Design and implementation of RFID-based anti-theft system

Abstract

High value assets loss due to theft can be reduced if the attempt to remove the asset is detected at once. This paper study the design of anti-theft system based on RFID technology. The RFID tag that attaches to an object is integrated with motion sensor. The interrupt function is utilized to the tag to automatically detect motion of an object. The real-time notification of theft is realized using multi-communication principle between a tag and a reader. The anti-theft system proposed here was tested for motion detection efficiency and RF communication performance in multi-floored building to verify its reliability.

Full text

Design and Implementation of RFID-based
Anti-Theft System
Shabinar Binti Abdul Hamid1
Anis Diyana Rosli2
Universiti Teknologi MARA
Permatang Pauh, Malaysia
1shabinar@ppinang.uitm.edu.my
2anis.diyana@ppinang.uitm.edu.my
Widad Ismail
School of Electrical &Electronic
Engineering
Universiti Sains Malaysia
eewidad@eng.usm.my
Aimi Zulliayana Rosli
Universiti Teknologi MARA
Machang, Malaysia
zullia068@kelantan.uitm.edu.my
Abstract—High value assets loss due to theft can be reduced
if the attempt to remove the asset is detected at once. This paper
study the design of anti-theft system based on RFID technology.
The RFID tag that attaches to an object is integrated with
motion sensor. The interrupt function is utilized to the tag to
automatically detect motion of an object. The real-time
notification of theft is realized using multi-communication
principle between a tag and a reader. The anti-theft system
proposed here was tested for motion detection efficiency and RF
communication performance in multi-floored building to verify
its reliability.
Index Terms--- motion sensor, UHF RFID, active RFID,
anti-theft device
I. INTRODUCTION
RFID (Radio Frequency Identification) is a technology
that employed the storing and retrieving data remotely and
provides identity codes to the monitored object. It comprises
of RFID reader and RFID tag. The RFID tag is attached to the
monitored object. The identity codes in the form of unique
identification number is stored in the RFID tag, contains all
information about the monitored object. In common
application, the RFID reader queries the tag to transmit
identification data when in the interrogation zone. The process
of capturing, processing and transmission of data are executed
in real-time. The function of RFID is further enhanced by
integrating the sensor to the tag.
One of the application benefits from the integration is an
anti-theft solution [1],[2]. In common scenario the owner will
only realize the asset missing when it was already out of sight.
A clear trend indicates that users are looking for the anti-theft
systems that not only capable to protect asset but also to
prevent loss. The prevention can only be realized if the
subscriber is alerted in real-time and the details of occurrence
are recorded in the monitoring system for tracking.
This paper describes the design of anti-theft system based
on RFID technology to protect high-value asset such as items
in the museum. The items such as artwork, museum artifacts
and rare book are seldom moved by curatorial staffs. The tag
designed in this project is used to monitor high value asset that
rarely moved.
II. BASIC OPERATING STRUCTURE OF THE SYSTEM
The anti-theft system consists of RFID reader, RFID tag
and interfacing unit. The setup of proposed RFID-based
anti-theft system is illustrated in Fig. 1. The proposed system
is functioning such that when the tag which is attached to the
monitored asset senses the tampered event, it will
immediately send the alert signal to the RFID reader. Upon
successfully receiving the alert signal, the reader will send the
information to the interfacing unit to circulate the warning.
The transmission power of the reader is constant throughout
the process. With immediate alert from the tampered asset, the
curators may have enough time to react before the stolen asset
is going out of sight. This enables an active communication
path between the tag and the user, as long as the ends are
within the operating range of 30 to 100 meters, depending on
type of the reader antenna.
Fig. 1. Illustration of the proposed anti-theft system
The functional block diagram of RFID reader is given in
Fig. 2. The main components of reader are controller unit,
CC1100 RF transceiver module and MAX3222 transceiver.
There are two micro-switches used in the RFID reader. One is
functioning as a reset switch and the other one used as a
request button. After the reader is powered up, a request
button is pressed in order to start the tag collection process.
The tags that are located in the reader interrogation zone will
respond to the reader request by transmitting their
2012 IEEE International Conference on Control System, Computing and Engineering, 23 - 25 Nov. 2012, Penang, Malaysia
978-1-4673-3143-2/12/$31.00 ©2012 IEEE
452

identification data. Then, the reader sends an
acknowledgment signal to the corresponding tag as a delivery
confirmation. The tags turn to sleep mode after received an
acknowledgement signal. If the tag senses the tamper event, it
automatically wakes up and transmits an alert signal to the
reader.
Fig. 2. RFID reader functional block diagram
The main components assembled in the tag are not much
different than the one built in the reader except the motion
sensor and power supply circuit. The tag is integrated with
passive motion sensor to minimize the power consumption
during continuous active mode. The basic functional block
diagram of RFID tag is shown in Fig.3.
Fig. 3. RFID tag functional block diagram
.
III. MOTION SENSOR SIGNAL PROCESSING
The motion sensor is integrated to the tag. To enable the
tag instantly detects the tamper event, the motion sensor is
connected to an interrupt pin of the microcontroller. Any
changes to the signal pattern will immediately trigger
microcontroller to analyze the input signal. The motion sensor
built in with a number of balls that act as a switch as shown in
Fig. 4. When at rest, the balls make contact with the
conducting body terminal which allows electric current to
flow through. While in motion, the balls move around and
intermittently make contact with the body terminal. This
triggers the switch to be in high or low state. The continuous
changes of the switch state produce the waveform as shown in
Fig. 5. The status of asset motion state is determined by
calculating the number of high state for a certain period of
time. Since switch state changes very fast while moving, the
suitable sampling rate need to be evaluated to ensure the
controller unit can capture these changes. The method used to
determine the suitable sampling period is explained in the next
paragraph.
(a) (b)
Fig. 4. Motion sensor behavior when (a) at rest and (b) in motion.
Sampling period is required to process the continuous data
of motion sensor at a discrete time. To determine the sampling
period, output signal of motion sensor was analyzed using the
oscilloscope. The waveform is captured during the sensor in
continuous movement. Analysis was conducted on a few
samples of the waveforms to measure the switch closure
period. The Nyquist-Shannon sampling theorem states that the
perfect reconstruction of signal is possible when the sampling
frequency is greater than twice the maximum frequency of the
signal being sampled [3]. The most important parameter for
sampling period measurement is the minimum switch closure
period, σmin. The lower range of sampling period is
estimated as σmin/2 [4]. Research from [4] found that the
switch closure must be sampled at least three times within the
maximum switch closure, σmax /3. Therefore, the range of
sampling period can be denoted as follows.
σmin/2 ≤sampling period ≤ σmax /3 (1)
σmin/2 is lower range of sampling period while σmax /3 is upper
range of sampling period. The minimum and maximum
interval of switch closure captured using oscilloscope is
shown in Fig. 5.
2012 IEEE International Conference on Control System, Computing and Engineering, 23 - 25 Nov. 2012, Penang, Malaysia
453

Fig. 5. Minimum and maximum interval of switch closure
From the output signal measured, it is found that the lower
range of sampling period is 25 µs while upper range is 250 µs.
Test was conducted to determine the best sampling period
should be used in the motion sensor algorithm to determine
the motion status of the tag within that range. The sensor was
moved horizontally and vertically to calculate the number of
detected movement data. Both direction was tested to ensure
the motion can be detect when the asset move slowly at any
directions. As shown on Table I, the sampling period of 25 µs
has better sensitivity to detect motion event.
Tab le I. Average detection data captured in horizontal and vertical direction
Sampling
Period
(µs)
Stationary Sensor
move
horizontally
Sensor
move
vertically
Average
#
motion
detected
slow fast slow fast
25 0 63 184 436 323 252
50 0 45 67 131 324 142
100 0 38 128 74 270 128
150 0 11 131 97 219 115
200 0 19 66 26 127 60
250 0 3 52 46 87 47
IV. IMPLEMENTATION AUTO SWITCHING MODE
Besides capable to process motion signal effectively, the
tag must also be able to send data efficiently to the reader. The
RTF mode is implemented during tags collection process
which executed at the beginning of the system activation.
Then the TTF mode is activated for tag to send subsequent
data to reader. The method of switching from one mode to
another can be many ways. For example, active RFID tag
listens to RFID reader broadcast command before switching
from RTF to TTF mode [5]. Meanwhile, in [6] design, the
RFID tag is switchable between TTF to RTF mode within a
switching time frame based on RFID tag’s transceiver
configuration. In this project, the mode switching method is
controlled two interrupt sources; real-time clock alarm and
motion sensor of RFID tag. During initial cycle, the system is
programmed to communicate in RTF mode to identify all tags
within reader read range. After tag identifies itself to the
reader, it will automatically turn to sleep mode. If the interrupt
source either from the real-time clock or motion event
triggered the tag to wake up, the tag will automatically switch
from RTF to TTF mode to send data to the reader. The
implementation of auto switching mode from RTF to TTF is
described in Fig. 6.
Fig. 6. Implementation scheme of RTF and TTF modes
V. FIELD TEST RESULTS AND DISCUSSION
A. Testing for motion detection efficiency
The motion detection efficiency was tested by moving the
laptop with the tag attach on it. The tag is properly sealed to
the laptop to ensure it will detect the laptop’s movements
rather than itself. The proper setup is crucial for anti-theft
device to function effectively. The arrangement for the test is
shown in Fig. 7. The laptop was moved up and down for 50
times. Once the tag senses the movement, the alert signal will
be transmitted to the reader. The information is displayed on
the Hyper Terminal as shown in Fig. 8. Upon successful
transmission, the tag transmits information which contains the
reader identification, tag identification, asset identification,
motion status and RSSI readout (dBm). The reader processes
the received data, add timestamp to the data and send them to
2012 IEEE International Conference on Control System, Computing and Engineering, 23 - 25 Nov. 2012, Penang, Malaysia
454

the host computer. The RSSI is negative integer expressing in
–dBm and is a measurement of signal strength. This value is
used to estimate the distance between the tag and the reader.
The number of detected motion is recorded based on the result
displayed on Hyper Terminal screen.
The efficiency of tag to detect motion is calculated by
dividing total number of received data with total number of
occurrences times a hundred. The result obtained shows that
the tag can detect all the movement created. It is learned that
the tag must properly attach to the asset in order to avoid false
alarm so that the motion status sent is belong to the asset and
not the tag.
The result shows that the proposed RFID-based anti-theft
system is reliable to detect motion of tagged objects
efficiently. It can be seen from the timestamp information
shown in Fig. 8 that the RFID tag is reliable to be used in
anti-theft application as the user or monitoring station could
receive motion alert signal at less than 1 second from the
moment the asset set in motion.
Fig. 7. The experimental arrangement for tag accuracy test in detecting
motion of tagged object
Fig. 8. Information received from tag
2012 IEEE International Conference on Control System, Computing and Engineering, 23 - 25 Nov. 2012, Penang, Malaysia
455

B. Testing for RF Communication Performance in
Multi-Floored Build
It is often desirable to use anti-theft RFID system in the
confined areas such as within multi-floored buildings to
ensure the asset can be monitored in all types of environment.
The objective of this experiment is to verify the read range
coverage of the proposed RFID reader in multi-floored
building that can detect the movement of tagged-asset. The
experiment is conducted in four levels building.
The test bed arrangement is shown in Fig. 9. The reader
antenna is positioned at vertical polarization. In order to find
the maximum read range, the reader has to be placed at the
highest floor, L1. The measurement setup is repeated at the
reader position L2, L3 and L4. The reader was tested with
Whip and Vivaldi antenna.
Fig. 9. The experimental arrangement for multi-floored read range
measurement
The angle is calculated using tangent function which equal
to the opposite length, h divide the adjacent length, δ. From
the experiment, it is found that the reader has longer read
range when the tag is in the same level. As the tag moves to
higher level or lower level, the received signal deteriorates.
The reader with Whip antenna cannot receive the tag’s data
when the tag is directly above it. The measured distance
versus angle is plotted as shown in Fig. 10 and Fig. 11. Zero
degree angle represents the tag location is directly above the
reader. 90° and 270° angle represent the tag is in the same
level as the reader. By using Whip antenna to the reader, the
tag can be detected in omni-directional but has smaller read
range as compare to Vivaldi antenna. Vivaldi antenna has
better read range but only focus at one direction.
This experiment is useful to optimize the placement of the
readers and the number of readers required in the building.
The reader that has long read range could minimize the
required number of the readers hence the maintenance cost
can be reduced.
Fig. 10. Measured radiation pattern of Whip antenna as a function of read
range (meter) in multi-floored building
Fig. 11. Measured radiation pattern of Vivaldi antenna as a function of read
range (meter) in multi-floored building
VI. CONCLUSIONS
The proposed RFID-based anti-theft system is capable to
detect movement of the asset that attached to the RFID tag.
The design of the motion-sensitive RFID tag with interrupt
function is capable to send the motion alert signal to the reader
in real-time. This feature is reliable to be used in anti-theft
system for efficient high-value asset monitoring.
ACKNOWLEDGMENT
The authors would like to thank University Technology
MARA (UiTM), Excellence Fund for sponsoring this work.
REFERENCES
[1] B. Khoo,, Y. Cheng, “Using RFID for Anti-Theft in a Chinese
Electrical Supply Company: A Cost-Benefit Analysis,”
Wireless Telecommunications symposium (WTS), pp. 1-6,
2011.
[2] G. Jayendra, S. Kumarawadu and L. Meegahapola,
“RFID-Based Anti-Theft Auto Security System with an
Immobilizer,” Confererence on Industrial and Information
System, pp. 441-446, 2007.
2012 IEEE International Conference on Control System, Computing and Engineering, 23 - 25 Nov. 2012, Penang, Malaysia
456

[3] C. E. Shahnon, (1998) Communication in the Presence of Noise.
Proceedings of the IEEE. 86(2), 447-457.
[4] M. Moy, D. B. Stewart, “An Engineering Approach to
Determining Sampling Rates for Switches and Sensors in Real
Time Systems,” Symposium on Real-Time Technology and
Applications, pp. 34–35, 2000.
[5] H. L. Van Eeden, (2009) Wireless Communication System. US
Patent 0045923.
[6] H. Haar, K. Bischof, & H. Scharke, (2006) RFID transponder.
WO Patent 109212.
2012 IEEE International Conference on Control System, Computing and Engineering, 23 - 25 Nov. 2012, Penang, Malaysia
457