In a reversal of the wireless norm, femtocells have gained most of their high pre-market profile from actual operator interest and need, rather than from pushes by major vendors, which have until recently stayed somewhat ambivalent about the miniature indoor cellular base stations. Their hesitancy is unsurprising ‘ femtocells introduce a consumer electronics cost base and supply chain, with accompanying low margins, to the once fat wireless infrastructure sector. But the strength of carriers’ attraction to the technology ‘ which can support improved indoor penetration, homezone pricing models and whole new approaches to mobile broadband build-out and services ‘ is pulling the majors off the fence, with Qualcomm and Cisco the latest examples, both working with UK start-up ip.access.
As always when the tier one suppliers invade a start-up market, the pioneers need to win the big names as partners, if not acquirers, to stand much chance of survival. In femtocells, the large RFPs and even contracts that are being issued by many large cellcos ‘ AT&T, Vodafone, Sprint and others ‘ will accelerate this process. In the first generation, the urgency of the need to get products to market should benefit the start-ups, since the larger vendors will need to partner with them to have femtocells ready and working in time. So, ip.access is working with Cisco on a deal to supply AT&T with a reported 7m units, worth around $500m, and also supplies units to Thomson and others; while fellow UK vendor Ubiquisys has agreements with NEC, Motorola’s home networking unit and Netgear (see Wireless Watch April 30 2008). If femtocells live up to their early promise in terms of real operator volume orders, and strategic importance to next generation wireless roll-outs, many vendors may decide they need to control their source by acquiring a favorite start-up.
The same process will be seen in silicon, where picoChip, also from the UK, has been the frontrunner. It has had the merchant market for W-CDMA and WiMAX femtocell chips pretty much to itself, and supplies both the UK start-ups among others, and its headstart is sure to make it the focus of potentially acquisitive interest from semiconductor majors in the near future. It will also have to face competitive pressure from larger chipmakers that decide to develop their own merchant silicon, and Qualcomm’s decision to take an unquantified but ‘substantial’ investment in ip.access suggests that the CDMA giant may be one of those, at least at the LTE stage. Taking an early march in all things LTE is vital to Qualcomm’s strategy to compensate for the inevitable decline in its CDMA heartland, and it particularly needs to keep close to the carriers and vendors focused on CDMA-LTE migration/coexistence (flagshipped by Verizon Wireless, which has recently waxed lyrical about femtocells as part of its 4G plans).
Some analysts believe the ip.access deal ‘ by all accounts heftier than a mere ‘eyes and ears’ stake – shows that Qualcomm wants to break into the W-CDMA femtocell market itself (something that could be threatening to picoChip, but could also bring the two chip designers closer), but insiders within the larger company suggest that it wants to ensure a position in the AT&T roll-out, and other major deals, only at the LTE stage. The official line was expressed by Frederic Rombaut, head of Qualcomm Ventures Europe, in a statement: ‘3G femtocells will have a very important role in future mobile network architecture. Ip.access has an innovative approach to the 3G femtocell market that will enable our customers to enhance the delivery of 3G services to mobile users.” Others are more aware of the impact of femtocells on the traditional Qualcomm model, demanding very low BOMs and likely to attract the sort of competitors with which the San Diego powerhouse does not usually have to deal. ‘We hated the idea of Wi-Fi router chips and these are just as bad for us,’ said one source. Texas Instruments’ infrastructure business has also expressed doubts about whether it can make acceptable margins from the femtocell market once the devices hit the sort of price points carriers are demanding (sub-$100).
These private misgivings at the silicon majors reflect the realities of a wireless infrastructure market that is moving towards pricing and value chain models more familiar in consumer electronics – but femtocells also represent a major new source of revenue that may accelerate 3G expansion and 4G build-out in some markets, by making the economics more viable for operators. In a sector where new revenue streams and accelerated operator purchasing cycles are increasingly hard to find, it will be impossible for the tier one chip or equipment makers to stay on the sidelines for long.