Title :
The Study of Feature Extraction Using Local Energy of Frequency Bands Based on Wavelet Packet Decomposition
Author :
Wang, Fengtao ; Chen, Jianguo ; Ma, Xiaojiang
Author_Institution :
Dalian Univ. of Technol., Dalian
Abstract :
A new feature extraction method based on local energy of frequency bands is presented in this paper. Wavelet packet decomposition is studied and the concept of local energy of frequency bands based on wavelet packet decomposition is used to measure the signal energy in selected frequency bands and certain time-interval. The proposed feature extraction method was tested on simulated signals. The results showed that the developed approach was effective.
Keywords :
feature extraction; signal processing; wavelet transforms; feature extraction; frequency band; signal energy; wavelet packet decomposition; Discrete wavelet transforms; Educational technology; Fault diagnosis; Feature extraction; Frequency; Laboratories; Machining; Signal analysis; Wavelet analysis; Wavelet packets;
Conference_Titel :
Natural Computation, 2007. ICNC 2007. Third International Conference on
Conference_Location :
Haikou
Print_ISBN :
978-0-7695-2875-5
DOI :
10.1109/ICNC.2007.749
