Research On WiFi Indoor Location Based On Ap Selection Algorithm

In modern society,the progress of wireless technology has brought unprecedented convenience to human beings,and services based on location information have been booming.People have higher and higher requirements for positioning,and map navigation has gradually become an essential software for people to travel.In the field of indoor positioning,many scholars have studied different technologies to achieve the location estimation of target and improve positioning accuracy.Different positioning technologies have different application scenarios.Thesis mainly studies WiFi indoor positioning using different algorithms,which based on the optimization of AP selection algorithms,Received Signal Strength(RSS)and Channel State Information(CSI).Thesis firstly analyzes the basic concept of indoor positioning technology and the basic theory of positioning algorithm,and chooses WiFi positioning technology based on location fingerprint according to the actual application scenario.WiFi positioning technology is inseparable from the installation and application of aps.Deploying a small number of aps will cause signal dispersion,and a large number of aps will cause data redundancy and calculation difficulties.To solve these problems,thesis proposes a new AP selection algorithmbased on information entropy and mutual information,which measures not only the discriminant ability of a single AP,but also the joint discriminant ability of two AP.In addition,regional growth algorithm is introduced to improve the role of AP in the whole region.Then,the WiFi location technology is studied based on AP selection algorithm,since RSS only provides coarse-grained information,CSI embodies fine-grained physical layer information,so the characteristics of RSS and CSI are organically combined.In the offline phase,dimension reduction of RSS and CSI were respectively processed to reduce the computational complexity and fingerprint update of the database.In the online phase,fusion algorithm RSCS was used to estimate the target location.Finally,the experimental simulation and analysis of the research content are carried out.The results show that the proposed algorithm can effectively reduce the redundancy of fingerprint database and improve the positioning accuracy.