Radar Forward-Looking Super-Resolution Imaging Based on Joint Projection and Non-Convex Two-Step Regularization

Regularization methods have been widely applied to enhance the performance of radar forward-looking super- resolution imaging, including single constraint and joint constraints regularization. However, the ill-posedness of the antenna measurement matrix hinders imaging performance under low signal-to-noise ratio (SNR) conditions. In this paper, we proposed a new two-step regularization method to improve the imaging performance. Specifically, a projection regularization is introduced in the first step to mitigate the ill-posedness of the antenna measurement matrix by truncating and modifying the singular values to suppress the noise level. Furthermore, a non-convex constraint is utilized to improve the radar azimuth resolution based on the newly obtained convolutional model. Finally, the alternating direction method of multipliers (ADMM) is used to solve the optimization problem. Simulation results demonstrate the effectiveness of the projection-non-convex two-step regularlzation method.