AN ORTHOGONAL NEURAL-NETWORK FOR FUNCTION APPROXIMATION

This paper presents a new single-layer neural network which is based o n orthogonal functions. This neural network is developed to avoid the problems of traditional feedforward neural networks such as the determ ination of initial weights and the numbers of layers and processing el ements, The desired output accuracy determines the required number of processing elements, Because weights are unique, the training of the n eural network converges rapidly, An experiment in approximating typica l continuous and discrete functions is given, The results show that th e neural network has excellent performance in convergence time and app roximation error.