The Race for White Space
By Kevin Werbach, Mon Sep 27 08:00:00 GMT 2004

In recent years, governments believed the best thing to do with wireless spectrum was to make it expensive. Now, regulators are realizing the value of making spectrum cheap, or even free.

The implications could be profound. This shift is clearly evident in the battle over the so-called “white space” in US television broadcast bands. The outcome of this obscure FCC proceeding may help shape the wireless landscape for decades to come.

Access to wireless capacity is the essential gating factor for all the exciting developments in the mobile world. Without rights to use spectrum, the best technologies and business models are worthless. And for nearly a century, governments have controlled and allocated spectrum with a heavy hand that would put the old central planners of the Soviet Union to shame. The fortunate few who benefited from this command-and-control system have every incentive to preserve their government-granted fiefdoms against potential competition, keeping wireless capacity artificially scarce.

Spectrum auctions, the rage of the 1990s, were supposed to be an improvement. Yet auctions quickly turned into a means for raising revenue rather than maximizing the benefits of communication. The infamous European 3G auctions nearly killed the markets they were supposed to create, by saddling firms with enormous debt. The problem is simple supply and demand. To extract the most money from spectrum access, governments have incentives to keep it scarce. That’s exactly the opposite of what they should be doing.

What if, instead of limiting wireless access, regulators saw it as their primary mission to expand it? And what if technology made it possible to unlock new capacity without taking anything away from existing licensees? And, best of all, what if that opportunity was in the most valuable and seemingly most heavily populated portion of the spectrum?

Fast forward to late 2002. The US Federal Communications Commission released a major task force report proposing to move away from command-and-control spectrum management. For the first time, the Commission provisionally endorsed the unlicensed “commons” model, in which companies are free to build devices and deliver services without seeking permission from the government or a private licensee.

Among the first fruits of the FCC spectrum policy reform effort was a notice of inquiry to open up the “white space” in the TV broadcast bands. The FCC followed up with a formal proposal in May 2004, which is now open for public comment.

The FCC’s proposal would mean more opportunities for both fixed and mobile wireless technologies, spanning a range of capabilities. What exactly will be deployed is impossible to know ahead of time, but the experience so far in higher-frequency unlicensed bands is encouraging. And adding more capacity in the broadcast bands below 1 GHz would be particularly significant.

Generally speaking, the lower the frequency, the better a signal propagates. A lower-frequency system tends to have better range and better ability to penetrate trees and walls. That makes low-frequency capacity particular valuable for applications like residential broadband connectivity, and for situations in which deploying a large number of antennas is prohibitive.

Because they came early to the wireless party, over-the-air television broadcasters control the lion’s share of low-frequency “beachfront spectrum.” Shockingly, most of that spectrum isn’t even being used. A technical survey last year found that two-thirds of UHF spectrum (TV channels 14-69) was effectively idle in downtown Washington, DC, a relatively congested urban area. Many of those channels are simply empty, maintained to give their owners guaranteed retransmission on cable and satellite networks, which serve 85% of US households. Others were officially in use, but their signal was undetectable.

Even within the primary broadcast frequencies, channels 2-13, there are substantial gaps. For example, every major city in the US has stations either on channels 2 and 4, or channels 3 and 6, but never both. Television receivers in the 1940s, when the bands were allocated, couldn’t distinguish a local station on channel 2 from a distant one on the same channel. Today’s sets have no such limitations. Yet the constrained channel allocation remains. Elsewhere in the world, especially in developing countries, even less of the spectrum allocated to broadcasting is actually in use.

The FCC is proposing to allow unlicensed wireless devices to operate in the un-used portions and buffer zones of the broadcast bands. Technical standards and techniques such as software-defined radio can be used to ensure these devices do not impinge on existing broadcasters. For example, a device could incorporate a global positioning system receiver and check a database of spectrum allocations to identify the available white space where it is located.

The battle for the white space has now been joined in earnest. And the combatants are no longer the usual suspects in telecom policy debates. High-tech companies such as Intel and Microsoft, eager to sell gear and software that would power a new wave of unlicensed broadband activity in the broadcast bands, have lobbied the FCC in favor of the white space proposal. Public interest groups, consumer electronics companies, and network equipment vendors have also expressed support for the proposal.

As one would expect, though, not everyone likes the white space idea. Incumbent broadcasters have raised concerns about interference, especially in light of the impending transition to digital television. Just before the deadline for filing comments in the proceeding, a standards body requested a six-month delay, saying that it needed time for technical due diligence. The fact that two trade associations for TV broadcasters filed in support of the motion, alluding to the participation in the standards process, suggests there may be more going on. [Note: I filed comments with the FCC opposing the extension. The Commission split the difference, granting a 90-day delay.]

With a Presidential election approaching and major turnover at the FCC likely, there is no guarantee the Commission will vote on the white space proposal any time soon. If and when it does, it would have powerful implications. In the near term, opening up the broadcast white space would allow new kinds of wireless devices to be deployed cheaply. This would create incentives for innovation in software-defined radio and related technologies, leading to more robust and flexible systems. If other countries follow the US lead, we could see an explosion of broadband wireless data services.

Longer-term, by establishing the idea that unused capacity should be opened up rather than hidden away, the white spaces decision would create a valuable precedent. Just because something was important fifty years ago doesn’t mean today’s applications and technologies should be shut out of the market, especially when both uses can coexist.

Demand for wireless capacity will only increase as users, usage and applications for mobile communication grow. The frontiers of unused spectrum were allocated long ago. The best way to meet the needs of the future is to exploit under-utilized capacity. The FCC’s white space proceeding could help move us down that road.