Research Of Enterprise Application Deployment And Simulation Based On RFID Middleware

At the rise of the Internet of Things,RFID technology has become the basis of the creation of the information society as one of technologies with the increasing application of RFID technology.To improve the utilization of RFID devices and reduce the cost of RFID devices deployment,the deployment issue of RFID device in complex environment has become the focus of research in industry and academe.Most existing methods simplify and abstract and simplify the deployed environment as point-coverage surface with random distribution of label and abstract read range of RFID reader with the binary sensing model,and then obtain optimal deployment schemas with intelligent optimization algorithms.However,in real deployed environment,existing studies always create binary sensing model based on the information about parameters provided by equipment manufacturers is obtained in ideal environment only.Therefore,value of these parameters can have errors in real environment and lead to incorrect deployment schemas.To overcome these shortcomings above,in this paper,we propose a more suitable abstract model in real deployed environment—grid sensing model.Besides,we translate the deployment issue of RFID application into optimization problems.Since the current RFID middleware is commonly implemented in the basis of EPCGlobal which mainly focuses on the function of data collection and pre-processing and lack the function of deployment optimization.In the end,we integrate the function of deployment optimization into existing RFID middleware.Specific tasks of this paper are as follows:1.For the disadvantage of binary sensing model,we propose a grid sensing model.In this model,grid domains are assigned with attributes by defining the RFID devices,sensing range of RFID device and divide sensing domain into several grid domain.And then,combining the real test results and interpolation method,the sensing ranges which correspond to sensing distances are obtained by the grid sensing model.2.In this paper,we propose an optimization strategy for RFID application deployment.This strategy abstract and simplify the application scenario using surface coverage and barrier coverage.After getting concrete deployment parameters,we adopt the proposed sensing model of RFID device and select coverage and overlapping ratio as the optimizing indexes.Considering both deployment parameters and optimizing indexes,the deployment schema is obtained using improved Particle Swarm Optimization algorithm(PSO).3.The design and implementation of the optimization-deployment function in RFID middlewares.Based on the proposed optimization strategy,we integrate a deployment-optimization function which includes test module and optimization module into our RFID middleware.Aim at the deployment issue of RFID devices in complex environment,we propose a grid sensing model and an optimization strategy for deployment of RFID application and apply them into a real deployment scenario,i.e.logistic storage.The actual tests show the effectiveness of optimization strategy.