Development of Wi-Fi localization tags with on-board sensors for indoor human activity sensing

• Main Author: Zhao, Yuanyuan
• Other Authors: Xie Lihua
• Format: Final Year Project
• Language: English
• Published: 2014
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/60135
• Institution: Nanyang Technological University

The rapid development of mobile devices and the popularity of social networks have aroused extensive demands on Location Based Service (LBS) in recent decades. LBS is an ubiquitous application whose functions are based on the locations of clients. The core of LBS is an effective positioning system. As wireless LAN (WLAN) is widely used and very easy to access, it has been extensively studied for indoor positioning recently. Received signal strength (RSS) in Wi-Fi networks is commonly employed in indoor positioning systems (IPS); however, device diversity is a fundamental problem in such RSS-based systems. The variation in hardware is inevitable in the real world due to the tremendous growth in recent years of new Wi-Fi devices, such as iPhone, iPad, and Android devices. Different Wi-Fi devices performed differently in respect to the RSS values even at a fixed location, thus degrading localization performance significantly. In this project, Procrustes analysis method is adopted to transform the Wi-Fi RSS to a new type of standard location fingerprints which can tolerate the heterogeneity of devices. The similarity between fingerprints is defined as Signal Tendency Index (STI). Then, by combining STI with Weighted Extreme Learning Machine (WELM), an indoor localization algorithm is developed which takes advantages of both algorithms. The proposed algorithm was evaluated in an indoor Wi-Fi environment, where realistic RSS measurements were collected through heterogeneous laptops, smart phones and tablets. Experimental results demonstrate the effectiveness of STI-WELM which outperforms previous positioning features for heterogeneous devices.