Army turns to artificial intelligence to counter electronic attacks
WASHINGTON — The Army offered a $100,000 prize for a solution to an increasingly tough problem for commanders in the field: In a battlefield dense with electromagnetic signals, is there a better way to distinguish friendly transmissions from hostile attacks?
There is, according to a team of eight engineers from Aerospace Corporation, based in El Segundo, Calif. They won the prize by correctly detecting and classifying the greatest number of radio frequency signals using a combination of signal processing and artificial intelligence algorithms.
The competition, known as the “Blind Signal Classification Challenge,” was sponsored by the Army’s Rapid Capabilities Office, a small organization that looks for ways to apply commercial technology to solve military problems.
When the challenge kicked off in April, the Army gave all 49 competitors a large amount of recordings of various types of radio signals to use as “training data” so they could develop their algorithms. In early June, the Army put out a new data set that had no labels, so contestants had to blindly analyze and identify the signals. The Aerospace team learned on Aug. 27 that it had won the challenge.
Bradley Hirasuna, who oversees technology programs at Aerospace, said the application of AI in electronic warfare could help the U.S. military thwart attempts by enemies to interfere with military GPS or communications satellite signals. Identifying friendly and hostile signals is a constant challenge, he said. “Because these signals are becoming more complex, electronic warfare officers are becoming overwhelmed, not able to keep up with the threat environment.”
Russian forces reportedly have deployed jammers to disrupt GPS-guided unmanned air vehicles in combat zones like Syria and eastern Ukraine. The Pentagon also worries about electronic attacks against satellite-based communications systems.
Traditionally the Army’s electronic warfare units deploy large armored vehicles bristling with antennas to scan the area and search for radio-frequency signals. With AI-based tools, the commander can pull up a picture of the RF spectrum on a laptop screen and see what is going on. “If they need to suppress a signal, they can identify the signal to suppress. If they need to make sure friendly forces’ communications can happen, they know not to suppress those signals,” Hirasuna said. “This is the Army’s way of aiding the electronic warfare officer as the RF spectrum becomes more congested and more complex.”
Hirasuna said the company became interested in applying AI technology to signals intelligence long before the challenge was announced. “We have seen the advent of software defined radios,” he said. “Machine learning and AI are now used to exploit the power of software defined radios.”
Andres Vila Casado, one of the engineers on the team, said Aerospace benefited from its decades of experience as a contractor to the U.S. Air Force in areas like satellite communications and information technology. “The military sees what Google and Facebook have achieved in AI and says, ‘Let’s use some of that to improve our electronic warfare tools.’”
The Army informed the team it plans to continue to explore ways to turn the “blind signal” challenge into an operational system, Vila Casado said. “We want to develop our solution into real products they could use in the field.”