Significance: Small cells are no longer the preserve of the residential or enterprise femtocell, but are moving into public access carrier networks, indoors and outdoors. That brings hope of a new way of thinking in the RAN ‘ new cost structures; competitive supply chains; above all better ways to increase mobile data capacity even in limited spectrum. But before the benefits can be realized, there are many technical, economic and political issues keeping carriers awake. At this week’s Small Cells World Summit, we identified the big challenges, and in the next issue will look at them in greater depth.
The Small Cell World Summit this week was the annual event’s first outing under its new name ‘ like its associated industry body, the Femto Forum, it changed its branding to reflect a broader reach for small base stations in the future. While most of the case studies still centered on the residential femtocell, the debate has moved to the issues of incorporating miniaturized cell sites into every part of the network ‘ indoors and outdoors, closed or with public access, in various spectrum bands, and eventually as an essential ingredient of the evolving HetNet (heterogeneous network).
Deploying large numbers of small cells, in the same frequency as the macro network or as a separate layer, and potentially with Wi-Fi integrated too, raises unprecedented challenges in terms of network planning and management, not to mention site acquisition and location, power management and regulatory approvals. There were some bullish roadmaps shared by certain carriers and vendors, outlining radical and ambitious HetNet plans, as well as some big numbers in terms of equipment shipments.
But generally the tone was a more sober one, with a strong focus on the practical issues of making small cell deployments work in reality. The notion that small cells are the only mainstream way to achieve the capacity required for growing mobile broadband loads has moved, as Small Cell Forum chairman Simon Saunders said, from heresy a few years ago, to ‘almost orthodoxy’. But with the acceptance of the approach has come another recognition, that just throwing small, low cost, low power base stations at the problem will not deliver the desired results. These will need to be planned and managed in the optimal way to ensure they don’t deliver more problems than capacity, and that includes the thorny issue of backhaul, not just the cell site itself. That will involve some crucial decisions to be made by the industry.
In our view, five key discussions emerged as the most pressing ones for operators at the event, all of them crucial to maximizing the capacity and business return. We will examine these in detail in a special small cells issue next week, but in outline they are:
Edge-based or centralized intelligence?
The highly centralized approach of Cloud-RAN, where cell site tasks are virtualized in the cloud, is popular with a few fiber-rich carriers like China Mobile, but generally seen as futuristic. There is growing interest in pushing network intelligence, cell site and even applications processing to the edge, as seen in the Intel’s Edge Cloud collaboration with Ubiquisys.
Indoor or outdoor?
Small cells will be used in both scenarios and many of the debates and product announcements were geared to the outdoor metrozone. However, with the bulk of mobile data being consumed indoors, there are likely to be clearer short term returns in deploying indoor zones, whether private for the enterprise or in public access areas. The former was highlighted by two partnerships between a start-up and a giant ‘ Quortus, the embedded core specialist, with Cisco; and Spidercloud, pioneer of enterprise-RAN, with NEC. In the latter space, Ubiquisys was one small cell provider which has seen the greatest growth coming from public access indoor deployments, especially in Asia.
Managed services and shared RANs:
These are two of the most controversial topics and often interrelated. The managed service providers aim to expand this lucrative business by playing on the huge complexities of deploying a major small cell network, from site identification to installation to management and interoperability. But companies like Alcatel-Lucent made some compelling points, appealing to carrier nervousness about the new topologies, while fiber operator Colt Telecom had an interesting take, proposing a ‘femtocell as a service’ offering which could support many cellcos’ services. The idea of a single wholesale big pipe is gaining traction in many countries for the LTE market, and sparks the debate over whether small cells are the carrier’s way to preserve differentiation if it is sharing the big pipe; or conversely, whether simplicity and local regulations will make a shared small cell network the only practical option.
End-to-end or multivendor?
This was the highest profile debate of all during the week, and the most political, since it touches on the fundamental split which has affected the wireless sector since it started to carry data and IP ‘ between the mix-and-match traditions of the PC and IP worlds, where control lies in the software platforms and the back end services; and the closed norms of the cellular industry, which used to see each vendor supplying its own end-to-end platform, hardly considering interoperability except with devices. The Femto Forum contributed a significant psychological turning point in this debate when it succeeded in gaining powerful industry support for Iuh, a common interface for residential small cells which allows operators to connect products from multiple vendors to their cores and gateways. The Forum, along with companies from outside the RAN like Cisco, want to extend that vision to the wider small cell market, but in the public carrier network, it comes up against the entrenched power of the big OEMs. As ip.Access and others argued passionately, many of the benefits and economics of the small cell idea will be lost if the new networks are single-vendor, even though some carriers are daunted by the prospect of integrating kit from multiple sources. But all agreed that the 3GPP standard underpinning interoperability, X2, will be unsufficient on its own, and the Forum has initiated a new workgroup to examine additional required techniques. These will be important if the end-to-end view is not to win by default, because multivendor interoperability cannot be delivered to the required level of performance and trust.
Will SON deliver?
Another area where basic standards will only deliver some of the required capabilities to reap the benefits of small calls is SON (self-organizing or self-optimizing networks). Suppliers are increasingly incorporating SON features ‘ the latest example being a partnership between specialist Airhop and base station chip giant Texas Instruments. But additional capabilities, whether standards-based or not, will be needed, and those will raise new interoperability issues.
While the debates over deployment practicalities proceeded, few disagreed that small cells will deliver strong growth for the equipment industry, even if there are still doubts about how quickly public access cells will gain mass scale, and where the best profit model will lie. In a new report, Informa analysts raised their forecast for global small cell shipments to 91.9m by 2016 (from 62.4m in an earlier prediction). In its latest quarterly update, the firm said small cells would outnumber macrocells by the end of 2012, with 6.4m deployments versus 6m (86% of that 64m figure will be femtocells).
“We are starting to see vendor led public access small cell R&D turning into concrete operator deployment commitments,” said Dimitris Mavrakis, principal analyst at Informa, in a statement. He also pointed to the valuable experience some carriers had gained from rolling out residential femtocells ‘ while simpler in terms of network planning complexities, they did highlight some of the issues associated with provisioning and managing large numbers of small base stations. He said most operators in the first wave of public access deployments had already supported femtocells, ‘so are accustomed to small cell challenges and how they can be overcome’.
Indeed, the residential market has provided many lessons which will be valuable in the broader small cell space, and which have changed many of the assumptions of the traditional RAN. One is the very idea that a base station can be built around a highly integrated system-on-chip, delivering the kind of cost and power levels more associated with client devices. While the platforms underpinning the public access cells will be more complex, the new thinking on price, power and supply chain will not be unthought. The consumer electronics norms which pioneers like Picochip (now Mindspeed) applied to home base station architectures are filtering up the chain into the more heavyweight models, and have shifted the behavior of established giants like TI and Freescale, both now selling base stations on a chip. This has forced creative debate on the economics of the new base stations, with TI’s head of product marketing, Josef Alt, pointing out that, while the overall cost of a small cell unit might be low, a far higher percentage of its value lay in the silicon than in previous architectures. But those kinds of shifts in balance affect the whole ecosystem, and have seen the emergence into the once rarefied BTS space of Taiwanese and Korean suppliers more used to producing consumer access points.
The excitement, and the concerns, may have moved up the food chain, but as the Informa figures show, the humble residential cell will still deliver the bulk of the volumes ‘ if not the profits ‘ for 2012 and 2013 at least. These devices are reaching critical mass as operators become more sophisticated about the value proposition for users, and as prices come down. There are now two markets ‘ the UK and France ‘ where all the mobile operators have deployed femtocells commercially, making the technology a mainstream element of the carrier’s toolbox. Two UK cellcos, O2 and Orange UK (part of the Everything Everywhere joint venture) have announced femto services, joining existing offerings from Vodafone, T-Mobile and 3UK. And Bouygues Telecom has completed a French quartet, adding its new service to those of Orange, SFR and Free Mobile. O2 claims to have achieved the densest femtocell deployment in the world (1,500 cells in one apartment complex in London).
‘Femtocells have now been deployed by 43 operators globally and France and the UK have become the first markets where all operators have announced services. With nine of the top 10 operators by revenue offering femtocells, the technology is proving to be a vital tool for improving mobile coverage and increasingly importantly capacity too. This comes at a time when the key femtocell technology is being employed in outdoor small cells which will be crucial for providing much needed extra capacity in urban hotspots as well as rural areas,’ said Saunders.