Ultrawideband: Still a Player?
Page 1 of 2
The notion that ultra-wideband (UWB), the sometimes controversial unlicensed wireless technology approved for commercial use by the FCC in 2002, could be a competitor for Wi-Fi gained credence last October when Pulse-LINK demonstrated a prototype UWB-based wireless LAN.
The Pulse-LINK technology is interesting, to be sure, and will get even more interesting, but the company's president and chief operating officer Bruce Watkins hastens to reassure Wi-Fi vendors and others with a vested interest in 802.11.
"I'm not somebody who says that just because we're doing ultra-wideband, we're going to take over the world of Wi-Fi," Watkins says. "I don't begin to believe that for an instant."
"There are economies of scale in place with Wi-Fi, it's a mature technology. And it's just going to keep getting cheaper. Ultra-wideband will not be there from a price standpoint."
Watkins sees his company's products capturing maybe one to two percent of the current Wi-Fi market by the end of 2007, but it will never be a competitor across all of Wi-Fi's application markets.
"Is there a short-term threat [to Wi-Fi from UWB]? I don't think so. Is there a long-term threat? I wouldn't use the term threat," Watkins says.
The interesting thing about the Pulse-LINK WLAN demo was that despite a widely held view that UWB is only suitable for very-high-data-rate Personal Area Networks (PANs) with a range of no more than 10 meters, Pulse-LINK showed a prototype with a range of 20 meters and a data rate of 125 Mbps.
"The notion that there is something in the rules of physics or in the FCC regulations that means ultra-wideband can only go 10 meters is just wrong," says Watkins.
That was the point of the October demo -- to serve notice that what Pulse-LINK had been telling its investors and strategic partners was for real. It could extend the range of UWB and still remain compliant with FCC regulations.
"Rather than try and argue the point in white papers, we decided to just build something and prove it," Watkins says.
The first-generation prototype only operated in point-to-point mode, though, and the range -- about 65 feet -- is not enough for a practical LAN. Pulse-LINK expects to have a second prototype in the spring, however, that will operate point to multipoint at up to 100 meters, and support up to 250 client devices -- that's mobile devices -- with a data rate of 7 to 8 Mbps.
If you were about to say, 'Yes, but Wi-Fi is faster,' think again. The Pulse-LINK technology will deliver 7 to 8 Mbps at 100 meters. At that distance in 802.11a and 802.11g networks, data rates fall off to lower values.
"Data rate and range is an advantage," Watkins says, "but it's not the advantage."
He points to three other crucial factors that will give UWB -- or, rather, Pulse-LINK's proprietary implementation of UWB -- a leg up over Wi-Fi in very specific application markets: improved security, more precise positioning and non-interference with other networks.
The most important is security. Watkins says the first applications of Pulse-LINK's WLAN technology -- it could be a commercial product as early as the first half of 2005 -- will be in homeland security and other areas that require high security, such as airports.
Pulse-LINK claims there are seven ways in which UWB offers better security than Wi-Fi, even before applying encryption -- and because UWB offers higher data rates, it can more easily afford the high overheads of encryption.
It would be nearly impossible for would-be evesdroppers to intercept data on a UWB network. This is because of the way UWB radios operate -- by sending very short, randomly-spaced pulses over a broad range of spectrums. A UWB network is only sending pulses four percent of the time.
"The engineering challenge to even find our signal is monumental," Watkins says.
Because Pulse-LINK is pursuing a proprietary approach, even if an interceptor could find the signal, it would still need to know about proprietary channel definitions and modulation techniques before it could capture signal.
The Pulse-LINK technology, unlike some other CDMA-like implementations of UWB, requires the receiver to capture every pulse sent out by the transmitter. So even if an interceptor could hone in on a signal and capture it, missing just one sub-nano-second pulse would mean lost -- and therefore unreliable -- data.
Advantage three? "With UWB, we have highly accurate information on where client devices are located in the coverage area," Watkins says. "That has security implications and also makes possible some neat functionality."
Finally, early concerns from the cellular industry about UWB systems interfering with their networks have now been conclusively disproved, Watkins claims. Like others in the UWB space, he says established wireless players hyped the interference issue for mainly competitive reasons.
Pulse-LINK has shown that even when the antenna from a UWB system is touching the antenna of another system, the UWB system doesn't interfere. In fact, it's much more likely that other RF systems will interfere with UWB.
"Imagine if you wanted to use wireless technology in a security setting in an airport," Watkins says. "You've already got Wi-Fi there in the Admiral lounge. You've got people walking around with Bluetooth headsets. You've got tenants using freespace optics, LMDS -- there's a plethora of spectrum already in use. If you want to introduce something else, it's got to be something that doesn't interfere."
That's UWB, he says.
At the same time Pulse-LINK is developing the second prototype, it's also working on ASICs (Application Specific Integrated Circuits) -- a chipset -- for its UWB radio. That work should be complete by the end of 2004, Watkins says.