Design And Implementation Of An Embeddedbased Wearable ECG Monitoring System

Traditional ECG monitoring devices have problems such as large size,high cost and complicated operation,which cannot meet people’s needs for daily monitoring of their own heart functions.And the existing portable monitoring devices often have a single monitoring function and lack data storage and playback function,which can hardly meet the demand of long-term monitoring for patients with chronic diseases.In this thesis,a wearable ECG monitoring system is designed to address the drawbacks of traditional ECG acquisition devices.The embedded-based wearable ECG monitoring system designed in this thesis consists of a hardware physiological signal acquisition terminal as well as a Lab VIEW software system.The system uses a highly integrated and interference-resistant AD8232 chip as the front-end acquisition chip,and also integrates a temperature acquisition chip,a fall detection module and a WIFI communication module,and uses a microdata processor STM32F103 chip to control the whole hardware acquisition terminal.The acquired signals are sent to the PC upper computer for display and analysis through the WIFI module.The upper computer mainly completes waveform reading,QRS wave detection,heart rate calculation,power spectrum analysis,data storage and other processing,and can call MATLAB noise reduction algorithm for filtering processing through the node and display it through the visualization interface.The focus part of this thesis is on the optimization of the ECG denoising algorithm.For the problem that the modal number K and the penalty parameter α are difficult to determine when using the variational modal decomposition algorithm for signal decomposition,a VMD algorithm based on pelican optimization(POA-VMD)is proposed,and the envelope entropy is selected as the value of the fitness function of the pelican optimization algorithm to optimize the parameter combination(K,α)with the goal of minimizing the envelope entropy.And the experimental comparison with three classical denoising algorithms is carried out.The results show that the POA-VMD denoising algorithm has significant superiority in both graphical and data performance compared with the traditional algorithms,and has the best signal-to-noise ratio and mean square error in terms of human ECG signal denoising effect.The thesis concludes with a functional test of the whole monitoring system,and all the functions meet the expectations,realizing an ECG monitoring system with the features of portability and comfort,meeting the monitoring requirements and data management.