AT&T Rolling out Femtos Next Year


There is a pot of wireless voice and data gold at the end of the rainbow, and it's a big one. The question for carriers and service providers is how to get from here to there.


The telecommunications network is in transition. There is a confusing sea of approaches to combining wired and wireless networks in a way marries the exciting features of the new network with the time-tested stability of the old. Carriers want to do this in a way that takes advantage, whenever possible, of the cheaper IP network.


Femtocells are a key player on the wireless side. One annoying limitation that planners have to deal with is that the cellular signals today tend to have a hard time permeating structures. This is worse in 3G than 2G and will be worse still in the 4G world. Femtocells-femtos, to their friends-are small base stations that can attach to the end of a DSL or cable line. Femtos can do two things that carriers love: Eliminate coverage problems (since they operate within the structure) and bypass expensive cellular spectrum.


AT&T yesterday announced that it will roll out femtos next year. The story says broad customer testing will begin during the second quarter. This, the piece says, will be proceeded by a smaller employee test. The company has an existing agreement with ip.access. The five-year, half-billion dollar deal aims to drive per-femto prices as low as $100. The story does a good job of outlining the potential of-and significant hurdles to-femto technology.


Here's one reaction to the AT&T announcement, from Michael Finneran of dBrn Associates: Not so good. Finneran sees femtos as a way for the carrier to have customers pay-via their DSL or cable modem subscriptions-to make up for the shortcomings of their wireless networks. A writer for The Dallas Morning News makes the same point, and suggests that carriers eventually will offer it for free. He doesn't use the term, but adding goodies such as this generally is referred to as a "sticky" service. While it doesn't generate revenue in in its own right, it makes it more likely that people will stick with his or her current provider.


Finneran offers a good comparison of femtos and a similar service, HotSpot@Home, from T-Mobile. The key differentiator, he says, is that the wireless element of the T-Mobile service is Wi-Fi and the service thus falls into the Unlicensed Mobile Access (UMA) category.


Femtocells make intuitive sense, but it is important to quantify their real-world benefits. The Femto Forum, hardly an objective observer, says that they are potent, according to this PC World report. (It would be a headline if it didn't.) For what it's worth, the organization just released a report saying that they offer a greater than 10-fold increase in capacity over equivalent non-femto enhanced networks.


This, according to Femto Forum Chairman Simon Saunders, is about the increase in capacity experienced in the switch of mobile phones from analog to digital. The story says that femtos haven't taken off in a way anticipated by observers, and suggests that the fears about interference with the macro base stations is behind the reticence. Saunders says that coming standards will address the concerns. Marketing issues still remain, however, says Infonetics analyst Richard Webb.


This is a nice overview of the benefits of femtocells at the British site SME Web, though it does gloss a bit over the challenges. The writer suggests that small- and medium-size businesses will be an early market for femtos, even though the consumer side of the equation has gotten most of the attention to date. While femtos can be a substitution for Wi-Fi-and eliminate the need for dual-mode phones in fixed mobile convergence (FMC) scenarios-the two technologies will coexist well into the future, according to the writer.