Reseach And Application Of Home Remote Monitoring System Based On Embedded Middleware

With the continuous progress and development of network and communication technology,more and more IoT technologies are used in remote monitoring system,which brings convenience to people’s life and work.However,there are still some problems in the remote monitoring system,such as the heterogeneity in the front-end of information collection and the back-end of data display,difference between different data and the poor compatibility between platforms,etc.Aiming at the above problems existing in the current home remote monitoring system,based on the embedded middleware technology,a middleware system for the interaction between wireless sensor devices and remote monitoring terminals is proposed in this thesis.The system uses excellent cross-platform function of Qt,making the transplanted gateway apply to the ARM11 architecture on the piece of Linux system.By using ZigBee wireless communication to manage terminal,the coordinator interacts with the home gateway thtough serial commnunication and uses the cloud database to store data collected,and then by using the smartphone,people can remotely monitor and query history,finally realizing the functions,such as the collection of data,upload,query,etc.In the 4G or WiFi network environment,the cloud server responds to the query command issued by the remote monitoring terminal.Through the query operation of the cloud database,the query result is responded to the remote monitoring terminal in JSON format.The test results show that the middleware system can effectively solve the interaction between wireless sensor equipment and remote monitoring terminal under a variety of operating system platforms.The overall implementation efficiency is high,and the data query is convenient.It can be successfully applied in the home remote monitoring system.Based on real-time data from the terminal nodes,in order to avoid false alarm and missed alarm caused by a single sensor single,multisensor data fusion technology is added.Via training of two types of sample data to extract the characteristic parameters.What’s more,people can judge whether there is a fire by Fisher linear discriminant,Finally,the simulation results prove the effectiveness of this method.