QuickStudy: Ultrawideband

Computerworld -

Listen to the Computerworld TechCast: Ultrawideband.

Origins of UWB

Gerald F. Ross first demonstrated the feasibility of UWB waveforms for radar and communications applications in the late 1960s and early 1970s. Originally developed by the Defense Advanced Research Projects Agency, the technology was called baseband, carrier-free, impulse communications or time-domain signaling, until the U.S. Department of Defense named it ultrawideband in 1989.

In some respects, UWB technology goes back to the dawn of radio and Guglielmo Marconi's early spark-gap transmissions. UWB is also a successor to spread-spectrum radio (also called frequency-hopping), a World War II technology that splits a broadcast across many different radio frequencies, using one at a time to avoid jamming. (Curiously enough, spread spectrum was invented -- and patented -- in 1942 by actress Hedy Lamar and composer George Antheil.) In contrast, UWB uses every frequency available to it, all at the same time.

UWB isn't a direct substitute for any other form of wireless communications, but it does some things that no other technology can match. A UWB transmitter sends billions of short-duration pulses across a wide spectrum of radio frequencies. These RF bursts come so fast -- lasting only from a few trillionths of a second to a few nanoseconds -- that each actually uses only a few cycles of an RF carrier wave.

This short duration gives UWB waveforms some unique properties. They are relatively immune to multipath cancellation effects, such as when a strong reflected wave arrives out of phase with the direct path signal, reducing the signal strength in the receiver. UWB pulses are so short that the direct signal has come and gone before the reflected path arrives, so no cancellation takes place. Because UWB pulses are so short, they can use very wide frequency spectra; this allows signals to use very low power, which minimizes interference with and from other radio frequencies, reduces health hazards and often falls below the normal noise floor, thus making it harder to detect.