Preparing for ultra wideband test

Richard A. Quinnell, Contributing Technical Editor -- Test & Measurement World, 10/1/2004

In 2002, the Federal Communications Commission (FCC) approved the rules for ultra wideband (UWB) technologies in Part 15, Subpart F (Ref. 1). Included with imaging, radar, and other UWB uses were two communications applications: indoor and handheld. The standards for UWB communications are still being debated. While waiting for the resolution, test engineers can begin preparing for the challenges of UWB test. UWB communications seek to take advantage of the broadened power spectrum that occurs when RF energy is produced in brief bursts. The spectral broadening allows communications devices to operate "underneath" existing users of a frequency band without causing interference. By controlling the timing and frequency of the bursts, the communications devices can also avoid being subjected to interference from the existing users.

The exact modulation scheme that will be used for UWB communications is still under discussion. The FCC has deferred to the IEEE wireless personal area network working group (IEEE 802.15.3a) to define the standards, and that group has yet to reach an accord. Two competing approaches, one based on orthogonal frequency division multiplexing (OFDM) and one based on binary phase shift keying (BPSK), are locked in battle, with neither garnering the required 75% support needed for acceptance as of the July 2004 group meeting.

FCC Part 15, Subpart F, does give test engineers a place to start, however. It specifies the emissions limits for UWB transmission (see table). This specification includes notches to protect the frequencies that aviation and other safety-critical applications use and serves as the final authority on UWB signal-level testing.

"It's a strict test that is difficult to do," says Keven Trach, supervisor of engineering services at UWB vendor Time Domain (Huntsville, AL). "Emissions are so low and the ambient signals so great that it is nearly impossible to perform outside of a semi-anechoic chamber."

Trach also notes that even inside a chamber, many companies are finding that they must move to within 1 m of the device under test (as opposed to the typical 3-m distance) in order to get enough sensitivity to make accurate measurements.

There are other testing challenges, too. Trach points out that the FCC specification also calls for a measurement of peak emissions using RMS detection with a 50-MHz bandwidth. The requirement for RMS measurement is not typical of FCC Part 15 testing, so test labs may not have an appropriate detector. The alternative is to take multiple data points and hand-calculate the RMS value, a time-consuming task that adds to the time and expense of UWB testing.

There is also a bandwidth challenge. The required 50-MHz bandwidth is beyond the performance limits of most spectrum analyzers. Instead, test engineers may have to use a 1-MHz bandwidth and extrapolate the results in order to verify compliance.

Some recently introduced spectrum analyzers claim to have the required bandwidth, but Trach urges caution against blindly employing these tools at full bandwidth. He recommends that engineers compare the results obtained by the more traditional scaling approach to the full-bandwidth measurement to make certain that the two are in agreement.