Inverse Electromagnetic Scattering for Doubly-Connected Cylinders using Convolutional Neural Networks

TL;DR

This paper presents a neural network-based method for solving the inverse electromagnetic scattering problem involving complex-shaped cylinders, enabling shape classification and boundary reconstruction with high robustness to noise.

Contribution

It introduces a divide-and-conquer framework using specialized convolutional neural networks for shape classification and boundary reconstruction in electromagnetic scattering.

Findings

Effective shape classification and boundary reconstruction demonstrated.

Robust performance under noisy measurement conditions.

High efficiency of the proposed neural network approach.

Abstract

In this work, we consider the inverse electromagnetic scattering problem for a magneto-dielectric cylinder covering an impedance cylinder of arbitrary shape. We solve it by introducing a divide-and-conquer framework using specially designed 1D multi-channel, circular-padding Convolutional Neural Networks. The solution of the direct problem provides us with the real and imaginary components of the far-field measurements representing the input data. We first classify the shape of the impedance cylinder and then reconstruct the unknown boundary curve and the impedance function. Through extensive numerical experiments, including noisy scenarios, we demonstrate the efficiency and robustness of our approach.