Synthetic aperture radar autofocus based on time-frequency transform

Phase gradient autofocus (PGA) is a robust tool for high resolution synthetic aperture radar (SAR) phase estimation and correction under the assumption that some strong scatterers with high signal-to-clutter ratio (SCR) are available within the unfocused image. In this paper, an advanced PGA algorithm is proposed, where the short-time Fourier transform (STFT) firstly is used to estimate the quadratic phase error (QPE) and the PGA is applied with a more effective filtering in the short-time Fourier domain (STFD). The proposed algorithm can relax the limitation on the scene content and achieve well-focused images without iteration. The validity of the proposed algorithm is demonstrated with the experimental results of simulation data and real radar data.