Artifact elimination using fuzzy rule based adaptive nonlinear filter

Myoelectric (ME) signals during dynamic movement suffer from artifact noise caused by mechanical friction between electrodes and the skin. The frequency components of artifact noise are similar to those of ME signals. Thus it is difficult to reject artifact noise using linear filters. the authors have proposed a nonlinear artifact elimination method that consists of an inverse autoregressive (AR) filter, a nonlinear filter, and the AR filter. To improve the performance they installed adaptive filters so as to track time-varying signals. However, fuzzy rules were employed for fine control of the nonlinear parameter. As a result, the authors achieved a better artifact elimination performance. The fuzzy rule based adaptive nonlinear filter will be useful in sport, science and rehabilitation