Design of metamaterials using artificial neural networks

This paper presents an analysis of the resonant characteristics of a composite medium based on a periodic array of interspaced conducting nonmagnetic split ring resonators (SRR) and continuous thin wires (TW). The medium exhibits simultaneously negative values of effective permeability (μ_eff) and permittivity (ϵ_eff) within a microwave frequency band, characterizing a metamaterial. An analysis using Artificial Neural Networks (ANN) is built to obtain the permeability and permittivity as a function of resonant frequencies for a given geometrical dimensions of the metamaterial, for an optimization of the development medium.