High-resolution image forming with bispectral data processing

Abstract

One of the promising methods for improvement of radio astronomy image forming systems is the method of analytical continuation of the object spatial spectrum. However, atmospheric turbulence and internal system noises restrict the real resolution of this method. In the present work, the authors propose that the analytical continuation method be supplemented by such effective noise suppression means as statistical signal processing utilizing triple correlation functions and their Fourier spectra which are called bispectra. The most important advantage of the triple correlation function is its virtual insensitivity to Gaussian noise.<>