Commercial and military wireless systems, radio astronomy observatories, weather radar systems, and other applications, frequently referred to as “passive users,” need to operate in quiet electromagnetic environments with limited interference. Passive users commonly strive to observe faint signals emitted by distant non-coordinating transmitters and their frequency or time separation from other sources may not be possible as they often continuously utilize large swaths of radio spectrum to detect and monitor physical processes that have electromagnetic presence at different frequencies. To fully reap the benefits of spectrum coexistence in 5G and beyond, there is a need to employ passive-user protection approaches that strike an optimal balance between two key objectives, namely reliably protect passive users from interference and maximize spectrum access opportunities for active users. This project addresses foundational challenges in spectrum coexistence by developing a novel low-cost hardware-reduced and multi-parameter reconfigurable ultra-wideband transceiver that optimizes passive-active spectrum sharing across a broad frequency range. The portability and adaptability of this new transceiver makes it attractive for next-generation mobile wireless platforms, including a variety of autonomous ground and/or airborne platforms, cellular base-stations, unmanned airborne and satellite communication systems and specifically impact 5G, Wi-Fi and future applications of connected autonomy. Through this project the PIs propose to train undergraduate, graduate, and postdoctoral students targeting Hispanic, women, and other underrepresented groups in science, technology, engineering and mathematics (STEM) through specialized outreach efforts and curriculum development.The project brings together an interdisciplinary team of researchers with complementary expertise ranging from Radio Frequency (RF) front-end hardware design, to Physical (PHY) and Medium-Access Control (MAC) layer optimization, to artificial intelligence (AI) theory and practice. To avoid hardware constraints and lack of reconfigurability imposed by analog and hybrid beamforming architectures the project develops a novel digital ultra-wideband beamforming architecture that enables: 1) Efficient spectrum utilization and interference-free passive-active coexistence through robust space-time-frequency sensing and cross-layer optimization at the PHY and MAC layers; 2) autonomous hardware multi-parameter tunability for ultra-wideband operation across 5G bands; and 3) practical realization of low-cost, versatile hardware-reduced wireless systems through new artificial-neural-network-aided code multiplexed array front-ends. Robust spectrum sensing involves new L1-norm principal-component analysis to assess the quality (validity and completeness) of the collected/sensed spectrum data and produce high-confidence power propagation and spatial coordination maps of active spectrum users. The project leverages high-quality radio maps, autonomous sub-arraying, multi-band, and multi-parameter reconfigurability across large bandwidths and incorporates reinforcement learning strategies to address the cross-layer PHY/MAC and front-end control problem for harmonious passive-active spectrum coexistence. The complete outcome of this project is expected to be an autonomously reconfigurable hardware-reduced platform that will allow integration of ultra-wideband sensing and beamforming functionalities in small-form-factor software-defined radio platforms. This new class of low-cost, versatile, hardware-reduced wireless transceivers that integrate multiple radio chains into a practical single lightweight package is enabled by the application of neural networks to cancel both linear and non-linear components of inter-channel interference that arise during the multiplexing of multiple non-orthogonal spread signal paths.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.