Enhancing WLAN-based Indoor Localization with Channel State Information

The Hong Kong University of Science and Technology
Department of Computer Science and Engineering


PhD Thesis Defence


Title: "Enhancing WLAN-based Indoor Localization with Channel State 
Information"

By

Miss Jiang XIAO


Abstract

With marvelous development of wireless techniques and ubiquitous 
deployment of wireless systems indoors, myriads of Indoor Location-Based 
Services (ILBS) have permeated into numerous aspects of modern life. The 
most fundamental functionality, is to pinpoint the location of the target 
via wireless devices. According to different application scenarios, we can 
classify the existing indoor positioning techniques into two categories: 
device-based and device-free. In general, the applications that requiring 
specific devices on the entities to fulfill the localization function 
belong to the device-based category. Otherwise, the ones whose subjects 
carry no device pertain to device-free.

WLAN has been witnessed to be a promising technique for indoor 
localization owing to its wide availability and prevalent infrastructure. 
Most WLAN-based positioning systems depend on received signal strength 
(RSS). However, RSS value is not reliable due to its coarse measurement 
and high temporal variability. In this paper, we first propose a new 
alternative called channel state information (CSI) which processes 
beneficial properties for accurate localization, including: frequency 
diversity and temporal stability. We then leverage CSI for device-based 
positioning and design three systems FILA, FIFS and NomLoc. FILA applies 
ranging approach to effectively compensate the multipath effects in 
complicated indoor environments. FIFS is a fingerprinting system that 
explores CSI to manifest a unique location. The NomLoc system leverages 
the mobility of nomadic APs and fine-grained CSI to dynamically adjust the 
WLAN network topology without calibration efforts. Afterwards, we exploit 
the possibilities of employing CSI for device-free application scenarios, 
and design an indoor motion detection system FIMD, which is an essential 
primitive for localization. We continue to further realize a device-free 
fingerprinting system Pilot based on the observation that CSI is capable 
of distinguishing the environment variances when the object presents in 
different positions. We conduct experiments in several typical indoor 
scenarios with commercial IEEE 802.11 NICs. Extensive experiments 
demonstrate that CSI is superior to RSS for WLAN-based indoor localization 
in both device-based and device-free circumstances, and the performance 
gain can be over 75 percents.


Date:			Thursday, 22 May 2014

Time:			4:00pm – 6:00pm

Venue:			Room 4480
 			Lifts 25/26

Chairman:		Prof. Kai Tang (MAE)

Committee Members:	Prof. Lionel Ni (Supervisor)
 			Prof. Shing-Chi Cheung
 			Prof. Qiong Luo
 			Prof. Zongjin Li (CIVL)
                         Prof. Jiannong Cao (Computing, PolyU)


**** ALL are Welcome ****