4. Integration Context

In a telecommunication network, MGCP does not operate in isolation – it is part of a larger signaling architecture that connects traditional telephony and IP-based voice services. The integration context of MGCP is best understood in a scenario where a call originates or terminates in the Public Switched Telephone Network (PSTN) but is carried over an IP network. MGCP’s Call Agent (controller) sits at the crossroads of these networks. On the telephony side, the Call Agent connects to SS7/ISDN call control (either directly or via a Signaling Gateway), and on the IP side, it communicates with other call control entities via protocols like SIP or H.323.

In practical terms, a Call Agent using MGCP will often perform signaling interworking: for example, converting between SIP messages and PSTN signaling (ISUP/BICC). The Call Agent controls media gateways via MGCP to actually cut through voice paths. [RFC 3435] shows that a distributed MGCP gateway system enables PSTN users to participate in sessions set up by IP signaling protocols (SAP, SIP, RTSP), with the Call Agent performing all necessary signaling conversion. In that example, the Call Agent receives SS7/ISUP signals from the circuit network, translates them into, say, SIP session setup on the IP side, and uses MGCP internally to command the media gateway to open voice streams. The MGCP messages are internal control commands – the outside endpoints only see the higher-level signaling (SIP, etc.) and media RTP streams, not the MGCP itself [RFC 3435].