Femtocells held back by technical difficulties: Ericsson

While offered as a solution for mobile broadband blackspots, femtocells would prove too difficult technologically and in a business sense, according to Ericsson

The technological and business complexities of femtocell deployment has kept the market from flourishing locally, according to Ericsson head of strategic marketing, Kursten Leins.

Though the miniature mobile base stations have long been touted in the global telecommunications industry as a solution to indoor mobile blackspots, Leins told Computerworld Australia that the devices would ultimately prove a headache for telco operators.

"The big challenge around femto is partly the spectrum planning and use," Leins said. "As soon as you introduce a signal [to] noise ratio you start shrinking the cell size, which has the effect of reducing your throughput to customers, and it's a spiral.

"You could have very clustered use of these services which then effectively means your benefit for in-building coverage is diminished; the end outcome may not be as nearly as good as you expect for customers."

Femtocells are designed to boost mobile coverage in a small area - within 100 metres - by using a home or business' fixed line broadband as backhaul infrastructure for voice and data. The devices have been available as basic GSM voice boosters or 3G data access points since 2006, and are expected to continue to grow, with 29 deployments expected to reach fruition in coming years and femtocell access points forecast to grow to 49 million by 2014, according to research from Informa Telecoms & Media.

Some markets have begun to see second iterations of the device, but the capability is yet to take off in Australia. That is despite the country coming third to Austria and Sweden in number of mobile broadband subscribers per capita, with almost 3.2 million customers to date.

There has been some interest among local carriers; Vividwireless' predecessor, Unwired Australia, became a member of representative organisation Femtoforum in 2007, but never announced any products or trials. A spokesperson for vividwireless confirmed the company had explored the technology, but had no current plans to release it commercially.

NEC Australia recently entered the femtocell market, commencing trials with one telco six to nine months ago in the 3G and HSPA sector.

Vodafone Hutchison Australia (VHA) is also believed to be in talks with Alcatel-Lucent about bringing the technology to the Australian market, following a deployment of Vodafone UK's Sure Signal 3G femtocells earlier this year. 

While Ericsson initially released a GSM femtocell in 2007, it failed to follow with a 3G version, citing poor demand for such a product. Leins said the company was always exploring femtocell products and could reconsider re-entering the market when LTE is widely deployed.

"Standardisation of femto in LTE is potentially interesting, but I think at this stage it's too early to tell," he said. "If we see operator demand for products in LTE in femto then of course we'll develop something to meet that need. Being active in standards means we can be developing protypes in the labs; it's not to say we can't commercialise, we just don't put product out there if there's no demand for it."

However, the technological issues remain a key deterrent for Ericsson.

One reason for the delay of femtocells in the Australian market has focussed on legislation around spectrum use. Unlike the 2.4Ghz and 5Ghz bands, which are unlicensed with power restrictions for use by Wi-Fi access points, femtocells would have to operate in common spectrum bands already in use by major carriers.

In Australia, the 850, 900, 1900 and 2100MHz are most often used by major telco for GSM and HSPA transmission, while LTE is expected to operate on the 700, 1800 and 2600MHz bands. Of those planned for use in LTE deployments, 700MHz is yet to be auctioned off, while the Federal Government also intends to re-plan the 1800MHz spectrum in preparation for the digital dividend sell-off.

The rollout of the National Broadband Network (NBN) has heightened speculation of a positive case for femtocells, as the network's relatively high bandwidth would decrease the strain of the femtocell's backhaul on a home's Internet connection. But even if the industry were to overcome the spectrum hurdles, the technological difficulty of planning for femtocell use - and possible cannibalisation of existing coverage - would, according to Leins, prove overwhelming.

"You have to literally start mapping where the femtocells are," he said, adding that as "you start going down to a more granular level, it adds cost to the equation."

The telco services provider, which has championed a vision of 50 billion mobile connections by 2020, has continued to argue that the lack of a business case for femtocell use would push back.

"It's a challenge for our market where we have an expectation of 'bucket plans' already for most consumer mobile broadband users," Leins said. "Cost is a major factor, and that's not just us, that's simply doing the maths on a generic product offering."

Femtocells have dropped in price over the past few years, with US telco AT&T even beginning to selectively provide the devices for free. However, they are often subsidised by the carrier, and continue to attract a one-off fee or ongoing monthly cost in addition to existing mobile or Internet connection plans. Some operators, like Huawei, have attempted to tackle this by integrating femtocell technology in common consumer electronics like a media streamer, though a spokesperson told Computerworld Australia that it wasn't currently in talks to bring such a product to Australia.

Instead, Ericsson has pushed mobile operators to address blackspot problems by increasing overlying cell capacity, particularly in dense, urban areas. Coverage issues within office buildings would be better served by in-building deployments, according to the company, as the femtocell's average capacity of four to five users would prove a problem for even small offices.

Vodafone UK's Sure Signal device is capable of registering 32 mobile phones for use, but can only connect to four phones simultaneously. While the telco does offer the femtocell to all customers, it does notably recommend the device for small businesses rather than medium or large enterprises. 

Leins conceded femtocells could address a niche market in more suburban areas where houses are further apart, but said that increasing issues with Wi-Fi density in those areas was a sign of things to come should femtocell be introduced.

In regional areas, femtocells would possible serve a single household but lack the capacity and coverage range to sufficiently address mobile blackspot problems. That issue has been tackled by researchers at Flinders University, who recently demonstrated the use of mobile phone Wi-Fi receivers to act as an internal mobile frequency transmitter. The makeshift capability is capable of transmitting a few hundred metres, but could conceivably harness other phones and inexpensive Wi-Fi transmitters in the area to provide more coverage, even if hundreds of kilometres away from a mobile phone tower.

"We are actually carrying voice over that but in a way that doesn't need to go back to a central repository anywhere," Flinders University researcher, Paul Gardner-Stephen, told ABC Local Radio program, AM.

"What research has actually shown is that the vast majority of the response to a disaster is actually form [sic] the local people there so if we can provide them with ease of communications as soon as possible after the earthquake, not 48 hours, not 72 hours but potentially minutes after a disaster, then we can help them to start rescuing people from rubble and generally rebuilding, maintaining law and order. All of these things that are really vital for the response to these kinds of things."

Join the Computerworld newsletter!

Error: Please check your email address.

Tags Ericssonfemtocells

More about ABC NetworksABC NetworksAlcatel-LucentAT&TAT&TEricsson AustraliaetworkFederal GovernmentFlinders UniversityFlinders UniversityHutchisonInforma AustraliaLucentNECNECNECUnwired AustraliaVHAVodafone

3 Comments

Keith Day

1

Ercisson comments are right: you can't deploy passive femtocells in shared spectrum without interference problems that might affect all customers.

But how come SFR in France and SoftBank in Japan have succesfully launched femtocells in shared spectrum?

The answer is that they have launched with a new generation of intelligent femtocells, which constantly monitor their environment and adapt to any changes, like your neighbour installing a femto, a macro user entering your apartment, or simply opening a window.

I work for the company that invented and patented this technology (Ubiquisys). I'm hoping that as people become aware that some femtocells are more than miniature base stations, that a more informed debate will emerge.

Andy Gothard

2

Keith is right - today there is no technological barrier to deploying femtocells. There is a whole ecosystem of companies offering successful femtocell solutions for home and business use, most of them using semiconductor devices made by the company I work for (picoChip - that's the plug over with:)
Softbank in Japan, SFR in France, Vodafone in the UK and Spain, AT&T in the US and Movistar in Spain are all offering fully commercial femto services. So if femtos are a headache, they're a very unusual type of headache - one that operators seem to like having.
Much of the technology needed to build femtocells into the mobile network is already enshrined in international standards via 3GPP.
And by the way, it's not just me and Keith! Our industry association, the Femto Forum, has over 120 members including 55 operators accounting for 1.4bn subscribers; and around 70 technology providers, including some of the industry's biggest names such as Alcatel Lucent, Cisco, Huawei and NSN.
Perhaps the most interesting effect of this for old lags of the telecoms industry (like me), is the need to learn new tricks fast. For us, making a compact basestation that works "out of the box" is nothing short of a revolution. Sometimes progress is uncomfortable, but we know it'll be worth it in the end (when we can use our mobiles in the basement!).

Mark

3

James, the thing that is missed here is how is this massive wireless broadband load increase, doubling every year till 2014 (according to Cisco) is going to be provided? Sure WiFi can do internet but it is much more than that. People want simple to install wireless, they want SMS, MMS, no hassle installation, they want a single number to call when there is a problem too. I've personally had experiences with a VPN setup when I don't know if it is the router, WiFi, my ADSL provider, or my work was the problem- who wants all that hassle. Actually I think if you look around you will see there is much more going on in Australia than you think - even world-first research results.

Comments are now closed

Get ready for IoT, mobile cybercrime in 2015: industry

READ THIS ARTICLE
DO NOT SHOW THIS BOX AGAIN [ x ]