A neural network for direction of arrival estimation under coherent multiple waves

This paper addresses the problems for the DOA estimation of narrowband emitter signals impinging on an array of sensors by the modified Hopfield neural network. We show that this network provides good resolution capability about closely located sources without any complicated computation of the matrix given from array response vectors. We also discuss the principle of the DOA estimation by the covariance fit estimator. In both cases the DOAs estimated through computer simulation are compared. Consequently we show that the operation of the neural network has an analogy with the covariance fit estimator in estimating the DOA.