Wavelet transform and independent component analysis application to multi-channel SEMG processing
暂无分享,去创建一个
Multi-channel Surface Electromyography (SEMG) always interfere each other while being acquired, which would distort the source SEMG feature. The new method of SEMG processing is put forwarded to eliminate the signal interference between multi-channel SEMG by combining the wavelet transform with the independent component analysis (ICA). In this method, some noise is removed from the raw SEMG first which are reconstructed again for the observed signals of ICA. Then, the observed signals are separated blindly by using Infomax algorithm. Finally, this paper induces correlation coefficient to judge the consistency of the outputs in ICA with the source SEMG. The experimental results indicate that this method is an effective way to separate multi-channel SEMG.
[1] Luo Zhizeng. Study of Tactile Sensor In Bionical Artificial Hand , 2003 .
[2] Erkki Oja,et al. A class of neural networks for independent component analysis , 1997, IEEE Trans. Neural Networks.