一种改进的经验模型分解方法

A Method of Modified Empirical Mode Decomposition Method

在对复杂信号进行分析中,常把它展开成一系列基本信号,然后,通过研究每个基本成分或者相应系数的特点来分析复杂信号。Huang等人提出经验模型分解方法(Empirical Mode Decomposition,EMD),通过筛选,将复杂信号中分解成一系列内在模型函数(Intrinsic Mode Function,IMF)。在本论文中,作者对经验模型分解中的一个重要的筛选过程作了部分改进,提出了一种改进检验模型分解法(Modified EMD,MEMD)。利用改进检验模型分解法,能够既快又准确地获得内在模型函数,而且,得到的内在模型函数能保留原信号中各成分的瞬时频率的规律。

During analyzing a complex signal, the signal was often decomposed into a series of basis signals, and then the complex signal was analyzed by searching the basis signals or their coefficients, Huang proposed an empirical mode decomposition （EMD）, which complex signal was decomposed into a serial intrinsic mode functions （IMFs） by the sifting process. In this paper, the author modified the important sifting process on EMD, and then put forward a modified （ MEMD ） which could gain quickly and precisely IMFs which kept the character of instantaneous frequency of original signal＇ s components.