基于相关系数的经验模态分解算法的改进

由美国国家航空航天局（NASA）的Huang等提出的Hilbert-Huang变换（HHT）是一种先进的时频分析方法,但在经验模态分解（EMD）过程中存在虚伪IMF分量针对这一问题,提出利用分解出的各个内建模态函数和原信号的相关系数去除虚伪分量的方法。并计算去除虚伪分量后的重构信号与原信号的相关系数,仿真和实验结果表明,采用该方可以有效的除去虚伪IMF分量。

Improvement on Empirical Mode Decomposition Based on Correlation Coefficient

The paper discusses the problem of illusive component due t o HHT,uses t he correlation coefficient of IMFs and t he signal as a criterion t o decide which IMFs should be retained and which IMFs should be eliminated.The signal simulate test proved that this method to deal with the illusive component has an obvious advantage and reasonability.