Design And Implementation Of Indoor Positioning Early Warning System Based On Millimeter Wave Radar

With the development of social economy and the increase of family income,people's demand for security system is higher and higher.Traditional security mainly depends on the combination of manual and hardware to solve the problem,but its real-time performance is relatively low and the cost is too high.In contrast,the intelligent security system realized by the intelligent indoor positioning early warning system can achieve 24-hour comprehensive monitoring,timely alarm and not rely on human resources.In view of the above situation,this thesis designs and implements an indoor positioning early warning system based on millimeter wave radar.The core of the system is the indoor positioning technology.The positioning and tracking technology of unbound devices can adapt to more complex scenarios,because it can complete the positioning and tracking task with the help of signals.This thesis chooses millimeter wave radar as the monitoring equipment to achieve precise positioning of the target through the radio frequency signal of the radar.Considering that the millimeter-wave radar can detect the position information of multiple targets indoors but cannot distinguish these targets,it is necessary to realize the multi-target tracking of the millimeter-wave radar at the software level.The core problem of multi-target tracking is the data association between the target and the trajectory.First of all,this thesis studies the joint probability data association(JPDA)algorithm of the common data correlation algorithm.Secondly,considering the high time complexity of the algorithm in the case of dense multiple targets,the improved joint probabilistic data association algorithm(IJPDA)is used to realize the data association of multi-target tracking.The IJPDA algorithm proposed in this thesis keeps the original tracking accuracy of the JPDA algorithm and greatly reduces the calculation amount of the algorithm.The algorithm has the advantages of high accuracy and strong realtime performance.Then,considering the multi-target tracking of millimeter-wave radar is also a nonlinear problem,this thesis combines the IJPDA algorithm with the Gaussian particle filter algorithm(GPF)to improve the system's multi-target tracking accuracy under non-linear conditions.Finally,in order to solve the problem of path point redundancy in the trajectory,this thesis adopts the Douglas-Peucker algorithm based on node importance segmentation to achieve trajectory compression.On the basis of the above research,the system splits services according to different business functions,and uses a micro-service architecture with strong scalability and low coupling for system design.The system is mainly composed of communication module,data processing module,user center module,data display module and alarm module.The front-end Web page of the system is implemented using the Vue framework,and the mobile terminal is implemented through a small program.The system uses Spring Boot as the back-end application development framework.The Netty framework is used in the communication between the millimeter wave radar and the server.My SQL is used as the entire system database,and Redis is used for temporary storage and caching.After the development is completed,the system has been tested by deploying relevant equipment in the field.The test results show that the system can meet the actual needs in terms of functionality and nonfunctionality.To sum up,based on the research of indoor positioning technology and multi-target tracking technology,this thesis applies the current mainstream software development technology to realize a simple layout,sensitive response,comprehensive monitoring,real-time alarm and reliable operation of indoor positioning early warning system,which can meet the needs of people for intelligent security.