The Berkeley Wireless
Research Center (BWRC) is rich with ambient intelligence. And the
smarts aren't limited to the dozens of researchers toiling over
their keyboards and lab benches inside the 12,000-square-foot downtown
Center.
"Ambient intelligence is literally about making the fabric of
everyday life smarter through wireless computing," says Jan Rabaey,
associate chair of Berkeley's Department of Electrical Engineering
and Computer Sciences (EECS) and scientific co-director of the
BWRC with EECS professor Bob Broderson.
Born from Rabaey and Broderson's InfoPad
project, which in the early 1990s introduced the concept of
a wireless Internet device, the BWRC's goal is to integrate and
implement technology to bring the power of high-bandwidth and
ubiquitous communication into the built environment.
Imagine living in a smart home instrumented with hundreds of
tiny sensors that monitor energy, light, temperature, and motion.
An ID badge in your pocket automatically triggers a central computer
to tailor each room's environment ‚ from lighting to air conditioning
‚ to your preference. You step into the home office to take a
videoconference call that's beamed from the central high-bandwidth
Internet connection and projected life-size on the wall. At dinnertime,
you walk into your kitchen where a screen on the refrigerator
glows with a digital broadcast of a cooking show streamed from
the television set-top box in your living room.
"In the home, you want to shuttle your data flexibly," Rabaey
says. "You don't want to have to run a coaxial cable from your
main computer to your refrigerator."
To make this wireless vision become a reality, the BWRC works
on two main fronts: packing more bits into the limited bandwidth
of wireless networks and building cheap and tiny wireless sensors
that don't hog power.
"The wireless spectrum is heavily regulated and trying to jam
more bits through that regulated space is no simple task," Rabaey
says.
Indeed, it takes a multidiscplinary approach to develop integrated
wireless systems that satisfy the seemingly insatiable hunger
for more bandwidth. One way to squeeze more data into existing
systems, Rabaey explains, may be to exploit what historically
have been considered limitations in wireless communications. [See
In Favor of Fading Channels for more
information.] Another is to transmit data over much higher frequencies,
where there are very little traffic and no FCC regulations.
Peg
Skorpinski photo
Rabaey
inspects prototype radio boards for a miniscule smart radio
that will allow 20 people at a time to teleconference with
each other seamlessly. (Click for larger
image.)
|
That's where Berkeley's expertise in CMOS, the semiconductor technology
that's the basis of today's microchips, comes into play. While radios
are commercially available that can transmit at 60 gigahertz —
the part of the radio spectrum Rabaey calls mostly "unharvested"
— they run upwards of $20,000 because they're manufactured from
non-standard materials.
"Can we at Berkeley push CMOS into that bandwidth? Probably,"
says Rabaey, pointing out that the CMOS transceiver technology
at the core of most cordless telephones and wireless local area
networks on the market today was born at Berkeley.
The other pioneering element in BWRC's big picture of ambient
intelligence is the PicoRadio. Expected to cost only 50 cents
when built in bulk, PicoRadios act as the neurons in a wireless
network. Outfitted with combinations of sensors and ultra-low
power transceivers, the devices keep tabs on the world around
them and bounce the data through a self-assembled network until
it reaches a central computer where it's processed. Ultimately,
the miniscule amount of power they need to operate could come
from subtle vibrations in a building's walls and heating ducts.
According to Rabaey, the third-generation PicoRadio with all
its components on a single quarter-sized chip is expected to be
functional before the end of this year. Meanwhile, in close collaboration
with Berkeley's Center for the Built Environment and the Center
for Information Technology Research in the Interest of Society
(CITRIS), the researchers are exploring how wireless ambient intelligence
might impact our everyday lives.
"Without the crucial synergy between implementation and application,
you come up with a lot of useless technology," Rabaey says. "But
by understanding how technology may be applied, you come up with
better implementations."
Berkeley Wireless Research
Center
Jan
Rabaey's home page
CITRIS
Center for the Built
Environment
