Start Date/Time: Thursday, August 27, 2009, 12:30 PM
Ending Date/Time: Thursday, August 27, 2009, 1:05 PM
Location: ATG 310C
Speaker: Norden Huang, TMSC Chair Professor, National Central
Title: A Tutorial on Ensemble Empirical Mode Decomposition (EEMD)
EEMD is built on the Empirical Mode Decomposition (EMD), a method emphasizing the adaptiveness and temporal locality of the data decomposition. Many traditional decomposition methods, including the Fourier Transform and wavelet decomposition methods, utilize a priori determined basis functions, which may faithfully represent the characteristics of a time series in some subsections but not in other subsections if the time series is non-stationary. Other methods, including empirical decomposition methods that rely heavily on autocorrelations, involve implicitly global domain integrals and therefore, are non-local and not effective in extracting physically meaningful information from non-stationary data. EMD, which uses extrema information of the riding waves in non-stationary data, is an adaptive and temporally local decomposition method that extracts successively the riding amplitude-frequency modulated oscillatory components without using any a priori determined basis functions. Since its discovery about ten years ago, EMD has found numerous successful applications in many different scientific and engineering fields and has accumulated thousands of citations. EEMD is a recent major refinement of EMD that improves the stability of decomposition of noisy time series and overcomes the difficulty and ambiguity in interpreting the physical significance of the results derived from the products of EMD.