It's voice's turn to tackle 3G

Ever since NTT DoCoMo launched the first 3G network in Japan in 2001, every year was supposed to be the year of 3G: 2002, 2003 and our present annum. Despite the over-inflated hype and the accompanying disappointment, there is little argument that 2005 will legitimately be the year of 3G in the U.S.

Both Sprint PCS and Verizon Wireless have committed to nationwide rollouts of CDMA 1X EV-DO in 2005, and while the newly engorged Cingular Wireless hasn't exactly laid out a deployment plan, it already has a head start with the six UMTS market rollouts it inherited from AT&T Wireless during the acquisition. With Verizon and Sprint breathing down its neck, Cingular can't hold out for too long without a rollout of its own.

That much activity in the world's largest market naturally has vendors in a tizzy. They stand not only to benefit from the nationwide deployments to come next year but also a base access platform from which to launch the next generation of multimedia applications and infrastructure. Vendors are producing enhanced multimedia messaging centers, push-to-view infrastructure and video streaming servers, all of which require the high-data rates of UMTS and DO networks to thrive. But oddly enough, what the application vendors seem to be talking up most is voice.

“2005 will be the year that voice technology will evolve,” said Raghu Rau, Motorola corporate vice president and director of marketing for the global telecom solutions sector.

In its 11 predictions for 2005, analyst group inCode divined that the wireless industry will finally give up its quest for the elusive killer app and focus on differentiating the voice service already representing 95% of its revenues. The trick, however, is to shed the commodity status of voice service itself. According to inCode analyst Jorge Fuenzalida, no differentiated voice service has been introduced since Nextel ran off with to push-to-talk. So expect carriers to start breathing new life into voice, offering applications like priority calling, group voice messaging and network-based cellular conferencing. In much the same way wireline broadband operators are tiering their data speeds, voice operators will look to offer premium levels of voice service, Fuenzalida said.

Not surprisingly, vendors are keying in on the trend, pushing advanced voice products and paying particular attention to IP Multimedia Subsystems (IMS) technology. IMS is the bridge between wireless and wireline networks and is critical between enabling dual-mode Wi-Fi and cellular handsets and eventually total convergence. While convergence is gathering a lot of momentum in the industry, vendors are also touting IMS's other charms. IMS provides a common platform, off of which carriers can deploy multiple solutions, discarding the old one-box-one-service notion of wireless infrastructure. Because it is SIP-based, an IMS platform can introduce enhanced services customers have become accustomed to in wireline VoIP over the cellular network, said Mark Morell, director of carrier network strategic marketing for Nortel. For instance, presence services could allow customers to initiate a voice call from an instant messaging client or select numerous contacts from an address book and initiate an on-the-fly conference call.

“Carriers will have IP and SIP in their networks,” Morell said. “There's no limit to the types of services they can hang off of those technologies.”

IP is already penetrating the wireless network. Carriers have used IP transport for years, but it's starting to make its way to the edge and core. UTStarcom has had success internationally selling its IP basestation portfolio, and according to that vendor, one U.S. carrier has committed to deploying its gear in 2005 for new rural builds and enterprise pico cell deployments. Motorola's Rau said interest in softswitch mobile switching centers is starting to pique. Brazil's Vivo purchased an astonishing 25 softswitches, and Rau said he expects others carriers to follow suit in 2005.

The IP holy grail, however, is end-to-end VoIP, which would finally end the division between voice and data over the airwaves. While end-to-end VoIP seemed a distant hope just last year, carriers and vendors have become more enthusiastic as the VoIP over broadband networks have gained prominence and focus has intensified on CDMA 1X EV-DO technology.

With both Sprint and Verizon Wireless committed to nationwide rollouts of EV-DO, the industry attention has turned away from EV-DV — which sports a voice channel similar to that of other circuit-switched CDMA technologies — to EV-DO revision A, which incorporates VoIP directly into the packet stream. In fact, if vendors' initial trials of Rev A live up to the specifications set for it, EV-DV could fall to the wayside, said Hakan Eriksson, chief technology officer for Ericsson.

“The prerequisite for taking EV-DV out of the equation is getting VoIP to work over revision A,” Eriksson. “VoIP will be latency-challenged. I don't think we can guarantee it will work well.”

While vendors usually tout the peak data capacities of their networks, latency is another story entirely. EV-DO may offer superior overall bandwidth to UMTS, its latency is much higher, 300 milliseconds compared to UMTS's 170 milliseconds, Eriksson said. Latency, simply put, is the amount of delay in the network, which is why when performing actions such as loading Web pages, UMTS can seem to be just as fast as EV-DO. Neither technology's latency, however, is low enough to support VoIP. Today's circuit-switched networks latency is about 100 milliseconds, so any IP network would have to match or exceed that benchmark to be viable. Both Rev A and the next upgrade to UMTS, high-seed downlink packet access (HSDPA), have projected latency's of 100 milliseconds. Ideally the industry would like to bring that number down to 70 milliseconds to fully support VoIP, Eriksson said.

Of course, the industry won't be spending all of 2005 merely implementing the current generation of network technologies. They'll also spend plenty of time hyping the technologies of 2006, 2007 and beyond. Vendors are already tossing the term 4G around, listing off some rather incredible, albeit still theoretical, numbers. Ericsson, for one, is defining a future 4G network as one that could support 100 Mb/s while mobile and 1 Gb/s in a fixed wireless environment. Other vendors are already talking up orthogonal frequency division multiplexing (OFDM) and its successor multiple input/multiple output OFDM (MIMO).

The first standards-based OFDM implementation will come out of the IEEE's 802.16e specification, and products will hit the market in 2006 and 2007 with MIMO a year or two further down the road. The technologies claim to support speeds as high as 600 Mb/s. But with the lack of a business model today that demands the need for sub-1 Mb/s bandwidths, vendors are focusing on other benefits of the technology.

“These will plug right into existing 3G technologies,” Morrell said. “OFDM will bring one and half to two times the spectral efficiencies of existing 3G technologies. MIMO adds four times the spectral efficiency to those same technologies. They give operators the best possible efficiency and throughput.”

InCode's Top 11 Wireless Predictions for 2005

1. Carrier consolidation — 2005 will see at least one more merger of major wireless carriers, though InCode wouldn't hazard to guess which carriers.

2. Wireless public safety — Wireless will take a critical role in national security and public safety, integrating emergency services directly into the network.

3. Wireless 411 gets a wobbly start — A national cell phone directory is dead in the water, but businesses will participate in a cellular Yellow Pages.

4. Carriers focus on differentiating voice — Voice will be the key service in 2005, and carriers will find ways of offering tiered and advanced voice services a la VoIP.

5. Broadband integration — Carriers will begin bridging networks in 2005, allowing customers to roam from Wi-Fi to 2.5G to 3G to broadband wireless.

6. Enterprise makes itself heard — With wireless becoming critical to the daily business of enterprises, corporations will begin demanding special treatment from the carriers.

7. Wireless gets bundled with everything — The wireless bundling trend will expand to multimedia and cable companies.

8. Carriers fight to control content — U.S. carriers tightly control content over their networks today, but in 2005, they'll start losing that grip as vendors and outside content providers sell content directly to consumers.

9. Spam spares wireless no longer — High-profile spamming incidents will drive the issue of wireless spam into the limelight.

10. Coverage moves indoors — In-building coverage will become a big differentiator for wireless carriers as consumers and enterprises become less and less tolerant of spotty coverage and dropped calls.

11. Adult entertainment stays mainly in Europe — Domestic carriers will be reluctant to launch it due to the more conservative social landscape in the U.S.

Source: InCode