A neural network based adaptive correlator in DS/SS mobile communication system

Abstract

A neural network adaptive correlator (NNAC) based on the complex multilayer perceptron (CMLP) is presented for detecting the received signals in CDMA DS/SS mobile communication system and a new complex algorithm, called the complex recursive least square with backpropagated errors, is derived which provides faster convergence rate and better performance than those of the complex BP algorithm. Computer simulation results show that the NNAC has lower bit error probability than that of the traditional RAKE receiver in the presence of tone, narrowband and cochannel interference (CCI).