure 1: General NGN network model Interface e1 is an access-network-specific interface, and is dependant on the access technology being used (xDSL, WLAN, and so on). Interface e5 is a roaming interface, and is independent of the access technology. Interface e5 is used to provide a consistent method for the visited NGN network to communicate with the home NGN network. Figure 2 depicts the functional composition of the access network and the NGN core for the roaming scenario, where a UE obtains network access via a visited NGN network and authenticates back with the home NGN network. Details of this model may be found in TS 124 234 [2]. ETSI ETSI TS 183 019 V1.1.1 (2005-12) 8 Home NGN Access Network UAAF Visited NGN Access Network PDBF CPE (CNG/TE) e5 ARF a4 Service control subsystems a3 a1 a2 e1 e2 e4 e3 Resource and Admission Control Subsystem CLF NACF CNGCF AMF UAAF Home NGN Access Network UAAF Visited NGN Access Network PDBF UE e5 ARF Visited NGN network Home NGN network Access Transport network e1 Figure 2: NASS mapped onto functional network roles - roaming scenario
e142e055458ad4dd142e50ecfb0407b8
183 019
4.1 Overview of interface e1
The present document details the protocols and profiles for e1 - the interface for authentication, authorization and IP address allocation. This interface enables the UE to initiate authentication and authorization requests, as well as initiate requests for IP address allocation, DNS allocation, and other network configuration parameters in order to access the network. These requests are received by the AMF (Access Management Function). It is assumed that the IP edge in the transport plane includes an ARF (Access Relay Function), which communicates with the UAAF (User Access Authorization Function) and NACF (Network Access Configuration Function) via the AMF. This interface enables the user equipment to provide user credentials (username/password, token, certificates, etc.) to the Network Attachment Subsystem (NASS) in order to perform network access authentication and authorization. This interface may also enable the NASS to provide authentication parameter to the UE to perform the network authentication when mutual authentication procedure is required. Based on the result of the authentication, the AMF authorises or denies the network access to the user equipment.
e142e055458ad4dd142e50ecfb0407b8
183 019
4.1.1 Authentication
This interface specifies the protocols and profiles used for authentication of the UE to the network, and vice-versa. The credentials used for authentication depend on the authentication method used as well as the preferences of the network service provider. Possible credential types include, SIM, certificates, and username/password.
e142e055458ad4dd142e50ecfb0407b8
183 019
4.1.2 IP address and network configuration
Once the UE is authenticated successfully, it shall be able to obtain an IP address in order to access the different services offered by the service provider. This interface also provides a method to transport IP address and network configuration information to the UE to enable IP access. ETSI ETSI TS 183 019 V1.1.1 (2005-12) 9
e142e055458ad4dd142e50ecfb0407b8
183 019
5 PPP-based access network configuration
e142e055458ad4dd142e50ecfb0407b8
183 019
5.1 Authentication phase
5.1.1 PPP link establishment phase (LCP) with authentication (PAP/CHAP/EAP) A generic description of the PPP-based authentication model is provided in TS 124 234 [2], clause 7.2 and annex B. There are currently no specific requirements on PPP-based authentication scenarios. This may be further studied and defined in Release 2 of the TISPAN specifications.
e142e055458ad4dd142e50ecfb0407b8
183 019
5.2 Network Configuration Phase (NCP)
e142e055458ad4dd142e50ecfb0407b8
183 019
5.2.1 PPP Network Configuration Phase for IP Networks (NCP/IPCP)
e142e055458ad4dd142e50ecfb0407b8
183 019
5.2.1.1 PPP and ARF
In NASS of WI02021, The Access Relay Function (ARF) acts as a relay between the user equipment and the Network Attachment Subsystem (NASS). It receives network access requests from the user equipment and forwards them to the NASS. Before forwarding a request, the ARF may also insert local configuration information and apply protocol conversion procedures. In PPPOE case, the ARF should implement a PPPoE intermediate agent function in order to insert access loop identification. As a PPPoE Intermediate, The ARF intercepts all PPPoE discovery packets, i.e. the PADI, PADO, PADR, PADS and PADT packets, but does not modify the source or destination MAC address of these PPPoE discovery packets. Upon reception of a PADI or PADR packet sent by the PPPoE client, the ARF adds a PPPoE TAG to the packet sent upstream. The TAG contains the identification of the access loop on which the PADI or PADR packet was received in the Access Node where the ARF resides. If the TAG containing the access loop identification is present in PADO or PADS packets sent by the AMF, the ARF must remove the TAG before sending the packet downstream. If a PADI or PADR packet exceeds 1 500 octets after adding the TAG containing the access loop identification, the ARF must not send the packet to the AMF. The ARF should then return a Generic-Error TAG to the sender in the appropriate PPPoE discovery packet (i.e. PADO or PADS). The required syntax for access loop identification is depicted in following table defined in RFC 2516 [22]. ETSI ETSI TS 183 019 V1.1.1 (2005-12) 10 +--------------+--------------+--------------+--------------+ | 0x0105 (Vendor-Specific) | TAG_LENGTH | +--------------+--------------+--------------+--------------+ | 0x00000DE9 (3561 decimal, i.e. "ADSL Forum" IANA entry) | +--------------+--------------+--------------+--------------+ | 0x01 | length | Agent Circuit ID value... | +--------------+--------------+--------------+--------------+ | Agent Circuit ID value (cont) | +--------------+--------------+--------------+--------------+ | 0x02 | length | Agent Remote ID value... | +--------------+--------------+--------------+--------------+ | Agent Remote ID value (cont) | +--------------+--------------+--------------+--------------+ The first four octets of the TAG_VALUE contain the vendor id. Specifically, the enterprise code here is that of the DSL Forum, (i.e. 3561 in decimal, corresponding to the IANA "ADSL Forum" entry in the Private Enterprise Numbers registry). The remainder of the TAG_VALUE is unspecified in RFC 2516 [22]. Note that the sub-options do not have to be aligned on a 32-bit boundary. "Agent Circuit ID" sub-option (sub-option 1). This sub-option uniquely identify the Access Node and the access loop on the Access Node on which the PPPoE discovery packet was received. "Agent Remote ID" sub-option (sub-option 2). The sub-option contains an string that uniquely identifies the subscriber on the associated access loop on the Access Node on which the PPPoE discovery packet was received. For encoding the sub-option, the same sub-option based encoding as in DHCP option 82 is used (see clause 7.1.2).
e142e055458ad4dd142e50ecfb0407b8
183 019
5.2.1.2 PPP and AMF
The Access Management Function (AMF) translates network access requests issued by the UE (TISPAN WI 02021). It forwards the requests for allocation of an IP address and possibly additional network configuration parameters to/from the NACF. In case PPP is applied, the AMF terminates the PPP connection and provides the inter-working with the interface to the network attachment subsystem e.g. using an AAA protocol (RADIUS or Diameter). The AMF acts as a RADIUS client if the UAAF is implemented in a RADIUS server. When ARF implements a PPPoE Intermediate Agent and adds access loop identification TAG. The AMF should be able to support access loop identification carried over PPPoE. ETSI ETSI TS 183 019 V1.1.1 (2005-12) 11 The AMF shall accept PADI and PADR packets containing a TAG that is used to convey the access loop identification to AMF. The access loop information present in a TAG in the PADI and PADR packets may be used by the AMF to check whether PPPoE discovery is allowed for the identified subscriber line. This procedure is independent of the PPP authentication phase performed later on. The AMF shall be able to use the access loop identification to construct the proper RADIUS Attributes (e.g. NAS-Port-Id, NAS-Port or Calling-Station-Id) during the PPP authentication phase. These Attributes are sent in a RADIUS Access-Request packet to the UAAF which is a RADIUS server. This allows the UAAF to take the access loop identification into account when performing authentication, authorization and accounting.
e142e055458ad4dd142e50ecfb0407b8
183 019
6 Ethernet-based access network configuration
e142e055458ad4dd142e50ecfb0407b8
183 019
6.1 Authentication phase
e142e055458ad4dd142e50ecfb0407b8
183 019
6.1.1 EAP over Ethernet (802.1X)
Figure 3 depicts a typical protocol stack for 802.1X-based [7] authentication. Further details may be obtained from [3]. The EAP messages are carried over EAPOL (EAP over LAN) frames between the UE and the AP and then reencapsulated in RADIUS [8] or Diameter messages when sent from the AP to the home AAA Server (via zero or more AAA proxies). In figure 3, for the WLAN scenario, the UE (mobile station) acts as the 802.1X supplicant, the AP acts as the 802.1X authenticator, and the AAA server (which implements the UAAF functionality) acts as the authentication server. For security reasons, RADIUS is sometimes also carried over IPsec (RFC 3162 [23] describes use of RADIUS over IPv6-IPsec, and RFC 3580 [24] also recommends use of IPsec to protect RADIUS). Diameter may also be used instead of or in addition to RADIUS. Figure 3: EAP authentication protocol stack ETSI ETSI TS 183 019 V1.1.1 (2005-12) 12 Figure 4 depicts a typical 802.1X-based authentication scenario. The UE (in this case, the wireless station) attempts to associate with an AP and is challenged to authenticate. At this point, the wireless station needs to indicate its user identity. There are usually two parts to this identity: the user name and the realm. Typically, these are combined into a Network Access Identifier (NAI) of the form user@realm. The realm part of the NAI is used to establish a connection with the appropriate AAA-H for that user. This presumes that the visited network recognises that realm name. If this is not the case, then the visited network will signal an authentication failure back to the wireless station. The wireless station can then either try a different NAI (with a different realm) or can try to establish a new NAI on the visited network. If those alternatives also fail, the wireless station will be denied access or will be granted only limited guest access. Figure 4: 802.1X-based authentication with RADIUS as AAA protocol To avoid revealing the true user identity to an entity other than the home service provider, especially across the WLAN radio link, the wireless station can use a generic user name like "anonymous" or "user" in the NAI given in the initial identity exchange. The realm part of the NAI is the only information the visited network needs to know at this point. If PEAP or TTLS are used to establish a secure tunnel between the STA and the AAA-H, then the protected identity exchange will not be visible to the visited network or to any eavesdroppers. The visited network will eventually need to obtain some identity value for charging and billing purposes if the authentication is successful. The home network can provide the identity that identifies the account for charging. This account is used between the visited network and the home network for inter-operator charging purposes only. This account need not be the same as that used by the home network to bill the subscriber. Furthermore, this identity can be an alias specified by the home provider rather than information that might compromise the true identity of the wireless user. The identity used for charging can be shared only with the AAA infrastructure and never needs to be sent unprotected across the WLAN radio link. ETSI ETSI TS 183 019 V1.1.1 (2005-12) 13
e142e055458ad4dd142e50ecfb0407b8
183 019
6.1.2 802.1X-based WLAN access network functional architecture
The WLAN Access Network will physically consist of at least one Access Point (AP), which will provide the radio connectivity for the WLAN UE devices. The Access Network or the core network may also contain an Access Controller (AC), that may manage a number of APs. A generic model for access of WLAN to NGN networks is depicted in figure 5. Figure 5: Generic model for WLAN access to NGN The numbers shown in figure 5 correspond to the following steps of a typical 802.1X-based authentication scenario. 1) The wireless station (UE) discovers an 802.11 access point (AP) and initiates a connection request. The AP (or a network authenticator) responds with a request for the UE identity. 2) The AP in the access network forwards the UE identity as an authentication request message to the local authentication server/proxy (AAA-Proxy) that implements the UAAF functionality. This may be forwarded via an Access Controller (AC). Either the AP, or the AC, or a combination of the two could implement the AMF functionality. 3) If the AAA-Proxy is able to authenticate the user credentials, it does so locally. If the AAA-Proxy examines the wireless station identity and decides that this is a roaming user, it forwards the authentication request on to the AAA server of the home provider of that user (AAA-Server) based on the realm name specified in the wireless station identity. 4) The AAA-Server (which also implements the UAAF functionality) authenticates the user via an EAP-based challenge-response method that runs end-to-end between the AAA-Server and the wireless station. A local user database (PDBF) is consulted by AAA-Server to verify the credentials provided by the wireless station. The result of the authentication and session key material are communicated back to the AAA-V and AP respectively. 5) The AP configures link-layer session keys and signals that the wireless station has been successfully authenticated. Prior to this time, the AP blocks any attempt by the wireless station to obtain an address or access the Internet. 6) The AP, through the AAA-Proxy, sends accounting messages to the AAA-Server. When the wireless station disconnects, an accounting stop message is sent as the last message for that session. The AAA-V and AAA-H generate charging records. The AAA-H sends these record to a billing center.
e142e055458ad4dd142e50ecfb0407b8
183 019
6.2 Network configuration phase
e142e055458ad4dd142e50ecfb0407b8
183 019
6.2.1 DHCP procedures
The Network Access Configuration Function (NACF) is responsible for the IP address allocation to the UE. IT may also distribute other network configuration parameters such as address of DNS server(s) and address of signalling proxies for specific protocols. A NACF may be realised as a DHCP server. ETSI ETSI TS 183 019 V1.1.1 (2005-12) 14 Once the UE has been authenticated as described above, it may use DHCP to request an IP address. A local DHCP server that functions as a NACF may respond to this request and assign an IP address with a local subnet prefix to the UE. It maintains a mapping between the UE and the IP address that has been assigned to it, and may forward this information to the CLF. Further specifications are provided in clause 7.
e142e055458ad4dd142e50ecfb0407b8
183 019
7 DHCP Support for Interface e1
e142e055458ad4dd142e50ecfb0407b8
183 019
7.1 Access Network based on IPv4