Wi-Fi Localization and Navigation for Parking Assistance System

Abstract

Over the years, Wireless Fidelity (Wi-Fi) based localization has attracted considerable attentions in practical applications by providing the localization and navigation functions in outdoor or indoor environments. Because of its widespread and overwhelming influence in public door contexts, Wi-Fi-based technology shows a lot of promise for applications. This work developed an android application (App) that uses equations generated via matrix laboratory (MatLab) software. The experimental testbed was chosen and some software platforms which include Wi-Fi Analyzer software, MatLab and the Android software were also used for data gathering, analysis and development of the android App. The developed android App. relates the grid locations within a classroom to the average received signal strength within each grid from specific access points.  This was then used to determine the user’s grid location within the classroom. Data were collected at different locations within each classroom. The data was then put in an appropriate format and the signal strength at each point was plotted against the reference points within the classroom for selected Access Points (Aps) in order to observe and analyze the pattern. The equations which were generated were written and implemented accordingly in the Java code in the main activity of the android Wi-Fi Localization App. The first step towards achieving accuracy was done by adjusting the position of the access points, while the second step was by improving on the current access point technology and also changing the channel of broadcast for each access point so that it is less susceptible to noise which causes the signal strength to fluctuate often over time. To validate the built Android Wi-Fi localization app, extensive experimental tests were conducted in the classroom for various access point setups. The test results show that the equations used in the app can give reliable estimations of a user’s location if the grid locations have distinct Signal Strength (SS) values and the values are somewhat stable over time. it could be concluded that our technique can achieve great localization and navigation performance in indoor environments.