Electronic Warfare (EW) systems have been used for decades, but are now as prolific as ever. In 2021 alone, Congress provided $5.75 billion in funding for the research and development of EW systems. Their primary functions are traditionally grouped into Electronic Attack (EA), Electronic Support (ES), and Electronic Protect (EP). ES typically serves Intelligence, Surveillance, and Reconnaissance (ISR) focused missions. Due to the elevated level of efficacy for deploying machine learning (ML) models in recent years, ML is now being considered as the top contender to handle mode-agile threats. In this work, ML is used as a vehicle to extract additional information, improve accuracy, and reduce required resources for ES systems. The concept is introduced by augmenting an ultrawideband spectrum sensor with polarimetry capabilities. Then, elements and techniques for tightly coupled, ultrawideband phased arrays are explored. A fully integrated active array is presented with discrete beam direction finding functionality. Both these concepts are combined to retrofit an ultrawideband passive array with a novel compact machine learning architecture to get both azimuth and elevation information from a planar structure. Finally, the contributions from these works are synthesized to develop a reduced-complexity direction finding system with integrated receiver.

Gaeron Friedrichs received the BS degree in electrical engineering from the Virginia Polytechnic Institute and State University (Virginia Tech) in 2019 with highest honors (Summa Cum Laude). He joined the Antenna Research Group at the University of Colorado Boulder in 2019, where he is currently an NSF Graduate Research Fellow and Lockheed Martin Endowed Graduate Fellow pursuing a PhD. His research interests are machine learning, tightly coupled arrays, and direction-finding systems.