• DocumentCode
    3056037
  • Title

    Modified Lifting Scheme Characterization for the Compression of EMG Signals

  • Author

    Ele, Pierre ; Ntsama, Pascal Eloundou ; Tonye, Emmanuel

  • Author_Institution
    ENSP of UYde, Yaounde, Cameroon
  • fYear
    2009
  • fDate
    Nov. 29 2009-Dec. 4 2009
  • Firstpage
    73
  • Lastpage
    80
  • Abstract
    This article discusses issues on data compression of electromyographic (EMG) signals. A new coding scheme was proposed in our previous work, and a summary of the first experimentations had been given. The new aspect presented in this paper is a deepening of this method followed by a N-order general decomposition scheme. From this generalization, a more complete characterization of the method is made. Tests for order N equal to 1, 2 and 3 carried out showed some improvements compared to. The technique helps to show that the decomposition of Modified Lifting Scheme (MLS) to the order N can solve the recurring problem of storage and/or duration of transmission with more efficiency as compared to wavelet packet transform and lifting scheme. Each of the methods is shortly presented, along with results obtained from actual signals. Signal to noise ratio, compression ratio, distortion of the mean frequency, entropy and computation load allow objective comparisons.
  • Keywords
    data compression; electromyography; medical signal processing; wavelet transforms; EMG signal compression; N-order general decomposition scheme; data compression; entropy; mean frequency distortion; modified lifting scheme characterization; signal-noise ratio; wavelet packet transform; Electromyography; Image coding; Indexes; Signal to noise ratio; Wavelet packets; EMG; Modified Lifting Scheme; compression; wavelet packet;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signal-Image Technology & Internet-Based Systems (SITIS), 2009 Fifth International Conference on
  • Conference_Location
    Marrakesh
  • Print_ISBN
    978-1-4244-5740-3
  • Electronic_ISBN
    978-0-7695-3959-1
  • Type

    conf

  • DOI
    10.1109/SITIS.2009.22
  • Filename
    5633995