Fast and Accurate Model to Determine the Resonant Characteristics of Elliptical Microstrip Patch Antenna

In this work, artificial neural networks are used in the design of a microstrip antenna of elliptical geometry. We propose two neural models. In the first one, the outputs of the problem are the patch size and the resonant frequency for even modes. In the reverse side of the problem, the resonant frequencies for both even and odd modes are obtained in terms of the effective semi-major axis, eccentricity of elliptical patch, the thickness and the effective relative permittivity of the dielectric substrate. To enhance the accuracy of the two models, measurements have been included in the training stage. The two proposed models are very adequate for a fast and accurate estimation of the characteristics of the elliptical microstrip antenna.