Optimal Wavelet Functions in Wavelet Denoising for Multifunction Myoelectric Control

Wavelet analysis is one of the most important methods for analyzing the surface Electromyography (sEMG) signal. The aim of this study was to investigate the wavelet function that is optimum to identify and denoise the sEMG signal for multifunction myoelectric control. This study is motivated by the fact that there is no universal mother wavelet that is suitable for all types of signal. The right wavelet function becomes to achieve the optimal performance. In this study, the optimal wavelets are evaluated in term of mean square error of two criterions, namely denoising and reconstruction. Fifty-three wavelet functions are used to perform an iterative denoising and reconstruction on different noise levels that are added in sEMG signals. In addition, various possible decomposition levels and types of wavelets in the denoising procedure are tested. The results show that the best mother wavelets for tolerance of noise in denoising are the first order of Daubechies, BioSplines, and ReverseBior but the classification results are not recommended. The fifth order of Coiflet is the best wavelet in perfect reconstruction point of view. Various families can be used except the third order of BiorSplines and Discrete Meyer are not recommended to use. Suitable number of decomposition levels is four and optimal wavelets are independent of wavelet denoising algorithms.