RFID BASED LOCATION IN CLOSED ROOMS

Implementation of a Location Algorithm using a Passive UHF-RFID System

Christoph Schönegger, Burkhard Stadlmann

University of Appled Sciences Upper Austria, Stelzhamerstr. 23, 4600 Wels, Austria

Michael E. Wernle

Meshed Systems GmbH, Alte Landstrasse 21, 85521 Ottobrunn, Germany

Keywords: Passive RFID, Location, RFID-standard-components, Positioning system, RSSI value.

Abstract: This paper presents a new concept for determining the location of an RFID-tag without any additional

hardware. For this positioning system standard RFID components with passive RFID-tags within the UHF

range are used. The measurement is based on a location algorithm which makes use of the RSSI value of the

UHF reader. The RSSI value is the return signal strength indicator and, as it is shown in the paper in hand,

this signal correlates to the distance between the RFID tag and the antenna of the reader. This positioning

system is especially useful indoors, where other positioning systems may not work. For this reason it could

prove very useful in various logistics applications. The maximum distance from antenna to the tag is

approximately between 0.5 m and 3 m. To this end a special algorithm is used to obtain stable calculation

results. A minimum of two antennas is needed to get a two-dimensional location.

1 INTRODUCTION

Identification using RFID (radio frequency

identification device) is more or less standard in

many industrial applications and in many logistics

processes. There are a variety of applications where

the combination of identification and location is very

useful.

This paper presents a solution for the location of

UHF-RFID tags within the range of a reader

antenna. For this positioning system only standard

RFID equipment is used. The concept, the

algorithm, and known limitations of the system are

presented.

2 THE BASIC IDEA

The basic concept of the positioning system is to

measure the distance between an RFID antenna and

the RFID tag using only standard RFID equipment.

The measurement of the distance is done by

interpreting the signal strength of the UHF signal.

The proposed system is especially useful for indoor

use.

2.1 Location Algorithms

There are a variety of different location methods

proposed in literature and in practical use nowadays.

Examples with RFID or WLAN can be found in

(bekkali 2007, chon 2004, ekahau 2007, geroldt

2007, ibach 2005, lionel 2004, tomberge 2004,

tsukiyama 2007). This chapter provides a

comparison between these methods. The main focus

is on determining location within buildings. This

ability could be very useful for numerous

applications in logistics.

2.1.1 Cell-of-Origin Concept

This type of positioning system makes use of an

algorithm using a mobile tag and a fixed reader.

Upon detecting a nearby reader, the tag determines

its position to be nearly equal to that of the reader.

For this the tag must be within the range of the

corresponding reader.

135

Schönegger C., Stadlmann B. and E. Wernle M. (2008).

RFID BASED LOCATION IN CLOSED ROOMS - Implementation of a Location Algorithm using a Passive UHF-RFID System.

In Proceedings of the Fifth International Conference on Informatics in Control, Automation and Robotics - ICSO, pages 135-140

DOI: 10.5220/0001484201350140

 SciTePress

This type of positioning system is frequently

used with RFID systems in the 125 kHz and the

13.56 MHz range. Practical applications of this type

can be found on AGV systems (autonomous guided

vehicles) or mobile robots.

A commercially available example of this type

of positioning system which uses WLAN or

Bluetooth technology is the Ekahau Positioning

Engine (ekahau 2007).

An “inverse cell-of-origin concept” is used if the

RFID reader is on the mobile unit and different tags

have a fixed position. If the reader can communicate

with a specific tag, then the position of the mobile

unit can be determined in relation to the position of

the fixed tag. An example for this concept can be

found in (tsukiyama 2007).

2.1.2 Triangulation Method

The location of a tag can be calculated by

triangulation if the distance between the tag and

several known reference stations can be determined.

The measurement of the distance can be achieved by

detecting the runtime of the radio signal or laser or

by measuring the signal strength of the radio signal.

The best known positioning system of this type

is the global positioning system (GPS). This method

also finds use in RFID systems which use active tags

and operate in the 2.45 GHz range.

2.2 RFID

An RFID system consists of two components which

communicate via radio:

 a device called a “tag” or “transponder” which

is capable of storing data

 a so called “reader” (providing read and write

functionality) communicates with the tags

using an appropriate antenna-system,

controller and amplifier.

Nowadays tags are very cheap as they are

produced in large numbers. Therefore they are

widely used in many goods and logistics devices.

One reader can be equipped with several

antennas. Within the range of one antenna multiple

tags can be detected and communication is organised

into a sequence of the different tags.

2.2.1 Types of RFID Systems

There are various types of RFID systems available.

Different types operate at different frequencies, have

different couplings and differ as well in the energy

supply of the transponders.

The different frequencies are

 “low frequency” (119 … 148.5 kHz)

 “high frequency” (13.56 MHz)

 “ultra high frequency” (865 … 955 MHz)

 “microwave” (2.4 … 2.5 GHz)

The different coupling technologies are inductive

coupling and modulated backscatter coupling.

Both the coupling method and the operating

frequency influence the range of operation

 “close-coupling” – distance < 1 cm

 “remote-coupling” – distance < 1 m

 “long-range” – distance > 1 m

Tags can operate as passive tags, which get their

power from the reader via the electromagnetic field

of the antenna, or as active tags, which are powered

by a remote battery or some other power supply.

2.2.2 Applications

As mentioned above logistics applications need a

location algorithm to find a specific tag. Location,

used in combination with RFID, has the advantage

of combining identification and location using the

same hardware.

One typical example is the localisation of parcels

on a conveyor belt, persons walking through an

RFID gate carrying a transponder or the localisation

of a palette carried by a forklift. All these

applications have the need for optimised

performance, better process control and supervision.

For proper use it is important to heed the basic

limitations of this concept.

2.3 The RSSI-Value

The RSSI value (received signal strength indicator)

is a commonly used value within radio

communication systems. Modern RFID readers

within the UHF frequency range have the ability to

determine this value as a measure of the reflected

UHF signal from the tag. That is where RSSI value

gets its name “reflected signal strength indicator”.

one must consider that up to now this RSSI value

has not been standardised and is therefore

manufacturer dependent.

2.4 Used Technology

To achieve the goal of this location procedure, RFID

technology is used which has a long distance range

and is widely used in logistic systems. According to

VDI 4472 the recommended frequency for logistics

applications is 868 MHz (UHF range), which is

indeed very commonly used. UHF systems have

ICINCO 2008 - International Conference on Informatics in Control, Automation and Robotics

136

long-range readability with passive tags and the

reader normally has an output for the RSSI value.

Therefore a passive RFID system with a

frequency of 868 MHz was chosen for this particular

location system.

3 MEASURING DISTANCE

USING THE RSSI VALUE

During the project a theoretical and practical

analysis was undertaken to determine if the RSSI

value is a well-working solution for measuring

distance. Figure 1 shows the schematic design of the

lab equipment for the practical investigations.

Figure 1: Schematic view of the lab equipment used for

distance measuring.

Figure 2: Measured RSSI-values varying only the y-

coordinate of the tag-position.

For the practical measurements three different

antennas and two different kinds of tags were used.

Figure 2 and Figure 3 depict the mean value of the

RSSI value of a 10dBi antenna. Figure 2 shows the

variation of one coordinate and figure 3 the variation

of two coordinates which leads to a three-

dimensional radio-map. Each measured value is a

mean value of 1000 datasets.

If the distance between the antenna of the reader

and the tag is too low, the RSSI value may not be

obtained due to saturation, which causes a heavy

non-linear behaviour.

Figure 3: Three-dimensional radio map of the RSSI

values, varying the x and the y coordinates of the tag

position.

Three different types of antennas and two

different types of tags were investigated (see Table 1

for the tags and Table 2 for the antennas of the

reader). All antennas used have a circular right

polarisation.

Table 1: Comparison of different types of tags.

Philips TI

Chip Philips U-Code HSL RI-UHF-00C02-04

Protocol ISO 18000-6B EPCUHF Gen 2

Antenna

Dipole

“dog bone”

Table 2: Comparison of different antennas of the reader.

GP-ANTU RH-ANTU RH-ANTU

VSWR < 1.3 : 1 < 1.5 : 1 < 1.5 : 1

Gain [dBi] 6 > 8.5 > 10

3 dB

beamwidth

horizontal

70° 63° 55°

Max. input

power [W]

10 6 6

Due to the characteristics of the antenna on the

tag, the best result was achieved using the

combination of a

λ/2-Dipole tag and the 10dBi antenna

on the reader side.

Furthermore the influences of temperature were

also investigated. This influence must be

polynomial function 3

degree

measured RSSI values

RFID BASED LOCATION IN CLOSED ROOMS - Implementation of a Location Algorithm using a Passive UHF-RFID

System

137

compensated by adequate correction algorithms

according to the measured temperature.

Additional influences are electromagnetic

disturbances caused by fluorescent lamps and, of

course, atmospheric humidity. As this system is

proposed for indoor use humidity will not influence

the system dramatically.

4 CALCULATION OF POSITION

In chapter 3 the strong correlation between the RSSI

value and the distance between antenna and tag has

been presented. To obtain the location of the tag, the

results of more than one antenna have to be

combined. During experiments in the project a two-

dimensional location was tested. To determine this

location at least two antennas are necessary which

have contact to the same tag simultaneously.

Knowing the RSSI value, it is possible to calculate a

set of positions relative to the antenna where the tag

might be based on the radio map.

Figure 4: Schematic view of the enhanced lab equipment

for location using two antennas.

Three different calculation algorithms have been

investigated:

 Numeric iteration with finite differences

 Geometric intersection of polynomial

approximations

 Weighted position determination

4.1 Numeric Iteration

This situation, where two antennas which

communicate with the same tag is depicted in Figure

5. Antenna A has an RSSI value which can be

located at the positions marked by red points (left

curve in figure 5), antenna B has an RSSI value

which can be located at the positions marked by blue

points (right curve).

A scenic analysis (see also bahl 2000) yields the

possible tag positions represented as a set of discrete

points from each antenna in accordance with the

measured RSSI value. The most probable tag

position is the minimum distance between the

possible locations of the two antennas within the

overlapping area. If there is only a non-zero

solution, the most probable position of the tag can be

calculated by calculating the mean value. The

Accuracy of this algorithm is, however, low.

Figure 5: Scenario of two antennas for the algorithm

“Numeric iteration”.

4.2 Polynomial Approximation

This algorithm is based on a polynomial

approximation of the line of constant RSSI value

using a least square algorithm. For each antenna one

polynomial exists for the measured RSSI value.

Figure 6 depicts the appropriate situation.

The position of the tag is the intersection of the

two polynomial functions. If this polynomial

function is of second order the error of the calculated

position is rather high. If the polynomial function is

of 5

order the error is very low but the calculation

effort is very high.

ICINCO 2008 - International Conference on Informatics in Control, Automation and Robotics

138

Figure 6: Scenario of two antennas for the algorithm

“polynomial approximation”.

4.3 Weighted Position Determination

Based on the same possible tag position as in chapter

4.1 the position of the tag is calculated with a

“centre of gravity calculation”. A similar algorithm

is presented in (bulusu 2000).

Figure 7: Scenario of two antennas for the algorithm

“weighted position determination”.

The position of the tag S is calculated by the

following equation:

)(

∑

→→

⋅=

OPm

(1)

Hereby i are all points within the overlapping

area and m

are the appropriate weighting factors.

This algorithm requires fewer computing

resources and achieves a higher level of accuracy. It

combines easy computing and the possibility of

adapting the radio map, allowing for the

compensation of atmospheric humidity or other

influences.

4.4 Recommendations

Due to the previously mentioned advantages and

disadvantages the “weighted position determination”

algorithm can be considered best.

This algorithm is very stable and returns a

calculated position of the tag with respect to the

available discrete relations between RSSI value and

possible positions relative to the two antennas.

Accuracy has been checked too. In a wide part of

the space it is quite good (less than 5%) but there are

single points of rather high inaccuracy. Further

investigations have to be carried out to determine

their cause and improve this situation. Further

details can be found in (schoenegger 2007).

5 CONCLUSIONS

The paper presents the results of having investigated

an RFID based location system. Only standard RFID

equipment with passive tags operating within the

UHF range was used. Determination of location

works within a range of approximately 3 meters and

is based on the use of the reader’s RSSI value. A tag

can be located two dimensionally if it is situated

within the range of at least two readers. This

location algorithm might be used, for example, in

combination with a fork-lift, whereby the forks are

equipped with antennas. The algorithm is capable of

providing a good notion where a specific tag is

located relative to the fork.

In addition to logistics application this algorithm

may be useful for positioning of mobile robots

within production automation.

As the presented location system uses only

standard hardware and is based on a simple

calculation algorithm, it might be considered as “a

new concept”. No similar solution is known to the

authors.

RFID BASED LOCATION IN CLOSED ROOMS - Implementation of a Location Algorithm using a Passive UHF-RFID

System

139

REFERENCES

Bahl, P, Padmanabhan, V.N., 2000. RADAR: An In-

Building RF-Based User Location and Tracking

System. In Proceedings of the 19

International

Conference on Computer Communications

(Infocom2000), Vol. 2, pp 775-784, Tel Aviv, Israel,

March 2000.

Bekkali, A., Sanson, H., Matsumoto, M., 2007. RFID

Indoor Positioning Based on Probailistic RFID Maps

and Kalman Filtering. In 3rd IEEE International

conference on Wireless and Mobile Computing,

Networking and Communications, October 2007,

White Plains, USA, October 2007.

Bulusu, N., 2000. GPS-less low cost outdoor localization

for very small devices. In IEEE Personal

Communications Magazine, 7:28-34, October 2000.

Chon, H. D. et.al., 2004. Using RFID for Accurate

Positioning. In The 2004 International Symposium on

GNSS/GPS, Sydney, Australia, December 2004.

Ekahau, 2007. Ekahau Positioning Engine 4.0 User Guide,

http://www.ekahau.com (August 8

, 2007).

Finkenzeller, K., 2006. RFID-Handbuch; Grundlagen und

Anwendungen induktiver Funkanlagen, Transponder

und Kontaktloser Chipkarten, Hanser Fachbuchverlag,

edition.

Geroldt, C., Uckelmann, D., 2007. Tracking and Tracing

in Production Scenarios with Passive RFID

Transponders. In 3rd European Workshop on RFID

Systems and Technologies, Duisburg, Germany, VDE-

Verlag, June 2007.

Ibach, P., Stantchev, V., Lederer F., Weiss A., 2005.

WLAN-Based Asset Tracking for Warehouse

management. In IADIS International Conference e-

Commerce, Porto, Portugal, December 2005.

Lionel, M.Ni, Yunhao L., Yiu C.L., Abhishek P., 2004.

LANDMARC: Indoor Location Sensing Using Active

RFID. In Wireless Networks, Volume 10, Number 6,

Springer Netherlands, November 2004.

Schoenegger, C., 2007. Untersuchung der Eignung eines

passiven UHF RFID Systems für die

Positionsbestimmung in geschlossenen Räumen. In

Diploma Thesis, University of Applied Sciences Upper

Austria, School of Engineering, Wels, September

2007.

Tomberge, P., 2004. Navigation mittels RFID –

Betrachtung der Navigationsmöglichkeiten durch

RFID-Eintrittskarten bei der WM 2006. In Diploma

Thesis, University of Münster, Germany, December

2004.

Tsukiyama, T., Suzuki, A., 2007. Navigation System for

indoor Mobile Robots based on RFID Tags,

Proceedings of the 4

International Conference on

Informatics in Control, Automation and Robotics –

ICINCO, Angers, France, May 2007.

ICINCO 2008 - International Conference on Informatics in Control, Automation and Robotics

140