Guidance
Rusen Oktem, Elif Uray Aydin Nergiz Ercil Cagiltay
Electrical and Electronic Engineering Department Software Engineering Department
Atilim University Atilim University
Ankara, Turkey Ankara, Turkey
E-mail: {rusen, eaydin}@atilim.edu.tr E-mail: nergiz@atilim.edu.tr
Abstract—This paper tackles an RFID based location finding and mathematical models at indoor environments. Location
tracking system. The system is an integral part of a navigation estimation algorithms have to offer either statistical methods or
aid being developed for guiding visually disabled people in a have to rely on vast amount of data in order to obtain a
store. The aid is composed of a portable hardware interface unit, sufficiently accurate positioning.
a standalone RFID unit, and a central processing unit. The units
interact via wireless communication to locate the position of the In the proposed system, a medium scale obscured indoor
user in a known indoor environment and tracking his/her environment is focused on, and an RFID based positioning
movement. An active RFID tag helps to estimate the location of a method is developed. RSS levels from three UHF transmitters
user and the user is guided to follow a route accordingly, via a are processed for estimating the location of the RFID tag. The
tactile compass. The system uses RF signal strengths and is based system is an intagrated navigation and information access
on Bayes Decision Theory. Initial simulation results with the system which enables not only navigation of visually impaired
system prove promising for location finding and tracking, people through aisles of a shopping store but also access the
especially when the tracked person is guided by a system. information about products. Instead of communicating through
sound, a special compass is used to direct the user. The system
Keywords-location estimation, RF signals, Bayes, RFID, also does not require the user to carry a computer system. In
classification, tracking. this system, the communication between the system and the
user through sound system is very limited to minimize the
interference with the visually impaired shopper’s perception of
I. INTRODUCTION environmental sound. The navigation system incorporates a
Navigation aids generally comprise either an outdoor or novel compass design, RFID tags, and wireless network to
indoor positioning system or both, for detecting the position of direct the user to the desired location in a building. The
the user [1-8]. The technology used in location estimation information access system includes a camera to read barcodes
depends on the type of the subject environment (indoor, of a product and wireless communication to inquire
outdoor, large scale, etc.) and also on the application. For information about that product through the bar code from the
example, GPS or GIS based systems are often utilized for large centre’s database. The system differs from existing projects in
scale outdoor environments [1,2,3]. On the other hand, a pair of that field in terms of usability and cost. Its main aim is to help
AM radio signal transmitter and a receiver, producing beep the visually impaired people do shopping without getting help
sounds when triggered, is preferred for the sake of simplicity from the environment.
and low cost, when a person wants to know the location of a
particular object in a small scale outdoor environment [4]. For II. MAIN COMPONENTS OF THE NAVIGATION SYSTEM
indoor environments, multiple ultrasound or infrared The navigation system consists of three main units: a
transmitters mounted on the walls or at the ceiling can be used
hardware interface unit (HWIU), a data collection and wireless
[2,5,6,9]. Such a system enables estimation of the location of
the receiver, by using received signal strength. However, line- communication unit (DCWCU), and a central processing unit
of-sight requirement of ultrasound and infrared sensors limit (CPU). Fig. 1 summarizes the general structure of this system.
their use in obstructed environments. Use of passive RFID tags
in a grid like structure is another alternative to estimate the
location with respect to the nearest transmitter [7]. The radio
signals do not require line-of-sight, and radio signal
transmitters are easy to integrate into wireless networks.
Hence, they have been particularly used for network based
location estimation purposes [10-12]. Radio signal based
location estimation algorithms exploit either received signal
strength (RSS), time of arrival (TOA), or related observations.
However, indoor environments possess various structures
which result in reflection, refraction, and diffraction of radio Figure 1. General Structure of the System
waves. Hence, such observations often deviate from assumed