Efficient Prediction of the EM Response of Reflectarray Antenna Elements by an Advanced Statistical Learning Method

Abstract

An innovative strategy based on an advanced statistical learning method is introduced to efficiently and accurately predict the electromagnetic response of complex-shaped reflectarray elements. The computation of the scattering coefficients of periodic arrangements, characterized by an arbitrary number of degrees of freedom, is first recast as a vectorial regression problem and then solved with a learning-by-example strategy exploiting the ordinary kriging paradigm. A set of representative numerical experiments dealing with different element geometries is presented to assess the accuracy, the computational efficiency, and the flexibility of the proposed technique also in comparison with state-of-the-art machine learning methods.