TY - GEN
T1 - Near-Field 3D Phase Compensation Method for Distributed Radar System based on Parallel Implementation Architecture
AU - Zhang, Zhengyi
AU - Su, Jia
AU - Tao, Mingliang
AU - Fan, Yifei
AU - Wang, Ling
N1 - Publisher Copyright:
© 2023 International Union of Radio Science.
PY - 2023
Y1 - 2023
N2 - Distributed coherent radar system is emerging to improve the radar target detection performance. However, due to the sparse arrangement of distributed radar nodes, the traditional far-field signal processing methods may fail in this scenario. To realize the near-field coherent accumulation, a 3-dimensional (3-D) phase compensation method via GPU device is proposed in this paper. Firstly, according to the geometry of the target and distributed nodes, the near-field target detection model is constructed. Then, the near-field matching function is designed to compensate for the phase difference via 3-D spatial grid search. Considering the high computational complexity of the 3-D search, Graphics Processing Unit (GPU) devices are used to realize the parallel acceleration of the proposed method. The experiment results show that the computational speed of near-field target detection is about 2000 times faster than that of the CPU serial method.
AB - Distributed coherent radar system is emerging to improve the radar target detection performance. However, due to the sparse arrangement of distributed radar nodes, the traditional far-field signal processing methods may fail in this scenario. To realize the near-field coherent accumulation, a 3-dimensional (3-D) phase compensation method via GPU device is proposed in this paper. Firstly, according to the geometry of the target and distributed nodes, the near-field target detection model is constructed. Then, the near-field matching function is designed to compensate for the phase difference via 3-D spatial grid search. Considering the high computational complexity of the 3-D search, Graphics Processing Unit (GPU) devices are used to realize the parallel acceleration of the proposed method. The experiment results show that the computational speed of near-field target detection is about 2000 times faster than that of the CPU serial method.
UR - http://www.scopus.com/inward/record.url?scp=85175208257&partnerID=8YFLogxK
U2 - 10.23919/URSIGASS57860.2023.10265435
DO - 10.23919/URSIGASS57860.2023.10265435
M3 - 会议稿件
AN - SCOPUS:85175208257
T3 - 2023 35th General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2023
BT - 2023 35th General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2023
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 35th General Assembly and Scientific Symposium of the International Union of Radio Science, URSI GASS 2023
Y2 - 19 August 2023 through 26 August 2023
ER -