Scientists at the U.S. Defense Advanced Research Projects Agency (DARPA) in Arlington, Va., are asking researchers in the Lockheed Martin Advanced Technology Laboratories (ATL) in Cherry Hill, N.J., to work with EW experts at the Raytheon Co. Space and Airborne Systems segment in McKinney, Texas, to field test EW machine learning software written to detect and jam enemy adaptive communications automatically.
The sole-source contract to Lockheed Martin ATL is part of the third phase of the Behavioral Learning for Adaptive Electronic Warfare (BLADE) program, which is developing technology to defeat adaptive wireless communications threats.
Wednesday's modification increases the value of DARPA's BLADE contract to Lockheed Martin to $29.4 million, DARPA officials say. Lockheed Martin has been working on the DARPA BLADE program since its inception in 2010.
Related: Siemens, Carnegie Mellon, and HRL pursue artificial intelligence to analyze images and text
Lockheed Martin has developed software algorithms and techniques that could enable U.S. EW systems to learn how to jam new radio frequency threats automatically. Now Lockheed Martin will work with Raytheon to test this software in a tactical EW system under realistic conditions in the field.
The third phase of the BLADE program is to expand the operational envelope of the BLADE system by integrating the algorithms into a Raytheon tactical communications jamming system, addressing real-world problems like multipath, complex electromagnetic environments, moving platforms and targets, and field testing the system against representative threat environments at a government test range.
Lockheed Martin and Raytheon will integrate BLADE software into an EW system and perform ground and flight tests on the developmental unit. The BLADE program focuses on developing an electronic attack system capable of automatically jamming new wireless communication threats in the field. The system is designed to detect and characterize new threats, learn to jam it, and assess jamming effectiveness in the field.
Lockheed Martin experts have demonstrated the feasibility of machine learning algorithms to separate and characterize communications networks, optimize jam waveforms, and provide real-time battle damage assessments.
Related: Programmable radar and adaptive electronic warfare take center stage
Lockheed Martin also worked with DARPA to achieve closed-loop machine learning and prosecution of jam waveforms for the first time against known and unknown radio frequency communication systems. For the program Lockheed Martin experts developed software for a digital radio transceiver that provides front-end real-time analog processing and back-end digital processing to prosecute a limited number of signals over a limited number of bands. Now company experts will test their software in the field against realistic threats.
The DARPA BLADE program seeks to push the bounds of machine learning -- a kind of artificial intelligence -- to counter enemy threats from wireless adaptive communications such as battlefield radios, command and control networks, and RF triggers like cell phones used to detonate improvised explosive devices (IEDs).
Adaptive communications automatically adjusts to conditions that degrade its performance, such as environmental conditions, or from intentional or inadvertent signals jamming.
On Wednesday's contract modification, Lockheed Martin will do the work in Cherry Hill, N.J.; Blacksburg, Va.; Fort Wayne, Ind.; and California, Md., and should be finished by October 2015.
For more information contact Lockheed Martin Advanced Technology Laboratories online at www.lockheedmartin.com/us/atl, or DARPA at www.darpa.mil.
