Joint Sparse Super-Resolution Method for Azimuth-Range-Velocity Estimation on Array Radar

This article addresses the sparse super-resolution (SSR) problem for azimuth-range-velocity estimation on array radar. Existing SSR methods based on compressed sensing achieve either one-dimensional or two-dimensional SSR for azimuth-range-velocity estimation. However, most methods perform either two-dimensional joint SSR along a specific dimension or single-dimensional SSR sequentially across different dimensions, rather than combining all three dimensions simultaneously, which limits target energy accumulation and affects reconstruction performance. To address the above challenges, we first design a range dimension steering vector based on the signal model and develop a three-dimensional (3D) joint steering vector by combining azimuth and velocity dimension steering vectors. Then, we establish a 3D joint sparse signal model using the 3D joint steering vector and solve it with a greedy method to accurately estimate azimuth, range, and velocity. Various simulation experiments are designed to proof the efficacy of the proposed methods, achieving accurate reconstruction even with fewer array elements, low signal bandwidth, and limited pulses.