Femtocells: Reinventing the wheel?

By Sean Jackson

femtobox

Reinventing The Wheel

For mobile operator and consumer alike, the potential attractions of femtocells include improvements to both coverage and capacity, especially indoors. There may also be opportunities for new services and reduced cost. We’ve heard something like this before of course; the question remains, are femtocells a fixed mobile convergence dream come true or a recurring technology nightmare?

In August, US CDMA carrier Sprint became the first carrier in the world to roll out a nationwide femtocell offering. Under the banner Airave, Sprint subscribers can get hold of their very own home access 1XRTT base station for an initial investment of $99.99 plus $4.99 per month for device rental. For an additional $10 or $20 per month (depending on the number of lines required) subscribers can get unlimited calling plans.

As world firsts go it is somewhat uninspiring. To recap, Sprint customers are being given the opportunity to pay, in addition to their existing plan, just shy of $160 in the first year for improved voice only coverage, while at home. You could argue, fairly reasonably, that Sprint should already be providing decent coverage for its customers.

You’d be well placed to call into question the wisdom of Sprint customers that sign up for this service rather than opting to churn to a supplier with better coverage. You might like to ponder over why, exactly, our loyal yet inadequately served, theoretical Sprint customer will be persuaded that the femtocell on offer is anything more than a repeater in disguise. If you know femtocells, you’ll know the technologies are far removed, but consumers have never heard of femtocells. They’re familiar with repeaters though, and they’re not well liked.

“Coverage is a problem, but it is a hard one to really promote a femtocell for,” says Allen Nogee, principal analyst for Wireless Technology & Infrastructure at In-Stat. “In the US with Verizon and AT&T, the two largest operators, every TV commercial you see is telling us how great their network is and how great their coverage is, and how now here is a device that is meant to improve coverage. So how do you advertise that? Coverage is great, but you may still need one of these?”

“The unlimited calling plans can be added only in addition to a mobile plan,” says Steven Hartley, senior analyst, Ovum. “Users pay Sprint once to subscribe to Sprint’s mobile service, again to improve indoor coverage and a third time to benefit from Sprint’s backhaul cost savings. This is far from compelling and suggests to customers that Sprint’s indoor coverage must be very weak if it expects customers to pay extra.”

To all intents and purposes Sprint’s offering resembles T-Mobile’s @Home service launched one month earlier. Sprint is using a femtocell, while T-Mobile uses wifi. “To a consumer they are essentially the same,” says Hartley. “Both divert fixed line minutes to the mobile operator by routing calls via a box.”

In fairness to the future potential success of ‘femtos’, the Sprint case study does not paint the full picture. It raises some important issues though, not least of which is consumer buy-in.

Not for the first time the vendor pitch suggests a world in which everyone wins. Consumers will leap at the chance of improved data coverage, reduced calling charges and some potentially compelling new data services. For operators, say the suppliers, femtos offer a twin cost saving; on the one hand they place part of the network rollout burden on its subscribers to subsidise a network roll out and on the other they shift the expected tidal wave of backhaul onto residential DSL. Not only that, subscribers are much more likely to opt for family packages which will help improve stickiness, say the vendors.

Thanks to an increased number of flat rate data tariffs and the boom in mobile broadband, mobile data usage is on the sort of upward trajectory that might well give operators some serious coverage and backhaul headaches. “We’re now seeing the flat rate tariffs stimulating traffic growth to such an extent that the network will eventually and finally be operating at something like a sensible capacity,” suggests Tim Lunn, managing consultant, PA Consulting. “So you’re going to have smaller and smaller cells and more and more base stations to support the traffic, which will be a cell planning and deployment issue in its own right.”

According to Lunn, femtocells provide the operator with a cost effective way of managing the increased demand being placed on their networks. “There are a number of reasons, from the operators’ point of view, why femtos are attractive and that will incentivise them to provide good deals for customers and so the customers will be incentivised to accept the femtocell and better data rates, coverage and services as part of what they expect,” he says.

From the consumer perspective, however, there is nothing particularly new about the headline benefits of femtocells. Offering improved prices in the home sounds like a winner. Try conducting a straw poll of your own, asking: ‘would you like cheaper calls when you’re at home?’ You’ll get a resounding yes from almost everyone you ask. Which is almost certainly something BT found prior to the launch of its UMA (Unlicenced Mobile Access) BT Fusion flop, which was taken off the market after attracting about 45,000 customers.

Orange’s Unik, launched November 2006, is widely regarded as one of the best examples of a successful UMA FMC offering. Recent subs numbers are hard to come by, but according to In-Stat’s Nogee, the firm has about one million Unik subscribers, which sounds fairly impressive until you realise the carrier has 176 million subscribers in total. Consumers might like the idea of cheaper calls at home, but they’re not always motivated enough to do anything about it.

Returning to Sprint, Nogee raises further questions about the business case: “Minutes in the US are so inexpensive that one of the big benefits of the femtocell, the unlimited minutes that you could offer or the low cost, really doesn’t come into play all that much. People don’t need these extra minutes.” And, incidentally, there is a good chance that offering your subscribers cheaper mobile calls at home will translate into mobile voice revenue cannibalisation rather than mobile for fixed substitution.

There are also some interesting issues with regard to the demand for improvements to coverage. Mobile users are well used to experiencing lacklustre indoor 3G coverage and those same mobile users will have experienced decent fixed internet connectivity, so why would they will be keen on paying an additional charge to install a femtocell in order to connect to the internet via a mobile device that suffers from some non-trivial user interface hurdles, when they’re more than likely to have a perfectly serviceable PC or wifi-enabled laptop?

“It is curious, I had the same thought I must admit,” says Dr Nick Johnson, CTO at femto provider ip.access. “But the truth is people like having [internet connectivity] on their phone. It’s in their pocket, they don’t have to go to a desk and turn the laptop on. The studies and the headlines about people who are interested in data services are interested in doing that from the device in their pocket,” he continues.

But, say the cynics, if they’re interested in connecting wirelessly via a phone then they already can, it’s called wifi. The number of wifi-enabled handsets is on the rise. As more and more wifi-enabled phones come to market, it will surely dampen demand for improvements in 3G mobile coverage in the home. Why buy a femtocell when you can get a wireless router?

“The principal reason why a licensed technology will win out over wifi is that there is a very strong business model behind cellular services in terms of your tariff and your bundle that you just don’t have with wifi,” says Johnson, adding. “There is a strong counter argument that says if you are subsidising the handset then they don’t really want to make it easy to use wifi because you lose the revenue that goes with making calls in the home.”

Peter Jarich, wireless research director, Current Analysis agrees: “Wifi will address things like data, but until you get wifi in every phone it is not going to address the access to your phone, it is not going to address the voice access and ultimately it won’t address those advanced applications that operators are looking to develop to make femtocells more compelling.”

Going back to mobile broadband’s chief competitor, the wifi-enabled laptop, Jarich suggests the carriers could use femtocells to muscle in on the space: “Part of the reason we have wifi in laptops is because a) Intel put it there and b) we haven’t had good indoor 3G coverage. I think ultimately operators wouldn’t mind seeing laptops running over 3G, especially if you had a femtocell in the house. Ultimately, operators would like to see-although no one will say it out loud-3G take on wifi.”

According to Informa Telecoms & Media, in 2008 the home environment will be responsible for 35 per cent of total mobile data traffic but this traffic is expected to predominate within five years, with an overwhelming 60 per cent in 2013.

“This growth will be driven by the fact that users will increasingly initiate longer and richer data sessions, in the relaxed environment of their home, through browsing internet, watching longer and richer video clips, downloading music video applications, exchanging pictures, or using VoIP chatting online,” says Malik Saadi, principal analyst, Informa Telecoms & Media. “This does not mean that mobile broadband services will fully substitute fixed broadband, but users would prefer to keep some applications in the vicinity of the private and intimate zone of their mobile devices.”

Sprint’s 2008 debut femto roll out came later than many analysts had predicted. AT&T, Vodafone, Taiwan’s Chunghwa Telecom and Telefonica O2 are all trialling femtocells at the moment, and there is consensus that we’ll see some more launches in 2009, but we might have to wait until 2010 for anything like common adoption.

“Some operators will have a wait and see attitude, the second mouse eats the cheese,” says Rupert Baines VP of marketing for the Femto Forum. “The others are fast moving and their attitude is that the market share among carriers changes very rarely, and a few times you get a significant shift is when you launch a brand new service.”

Baines suggests that the last example of a comparably disruptive service in Europe was pay as you go, where the operators that launched first did very well. “One operator I’ve been talking to thinks femtocells will be similarly disruptive,” says Baines. “Their logic is that subscriptions at the moment are sold to individuals whereas a femtocell will be sold to a household. So selling one service will snag additional subscribers.”

Informa Telecoms & Media expects the number of femtocells deployed by the end of 2013 exceed the 40 million mark. “This installation base could help operators to offload up to eight per cent of total mobile traffic to fixed networks via the end-user subscriber line. Obviously this could imply mobile operators will make significant savings by reducing the need to create additional macrocell capacity to cope with this traffic,” says Saadi.

A number of technical challenges exist, however, before we’ll see a mass-market rollout. Interference and handoff are the two headline issues. Both of which the femto vendors claim to have solved, but no one has rolled out a full femto network, so in practice they may still cause significant problems.

Sprint’s femto solution has sidestepped some of the handoff challenges: “If a user does take a supplementary Airave calling plan, calls initiated on the femtocell are charged at the discounted rate even if the user roams on to the macro network,” says Ovum’s Hartley. “Sprint told us that it saw little revenue leakage during trials and wanted to prevent bill shock. This is reasonable; however, potential for abuse remains.”

Slightly less reasonably though is the fact that there is also no handover to the femtocell for a call initiated on the macro network. “This is unhelpful to a user if they paid extra for an Airave calling plan, or bought the femtocell for improved indoor coverage. The call must be completed outside and restarted on the femtocell to benefit,” says Hartley.

A traditional network might include thousands of base stations, each of which must be monitored and maintained centrally by the carrier, a femto network could extend to the millions. Again, operators must be able to drill down to each and every point on their network. Again the vendors say this will all be possible, but it could take a huge leap of faith before we find out how these networks are managed in practice.

“It will be different from wifi to the extent that you need to deploy it as the end user and it is part of the operator’s network, so [the operator] has to have a view into that thing, they have to be able to shut it down and do the diagnostic,” says Current Analysis’s Jarich.

There are management solutions, but the fact that no one has seen them up and running is one of key the reasons they haven’t taken off yet. “Operators want to see stuff work. I can tell you that I have a really cool solution, but operators will want to make sure it works first,” says Jarich.

“As part of the Femto Forum I co-chair one of the working groups which is specifically interested in the radio physical layer,” says ip.access’ Johnson. “One of the activities that we have been conducting over the last quarter or so is an interference management study looking at various scenarios where you have a femto sitting on a table in direct line of site of a macrocell with a nearby femto user and a nearby macro user.” The results, according to Johnson, are overwhelmingly positive.

Rupert Baines, also of the Femto Forum, plays down the interference issue: “Femtocells are smart devices. Putting a repeater in is just like putting a loud speaker into your back garden, it is great if you want to listen to your music, but your neighbours gets to listen to your music whether, they would like to or not. A femtocell is more like a pair of headphones, it will only send information to the people who want it.”

There is one final issue that operators will need to overcome, and it is an issue that has dogged the positioning of traditional cell towers for years: Health. So much research has been done to disprove the supposed link between cell phones and sites and cancer that the operators are well versed in tackling what could be a challenging piece of PR. But persuading consumers to put a base station in their living room might stretch even the most silver-tongued of public policy professionals.

“This is a matter of perception, and very difficult to predict,” says Peter Thornycroft, product manager, Aruba Networks. “Given the widespread market for wifi access points, one might expect it will not be an issue. But femtocell devices might be seen as different from wifi and it might only take one or two newspaper articles or TV programmes to set off a backlash.”

Tim Lunn of PA Consulting says: “The argument that wifi routers don’t cause cancer is not a good argument, even raising it as an issue would be unwise. I don’t know how that argument is going to be addressed, it may be raised by some groups.”

Indeed, Lunn (and others) suggest there is a potential health upside to femtocell deployment: “Having a femto base station means that your handset is transmitting at a lower power. This is being used for the argument that it will save your battery, which is a nice feature but I don’t think it will make the consumer decide one way or another. And likewise, I don’t think people will be pushing the health reasons as a reason to buy a femto, though in reality it is a benefit.”

It will be interesting to see how operators build business models around femtocells. And how the tier one network infrastructure vendors integrate the wide range of smaller, as yet un-standardised, solutions into the core network. It’s likely that we’ll see some bold first moves being made by disruptive carriers-like Sprint.

Buy-in from the market leaders might be slower to come. Like a good number of bright new technologies, femtocells could end up being something that gets rolled out because your rivals have rolled it out. Femtocells might eventually enable FMC, and they will surely improve the uptake of mobile broadband. But they are not proven to be a consumer’s mobile dream come true even though they might be a solution for carriers’ coverage worries.