2-D target imaging of spotlight SAR based on higher-order statistics

This paper analyzes the imaging algorithm of spotlight SAR. It is shown that the spotlight SAR imaging can be viewed as 2D harmonic retrieval when the target is mainly composed of a small number of distinct point scatterers. A new approach based on higher-order statistics for radar imaging is presented to reduce the influence of Gaussian colored noise. A new method for harmonics pairing is also presented. Compared with the Fourier transform and algorithms based on autocorrelation, such as MUSIC, the new method can effectively reduce the influence of Gaussian colored noise.