It’s an old idea finally seeing the light of day: a multi-zone wireless phone service with handsets that work at home, on the go, and in the office. The question is, which set of technologies will ultimately drive it forward?

Will it be Wi-Fi and dual-mode 3G/Wi-Fi phones, or femtocells, the 3G home base stations that Sprint has already begun to deploy and most other mobile carriers are now, or soon will be, testing? 

And which players in a complex emerging ecosystem—mobile carriers, infrastructure equipment suppliers, femtocell developers, even home network vendors—will ultimately determine the shape of the market?

Surveying the battle
Network equipment maker Kineto Wireless has an interesting vantage point from which to watch it all unfold. The company pioneered the use of UMA (unlicensed mobile access), the technology that lets GSM-based mobile carriers hand off calls on dual-mode phones from a 3G base station to a home or office Wi-Fi router and then send the call over the Internet. 

Since UMA was originally developed to enable fixed mobile convergence (FMC) using dual-mode handsets, you would expect Kineto to be solidly in the Wi-Fi camp. But not so. UMA, it turns out, can also play in the femtocell world.

Kineto is partnering with Japanese electronics giant NEC, Corp. and network equipment maker Motorola, Inc. to develop femtocell solutions, and has already shown that its UMA network equipment can interoperate with femtocells from Ubiquisys, a pioneer in the femtocell equipment market—and one in which Google recently invested.

“A home-zone [FMC] service is something mobile operators have to do,” says Kineto’s associate vice president of marketing, Steve Shaw. “We don’t care whether it’s Wi-Fi or femtocells. We’re playing both sides. It’s a great position to be in.”

There seems little doubt now that most North American operators will eventually offer FMC services. The evidence that consumers want them is mounting. According to various studies, up to 40% of mobile calls today are made from home or inside other buildings—where cellular networks traditionally provide poor coverage or only provide good coverage at great expense.

The big guns
T-Mobile fired the first shot, launching its Wi-Fi-based Hotspot@Home service in June. In September, Sprint Nextel started selling Samsung Airave femtocells in Denver and Minneapolis. They let users with any Sprint phone—dual-mode handset not required—make unlimited calls at home for $15 to $30 a month.

But UMA’s place—and therefore Kineto’s—is by no means assured in the femtocell world. It’s only a lock in GSM networks that opt for the Wi-Fi approach. If there.

The UMA client on a dual-mode handset finds authorized Wi-Fi networks and connects to them. It then converts the cellular stream to IP packets and sends them over the Internet through a secure tunnel to a mobile carrier’s RAN (radio access network) gateway, which passes them to the core mobile network.

Femtocells eliminate the need for dual-mode phones and Wi-Fi because they use the same technology as traditional cellular base stations. But they do need a mechanism for turning cellular calls into IP packets and sending them securely over the Internet. Kineto thinks it should be UMA.

“UMA is already a defined standard,” Shaw points out. “It has already been proven out by operators, it’s already deployed, it’s got all the bells and whistles you need. So we’re saying, ‘Look Mr. Operator, why don’t you take something you already know and love and just use it for backhauling femtocells as well.’”

But UMA is no slam dunk. Infrastructure equipment makers Alcatel-Lucent and Nokia Siemens Networks, and start-up femtocell maker ip.access are all working on their own very similar, but proprietary technologies.

Why don’t they just use UMA? “They’re idiots,” Shaw jokes, then turns serious. In the case of the traditional infrastructure equipment makers, he speculates, it’s because they want to keep control of the market.

“If you’re in the business of selling RAN gateways, you really don’t want an open standard interface [like UMA] between the gateway and the femtocell because then anybody can build femtocells,” Shaw says.

Swearing allegiance
With a proprietary approach, only Alcatel or Nokia femtocells would work with Alcatel or Nokia network gateways. For operators, he warns, opting for a proprietary solution could mean getting locked into one femtocell maker. This would likely keep costs high—which in turn could impact the economics of the entire femtocell market.

But, the carriers are not dismissing proprietary solutions out of hand. When they put out requests for proposals, they are typically selecting one UMA and one non-UMA solution to test, Shaw says. UMA will just have to duke it out. He believes it has a key advantage, though.

UMA already has the features in place that operators will need—the mechanisms for finding and connecting to femtocells, enabling and disabling users and femtocells, and the ability to comply with regulations around location, lawful intercept, and emergency calling. The proprietary solutions do not—not yet.

Upcoming carrier femtocell trials, in which Kineto will be involved—five different lab trials starting in the fourth quarter of 2007, none of them officially announced yet—will in part be “bake-offs” between UMA and other solutions, Shaw hopes. And UMA will win because it’s ready for use in commercial systems now and the others are not. 

“At the end of the day, these guys [the carriers] are in business to make money, and the faster they can get their femtocell services to market, the better off they’re going to be. I think that’s really where UMA is going to shine.”

So where does all of this leave Wi-Fi? Shaw is cagey with his answers. As he says, only half-joking, “We love them both equally.”

But the femtocell approach has two weaknesses. Notwithstanding Sprint’s early entry into the market with the Samsung Airave units, Shaw believes the other carriers likely won’t follow suit until late 2008 at the earliest when other femtocell technologies have gone through the lab and field trial process. So timing is an issue.

Funding the war
The second problem is cost—predicted to be in the $200 neighborhood. The carriers will probably have to subsidize because it’s unlikely in the early going that consumers will be willing to pay up front. “That’s a pretty big nut for the mobile operator to swallow,” Shaw says, adding, “Wi-Fi is undoubtedly cheaper.”

Wi-Fi has another key advantage. It’s already ubiquitous in offices, hotspots, airports, friends’ homes—including overseas. “So a home-zone service based on Wi-Fi has much broader availability to the consumer than a home-zone service based on femtocells,” he says.

The only kick against the Wi-Fi/dual-mode approach is that there is little selection of dual-mode handsets—only three right now, according to Shaw. But that will change fairly quickly. He predicts there will be a dozen handsets available by the end of 2007, 30 by the end of 2008. And some will be priced at or near par with conventional phones.

So that means Wi-Fi will win? “I think the femtocell opportunity is short to medium term,” Shaw says. “But in the long term, dual-mode handsets are not going to go away.”

Right. So Wi-Fi wins. Well, not so fast. “There will always be an opportunity for femtocells,” Shaw amends. “But there will also always be an opportunity for dual-mode handsets.”