Getting Back to Basics: Why Aren't More Carriers Deploying Softswitches?

By Laurie Gooding
Last year at Supercomm, I was dumbfounded by the virtual nonresponse I received from vendors when I excitedly introduced them to a new quality-of-service technology that can dramatically improve voice quality by actually looking inside an IP packet and actively monitoring call quality. I believed (and still do) that technological enhancements such as this have the potential to change the whole packet telephony industry.
The equipment vendors-and even a few of the service providers-I spoke to last year were lukewarm about improving voice quality. One softswitch vendor even said "That's cool but I really don't see the demand for it right now.All of our (carrier) customers ensure quality of service by overbuilding their networks."
I was flabbergasted. Didn't he understand this was my whole point? Why migrate to a packet-based infrastructure if you are going to eat your own profit margins with higher operational expenses? "These guys," I thought, "need to get back to basics."
Disillusioned by the slow uptake of truly distributed multivendor softswitch networks, several equipment vendors have approached me recently. Despite a break in the clouds looming over the telecom industry, they are scratching their heads pondering the question: When will carriers buy into the multivendor approach?

Multivendor or No?
Carriers have said they prefer a multivendor approach, but the majority of softswitch deployments to date have been for Internet offload applications supported by single-box, convergence-switch products such as the Sonus Networks Inc. GSX 9000, the Telica Inc. Plexus 9000 and the Unisphere Networks SMX 2100. When first launched, most of these products offered cost savings with Internet offload today and the ability to support full-featured, packet voice in the future.
Ironically, convergence switches (physically) look more like a regular TDM switch because they have physical terminals and switching intelligence collocated on the same chassis. The International Softswitch Consortium's (ISC) softswitch framework places switching intelligence in the softswitch and terminations in the media gateway.
One reason the ISC's softswitch framework is taking longer to deploy is the slower-than-expected evolution of the protocol standards that are required for communications between media servers and gateways, such as MEGACO/H.248. The prospect of SIP (Session Initiation Protocol) becoming an IETF (Internet Engineering Task Force) standard later this year doesn't solve these problems.
Other reasons include concerns over voice quality and the need for improved latency, which gets back to my point at Supercomm last year. Overbuilding and overengineering a network to compensate for lack of quality will drain future revenues just like a high-interest "buy back your credit" credit card.

ISC Survey
Sure enough, recent conversations with a number of ILECs and IXCs, and some recent findings by the International Softswitch Consortium confirm that concerns about voice quality, latency and the challenges of multivendor interoperability are key technology hurdles the industry needs to overcome before widespread deployment of softswitches can begin.
According to the ISC's survey of some 40+ service providers based in the United States, none of the ILECs or IXCs deploying softswitches have future plans to support Class 5 voice. They cited numerous deployment hurdles, such as:
* Handing off calls between Class 4 and Class 5 offices is difficult
* Multivendor interoperability is a challenge
* There may be diminished quality of service because of issues, such as echo cancellation or latency, once implemented.
Once the equipment vendors overcome these technical challenges, softswitches will find their rightful place in the voice world. Although economic and political issues have suppressed the telecom industry as a whole, and investors are still generally wary of the technology sector, carriers are still motivated to deploy solutions that lower costs associated with delivering profitable services. This dynamic is key to the ongoing success of Internet offload and will enable the future success of softswitches.

Laurie Gooding is a senior market analyst with Pioneer Consulting covering the IP network convergence industry and author of the "Carrier IP Convergence: Emerging Opportunities for Softswitches" and "Next-Generation Core Wireless Networks: Bridging the Gap to 3G" reports. Gooding is widely recognized as a market leader in the areas of wide-area networking (WAN), packet telephony and carrier IP convergence, as well as other broadband carrier technologies and trends.