A THz SAR Autofocus Algorithm Based on Minimum-Entropy Criterion

The minimum entropy autofocusing for terahertz SAR motion compensation is proposed. Minimum entropy is a nonparametric autofocusing algorithm, which estimates the phase errors by minimizing image entropy. It works better in sparse scenes, such as a collection of point scatterers, which is often encountered in terahertz SAR processing. In this paper, the minimum entropy and Newton method are combined to compensate the platform motion errors. The experimental results show that the minimum entropy autofocusing method commonly used in microwave SAR is also applicable to the terahertz band.