The cons and pros of 2.4GHz wireless connections

THERE'S GOOD NEWS and bad news about the growing popularity of 2.4 GHz for wireless applications. Good news is there are plenty of network schemes around the globe using it, including Bluetooth, WiFi/802.11b, and the fast 54 Mbps blasting (almost finalized standard) 802.11g. It's popular enough that even Microsoft is cashing in on the wave with their own line of WiFi gear, just the sort of thing to connect a PC or the X-Box to the Internet.

Bad news is that there are plenty of other devices also transmitting on the same frequency, ranging from wireless phones to microwave ovens and plenty of non-standard long haul wireless gear. While you may be able to travel in most places in the civilized world by the end of the year for the potential of a stock 11 Mbps 802.11b WiFi connection, the airwaves may be cluttered up by electronic fog inhibiting the full speed of that connection. And did we mention the muddle with 802.11b, "draft" 22 Mbps wireless devices, and the (almost, but not quite done) 802.11g standard.

Interference on 2.4 GHz has been a problem for years. I was personally a witness to a wireless 2.4 GHz data circuit trial, the $40,000 non-WiFi gear designed to be a T1/1.5 Mbps replacement at distances of a couple of miles. The distance for this particular installation was less than three blocks, call it under 300 feet, clear line-of-site between the two antennas. Most of the time, it worked fine, but the link would intermittently cut out in the mornings, afternoon, and evenings. After a week or two, the downtimes were mapped to the up times of the one site's microwave oven, a sturdy early '90s model. Needless to say, the gear got dumped and replaced with a glorified DSL connection.

Mix in microwave ovens, wireless telephones, WiFi networking gear, Bluetooth, non-standard long-haul 2.4 GHz gear and it'll be pretty interesting to see how well 802.11g actually cranks in the real world. I'm not saying that 802.11g won't work for most users, but I'm not sure how well it will work in a non-pristine RF environment, especially if you don't talk to your neighbors. Another fun area will be the (eventual) overlap and conflict of 802.11b "hotspots". Take a large mixed-use office building, put a Starbucks on the first floor, a couple of businesses on the each floor, and figure out the coverage overlaps. If a 802.11g network on the second doesn't work at a full 54 Mbps, is it the fault of the Starbucks below them or the wireless network above them? Or the microwave oven in somebody's kitchenette? [Note to self: Good market here for "shielded" microwave ovens in the office environment.] Apartments and townhouses will also be nice places for 2.4 GHz fuzz. It's almost enough to make one run for the Cat-5 for GigE and be done with it.

On the other hand, this may turn out to be a blessing in disguise for 5 GHz land if not 802.11a specifically. After all, with a lot of squeezing and arguing, 802.11g is coming out the door at 54 Mbps, but there's not a lot of technical tricks left for another speed doubling in the near term. The 802.11a standard came out of the box at 54 Mbps, but wireless chip-maker Atheros already has tweaked their chip sets to deliver 108 Mbps in 5 GHz. Hopefully, the U.S. and Europe can work out their differences on the use of bandwidth between 5 and 5.8 GHz so we can get some economies of scale going.

Another trick hopefully, maybe, possibly on the horizon of wireless networks is UWB, ultra-wideband. According to advocates (i.e. UWB chips manufacturers), first generation UWB wireless networks have the potential to deliver 500 Mbps or faster within the typical home, dropping down to 100 Mbps if you have to link at distances of oh, say, the mansion with the stables and the pool boy's residence. In addition, there's much lower power consumption with UWB than a typical WiFi device, resulting in significant battery life for portable devices. UWB does have some political hang-ups; limited real-world testing with UWB simulation in commercial aircraft has shown that under some conditions (enough hedging here, hm?) the technology can play hell with navigational devices. µ