A Novel Boundary Extension Approach for Empirical Mode Decomposition
The Empirical Mode Decomposition (EMD) proposed by Huang et al. in 1998 shows remarkably effective in analyzing nonlinear signals. However, the boundary extension is one of the theoretical problems unsolved in EMD. In this paper, a novel boundary processing technique is proposed to deal with the border effect in EMD. An algorithm based on the sigma-pi neural network is used to extend signals before applying EMD. By virtue of this method, the frequency compression near the end is eliminated and errors caused by end effect are reduced. Verifications of the experimental signals show that the newly proposed boundary extension method is useful in practice.
Unable to display preview. Download preview PDF.
- 1.Huang, N.E., Shen, Z., Long, S.R., et al.: The Empirical Mode Decomposition and the Hilbert Spectrum for Nonlinear and Nonstationary Time Series Analysis. Proc. R. Soc. Lond. A 1, 454–899 (1998)Google Scholar
- 2.Anna, L.: Variable Sampling of the Empirical Mode Decomposition of Two-Dimensional Signals. International Journal of Wavelets, Multiresolution and Information 4738, 1–8 (2002)Google Scholar
- 3.Zhao, J.: Improvement of the Mirror Extending in Empirical Mode Decomposition Method and the Technology for Eliminating Frequency Mixing. High Technology Letter 4, 40–47 (2002)Google Scholar
- 4.Kan, Z., He, M.: A Simple Boundary Process Technique for Empirical Mode Decomposition. IEEE Trans. PAMI 23, 234–241 (2004)Google Scholar
- 9.Courrieu, P.: Solving Ttime of Least Square Systems in Sigma-pi Unit Networks. Neural Information Processing 4, 39–47 (2004)Google Scholar