Research And Performance Analysis Of Multiuser Detection Algorithms In DS-CDMA System

WCDMA,CDMA2000 and TD-SCDMA are the scenarios on the third Generation mobile communications, some key techniques are consistent in these schemes. And including : highly effective channel encoding-decoding,software radio,smart antenna,multiuser detection and interference cancellation etc. In direct-sequence(DS) CDMA system, multi-access interference is one of the important factors that affect the systematic performance and capacity. Multiuser detection suppresses the interference at the receiver. The problem that it should solve is: How demodulate the bit information reliably of some specific user or all of the users from digital bits sequence that interference effectively by joint detection or subtract the MAI from the receiving signal according to the knowledge of sequences of all users basing on traditional detection of filter bank. Thus it can improve the performance of the system and increases capacity of the system.This paper educed the necessity of multiuser detection by analysis the advantages and problems of the DS-CDMA system in existence. On the analysis and contrast based on existing typical algorithms on multiuser detection, two new MUD algorithms are presented for improving computing complexity: "Low complexity polynomial expansion closed" and "Fast equalization by using frequency domain block constant modulus". The performance of the new detectors is analyzed and researched in detail in this paper. Theoretical analysis and computer simulation results indicate that these two algorithms have shown a good performance. Among them, former polynomial expansion detection algorithm based on the classic conclusion ,Gauss—Seidel iterative method which will introduce the same time. Inverse matrix will eventually form into a Taylor series approximation. This not only makes a huge amount of computation matrix inversion have been significantly reduced, but cascade structure clear, simple realization. The latter algorithm is a "blind channel balancing" areas, from the analysis of the cost function and constraints optimization proceed. IDFT introduction to the time-domain-defined output expression updating filter coefficients derived ultimately tap "input symbols" and "constant modulus" the relations between them. The algorithm guarantee system performance (BER) on the premise that Tap the filter coefficient update has been significantly reduced complexity. These two algorithms are linear detection algorithm in a relatively poor channel environment, still be able to maintain good performance and for the "Near-Far" of the resistance.