Enhancing motion compensation in spaceborne SAR imaging

Abstract

Synthetic Aperture Radar (SAR) is a widely utilized remote sensing technology, offering robust operational efficiency under all weather conditions and independent of daylight. Ideally, the SAR platform maintains a linear trajectory at a constant altitude and velocity. However, this idealization is compromised for spaceborne SAR systems, such as those in low Earth orbit (LEO), due to the satellite’s elliptical orbit, which introduces motion errors that degrade image focusing quality. This paper presents a novel approach to enhance first-order motion compensation (MOCO) by addressing the motion errors caused by elliptical orbital dynamics and perturbations. The proposed methodology involves applying three distinct fitting techniques to the invariant range error, a critical parameter in first-order MOCO, and optimizing phase gradients to determine the optimal coefficients for improving image quality metrics. Real-raw SAR data from the Sentinel-1 Level-0 dataset is processed to validate the proposed techniques, and the results are benchmarked against the corresponding Sentinel-1 Level-1 Single Look Complex (SLC) image. The validation is conducted through two approaches: first, image quality assessment using sharpness, contrast, and entropy metrics; and second, quantitative evaluation of azimuth-integrated sidelobe ratio (AISLR), azimuth peak sidelobe ratio (APSLR), and impulse response width (IRW) at two prominent reflective points. The findings indicate a marked enhancement in the image quality parameters, demonstrating the efficacy of the proposed motion compensation and optimization framework.