Heart rate acquiring method, device and system and medium

Abstract

Embodiments of the present invention disclose a heart rate acquiring method, device and system and a medium. The heart rate acquiring method includes: preprocessing pulse wave signals and accelerationsignals in all time periods; performing fast Fourier transform on the pulse wave signal in the current time period and the pulse wave signal preprocessed in the previous time period to obtain a pulsewave frequency domain set; performing fast Fourier transform on the acceleration signal in the current time period and the acceleration signal preprocessed in the previous time period to obtain an acceleration frequency domain set; and adjusting each frequency domain weight coefficient contained in a frequency domain weight coefficient set in the current time period in a preset LMS iteration wayaccording to the pulse wave frequency domain set, the acceleration frequency domain set, the frequency domain weight coefficient set in the current time period, a frequency domain weight coefficient set in the previous time period and an input power set in the previous time period to obtain a target frequency domain weight coefficient set, and performing fast Fourier transform on the target frequency domain weight coefficient set and the acceleration frequency domain set to obtain a heart rate signal without motion disturbances.

Description

technical field [0001] The present invention relates to the technical field of heart rate monitoring, in particular to a heart rate acquisition method, device, system and computer-readable storage medium. Background technique [0002] During the beating cycle of the heart, the blood flowing through the arterioles, capillaries and venules in the peripheral blood vessels is correspondingly pulsating. The blood volume is greatest when the heart contracts and smallest when the heart relaxes. This pulsating change of blood volume can generally be obtained by a photoelectric volume sensor, and the obtained waveform contains volume pulse blood flow information. Therefore, systolic blood pressure and diastolic blood pressure can be obtained through the relationship between volume pulse blood flow information and blood pressure signal, and this method is called photoplethysmography (Photo Plethysmo Graphy, PPG). [0003] In the wearable heart rate detection scheme, heart rate monit...

AI Technical Summary

Problems solved by technology

However, in the static state of human beings, the PPG signal is easily affected by the motion interference signal, and even annihilated in the motion interference signal. It is difficult to separate the pure PPG signal through an effective method.

[0004] In the traditional way, using the time-domain method to analyze, the amount of calculation is small, and the real-time performance is strong, but the stability and anti-interference are very poor, and it is almost powerless against motion interference.

Through discrete Fourier transform, using the difference of each frequency component in the PPG signal and its energy distribution, extract the frequency component corresponding to the heart rate of the human body from the spectrum, and finally get the heart rate. This method has strong anti-interference, but When the heart rate signal and the exercise signal overlap each other in the normal heart rate range, this frequency domain method is inefficient

Images