Single aperture multiple beams of array antenna using Radial Basis Function Neural Network

In the present work, an optimized approach is described in predicting the current excitations of a planar array geometry for obtaining multiple beam widths in a single aperture using Radial Basis Function Neural Network (RBFNN). The approach utilizes current distributions of uniform, binomial and triangular forms for a 5×5 planar array. The radiation pattern values are given as input to the neural network. The output of the neural network is current excitations of the planar antenna array elements. RBFNN is initially trained with the input-output data pairs and tested and explored for the estimation of current excitations. The network showed a high success rate.