IP Multimedia Subsystem: An Architecture for the Future
Executive Summary
IMS holds much promise for revolutionizing service-delivery architectures. An Emerging set of foundational network standards based on the Session Initiation Protocol (SIP), IMS will eventually allow network providers to deploy converged voice, data and video services that work seamlessly across multiple wired and wireless network types. Subscribers will be able to maintain collaborative sessions as they transition across networks, adding participants and applications without breaking their initial connections. A full migration to IMS standards should occur over a period of several years.
The IP Multimedia Subsystem (IMS) is an open, international standard for Next Generation Networking that supports the delivery of Voice over IP (VoIP) and a range of fixed mobile convergence and multimedia services. It is not in itself a service but an architecture for service delivery. IMS was originally developed by wireless carriers within the context of the Third Generation Partnership Project (3GPP) and was first defined in 2002. Since then, its particular advantages have led to the adoption of IMS by wireline carriers. It provides for person-to-person and person-to-content communication in a variety of modes. Many industry observers believe it will become the standard protocol for telecommunications in the years ahead.
An important driver for IMS is the general trend toward converged networks. Wireline and wireless carriers have been independently converging on IP standards for both voice and data transmission. Some of the reasons include IP’s increased flexibility and the potential it offers for new telecommunications services. IMS also supports the convergence of wireline and wireless access. The true vision of IMS is to be a convergence platform to enable service providers to offer a common user experience with converged feature sets that are accessible from anywhere, regardless of access technology. “Anytime, anywhere, platform independent, device independent” communications will be a fundamental shift in the way that people interact with telecommunications, with less emphasis on the device and more emphasis on individual preferences.
There is a great deal of interest and support for IMS throughout the telecommunications industry. Since IMS is a standard, its broad acceptance should help drive innovation and economies of scale, although its actual effect on the market is just now beginning to be seen. It holds at least the promise for enabling “seamless” mobility with the full integration of wireline and wireless services. AT&T is playing a leading role in the development of the IMS specification, working with various standards bodies and trade groups.
The IMS architecture as defined in the standard alone does not go far enough to enable the creation of secure, flexible and highly reliable services. Building on the foundation of the IMS standard, AT&T’s IMSbased architecture, called the Common Architecture for Real-Time Services over IP (CARTS), includes many additions and enhancements,such as improved security and a rich service development platform.
Technology Overview
The foundation element of IMS is Session Initiation Protocol (SIP), a standard promulgated by the Internet Engineering Task Force (ITEF), with the addition of some 3GPP extensions. The SIP standard is used to establish and control voice and multimedia sessions, providing a single common internal signaling protocol for all infrastructure components. This greatly simplifies the sign-on and authentication process for operators and users. It will let carriers offer networkcontrolled multimedia services, control the hand-off of a session from one network to another or transfer control between different kinds of media and application servers.
IMS has a layered architecture. In the first layer, a “Session Border Controller” (SBC) manages the specifics of each access protocol involved. SBCs not only provide protocol conversion, but also enforce the policies needed for call admission control and security. On a separate layer, intelligent application servers support various kinds of applications for voice, messaging, presence, etc. By separating the session control from the services provided, IMS allows various devices to communicate with each other in a reliable and flexible manner. The interconnect relations, routing and security architecture thus supports any potential service that is built on the IMS infrastructure.
The IMS Architecture
As a layered architecture, IMS increases network flexibility by integrating access methods and applications in a common control and management infrastructure.
The basic IMS idea is shown in this illustration. Each access technology communicates via a specific SBC which provides a uniform internal view. To provide a service, the corresponding application server is "plugged-in." Once a new access technology is supported by a SBC, it can be used by all existing and future application servers. When a new application server is deployed, it can support all existing and future access technologies.
With this flexible "plug and play" type of architecture, IMS offers some inherent advantages. All transactions from any device potentially can access the same set of features, since identities are not tied to the device or service involved. This gives end users greater control, regardless of how they connect. Depending on the service involved, users could control and interact with it via the Web, a GUI device (such as a cell phone, laptop or PDA) or a voice portal. For example, multiple devices could be associated with a single telephone number and be linked to a single voicemail, which would then use user-defined rules to control the system’s behavior. Calls could be forwarded to other voice devices, audio messages routed to email systems and all decisions made with a consideration of factors such as originating number, time of day, day of week and the user’s availability status.
Why Does it Matter?
The general business drivers for IMS include improvements in efficiencies that would flow from an environment of converged devices and services. Traditionally, bringing a new offering to market generally requires an end-to-end solution, with much duplication of effort. Within an IMS environment, the ability to reuse functional modules should benefit both service providers and users. It will also support the trend toward access-independent services and global mobility, by laying the foundation for a flexible, always-available and customer-focused network. That’s why AT&T has chosen to adopt IMS as the foundation for its multimedia network architecture, and to deploy IMS-compliant functions as business needs require. AT&T’s architecture goes far beyond the IMS specification, adding a robust security layer, as well as capabilities for service delivery, service creation, service building and orchestration. Generic support for various access technologies, including Government-mandated capabilities, is also provided.
IMS should also mean faster, less expensive development of innovative offerings. With a common telecommunications standard, calls could be transferred easily from wired to wireless networks, and vice versa. Voicemail systems could be converged and supported by a common address book. Users would have better access to network information from remote locations.
IMS may improve productivity for businesses. By delivering the identical functionality at home, on the road or in the office, IMS is an architecture that could bring full empowerment to remote workers. IMS applications could help bridge collaboration gaps with file sharing and conferencing. This could lead to a general improvement in communications, whether with customers, business partners or employees. By allowing individual users to control their own definition of “reachability,” IMS can support the creation of public and private identities, as well as the concept of “work” personas versus “home” personas. Logic-based service management will control how communications are handled, giving end users an unprecedented degree of control over how, when and to whom they connect.
Consider a typical business day in a pre-IMS world. An employee connects with a co-worker while on route to the office using a wireless device. Reaching her desk, she terminates the wireless call and redials from her land line. She wants to share a presentation file, and sends that to the other party via email. Upon further discussion, she and her colleague decide a teleconference is needed, so they terminate their two-party connection and log into a bridge. Using an Instant Messaging tool (possibly an unsecure, unofficial channel), additional parties are located for the discussion. Not everyone can be found, so some additional calls are placed, emails sent and voice messages left.
This is a fairly routine set of circumstances, and IMS can support the entire scenario in a seamless way. The call can be handed off from the wireless device to the desk phone without breaking contact. The presentation file can be shared on-line via a media server, and the call can be expanded into a teleconference without connecting to a separate bridge. A presence management tool can locate everyone on the team who’s available to participate, and add them to the conversation with the click of a mouse.
The Road to Full Deployment
There is broad agreement that IMS is the way of the future, since it is a control architecture that provides a common standard for both wireline and wireless. It can deliver on the potential of access independence, support seamless services, as well as reduce costs and speed time to market. As an emerging technology, the potential benefits are just now becoming apparent. AT&T believes that IMS will add tremendous value for its customers. It enables a rich new set of services that enhance productivity, provide both lower cost solutions and allow faster feature introductions.
It’s also true that challenges remain. The IMS standard is a work in progress, and there are still issues to be resolved in both functional coverage and interoperability. Many carriers have declared adoption of IMS standards, but are still working through architecture and service implementation plans. That means the full migration to IMS standards will take place over several years. AT&T is in the process of deploying IMS-based service implementations.
For AT&T, IMS is a natural next step in the evolution of the company’s network, which has been building IP-based systems since the mid 1990s. IMS is recognized as a strategic architecture that will bring benefits for AT&T as a company, and for AT&T’s customers. With its growing acceptance by the telecommunications industry, AT&T is confident that IMS will become the standard architecture to support the multi-application networks of the future.