Font Size: a A A

Based On The Android Platform WiFi Positioning System Research And Design

Posted on:2018-05-12Degree:MasterType:Thesis
Country:ChinaCandidate:F GaoFull Text:PDF
GTID:2348330536479754Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
With the rapid development of mobile Internet industry,the demand for location-based services(LBS)is increasing.LBS is based on the location of mobile user.However,GPS signal is weak under the indoor,which can not satisfy the requirement of indoor positioning.Indoor localization method based on WiFi is quickly becaming one of the hot spots.This paper is based on the WiFi location fingerprinting positioning system.Specific work is as follows:First,a set of indoor positioning system is developed based on Android platform,which is based on the WiFi location fingerprint technology,using WKNN matching algorithm,and combined with SVG indoor map.The experiments were carried out on 150 square meters of indoor environment.In the experiment,the average positioning error of the system is 1.42 meters,which can meet the requirements of indoor positioning accuracy.Second,this paper optimizes the parameters of the positioning system.Giving the reasonable sampling number and AP number: the deployment should be integrated into the database complexity and accuracy requirements,with an interval of 2.5 meters WiFi information collection establishment location fingerprint database is the most reasonable;as far as possible to meet the location within the region can search for at least 5 AP signal.Third,this paper puts forward a solution to the problem of AP in the positioning stage: according to the AP which is searched in the real-time positioning phase,the temporary fingerprint data sub library is established,and then the matching operation is carried out.This method improves the error caused by the method of assigning AP to the fault-100 dBm.Fourth,this paper improves the matching algorithm WKNN.In the improved algorithm,the actual coordinates matching K fingerprint correlation operation,shooting and other positions far apart the fingerprint coordinates,and then calculate the positioning results,the positioning system was promoted to turn at the corner,near the doors and windows and other special areas of precision.The research of this paper provides theoretical and practical support for the WiFi indoor positioning system,which lays the foundation for the practical application of WiFi positioning system.
Keywords/Search Tags:Android Platform, WiFi Indoor Positioning, Location Fingerprinting Positioning, Weight K-Nearest Neighborhood
PDF Full Text Request
Related items