Robust Phaseless Multistatic Synthetic Aperture Radar Imaging

We present a robust phaseless imaging approach for multistatic synthetic aperture radar (SAR) in the presence of additive noise and outliers. Phaseless SAR is a novel imaging modality that offers advantages in reduced hardware complexity, operability at high frequencies, and robustness to phase errors. Additionally, robustness to additive noise and outliers is a desirable feature of the reconstruction algorithms in practical remote sensing applications. In this paper, we utilize stochastic transmitted waveforms and spatial diversity in a multistatic configuration to enable phaseless SAR. We use reverse Kullback-Leibler divergence (RKLD) as a loss function within the Wirtinger Flow (WF) framework as a robust reconstruction algorithm in the presence of noise and outliers. The RKLDbased loss function does not assume any statistical distribution of noise. Additionally, the logarithmic processing of data by RKLD suppresses large magnitude outliers. Our numerical simulations demonstrate the superior reconstruction performance of RKLDWF at low signal-to-noise ratios with respect to ambient noise, and in the presence of sparse outliers.