C.4 P_CSCF procedures to support media traversal through hosted NAT without ICE

29.2133GPPPolicy and charging control signalling flows and Quality of Service (QoS) parameter mappingRelease 17TS

Both the media flows and the SIP signalling can traverse a NA(P)T device located in the customer premises domain, or “hosted NAT”, as described in Annex F of 3GPP TS 24.229 [5]. The hosted NAT will modify the source IP address and source port of uplink IP packets, and the destination IP address and port of downlink IP packets. The IP addess and port information provided by the UE are thus not appropriate to configure PCC rules.

The UE may use ICE procedures for hosted NAT traversal, and related PCC procedures are described in clauses C.1 to C.3. The present clause provides procedures to cover the case where ICE is not used.

If the P-CSCF determines that the UE is located behind a hosted NAT (using procedures in Annex F of 3GPP TS 24.229 [5]) and that ICE is not used (using procedures in clause C.2.1), the P-CSCF shall, within the service information sent to the PCRF:

– provide the source IP address of IP packets transporting incoming SIP messages from the UE as destination IP address of downlink media streams;

– for Ipv4, provide the source IP address of IP packets transporting incoming SIP messages from the UE as source address of uplink media streams;

– for Ipv6, derive the source address of uplink media streams from the prefix of the source IP address of IP packets transporting incoming SIP messages from the UE;

– wildcard source ports of uplink media streams; and

– wildcard destination ports of downlink media streams.

– provide the port information within SDP sent towards the served UE as source ports of corresponding downlink media streams.

– provide the port information within SDP sent towards the served UE as destination ports of corresponding uplink media streams.

Annex D (normative):
Access specific procedures for GPRS