Sign In

Communications of the ACM

ACM TechNews

DARPA $2m Contest Looks to Bring AI to Wireless Spectrum Provisioning


Wireless chips.

The U.S. Defense Advanced Research Projects Agency recently announced a $2-million Grand Challenge.

Credit: Reuters

The U.S. Defense Advanced Research Projects Agency (DARPA) recently announced the $2-million Spectrum Collaboration Challenge (SC2), which aims to infuse radios with advanced machine learning capabilities so they can collectively develop strategies optimizing the use of the wireless spectrum in ways that are not possible with conventional technology.

The current method of assigning fixed frequencies for various uses irrespective of actual, moment-to-moment demand is too inefficient to keep up with demand and threatens to undermine wireless reliability in the military as well as civilian applications, according to DARPA.

The SC2 will leverage recent progress in artificial intelligence and machine learning to facilitate new developments in those fields.

"We want to radically accelerate the development of machine learning technologies and strategies that will allow on-the-fly sharing of spectrum at machine timescales," says SC2 program manager Paul Tilghman.

SC2 will include three year-long phases starting in 2017 and finishing in 2020 with a live competition of finalists who have advanced from the two preliminary contests. The team whose radios collaborate more effectively with various types of other radios to dynamically optimize spectrum usage will win the grand prize of $2 million.

DARPA also will build a wireless test bed, called the Colosseum, for evaluating spectrum-sharing strategies, tactics, and algorithms for next-generation radio systems.

From Network World
View Full Article

 

Abstracts Copyright © 2016 Information Inc., Bethesda, Maryland, USA


 

No entries found