The Analysis of Mode Mixing Problem in Empirical Mode Decomposition

Abstract

After defined the basic concept of mode or intrinsic mode, EMD as a data-driven adaptive decomposition method, its purpose is to decompose arbitrary non-stationary and nonlinear signal into different intrinsic mode functions (IMFs) on the time series. However, in some cases, the  EMD method often cannot get ideal decomposition results which can result mode mixing problem. once mode mixing phenomenon appear, mode mixing will affect the subsequent decomposition components, the time-frequency distribution of subsequent IMFs are confusion, eventually the EMD decomposition process lose physical meaning. The EMD mode mixing problem directly affect its application in various fields. This paper discuss the basic cause of mode mixing, the method used to overcome this problem. Results show that ensemble Empirical mode decomposition reduces the problem of mode mixing and the signal can be reconstructed