RF Machine Learning Systems (RFMLS)

Abstract

The RF Machine Learning Systems (RFMLS) program will address the performance limitations of conventional radio frequency (RF) systems such as radar, signals intelligence, electronic warfare, or communications. Currently, the capabilities of these systems are fixed at the time of design and limited by their designer's vision. Conversely, a generic RFMLS system would learn how to reconfigure its circuits and processing to meet the requirements of a desired application in a specific environment. The relevant RF features are hand crafted and human specified today, and would instead be learned through machine learning algorithms applied within the RF system itself. The RFMLS system would later learn to adapt to changing conditions and requirements, making for a much more robust RF system solution. This flexibility should reduce the time and cost of continually re-designing and upgrading new systems and extend RF system performance beyond the limits of human designers. RMFLS exploits recent advancements in machine learning that have not previously been applied to RF systems.

Document Details

Document Type
Accomplishment
Publication Date
Oct 01, 2018
Source ID
5d34bfaf89e05f57d8623994ac271243

Tags

Fields of Study

  • Computer science

Readers

  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • AI & ML - Neural Networks
  • Microelectronics

Related Documents