Rapid Three-dimensional Reconstruction for Micro-motion Target Based on the Low-rank State-space

This paper proposes a rapid three-dimensional (3D) reconstruction method for micro-motion target based on the low-rank state-space (LRSS). Firstly, considering the low-rank characteristics of the target echo data, an improved state激space algorithm based on randomized singular value decomposition (rSVD) is utilized to efficiently extract micro-range and micro-velocity information of the target. Then, the labeled multi-Bernoulli filter (LMB) algorithm accurately associates micro-range information from multiple scattering centers. Subsequently, a low-rank characteristic-based factorization 3D reconstruction algorithm is proposed for achieving 3D reconstruction of target scattering center positions. Finally, the experimental results on the real scenario verify the effectiveness and superiority of the proposed method.