摘要
针对随机噪声对经验模态分解(Empirical Mode Decomposition,EMD)的影响,本文在分析现有处理方法的基础上,提出一种基于小波包和奇异值分解(Singular value decomposition,SVD)的改进算法。首先选取合适的小波基函数进行小波包分解,再重构奇异值矩阵进行降噪处理,对降噪后的信号进行EMD。通过仿真实验,证明该算法可以有效的分离含噪信号中的有用分量,算法简单易于实现。
The impact of random noise on the empirical mode decomposition based on the analysis of the existing approach, an improved algorithm based on wavelet packet and singular value decomposition is put forward. An appropriate wavelet basis function is selected for wavelet packet decomposition to deal with noise in the reconstruction of the singular value matrix EMD of the denoised signal. Through simulation and engineering experiments, the algorithm can effectively separate the useful component of the noisy signal, the algorithm is simple and easy to implement.
出处
《科技视界》
2012年第27期209-211,276,共4页
Science & Technology Vision
关键词
经验模态分解
降噪
小波包
奇异值
Empirical mode decomposition
De-noising
Wavelet packet
Singular value