Genetic Algorithm based multiuser detection in DS-CDMA: A comparative analysis

The efficiency and the strength of Direct Sequence Code Division Multiple Access (DS-CDMA) systems is always effected by Multiple Access Interference (MAI). Many MUD (Multiuser detection) techniques have been introduced to handle this MAI problem. All these MUD techniques have certain advantages and disadvantages over each other. ML Detector has best implications among all MUDs. But with the increase of the number of users the Computationally complexity of the ML detector increases exponentially. In this research work the Problem of computational complexity is handled employing two Evolutionary Techniques. Theses two evolutionary techniques are Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). Simulation and comparison shows that these two Evolutionary Techniques GA and PSO based MUD works better as compare to other detectors. Both techniques remarkably reduce the computational complexity of optimum ML detector. Results shows that that in the beginning PSO based MUD in DS-CDMA has edge over GA-MUD. When number of users are increased more and more GA-MUD appears to be more robust than PSO-MUD.