Study of Wi-Fi Fingerprint-Based Indoor Positioning on a smartphone

Abstract

Nowadays, there is a high demand of Location Based Services (LBS) for indoor environments. Because the widely used Global Positioning System (GPS) is unavailable indoors, many technologies and methods have been investigated to address this problem. Indoor positioning based on Wi-Fi fingerprinting has attracted significant interest due to its potential to obtain high accuracy at low costs. It can be applied to any indoor scenario where Wi-Fi networks are deployed without any additional hardware. This thesis first examines the current developments in the field of indoor positioning and it investigates the problems with Wi-Fi fingerprinting in particular. In a second stage an Indoor Positioning System (IPS) is developed based on a novel implementation using a modified Weighted K-Nearest-Neighbors (WKNN) algorithm with prior Spearman's Rank Correlation (SRC) calculation. The proposed positioning algorithm also takes into account the number of signals being omitted during localization. Therefore, unreliable results have a smaller impact on the final result. The proposed system consists of two parts: an Android smartphone application and a webserver provided with the proposed algorithm written in Erlang. The proposed IPS achieves an exceptional accuracy with an average positioning error of approximately 80 cm using an up-to-date fingerprint database. On the basis of the results of this research, it can be concluded that it is possible to use Wi-Fi fingerprinting for indoor positioning to obtain a state-of-the-art accuracy.