Jan Rabaey, University of California, Berkeley

In the late 1980s, Jan Rabaey was working on the Infopad, an early wireless Internet access appliance notable for storing computing power in the network rather than the device. A group of industry companies assisted with the project, and while all involved found the collaboration fruitful, they realized they could do more.

“You get something from a consortium like that, but you don't get its full value,” Rabaey said. “So the idea was, ‘Let's have something that's much more interactive.’”

The result was the Berkeley Wireless Research Center at the University of California, Berkeley. Its dual mission was to research wireless systems and encourage collaboration between academia and industry. Intel, Agilent, Texas Instruments, Ericsson and Qualcomm are among those that contribute to BWRC's budget, which is supplemented by government grants.

According to Rabaey — one of the center's two scientific directors, along with Robert Broderson — other university wireless research centers focus mostly on services and high-level system development. BWRC concentrates on nuts and bolts like lower power requirements, smaller components and cheaper parts.

Rabaey's latest project is the development of pico nodes, which can be used in any environment where data needs to be gathered from multiple points, such as home networking. What sets these nodes apart, however, is their power requirement — only 100 microwatts. At that level, they require no outside power sources, instead harnessing ambient light and energy to generate their own juice.

The pico project brings Rabaey full circle: Early in his career, he also focused on power implementation issues, a path that led to present pursuits. “I saw the big growth was in wireless. It created a set of challenges that were not existing in wireline, and the opportunities were a lot bigger, too.”