Multi-user Detection In The Dynamic Environment Under Study

The Multi-user Detection (MUD) is the key technique for Code-Division Multiple-Access (CDMA) mobile communication system. In classic MUD theory, it detects single user's data using all the users' signal data which cause the Multiple-access interference based on traditional detection technology, so the detection has good performance for anti-interference. Meanwhile, the capacity of communication system has been greatly enhanced for using up-link frequency spectrum resource effectively.And then, many studies focus on how to enhance the capacity of communication system. In recent years, people realized that it's very important to know the number of active users in system for MUD. For conventional receivers, this dearth of information does not affect performance because we can assume that all users are active in the system. But for the random access communication system that's a real system, knowing this information will be help the system to promptly process requests and efficiently allocate channels. In such a way, system capacity can be increased.In this thesis, by establishing a dynamic model of the users in system, the Particle Swarm Optimization (PSO) strategy was applied to the search process of the Maximum A Posteriori Probability (MAP) MUD based on the fact that MAP-MUD can be viewed as a combinatorial optimization problem. According to this idea, a multi-user detectors based on Standard-PSO strategy was proposed, numerical simulation results have shown that it's a good performance. Considering the SPSO is easy to get into the part best solution, other MUDs based on PSO strategy is proposed. Simulation indicates that the performance of these detectors can be quite close to MAP-MUD, while they have perfect performance on computational complexity which is crucial for practical application. They will also provide a new thinking for implementing MUD which is in dynamic environment in CDMA communications systems.