Understanding IMS (IP Multimedia Subsystem)

IMS (IP Multimedia Subsystem) is an emerging architectural framework, based on SIP and IP, for offering multimedia and VoIP services, as well as fixed mobile convergence. IMS has been standardised by the 3rd Generation Partnership Project (3GPP). The 3GPP specification covers all GSM wireless services.

IMS is being adopted by both wireless and wireline providers as a mechanism for enablling "anytime, anywhere" applications. IMS provides support for the delivery of multimedia applications and voice service over IP transport. The IP transport is the key in that it allows IMS to communicate over any network that supports IP.

Think of how you communicate today. Your voice calls travel over a traditional voice network, a cellular network or a data network enabled for VoIP. The same goes for data, because it can be delivered over both data and cellular networks. IMS provides a common framework and protocols that define how IP networks handle voice calls and data sessions. IMS is an overlay solution that sits above the underlying networks to provide true interoperability among IP enabled devices.

Architecturally speaking, IMS consist of three separate layers: the connectivity layer, the control layer and the service layer. Each of these is discussed below:

1. Connectivity Layer: This layer is composed of the routers, switches and access elements that sit at the edge of a provider's network. The connectivity layer is the layer at which individual users are connected to the network via phones, laptops or PCs.
2. Control Layer: This layer comprises control servers. The control servers manage call or data session set-up, modification and disconnect/release. Most important in this layer is the Call Session Control Function (CSCF). The CSCF is the SIP server that handles call set-up between the end devices.
3. Service Layer: This layer (sometimes called the Application Layer) is made up of content or application servers providing a multitude of enhanced service features for IMS-enabled networks.

With this architecture in place, providers can offer what I like to call session-based services. Simply put, with IMS enabled, any IP/SIP device can establish a session with the control servers and then establish connections with other IP/SIP devices to deliver voice, video and data sessions amongst the two end clients. This breaks the traditional barriers affecting end devices. With IMS enabled, I can initiate a call from my cell phone and communicate with my co-worker over his WLAN PC at his home. The ability to decouple the underlying infrastructure from the services delivered is a major benefit of IMS.

From a service provider perspective, IMS allows for new revenue opportunities because multimedia services can be offered over the current IP infrastructure. For end users, the ability to initiate voice, video and data sessions from any access mechanisms provides the benefit of a much richer end-user experience, as well as business capabilities. Videoconferencing, collaboration, presence and "one" phone are great examples of the way in which IMS can enable rich media content actually to be delivered to an end user.

IMS is gaining significant traction. Many carriers are in development and deployment modes, and they may begin offering IMS services as soon as this year (in a limited fashion). From the enterprise perspective, IMS can provide the framework for deploying truly converged IP networks capable of supporting a rich set of applications. The 3GPP working group Web page can help you keep up with this exciting new technology.

About the author: Robbie Harrell (CCIE#3873) is the National Practice Lead for Advanced Infrastructure Solutions for AT&T. He has more than 10 years of experience providing strategic, business and technical consulting services. Robbie lives in Atlanta and is a graduate of Clemson University. His background includes positions as a principal architect at International Network Services, Lucent, Frontway, Callisma and SBC Communications.