BARCELONA--When the industry began laying down its stakes in 3G, Chelmsford, Mass.-based Airvana put all of its bets on a single technology, CDMA 1X EV-DO. It turned out to be a smart move. Its software, chassis card and radio network controller technology are the standard in a good deal of the world’s 3G CDMA networks—publicly acknowledged by Nortel Networks as its base station technology for EV-DO and Revision A and not-so-secretly used by other Tier I vendors.

But as CDMA makes its way up the 3G evolutionary path toward Revision B and Revision C--which is not even CDMA-based--technologies, Airvana is hedging its bets. The company has refocused its efforts on fixed/mobile convergence and radio-agnostic technologies like its new Universal Access Gateway, which would regulate traffic and hand-off between any IP mobile network—EV-DO, UMTS, WiMAX or otherwise.

Airvana vice president of business development Paul Callahan was quick to say it hasn’t given up on CDMA technologies--far from it. EV-DO is its bread and butter, and it will continue to support it long after the final CDMA carrier upgrades to Rev. A. But Airvana is a small company that can do little to move the industry in one direction or the other. It’s dependent ultimately on the portfolios its vendor partners choose to develop and the technologies those vendors’ carrier customers choose to deploy. Callahan said Airvana has decided to make itself less dependent on a particular technology and apply its expertise to a technology that can be used across all standards and networks.

While a universal access gateway may seem a strange choice for a company whose prime focus has been on radio access networks, Callahan said it’s not as much of a stretch as people might think. Airvana’s primary contribution to EV-DO was not in the radios themselves, but in the management of IP call sessions, VoIP and the peculiarities of handing off real-time IP traffic on a mobile network.

“The heart of our intellectual property is in not in the base station, but in the [Radio Network Controller],” Callahan said. “With Rev. A we learned a lot about QoS and engineering around VoIP and VoIP-based push-to-talk. We can apply that knowledge to other technologies.”

While the CDMA vendor communities are on track to have Rev. C or Ultra Mobile Broadband (UMB) commercially ready before its GSM alternative Long Term Evolution emerges in 2009, a big question mark still hangs over the technology. Unlike previous CDMA generations, UMB is not a simple software upgrade to 1X base station. It requires a completely new radio access technology based on Orthogonal Frequency Division Multiplexing (OFDM) rather than CDMA. That makes the migration choice for existing CDMA carriers less than obvious since they will be required to build new networks. Most vendors have already written off the intermediate technology Rev. B, which essentially merges EV-DO channels into a single a high-capacity channel, and now some are starting to question Rev. C.

CDMA rising star ZTE is chugging ahead with its UMB research, contributing to the 3G standard and developing the radio technology in its labs, but ZTE vice president of radio wireless technology Shen Donglin said it has not yet committed to building a UMB product.

“Research and development into the technology is relatively inexpensive,” Shen said, but not so for developing a commercial track product, he added.