Research On Fall Detection And Heart Rate Measurement Algorithm Under Intelligent Medical Monitoring

With the acceleration of China’s aging process,the proportion of the elderly population is increasing year by year.The health monitoring of the elderly has attracted many scholars’ research,especially the fall detection and cardiovascular disease prevention monitoring system for the elderly living alone has been put forward,which has achieved good monitoring effect.However,the existing monitoring system often transforms the hardware and does not improve the core algorithm,so there are some problems,such as failure to detect the fall posture,false detection of fall posture,large error between the pulse wave measurement and the actual situation and so on.In order to further improve the accuracy of fall posture detection and pulse wave measurement,this paper improves the existing fall posture detection algorithm and pulse wave measurement denoising algorithm,and builds a simple monitoring system based on STM32 and FPGA.The specific work is as follows:First of all,the existing sensor-based fall attitude detection algorithms only rely on accelerometer and gyroscope,and the collected data has zero drift phenomenon,which seriously affects the accuracy of fall behavior detection.Therefore,in order to solve this problem,this paper proposes to add AHRS heading and attitude reference system algorithm and geomagnetic sensor,which can further improve the attitude angle accuracy of fall attitude detection and largely correct the zero-drift phenomenon of the collected data.Through the simulation analysis of the proposed algorithm,the maximum acceleration correction accuracy is 0.171 g,and the yaw angle,pitch angle and roll angle are reduced by 2.934 degrees,1.102 degrees and 2.807 degrees respectively.Secondly,pulse wave is one of the most important reference basis for measuring heart rate,which is easily affected by interference noise in the process of acquisition,affecting the clinical diagnosis results.In order to solve the problem that there is a lot of mixed noise interference in the pulse wave signal,CEEMDAN combined with wavelet algorithm is proposed to remove the inaccuracy of heart rate calculation caused by mixed noise signal,which can alleviate the baseline drift phenomenon to a certain extent.Through the simulation and analysis of the improved algorithm,a larger signal-to-noise ratio is achieved,and the mean square error is further reduced.Finally,this article builds a simple monitoring system based on the dual main control system STM32 and FPGA,and integrates the improved algorithm into the system.Through many experiments and comparative analysis,compared with the monitoring system without the improved algorithm,the monitoring system with the improved algorithm has higher accuracy in detecting fall posture,and the obtained heart rate is closer to the real value.