Synthetic aperture radar system design for random field classification

Abstract

An optimum design study is carried out for synthetic aperture radar systems intended for classifying randomly reflecting areas (such as agricultural fields) characterized by a reflectivity density spectral density. The problem solution is obtained, neglecting interfield interference and assuming areas of known configuration and location, as well as a certain Gaussian signal field property. The optimum processor is nonlinear, but includes conventional matched filter processing. A set of summary design curves is plotted, and is applied to the design of a satellite synthetic aperture radar system.