Hearing voices (of the next generation)

Mar 29, 2002 12:00 PM, Kevin Mitchell

These voices I hear are from the service provider respondents of Infonetics Research's recently completed study entitled “The Next Gen Voice Service Provider Opportunity, US/Canada 2002,” for which we interviewed technology executives from 17 service providers deploying next-gen voice by January 2003. According to our respondents, there is a healthy mix of next-gen voice applications--such as softswitch-based voice services and voice-over-packet access--in the works now and next year. Over time, some service providers will adopt many or all applications of the technology. Service providers may be focused on cost avoidance using Internet offload or voice-over-packet transport, but their long-range plans call for using next-gen voice equipment as integral parts of their voice services. The particular application mix implemented depends on the type of service provider.

The major reasons for service providers' interest are not surprising, especially in today's environment. Cost savings is a major driver to buy next-gen voice equipment; the cost per port is lower than for legacy voice switches, and the benefits of floor space and power savings will contribute to lower costs of ownership. Despite some recent cost-cutting, legacy voice switch equipment remains expensive, and service providers are seeking to cut total costs to boost profitability. Service providers are also aware of the new revenue opportunities possible with next-gen voice equipment.

Next-gen voice products can be used for applications in which the service provider uses packet-based equipment to transport voice traffic, offload Internet dial-up calls, or offer both traditional and enhanced voice services using a softswitch architecture. Next-gen voice equipment can act as bridge between the public switched network and data networks, or it can be deployed solely in a pure packet network as it derives and delivers telephony services without interfaces to the public network. According to our definition, next-gen voice currently includes the following product types:

• Softswitches: Provide call control and SS7 signaling, and control next gen voice hardware using H.323, MGCP, or MeGaCo; they may also provide Class 5 features or enhanced packet voice services; they reside on a server or a dedicated hardware platform and may be distributed over several platforms

• Voice over broadband gateways: Provide gateway function between data access networks and the public network; connect to Class 5 switches using GR-303 or V5 interfaces

• Media gateways: Provide switching and adaptation between the public network and data networks for voice switching, Internet offload or voice over packet; they have both packet/cell and public network interfaces with a high density of DS0s; some gateways may also deliver packetized voice services to customer premises by providing a subset of Class 5 switch functions and SS7 signaling via an integrated softswitch

• RAC VoIP gateways: NEBS level 3 compliant, remote access concentrator-based media gateways using TDM interfaces to connect to the public network, and packet interfaces to connect to the data network; support H.323 with future migration to SIP and H.248

• Voice application servers: Provide traditional Class 5 switch services or new IP-based voice and multimedia applications; they communicate with softswitches using an SIP interface

• Broadband loop carriers: Provide POTS, packetized voice, and broadband data services from a platform designed for remote terminals.

Despite service provider interest and inevitable deployment, the next-gen voice equipment market remains immature and crowded with numerous competing and complementary approaches and strategies. Many of the current vendors will eventually succumb to classic market survival-of-the-fittest and consolidate to perhaps a half dozen companies. Some equipment vendors have spent the past 5 years developing next-gen voice products and frameworks, and others are immersed in or are just now emerging from the development cycle. Legacy voice switch vendors have entered the fray as well, realizing that the time is nigh for them to be serious about packet voice. We'll see some major changes in the vendor landscape over the next two years.

In these early stages of deployment, more next-gen voice hardware is sold than software (the worldwide equipment revenue mix for 2001 was 87% hardware and 13% software) because initial builds focus on footprint and tactical cost-saving measures such as packet tandem and Internet offload, which have fewer softswitch requirements. Software-like softswitches and voice application servers, which are earlier than gateways in their development cycles but are the foundation of strategic applications such as enhanced voice services, will make up a larger portion of next-gen product revenues as later phases of deployment are implemented.

Due to overall product immaturity and incomplete solutions, service providers are forced to act as systems integrators to get the diverse elements to work together, which is not the ideal solution for them. Ideally, the softswitch architecture allows for and promotes total decoupling of call processing, control and management, and service creation. This disaggregation is the polar opposite of the monolithic Class 4/5 voice switches, so that service providers can mix and match products from different vendors for a total solution. Reality, however, calls for proprietary interfaces, dashes of vendor secret sauce and lots of interoperability work between products and vendors. That standards are in flux adds to the problem, and until standards gel, there is much integration work required of service providers and vendors to get next-gen voice up and running.

So, vendor partnership menus abound, and service providers are presented with a plethora of vendors offering products with only a degree of interoperability. As the market consolidates, solutions will become more pervasive and interoperable, but until then service providers making early investments in next-gen voice will feel the pain.

Figure 1 shows today’s major service provider segments across the 3 tiers, and the biggest spenders will be U.S./Canadian ILECs, IXCs and MSOs. Their notoriously long selection process combined with the relative immaturity of the next-gen voice equipment market will retard their spending in the near term. The rollout will be glacial--it will be slow, but it will change the telephony landscape when it arrives. Other service provider types may jump-start innovation or charge ahead of the big players but will have a much smaller effect on the overall next-gen voice market.

We expect that U.S./Canadian service providers will grow their expenditures on next-gen voice equipment 468%, from $775 million to $4.4 billion between 2002 and 2006. Despite shrinking capital budgets exacerbated by dwindling numbers of service providers, next-gen voice equipment lays the foundation for the all-packet future, and investments will be made in the equipment and service rollout.

However, the replacement of legacy voice switches is years away, as many billions of unamortized dollars are tied up in legacy equipment. Next-gen voice equipment will be used to extend the life of legacy TDM and will be installed in new locations as incumbents expand out of region. This replacement of the current generation of equipment with the next will be a 15- to 20-year cycle.

For service providers, the next few years will be focused on buildout and tactical cost-saving approaches. The long-term strategic deployment of equipment for consumable services will follow. During this time the vendor community will consolidate to 4 to 6 worldwide suppliers with a few remaining niche players. As end users, we must exercise patience before we will see the much discussed and demonstrated next generation of voice networks and services. I'm confident that the voices I hear are telling the truth about the future.

Kevin Mitchell is Directing Analyst, Service Provider Networks for Infonetics Research, Woburn, MA. He can be reached at Kevin@infonetics.com.

Visit Infonetics online.