Jupiter Research Exclusive Research: Keeping Bluetooth Simple - Wireless Internet Magazine

Topics

Features
Above the Fray
Follow the Money
Market Research
Big Ideas
Technoptions
Developer Insights
Q & A
Opinions

Tools and Services

Archives
Calendar of Events
ASP Directory
Other Cahners Pubs

About Us

Submit a Release
Meet our Staff
Contact Us
Advertise
Magazine Subscription

Jupiter Research Exclusive Research:
Keeping Bluetooth Simple

by dylan brooks
January 21, 2002

With technical discussions intensifying around propagating Bluetooth to 100 meters and increasing speed from 1 to 2 megabytes and beyond, one must ask, why bother? The allure of true, seamless ad hoc networking inflated the hype surrounding Bluetooth and pushed it too far toward becoming a wireless local area network. While some ventures, such as Intel-backed Red-M and Classwave, are experimenting with Bluetooth-based LANs, the more common use for Bluetooth will remain simple device-to-device communication.

Bluetooth Takes The Front Seat

In the United States, two-thirds of wireless minutes are used in automobiles. Because earpieces and headsets have become increasingly common, wireless kits already are in high demand. While phone-to-peripheral connections certainly will exist in the car, automotive manufacturers may have the car serve the role of both peripheral (e.g., allowing the sound system to communicate via Bluetooth with a mobile phone) and terminal, (e.g., receiving data from General Motor's OnStar system and sharing data updates with a Bluetooth-enabled PDA or MP3 player). Given the slow, last-generation technology focus of most automotive companies, such a risky effort is a very remote possibility as a factory-installed option.

While it is technically feasible to use 802.11 and Bluetooth interchangeably, each should be used for what it is best at (and ideally not near each other because of interference). The Wireless Ethernet Compatibility Association has wisely focused on an existing global standard for networking – Ethernet – rather than creating a new one, making 802.11b the best current bet for connecting computers and servers (security issues notwithstanding). However, the 802.11b, or Wi-Fi, standard was not designed as a low-power cable replacement technology and generally is a poor choice for such menial tasks as replacing serial and audio cables. Bluetooth will excel in this area, delivering some of the original promise of IR in creating home PC setups with fewer wires. Successfully implemented, Bluetooth will be considered the USB of wireless networks.

Currently, Bluetooth often plays a me-too role with 802.11 gear, finding its way into external peripherals to attach to PDAs and even in PC cards for laptops. By the second half of 2002, embedded Bluetooth will face its litmus test: Can Bluetooth make more of embedded mass distribution than IR has? The Bluetooth Special Interest Group has failed to temper expectations of Bluetooth's ability to work seamlessly on the fly, in part creating the hype bubble that burst about the same time as CeBIT 2001 was held. In reality, many devices will be configurable to talk to one another, but true out-of-the- box ad hoc networking should not be expected until 2004.

The U.S. market has yet to see general retail availability of mobile handsets with Bluetooth connectivity – the standard is relegated to expensive PC cards. Of note, Compaq Computer Corp. recently included Bluetooth on its highest-end iPAQ. In Europe, wireless headsets currently are in use, though limited to a few handsets such as Ericsson's T39 GSM phone or the R520 GPRS unit, each of which has Bluetooth embedded. In Japan, Bluetooth is embedded in laptops from Sony, while Samsung's Bluetooth video camera aims to allow video-clip transmission via handsets to 3G networks. Such high bandwidth operations are not a good idea for use with the low-speed cellular infrastructure in the United States. However, in the home and enterprise, video may yet get some play because the ease of setup and high speeds – relative to current wireless connectivity – will work well for monitoring and temporary use. Bluetooth's strength will be the ability to connect non-computing devices such as earpieces, cameras and MP3 players.

Chipsets for Bluetooth are finally being offered at $5 or less from Broadcom, among others, but the total added cost of putting Bluetooth on a device still hinders Bluetooth's global distribution. In order to save on subassembly, which costs more than $25, manufacturers are looking to re-use existing processors and other parts of a mobile phone's assembly. Because of the customization involved, few handsets have embedded Bluetooth so far, and most of these are GSM phones. Adding any cost to handsets currently is problematic in the U.S. market. Mobile operators are subsidizing handsets less, and consumers are increasingly favoring the lowest-cost handset. In a recent Jupiter Consumer Survey, mobile buyers were concerned more with cost than with functionality, likely relegating Bluetooth to a niche role in the U.S. mobile market for the next two to three years.

Dylan Brooks is senior analyst with Jupiter Research.