Open AccessHeart Rate Estimation Algorithm Based on Normalized Least Mean Square combining Complete Ensemble Empirical Mode Decomposition with Adaptive Noise
Author(s) -
Jie Cheng,
Chang-shui Yu
Publication year - 2022
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2224/1/012126
Subject(s) - hilbert–huang transform , adaptive filter , noise (video) , signal (programming language) , mathematics , filter (signal processing) , algorithm , noise reduction , white noise , computer science , statistics , artificial intelligence , computer vision , image (mathematics) , programming language
Aiming at the difficulty of extracting heart rate due to the relative movement between human body and acquisition equipment, this paper proposes a Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDA) based on Normalized Least Mean Square (NLMS) adaptive filtering combined with fully adaptive noise sets for heart rate estimation. Firstly, an adaptive filter was used to filter the motion artifacts in the original signal by taking the triaxial acceleration signal as the reference signal. Then, the PPG signal was decomposed by CEEMDAN to obtain a series of Intrinsic Modal Functions (IMF) from high to low frequency. The Permutation Entropy (PE) criterion was used to determine the threshold range of the signal, so as to filter out the high-frequency noise and baseline drift. The results show that the Pearson correlation coefficient between the computed heart rate of PPG signal after noise reduction and the standard heart rate based on ECG signal is 0.742, and the average absolute error percentage is 6.08%, which indicates that the proposed method can accurately calculate the heart rate in the exercise state, and is beneficial to human physiological monitoring under the exercise state.