BAE to integrate machine-learning technology into RF platform

9 July 2019 (Last Updated July 9th, 2019 14:43)

BAE Systems has secured a contract from the US Defense Advanced Research Projects Agency (DARPA) to integrate machine-learning (ML) technology into platforms that decipher radio frequency (RF) signals.

BAE to integrate machine-learning technology into RF platform
CHIMERA provides a reconfigurable hardware platform for ML algorithm developers to make sense of RF signals in the evolving dense spectrum environments. Credit: BAE Systems.

BAE Systems has secured a contract from the US Defense Advanced Research Projects Agency (DARPA) to integrate machine-learning (ML) technology into platforms that decipher radio frequency (RF) signals.

The Controllable Hardware Integration for Machine-learning Enabled Real-time Adaptivity (CHIMERA) solution will provide a reconfigurable hardware platform that will allow ML algorithm developers to decipher RF signals.

CHIMERA will prove to be an effective solution in increasingly crowded electromagnetic spectrum environments, BAE Systems noted.

The hardware platform provided by the solution will ensure greater automated situational awareness to commercial or military users.

DARPA awarded a contract worth up to $4.7m under the Radio Frequency Machine Learning Systems (RFMLS) programme, which aims to facilitate the application of data-driven ML to the RF spectrum domain.

This is the second contract secured by BAE Systems under the RFMLS programme. The previous contract was for the development of data-driven ML algorithms.

RFMLS seeks the development of a hardware solution with control surfaces to enable improved discrimination of signals.

The company will be responsible for hardware delivery, integration and demonstration support.

BAE Systems command, control, communications, computers, intelligence, surveillance, and reconnaissance (C4ISR) systems vice-president and general manager Dave Logan said: “CHIMERA brings the flexibility of a software solution to hardware. Machine-learning is on the verge of revolutionising signals intelligence technology, just as it has in other industries.”

The CHIMERA setup is designed to adapt the hardware’s RF configuration in real time for optimised mission performance.

The presence of multiple control surfaces allows for on-the-fly performance trade-offs.

“Machine-learning is on the verge of revolutionising signals intelligence technology, just as it has in other industries.”

In addition, open architecture interfaces support third-party algorithm development.

This makes the CHIMERA system easier to upgrade upon deployment.

The company added that the hardware platform can support other RF functions such as communications, radar and electronic warfare.

BAE Systems will perform contracted work at its US facilities in Hudson and Merrimack, New Hampshire, and Dallas, Texas.

SIGINT systems developed by the company are used to gather information from adversaries’ electronic signals to secure actionable intelligence.