Entropic Empirical Mode Decomposition

TL;DR

This paper introduces a novel approach combining Permutation Entropy with Empirical Mode Decomposition to address the mode mixing problem in analyzing nonlinear, nonstationary data, enhancing the interpretability of intrinsic mode functions.

Contribution

It proposes a new method integrating PE with EMD to improve the quality of IMFs by reducing mode mixing issues.

Findings

Reduced mode mixing in IMFs

Enhanced analysis of nonlinear data

Improved physical insight into data

Abstract

Empirical Mode Decomposition(EMD) is an adaptive data analysis technique for analyzing nonlinear and nonstationary data[1]. EMD decomposes the original data into a number of Intrinsic Mode Functions(IMFs)[1] for giving better physical insight of the data. Permutation Entropy(PE) is a complexity measure[3] function which is widely used in the field of complexity theory for analyzing the local complexity of time series. In this paper we are combining the concepts of PE and EMD to resolve the mode mixing problem observed in determination of IMFs.