A Rule-Based Approach to Indoor Localization based on WiFi Signal Strengths

PhD Thesis Proposal Defence


Title: "A Rule-Based Approach to Indoor Localization based on WiFi Signal Strengths"

by

Miss Qiuxia Chen


ABSTRACT:

Location plays a very important role in location-aware computing systems,
in which queries are expressed on the locations of physical objects. For
example, finding the nearest objects around a person requires knowledge
about the locations of the objects and the location of the person. The
proposed research investigates methods for identifying the location of an
object within a space. The process is known as localization. The proposed
investigates localization methods based on WiFi signal strengths for
indoor environment.

A major challenge for indoor localization is that GPS (Global Positioning
System) is not available indoor. Thus, a sensor infrastructure must be
available to make indoor localization possible. This proposal focuses on
approaches based on analyzing the Received Signal Strength (RSS) of WiFi
signals. Specifically, RSS are measured at each location and stored in the
server. The measurements are called location signature of the space. When
a user requests localization service, he/she obtains the RSS signature and
compares with the location signatures at the server. These approaches have
low setup cost due to the high availability of WiFi clients on mobile
clients and access points (APs) inside most buildings.

Traditional localization methods aim to improve localization accuracy.
That is, the error between the estimated location and the actual location.
However, they assume that the location signatures are accurate, but this
is not true because RSS changes due to noise, obstacles and environmental
changes, causing localization accuracy to deteriorate quickly. Thus, they
require regular calibration on the location signatures to maintain
localization accuracy.

This proposal aims to improve both the accuracy and stability of indoor
localization. Instead of using absolute RSSs in comparing the signatures,
we propose a rule-based approach, which can achieve high localization
accuracy and stability. The main idea is to maintain the relations (i.e.,
"less than", "equal to", and "greater than") of the RSSs of the access
points received at a location and to set up rules to match the RSS
signatures based on the relations. Rule-based approach enhances stability
because the relation of two RSS signals could remain the same even when
their values are changing constantly.

To further address the stability problem, we introduce two important
notions, the stability and sensibility of APs, at a certain location. We
note that although the RSSs from APs change over time, some APs change
less than others, thus having higher stability, while some APs have
stronger signals than other, thus having higher sensibility. Based on the
stability and sensibility of APs, we introduce a method to estimate the
stability of a rule to measure the trustworthiness of the rule for
localization. We present an effective and simple approach to create the
relations and rules, as well as heuristics to select the rules for use in
localization. We develop a suite of rule-based localization methods based
on different combinations of the techniques, including pure matching of
location signatures, rule-based system with and without AP stability and
rule-based systems with and without rule stability. We implemented the
location methods and tested them in the Department's Lab area and the
results show that rule-based system with or without stability measures
perform much better than pure signature comparison and rule-based system
with stability consideration perform better that those without stability
consideration.


Date:                   Wednesday, 9 May 2012

Time:                   2:00pm - 4:00pm

Venue:                  Room 5510
                         lifts 25/26

Committee Members:      Prof. Dik-Lun Lee (Supervisor)
                         Dr. Wilfred Ng (Chairperson)
 			Dr. Lei Chen
 			Dr. Qiong Luo


**** ALL are Welcome ****