脉搏波信号降噪和特征点识别研究

Research on pulse wave signal noise reduction and feature point identification

对脉搏波的完全分析是建立在含有少量噪声且较为清晰的脉搏波信号中,然而在采集脉搏波信号时容易受到多种干扰的影响,使其提取出来的脉搏波含有大量的噪声,因此降噪处理显得尤为必要。同时,脉搏波中含有人体生理病理信息,不同的人将表现为不同的特征,可以看出确定脉搏波特征点对于分析人体生理健康很有意义。针对信号去噪问题采用小波变换和多分辨率分析的方法,该方法在时域和频域都能表征信号局部信息的能力,且具有对信号具有自适应性。运用极值法确定出脉搏波的峰值点,然后再根据峰值点确定出其他特征点的位置,实验证明该方法能够增加特征点的检出率。

The full analysis on the pulse wave is in the collection of the pulse wave signal in the signal containing a small amount of noise and more clear, however vulnerable to a variety of the influence of interference, so that the extracted pulse wave contains a lot of noise, thus loweringnoise handling is particularly necessary.At the same time, pulse wave contains human physiological and pathological information, different people will exhibit different characteristics, can be seen to determine the characteristic point of the pulse wave analysis of human physiological health makes sense.Method for signal denoising using wavelet transform and multi-resolution analysis, the method in the time domain and frequency domain can characterize the local signal information, and has a self-adaptive signal.The use of a the extremum method to determine the peak point of the pulse wave, then determined based on the peak point out the location of the other features, the experiment proved that this method can increase the detection rate of feature points.