High Resolution FMCW SAR Imaging Based on Compressive Sensing Framework

This paper presents an innovative Compressive Sensing (CS) based algorithm for synthetic aperture radar (SAR) imaging, aimed at achieving high-resolution imaging. Traditional imaging algorithms often need large amounts of raw data and substantial storage capacity to generate clear target images. In this context, we propose the utilization of a Linear Frequency Modulated (LFM) signal, which offers superior range resolution while reducing complexity even under low Signal-to-Noise Ratio (SNR) conditions. Initially, our proposed work formulates the LFMCW SAR imaging model with the CS framework. Subsequently, we apply both 1D and 2D Orthogonal Matching Pursuit (OMP) methods to reconstruct the SAR images. Additionally, we extend this approach to handle the reconstruction with non-uniform sampled signals, leveraging their inherent sparsity. By adopting our proposed algorithm, we achieve clear target localization and enhanced image quality by effectively mitigating computational complexity, particularly in noisy environments. These results surpass the capabilities of existing methods documented in the literature.

I. Introduction

Synthetic Aperture Radar (SAR) is a trending imaging technology that utilizes a moving antenna along a platform to create a large aperture [1]. It is widely recognized for its ability to produce high-resolution images of stationary targets even in challenging weather conditions [2]. SAR finds diverse applications, including but not limited to oil spill detection, monitoring changes in the Earth's surface, military surveillance, flood detection [3], environmental monitoring, and drought mapping [4]. To ensure optimal performance across these various applications, it is crucial to generate high-resolution images. Image reconstruction plays a vital role in achieving this goal.