Blind Multi-user Detection Technology In CDMA Mobile Communication System

Code Division Multiple Access (CDMA) is key technology for the third generation (3G) and the future mobile communication system, but it is an interference-limited system. As it is difficult for spread spectrum codes of users to simultaneously acquire excellent self-correlation characteristic and cross-correlation characteristic, and performance of traditional matched filter receivers is deficient, which could lead to Multiple Access Interference (MAI) between users. Based on traditional detection technique, the MUD technique was first proposed by Schneider in 1979, which makes full use of information from MAI caused by all users to combine with the detection to multiple users. It is not only resistant to MAI, but also effectively solves Near-far Effect and the problems of inter-symbol interference, lowers the requirement of system to power control, and increases the capacity of CDMA system. However, the existing MUD technique isn't a compromise between performance and complexity, the high complexity is still the greatest factor for MUD in practical. The blindness adaptation of MUD is an especially attractive research hotspot because it doesn't require training sequences and interferential information of other users. The main work of the paper is summarized as follows:(1) Based on the research of spread spectrum communication and synchronized CDMA system, BER performance of both DS-CDMA system and MC-CDMA system is being obtained by simulation analysis. The experimental result shows that BER of the two systems decreases with the increase of SNR.(2) Based on the research of the principle of MUD and some classical MUDs, BER performance of Conventional Detector (CD), Decorrelating Detector (DD), MMSE Detector, Successive Interference Cancellation (SIC) Detector, Parallel Interference Cancellation (PIC) Detector, Decision-Feedback Detector (DFD) and Single User Detector in different conditions is compared by simulation analysis. (3) Based on researching on blind algorithm of LMS, RLS, Kalman, the convergence of the three blind multiuser detection algorithms is analyzed by simulation. And then both SIR and Mean Error of the three algorithms are compared and analyzed by simulation in different channel environment. The experimental result shows the performances of the three algorithms: the effect of Kalman is the best, followed by RLS.