Small cells may be relatively simple beasts when taken individually, but deploying and managing a network of thousands of them ‘ indoors and outdoors; in different bands; perhaps in uneasy proximity to other networks ‘ is daunting. The complexities of rolling out and managing large-scale small cell systems were the overriding theme of last week’s Small Cells World Summit in London, and gave rise to two related debates ‘ whether these dense networks of low power, limited-range base stations will be most efficiently deployed as a managed service; and whether carriers will end up sharing a common infrastructure in most cases. Those lead to another stage in the thinking ‘ the concept of a wholesaler which can offer femtocells as a service, even on an on-demand basis, to support multiple and often disruptive providers.
The issue of network sharing is an important one across the 4G market. Many carriers have deals to share elements of 3G infrastructure and these will become more common in LTE as operators look to reduce upfront cost and sometimes to pool spectrum capacity. There are various models ‘ sharing of passive infrastructure, but not base stations; a joint venture covering the entire network roll-out, like that of Everything Everywhere in the UK; or a single wholesale RAN providing services to multiple retail providers. In the last case, the wholesaler may be an independent business, offering RAN as a service, as LightSquared planned to do in the US; it may have investment from its carrier customers, like Clearwire or Russia’s Yota; or it may be initiated by a government agency, as in Kenya’s LTE plan.
When it comes to small cells, some argue that they are the best way for carriers which are sharing the macro network to retain control and differentiation for their own subscribers ‘ by securing the best locations, and by taking advantage of the small base station’s inherent location and presence awareness to support new applications. This was much discussed by the US cable operators which invested in Clearwire, although they have since backed away from the venture.
Conversely, some believe the efficiencies of the shared RAN will be diluted if separate small cell networks need to be maintained, and that differentiation should be driven by the services and tariffs offered, rather than the topology of the network. And a third group think that, even in situations where there are separate macro networks, carriers may be forced to share dense small cell deployments to avoid seeing excessive numbers of base stations, potentially interfering with one another, in certain neighbourhoods. This may also be necessary to satisfy local authority guidelines.
The UK provides a useful case study of the complexities of 4G network planning and sharing, with its various providers supporting many of the variations outlined above. Everything Everywhere is a full joint venture which also has a RAN sharing deal with 3UK, but sources indicate this is unlikely to extend to small cells. By contrast, the other two carriers ‘ Vodafone and O2 ‘ recently expanded their own network sharing agreement and are said to be more interested in pooled metrocell roll-outs. Meanwhile, UK Broadband is rolling out TD-LTE in the 3.5GHz band, which many want to earmark for small cell roaming; while the main pay-TV providers, Sky and Virgin Media, both have activities which could point to a wholesale small cell offering in future. Both firms have extensive networks of Wi-Fi hotspots, which not only support branded consumer access but also offload services for cellcos. Virgin is building a Wi-Fi network for the London Underground subway, which will be open to the mobile operators to use for their own services. And both media players have trialled small cells, leading to speculation that they could seek to acquire licensed spectrum in the forthcoming auctions. Then there is British Telecom, which has a significant wholesale Wi-Fi business, and Arqiva, which has expanded its traditional business in cellco infrastructure such as towers, to provide indoor and Wi-Fi networks on a white label basis (it recently bought white label hotspot provider Spectrum Interactive).
This kind of patchwork of operators and networks will be increasingly common, and small cells ‘ often integrated with Wi-Fi ‘ will both complicate the picture, and create new opportunities for wholesale models.
In many instances, especially in Europe and parts of Asia, the trend towards RAN sharing has gone hand-in-hand with a rising reliance on outsourcing or managed services. The major infrastructure vendors are looking to these services for much of their growth as hardware margins come under pressure. So even firms, like Ericsson, which are wary of the small cells trend because it breaks down the economics of their traditional macrocell business, will welcome the opportunity to play on operators’ fears about the complexities of deploying and managing metrozones. Operators, too, can see the opportunity for managed services, especially with regards to the enterprise ‘ Orange France, for instance, was the first cellco to announce such an offering for business customers last year.
Colt Telecom has come up with an interesting approach, which it dubs ‘femtocell as a service’ (FaaS). Len Schuch, co-founder of femtocell specialist Ubiquisys and now Colt’s small cell adviser, believes operators understand the benefits of small cells, such as cost savings and market share protection well ‘ but are still hesitant to take the plunge. Fewer than one in five European operators has a commercial femtocell deployment, he says, but FaaS could accelerate uptake. Colt provides a centrally hosted small cell core, which can be in the cloud, and can service multiple operators across Europe, harnessing Colt’s fiber network. The firm operates a monthly per-femto subscription model for those carriers, so there are no upfront launch costs. Most importantly perhaps, the thorny issue of backhaul is avoided.
Schuch believes it will appeal particularly to small to medium operators, and to tier one carriers targeting specific sectors such as an enterprise roll-out. He also believes MVNOs have been largely ignored in the small cell debate so far, but this enables them to offer a femtocell service even if their host carrier does not ‘ or does not want to share its femtos. This type of thinking could be a significant step forward in wireless services competition. The dream of hundreds of different providers offering differentiated and targeted applications is always held back by the high barrier to entry in the market, even as an MVNO ‘ and the control that the big carriers usually continue to exert over their virtual partners. Fat pipe wholesalers, as envisaged by Google ‘ which goes so far as to suggest an on-demand model where capacity is used only when and if required ‘ may emerge one day. In the meantime, a fat fiber/core offering like Colt’s, enabling almost any provider to support small cells on a broad or narrow basis, is not a bad start in that direction.
NEC targets enterprise small cells with SpiderCloud
Start-up SpiderCloud Wireless has signed NEC as the biggest partner yet, for its Enterprise-RAN small cell technology. The Japanese firm, perhaps the most aggressive of the wireless infrastructure players about shrinking base stations, will offer the SpiderCloud system to carriers round the world.
The smaller company offers a ‘small cell cloud’ targeted at medium to large enterprises. Its ‘controller-to-radio-nodes’ architecture can scale from 250 to 25,000 employees and can be installed in a few weeks or even days, depending on the requirements.
The deal will complement NEC’s existing alliance with Ubiquisys, for residential or small business femtocells. The Japanese supplier is also banking on the emerging class of public access small base stations to target 4G roll-outs and boost its presence in the RAN. NEC’s small cell gateways support about 1m connections, with customers including Softbank Mobile, France’s SFR, 3UK and Russia’s MTS and MegaFon.
Its new partnership will extend its reach to the corporate world. Martin Guthrie, head of NEC’s business development team in Europe, told GigaOM: ‘We were never really able to play in the very large enterprise space. For us that makes SpiderCloud a perfect fit.’ For its part, SpiderCloud’s main carrier deal is with Vodafone, which effectively incubated the E-RAN technology before it was launched in its full commercial version.
In April this year, SpiderCloud raised a $35m round of funding which valued the four-year-old firm at just under $200m and brought its total venture capital to date to $100m. Investors included Charles River Ventures, Matrix Partners, Shasta Ventures and Opus Capital.
SpiderCloud’s CEO Mike Gallagher – whose resume includes Flarion (sold to Qualcomm), FiberTower Networks and Nortel ‘ said in an interview at the time of the new round that he did not expect to raise more VC funds. ‘Revenue is coming in now. Small cell technology is the architecture of the future,’ he said.