Noise reduction of surface electromyography signal using spectrum interpolation and empirical mode decomposition
基于谱插值与经验模态分解的表面肌电信号降噪处理

This paper introduced a combined method of spectrum interpolation and empirical mode decomposition (EMD) to explore the noise reduction of surface electromyography (sEMG) signal. According to the spectrum interpolation, subtracted the power line interference and could hold the useful information of sEMG signal at the power line frequency. Based on the EMD algorithm, selected some proper intrinsic mode functions (IMFs) to be analyzed by the wavelet soft threshold, removed the residual component and several noised IMFs dominated by low frequency, then, reconstructed the denoised sEMG signal by the processed IMFs. Using the simulated and real sEMG signals, the experimental results show that the signal quality is improved. And the noise in the sEMG signal can be suppressed by the proposed method.