Adaptive Step Size RAKE Receiver for DS/CDMA Communications Over a Rayleigh Fading Channel

Despite its simplicity, the stochastic gradient method used by the majority of adaptive algorithms, e.g. the LMS algorithm, has a slow convergence speed, thus limiting its use in mobile applications operating in a time-varying channel. A blind adaptive step size algorithm has been proposed in [3] in order to enhance the convergence speed of a stochastic gradient based adaptive algorithm in an AWGN channel. In this paper, the blind adaptive step size algorithm is incorporated into a RAKE receiver in order to harness the energies of the multipath signals in a fading channel. The performance of the proposed multipath combiner with respect to detector complexity in terms of the number of fingers involved in the detection process is investigated. Simulation results show that the proposed RAKE receiver with reduced number of fingers achieves improved frame error rate over the standard RAKE receiver whose fingers consist of the classical match filters.