Sign In

Communications of the ACM

ACM TechNews

Photonic Neuron May Compute a Billion Times Faster Than Brain Circuits

Paul Prucnal

Paul Prucnal, a Princeton University professor of electrical engineering, is leading a collaboration between researchers at Princeton and Lockheed Martin to develop a device that uses fiber-optic circuits that transmit information a billion times faster t

Photo courtesy of Brian Wilson

Princeton University researchers and Lockheed Martin are developing fiber-optic-based computational devices that work 1 billion times faster than human neurons. The researchers say the technology could lead to computer circuits that can make instantaneous decisions in life-or-death situations, and process huge amounts of data, such as video signals that guide robotic cars.

The researchers, led by Princeton professor Paul Prucnal and Lockheed Martin researcher David Rosenbluth, used clear fiber-optic cables instead of electrical wires, enabling the information to travel at nearly the speed of light. Unlike conventional fiber-optics, which converts information back into electrons before processing, the Princeton technology processes the information while it is still encoded in light.

Before the project started, the researchers noticed that the mathematical equations used to model neural and fiber-optic networks were very similar in their overall formation, despite using different variables. "We are transposing learning, inhibition, and other behaviors typical of neural processing onto fiber-optic circuits," Rosenbluth says.

From Princeton University
View Full Article

Abstracts Copyright © 2011 Information Inc. External Link, Bethesda, Maryland, USA 


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account