DS/CDMA系统中基于自适应并行次梯度投影的多址干扰抑制算法  

MAI suppression algorithm based on adaptive parallel subgradient projection in DS/CDMA systems

作　　者：魏昕[1] 赵力[1] 邹采荣[1] 王青云[1] 

机构地区：[1]东南大学信息科学与工程学院,南京210096

出　　处：《东南大学学报（自然科学版）》2009年第5期873-877,共5页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(60872073);江苏省自然科学基金资助项目(BK2008291);教育部博士点基金资助项目(20050286001)

摘　　要：为了抑制直接序列码分多址(DS/CDMA)系统中的多址干扰,提出了一种基于自适应并行次梯度投影的多址干扰抑制算法.首先,根据接收端信号模型建立包含最优干扰抑制滤波器系数的随机属性凸集,并运用并行次梯度投影的思想将对该凸集的投影转化为对多个闭合半平面的投影.然后,分析了影响系统收敛速度和稳定状态下收敛性能的重要参数——膨胀系数,设计了在不同迭代阶段下自适应调节膨胀系数的机制.最后,将更新后的干扰抑制滤波器系数矢量投影到限定集合上.理论分析和仿真结果表明,该算法具有快速收敛性和稳定的干扰抑制性能,在不同的噪声强度下均具有较低的误码率和较高的鲁棒性;与同类算法相比,该算法在计算复杂度上具有一定的优势.To suppress multiple access interference （MAI） in direct sequence/code division multiple access （DS/CDMA） systems, an algorithm based on the adaptive parallel subgradient projection is proposed. First, a stochastic property convex set containing the optimal interference suppression filter coefficients is established based on the receiving signal model. The projection onto the convex sets is converted to that onto multiple closed half-spaces by the parallel subgradient projection. Then, the inflation parameters which influence the system convergence speed and the convergence performance in steady state are analyzed, and the mechanism which adaptively adjusts inflation parameters in different iterations is designed. Finally, the updated interference suppression filter coefficients vector is projected onto a constrained set. The theoretical analysis and the simulation results show that this algorithm has fast convergence, steady interference suppression performance, and exhibits lower bit error rate and higher robustness in different noise intensities. Besides, it has lower computation complexity than other algorithms.

关 键 词：DS/CDMA系统 自适应并行次梯度投影 多址干扰抑制 

分 类 号：TN914.4[电子电信—通信与信息系统]