U.W.Be Careful What You Wish For
By Kevin Werbach, Wed Oct 22 14:30:00 GMT 2003

A funny thing happened on the way the wireless revolution. The US government actually did what the entrepreneurs asked for.


I'm talking about ultra-wideband (UWB), a promising technique that spreads very low-power signals across huge swaths of bandwidth. UWB proponents tout several extraordinary properties of the technology. It can "underlay" licensed signals without interference. It can penetrate walls and other obstacles. Its transmissions resist detection, jamming, and the fading effects that plague other wireless systems. And it allows for radios that are smaller, cheaper, and less battery-draining than anything before. Unfortunately, those benefits haven't yet shown up in real-world products.

Four years ago, USA Today called the potential impact of UWB "astounding" and said the technology "might be as important as the transistor or electric light bulb." The Economist, Business Week, and the New York Times also trumpeted UWB's potential.

The storyline then was that UWB systems were illegal under FCC rules. It was a classic David-and-Goliath conflict: determined entrepreneurs fighting entrenched bureaucrats, military agencies worried about interference with their radars, and the established communications industry worried about competition. David won. In February 2002, the FCC adopted rules permitting UWB communications systems. The rules were a compromise to ensure that existing systems were not subject to interference. Yet the nascent UWB industry hailed the FCC action. Now at last they would conquer the wireless world with their superior technology.

It hasn't quite worked out that way. Eighteen months after the FCC decision, no mass-market UWB devices are available. Only one startup in the space has venture funding. Competing vendor alliances are squabbling in standards bodies, even trying to drag the FCC back in as an arbiter. The early UWB leaders are struggling, with some already shut down. What happened?

One issue is market positioning. UWB's cheerleaders described it as a technology that could do everything. When it comes to building and selling products, though, one size doesn't fit all. The laws of physics and business mean that a new communications technology must secure a market niche that gives it visibility and adoption. By the time UWB was approved, the wireless local-area networking niche was already taken by WiFi. So the leading UWB companies positioned UWB as the technology for sending rich media between consumer electronics and mobile devices. They formed the WiMedia Alliance to promote this application.

The problem is that the networked home media market doesn't really exist yet. And there is already significant confusion among existing home-networking technologies. UWB may be the best solution for wirelessly streaming audio and video, but it has several hurdles to overcome.

A second challenge for UWB involves standards. At the root of the problem is a barely-noticed element of the FCC's decision. UWB traditionally meant impulse radio, otherwise known as carrierless communication. Most wireless systems impress a signal on a carrier wave of a specified frequency. Impulse radio uses extremely short, highly synchronized electrical pulses, with no carrier wave involved. Impulse radio is what made UWB so novel, and gave it many of its beneficial properties.

The FCC's order didn't require impulse radio. It defined UWB as essentially any wireless system that spreads signals across at least 500 Megahertz of bandwidth. This change has major consequences. Says Roberto Aiello, founder of Staccato Communications and a leading UWB researcher: "My view of UWB changed after the FCC Report and Order. That's when I started thinking of UWB as available spectrum instead of technology."

Most current UWB vendors are using conventional spread-spectrum techniques, similar to those employed in 3G phone networks, because they are better understood than the exotic impulse radio. As a result, their initial offerings are more evolutionary than revolutionary. And several different approaches are competing in standards bodies. The FCC's decision means the UWB spectrum can be partitioned into 500 MHz chunks, with transmissions sent simultaneously in more than one. One group of vendors is pushing this "multiband" approach as a standard, with others favoring a single undivided band. Whatever the outcome, the debate is splitting the industry and potentially slowing product development.

The FCC probably did the right thing. The private sector, not the government, should decide the best technical approaches to use. The FCC isn't running a research project; it's trying to facilitate new communications systems that can succeed in the market.

That leads to the last challenge for UWB, which may be the most serious. While there's no question that UWB works in principle, there are increasing doubts about the current generation of UWB devices. Though vendors continue to issue optimistic press releases, few UWB systems have been certified by the FCC. Rumors abound that laboratory success isn't translating to real-world products.

It turns out that the current commercial UWB efforts differ from earlier military and research systems. The military was interested in UWB because of its low probability of detection and its ability to function in low-visibility conditions such as desert sandstorms where conventional radios and radars have trouble. Sending large amounts of data quickly wasn't essential. Says Robert Fontana, who spent years building classified UWB systems, "I would never build a high data rate system for the military, because it sticks out like a spare thumb."

In the commercial market, however, high data rates are important. This is especially true for the home media applications where UWB vendors are focused. Though UWB can support high-speed transmission, it is proving more difficult than expected to make the transition. "The thing people forgot is that all these great properties were only valid at low data rates," says Fontana. He now runs a startup that builds UWB systems for location detection and other functions closer to the traditional military applications.

The technical hurdles to high-speed UWB will eventually be overcome, or vendors will find applications that don't involve high-resolution video streams. UWB's fundamental advantages in power consumption, spectral efficiency, and cost are bound to give it a role in the market. The eventual future of UWB remains bright. It just may take a bit longer to get there.