Comparison between the performance of conventional and selective RAKE receiver schemes

In this paper, we determine the effects of different diversity schemes on the behavior of a direct-sequence spread-spectrum RAKE receiver that operates with random sequences in a multipath fading channel with an arbitrary power-delay profile. Two different receiver architectures, the former employing the L instantaneously strongest paths out of the N/sub r/ resolved ones and the latter simply the L strongest on average paths out of the received N/sub r/, are confronted in terms of symbol-error probability, when an M-level phase-shift keying system is assumed. The analytical approach devised for assessing such a comparison gives the symbol-error probability in terms of simple elementary integrals and immediately evidences the influence on system behavior of the number of resolved paths, bandwidth occupancy of the signal- and power-delay profiles with a remarkably moderate computational effort.