Spectral analysis of myoelectric signals by wavelet methods

The wavelet packet method is introduced as an alternative method for spectral analysis of surface myoelectric (ME) signals. Both computer synthesised and real myoelectric signals during static contractions are investigated, and the results show that the wavelet packet method performs equally well as the Fourier method. Moreover, wavelet packets give one some advantage over the traditional methods such as multiresolution of frequency, as well as its potential use for effecting time-frequency decomposition of the nonstationary signals such as the ME signals during dynamic contractions. The wavelet shrinkage method is also introduced for improving spectral estimates by reducing the mean square errors for both Fourier and wavelet packets estimates.