一种高精度的非平稳随机信号实时建模算法研究  

Research About An High Accuracy Algorithm of Non-stationary Random Signals Timely Modeling

作　　者：王雪梅[1] 倪文波[2] 李芾[2] 

机构地区：[1]西南交通大学机械工程学院,四川成都610031 [2]西南交通大学牵引动力国家重点实验室,四川成都610031

出　　处：《信号处理》2006年第2期198-202,共5页Journal of Signal Processing

基　　金：教育部博士点基金项目(20020613016)

摘　　要：针对传统的非平稳随机信号时变参数筹分模型算法的不足,论文把非平稳随机信号时变参数差分模型与一种带自适应遗忘因子的RLS算法结合起来,形成一种有效的非平稳随机信号实时建模算法,仿真分析表明该算法不仅能获得快的收敛速度而且能获得高的建模精度。In view of the disadvantages of traditional time-varying parameters differential modeling algorithm about non-stationary random signals, an effective modeling algorithm about non-stationary random signals is put forward, which combines the time -varying parameters differential model and a RLS algorithm with adaptive forgetting factors together. Simulations show that this algorithm not only can acquire quick rate of convergence, but also can obtain high modeling accuracy.

关 键 词：非平稳信号 实时建模 自适应遗忘因子 

分 类 号：TN911.23[电子电信—通信与信息系统] TP872[电子电信—信息与通信工程]