A New Signal Feature Extraction Method - Fractal dimensions of TimeFrequency Domain

Authors

  • Yuan Yu School of mechanical engineering, Dalian JiaoTong University
  • Wang TingYu School of mechanical engineering, Dalian JiaoTong University
  • Li Baoliang
  • Yao Shixuan

Keywords:

multifractal; Hilbert-Huang transform; general dimension

Abstract

In this paper, vibration signal multifractal method was researched, multifractal on time-frequency domain was an energy fractal and the signal feature extraction was based on the analysis of its energy distributing. The method analysed signal on timefrequency domain to characterize the distributing of its frequency or energy, and the signal’s feature was extracted by fractal dimension. After the signal was changed to time-frequency domain by Hilbert-Huang transform, general dimension Dq would be calculated from the signal in time-frequency domain by least square method. In the end, examples of emulator and practical application proved that this integrated method was feasible.

Downloads

Download data is not yet available.

Downloads

Published

2010-04-01

How to Cite

Yuan Yu, Wang TingYu, Li Baoliang, & Yao Shixuan. (2010). A New Signal Feature Extraction Method - Fractal dimensions of TimeFrequency Domain. International Journal of Computer Information Systems and Industrial Management Applications, 2, 8. Retrieved from https://cspub-ijcisim.org/index.php/ijcisim/article/view/44

Issue

Section

Original Articles