FRAMINGHAM (01/16/2004) - Among other things over the recent Christmas break, I read Michael Crichton's techno-thriller Prey about malevolent technology and also got involved in testing proprietary, high-speed wireless LANs. The former is (no surprise) a story of good technology "gone bad" -- but wait . . . so was the latter. Such is the world on the wild side of WLANs.
I know an explanation is required, and, no, I'm not implying that today's wireless technology has anything in common with Crichton's self-reproducing nano-technology but it has been rare that one test we've run has spelled trouble for another. With wireless, those days are over.
We all know that the "wall-less" nature of wireless is a double-edged sword. It's great for the flexibility that it provides us, but interference is the edge that cuts against us. In the world of 802.11b and 802.11g -- the 2.4-GHz spectrum -- the interference issue looms larger every day.
It is no surprise that proximity is a factor. After all, there is nothing new about businesses or individuals hearing "noise" from their neighbors. Only now that noise is radio frequency.
Our walk on the wild side began with one vendor that said the technology from a competing vendor made for an especially "bad neighbor." Imagine the guy in the apartment next door who cranks his music up to ear-banging proportions and you get the idea.
Just running the test was a challenge because of -- you guessed it -- interference. Not surprisingly, the research park in Boca Raton, Fla., that houses our lab also houses many other high-tech companies. These companies also avail themselves of WLAN technology.
Ultimately, we took our field test to, well, a field. To get away from it all (radio frequency, that is), we decided to test in an airplane hangar. Because we wanted to gauge interference and needed to be able to move access points around, the hangar adjacent to a private, grass airstrip in the Florida Keys worked out perfectly.
We proved the so-called "bad neighbor" theory. By itself, an access point using industry standard 802.11g (54M bit/sec) wireless sustained more than 22M bit/sec to a single matched wireless client. When the access point built using the competing technology was cranked up -- just 30 feet away -- to move files to its matched client using its proprietary high-performance mode, our standards-based pair saw its throughput drop close to 95 percent to about 1.27M bit/sec.
Network managers certainly have new challenges in managing, auditing and benchmarking 802.11 wireless. Fortunately, though, vendors have been preparing for this for a while now. Thankfully, we don't have to wait for test tools to catch up to the market need.
We are fortunate to have Agilent Technologies Inc.'s WLAN Advisor (aka AirMagnet) to help sniff out the existence of other networks. Recently, we took a big step forward by being one of the first labs to boast an Azimuth Systems Inc. W800 Wireless Test System. This provides us with a sophisticated test platform providing absolute control over our test environment and, most importantly, isolation from outside radio frequencies. Like it or not, it is time to start adding a line item for wireless test gear to your IT budget.
Tolly is president of The Tolly Group, a strategic consulting and independent testing company in Boca Raton, Fla. He can be reached at firstname.lastname@example.org.