This article may contain URLs that were valid when originally published, but now link to sites or pages that no longer exist. To maintain the flow of the article, we've left these URLs in the text, but disabled the links.


MIND

Flux
flux@microsoft.com
Douglas Boling
Bluetooth
I
love this industry. You hear about a new technology being proposed by a group of companies and all you have to do is go to Yahoo, AltaVista, or another search site to get everything there is to know about the new technology. The other day I heard about a technology called Bluetooth. To be absolutely clear, I am not talking about Gorm the Old's son, Harald Bluetooth (940-986), the Danish king.
      Bluetooth is a new wireless standard being proposed by Intel, IBM, Ericsson, Nokia, and Toshiba. Bluetooth is a specification of a short-range radio link that is in effect a desktop radio LAN. http://www.bluetooth.com provides an intriguing level of detail, but it's more a marketing device than a technical disclosure.
      The group currently working on Bluetooth is a mix of computer and cellular phone companies. That seems logical. We've all wanted a cell phone that was smart enough to turn into a cordless phone when we were at home or in the office. That'd go a long way toward reducing cell phone costs.
      On the computer side, while IrDA is okay, it'd be better if you didn't have to aim and shoot data from one infrared device to another. With a radio LAN, you
Bluetooth is a specification of a short-range radio link—in effect a desktop radio LAN.
just put two devices in the same room and they can talk. Those of you who've had to go out and buy that special cable that links a PCMCIA modem with a specific cell phone can see why having your laptop and cell phone link automatically would be a godsend. Sounds like connectivity nirvana, doesn't it?
      From a hardware standpoint, each Bluetooth device is a peer. Once a link is established, in what the companies refer to as a "piconet," one device becomes a master for that net. Each piconet can support up to eight devices, but multiple piconets can exist across an area, something referred to as a "scatternet." The range of a Bluetooth link is 10 meters. (That's around 33 feet for us Americans.) That's plenty of range for a desktop LAN. With more power, the link can be maintained up to a planned 100 meters.
      One plus for Bluetooth is that the technology uses a globally available set of frequencies, which eliminates the "different hardware for different countries" problem that plagues cell phones. I said frequencies because Bluetooth uses a spread-spectrum technology in which the transceivers hop from one frequency to the next at fixed time intervals. One of the advantages of a spread-spectrum design is that the link is more robust, especially in electrically noisy environments.
      The bandwidth of a piconet link is one megabit—pretty good for a low-power wireless network. That one-megabit bandwidth is parsed into three 64KB synchronous channels designed for voice communication, with the remainder dedicated to symmetric or nonsymmetric packet data communication. The packet bandwidth can be configured as a balanced 432 Kbps link or favoring one direction of the link up to a 721 Kbps–57.5 Kbps split. The packet bandwidth allocation can be reassigned on the fly.
      As a Bluetooth unit enters a piconet, it must respond to a challenge response interrogation that provides access-level security. Link security is provided by streaming hardware encryption with a 0-, 40-, or 64-bit key. While certainly not bulletproof, this level of encryption together with the frequency hopping and low power of the transceivers should make the basic link secure for most purposes. You can always encrypt the data at a higher level for more security.
      It'd be nice if my PC, my Palm-sized PC, and my laptop could all link just by being in the same room. The advantages of this type of ubiquitous connectivity reach further, though. With range cards that could go beyond 100 meters, you could have painless home networks with trivial file and printer sharing without having to run any wire. In an office, Bluetooth doesn't have the bandwidth to be the primary link for all PCs on a LAN, but in the home one megabit would do just fine.
      All these Bluetooth devices are only vaporware at this point. The group is actively seeking users of this technology, and considering how handy it'd be, I expect to see a fair number. The specification is supposed to be finalized this year, with products appearing sometime late in 1999. When this article went to press there wasn't any information about cost on the marketing site. But if these things have any chance of being placed in PDAs, the cost of the transceiver will have to be pretty low.
      As for the original Bluetooth, when I first went to Yahoo to search for information on this subject, almost all of the hits were pages discussing Harald. Isn't the net wonderful? I go looking for a wireless technology and I end up spending an hour reading about the Danish royal family of a thousand years ago. By the way, did you know that Harald Bluetooth was killed in a battle he fought against his own son? Talk about ungrateful offspring. Let's hope the Bluetooth radio LAN technology has a brighter future.
 

From the August 1998 issue of Microsoft Interactive Developer.