摘要
通过HHT理论将信号分解成若干IMF分量,直接滤除含有噪声成分的分量进行信号的降噪处理出现不同程度的信号失真。文中提出通过对每个IMF分量进行频谱分析,结合噪声信号的特征,对半数的高频IMF分量单独进行小波阈值滤波处理后重组信号,进而完成对原始信号的滤波处理,在提高了信噪比的同时避免了有效信息的丢失。
A signal can be decomposed into several members of IMF components by the HHT theory, filtering out the IMF components containing noise components directly maybe appear different degree of signal distortion.In this paper,through the spectrum analysis of each IMF and characteristics of the noise signal,about half of the high frequency IMF component was carried on the wavelet threshold de-noise processing,and then recombined the IMF to complete the original signal filter processing.This method improves the signal-to-noise ratio and avoids the losing of effective information.
出处
《仪表技术与传感器》
CSCD
北大核心
2015年第5期84-85,89,共3页
Instrument Technique and Sensor
基金
辽宁省科学事业公益研究基金项目(2012004002)
辽宁省科学计划项目(2012217005)
关键词
经验模态分解
本征模态函数
HHT
信号降噪处理
empirical mode decomposition
intrinsic mode function
HHT
signal noise reduction
