Design And Research On An Adaptive Parameter Acquisition Device For Respiration And Heartbeat

At present,with the improvement of living standards,the degree of health concern is also improving.Due to sleep as a physiological behavior that occupies one-third of a day’s time,the basic physiological parameters appear in the process will greatly affect the prediction on people’s physical condition.Therefore,the design and research on a physiological parameter extraction device to monitor the physiological parameters of human sleep and the extraction and analysis of them via the device will be of great significance.The human body physiological parameters extraction device proposed in this paper is mainly for the human body’s sleep process.In the design of device hardware,the reliability,portability and durability are taken into accounted.Similarly,the accuracy,adaptability and timeliness are also taken into accounted during the process of algorithm design.Meanwhile,the overall design of the device including front-end analog circuit design,digital signal processing,back-end signal output and other multiple modules was also presented.By considering the proposed scheme of the mechanical parts of the device,a suitable signal acquisition sensor is selected for signal acquisition to ensure that the target signal is completely collected.It is inevitable to take into account that the collected signal by using the piezoelectric film is relatively weak in this paper,the front amplifier filter circuit was designed in the hardware circuit design process.At the same time,considering individual differences,the variable gain amplifier circuit was put forward,which greatly improve the efficiency of the subsequent signal processing.In addition,the design of chip peripheral circuit,charging circuit,Bluetooth module and other circuits were carried out for the selected low-power high-performance chip and finally achieve the design and development of the PCB experimental board.In this paper,an EEMD algorithm for signal separation based on fixed number of “sifting” as stop criterion was proposed to improve the timeliness of the algorithm by studying the stopping criterion.Also,due to the heartbeat signal is weak,two times’ EEMD decomposition was utilized in this paper to improve the accuracy of the algorithm.The method of signal identification and extraction by using correlation analysis for the signal after extraction was proposed in this paper.This new idea can achieve the accurate identification of the signal for the physiological characteristics of different individuals,and the adaptability of the algorithm is improved.In order to prevent the interference and misjudgment of the peak signal in the slight body movement state,a variety of operating conditions and slight body movement state signal characteristics are considered.By comparing the signal prediction scheme based on BP neural network and the one based on ARMA model,a signal processing scheme based on quadratic EEMD-ARMA prediction in slight body movement state was proposed in this paper.The scheme can realize the accurate prediction of the signal and improve the accuracy of the algorithm.Experiments show that the adaptive physiological parameter extraction device designed and researched in this subject meets the above mentioned design requirements.By studying the simulation signal and the experimental signal,the feasibility of the proposed extraction algorithm for respiration and heartbeat parameters in two different states is verified.The above research will be valuable to the development of physiological signal acquisition device and the separation and extraction of such mixed signal.