SSRI-NET: AN ITERATIVE NETWORK FOR SAR SUPER-RESOLTION IMAGING

SAR imaging is a widely used technique to solve the remote sensing problem. However, the image resolution of traditional imaging methods is limited by signal bandwidth. In order to solve this problem, a new iterative network is proposed to obtain SAR super-resolution images, i.e., SAR Super-Resolution Imaging Network (SSRI-Net). The proposed SSRI-Net is designed to implement SAR super-resolution imaging. In SSRI-Net, the SAR echo is transformed to the SAR reconstruction image, while an additional decoder part does the opposite operation. The SSRI-Net is composed of the unfolded Alternating Direction Method of Multipliers (ADMM), and the decoder part is formulated into a linear mapping to achieve self-supervised imaging. Numerical simulation and real-data experiment in microwave anechoic chamber demonstrate the robustness and effectiveness of the proposed method.