Tabu Search Based Multi-user Detection In CDMA

Code-Division Multiple Access (CDMA) is known as the main technique of the third generation and future mobile communication systems. Regardless of its benefits on large capacity, soft switch, high-quality tone and excellent secrecy performance, CDMA mobile communication systems are interference-limited systems. When multi-users share the transmission channel, the communication performance of CDMA systems decay for the multiple access interference (MAI). An efficient method suppressing MAI is multi-user detection (MUD) which views MAI as useful resource and makes full use of the relationship between users to increase the detection performance. Thus, the MUD is one of the key techniques in future CDMA communication systems.Computational intelligence (CI) is a novel methodology for information processing, which was proposed and studied widely in recent years. Tabu search is a computational intelligence method for global optimizing, which tries to obtain the global optimum by its memory structure and tabu strategies. Another CI method is swarm optimizing, which inclines to the global optimum by the cooperation of particles in the swarm.Since the problem of MUD can be transformed into a combinatorial optimization problem, the strategies of CI for optimization can be applied in MUD. This thesis focuses its work on tabu-search based multi-user detection. An adaptive tabu search algorithm and new techniques based on swarm intelligence and tabu search were proposed and applied in MUD.