IMS Access Media Gateway

Description

The IMS Access Media Gateway (IMS-AGW) is a media processing entity defined in the IMS media plane architecture. It is not a signaling element but a bearer plane function responsible for the manipulation and forwarding of user media streams (voice, video). The IMS-AGW operates under the control of a Media Gateway Control Function (MGCF) or, more specifically in later architectures, an IMS Media Gateway Control Function (IMS-MGW-CF) or a Media Resource Function Controller (MRFC). Its operation is governed by protocols like H.248 (Megaco) or its IMS profile, which allows the controller to instruct the IMS-AGW to create, modify, and delete media processing contexts and terminations.

Functionally, the IMS-AGW performs several key roles. First, it acts as a media anchor point. This is crucial for services like Voice Call Continuity (VCC) or Single Radio Voice Call Continuity (SRVCC), where a call needs to be seamlessly transferred between different access networks (e.g., from LTE/VoLTE to 3G circuit-switched). By anchoring the media in the IMS-AGW, the remote party's media path remains unchanged, and only the leg to the moving UE is switched, minimizing interruption. Second, it provides media interworking functions, such as transcoding between different voice codecs (e.g., AMR-NB used in legacy networks to EVS used in VoLTE) and packetization protocol conversion (e.g., RTP to circuit-switched TDM bearers).

Third, the IMS-AGW enables media-based services like conferencing (mixing multiple audio/video streams), playing tones and announcements, and facilitating lawful interception by providing a controlled point for duplicating media streams. Architecturally, it sits at the boundary between the all-IP IMS domain and legacy circuit-switched networks or other IP networks with different capabilities. It contains resources like Digital Signal Processors (DSPs) for codec operations and packet processing engines. The separation of control (MGCF/MRFC) and media (IMS-AGW) follows the gateway decomposition model, promoting scalability, flexibility, and the use of standard interfaces.

Purpose & Motivation

The IMS-AGW was introduced to address specific challenges in the evolution towards a pure, end-to-end IP multimedia network. Early IMS visions imagined direct IP-based media paths between endpoints. However, practical deployment revealed several scenarios where intelligent media plane intervention was necessary. Legacy networks (2G/3G circuit-switched) would coexist with IMS for decades, requiring seamless interworking for voice calls. Furthermore, not all user equipment or access networks support the same set of advanced codecs, necessitating transcoding to ensure interoperability.

A key driver was the need for service continuity during mobility events, particularly SRVCC. Without a media anchor, handing over a VoLTE call from LTE to a 3G circuit-switched network would require re-establishing the entire media path with the remote party, causing unacceptable audio gaps. The IMS-AGW provides a stable anchor point in the home IMS network, allowing for a swift update of only one leg of the call. This solved a major obstacle for the commercial launch of VoLTE, as it guaranteed voice service continuity even when users moved out of LTE coverage.

Additionally, regulatory requirements for lawful interception mandated a network-controlled point where media could be copied. A distributed, peer-to-peer media model made this difficult. Concentrating media processing in a managed network function like the IMS-AGW provided a natural point for such intercepts. The decomposition into a separate control function (MGCF) and media gateway (IMS-AGW) also aligned with broader industry trends towards softswitch architectures, allowing operators to source control and media hardware/software independently and scale them according to different traffic patterns.