A Hybrid Model for Ubiquitous Positioning System

  • 唐 怡君

Student thesis: Master's Thesis

Abstract

Internet of Things (IoT) refers to a scenario in which objects are uniquely identified and has the ability to connect the physical world over a network without requiring human-to-human and human-to-computer interaction Hence the most important part of IoT is how to make objects addressable and traceable As we know Global Positioning System (GPS) is a speed and precise positioning system However the situation is quite opposite when GPS positioning in indoors Due to the absence of Line of Sight (LOS) to satellites the performance of GPS is severely degraded As for indoors there are many mature positioning technologies such as WiFi RFID Bluetooth and so on Each positioning method has its advantages weakness and restrictions Thus combining more than one type of positioning algorithms at the same time could get better performance Through inspired by IoT there is a need for ubiquitous positioning system that can work both indoors and outdoors Due to unique properties of different positioning systems drawing on the strength of each to offset the weakness of the other is necessary Therefore we propose a ubiquitous positioning system combining GPS WiFi BT 4 0 and Dead Reckoning (DR) Also with the use of Complementary Extended Kalman Filter (CEKF) for the data fusion and correcting the trajectory; the proposed system implements a ubiquitous positioning system which can position objects anytime and anywhere The result shows the proposed hybrid positioning system performs better than other individual positioning methods
Date of Award2014 Aug 28
Original languageEnglish
SupervisorMing-Shi Wang (Supervisor)

Cite this

'