Fast deconvolution method for angular super-resolution imaging based on sub-space embedding

Abstract

This study presents a fast deconvolution method based on the sub-space embedding for angular super-resolution in radar forward-looking imaging area. The ill-posed character of convolution matrix causes the difficulty to improve angular resolution and the redundancy of matrix increases computational complexity. In this study, the sub-space embedding theory is applied to reduce the redundancy of convolution matrix. Through the sketching matrix, the effective space of convolution matrix is extracted, which improves the computational efficiency and modifies ill-posed character. Simulations and experimental results demonstrate that the proposed method offers a time complexity reduction without loss of performance.

Keywords

• image resolution
• deconvolution
• computational complexity
• radar imaging
• matrix algebra
• convolution
• fast deconvolution method
• convolution matrix
• redundancy
• sub-space embedding theory
• angular super-resolution imaging
• computational complexity
• radar forward-looking imaging
• ill-posed character