Femtocells: Obscure name, bright future

For all the speed mobile operators are building into their networks, a lot of it doesn't reach subscribers trying to make calls or surf the Web indoors. While some carriers look to Wi-Fi and dual-mode phones to bridge the gap, another technology is coming along that could work with the handsets customers have today.

Femtocells, named after a small order of size in physics, are cellular base stations for homes and offices. The concept is not new, but the constant smaller-and-cheaper trend in processors has finally made it feasible. Several carriers are considering the technology, according to vendors and analysts, and the first large commercial deployments may hit the ground next year.

Mobile operators have invested billions of dollars in licenses and infrastructure for fast 3G (third-generation) mobile data networks. Yet subscribers often can't get the throughput they've paid for when they're indoors -- exactly where people are most likely to use network-intensive multimedia services, said ABI Research analyst Stuart Carlaw. Picocells, often used in office buildings and campuses, are too big for home use.

To get the bandwidth where it's needed, mobile providers are looking to the DSL (digital subscriber line) and cable modems most subscribers have in their homes. This also saves them money on backhaul, the links that tie the cellular to the wired network, which carriers typically have to lease. At the moment, giving better in-home coverage means using Wi-Fi. T-Mobile USA, BT Group and France Telecom all have services that let subscribers use the cellular network outside and automatically switch to Wi-Fi indoors.

Wi-Fi router options are plentiful and inexpensive. But when it comes to phones, subscribers who want to go this route have only a handful of models to choose from. Because femtocells are just like regular cellular base stations, any phone for the carrier's network can be used with them. From the carrier's perspective, a femtocell system is also preferable because it has the same security between the phone and the cell, according to Wen Tong, a research fellow at Nortel Networks.

The 3G bandwidth crunch is real. For example, EV-DO (Evolution-Data Optimized) 3G technology is supposed to deliver between 400K bps (bits per second) and 600K bps of throughput on average. But if just 20 subscribers try to access data services at the same time, a typical EV-DO base station can only provide 100K bps to each, Tong said. Supporting more users over the base station's range, which typically is several kilometers, requires expensive upgrades.

Cost has been the major barrier to femtocell adoption. Nokia and Telefonaktiebolaget LM Ericsson tried out the concept more than 10 years ago, but the equipment for homes and offices was too big and too expensive, ABI's Carlaw said. Moore's Law, which predicts that microprocessors will continually get faster and smaller, soon will make that customer equipment small and cheap enough to deploy commercially. Carlaw believes that in 2009, when most observers expect femtocells to hit the market in force, a cell with an integrated broadband modem will cost the carrier about US$120 (AU$150.8).

Paying for an entirely new device will be a tricky issue in the price-sensitive cellular business, according to In-Stat analyst Allen Nogee. U.S. carriers already subsidize handsets and will be reluctant to take on the additional cost, he said. And European and Asian operators aren't used to offering subsidies at all. It may be hard to persuade subscribers to pay for something that essentially saves the carrier from having to build a better network and deliver the performance it promised in the first place, Nogee said.

• Bookmark this page
• Share this article
  •  
  • facebook
  • slashdot
  • digg
  • Reddit
  • stumbleupon
  • linkedin
  • twitter
• Got more on this story? Email ARN
• Follow ARN on twitter

• Share
  Share this article

  •  google+
  • facebookfacebook
  • slashdotslashdot
  • diggdigg
  • RedditReddit
  • stumbleuponstumbleupon
  • linkedinlinkedin
  • twittertwitter
• print
• email
• Bookmark