An optimal method for phase error estimation in SAR imagery

A new method of phase error estimation that utilizes the weighted least square (WLS) algorithm is presented for SAR autofocus applications. The method has the capability to estimate all kinds of phase errors, no matter they are of low-order, high-order or random. The WLS estimation is optimal in the sense that it has the minimum variance of the estimation error. Excellent results have been obtained in autofocusing experiments.