基于小波分析的脉搏波信号去噪

摘要

In order to remove the noise signals such as the EMG interference and baseline drift from pulse-wave signals, an improved spatial correlation filtering algorithm was presented to target the poor performances of traditional filtering methods. After the signals were decomposed with wavelet, we added an adaptive adjustment factor to the wavelet transform values. At last, the signals were reconstructed with retained wavelet coefficients. This new algorithm keeps the main edges of signals well while getting rid of noises mostly. Simulation results indicate that the proposed de-oising method avoids pseudo-Gibbs phenomena and gives better SNR gains and RMSE performance.%脉搏波信号采集过程中存在引入的肌电干扰和基线漂移等噪声,引起脉搏信号不准确.针对传统滤波方法滤除脉搏波噪声性能较低的特点,提出了一种改进的空域相关滤波方法.将信号小波分解后,在小波变换值的相关量中加入一个自适应调整系数,最后用保留的小波系数对原信号进行重构；改进方法具有良好的自适应性能和显著的滤波效果,在有效去除噪声的同时,很好地保留了信号的主要细节.仿真结果表明,改进算法去噪效果好,同时减小误差,提高了信噪比,为脉搏分析提供了准确信息.