PSO-based multiuser detectors for high-order modulation DS/CDMA systems under spatial and multipath diversities

In this paper, the uplink direct-sequence code-division multiple access DS-CDMA multiuser detection MuD problem is studied from a heuristic perspective. The heuristic particle swarm optimisation PSO approach is due to the NP-complexity of the MuD optimisation problem. The challenge is to solve the associated hard complexity problem in a polynomial time. A complete parameter optimisation procedure for PSO applied to the MuD problem is carried out herein under different system configurations, such as high-order modulation and diversity exploitation. The PSO-MuD performance is analysed through Monte Carlo simulation MCS method. MCS results show that, after convergence, the performance achieved by the PSO-MuD is much better than that shown by the conventional detector, and somewhat closer to that given by the single-user bound. The flat Rayleigh channel is initially considered, but the results are further extended to the space-time wireless channels.