Femtocells were developed to aid 3G networks, but their real impact may be much greater on the successors: the 4G networks, WiMax and LTE.
Many of the issues facing femtos in the 3G world were addressed at last month's Femtocells Europe 2008,but the conference also considered how femtos could shape the delivery of fourth generation services.
Indoor base stations have obvious benefits - they improve coverage for users and provide the operator with cheap backhaul over the user's own broadband. But operator trials on 3G networks have been slow to emerge, for reasons we've recently discussed.
The case for femtos is much clearer on the data-centric 4G networks which are due to succeed today's 3G networks over the next few years. Femtocells have to be bolted onto today's 3G networks, but can be planned into 4G networks. It's widely expected that femtos will be a fundamental part of LTE networks. The Next Generation Mobile Network Alliance has publicly promised to work with the Femto Forum, for instance.
Meanwhile, there have been suggestions that early WiMax networks might do away with macro cells at the outset, and launch with indoor cells only (a model that looks strangely like public Wi-Fi).
But the picture will be more complicated because none of these developments is happening in isolation.
Better by design/
Femtos' fitness for 4G is not just a matter of timing. They will suit the technology well, and iron out difficulties with speed-boosting technologies added into 4G, points out Simon Saunders, chair of the Femto Forum. LTE includes MIMO, which uses reflections of radio signals to set up multiple paths between sending and receiving antennas - but this is much more of a benefit indoors, where femtos live, he points out.
"LTE and WiMax have a large number of modes," says Saunders, and the top data rates may only be achieved with femtocells in buildings.
This could mean that LTE will actually be led by femtocells before it is deployed on the wider network. Users could be sold a 4G handset and a femtocell, so they can use the new technology at home, and fall back to 3G outside while LTE is being implemented: "LTE won't necessarily need coverage everywhere," says Saunders.
LTE with a 3G femto?
To confuse things, though, there's an opposite possibility. With all this tricky MIMO ability, won't 4G femtos be costly to develop, and therefore unattractive? Saunders doesn't think so: "The thing about LTE or other 4G networks, is they will be building on the experience of the 3G femtocell. There are technical challenges, but the challenges of getting cellular infrastructure into consumer volumes will already have been done - 3G femto cells have cleaned the pipes for future femto cells."
The Femto Forum and the Next Generation Mobile Network Alliancewill be specifying 4G femtos, so they shouldn't suffer a delay or a price hike, says Saunders.
If for any reason LTE femtos were difficult, however, we could have the interesting possibility of LTE networks relying on 3G femtocells. Any LTE handset will be able to fall back to 3G - and using a 3G base station indoors should be preferable to connecting through the walls to an overloaded LTE base station outside. The data rates indoors on a 3G femtocell would be superior to those you'd get connecting to a remote LTE cell.
Will femtos delay LTE?
But there's another thought. Could LTE itself be delayed by the very existence of femtocells? Operators want to sell data services, and adding femtos will help them do that on 3G networks. What if those femtos make the 3G data services so good they don't need to implement LTE - or can put if off for a couple of years?
That's a possible scenario. Femto vendors are very fond of pointing out that most users of wide area data tend to consume their data indoors (they also tend to do it on devices that also have Wi-Fi, but that's another story). It might be that a femto-equipped 3G network meets users' needs in the short term: "Femtocells may delay the need for LTE," says Andrew Parkin-White, research director at Analysys Mason.
Realistically, however, the specialised applications that make strong use of femtocells will probably have to wait till LTE handsets and femtos arrive.
Femto vendors like ip.access are talking about clever applications that do things like downloading podcasts automatically when the user arrives home. It's been pointed out that current handsets cannot tell the difference between femto and macro cells (that's how the femtos are able to work with any handset). So those clever applications aren't going to work well, without some more or less clumsy way of telling the handset when it's on a femto.
There will be other aspects to handsets or devices that suit them to femto use, says analyst Dean Bubley of Disruptive Analysis. For instance, if people really do start to watch movies on their phones, they may want to have a kick-stand to prop it up on a table. Handsets will inevitably change to do the things that femtos allow - and given the development cycle of handsets, that won't happen till LTE and other 4G networks are well on the way.
Whatever the requirements of a femto-aware handset, groups like the Femto Forum and the NGMNA will be working on them, says Simon Saunders. "When we start to look at next generation there are some things we can do," he says. "We can give users a good experience in 3G without contextual awareness, but it will definitely be part of the next generation experience."