Regulating time
By Maury Wright, Editor in Chief -- EDN, June 24, 2004
I wouldn't normally advocate increased regulatory control on emerging technologies. In contrast, I deplore the fact that the FCC and other agencies have hampered deployment of such things as digital TV. But the UWB (ultrawideband) technology that's percolating in labs worldwide sorely needs a careful assist from regulatory agencies. Now, I'm not saying that UWB is a slam-dunk. We still haven't seen it in applications such as moving video outside contrived laboratory demos. But, even if the technology does prove capable and affordable, I believe it will struggle without some governmental guidance.
UWB is a different animal when it comes to use of the airwaves. Most wireless—and, for that matter, wired—communication schemes use a frequency band that agencies such as the FCC define. By definition, UWB spreads its power pulse over an ultrawide band. This burst led many, fearing UWB would trample on narrowband channels, to petition the FCC to not allow it for any commercial use. The naysayers predicted the demise of everything from reliable emergency radio communications to interference with upstart technologies such as WiFi.
Since the FCC approved a trial spectrum mask for UWB applications, however, it appears that the technology can coexist with narrowband channels. But incompatible UWB radios in the same market might make poor neighbors. Just as Bluetooth and WiFi can suffer degraded performance when in proximity, two nearby and incompatible UWB systems could drown one another out.
You could argue that the issue is as simple as establishing a UWB standard. Special Projects Editor Matt Miller recently covered the UWB standard impasse between the MBOA (Multiband OFDM Alliance) and the UWB Forum (see "Ultrawide division: Rival UWB groups tout support").
But setting a single UWB standard doesn't address the underlying issue. One standard may one day prevail, but we need a way to carve up the UWB spectrum, just as we carve up the frequency-domain spectrum. Carving up the UWB spectrum, however, must happen in the time domain, not the frequency domain.
Start-up UWB specialist Pulse~Link has proposed a spectrum-allocation scheme—essentially, an arbitration layer that underlies all UWB systems—that assigns time slots to UWB equipment. The company claims that its CSP (Common Signaling Protocol) would ensure that you could operate multiple incompatible UWB systems in the same physical space.
I'm far from expert enough to decide whether the CSP is the right answer. But I do believe that we need the regulatory agencies to ensure that incompatible UWB systems can operate in proximity to one another. This argument doesn't apply just to competing UWB schemes that promise to move video about a home. Pulse~Link, for instance, plans to use UWB PHY (physical) and MAC (media-access-controller) layers that differ from those in the standards battle for secure applications.
You might think that players pushing UWB standards would welcome a way to ensure that future products would work unimpeded. Some of the large companies fighting the UWB standards battle, however, want things their own way so that they can dominate the market. I hate to name names, but companies such as Intel and Texas Instruments are just plain wrong in ignoring spectrum allocation. Forget your precious standards fight for a minute, and make sure we end up with a good way to share the time domain in UWB the same way that we share the frequency domain in today's varied RF systems. Government-driven frequency allocation may not always suit everyone, but image one communication technology's taking up the entire frequency spectrum. If the UWB players don't immediately address the allocation problem, I hope that the government steps in.
Contact me at mgwright@edn.com.
