A neural network approach to the scattering from a spherical shell with a circular aperture

This paper presents a neural network approach to estimate the backscattering cross section for the scattering by a spherical shell with a circular aperture due to a normal incidence field using the radial basis function (RBF) network. The main idea of the proposed approach is to train the RBF network to model the nonlinear relationship between the electrical radius, half aperture angle, and the backscattering cross section using experimental data. Once the network is trained, it can be used to predict the backscattering cross section of a spherical shell with electrical radius and half aperture angle different from those used for training.