arrow
Volume 8, Issue 3
The Application of Frequency Slice Wavelet Transform in ECG Signal Feature Extraction

Nan Li & Zhaochun Yang

Journal of Fiber Bioengineering & Informatics, 8 (2015), pp. 461-472.

Published online: 2015-08

Export citation
  • Abstract
According to the difference of time-frequency characteristics of ECG (electrocardiogram) signal and jamming signal, FSWT (Frequency Slice Wavelet Transform) is used to deal with the ECG signal denoising and feature extraction. FSWT algorithm has a good time-frequency aggregation and can freely choose the frequency range for signal reconstruction to extract characteristic information flexibly and accurately. Firstly, ECG signal is decomposed to get the whole time-frequency distribution characteristic by using FSWT and carries on the detailed analysis. Frequency section interval is determined according to frequency distribution characteristics of the jamming signal, disturbance signal is refactored and isolated through the time-frequency filter and the inverse transformation of FSWT. So it can realize the ECG signal denoising and feature extraction. The proposed algorithm is compared with wavelet threshold denoising method, Empirical Mode Decomposition (EMD) and average empirical mode decomposition (AIMF). The simulation results show that, the denoising effect of FSWT is superior to other methods for ECG signal, and gives the time-frequency distribution characteristics of ECG signal.
  • AMS Subject Headings

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{JFBI-8-461, author = {}, title = {The Application of Frequency Slice Wavelet Transform in ECG Signal Feature Extraction}, journal = {Journal of Fiber Bioengineering and Informatics}, year = {2015}, volume = {8}, number = {3}, pages = {461--472}, abstract = {According to the difference of time-frequency characteristics of ECG (electrocardiogram) signal and jamming signal, FSWT (Frequency Slice Wavelet Transform) is used to deal with the ECG signal denoising and feature extraction. FSWT algorithm has a good time-frequency aggregation and can freely choose the frequency range for signal reconstruction to extract characteristic information flexibly and accurately. Firstly, ECG signal is decomposed to get the whole time-frequency distribution characteristic by using FSWT and carries on the detailed analysis. Frequency section interval is determined according to frequency distribution characteristics of the jamming signal, disturbance signal is refactored and isolated through the time-frequency filter and the inverse transformation of FSWT. So it can realize the ECG signal denoising and feature extraction. The proposed algorithm is compared with wavelet threshold denoising method, Empirical Mode Decomposition (EMD) and average empirical mode decomposition (AIMF). The simulation results show that, the denoising effect of FSWT is superior to other methods for ECG signal, and gives the time-frequency distribution characteristics of ECG signal.}, issn = {2617-8699}, doi = {https://doi.org/10.3993/jfbim00133}, url = {http://global-sci.org/intro/article_detail/jfbi/4727.html} }
TY - JOUR T1 - The Application of Frequency Slice Wavelet Transform in ECG Signal Feature Extraction JO - Journal of Fiber Bioengineering and Informatics VL - 3 SP - 461 EP - 472 PY - 2015 DA - 2015/08 SN - 8 DO - http://doi.org/10.3993/jfbim00133 UR - https://global-sci.org/intro/article_detail/jfbi/4727.html KW - ECG Signal KW - FSWT KW - Time-frequency Analysis KW - Feature Extraction AB - According to the difference of time-frequency characteristics of ECG (electrocardiogram) signal and jamming signal, FSWT (Frequency Slice Wavelet Transform) is used to deal with the ECG signal denoising and feature extraction. FSWT algorithm has a good time-frequency aggregation and can freely choose the frequency range for signal reconstruction to extract characteristic information flexibly and accurately. Firstly, ECG signal is decomposed to get the whole time-frequency distribution characteristic by using FSWT and carries on the detailed analysis. Frequency section interval is determined according to frequency distribution characteristics of the jamming signal, disturbance signal is refactored and isolated through the time-frequency filter and the inverse transformation of FSWT. So it can realize the ECG signal denoising and feature extraction. The proposed algorithm is compared with wavelet threshold denoising method, Empirical Mode Decomposition (EMD) and average empirical mode decomposition (AIMF). The simulation results show that, the denoising effect of FSWT is superior to other methods for ECG signal, and gives the time-frequency distribution characteristics of ECG signal.
Nan Li & Zhaochun Yang. (2019). The Application of Frequency Slice Wavelet Transform in ECG Signal Feature Extraction. Journal of Fiber Bioengineering and Informatics. 8 (3). 461-472. doi:10.3993/jfbim00133
Copy to clipboard
The citation has been copied to your clipboard