Input Parameters Optimization in Swarm DS-CDMA Multiuser Detectors

TL;DR

This paper introduces a particle swarm optimization approach for multiuser detection in DS-CDMA systems, optimizing input parameters to enhance performance in high-order modulation and diversity scenarios.

Contribution

It provides a comprehensive parameter optimization procedure for PSO in MuD problems, improving detection performance over traditional methods.

Findings

PSO-MuD outperforms conventional detectors after convergence

Performance approaches single user bound in simulations

Effective in Rayleigh flat and diversity channels

Abstract

In this paper, the uplink direct sequence code division multiple access (DS-CDMA) multiuser detection problem (MuD) is studied into heuristic perspective, named particle swarm optimization (PSO). Regarding different system improvements for future technologies, such as high-order modulation and diversity exploitation, a complete parameter optimization procedure for the PSO applied to MuD problem is provided, which represents the major contribution of this paper. Furthermore, the performance of the PSO-MuD is briefly analyzed via Monte-Carlo simulations. Simulation results show that, after convergence, the performance reached by the PSO-MuD is much better than the conventional detector, and somewhat close to the single user bound (SuB). Rayleigh flat channel is initially considered, but the results are further extend to diversity (time and spatial) channels.