Research On Indoor Location Algorithm Based On WiFi Position Fingerprint

In recent years,with the increasing popularity of intelligent devices,location-based service(LBS)has been widely used in real life,and the research of improving the accuracy of location recognition has become more and more important.The use of WiFi installed in indoor space for WiFi positioning does not require the creation of additional environments,and most intelligent devices can support WiFi,so it has the advantage of low cost and has been widely used.Therefore,based on the location fingerprinting method of WiFi indoor location,two optimization methods are proposed in this paper,which improve the positioning accuracy of the location fingerprint method.Specifically:1.In this paper,we propose an optimized stochastic forest model.The model is divided into two layers,the first layer collects information data,and the indoor space is divided into rectangular grids of certain size and stored in the server.The second layer makes use of the marked grid and the RSS to carry on the second contrast analysis,cuts off the decision tree of the region discriminant decision error,then applies it to the random forest learning model,improves the execution time and the precision of the localization method.2.In the indoor location recognition system based on Fingerprint,a phased processing RSSI,is proposed.Firstly,the simulated annealing algorithm is used to optimize the fingerprint database to improve the accuracy of the fingerprint database,and the comparison with the common mean value model method is carried out.The simulated annealing algorithm is used to optimize the parameters of the Elman neural network model(SA-Elman).Then on-line phase proposed to use the SA-Elman algorithm for positioning.The algorithm has high execution time and precision.The significance of this method is that no matter the amount of data of location information,the system has good performance,and this method can be applied to the fingerprint database constructed by crowdsourcing method.3.The two methods mentioned above are verified by collecting RSSI data.The experimental results show that compared with the traditional location fingerprint location algorithm,the positioning method proposed in this paper can effectively reduce the error and improve the positioning accuracy of indoor location.