基于修正加速度的对数归一化变步长自适应滤波的心率估计算法

针对运动过程的光电容积脉搏波(photoplethysmography,PPG)信号易受到运动伪影的干扰,使得心率测量困难的问题,提出了一种基于修正加速度的对数归一化变步长自适应滤波(log-normalized least mean square,Log-NLMS)算法.该算法利用角速度对加速度进行卡尔曼滤波修正,去除重力分量.然后将修正后的加速度作为对数归一变步长自适应滤波的参考信号,滤除PPG信号中的运动伪影,最后利用谱峰追踪估计心率.结果表明:该算法估计心率的平均绝对误差较传统算法平均降低了1.4次/min,平均相对误差降低了1.55％,可见本文算法能够有效地去除运动伪影的干扰,获得更为准确的心率.

Heart rate estimation algorithm based on logarithmic normalization variable step size adaptive filtering based on modified acceleration

Aiming at the problem that photoplethysmography (PPG) signals during exercise are susceptible to interference from motion artifacts, making it difficult to measure heart rate, a logarithmic normalized variable step adaptive filter algorithm based on corrected acceleration is proposed. the angular velocity was used by the algorithm to perform Kalman filter correction on the acceleration, which had removed the gravity component. The corrected acceleration was used as the reference signal of logarithmic normalization variable step size adaptive filtering to filter out the motion artifacts in the PPG signal, and finally the heart rate was estimated by spectral peak tracking. The results show that the average absolute error of the estimated heart rate of the algorithm is reduced by 1.4 BPM on average compared with the traditional algorithm, and the average relative error is reduced by 1.55%. It can be seen that the proposed algorithm can effectively remove the interference of motion artifacts and obtain a more accurate heart rate.