With Modern Optimization Algorithm Cdma Multiuser Detector

Code Division Multiple Access (CDMA) communications systems are interference limitation systems. Multiple-access interference not only affects the performance of the systems,but also decreases the capacity of the systems. Due to cancellation of multiple-access interference, multiuser detection (MUD) increases the performance and capacity.Multiuser detection is a nondeterministic polynomial hard combinatorial optimum problem.So in this thesis, three modernistic optimum algorithms for MUD in CDMA communications systems were proposed.They are decorrelator-based paiallel genetic algorithms(PGA), PGA-based Lagrange neural network multiuser detection, decorrelator-based tabu search algorithms.Theoretical analyses and numerical simulation results have shown that the three kinds of MUD are feasible and effective.They provide a new way for implementing sub-optimal MUD in CDMA communications systems.