Improved Empirical Mode Decomposition Based on Harmonic Wavelet Filter

Abstract:

The decomposition result of empirical mode decomposition (EMD) may be incorrect if some frequency components of original signal are close to each other. To avoid illusive result and get an exact one suiting for analysis, a harmonic wavelet filtering can be used before the EMD is performed. In the course of application of harmonic wavelet filtering, data was decomposed subsection overlapped to overcome the edge effect of usual digital filter that is caused by data truncation. The experiments showed its feasibility and validity in application of real-time processing.

• Periodical: Applied Mechanics and Materials (Volumes 239-240)
• Pages: 946-949
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.239-240.946
• Online since: December 2012
• Authors: Yang Liu, Yong Liu, Shun Jing Yu, Jie Feng Yang
• Keywords: Empirical Mode Decomposition (EMD), Filter, Harmonic Wavelet

References

[1] Norden E H, Zheng S, Steven R L, Manli C W, Hsing H S, Quanan Z, Nai-Chyuan Y, Chi C T and Henry H L 1998 Proc. R. Soc. Lond. A 903-995.

[2] Wei H, Zheng S, Norden E H and Yuan C F 1999 Proc. Natl. Acad. Sci. USA 96 1834-1839, (1999).

[3] Chin H L, Tsu C W and Norden E H 2001 BSSA Special Issue of Chi-Chi Earthquake, 91 5 1339-1357.

[4] Chunming H, Huadong G, Changlin W, Dian F and Huiyong S 2003 High technology letters (Chinese version) 6 13-17.

[5] David E N 1993 Proceedings: Mathematical and Physical Sciences 203-225.

[6] David E N 1994 Transactions of the ASME 409-416.

[7] Yu G 2003 Ph.D. Dissertation of Chongqing University 78-79.