In a recent exclusive interview with ThinkFemtocell.com, Manish Singh, VP Product Management for Continuous Computing revealed their company has an astonishing 23 active engagements in the femtocell marketplace. This suggests that many more "undeclared" vendors may be joining the fray having identified strong market potential. Continuous Computing (also known as CCPU for short) is the leading provider of off-the-shelf protocol stacks for femtocells and femto-gateways, offering the basic software used by femtocell vendors and OEMs to incorporate into their products. So who are these different vendors, why are there so many, and how will the market shape up in the face of such diversity? For obvious commercial and confidentiality reasons, Continuous Computing are unable to provide a list of their sales pipeline, but we can surmise some reasoning behind this trend.
Mainstream basestation vendors such as Ericsson and Alcatel-Lucent are used to continuous ongoing downward pressure on their product costs. Operators typically expect price reductions of 15% year-on-year for their equipment, and have sourced systems from low cost Far-Eastern manufacturers such as Huawei and ZTE to reduce costs. This has been one of the main drivers for industry consolidation such as Alcatel-Lucent and Nokia-Siemens-Networks, both of which continue to lose money in many areas. It is said that Ericsson no longer make money on their hardware shipments, instead their managed services division generates the profits by installing and operating networks for operators around the world.
Femtocells are very much more of a FMCG (Fast Moving Consumer Goods) product, similar to mobile phones and other domestic gadgets. Consumers will expect them to be extremely useable, fault-tolerant, efficient and fashionable. But most of all, they will need to be very low cost. Mr Singh expects product volumes to be high, margins to be low.
One way of reducing costs is to incorporate the femtocell within the customer's DSL broadband modem or Cable set top box. This avoids additional cost and complexity of a second power supply, case, display etc. It also removes issues due to incompatible equipment, remotely managing multiple equipment and prioritising the voice traffic. It also allows the operator to capture a wider range of services from the same household by providing a package of mobile, broadband and IPTV which it can deliver and manage better and for lower overall cost.
We've listed quite a few femtocell vendors on this website, including femtocell only, femtocell subsystem (including femto-gateway) and DSL broadband modem variants. Clearly, we're not up to 23 yet so its likely there are quite a few potential entrants who are quietly working on this technology in the background, prepared to invest further and launch product in the next year or two. Even Continuous Computing would not be so bold as to say that all femtocell vendors use their stack, so the total is likely to be higher. We've not found a direct competitor to CCPU in this space yet, although there are a few protocol stack companies able to offer some key parts of the solution and existing RAN vendors would have many (but not all) of the functionality in their systems already.
Continuous Computing have partnered with picoChip to create a reference design from which OEMs can relatively quickly and easily create their own 3G femtocell. This is architecture agnostic and can support all of the main options, allowing a quickstart development with reduced risk and time to market.
Femtocell vendors then add their special "secret sauce" to this standard platform to differentiate themselves by improved operation, manageability and reduced impact on the external network.
Although we have seen a wide range of mobile phone handset vendors emerge to meet demand for variety, I think it is unlikely that in the long term there will be such a diverse or fragmented set of suppliers. Operators typically use around two radio network equipment vendors, and over 80% of that market is now satisfied by just 3 vendors - there are now less than 10 serious vendors for this equipment worldwide.
In order to avoid technical problems with their existing networks, and managing/supporting and maintaining a femtocell customer base, it seems likely that operators will again be conservative and limit their femtocell vendors in a similar way (i.e. to two or three per network). The core femtocell modules from a few leading suppliers may be incorporated into a wider range of 3rd party equipment (e.g. DSL modems, Set Top Boxes etc), rather than each vendor developing their own (including its "secret sauce").
We will have to wait and see if this prediction turns out to be true or not.
