并联滤波的常数模与判决引导融合盲均衡算法  

The fusion blind equalization algorithm based on constant modulus algorithm and decision direction with parallel filter

作　　者：肖瑛[1] 董玉华[1] 

机构地区：[1]大连民族学院信息与通信工程学院,大连116600

出　　处：《声学学报》2012年第5期495-500,共6页Acta Acustica

基　　金：中央高校基本科研业务费专项资金资助(DC10040103;DC12010218);辽宁省教育厅科学技术研究项目(2010046)

摘　　要：针对常数模和判决引导双模式盲均衡算法切换时机选择困难问题,提出了一种并联滤波的双模式融合盲均衡算法。算法以并联滤波器作为盲均衡器,两路子滤波器分别以常数模算法准则和判决引导算法准则进行更新,通过加权因子实现两种算法模式自适应切换,完成两种算法的融合处理,加权因子依据归一化均方误差进行调整。为防止信道突发干扰,定义了归一化均方误差信息熵增量,以信息熵增量为判据适时实现均衡器和加权因子重置,使算法在信道突发干扰条件下能实现自适应跟踪。计算机仿真和水池试验处理结果表明:该算法有效地结合了常数模和判决引导算法的优点,具有较好的均衡性能。To solve the problem of switching time of dual-mode blind equalization algorithm for CMA （Constant Modulus Algorithm） and DD （Decision Direction）, a dual-mode blind equalization algorithm with parallel filter is proposed. The weights of subfilters update with CMA and DD algorithm respectively, the weighting factor is adjusted according to normalized mean square error with the iterative times that controls the dual-mode algorithm adaptive switch, which implement the fusion process. To prevent the channel sudden interference, Information entropy increment of normalized mean square error is defined as criterion for resetting of the weighting factor and the weights of filter, thus can ensure the stability of the algorithm under the channel sudden interference. Simulation analysis and the pool experiments show that the proposed algorithm can make full use of the advantages of the two algorithms and has better performance.

关 键 词：判决引导算法 盲均衡算法 常数模算法 并联滤波 自适应跟踪 加权因子 盲均衡器 均方误差 

分 类 号：TN911.5[电子电信—通信与信息系统]