WI-FI Based Indoor Positioning Using Fingerprinting Methods (KNN Algorithm) in Real Environment

Abstract

Global positioning systems have difficulties in finding positions inside buildings. Hence other techniques and methods are required for indoor localization. Since indoor positioning needs additional indoor infrastructures deployment, convenient techniques using Wi-Fi can be developed. In this research, indoor positioning using Wi-Fi access point is investigated as the main usage of Location Based Service (LBS) applications. We employed fingerprinting methods to increase the accuracy of positioning. The study had been done in real environment in Universiti Teknologi Malaysia (UTM). The models were designed using KNN algorithm for indoor positioning. The fingerprinting dataset contained received signal strength from different numbers of existing Wi-Fi access points in the real environment. Additional features were applied to the model in order to enhance the accuracy. The accuracy rate and mean square error were calculated. Evaluations of models had been done by conducting experiments to compare each model with different features. Analysis suggests that KNN method which achieved 77% of accuracy with K=7 is the most precise model for indoor positioning in this study. By applying signal strength from additional access points, more precise results had been achieved and distance errors had been eliminated.