The University of Central Florida Board of Trustees received a $500,000 Project Grant award from the National Science Foundation Division of Electrical, Communications and Cyber Systems under the NSF Engineering program (CFDA 47.041) for the period of September 1, 2022 through August 31, 2025.
The award will support development of an AI-assisted reconfigurable dual-input load-modulation transmitter array to enhance the energy and spectrum efficiencies of massive MIMO communications systems. A novel dual-input hybrid load modulated balanced amplifier architecture will be proposed and optimized through reinforcement learning to dynamically counteract antenna impedance mismatch and temperature variations during MIMO operation. A reinforcement learning framework based on nonstationary Markov decision processes will be formulated and implemented using reconfigurable FPGA technology for real-time low-latency optimization. The AI-assisted operation and multiband/multistandard capabilities will be extended from individual power amplifiers to the massive MIMO array through unique design of a wideband fractal-shaped antenna array.