Continuing our current tutorial series regarding softswitches, we have previously discussed:
- Concepts of telephone switching
- Architecture of a softswitch
- Functions of the media gateway controller, also known as call agent
- Capabilities of the media gateway,
- The Media Gateway Control Protocol (MGCP)
- The MEGACO/H.248 Protocol
Heretofore, most of our attention has been focused on the integration of VoIP systems with legacy circuit-switched networks, such as Private Branch Exchange (PBX) switches that are used in medium- to large-sized enterprises, or the central office (CO) switches that are deployed throughout the Public Switched Telephone Network (PSTN). Another option, however, is the integration of the VoIP network with a cellular network, and with that, the role that a softswitch would play within the cellular architecture. This concept is called the IP Multimedia Subsystem, or IMS, which many of the softswitch vendors are beginning to include within their own product architectures. Thus, before we look at specific vendor architectures in future tutorials, it behooves us to explore the IMS.
If we take a step back and look at the big picture, there are a number of large communication systems that require integration. The first is mobile (or wireless) telephony, which differs from fixed (or wireline) telephony in its ability for users to move between networks, and for those networks to hand off the users signal as they pass through a particular area. (Or at least, that is the way that it is supposed to work, but all of us have experienced dropped calls on our cell phones!). Call setups/disconnects, billing, and other technical information must be passed between the wireless and wireline networks as well. Secondly, we have applications from the Internet, such as web pages, video clips, and so on, that are part of the data-centric nature of the Internet Protocol.
Thus, IMS is envisioned to provide cellular users with access to Internet applications. In addition to VoIP, these could include video conferencing, and push-to-talk over cellular. If two carriers were both IMS-compliant, their respective subscribers would be able (at least in theory) to transparently share these communications. From the carriers perspective, IMS looks to provide some significant value-added services (translation: new revenue opportunities) that go beyond the subscriber wars that have become commonplace in the last few years. These concepts have been developed by the 3rd Generation Partnership Project (3GPP) and documented in Release 5 of the 3GPP specifications.
At the heart of IMS is the Session Initiation Protocol, or SIP, that has been discussed in previous tutorials, and well documented by the Internet Engineering Task Force (IETF). The IETF has also been very active with the IMS development, and has documented their work in RFC 4083, Input 3rd-Generation Partnership Project (3GPP) Release 5 Requirements on the Session Initiation Protocol (SIP), available at ftp://ftp.rfc-editor.org/in-notes/rfc4083.txt.
From a very high level, a cellular network can be divided into two network types: a subscribers home network, and a visited network. Since the users services are initiated in the home network, SIP server(s), such as a Registration or Authentication server, would be deployed within the home network to support IMS. The visited network would also contain a SIP server, which would provide communication with the user terminals, and manage any packet-network-based resources that would be accessed as part of the connection. The terminals themselves transport IP datagrams using the General Packet Radio Service (GPRS), using an IP version 6 (IPv6) address that is generated at that terminal. Terminals could include SIP-based devices, or laptop computers having the ability to communicate with a GPRS interface on the home network.
RFC 4083 also details some of the requirements that have been placed on SIP in order for it to operate within the IMS architecture. These include: an efficient use of the radio interface (thus minimizing signaling messages and bandwidth), a minimum call setup time, and a minimum overhead (memory, processing, power, etc.) at the terminal. As we have seen, the industry has developed a great interest in SIP, and further enhancements to that protocol are anticipated as the IMS architecture becomes more prevalent.
In our next tutorial we will continue our discussion of the IMS and softswitch architectures.
Copyright Acknowledgement: © 2005 DigiNet ® Corporation, All Rights Reserved
Mark A. Miller, P.E. is President of DigiNet ® Corporation, a Denver-based consulting engineering firm. He is the author of many books on networking technologies, including Voice over IP Technologies, and Internet Technologies Handbook, both published by John Wiley & Sons.