SAR imaging with noise waveform and low sampling rate based on sparse optimization

Abstract

In this paper, we apply sparse optimization method to synthetic aperture radar (SAR) imaging on the airborne data of a Ku-band SAR using noise waveforms. The SAR system transmits chaotic pulse waveforms at three carrier frequencies (13.7GHz, 13.9GHz and 14.1GHz). Each frequency channel has a same bandwidth of 220MHz and the total bandwidth covered by the three channels is 620MHz. According to the compressed sensing (CS) theory and the randomness property of noise signal, we can uniformly down sample the echo data with a rate below the Nyquist sampling rate. Effects of low-rate sampling on noise radar imaging are discussed with simulated 1-D data presented. The reconstruction of SAR image from low-rate samples is based on our recently proposed maximum a posterior (MAP) estimation method, which is developed from the sparse optimization techniques in CS. Experimental results are presented to show the effectiveness of our algorithm.