Femtocells Gain Data Abilities

Realising the benefits of femtocells
By Rupert Baines, picoChip
Thursday 14 January 2010

'More business models exist that could bring a variety of new players into the field...'
The recent launch of femtocells by AT&T follows Vodafone's entry into the femtocell market.
The first launches from Sprint and Verizon were relatively simple, with 2.5G service and a
rationale of voice coverage and customer retention. Vodafone and AT&T's moves are more
sophisticated; with High Speed Packet Access (HSPA) femtocells reducing churn while boosting
both voice coverage and data capacity.

We are starting to see more interesting and mould-breaking business models and it is likely
that more will emerge as players appreciate the variety of ways femtocells create value and
deliver unique capabilities.

To explore those we should consider the four areas of benefit from a femtocell:

First is voice coverage - surprisingly, still a significant problem in Europe. Recently U.K.
regulator Ofcom reported that more than half of consumers said they had experienced
difficulties with mobile reception; a third said this was a regular experience.

Coverage is often overlooked. If one carrier has good coverage in a village then residents switch
to that carrier and forget the others. But for the "failed" carriers, that represents lost subscribers.
We already see femtocells simply marketed as solving voice coverage, charged-for or free, but
within a few years simple competition will ensure this is a baseline part of any operator's
offering: if you cannot serve a customer, you cannot have their business.

The second area is data coverage. In the last decade carriers have spent billions of dollars on
3G licenses and networks, predicated on selling data services to consumers. Yet, carrier
coverage maps reveal that all too often indoor coverage is voice only. That's a consequence of
the higher frequency (2.1 GHz) that doesn't go through walls (interestingly, that helps femtocells
to work better, by isolating the indoors and outdoors). With 60% of all cellular data traffic sent
from inside a building there is a clear business case to improve this.

A third aspect is data capacity. Quite simply, cells are getting full up. The success of dongles,
the iPhone and other smartphones has driven explosive uptake of data. By 2014 global mobile
Internet users will send and receive 1.6 Exabytes of mobile data each month: that's more than
the 1.3 Exabytes transferred during the whole of 2008, according to ABI Research.

While the amount of data is doubling every year, basestation capacity is almost stagnant.
Increases like HSPA+ are important but not sufficient.

The only real way to increase capacity is to add more basestations, but it is impractical /
impossible to add more macrocells. So it is no surprise that AT&T, confronting the success of
the iPhone and explosive traffic growth has launched HSPA femtocells.

The most exciting revenue area is applications, based on the uniqueness and personalization of
a phone and the trusted billing relationship that exists with the individual subscriber. While
resisting the temptation to jump on the 'connected Appliance' bandwagon, there will be new
businesses using a technology that "knows" when people are home – home delivery services,
healthcare, "fridge magnet" family reminders, and so on. These could be from incumbent
operators, or new entrants using femtocells to cost-effectively offer consumers a compelling
package of voice and data as well as a series of "femto apps", without the high entry costs of
traditional infrastructure equipment, base station real estate or rental, power and other utilities.

As with many areas of technology it is impossible to predict what will capture customers'
imaginations or exactly what the business models will be. But the combination of
personalisation, context-sensitivity and always-on, always-available high-speed connection to
your individual phone will doubtless lead to attractive new services.

In the non-residential world, corporate-focused carriers are starting to offer enterprise
femtocells as a cost-effective alternative to DAS or microcells, with additional services such as
PBX integration. Also, "metro femtos" can be quickly and cheaply deployed to address specific
blackspots or capacity crunches in shopping malls, airports, hotels, subways or the like –
anywhere there are lots of users and poor service.

The reverse can also make sense: instead of "dense urban", some carriers are planning on rural
femtos. These are deployed on a case-by-case basis to put spot coverage just where it is needed,
to cover an isolated village or railway station more cheaply than installing an expensive base
station to cover a whole sparsely populated region.

Significantly, all of these systems are cheap (low capex and opex): benefitting from the
economies of scale that the mass-market residential devices enable and leveraging the same
core network investment.

More business models exist that could bring a variety of new players into the field. A wireline
broadband company, for instance, could deliver femtocells to its subscribers and sell
connectivity and backhaul to mobile network operators (MNOs) as a white label or wholesale
supplier. Or someone may enter with an application-focused product for the home, perhaps
analogous to Amazon's Kindle which emphasizes the service to such an extent that most users
don't care there is a cellphone inside.

In summary, there is no single strategy for femtocells. They offer providers – whether operators
or new entrants – so much flexibility that depending on the targeted customer segment, and the
features that the operator wishes to make available, the field remains wide open. It will be in the
interests of the customer and the market as a whole if a variety of completely new business
models appear and the best succeeds. What is clear that the variety of different facets (voice,
data, capacity, service) combine to deliver an extremely powerful business case that carriers
ignore at their peril.

Rupert Baines is VP marketing at femtocell technology company picoChip.