基于改进阈值消噪算法的小波包降噪  

Signal De-noising in Wavelet Packet Based on An Improved-threshold Function

作　　者：于文新 张谦[2] 

机构地区：[1]湖南中移鼎讯通信有限公司,湖南长沙410011 [2]长沙理工大学数学与计算科学学院,湖南长沙410011

出　　处：《通信技术》2010年第6期7-9,共3页Communications Technology

摘　　要：小波变换是一种把时间和频率结合起来的时频分析方法,D.L.Dohono在小波变换的基础上提出了小波软、硬阈值消噪法,此方法简单有效,得到了广泛的重视和研究。在经典的小波软、硬阈值消噪方法的基础上,提出了一种基于小波包分析的改进方法。Matlab仿真结果表明,此法同时克服了软、硬阈值方法的缺点,而且其消噪效果比小波消噪方法要好。Wavelet transform is a time-frequency analysis method.D.L.Dohono,based on the wavelet transform,proposes the wavelet soft-and hard-threshold denoising method,this method is simple and effective,and receives much attention and study.With the two traditional wavelet de-noising methods of soft-threshold and hard-threshold strategy,an improved method based on wavelet packet analysis is proposed.The simulation with the Matlab indicates that the improved method overcomes shortcomings of the traditional nonlinear threshold strategy and is better than wavelet threshold strategy in denoising.

关 键 词：小波包分析 小波包阈值降噪 阈值 阈值函数 均方根误差 信噪比 

分 类 号：TN911[电子电信—通信与信息系统]