Neural network modelling of electromagnetic field problems

This paper discusses the application of neural network (NN) technique in the modelling of a typical electromagnetic (EM) field problem consisting of a coaxially-driven dielectric-ring resonator antenna. The NN takes into consideration the geometrical and material parameters of the EM model, and the reflection coefficient values for different frequencies in the bandwidth of interest. The NN is trained off-line using the training sets generated by the finite element method. Once an acceptable NN model is found by simulation it can be implemented in hardware in order to provide the real time behaviour requested for further integration in an EM-CAD tool.