Research Of Human Upper Limbs Surface Electromyography Signal Acquisition And Processing

Surface electromyography with its faint and complex is a biomedical signal. It comes from contraction of muscle that is controlled by neuron. As a result, the surface electromyography analysis can reflect a tendency of neuromuscular control, and then make a study on clinical diagnosis, rehabilitated medical etc. This paper is based on the research and application of rehabilitation robots key technology which is the985project, The main research is building surface electromyography signal acquisition system, pretreatment of surface electromyography signal, extraction of surface electromyography signal feature and action classification. this paper successful built surface electromyography signal acquisition system, Solve surface electromyography signal de-noising problems, successful design mode classifier with a high recognition rate.In this paper, surface electromyography signal collection system as a physiological signal acquisition device named FlexComp Infiniti with10channel parameters is designed. And Biograph Infiniti software is used to analysis the collected signal. Through the signal acquisition experiment, a conclusion can be got that this system can well acquisition and processing of surface electromyography signalIn surface electromyography signal acquisition, there are many Jamming signals that are mixed up with wanted signals. Among them, the strongest interference is50Hz Power frequency interference, and with other noise signal.In this thesis, how to wipe off disturb and yawp is made under the MATLAB. IIR50Hz notch filter and FIR50Hz notch filter that are used to remove50Hz Power frequency interference and not affect the available signals are designed. Experimental results show that FIR50Hz filter can effectively remove50Hz power frequency interference. FlexComp Infiniti and Biograph Infiniti software can be evaluated for patients by collecting physiological signal, and can treat patients’mental illness and rehabilitation by biological feedback.In this paper, wavelet multi-resolution analysis to threshold de-noising is used to remove other noise. The wavelet de-noising methods, that using wavelet multi-resolution analysis to threshold de-noising and using wavelet transform digital filters to threshold de-noising, of surface electromyography signal is also researched. Experimental results show that the wavelet transformation of the digital filter can effectively remove other noise.And, furthermore, wavelet transform is chose to be used for feature extraction, and getting the characteristic value of each layer. BP neural network is used to Line pattern classification and four kinds of action recognition, and getting a high rate of experimental results.By the development of this paper, we proposed a series of algorithms and implementing approaches of surface electromyography signal acquisition. In the future, we will continue to study the various algorithms of pattern recognition and find which one is more suitable for the EMG. We will also try different kinds of classifiers and compare the performance of them and apply our system to the rehabilitation robot.