WiMAX and white spaces

The FCC decision to open up ‘unused’ TV frequencies in the 54-698MHz spectrum range for unlicensed usage has predictably provoked a wide range of responses.

On the one hand, there are those who have warmly and openly embraced the decision (such as Microsoft, Google and Intel) because it expands their addressable market.

And then there are the existing cellular operators in the US, which, while not appearing to say much in public about the FCC decision (as far as I can see), must be concerned about the prospect of greater broadband competition. Add in the fact that Verizon and AT&T have recently paid billions of dollars for licensed 700MHz spectrum, then the white spaces decision by the FCC may well be perceived in certain quarters – and with some justification – as grotesquely unfair.

Extra colour has been added to the white spaces debate by the pleas to the FCC from Dolly Parton (and the US entertainment industry in general) not to allow public access to this spectrum – which is used to ‘buffer’ different TV channels – on the grounds, among other things, of potential interference with wireless microphones.

Despite all the noise surrounded the white spaces debate, it’s still way too early to determine how the release of this spectrum for public use (which will become available in February 2009) will affect the US wireless broadband market, and which transmission technologies will gain market traction as a consequence.

Intel and Fujitsu are reportedly developing wifi and WiMAX radios in a single package for white spaces usage, where wifi is expected to be used in the LAN, and WiMAX for long-range connectivity. And, so one argument goes, because operators don’t have to pay for this spectrum, they can offer data services much cheaper than the cellular players and still make a profit.

If it sounds too good to be true then that’s because it probably is, at least in the short to medium terms. For one thing, the FCC insists that the fixed and portable devices that can be used in this unlicensed spectrum will have geo-location and spectrum-sensing capabilities (as well as power limitations), which are all designed to limit the potential for interference with TV channels (and wireless microphones). The devices also need to be FCC-certified. All this may prolong the device development cycle, of course, as well as increase costs.

Larry Page, Google co-founder, has reportedly said that chipsets in these devices should cost $5, but this seems a difficult target to meet as it implies very high volumes.

It’s also still very much open to debate how far device chipset vendors can software-programme the same silicon (which does promise to lower chipset costs) to successfully support different air interfaces and not jeopardise performance. There are, however, a number of smaller suppliers who claim this can be done (including Wavesat, Comsys and Altair).

What is promising from the WiMAX camp’s point of view is there’s growing evidence that the technology performs well in unlicensed spectrum, which should give it an advantage in the white spaces segment.

Towerstream, which offers T1/DSL replacement services for businesses using WiMAX in the unlicensed 5.8GHz frequency band, offers SLAs against 99.99 percent reliability. And looking at Towerstream’s average monthly churn rate during 3Q 2008 – a fairly impressive 1.22 percent – it seems to be doing a good job at satisfying its customers. (Jeff Thompson, Towerstream CEO, has been particularly vocal in supporting the FCC ruling.)

The FCC decision may not suit cellular operators in the US and New Clearwire, but WiMAX suppliers at least appear to be in a good position to fill in some of the white spaces.