Research On Multiuser Detection Of CDMA System

Code Division Multiple Access is a self-interfered system, where multi-accessinterference(MAI) and Near-far effect are existing as the main factors restrictingCDMA system capacity. Over the past decade, a significant amount of research hasaddressed various multiuser detection(MUD) techniques. MUD is an efficienttechnique for interference suppression that reduces the multi-access interference(MAI),solves the Near-far problem and improves system capacity. In practical applicationMUD algorithm should have good performance and low complexity, but there is noMUD algorithm which could balance well between performance and complexity so far.Therefore, it is necessary to research MUD algorithm further in detail.In this paper, several typical MUD algorithms are firstly discussed, and the formulaof system capacity is deduced, and then the influence of multi-user detection to CDMAsystem capacity is analyzed visually by the simulation. This paper gives a detailedanalysis of blind MUD algorithm emphatically, the MUD algorithms based on MOE,RLS and Kalman are discussed and compared. Thereafter, the theory deduction andconcrete simulation of linearly constrained constant modulus algorithm(LCCMA) aregiven. Finally, in order to overcome the shortcoming of LCCMA, a modified linearlyconstrained constant modulus algorithm is presented, which tracks the time-varyingchannel with adaptive step-size. The computer simulation results indicate that theperformance of the proposed algorithm is better than that of other similar algorithms,and the algorithm still work well in multi-path fading channel because of theadaptation of the step-size.Moreover, the algorithm can converge to the desired userunder discretional initial conditions, and the performance of convergence andsteady-state behavior of the algorithm are more better than that of otheralgorithms.Therfore the proposed algorithm is a practical and valuable method in thefuture.