The application of radial basis function networks to the reconstruction ofdirect-sequence spread spectrum signals in multipath

This paper addresses the problem of reconstructing a direct-sequence spread spectrum signal in multipath. Specifically, a new algorithm based on radia l basis functions is developed for blind channel estimation and data recove ry. Simulation results are presented to demonstrate the capability of the p roposed algorithm to identify the channel and restore the spread spectrum s ignal without training samples. Results obtained show that significant impr ovement in bit-error-rate performance can be achieved by using a spread spe ctrum receiver that employs the proposed reconstruction algorithm.