An Iterative Method With Variable Window Length STFT for Compensating Vibration Phase Error in ISAL Imaging of Satellites

Spaceborne inverse synthetic aperture ladar (ISAL) is an extension of inverse synthetic aperture radar (ISAR) in laser band. It can provide ultrahigh-resolution distance imaging of noncooperative satellites. However, the image quality will be severely degraded by radial and angular vibration phase errors, which are common in satellites. This article proposes an iterative method to jointly compensate for both types of vibration phase errors. Unlike traditional methods relying on prominent points on the target, our approach identifies prominent regions to reduce noise impact, enhancing ISAL imaging suitability. We use short-time Fourier transform (STFT) to generate time-frequency (TF) distributions of these regions. Doppler centroid tracking (DCT) yields instantaneous Doppler curves, independent of strong scattering points. Phase error estimations are derived using a weighted least-squares algorithm. An iteration process is designed to improve the estimation accuracy by adjusting the STFT window length matching with the residual phase errors during iterations. Extensive experiments on simulation and real data confirm the effectiveness and noise robustness of our proposed method.