Swarm, genetic and evolutionary programming algorithms applied to multiuser detection

Fernando Ciriaco, Leonardo Dagui de Oliveira, Taufik Abrão, Paul Jean Etienne Jeszensky


In this paper, the particles swarm optimization technique, recently published in the literature, and applied to Direct Sequence/Code Division Multiple Access systems (DS/CDMA) with multiuser detection (MuD) is analyzed, evaluated and compared. The Swarm algorithm efficiency when applied to the DS-CDMA multiuser detection (Swarm-MuD) is compared through the tradeoff performance versus computational complexity, being the complexity expressed in terms of the number of necessary operations in order to reach the performance obtained through the optimum detector or the Maximum Likelihood detector (ML). The comparison is accomplished among the genetic algorithm, evolutionary programming with cloning and Swarm algorithm under the same simulation basis. Additionally, it is proposed an heuristics-MuD complexity analysis through the number of computational operations. Finally, an analysis is carried out for the input parameters of the Swarm algorithm in the attempt to find the optimum parameters (or almost-optimum) for the algorithm applied to the MuD problem.


Multiuser detection; Genetic algorithm; Evolutionary programming; Particle.

DOI: http://dx.doi.org/10.5433/1679-0375.2005v26n2p195

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.