Heart beat information prediction based on a Kalman filter

In view of the difficulty of measuring the photoelectric volume pulse wave(PPG)signal during motion, the log-normalized step-step)algorithm is proposed. The algorithm uses angular velocity to correct the acceleration and removing the gravitational component via a Kalman filter. The corrected acceleration was then used as a reference signal for the log-normalized variable step-length adaptive filtering to filter out the motion artifacts in the PPG signal, and finally the heartbeat information was estimated by spectral spike tracking. The results showed that the mean absolute error of heart rate information estimated by the algorithm decreased by 1.4 times/min and the mean relative error by 1.55%.It can be seen that the algorithm in this paper can effectively remove interference from motion artifacts and obtain more accurate heartbeat information.