Location Fingerprint Based WiFi Indoor Positioning Technology And Implementation

With the popularity of smartphones and the rapid development of mobile Internet,the demand of location-based services (LBS) grows rapidly and it is widely used inmobile applications, such as: location-based check-in service, electronic coupon service,route planning and location tracking and so on. Although the most widely used GPS(Global Positioning System, Global Positioning System) is able to meet the mostdemand of the outdoor location service, however, GPS is not work as it signals areobscured in indoor environment. As an alternative solution, location-based fingerprintWiFi indoor positioning technology utilizes the widespread of wireless assess point toposition, with the advantage of wide application range and low cost, it has become oneof the most popular indoor positioning technology in recent years.Howerver, the WiFi received signal strength indication (RSSI) is not a fixed valuedue to it is vulnerable to various disturbances and RSSI of different devices differs atmost cases, therefore existing WiFi indoor positioning technology is unable to meet theneeds of LBS neither in accuracy nor in large-scale application. This dissertationstudies the characterization of received signal strength indication(RSSI) for indoorpositioning, the influence of equipment defference on positioning accuracy, positioningalgorithm, designs and implements a location-based fingerprint WiFi indoor locationsystem on Android platform. The mojor contributions and innovation of this thesis are:(1) With the WiFi RSSI sample data measured by android smart phone, the thesisanalyzes the characterization of RSSI for positioning purpose. The humanbehaviors are divided into five kinds of typical scenarios to analyze the impact ofhuman behaviors on the WiFi RSSI and the primary research explains the reasonsof bimodal phenomenon.(2) According to the distribution of RSSI sample data, a modified Gaussian model isproposed to fit the RSSI data. The test result shows that the proposed mothed leadto better fitting effect and better location performance. (3) Considering the difference of RSSI measured by different equipment, a RSSIdifference based location algorithm is proposed to mitigate its impact onpositioning result.(4) This thesis designs and implements a location fingerprint based WiFi indoorpositioning system and analyzes the impact of AP number and its layout onpositioning accuracy.