Noncontact Heart Rate Detection Method Based on Kalman Filter

Noncontact vital signs detection via mm-wave FMCW radar has attracted more attention due to more comfortable experience and lower burden. To enhance the stability and accuracy of the detected heart rate, this paper introduces an improved Kalman filter adaptive heart rate tracking algorithm. Firstly, the phase signal obtained by signal preprocessing is filtered to extract the heartbeat signal. After the initial value of Kalman filter heart rate tracking is obtained by using the frequency estimation method based on peak-to-valley difference for heartbeat signal in time domain, the frequency of heartbeat signal is accurately measured by the Rife algorithm in frequency domain as the observation value, and the optimal estimation is carried out to obtain the accurate heart rate at this time. According to this heart rate, the filter setting is updated to narrow the passband range, and the updated filter is used for the next moment of processing, so as to improve the detection accuracy. Experiments are conducted using a 77GHZ FMCW radar to prove that the function of the improved algorithm outperforms the existing method. The heart rates obtained by the improved algorithm are compared to the output of a chest heart rate sensor.