An indoor navigation application using Wi-Fi fingerprinting and inertial-based sensors for mobile devices

Date of Publication

2012

Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Computer Science

College

College of Computer Studies

Department/Unit

Computer Science

Thesis Adviser

Nellie Margaret Chua

Defense Panel Member

Gregory G. Cu
Raymund C. Sison

Abstract/Summary

Current GPS navigation systems have proven to be capable of navigating a user in outdoor locations. However, they are not fit for use within indoor locations, due to the blocking of GPS signals indoors. Therefore, Wi-Fi signals have been used to determine a user’s location indoors. However, these signals are not always reliable due to their constant fluctuations, significantly affecting the overall positioning accuracy of the signals. This paper presents a mobile application that integrated the inertial sensors embedded within a mobile device to navigate a user indoors using the particle filter algorithm. By combining these two distinct technologies, the localization errors that Wi-Fi technology produces were compensated. The proponents compared three different combinations in order to determine if inertial sensor and particle filter are necessary for increasing the accuracy of localization. These three combinations were Wi-Fi Fingerprinting only, Wi-Fi Fingerprinting + Inertial Sensors, Wi-Fi Fingerprinting + Inertial Sensors and particle Filter. Test showed that Wi-Fi Fingerprinting and Wi-Fi Fingerprinting + Inertial Sensors produced almost same results which are 7.79 meters and 6.96 meters and 6.96 meters respectively. Adding inertial sensors was not sufficient since the initial position determined through Wi-Fi Fingerprinting was not highly reliable. Inertial sensors gave only the velocity which helps in following the path of the user. On the other hand, combining particle filter with the previous combination increased the accuracy of localization to 2.68 meters. A test done with 30 respondents showed that the system was able to navigate the user to his/her chosen destination.

Abstract Format

html

Language

English

Format

Print

Accession Number

TU18513

Shelf Location

Archives, The Learning Commons, 12F, Henry Sy Sr. Hall

Physical Description

1v. various foliations : illustrations ; 28 cm.

This document is currently not available here.

Share

COinS