Equalization of DS-UWB Systems using Genetic Algorithm with Adaptive Parameters

Abstract

We use adaptive generation along with some other parameters to investigate their effects on the performance of Genetic Algorithm (GA) in comparison to a previous work, where the output of a RAKE receiver is utilized as the input to a GA so as to reduce the inter-symbol interference (ISI) due to the frequency selectivity of UWB channels because of the very high rate of transmission. The effects of two different scaling methods and two mutation types, on the performance of a GA when used with a receiver for the equalization of the channel of a direct sequence ultra wideband (DS-UWB) wireless communications system are presented. The results show that fitness scaling has effects on GA based optimization while mutation prevents local convergence.