Which way: Femtocells or UMA?

Discussions are again resurfacing about which technology should be used to improve coverage; Femtocells or UMA. Honestly, I have not been a big supporter of UMA (and you dont have to agree with me) and earlier this year when Nokia raised doubts about the technology, i stopped following it completely. Kineto has been the biggest supporter of UMA and is still carrying the torch.

Kineto recently received over $15 million in investment from Motorola that it plans to use to increase its reach in UMA. Last month it had received investment from NEC. One of the reasons for the resurgence could be because the UMA technology has matured since being used for some time. T-Mobile has already rolled Hotspot @ Home using UMA. Also more phones are now available supporting UMA.

One of the drawbacks that will always be present in case of UMA is that special handsets will be required that would support WiFi as UMA is based on 802.11. This means more expensive phones and higher consumption of power leading to smaller battery lives. One more problem with UMA is the interference due to other ISM band devices and there are many technologies like Bluetooth, etc that are competing for the band.

The drawback in case of Femtocells could be that their price is still quite high as complete Node B + RNC functionality is generally available in a Femtocell and at the same time all the aspects have not yet been standardised. Along with these, Femtocells that use the same spectrum as that of the operator can cause interference with the Macro cell. This would in turn require very clever management of spectrum frequencies, etc.

A possible long term solution could be (but I dont see anyone agreeing to it right now) that phones with UMA technology become more common and a combined UMA + Femtocell equipment is rolled out by the operator. At the end user premises, depending on the strength of rf reception, the equipment can either use UMA or normal Femtocell functionality. If this idea is agreed upon, then there would still be couple of years before all interworking and other technological problems are ironed out.