Study of pattern synthesis techniques using neural networks

A comparison study among three neural network algorithms for the synthesis of array patterns is presented. The neural networks are used to estimate the array element excitations for an arbitrary pattern. The architecture of the neural networks is discussed and simulation results are presented. Two new neural networks, based on radial basis functions (RBF) and wavelets neural networks (WNN), are introduced. The proposed networks make the synthesis procedure more efficient compared to other available techniques.