为了降低噪声对人体脉搏信号的干扰、提高采集精度, 提出了一种改进的滤波算法。从脉搏信号及其噪声特点出发, 采用与经验模态分解法结合的方法, 选择适当的小波基并改进小波阈值函数, 构造模态系数对脉搏信号进行滤波。经过理论分析与实验验证, 取得了理想的实验数据。结果表明, 改进的阈值算法不仅克服了软、硬阈值的局限性, 并能有效克服傅里叶变换后产生的边缘效应问题; 同时, 与经验模态分解法相结合, 削弱了低频噪声滤除的误差, 增强了小波变换的自适应性, 较传统的滤波方法能更好地抑制噪声, 有助于提高信噪比。

In order to reduce the disturbance of noise on human sphygmus signal and improve the precision of data acquisition, a new improved filtering method was proposed. Based on the characteristics of sphygmus signal and noise, by using empirical mode decomposition method, the appropriate wavelet basis was selected, the wavelet threshold function was improved and a modal coefficients was built suitable for the filtering of a sphygmus signal. And then, sphygmus signals were filtered by the structural modal coefficient. After theoretical analysis and experimental verification, ideal experimental data was obtained. The results show that the improved threshold algorithm not only overcomes the limitations of soft threshold and hard threshold, but also overcomes the edge effect problems effectively which is generated by Fourier transform. At the same time, combining with the empirical mode decomposition method, the filtering errors of low frequency noise are weakened and the adaptability of wavelet transform is enhanced. Compared with the traditional filtering methods, the new method can suppress noise effectively and help to improve the signal-to-noise ratio.