Share this:
02/09/2004 08:00:00

Caltech Engineers Design a Revolutionary Radar Chip

PASADENA, Calif. -- Imagine driving down a twisty mountain road on a dark foggy night. Visibility is near-zero, yet you still can see clearly. Not through your windshield, but via an image on a screen in front of you.

Such a built-in radar system in our cars has long been in the domain of science fiction, as well as wishful thinking on the part of commuters. But such gadgets could become available in the very near future, thanks to the High Speed Integrated Circuits group at the California Institute of Technology.

The group is directed by Ali Hajimiri, an associate professor of electrical engineering. Hajimiri and his team have used revolutionary design techniques to build the world's first radar on a chip--specifically, they have implemented a novel antenna array system on a single, silicon chip.

Hajimiri notes, however, that calling it a "radar on a chip" is a bit misleading because it's not just radar. Having essentially redesigned a computer chip from the ground up, the technology is revolutionary enough to be used for a wide range of applications.

The chip can, for example, serve as a wireless, high-frequency communications link, providing a low-cost replacement for the optical fibers that are currently used for ultrafast communications. Hajimiri's chip runs at 24 GHz (24 billion cycles in one second), an extremely high speed, which makes it possible to transfer data wirelessly at speeds available only to the backbone of the Internet (the main network of connections that carry most of the traffic on the Internet).

Other possible uses:

* In cars, an array of these chips--one each in the front, the back, and each side--could provide a smart cruise control, one that wouldn't just keep the pedal to the metal, but would brake for a slowing vehicle ahead of you, avoid a car that's about to cut you off, or dodge an obstacle that suddenly appears in your path.

While there are other radar systems in development for cars, they consist of a large number of modules that use more exotic and expensive technologies than silicon. Hajimiri's chip could prove superior because of its fully integrated nature. That allows it to be manufactured at a substantially lower price, and makes the chip more robust in response to design variations and changes in the environment, such as heat and cold.

* The chip could serve as the brains inside a robot capable of vacuuming your house. While such appliances now exist, a vacuum using Hajimiri's chip as its brain would clean without constantly bumping into everything, have the sense to stay out of your way, and never suck up the family cat.

* A chip the size of a thumbnail could be placed on the roof of your house, replacing the bulky satellite dish or the cable connections for your DSL. Your picture could be sharper, and your downloads lightning fast.

* A collection of these chips could form a network of sensors that would allow the military to monitor a sensitive area, eliminating the need for constant human patrolling and monitoring.

In short, says Hajimiri, the technology will be useful for numerous applications, limited only by an entrepreneur's imagination.

Perhaps the best thing of all is that these chips are cheap to manufacture, thanks to the use of silicon as the base material. "Traditional radar costs a couple of million dollars," says Hajimiri. "It's big and bulky, and has thousands of components. This integration in silicon allows us to make it smaller, cheaper, and much more widespread."

Silicon is the ubiquitous element used in numerous electronic devices, including the microprocessor inside our personal computers. It is the second most abundant element in the earth's crust (after oxygen), and components made of silicon are cheap to make and are widely manufactured. "In large volumes, it will only cost a few dollars to manufacture each of these radar chips," he says.

"The key is that we can integrate the whole system into one chip that can contain the entire high-frequency analog and high-speed signal processing at a low cost," says Hajimiri. "It's less powerful than the conventional radar used for aviation, but, since we've put it on a single, inexpensive chip, we can have a large number of them, so they can be ubiquitous."

Hajimiri's radar chip, with both a transmitter and receiver (more accurately, a phased-array transceiver) works much like a conventional array of antennas. But unlike conventional radar, which involves the mechanical movement of hardware, this chip uses an electrical beam that can steer the signal in a given direction in space without any mechanical movement.

For communications systems, this ability to steer a beam will provide a clear signal and will clear up the airwaves. Cell phones, for example, radiate their signal omnidirectionally. That's what contributes to interference and clutter in the airwaves. "But with this technology you can focus the beams in the desired direction instead of radiating power all over the place and creating additional interference," says Hajimiri. "At the same time you're maintaining a much higher speed and quality of service."

Hajimiri's research interest is in designing integrated circuits for both wired and wireless high-speed communications systems. (An integrated circuit is a computer chip that serves multiple functions.) Most silicon chips have a single circuit or signal path that a signal will follow; Hajimiri's innovation lies in multiple, parallel circuits on a chip that operate in harmony, thus dramatically increasing speed and overcoming the speed limitations that are inherent with silicon.

Hajimiri says there's already a lot of buzz about his chip, and he hasn't even presented a peer-reviewed paper yet. He'll do so next week at the International Solid State Circuit Conference in San Francisco.

Note to editors: Color pictures of the tiny chip, juxtaposed against a penny, are available.

Media Contact: Mark Wheeler (626) 395-8733 wheel@caltech.edu

Visit the Caltech Media Relations website at http://pr.caltech.edu/media

Written by Marcus Woo