Abstract: The blind multiuser detection algorithm of CDMA（Code Division Multiple Access） system is studied under the condition with ISI, MAI and FNP. A new BN-Kalman algorithm to multipath CDMA system is proposed. The algorithm can estimate the unknown noise statistics characteristics on-line while conducting state filtering, and an efficient iterative implementation that provides very fast convergence under this representation is presented. Simulation results show that the new algorithm has the better tracking capability under the dynamic environment and convergence ability than Kalman algorithm.

Key words: Kalman filtering, Blind Newton（BN）-Kalman filtering, Multiple Access Interference（MAI）, Inter-
Symbol Interference（ISI）

摘要： 针对ISI、MAI并存情况下码分多址系统盲自适应Kalman滤波算法检测稳定性不足的问题进行了研究，结合Newton迭代法提出了一种适于多径信道下DS-CDMA系统的盲自适应拟Newton迭代Kalman滤波MUD算法。该算法能够在进行状态滤波的同时对未知噪声的统计特性进行在线估计，确保算法能够有效收敛于期望用户提高了检测算法在动态条件下的用户跟踪能力。仿真结果表明，BN-Kalman算法具有更好的收敛性和动态性能。

关键词: 卡尔曼滤波, 拟Newton卡尔曼滤波, 多址干扰（MAI）, 码间干扰（ISI）