Target-Before-Detect Algorithm for Passive MIMO Radar Based on Improved Particle Filter

• Published in: 2022 Global Conference on Robotics, Artificial Intelligence and Information Technology (GCRAIT) pp. 110 - 119
• Main Authors: Zhou, Yawen, Zhu, Jiawei, Song, Tingsong, Yao, Siyi, Lai, Rui, Guo, Yuning
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.07.2022
• Subjects: Filtering algorithms; MIMO radar; Object detection; Particle Filter; Particle filters; Passive Multiple-Input-Multiple-Output (MIMO) Radar; Passive radar; Simulation; Target-Before-Detect (TBD); Time-domain analysis
• DOI: 10.1109/GCRAIT55928.2022.00032

In order to improve the target detection performance of passive multiple-input-multiple-output (PMR) radar system, a Target-Before-Detect (TBD) algorithm based on improved particle filter is proposed in this paper. As a classical TBD algorithm, particle filter algorithm can realize the gain accumulation of target signal in time domain and effectively improve the performance of target detection. Although this algorithm has performed well in passive single-input-single-output (SISO) radar (PSR) systems, it cannot be directly applied to PMR systems. Therefore, an improved particle filtering algorithm based on passive MIMO radar is proposed in this paper. This method takes advantage of the PMR spatial accumulation and proposes a novel particle weight calculation method, which uses the false alarm probability distribution of detector to calculate the weight of the particle, so as to obtain the gain accumulation of the target in time and space. The simulation results show that the detection probability of the proposed algorithm can reach 90% even when the target-path signal-to-noise ratio (SNR) is low. Meanwhile, the good detection performance of the proposed algorithm is verified by comparing with the single frame detection results.