Estimation of propagation structure by means of Hopfield neural network
T. Maruyama, Yoshihiko Kuwahara · 2002
In order to develop the spatial-temporal equalization technique for 3G/4G mobile communication systems it is necessary to understand the indoor or outdoor propagation structure. It has been proposed to apply an artificial neural network (ANN) to DOA estimation (see El Zooghby, A.H. et al., IEEE Transactions on Antennas and Propagation, vol.45, no.11, p.1611-17, 1997; Goto, K. et al., T. IEE Japan, vol.118-C, no.3, p.442-7, 1998). The purposes are to direct narrow beams to the terminals after detecting their directions. We propose an algorithm using an ANN (Hopfield neural network) to estimate coupled DOA and TOA of multiple incident waves.