Research And Implementation Of Medical Monitoring System Based On RFID Technology

With the development of China's economy,the continuous improvement of the national health consciousness and the status quo of China's gradual entry into an aging society,the concept of wireless Internet of Things for human health monitoring has been raised.Through the combination of medical system and Internet of Things,remote monitoring of human physiological information and real-time location can greatly improve the management atmosphere of existing hospitals.At present,indoor positioning technology is still in the research stage compared with mature outdoor positioning technology.Among them,RFID-based indoor positioning technology is favored by researchers because of its low cost,high precision and reliable performance.In this paper,the RFID-based indoor positioning algorithm will be researched.Based on the research results,the demand analysis and design of the medical monitoring system will be carried out,and the medical monitoring system will be realized.This paper first analyzes the current status of RFID indoor positioning technology development at home and abroad,and then introduces several RFID indoor positioning algorithms widely used in RFID indoor positioning systems and analyzes the shortcomings of these algorithms.Afterwards,in order to solve the shortcomings of the traditional positioning algorithm being severely affected by the environment and the cost of manual maintenance is too high,according to the characteristics of the Generalized Regression Neural Network(GRNN)with strong nonlinear fitting ability and few artificial adjustment parameters,this paper proposes a indoor positioning algorithm based on GRNN.In addition,based on the proposed GRNN neural network localization algorithm,this paper optimizes the classical Kalman filter algorithm and proposes an adaptive dynamic sliding window-extended Kalman filter data filtering algorithm.The particle swarm algorithm is used to replace the traditional parameter traversal-cross validation.Methods The GRNN neural network parameters were obtained,and according to the model structure of the GRNN neural network,the GRNN neural network output dimension was extended to meet the needs of this scenario.Afterwards,the effectiveness of the proposed algorithm and the optimization method is verified by experiments.The experimental results show that the data filtering algorithm implemented in this paper combined with the particle swarm optimization algorithm optimizes the GRNN neural network localization algorithm.The maximum positioning error is less than 1.5 meters and the average positioning error around 1 meter,the positioning accuracy is significantly improved compared to the traditional path loss model method.After that,the demand analysis of the medical monitoring system was carried out.The topology of the system was designed according to the requirements to meet the monitoring requirements of the location and physiological information in the medical system.The information management system was designed and constructed using the Django framework to complete information management system.Finally,this paper uses the developed information management system and hardware equipment to build an experimental verification platform in the laboratory(common office environment),and performs functional testing and performance testing on the system.The test results show that the system function is normal,and can accommodate 100 people to access the system at the same time.