The Research Of Multiuser Detection Technology Based On Intelligent Optimization Algorithms

Code Division Multiple Access technology is the mainstream standard which is adopted by the third generation mobile communications system. Because various users use spread spectrum codes collection which is not strict orthogonal, it has brought many problems such as multiple access interference and near-far effect. In that case, the communications system performance would be seriously affected and the system capacity increase would be limited. Multiuser detection is the key technology proposed to solve those problems. Optimal multiuser detection algorithm proposed previously is difficult to achieve in practice due to the exponential computation complexity. So the sub-optimal multiuser detection algorithm is more used in studies and applies.The intelligent optimization algorithm through the imitation of nature and the biological mechanism processing pattern refines biological optimization processing process as information optimization processing method. It has the global optimization performance, strong versatility and is suitable for parallel processing. This paper unifies the intelligent optimization algorithm and multiuser detection, explores easier way to achieve the sub-optimal multiuser detector design method in project. It has been a hot research topic of communication field in recent years.The major work of this paper is summarized as follows:(1) This paper summarizes the development course and the current research situation of multiuser detection, elaborates the intelligent optimization algorithm and its application feasibility in multiuser detection. On the basis of analyzing the multiuser detection's principle, it compares several kinds of typical multiuser detection algorithm's performance, characteristic and deficiency and does the computer simulation.(2) This paper analyzes the performance of genetic multiuser detection algorithm and the parameter selection influence. In view of the algorithm's deficiency, it proposes an adaptive genetic algorithm based on the Sigmoid function to improve crossover probability and mutation probability for multiuser detection. Besides these, it brings these adaptive changes of crossover probability and mutation probability into parallel genetic algorithm and then produces an adaptive parallel genetic multiuser detection algorithm. Computer simulation shows that the capacity of resisting multiple access interference and near-far effect of two algorithms have improved quite a lot.(3) On the basis of adaptive genetic algorithm, it introduces simulated annealing and changeable tabu length of tabu search algorithm, constructes two hybrid intelligent optimization algorithm and applies them to multiuser detection. Simulation showes that two new algorithms have the higher convergence rate and can obtain the close performance. The algorithm's computational complexity is below the optimal multiuser detector's as well.