ADMM with Spectral Conjugate Gradient Method for 1-Bit Compressed SAR Imaging

Abstract

This paper presents a novel approach for 1-hit compressive sensing (CS) synthetic aperture radar (SAR) imaging. The algorithm recovers SAR image from 1-bit quantized measurements. Logistic distribution is used as a consistency function between 1-bit quantized data and SAR image reconstructions. Alternating direction method of multipliers (ADMM) based approach is used for solving sparse optimization problem of 1-bit compressive SAR system. In ADMM, the alternating minimization problem due to the smooth consistency function is normally solved by standard Newton's method. But in this study, the minimization of that problem is achieved by spectral conjugate gradient method (SCGM). The effectiveness of proposed approach is shown in the simulations with synthetically generated radar data.