Tweeter Feeds

Monday, July 4, 2011

The bumpy road of IMS (part 2): Why the bumps exist and how to get over them

The bumpy road of IMS (part 2): Why the bumps exist and how to get over them.
Ali Kafel – July 4th, 2011

In Part 1 of this blog, I discussed the history of IMS and its current level of adoption.  To recap:  The IP-Multimedia Subsystem (IMS) standard is a framework for delivering multimedia (voice, data, video and messaging) services. It was initially defined in 2003 by the 3GPP organization in Release 5 (R5). To date, there are relatively few IMS subscribers: as of 1H10, or 3% of the approximately 6 billion phone subscribers (about 5B mobile and 1B Fixed) in the world today. This naturally raised the question: Why has IMS not taken off as originally planned?

After reading the feedback to Part 1 and reflecting on it, I believe there are four major reasons why IMS has not taken off. I’ll begin with the three most frequently cited reasons (weak business case, complexity, limited mobile handset support) and spend some time discussing the fourth, and probably most overlooked, reason: The lack of a new business model.

1.     The lack of a strong incentive/business case versus deploying a Next Generation Network (NGN).  Most of the current deployments of IMS (less than 200M subscribers) are focused on voice, a service that is already well served by NGN.  There is no doubt that, for some telcos, IMS has not yet shown a solid business case that would justify the evolution from NGN to IMS.  However, Rich Communication Suite (RCS) commercial launch commitments and Voice-over-LTE (VoLTE) announcements will once again create a positive trend.
2.     The cost and complexity.  From NGN to IMS, many network elements become separated. For example, the combination of the Softswitch and SBC (two network elements) in NGN become at least six network elements in IMS (P/I/S-CSCFs, MGC, BGCF, IBCF, etc). Although some of these functions may reside in the same hardware. Managing these various network elements and integrating them with existing BSS/OSS systems naturally increase cost and complexity in the IMS model.
3.     Lack of handsets for mobility. Because IMS is an all-IP network, voice and other services need overlay applications. In the fixed world (where most of the current IMS deployments exist), this is much simpler because many end users still connect to the IMS network through legacy ‘black phones’ going through an MSAN or ATA running a SIP client. For mobility, the need to support handoff (Voice Call Continuity) and roaming requires handsets that support these services.
4.     The lack of new business model. This is probably the most significant and complex  issue to address because it requires a new thinking in two ways:
a.     Deploy best-of-breed and demand interworking. This requires service providers to accept and promote the concept of best-of-breed rather than be locked in with one major switch vendor. To protect their base and keep control of the telco service providers, some of the BIG equipment vendors have been pushing IMS and  trying to convince the telcos that they must immediately upgrade their NGN switches to IMS to take advantage of newer services such as Fixed-Mobile Convergence and VoIP (over wireline or wireless broadband).  They also argue that, because IMS is so complicated, the telcos should buy everything from a single vendor—in other words, from them.  As I pointed out in part 1 of my blog, these are the same tactics used during the Intelligent Network (IN) days. With IMS, the need to have an ecosystem of interworking vendors is more critical than it was with IN because, unlike IN, IMS has competition from Over-The-Top (OTT) solution providers that use the cloud and smartphones to offer services which bypass the telecom service providers. To offer competitive services using IMS, service providers must push vendors to work together and offer best-of-breed sub-components such as the Access Gateway Control Function (AGCF), Breakout Gateway Control Function (BGCF), Interconnection Border Control Function (IBCF) and Telephony Application Server (TAS). For example, Ericsson or Alcatel-Lucent may provide a better Home Subscriber Server (HSS) or Serving Call Session Control Function (S-CSCF) while Sonus Networks may provide a better BGCF and TAS.
b.     Support Co-opetition and a two-sided revenue model.  As we all know, the Apple iPhone and iTunes brought the Web 2.0 business model to mobile phones. This is a model in which a service provider offers content and various services to their users as a proxy for other companies that may be complementary or competitive. In this model, the service provider earns revenue from both upstream and downstream customers. Traditional service providers (e.g., Sprint and Google, Verizon and Skype) are now starting to think differently and accept this model.  IMS must support this model, especially with adoption of LTE even if it means co-opetition. Otherwise many users may decide to just use an OTT VoIP provider and bypass the wireless provider unless the wireless provider can offer added value.

So, should service providers consider deploying IMS today?  For some, the answer is Absolutely!  For others, however, the answer may be that NGN is all they currently need.  After all, it’s less complex, cheaper, and supports the needs of most service providers today.  But . . . make sure the solution you choose can migrate easily to IMS.

The call to action: Network equipment vendors should continue to improve their IMS solution, and service providers should demand interworking between best-of-breed equipment providers and deploy solutions from the best vendors rather than be locked in with one major switch vendor.  This approach will empower service providers with the most creative and flexible means of  addressing the challenges of the future by giving them the ability to deploy innovative services with two-sided revenue model. If this change happens, I believe IMS will have a bright future. Otherwise, the forces of stagnation could really dampen the IMS movement.  To put it in perspective, there are already about 900M active worldwide smartphone subscribers running tens of thousands of multimedia apps.  Many of these apps offer multimedia services that are useful, offer good value, are simple to use and improve the way we live, work and play. This number is estimated to grow by 500M subscribers per year. By comparison, IMS has been around twice as long and has scored only 25% of the smartphone installed base (or 3.5% of total telephony users) and it is not growing nearly as fast. Clearly, IMS needs to enter the mainstream with its thinking or risk being relegated to the sidelines while the multimedia revolution happens without it.

For other thoughts on technology solutions and tools that increase business effectiveness, improve our lives and bring us joy, you can personally follow me at