Comparative Study of the Goldfarb Iterative and the Genetic Algorithm Methods for Solving Inverse Problems

摘要

In this paper, a comparative analysis of various approaches to the formulation and solution of inverse scattering problems is given. Problems for sensing with the aid of electromagnetic fields are considered. The informative parameters can be both the frequency of a monochromatic electromagnetic wave and the parameters characterizing its polarization state, and also the angle of the wave incidence for plane-layered media. The main attention is given to the comparative analysis of two methods - the Goldfarb method and the genetic algorithm. For carrying out computational experiments, a software is developed that allows to choose a method and thereby optimize the rate of convergence of the iterative process. The results of computational experiments are presented.