Hopfield neural network and Viterbi decoding for asynchronous MC-CDMA communication systems

Abstract

Hopfield neural network (HNN) multiuser detection in conjunction with Viterbi decoding of an asynchronous multicarrier code-division multiple-access (MC-CDMA) communication system is investigated. In this scenario, the noise is characterized as being the sum of other users interference and additive white Gaussian noise (AWGN). The optimal multiuser detector for CDMA has a complexity that is exponential in the number of users. Previous studies have shown that this detector can completely remove the multiple-access interference from the received signal. Convolutional encoding with Viterbi decoding is particularly suited to a channel in which the transmitted signal is corrupted mainly by AWGN. Simulation results indicate that the HNN detector with Viterbi decoding is a good alternative to matched filter detector.