TV White Spaces Could Mean Plenty For WISPs

The FCC decided to allocate unused TV spectrum to wireless use in its Order of November 4, 2008 (.pdf). The decision opens up channels 2 to 69 in four bands of frequencies in the VHF and UHF regions of the radio spectrum (54-72 MHz, 76-88 MHz, 174-216 MHz, and 470-806 MHz), notes WISPA’s FCC Committee Chair Jack Unger. (WISPA’s press release on the topic is here.)

The rules are not simple, however, and not all of that spectrum will be vacated. According to a study by the New America Foundation (summary .pdf), the more rural an area, the larger the amount of available spectrum. For WISPA’s members, this is generally good news.

This spectrum is valuable because it propagates well. It travels far, going through trees and even over hills. When I think about this spectrum, I like to think in wavelength rather than frequency. Wavelengths range from (nice chart here) almost 20 feet at 54 MHz to barely 4 inches at 806 MHz.

The spectrum you’re used to dealing with at 2.4 GHz (4.8 inches) is blocked by water and therefore also by trees (and makes microwave ovens work at power levels thousands of times greater than what you use). Those of you who use laser or LED have links that can be blocked by any weather condition that you cannot see through (generally fog). The longer wavelengths are blocked only by larger obstacles — not by water or fog or even the walls of a house.

Hotels might love this spectrum. As you know, many early Wi-Fi deployments in hotels required several radios per floor. Later applications placed radios outside the hotel, beaming in. White spaces equipment could be more efficient here, and in many other important applications.

Besides propagation characteristics, wavelength also determines antenna size. Generally, an antenna has to be at least 1/4 the wavelength of the spectrum. Peter Stanforth, CTO of Lake Mary, Fla.-based Spectrum Bridge, which is both an equipment manufacturer and spectrum database provider in this area, notes that cell phone companies cannot use spectrum that would require a large antenna.

(Imagine. Inventor: “I’ve just created the first 2 inch cell phone! Eureka!” Antenna maker: “And I’ve got a 2 foot antenna to put inside it!”)

Licensed vs. Unlicensed Spectrum

WISPA strongly advocates that this spectrum be subject to minimal license requirements (licensed lite) and not be auctioned off. Unger worries that the government will look only at the cash the auctions could bring in and will ignore the damage such auctions do when they enable companies to lock down spectrum and prevent others from using it.

The harsh truth of spectrum auctions is that buyers pay for the right to acquire spectrum for a fixed amount of time and for limited public benefit uses and end up owning the spectrum in perpetuity, a point made in 2007 in a key study by the New America Foundation (see How They Got $480 Billion in Spectrum Giveaways).

But others are arguing strongly that licensed is the only way to use this spectrum. George Ou, policy director of Digital Society, wrote, “54-698 MHz propagation is too powerful for unlicensed use since people will get interference from 50 neighbors instead of only 5.  High-powered unlicensed spectrum usage has never been technically feasible because we don’t have mobile radio technology that can get around the interference issues.”

That opinion might work if you took into account only the technologies used to manage Wi-Fi, but it’s out of date today. It’s the sort of mistake that a network engineer who is not versed in RF would make (and is why many WISPs have an RF engineer and a network engineer as separate jobs).

Today’s radios can whisper, avoiding interfering with each other, a point noted in the New America Foundation’s excellent Cartoon Guide to Spectrum Policy (see pp.4-5) that was published in 2004.

Many attribute this idea to Timothy Shepard, who filed the idea with the FCC in 2002 but came up with it before then. Shepard’s filing is well worth reading.

How Can We Share The White Spaces?

The FCC considered several methods for sharing white spaces, but settled on a geo-location database. Stanforth says that the database (Spectrum Bridge will offer one, as will other companies, including Google) will be very granular, to within a few thousand square feet (he said “a few hundred square meters” and you mutiply by roughly 9 to get square feet).

The system must work in order to show that the spectrum can be unlicensed.

WISPs and Jack Unger are worried about the role that unlicensed wireless microphones might play in all of this. WISPs are all too familiar with the many devices that clutter 2.4 GHz and 5.x GHz spectrum. Many now incorporate this issue into their customer service script. “Have you recently purchased a baby monitor, cordless phone …”

Wireless microphones could play the same role in white spaces networks — denying service to customers. WISPA is proposing that wireless microphones get to operate on 2 specific channels in any area. This would allow these devices to continue operating without interfering with or experiencing interference from WISPs — even though their manufacture may have gone against past laws.

Unger is concerned that if unlicensed wireless microphones are allowed to operate anywhere in the white spaces spectrum, their use could affect a WISP’s entire network. Remember that different parts of the white spaces spectrum have very different wavelengths. Where a WISP deploys at the lowest frequency, longest wavelength spectrum, it will be depending on the propagation characteristics of that spectrum to connect a low density network over a large area. If a wireless microphone came on the air and forced a WISP to switch to a shorter wavelength higher frequency spectrum, part or even all of the network could get cut off.

Stanforth believes it won’t be necessary to allocate specific channels to wireless microphones and he says that his company has not encountered any wireless microphone problems in its four white spaces test deployments to date. But he also admits that each deployment had only a few dozen radios.

When Will We Have Equipment?

Unger believes that equipment could be cheaper if designed for specific portions of the white spaces. A TV antenna is large because of the wide variety of wavelengths it needs to receive. Stanforth admits that no single radio could handle all of the white spaces spectrum, but argues that a two radio solution might work (and he would know).

Stanforth thinks the FCC might make its final decision on white spaces issues during the summer, and that equipment could be on the market quickly. It would initially be based on other designs, such as those for Wi-Fi. “I expect the price performance of this equipment to be better than many alternatives,” Stanforth says.

He notes that although there are no currently publicized plans to base white spaces equipment on WiMAX, such a solution would provide obvious benefits to WISPs by delivering QoS for demanding services such as VoIP.

As to databases, Stanforth says he expects there will be between three and nine commercial entities offering them, and the FCC hopes that this competition will ensure that the databases maintain high standards and affordable pricing.

Unger feels that this is the most important FCC proceeding right now. Details will matter. For example, the FCC is considering height restrictions for wireless antennas, perhaps as an attempt to limit the propagation of the longer wavelengths, but WISPs in rural areas could benefit from placing antennas high on existing cellular or radio towers.

If we get it right, then rural areas will get broadband and WISPs will have plenty of spectrum to use.