Development of wi-fi based indoor navigation system

• Main Author: Luo, Yiwen
• Other Authors: Xie Lihua
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation::Control engineering
• Online Access: http://hdl.handle.net/10356/64271
• Institution: Nanyang Technological University

This report documented the progress of a Final Year Project (FYP) at the Nanyang Technological University in partial fulfilment of the requirements of the degree of Bachelor of Engineering with the topic of a Wi-Fi based Indoor Navigation System. With the advantage of low cost and needlessness of extra installation of infrastructures, Wireless Local Area Network (WLAN, also known as Wi-Fi) was selected as the technology for the positioning platform. During the first part of the project, Weighted Path Loss (WPL) algorithm, which is a model based approach, was leveraged for the positioning due to its prompt response and simple implementation. It has been tested out in the real environment. In the second part of the project, fingerprinting based approaches were implemented as it can provide a higher positioning accuracy in general. Extreme Learning Machine (ELM) and Weighted K-Nearest Neighbor (WKNN) algorithms were leveraged for better performance, especially the accuracy and robustness of the navigation systems. The third part of project emphasized on the Access Point (AP) selections. Mutual Information (MI) based methodologies were leveraged to provide a lower computation requirement and a higher navigation speed with minimized reduction of the positioning accuracy. Online Mutual Information (OnlineMI) algorithm was proposed. Simulations and experiments were conducted to prove to feasibility and effectiveness. Matrix Laboratory (MATLAB) software was used for simulation of the algorithms. OpenWrt system, which is an open source embedded Linux system with high configurability, was deployed as the test platform of the positioning system.