Multiplicative regularized Gauss-Newton approach for three-dimensional microwave imaging

Abstract

In this work we applied the so-called multiplicative regularized Gauss-Newton inversion approach for solving the three-dimensional electromagnetic microwave inverse problem. This inversion approach adjusts automatically regularization weight and when it is combined with the total variation type regularization approach, they provide an excellent edge-preserving characteristic. We demonstrate the efficiency and robustness of this approach by using a simple synthetic test case and an experimental data from a biomedical object.