Target Localization Based on Multistatic Mimo Radar via Double Coupled Canonical Polyadic Decomposition

This paper considers target localization with a multistatic MIMO radar system of multiple transmit arrays and multiple receive arrays. We formulate the matched filtered output data into tensors that admit the double coupled canonical polyadic decomposition (DC-CPD) model, which efficiently characterizes the coupling between receive arrays and that between transmit arrays, and multilinear structure in each tensor. We propose a novel algebraic DC-CPD algorithm based on coupled rank-1 detection mapping, and provides analysis into the uniqueness conditions. A post-processing approach is also introduced to calculate and fuse the DOD and DOA information from the DC-CPD results to finally obtain the target locations. Simulations are provided to illustrate the merits of proposed method over CPD and C-CPD methods, with regards to accuracy and identifiability.