We've been waiting for enterprise solutions that will converge mobile phones with the PBX for as long as we've been doing Techworld, so it's nice to start seeing some proof that it will be done.

There's already at least one large customer for converged voice, Osaka Gas, in Japan, but the early adopters will be getting a system that doesn't let calls roam from one network to the other, or that falls down on battery life.

Nevertheless, every product announcement and every customer win is a step towards doing what we know is possible.

Enter a new player
Trapeze announced a partnership with new (to us) IP voice company DiVitas, and now Symbol and others have followed suit. The start-up is enabling the switch makers to migrate phone calls between the cellular network and a Wi-Fi system, without any involvement from the cellular operator.

A controller box on the network does the trick, working with client software on the handset. It is very much like other approaches already announced to handle the same process, most of which centre on connecting an IP PBX to the wireless LAN and using SIP. The DiVitas controller is currently a beta and should be available to buy, sometime this summer.

"The carrier is more-or-less relegated to being a pipe," comments Dean Bubley of Disruptive Analysis. "Completely the right approach for large enterprises, in my view."

Another way to look at it is pushing the frontier of the mobile operators' sphere of influence back, until they are more or less bottled up in the SIM card.

It's in contrast to the carriers' favourite solution for linking cellular phones to fixed networks, unlicensed mobile access (UMA), in which the mobile operator handles it all, the frontier moves a long way into the enterprise, and the WLAN is just a local pipe to improve coverage within the building. "Tight integration with the enterprise-grade WLAN network is totally absent in UMA and most forms of carrier-based SIP mobility management," says Bubley.

What's the drawback?
The scheme has some problems to overcome, of course. The demo uses a Windows Mobile 5.0 device, and there's no indication about what other platforms will be supported. "Doing a Windows Mobile 5.0 port on a dedicated handset is fine," says Bubley, "but they will also have to work with countless other Wi-Fi-enabled devices that corporations and their suppliers prefer."

That may not be as easy as it seems at first. Hardware manufacturers have given the impression that there are a host of dual-mode devices coming, but it may not be as simple as that.

In some cases, operators may nobble the devices to prevent them offloading paying calls onto "free" connections. This may be done by ensuring that the two networks use separate SIP communications stacks.

Disapproval from the operators may be behind the possibility that, even though adding Wi-Fi to a handset has become a cheap and obvious thing to do, not every handset vendor is going to jump to do it. Even when they put SIP on their phones, that would allow voice over Wi-Fi, handset makers may still hesitate, says Bubley: "Fewer than 15 percent of SIP-capable cellphones will have WiFi," he says. "Not just this year, but to 2010 and beyond."

Partner up with the PBX
With all this going, on vendors like DiVitas only have one option to sell their products, and that is to work with the people who are already selling the fixed side of the convergence equation: the IP PBX vendors.

These vendors want to wireless-enable their solution, and adding a cellular hand-off ability does a lot to improve their products. It lets the user save money on calls that would otherwise go on the mobile network, and it lets the vendor offer a choice of cellphones that can be used as cordless IP PBX extensions - depending, of course, on the availability of Wi-Fi enabled SIP cellphones.

"This type of dual-mode solution will have to fit in with the enterprise telephony world, where the IP-PBX vendors and their resellers are the kingmakers and gatekeepers," says Bubley.