Effects of Multiple Scattering on Resolution of Full-Wave Inverse-Scattering Solver

Abstract

Inverse-scattering techniques have wide applications in quantitatively determining either physical or geometrical properties in various fields. Nevertheless, inverse-scattering problem is very challenging due to its ill-posed and nonlinear properties. In this paper, we study and analyze the effects of multiple scattering on the resolution of full-wave reconstructions in inverse-scattering problems, where theoretical analyses and numerical simulations have been conducted. In the numerical experiments, twofold subspace-based optimization method is used to reconstruct permittivities of scatterers from scattered fields for weak, moderate, and strong scatterers under different noise levels. It is found that, for weak scatterers, inversion is unstable, especially when noise is high. For strong scatterers, one can hardly obtain a solution that is close to the exact one even when the noise level is low. Furthermore, for moderate scatterers, the inversion is more stable than weak scatterers and one is able to find a solution that is close to the exact one with the optimization method.