An approach for enhancing fingerprinting for indoor localization

Other Title(s)

طريقة جديدة لتحسين نظام البصمات لتحديد المواقع بالأماكن المغلقة

Dissertant

al-Misri, Hala Khalid Abd al-Munim

Thesis advisor

al-Mahadin, Bassam Muhammad Salim
al-Abadilah, Ahmad Hamad Hammud

Comitee Members

al-Jaafirah, Ahmad Falah
al-Hammuri, Awni Mansur
al-Tarawinah, Muhammad Mahmud

University

Mutah University

Faculty

Information Technology College

Department

Computer Science Department

University Country

Jordan

Degree

Master

Degree Date

2016

English Abstract

Indoor positioning systems attracted the attention of researchers in indoor localization field.

There are some drawbacks which still need to be resolved.

First, GPS technology has proven itself as a successful technology for outdoor localization, but it is not suitable for indoor localization, since the satellite signals cannot penetrate the walls or roofs of buildings.

Second, some of indoor localization approaches which depend on Wi-Fi signal have problems with signal attenuation which affects the system accuracy.

This thesis proposes an indoor localization approach, which depends on received signal strength from access points which are distributed inside the building, as well as, acceleration and direction from Smartphones sensors, 3-axis accelerometer and orientation respectively, in order to track the users and determine their positions.

The proposed approach addresses the limitations of a promising indoor positioning technique which is called fingerprinting technique, such as the effort of building the radio map, the size of database, the energy consumption and the calibration.

The proposed approach has two phases, offline and online phases.

In offline phase, all sections (Rooms and Corridors) inside the building are surveyed depending on people's movement pattern when they enter into the sections, in order to store the acceleration, direction and received signal strength readings as a unique signature for each section called offline signature in the signatures map.

In online phase, the users can detect their positions using Smartphone, through recording the acceleration, direction and received signal strength readings, which form the online signature, then comparing it with all offline signatures using new proposed algorithm called lookup algorithm in order to determine the users' positions.

The proposed approach is tested inside different types of buildings using real experiments, which illustrate that combining the direction and acceleration with received signal strength reduces the time and cost which are consumed when the users’ positions are determined, and minimizes the effects of calibration process that is usually required in fingerprinting approaches because of signal attenuation problems.

So, the proposed approach outperformed the fingerprinting approaches in terms the overall accuracy.

Main Subjects

Information Technology and Computer Science

No. of Pages

78

Table of Contents

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-726441