The linear predictive estimation in the RAKE for multipath-fading channel with white noise of general distributions

Abstract

The parameter estimation of the multipath and fading channel for a RAKE correlator is investigated. A linear predictor is proposed to get better estimation. Since the industrial interference is generally impulsive, the investigation is under a general distribution assumption for white noise. A parallel LMS adaptive algorithm is proposed to reduce the impairment of impulsive noise and decision feedback error on predictor weights adjustment. Performance analysis and computer simulations are presented.