Design And Research Of Piezoelectric Ballistocardiography Detection Device

Ballistocardiography(BCG)is the vibration signal caused by the blood pumping of heart.As a non-contact technology,BCG detection carries nice potential in cardiovascular condition monitoring.However,recent researches about BCG mainly focus on the design of signal processing algorithm and most BCG detection devices bear limited signal-noise ratio(SNR).Since the BCG signal is weak and its spectrum overlaps with some noise,most BCG detecting systems are vulnerable to motion artifacts.To solve the above problems,this thesis designed a BCG detecting system for high-quality BCG signal detection,and research was carried out about the noise removal for swing motion and the structure optimization for different sitting postures.The piezoelectric BCG sensor designed here consists of the external force transmitting structure and the internal sensing components which works through piezoelectric ceramics.The motion noises developed by swing motions in sitting state to the sensing components on different positions have been studied by waveform comparison and correlation analysis.Based on the features of swing noise,a symmetrical sensing structure was designed to remove the left-to-right swing motion noise.In addition,a 6-axis motion sensor was used to obtain the attitude angles of the upper force transmitting plate through coordinate transformation and attitude fusion,and the detection of tilting position was realized based on attitude angles.According to the analysis of signals detected in different positions with several sitting postures,for the left-right tilting or front tilting posture,the proposed symmetrical structure is capable of increasing signal-noise ratio and J-peak amplitude.As for front-rear tilting posture,BCG quality can be increased by using the signal detected from rear sensing component.In order to synchronously convert BCG and Electrocardiography(ECG)into analog voltage signal,this thesis designed the 3-stage BCG signal processing circuit as well as the 6-lay PCB of the pre-processing module.By embedded programming,this thesis realized the following functions of the detection system: synchronous collection of BCG signal,ECG signal and the motion sensor data,calculation of the attitude angles based on the data of 6-axis motion sensor,the low-energy Bluetooth communication with table computer,and the serial communication with PC.The static experiment was conducted with the participants sitting still.Result of static experiment shows that the proposed system can achieve a high accuracy of 99.7% in heart rate detection,a low error of 1.6 ms in cardiac cycle detection and a high SNR of 32 dB.As for the motion experiment,participants were asked to repeatedly swing from left to right.The result of swing test shows that the symmetrical structure can detect most J-peak waveforms by effectively removing the motion noises caused by left-to-right swings.Therefore,the proposed system processes high quality of BCG detection in static state as well as the capability of detecting heartbeat signal during left-to-right swings.