Improved MOCO Approach Based on 2-D Taylor Expansion for Spatially-Variant Errors in High-Resolution Airborne SAR Imagery

Abstract

Motion compensation (MOCO) is important for qualities of airborne synthetic aperture radar (SAR) image. However, in the case of high-resolution, the spatially-variant components of motion errors are significant and cannot be compensated by traditional MOCO algorithms. To address this problem, an improved compensation approach based on two dimensions (2-D) Taylor series expansion for high resolution airborne SAR imaging in the condition of large spatially-variant motion errors is proposed. First of all, an accurate slant range history is expressed in the geometric model, which is developed by the motion parameters measured by inertial navigation system (INS). Secondly, 2-D Taylor series expansion is performed to decouple the errors so that the spatially-variant components can be removed completely by means of the interpolation. Unlike previous MOCO methods, our approach can greatly decrease the spatially-variant effects on airborne SAR imaging in the condition of large motion errors. Finally, real data experimental results demonstrate the effectiveness of the proposed approach.