Cooperative localization techniques for Wi-Fi (WLAN) systems

Abstract

Localization for indoor environments has gained considerable attention over the last decade. The localization technology provides the fundamental basis for a myriad of locationenabled services in indoor environments such as locating personnel and objects in residential homes, guiding shoppers inside a mall for the latest discount offers, locating the elderly in nursing homes or children tracking in day-care centers. There are many localization techniques with varying degrees of accuracy and complexity. However, it is more convenient to take advantage of the existing wireless network infrastructure in indoor environments, such as WiFi/wireless local area network (WLAN)Access points (APs). Recently, IEEE 802.11v standard has been proposed to enable accurate Time Of Arrival (TOA) - distance - measurements between WLAN devices. Thus one promising approach to providing accurate localization in indoor environments is to integrate IEEE 802.11v with Cooperative Localization (CL) in WLAN networks. The main challenge facing WiFi cooperative localization in indoor/urban environments is the multipath and Non Line Of Sight (NLOS) problems. Hence, the main contribution in this thesis is to 1) measure and characterize TOA based ranging for WLAN systems 2) investigate NLOS identification for WLAN systems 3) integrate the TOA based ranging models into well-known centralized cooperative localization algorithms. 4) evaluate the performance of centralized cooperative localization algorithms in realistic indoor environments for IEEE 802.11 WLAN Systems. The results show that the TOA ranging measurement is affected by multipath and NLOS and it follows different distributions based on the environment structure and NLOS conditions. Also, new NLOS identification metric is developed based on Coherence Bandwidth (CB) which shows good identification performance of LOS/NLOS channel condition for WiFi based network. Moreover, the performance of the developed cooperative localization algorithm for Wireless Sensor Network (WSN) was tested and the results show that the accuracy of the used CL algorithm degrades because of NLOS and multipath problems.

Date of Award	2015
Original language	American English
Supervisor	Saleh Al-Araji (Supervisor)