Internet protocol multimedia subsystem (IMS) is an architecture that merges the applications and capabilities of the Internet with both wireless and wireline telephony and promotes fixed-mobile convergence. When realized, IMS will pave the way for true multimedia applications—both voice and video—to be used through multiple forms of access, including third generation (3G) wireless devices such as mobile phones and personal digital assistants (PDAs), as well as through traditional Internet access methods, including corporate local-area networks (LANs) and broadband connections.

Perhaps more important, IMS is also a blueprint for how carriers and service providers can build their networks to deliver these multimedia applications, defined in such a way as to make it much easier and faster for vendors to produce the underlying, interoperable infrastructure that enables service delivery. The interfaces and protocols required to deliver these IP multimedia applications and services are standardized, making it much easier for providers of these new services to develop and integrate new services that will be delivered over IMS infrastructures. It also provides a much cleaner mechanism for leveraging the vast wealth of services developed for the Internet using technologies such as hypertext transfer protocol (HTTP) and Java, allowing these services to be utilized over mobile communications in addition to "non-mobile" connections. Last, but by no means least, IMS even allows services to be transferred seamlessly among access networks when users switch among them (e.g., when users move from a fixed network connection such as a home broadband connection to a mobile network connection such as a mobile phone).

Clearly, IMS is a huge topic, and much has been written about many detailed aspects of it—so much so that it is not always easy to know where to start. This article aims to provide that starting point with a very basic overview of the most important aspects of the IMS specification.

It is very important to note that IMS details multiple functional entities (or service functions) that are the building blocks of the infrastructure. Each entity has well-defined, open interfaces associated with it to make it easy for vendors to create products that provide the necessary function and/or interface to various parts of the system. Vendors are free to create products that combine multiple functions of IMS or may simply choose to produce hardware or software components that address a single function of IMS. Figure 1 shows IMS architecture in detail.

Figure 1: The IMS Functional Architecture

As you can see, there is a lot to the IMS specification. Much like all telecommunications standards or specifications, it is a veritable alphabet soup of acronyms. Rather than define each of these acronyms, I will summarize some of the individual functions.

Perhaps the most powerful concept in IMS is the premise of removing any dependence of service delivery from the physical network itself (aside from the obvious physical constraints of bandwidth). This layered approach allows IMS to provide a truly access-independent service delivery mechanism that opens up the entire network for third-party service delivery. However, there has also been a lot of work put into service control in the network to ensure that access providers are not simply "bit-pipe" providers, with all the service revenue going to service providers.

The following is a closer look at the three main layers of IMS: the transport layer, the control layer, and the service layer.

• The transport layer comprises many types of access networks. Some examples of packet-based networks are general packet radio service (GPRS), Universal Mobile Telecommunications System (UMTS), code division multiple access 2000 (CDMA2000), wireless local-area networks (WLANs), PacketCable, and asynchronous digital subscriber line (ADSL). The traditional public switched telephone network (PSTN) is an example of a circuit-switched network. Users are connected to the IMS infrastructure through the transport layer, either directly through an IMS terminal (such as a 3G wireless handset) or most likely (at least for the near future) through a non-IMS device that interfaces the IMS infrastructure through a gateway. There are several gateway functions found in the transport layer that are primarily in place to provide interworking between legacy networking functions and IMS.
• The control layer is the functional area in IMS that provides all of the session and call control. The call session control function (CSCF) is the central routing engine and policy enforcement point for the network and uses session initiation protocol (SIP) for call control. The home subscriber server (HSS) is also found in the control layer. The HSS is a centralized database that contains all the pertinent user information such as home network location, security information, and user profile information (including the services for which the user has subscribed and may therefore participate in).
• The service layer is where the application server (AS) resides. This is where all of the services are delivered through the IMS interface to the control layer through standardized protocols, primarily SIP.

This is a very simplified overview of some of the main functions of IMS. Because IMS is broken out into these layers, with specific functions and interfaces defined for each layer, it makes it much easier for vendors to choose which functions they wish to incorporate into their equipment by combining these elements in any way they choose. Ultimately, the market will decide which elements are desirable in an integrated solution and which ones should be offered a la carte.

In today's world, multimedia applications are already being delivered over wireless and wireline access networks. Videoconferencing has been available for decades now; "podcasts" and other types of multimedia streaming are also becoming available for mobile phones and other similar devices; and Nextel practically invented push-to-talk services in the United States. This might leave the reader wondering how IMS will really change anything. Let us take a closer look at a potential real-world situation that exposes how IMS could potentially change the end-user experience.

One example is as follows: Imagine you start a videoconference from your home computer or laptop, complete with a live chatroom with a whiteboard. So far, this is nothing new. However, after half an hour, you need to head to the airport to catch a flight, so you log on to the videoconference using a 3G mobile phone, which allows you to join the session with all the same privileges you had from your laptop. However, because your mobile phone does not support the appropriate bandwidth to sustain the video stream, only the audio and text is delivered to the phone. Once you get to the airport, the 3G phone is able to attach to a WLAN (which offers more bandwidth) at the airport, and the video stream may now be added to the session. This is just a small example of how IMS can help users experience the "mobile" Internet.

IMS specification presents the most comprehensive blueprint to date for carriers in building their next-generation networks. While there is certainly a lot more work to be done in certain areas, many vendors and even carriers are aligning their products and solutions to support the IMS specification. Certainly more vendors will emerge to provide specific applications that will be delivered over IMS networks. While it will take several years for the full IMS vision to become reality, we are already starting to see early implementations of IMS elements in next-generation carrier networks.