Research On Sleep Staging Method And Sleep Monitoring System Based On Heart Rate Variability

Sleep,as a necessary physiological activity of human beings,has attracted widespread attention.Most of the current sleep research uses multi-channel(EEG,ECG,eye electricity,EMG)sleep monitor as a tool to study the sleep process combined with data.However,the collection process of this method is tedious,which interferes with the user's sleep process greatly,so it is not easy to apply the research results to daily life.Now we need a kind of sleep monitoring technology with convenient collection and little interference to users.Some studies have shown that HRV and sleep state are related to the activity of autonomic nervous system,and HRV has the advantages of simple collection and low data complexity.Therefore,from the perspective of heart rate variability,this paper studies sleep stage and sleep monitoring.This paper focuses on the research of sleep stage method and sleep monitoring system based on HRV.The research work is divided into the following two parts:1)Research on sleep stage method.Firstly,the pre-processing process extracts RRI sequence from ECG signal,and then uses RRI signal quality evaluation method proposed in this paper to evaluate signal quality and eliminate the abnormal values in the sequence.Then 13 features are extracted from time domain,frequency domain and non-linear.Five new features are obtained by principal component analysis.After normalization,the second-order hidden Markov model is established respectively.Finally,the final sleep stage results are obtained by weighted voting based on the results of these five models and the experiment is designed to verify.2)Development of sleep monitoring system.A portable ECG acquisition device is designed.After the functions of ECG signal preprocessing,RRI sequence extraction and RRI sequence preprocessing are realized,the key parameters are uploaded to the upper computer for staged processing.This paper introduces the design details of the sleep monitoring system from hardware design to embedded software design to upper computer software design.The sleep stage model in this paper achieves 71.60% of the total recognition accuracy under the three stage goal.Finally,a set of sleep monitoring system based on HRV is developed,which provides a reliable solution for the sleep staging technology based on HRV and the application of sleep monitoring technology in life.