Analyst Angle: The IMS effect on network bandwidth

Editor’s Note: Welcome to our Monday feature, Analyst Angle. We’ve collected a group of the industry’s leading analysts to give their outlook on the hot topics in the wireless industry. In the coming weeks look for columns from Current Analysis’ Avi Greengart, Compete’s Miro Kazakoff and Jupiter Research’s Julie Ask.
As I sat down to write this column, I was considering which aspect of the iPhone had not been explored ad nauseum and what I could discuss. Aside from a comparison of the relative effects on society of the iPhone and Paris Hilton’s prison term, I could not come up with much. So I decided to ignore the iPhone and move on-of course, ignoring the sweet looking Apple phone is a heck of a lot harder than ignoring Ms. Hilton.
So let’s talk about. IMS. Remember IMS? IP Multimedia Subsystem? All of the world’s major mobile operators are planning IMS deployments to maximize the returns on their 3G investments. AT&T Mobility introduced its IMS-based video service a couple of weeks ago. By implementing an IMS platform, service providers gain a common, central and secure service development framework and architecture that operates across multiple network domains. In essence, IMS allows for network and service evolution versus network and service revolution, and allows the mobile service provider to avoid simply providing a “dumb pipe” data service.
IMS also reduces the number of servers the operator needs to support, thereby reducing operating expenses. As opposed to having a dedicated server per service or feature, the IMS platform supports a range of applications and hence allows the operator to spread the processing load and cost.
So, what types of applications can be developed because of IMS? Well, the answer is really: “Whatever you can imagine and then develop.” By mixing and adding core IMS functions to existing apps and services, new IMS applications which have a higher perceived value than the original app are created. And since we area talking about mixing voice and data services, the bandwidth required will increase.
The question, of course, is how much extra bandwidth? To answer this question, iGR recently modeled IMS application usage for the next few years and the resulting effect on bandwidth. We started with models for application usage in 2007 for light, moderate and heavy users. Then we increased the adoption of the various applications over the next few years, assuming that some subscribers would use the various applications more than others.
The result of the models was that significant increases in bandwidth consumption begin in 2008, when iGR expects a fair number of IMS applications to become commercially available. The IMS applications which are the biggest bandwidth “hogs” are, of course, those related to video-video messaging, see what I see, push to video, etc. Note that the models and forecasts show the total bandwidth used relates to both the frequency and duration of usage-which isn’t surprising at all. Mobile operator networks need to be able to support longer duration data activities.
So how much extra bandwidth is iGR forecasting for the next few years? Using 2008 as a base, the model shows the total bandwidth used by mobile subscribers in the U.S. more than doubling by 2011. The “light” subscribers are expected to increase from about 6.3 MB per month in 2007 to over 22 MB per month in 2011.
Modeling the IMS applications that are likely to be implemented and adopted by consumers is important because the resulting bandwidth estimate provides a frame of reference within which mobile operators can evaluate the price of their mobile data packages. The models also provide a concrete way to discuss applications that have yet not yet been commercially implemented-it goes beyond the “IMS is push-to-talk” stereotype. And since IMS is not a wireless-only standard or effort, defining the applications that will be used across all broadband networks and then estimating the bandwidth they will consume are natural first steps in the march toward convergence. And finally, end-user devices are also impacted in terms of processor speed and memory-to ensure the user experience is maintained (or ideally improved) as the amount of data delivered to and sent from the mobile device increases, the processing and memory capabilities must be upgraded.
Questions or comments about this column? Please e-mail Iain at [email protected] or RCR Wireless News at [email protected].