TY - GEN
T1 - Intelligent Array Antenna in Complex Environment Based on Deep Deterministic Policy Gradient Algorithm
AU - Wang, Tong
AU - Deng, Jinshan
AU - Cao, Kaiqi
AU - Gao, Hongwei
AU - Jin, Cheng
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - To reduce the loss of phased array antenna gain in complicated electromagnetic environment, a phase regulation algorithm based on deep reinforcement learning (DRL) is investigated in this letter. At first, we analysis and design a DRL-based interactive system between the phased array antenna and inhomogeneous electromagnetic environment. A typical policy-based DRL, namely Deep Deterministic Policy Gradient (DDPG), which has excellent fitting ability for high-dimension continuous action space, is selected to deal with the radiation pattern synthesis problem of the phased array antenna located around metal obstacles. Then, the configuration of a 2 × 8 phased array antenna and the radiation environment is given, and the training process of DDPG-based phase regulation algorithm is designed. After completing the training, the phased array antenna with DDPG-based phase regulation algorithm in complicated radiation environment is achieved, which compensates the antenna gain loss by 60% on average.
AB - To reduce the loss of phased array antenna gain in complicated electromagnetic environment, a phase regulation algorithm based on deep reinforcement learning (DRL) is investigated in this letter. At first, we analysis and design a DRL-based interactive system between the phased array antenna and inhomogeneous electromagnetic environment. A typical policy-based DRL, namely Deep Deterministic Policy Gradient (DDPG), which has excellent fitting ability for high-dimension continuous action space, is selected to deal with the radiation pattern synthesis problem of the phased array antenna located around metal obstacles. Then, the configuration of a 2 × 8 phased array antenna and the radiation environment is given, and the training process of DDPG-based phase regulation algorithm is designed. After completing the training, the phased array antenna with DDPG-based phase regulation algorithm in complicated radiation environment is achieved, which compensates the antenna gain loss by 60% on average.
UR - http://www.scopus.com/inward/record.url?scp=85190120444&partnerID=8YFLogxK
U2 - 10.1109/APCAP59480.2023.10469688
DO - 10.1109/APCAP59480.2023.10469688
M3 - Conference contribution
AN - SCOPUS:85190120444
T3 - 2023 IEEE 11th Asia-Pacific Conference on Antennas and Propagation, APCAP 2023 - Proceedings
BT - 2023 IEEE 11th Asia-Pacific Conference on Antennas and Propagation, APCAP 2023 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 11th IEEE Asia-Pacific Conference on Antennas and Propagation, APCAP 2023 - Proceedings
Y2 - 22 November 2023 through 24 November 2023
ER -