In a recent interview, Lauren Town, Head of Marketing at Orange, said the operator was not looking to replace its current Unik WiFi UMA service with 3G femtocells. What is the likely evolutionary path for existing UMA services and can they co-exist with femtocells?

Some commentators have disparaged cellular/WiFi dual mode services, highlighting a lack of available handsets, reduced battery life and somewhat clunky handover between the two systems. Indeed, we have seen several UMA dual mode services introduced with varying degrees of success. France Telecom/Orange, which it provides in France under the Unik brand, has had good takeup. T-Mobile USA, with it's hotspot@home brand also seems to be making good headway. However, it is a mixed bag, with services in some countries being withdrawn due to lacklustre  takeup, and BT Fusion in the UK reportedly achieving only some 45,000 subscribers.

Orange's view was that the femtocell business case, at least in the domestic market, is very similar to that for WiFi/UMA. A wider range of UMA/3G handsets is now available (Orange will offer 25 by end 2008), and thus the additional benefits of longer battery life, easier configuration are not substantially different. The demonstrations of the 3G connected home were not sufficiently different from what might be achievable over WiFi, although I suspect the configuration/setup of these might still be fairly complex.

In the case of new opportunities, it was suggested that it is quite possible FT/Orange would consider using 3G femtocells in the future - once the price and maturity of the solution exceeds what their UMA solutions offer today. They clearly distinguish between business and domestic markets, and may have separate offers in each case.

There have been races to achieve market penetration of new technologies many times before, and will be many times again. VHS vs Betamax (actually Phillips 2000 was probably the best videotape spec of the lot), HD-DVD vs Blueray, CDMA vs GSM. So it is with Dual Mode WiFi UMA and 3G femtocells. If Dual Mode WiFi can achieve some good market share before 3G femtocells achieve volume, then they will be credible for some time. If not, handset choice will dwindle as vendors prioritise other features to grow their market share.

We've described a variety of femtocell competitor solutions, but a summary of tradeoffs between 3G femtocells and WiFi UMA as we see it today are:

A technical evolution of dual mode WiFi to femtocells would be to provide customers with integrated broadband modems that do both - indeed this is what we are seeing developed and demonstrated by several of the large DSL modem vendors today. This equipment could support both WiFi and 3G femtocells. We'd estimate the additional cost of femtocell functionality of at least $200 in volume (ex-factory) today, with an expectation this will reduce rapidly as volume increases. However, operators would need to see strong benefits/business case to implement both services simultaneously.

Another scenario is that operators may wait for 4G LTE femtocells and use this major change to justify upgrading from WiFi.

Our prediction is that existing operators with WiFi solutions will continue with them for quite some time and then either migrate to 3G femtocells in the medium to long term. If they are withdrawn due to lack of takeup, operators may be wondering whether the same issues and same business case may also falter for femtocells.

However for "greenfield" deployments, at least those planned for 12 months out or later, 3G femtocells pose a strong competitive threat to WiFi dual mode handsets, and we would expect them to be the preferred choice in the medium term.