While we continue to investigate our problem with deferred procedure calls (see last week's Gearhead) -- a problem that seems to have mysteriously vanished again -- we wanted to bring to your attention a tool that, if you are doing wireless stuff, you are going to want.

The tool is called Wi-Spy 2.4x and is manufactured and sold by the wonderfully named MetaGeek. The Wi-Spy is a USB (1.1 or 2) spectrum analyzer for the radio frequency range from 2400 MHz to 2483.5 MHz with a resolution of 328 KHz. It can detect signals in the range -110 dBm to -6.5 dBm with an amplitude resolution of 0.5 dBm and has a sweep time of 165 milliseconds.

In other words, this is a pretty snazzy piece of hardware, but two other features make the Wi-Spy outstanding. The first is its software, Chanalyzer, which runs on Windows 2000 or later with .Net 2.0 installed. Linux and Mac support are also available via third-party software.

Chanalyzer’s user interface shows three graphs that share a common X-axis of frequency; you can select this axis to be displayed as frequency or as Wi-Fi channels or Zigbee channels (I plan to cover Zigbee in more depth in a future Gearhead column -- for now, see this Network World story).

The top graph is the Spectral View. This is a “waterfall graph” (that is, one that scrolls down over time) that plots the selected time period (you can select from 15 seconds to one hour) against the frequency range, with each point color-coded by the signal’s amplitude. This highlights bandwidth use over time.

The graph below that is what MetaGeek calls the Topographic View. This display shows the popularity of the spectrum by plotting the percentage activity for each frequency and amplitude over the selected time period. In other words, it is a spectrum utilization map.

The bottom graph is the Planar View, which plots amplitude against frequency and shows current, average and maximum amplitudes. There are also two markers that you can place on the frequency axis to get data for a specific frequency.

What this deluge of data shows you is, for a given location, how the spectrum is being used and abused. For example, you can see where devices such as microwave ovens, Bluetooth devices and cordless phones might be adding noise and reducing Wi-Fi throughput.