Firing up those femtocells

15.10.2009
The folks over at the , in conjunction with the European telecommunications standards institute, recently announced a "plugfest" for March 2010. The plugfest will serve as a forum to for interoperability tests among network gateways, security gateways, femtocell access points and chipsets.

For those who haven't been following this space, femtocells are tiny base stations designed for use in residences and branch offices, capable of supporting roughly two to to five phones/mobile devices. Although the plugfest will focus specifically on universal mobile telecommunications (UMTS) technology, femtocells can be deployed with any flavor of wireless technology, including long-term evolution (LTE), the successor to UMTS, CDMA and WiMAX.

What's the big deal? For providers, femtocells offer the ability to deploy and manage wireless indoor networks, thus offering "managed wireless LANS" as a service offering. For users, the benefit is the ability to offload the management of WLANs and small-office/branch-office wireless support to carriers for a fixed monthly fee.

And for these reasons, femtocells seem to be taking off. Major U.S. and global carriers, including Vodafone, AT&T, and Verizon, have announced femtocell trials. This summer, Vodafone trialed femtocells in Italy, Spain, and the UK. Verizon introduced femtocell service as a consumer offering in the U.S. this past January, followed shortly by AT&T.

Femtocells raise interesting issues, though. First is the viability of managing a several-orders-of-magnitude increase in the number of base stations. Instead of worrying about thousands or tens of thousands of cell towers, carriers will need to worry about millions or tens of millions of femtocell stations.

That said, scaling is one of the few things carriers are actually good at (unlike, say, customer service or market innovation). So while I don't doubt there will be stumbles along the way, I'm confident that carriers will pull this one off — assuming, of course, there's enough demand to generate scaling problems in the first place.