Sorting Through The Femtocell Fray

The telecom market is no stranger to hype. Year after year, new technologies are introduced, all promising to meet a broad array of operator and end-user requirements. Marketing campaigns ensue. After a good deal of market ‘education’ the new technology is seen as an inevitable success. Technology development and deployment delays ensue…along with an inevitable backlash. Somewhere in this process, the new technology gets its own conference or tradeshow.

Femtocells are clearly following this formula. With UMA and other dual-mode FMC solutions preceding it, the market education phase has been a short one – but it has also resulted in some undeniably inflated expectations along with the inaugural ‘International Conference on Home Access Points and Femtocells’ held by Avren Events in London from last month.

The basic value proposition of femtocells is simple to understand: all the benefits of WiFi-based FMC without relying on a limited supply of relatively expensive dual-mode devices. To this end, Avren’s conference had no sessions dedicated to selling the femtocell vision. The focus was where the femtocell market stands today, where it’s going, what it will take to get there and which questions are still largely unanswered. And, with several topics and themes repeated multiple times over the course of several days, a clear snapshot evolved.

• Femtocell Rationale. With the first femtocell trials still underway, the specifics of actually rolling out a femtocell-based service are unknown. Surprisingly, however, there is limited insight into why an end-user would want a femtocell. Improved indoor coverage can be delivered by repeaters, dual-mode solutions or even additional macro-cell base stations. Improved capacity would follow. Femtocells remove the need for subscribers to purchase dual-mode devices, but this assumes they see the value in FMC services and are strong 3G users – 2G networks, after all, already provide strong coverage. The primary rationale, then, is operator-driven: transport offload, an enticement for subscribers to use 3G, improved macro-cell capacity, etc. In short, the customer draw (today) is cheaper voice services while the operator draw is cheaper data delivery. But, with other solutions delivering cheap voice, the operator rationale dominates.
  
  

• Ownership. Compact, low-capacity 3G base stations are not a new development. Femtocells, however, make them a residential reality by bringing costs down. But, this raises a critical question, “who actually pays for the femtocell?” If the end-user buys it, the service offer (and savings) will need to be very compelling. If the operator buys it, the costs of rolling out a femtocell solution quickly escalate. Most operators agree that they would need to, “own” the femtocell since it is an extension of their network – something that would leverage their spectrum assets and something they would want to manage. Sprint, with its vision of WiMAX devices (femtocells included) being completely unsubsidized, was the most vocal outlier.
  
  

• Marketing. Femtocell service launches would require their own, dedicated marketing efforts. Beyond the basics of FMC marketing, femtocells come with their own unique value proposition and obstacles. At a minimum, broader acceptance will require one common, accepted terminology for the device: femtocell, home access point, home base station, 3G home gateway, home node, ugly grey wireless box, etc. Without one, market confusion should be expected. Likewise, while fears of RF exposure are unfounded given low-power output, it will still be necessary to make sure end-users understand this point. Ultimately, if the femtocell value proposition is tied to mobile data usage, the success of the femtocell market depends most of all on basic 3G or WiMAX marketing, getting devices and services into the hands of users.
  
  

• Love of Existing, Open Standards. Operators could launch femtocells that operate as little more than WiFi access points, interfacing directly into an open Internet and improving the data performance of 3G (or even 2G) devices. That isn’t FMC, though. It doesn’t provide any integration with an operator’s mobile voice or data infrastructure. It doesn’t provide seamless connectivity with an operator’s OSS and BSS solutions. From an integration standpoint, then, there’s no shortage of solutions for connecting femtocells into an operator’s mobile core, the three front-runners being SIP/IMS, UMA and an ’RNC concentrator’ or gateway approach linking into the core via existing Iu interfaces. While too early to declare any one option a winner, operators are clear on two things. First, the use of Iu interfaces into the core is particularly appealing. Yes, most operators are moving on SIP and IMS. Iu, however, promises to integrate into existing kit in an easily understood way. Secondly, an open interface from the femtocell into the gateway is expected. 3G networks have been built with base stations and RNCs coming from one vendor. Femtocells, however, may take diverse forms with different capabilities serving different market segments. No operator wants to be locked into a single brand of home base stations simply based on their network gateway vendor.
  
  

• Defining “Open”. So, what is an ’open’ interface? Generally, it’s any interface whose specifications have been made public or, at least, released to interested parties. Unfortunately, this doesn’t mean that the interface is generally accepted or even easily adopted. Take the example of NSN’s new Femto Gateway. The vendor touted its reliance on standard and open interfaces. Yet, it also noted that it will, ‘certify’ the femtocells that interoperate with the gateway. Why is certification necessary? Because ‘open’ does not mean ‘standard’ and even where standards are employed, there are often various implementations of the standard. While it’s understood that Iu-based solutions are attractive, it’s also expected that a diverse set of open interface options from the femtocell to the gateway may fragment the market, opening up opportunities for UMA and SIP solutions.
  
  

• Costs, Costs, Costs. Noted earlier, the promise of low-cost home base stations – something end-users and/or operators could afford – has helped to make femtocells a viable FMC solution. The question of, ‘how cheap’ is a rhetorical one. Costs on par with WiFi APs would be ideal but difficult to obtain until scale ramps up significantly. Confounding scale, however, is the fact that there is no standard set of functions or capabilities a femtocell is supposed to deliver. Femtocell vendors are pursuing their own interface strategies. Most have diverse solutions for interference mitigation as well as security. Some will want to include WiFi in the node. DSL is the current expectation for backhaul. Cable and even fiber options are likely to follow…likely integrated into various forms of set-tops and residential gateways. A single chip femtocell solution, it was argued, is feasible and would enable a sub-$50 bill-of-materials (BOM). Until vendors and operators agree on common forms and functionalities, few vendors will risk developing such a solution and cost constraints will remain.
  
  

• Trials and RFPs. Of course, one of the biggest questions surrounding the femtocell space is, ‘when will it take off?’ There is little doubt that femtocells will get deployed. But, if this takes years, interest is bound to fade and competing solutions will gain ground. Luckily, vendors and operators, alike, leveraged the opportunity of the Femtocell conference to talk about their plans. Orange, already delivering UMA services, revealed its interest in a femtocell RFP (request for proposal). T-Mobile noted its commitment to femtos, even as its US arm launched a UMA offer. News of Vodafone’s RFP hit the wires, O2 confirmed its investigations and Sprint, once again, noted plans for WiMAX femtocells. Softbank even held proof-of-concept demonstrations in Japan.
  
  

• The Importance of RF Planning. The biggest technical issue keeping would-be femtocell operators up at night is RF planning. For the sake of capacity, operators will want to launch femtocells on the same channel as their macro cell network. This raises the specter of interference: indoor vs. outdoor not to mention indoor vs. indoor (e.g., between neighboring femtos). Separately, for the purpose of hand-in from the macro cell network, the existence of hundreds of new neighboring cells (IE, femtocells) could easily result in an overwhelming set of new cell global identifiers (CGIs) if standard cell planning procedures are followed. Ultimately, the result could be downgraded network performance, awkward handoffs and a burden on the core network in terms of cell management. Unlike obstacles which can be overcome by business practices or strategies, RF planning and interference worries will require specific, technical solutions from femtocell vendors.
  
  

• The Femto Forum. The day before the conference began, the Femto Forum launched. The Femto Forum is a group formed to promote the uptake of femtocells through standards development, ecosystem development and market education. How important is this? Take a moment to review the points above. Where marketing is important, the Forum could represent one unified, compelling voice. Where a standard, open interface out of the femtocell is required, the Forum (with the help of operator members) could help in defining the interface. Where operators pursuing diverse functionalities could keep femtocell scale in check, the Forum could help to coordinate requirements.

Peter Jarich is principal analyst with Current Analysis, a competitive analysis firm.