Fast Angle Estimation of NLoS Coherent and Noncoherent Targets via Tensor Decomposition in RIS-Assisted Bistatic MIMO Radar

In this correspondence, we propose a fast tensor-based angle estimation scheme for non-line-of-sight (NLoS) targets in reconfigurable intelligent surface (RIS)-assisted bistatic multiple-input multiple-output (MIMO) radar systems. First, the received signal is formulated in a general third-order tensor form, which is applicable to configurations where both the transmitter and RIS are either uniform linear arrays (ULAs) or uniform planar arrays (UPAs). Then, after applying spatial smoothing, we perform closed-form tensor decomposition to obtain the estimated factor matrices, thereby enabling the joint estimation of one-dimensional or two-dimensional directions of departure (DODs) and directions of arrival (DOAs) for both non-coherent and coherent targets. Besides, the Cramér-Rao bound (CRB) is derived and provided as a performance benchmark for angle estimation. Compared with existing algorithms, the proposed scheme adapts well to different array configurations and channel types, achieving superior angle estimation with low complexity even in the presence of NLoS coherent targets.