Northrop Grumman will develop machine learning algorithms to help the Boeing EA-18G Growler’s electronic warfare system pin down and jam fast-changing enemy radar signals.
Such machine learning algorithms would be needed “against agile, adaptive, and unknown hostile radars or radar modes,” according to the announcement of a $7.3 million cost-plus-fixed-fee contract by the US Navy on 25 April.
Modern radio frequency (RF) transmitters, including active electronically scanned array radar, can use a technique called frequency hopping to confuse systems that detect and jam their signals.
Countering such dynamic radar techniques is further complicated by operating in airspace that is increasingly crowded with civilian and commercial RF signals.
Machine learning could help the EA-18G Growler’s crew locate the hostile radar signals among the noise, and then direct the aircraft’s electronic attack units to jam those signals. Machine learning software uses statistical methods to find patterns in large data sets which would be difficult to analyze efficiently by hand calculations or other computational methods.
The US Navy’s EA-18G Growler is designed to blind an enemy by jamming or destroying their radar and communication systems.
Work on the contract is planned to be performed in Bethpage, New York, and is expected to be completed in December 2019.