Research on Heart Rate and Respiration Rate Recognition Algorithm in Multi-Channel Data Acquisition Scenario.

Sleep is an important part of human life. However, conventional polysomnography is not suitable for long-term home monitoring. Non-contact measurements, which measure sleep status in an unobtrusive manner, have received some attention in the sleep monitoring field. In this paper, the data of the human body on the bed is collected based on the piezoelectric sensor array. To extract the vibrational components of heartbeat and respiration from the original signals, discrete wavelet transform was applied to reconstruct the ballistocardiogram and respiration signals. On this basis, in order to solve the problem of signal channel selection and the identification of heart rate and respiration rate, all channels are screened in this paper, then identify heart rate and respiration rate using the method based on frequency domain and the method based on template. And the influence of different deployment scenarios on the recognition accuracy of the algorithm is discussed through experiments.