These are difficult times for mobile operators. After years of glamour and high growth, stocks are being re-rated as utilities. In many parts of the world markets are saturated and the only scope for growth is taking market share.
To complicate things there is the threat from new entrants using IP and broadband, with fixed-line service providers using UMA (universal mobile access), FMC (fixed mobile convergence) and voice-over-WiFi (wireless VoIP) to deliver cheap service indoors.
Just to add salt to the wound, 3G does not actually work that well indoors. Or to be more accurate, the combination of high frequencies (2.1GHz), high data rates (16QAM constellations), long range and attenuation from walls is not a good mix.
But there is growing discussion about technology that could shift the balance back towards spectrum owners – the 3G home-basestation. The femtocell, or residential access point, is simply a box that connects to a broadband line. But instead of Bluetooth or voice-over-WiFi, the connection is standard 3G connecting to any handset.
This would appear much like UMA with the same advantages. At home calls would be cheap or even free and this will work with any handset, so there is no need to get a new dual-mode terminal. And it is convenient because the consumer is using their standard mobile phone.
The benefits for a carrier are stronger – it delivers genuine differentiation, in a cost-effective but very profitable way. The operator has a powerful tool to sign up everyone in the household, driving “lock in” and reducing customer
turnover. The carrier will probably want to deliver the broadband service too and bundling is a proven way to keep customers and increase revenue per user.
The femtocell basestation also increases capacity at just the place people need it, which is especially good for data services.
Current mobile systems need to shout really loudly to blast a signal through the walls, which then causes noise that impacts the service of other users outside. By putting the basestation inside the walls, the user gets better quality and the users outdoors no longer have to cope with the shouting. So the femtocell increases network efficiency by
offloading the (expensive) macrocell.
This is the “Shannon meets Isenberg” logic. Isenberg famously predicted that intelligence moves to the edge, with lots of smart devices at the periphery.
Shannon’s law says that the best way to get efficiency and improve performance is to improve the signal-to-noise ratio, and what better way than to have basestations where they are needed at the edge, with short distances, less attenuation and less interference.
By using similar logic and similar architectures, this style of FMC becomes not a threat for a mobile carrier, but an opportunity to take customers away from fixed line competitors. But this is not easy to do.
One major problem is cost. The whole idea only works if the femtocell is cheap enough.
Even with fierce competition, a 3G macro-basestation is typically $20,000. For this idea to fly, it must be orders of magnitude cheaper – perhaps one per cent of that price. With the latest generation of multi-core DSP to deliver sufficient performance, this is now do-able.
The next challenge is to integrate this into the radio network and manage interference, in a scalable and completely automatic way. The precise techniques for addressing this are proprietary to each OEM, but a number of different approaches have been developed.
Lastly, how to integrate this into the carrier network? It must be plug-and-play, easy for the customer to install and use and very scalable.
Different manufacturers see different approaches to network architecture. Some wish to leverage their existing network, using broadband to connect to the standard 3GPP network. Others see this as the step to an all-IP network using UMA or IMS.
The basestation is better described as an access point, with a “collapsed stack” that integrates functionality from the RNC and SGSN. This then allows the connection to leap-frog from the basestation, through the internet directly to the core. There are now commercial stacks and complete reference designs to implement either approach.
Companies who excel at developing sophisticated macro-basestations may not be those best positioned to deliver consumer electronics as part of a broadband network. A likely consequence is the rise of reference designs from semiconductor suppliers, as seen in DSL and Wi-Fi markets.
In a time where competition in the mobile world often seems to reduce to price wars or who has the prettier logo, it is encouraging to see there are carriers looking to use new technology to drive genuine service innovation. And whichever approach wins, we can be sure that the increased differentiation and competitive pressure will drive better services for everyone.
Rupert Baines is vice president of marketing at PicoChip