Introducing libeemd: A program package for performing the ensemble empirical mode decomposition

摘要

The ensemble empirical mode decomposition (EEMD) and its complete variant(CEEMDAN) are adaptive, noise-assisted data analysis methods that improve onthe ordinary empirical mode decomposition (EMD). All these methods decomposepossibly nonlinear and/or nonstationary time series data into a finite amountof components separated by instantaneous frequencies. This decompositionprovides a powerful method to look into the different processes behind a giventime series data, and provides a way to separate short time-scale events from ageneral trend. We present a free software implementation of EMD, EEMD andCEEMDAN and give an overview of the EMD methodology and the algorithms used inthe decomposition. We release our implementation, libeemd, with the aim ofproviding a user-friendly, fast, stable, well-documented and easily extensibleEEMD library for anyone interested in using (E)EMD in the analysis of timeseries data. While written in C for numerical efficiency, our implementationincludes interfaces to the Python and R languages, and interfaces to otherlanguages are straightforward.