A PPG signal denoising method by combining morphological filtering and double-density dual-tree discrete wavelet

As an inherent physiological signal of the human body, the PPG signal contains a large amount of physiological information, but it is often interfered with by various factors during the detection, which affects the subsequent acquisition of physiological data. This paper proposed a denoising algorithm based on morphological filtering and double-density dual-tree wavelet thresholding to address interference problems such as baseline drift and high-frequency noise in PPG signal acquisition. The results of the control experiment with traditional discrete wavelets show that our algorithm can remove high-frequency noise and baseline drift more effectively in the presence of intense noise, with an average additional increase of 1.7483 in signal-to-noise ratio and an average additional increase of 38.86% in mean square error reduction, which can facilitate the subsequent signal processing and the establishment of blood component concentration models, etc. The algorithm structure is relatively simple and of low computational complexity, which can provide better technical support for real-time waveform analysis.