FOURIER-DOMAIN MULTICHANNEL AUTOFOCUS FOR SYNTHETIC APERTURE RADAR

We consider the problem of autofocus for synthetic aperture radar (SAR). We present a new autofocus algorithm, termed Fourier-domain MultiChannel Autofocus (FMCA), which is derived using a linear algebraic framework that allows the SAR image to be focused in a noniterative fashion. Motivated by the MutiChannel Autofocus (MCA) approach, the proposed autofocus algorithm also invokes an assumption on the underlying image support. However, it works with the collected Fourier data directly and is capable of accommodating, both wide-angle monostatic SAR and bistatic SAR. Previous SAR autofocus algorithms rely on the assumption that radar's range of look angles is small, and may depend upon characteristics of the SAR scene. The performance of these autofocus algorithms degrades if the assumptions are violated. The proposed algorithm has the advantage that its formulation makes no assumptions about the range of look angles or characteristics of the scene.