If one believes the tea leaves, we're soon to be drowning in Bluetooth devices. I've even seen a market research forecast that says more than 1 billion Bluetooth-equipped devices will be sold in the year 2005. Hyperbole aside, forecasts of even a fraction of that number could give network managers serious heartburn beginning as early as this year.
That's because Bluetooth occupies the same 2.4-GHz unlicensed spectrum used by most popular wireless local-area networks (most notably those based on the 802.11b standard), as well as many popular cordless phones. If all of those Bluetooth radios start beating against the poor, defenseless wireless LANs and cordless phones, no one is going to get a lot of work done.
One little detail left out by the doomsayers is that Bluetooth has a number of saving graces in the interference department. First, it's very low power-0 dBm, or 1 milliwatt. As a consequence, its negative impact will likely be limited to the range of a meter or so. Second, the typical usage scenario for Bluetooth is infrequent, ad hoc communications-synchronizations of notebooks and PDAs, exchanges of short documents, and the legendary wireless business-card swap. So how much interference can that innocent wittle Bluetooth cause, anyway?
Well, if Bluetooth stuck to the quick-and-dirty-cableless-connection scenario, likely not very much at all. But the marketing departments at Bluetooth companies have been working overtime, telling us about piconets, scatternets and voice capabilities, and therein lies the rub. We're seeing Bluetooth telephone headsets and access points. Gee, this is starting to look like a wireless LAN-and they're on all the time. And LANs, wireless or not, are usually mission-critical. We're back to the doom scenario: Bluetooth everywhere is going to cause interference to other devices. The net effect will be lower throughput, as opposed to a complete collapse, but who wants that?
The folks at the IEEE 802.15 committee, which is developing wireless personal-area network standards, are also working on interference management, among other issues. The result will be a "recommended practice," as opposed to a standard. In other words, the things network managers and users need to be aware of in order to deal with interference in the uncontrolled free-for-all that is unlicensed spectrum.
In the meantime, users will have to pick the best wireless tool for the job. And, if it hurts when they do that, wireless system designers will be forced to reply in the words of the doctor in the old joke: "Well, don't do that."