Research And Implementation Of The Indoor Location-based Services Method Based On IBeacon And WeChat

In recent years,with the development of mobile Internet technology and the popularization of intelligent terminals,location-based services has developed rapidly,and gradually integrated into people's daily life.But the current location-based services is mainly used in outdoor scene,while the application and development for indoor location-based services is not mature,and the related theoretical study is not perfect.Therefore,implementing indoor location-based services with high usability and practicability,and solving the key problems in the process of application are very important for integrating indoor and outdoor location services.In this paper,an indoor location-based services method based on iBeacon and WeChat is proposed for the actual project.There are several practical and basic problems to solve in the process of actual development.In the construction phase of the system,the devices need to be properly deployed.In the running phase of the system,a number of feasible paths need to be planned.And in the optimization phase of the system,user's behaviors need to be predicted.This paper will provide feasible solutions for above problems,and implement these methods in the real project,so as to provide a theoretical basis for the solution to above problems.The main work and contributions of this paper are as follows:First,for the problem of device deployment,an optimized deployment algorithm is implemented,so that the coverage rate of the actual region meets the actual demand.At the same time,this algorithm also minimizes the number of deployed devices and reduces the deployment costs.Because the actual region is usually large and includes various obstacles,manual deployment method will inevitably lead to serious overlap of device signals or the fact that the edge area of the region can't be covered by the signals or too many devices are deployed,resulting in poor deployment effect or increasing deployment cost.To solve these problems,this paper adopts the idea of optimization,transforms the problem of device deployment into an optimization problem by mathematical modeling,and applies a genetic algorithm to solve this problem.And the coefficients of the mathematical model are adjusted by experiments,and the advantages of genetic algorithms in solving the problem of device deployment are analyzed;On this basis,this paper realizes some functions,such as visual device deployment and user location.Second,aiming at solving the problem of multi-path planning,we realized a multi-path planning algorithm across multiple floors,so as to reduce the passenger flow pressure in each section of paths and improve the planning speed.Due to the narrowness of the interior space,it is easy to cause passenger traffic congestion,so we need to plan multiple paths for users in the navigation process to relieve the pressure of all sections of paths.At the same time,as there are multiple floors indoor,it is necessary to plan paths across multiple floors.To solve above problems,we use the idea of hierarchical paths planning to transform the problems of multi-path planning into multi-section planning in floor,then build the target path by splicing the inter sections.We analyze the time complexity of the algorithm in theory,and study the advantages of the algorithm in computing speed by experiment.On this foundation,visual path management and user navigation function are realized in this paper.Third,aiming at solving the problem of user behavior prediction,we implement a prediction method combined with model-based prediction method and rule-based prediction method,so as to achieve better prediction results.Location-based services can collect a large number of user data during running.By analyzing these data,we can predict user behavior and provide personalized location-based services in the process of location and navigation,thus increasing users' stickiness.Based on the historical data that can be collected in the actual application,we researched model-based prediction method and rule-based prediction method respectively.On this basis,the effective combination of two prediction methods is realized and the effect of prediction is further optimized.We test and compare the results of different prediction methods by experiments.Finally,on the basis of theoretical research,the functions of user data collection and data visualization are realized.