An RFID-enabled Library Management System using Low-
SAR Smart Bookshelves
I. Markakis1 T. Samaras2
Abstract − An intelligent Library Management System (LMS)
based on the emerging UHF passive RFID technology is
currently being designed in order to replace a traditional
barcode system at the university’s library. Primary objectives
of the design include maximization of tag readability,
localization of tagged items in the smart bookshelf, reduced
spill-over energy to nearby shelves, and minimization of the
Specific Absorption Rate (SAR) to nearby library users.
Different types of shelf antennas were used and tested
including evanescent-type antennas as well as far-field
antennas. The antennas were designed, simulated, built, and
tested in a realistic environment. HFSS and SEMCAD-X were
used to evaluate radiation parameters and illustrate the
radiation effects of near- and far-field antennas in the
presence of the cabinet and library users.
1 INTRODUCTION
Radio Frequency Identification (RFID) is a growing
technology with a plethora of useful applications in
major sectors of the economy ranging from the
healthcare [1] and pharmaceutical industry to
retailing, transportation and logistics [2]. The project
at hand is dealing with the design, implementation,
and deployment of a pilot RFID system for the
university’s library. The main idea is to use the
passive RFID technology at UHF frequencies in
order to develop an intelligent low-cost Library
Management System (LMS) that reduces labor and
running cost and improves productivity at workplace
[3]. The general objectives of the ongoing project
include: a) replacement of traditional time-consuming
processes for checking books in and out of the
library; b) automation of labor-intensive and time-
consuming stock-taking processes; c) improvement
of productivity at workplace by eliminating tedious
and paper-bound processes traditionally followed
during inventory taking; d) reduction of library
running cost due to a more productive and highly
efficient working environment; e) reduction of thefts
and of the number of missing or misplaced items; etc.
In order to achieve the aforementioned general
objectives of the project, the research team has
undertaken the task to design and build: a) a smart
RFID bookshelf interfaced with computer technology
via a Graphical User Interface (GUI) which is user-
friendly, portable, and versatile; b) a check-out
counter where the library user can check out books
through the use of an RFID-enabled system which is
fully interfaced with the existing library database
system; c) a book-return counter or drop box where a
________________________________________________________________________________________
1
Department of Physics, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greecİ, e-mail: jmarkakis@physics.auth.gr, tel. & fax:
+302310998069
2
Department of Physics, Aristotle University of Thessaloniki, 54124 Thessaloniki, Greecİ, e-mail: theosama@auth.gr, tel.: +30 2310998232
3
Department of Electrical & Computer Engineering, University of Nicosia, 46 Makedonitissas Ave., 1700 Nicosia, Cyprus, e-mail:
polycarpou.a@unic.ac.cy, tel.: +35722842514, fax: +35722367481.
4
Department of Electrical & Computer Engineering, University of Nicosia, 46 Makedonitissas Ave., 1700 Nicosia, Cyprus, e-mail:
sahalos.j@unic.ac.cy, tel.: +35722841740, fax: +35722367481.
978-1-4673-5707-4/13/$31.00 ©2013 IEEE
227
Authorized licensed use limited to: INSTITUTO FEDERAL DE SAO PAULO. Downloaded on April 25,2023 at 23:06:57 UTC from IEEE Xplore. Restrictions apply.
A. C. Polycarpou3 J. N. Sahalos4
library user is able to return books without the help
of a librarian; d) a portal RFID-enable system which
is activated when a non-properly checked-out book is
about to exit the library entrance.
In this paper, we focus our efforts primarily on
the design and evaluation of a smart cabinet
developed using existing state-of-the-art passive UHF
RFID systems interfaced with computer technology
(laptops, tablets, and smart-phones) via a user-
friendly and portable GUI. A properly designed
antenna is installed on each shelf of the cabinet
allowing maximum readability of RFID tags placed
somewhere in the interior of the books. Either a far-
field or a near-field antenna can be used. A far-field
antenna has a larger range allowing reading of tags
from a relatively large distance (4-7 m) depending on
the tag type, reader’s sensitivity, antenna input
power, and environment. A near-field antenna has a
shorter range – measured in tens of centimeters –
thus allowing readability of tags in the near vicinity
of the antenna. Design of far-field antennas is easy
and straight-forward; however, there is often
significant spill-over of energy to nearby shelves
creating tag localization problems and undesirable
readings. In addition, it generates excessive radiation
toward unsuspected library users in the vicinity of the
shelves. On the contrary, near-field antennas are
extremely difficult to design and have a very short
readability range creating tag identification problems
of books on the shelves. In the current work, we
decided to test two well-known and previously
published antennas – a near-field microstrip line and
a far-field rectangular patch antenna – in terms of
coverage on the shelf as well as in terms of the
Specific Absorption Rate (SAR) for nearby library
users. An optimum solution is the one that provides
adequate coverage and low SAR readings.
Section 2 presents the underlying problem setup.
Section 3 presents the simulation results obtained
using SEMCAD-X and HFSS and a discussion of the
results. Section 4 provides concluding remarks on the
findings of the current research work.
2 PROPOSED RFID-ENABLED LMS
2.1 Problem Setup
In this paper, we will focus primarily on the radiation
characteristics of the smart cabinet assuming certain
 types of shelf antennas. A tyypical cabinet is
illustrated in Figure 1. Each of thesse shelves consists
of a shelf antenna positioned eitheer at the bottom of
the shelf, the back of the shelf, or even the side-wall
of the shelf. Such antennas aree either near-field
antennas or far-field antennas. Thhe objective is to
allow 100% readability of the bookks in the particular
shelf without reading other boooks from nearby
shelves.
Another objective is to avoiid excessive radiation
toward the library user due to spill-over energy. For
this reason, we considered to examine two types of
antennas: a) a simple microstriip line terminated by a
matched load which was recentlyr proposed by
Medeiros et al [4], and b) a linearly polarized far-
field patch antenna designedd for the UHF RFID
frequency band. Both these anttennas are illustrated in
Figure 2.
The microstrip line generaates an evanescent field
in the transverse plane wheereas the voltage and
current waves are propagatinng undisturbed in the
longitudinal direction. This tyype of antenna can be
placed either on the bottom off the shelf, as shown in
Figure 3(a), or at the back of the shelf, as shown in
Figure 3(b).

(a) (b)
Figure 1: CAD model of a typical smart cabinet consisting of
six shelf patch antennas mounted on o the side walls
(without (a) and with boo
oks (b)).

(a) (b)
Figure 3: (a) Microstrip line mountted on the bottom of the
Foam ( ε r  1 ) h shelf; (b) Microstrip line mounted
d at the back of the shelf.

50 Foam ( ε r  1 ) h The patch antenna printed on the FR-4 substrate,

shown in Figure 2(b), was deesigned as an isolated
radiated element using the HFS SS. The corresponding
(a) return loss as a function of freqquency is illustrated in
Figure 4. As observed, the prinnted patch antenna has
a good return loss and impedaance bandwidth within
the RFID band of interest. A maximum
m realized gain
Dx of approximate 1.65 dBi iss obtained, which is
Wy considered satisfactory bearinng in mind that the
Dy substrate is highly lossy. This type of antenna is
installed on the side-walls of the shelf, as shown in
Wx Figure 1(a).
In the following section, SEEMCAD X (Schmid &
Partner Engineering AG, Zurich, Switzerland) will be
(b)
used to test the field distributtion inside the cabinet
Figure 2: (a) Microstrip line (width==5 cm, h = 1 cm)
for the different mounting possitions of the antennas
terminated by a 50-ȍ load; (b) Linearrly polarized patch
proposed in here. In addition,, SEMCAD X will be
antenna printed on an FR-4 substraate ( ε r = 4.55 , used to evaluate the radiationn effects of these two
tan δ = 0.0175 , Wx = 110 mm, Wy = 79.5 mm, Dx = 55 antennas on library users who happen to be standing
mm, D y = 14 mm)). in the vicinity of the smart cabinet.
c The numerical
results of the patch antenna were
w validated against
those obtained by HFSS and theret was a very good

228
Authorized licensed use limited to: INSTITUTO FEDERAL DE SAO PAULO. Downloaded on April 25,2023 at 23:06:57 UTC from IEEE Xplore. Restrictions apply.
 agreement with less than 0.05dB
B deviation in the Table 1: Relative value of maaximum electric field
antenna gain. observed in non-scannedd nearby shelves
Spill-over E-field
2.2 Numerical human models Antenna type/case
(dB)
Patch antenna -28.5
Three human models from the virttual population [5]
Patch antenna
were used in the simulations. The models were both -29.5
(no books)
male and female and of different age
a and body mass
index (BMI). The resolution of thee models was 1mm Microstrip horizontal -30.6
× 1mm × 1mm. Specifically, the models used were Microstrip horizontal (no bookks) -38.9
Duke (male 34 years old), Ella (femmale 26 years old) Microstrip vertical -31.9
at 7 months gestational age and Thelonious
T (male 6 Microstrip vertical (no books) -38.8
years old). Furthermore, the dielectric properties of
the tissues were calculated according
a to the
parametric model proposed by Gabriel
G et al. [6],
taking the approach of Dimbylow w [7] for pregnant
woman tissues.
Two different simulation scenaarios were studied.
In the first scenario the human model
m is facing the
cabinet at a distance of 10cm, wheereas in the second
case the human model is nearlyy touching (5mm
away) the cabinet with its arm. Simulations were
performed with or without the preseence of books.
0.00

-5.00

-10.00

-15.00
dB(S(1,1))

-20.00
(a) (bb)
-25.00 Figure 5: Side view of the cabinet with
w the microstrip (a) at
the back of the shelves; (b) on thee bottom of the shelves.
-30.00

-35.00
850.00 855.00 860.00 865.00 87
70.00 875.00 880.00
Freq [MHz]
Figure 4: Return loss of patch antennaa versus frequency.

3 SIMULATION RESULTS

The evaluation of the library systemm proposed in this

paper is based on two factors. Firstlly, we examine the
spill-over of the energy to the neearby shelves and
secondly the maximum exposure to human models
from the antennas.

3.1 Spill-over electric field

In order to assess the spill-over, wee have compared in
dB the maximum E-field in nearbyy shelves with the
maximum E-field in the space off scanned shelves.
Table 1 contains this comparison foor each antenna.
In Figures 5(a) and 5(b) the spill-over is Figure 6: Electrical field distribution in the vicinity of the
presented in the case of the microsttrip antenna placed cabinet in the presence off a human model.
at the back and on the bottom m of the shelves,
3.2 Peak spatial SAR10g
respectively.
Furthermore, in Figure 6 the E-field
E distribution For the worst case exposure, thhe value of peak spatial
in the proximity of the cabinet in the presence of a SAR averaged over 10g of human tissues mass
human model (Duke) is shown. (denoted as “psSAR10g”) haas been calculated. In

229
Authorized licensed use limited to: INSTITUTO FEDERAL DE SAO PAULO. Downloaded on April 25,2023 at 23:06:57 UTC from IEEE Xplore. Restrictions apply.
 Table 2 psSAR10g for each simulation case is
presented.
4 CONCLUSIONS
In this work, we have shown that it is possible to
design a smart library cabinet taking into account
both its efficiency in books scanning, as well as its
compliance with environmental factors, like human
exposure.
The numerical results indicate that using a
microstrip antenna placed at the back of the cabinet
shelves results in reduced spill-over electric field in
the space of non-scanned shelves. This configuration
is also the second best in terms of energy absorbed in
the body of library users, whereas the patch antenna
leads to the lowest exposure. However, the values of
psSAR10g are significantly lower than those in
exposure guidelines (2W/kg), so that choosing the
configuration of vertical microstrip antenna can be
justified.

Table 2: psSAR10g for each simulation case

distance books psSAR10g (W/kg)

model name
10cm 5mm present absent microstrip vertical microstrip horizontal patch
x x 0.030 0.121 0.024
x x 0.026 0.027 0.014
Duke
x x 0.004 0.007 0.002
x x 0.005 0.009 0.014
x x 0.040 0.085 0.039
x x 0.012 0.015 0.026
Ella 7 months
x x 0.012 0.016 0.006
x x 0.010 0.011 0.013
x x 0.041 0.146 0.025
x x 0.009 0.014 0.021
Thelonious
x x 0.009 0.009 0.011
x x 0.010 0.012 0.025

Acknowledgments [4] C. R. Medeiros, J. R. Costa, and C. A. Fernandes,

This project is funded by the Cyprus Research
Promotion Foundation (RPF) under the grant with
protocol number ȉȆǼ/ȅȇǿǽȅ/0311(ǺǿǼ)/03.
References
[1] A. C. Polycarpou, A. Dimitriou, A. Bletsas, P. C.
Polycarpou, L. Papaloizou, G. Gregoriou, and J.
N. Sahalos, “On the Design, Installation, and
Evaluation of a Radio-Frequency Identification
System for Healthcare Applications,” IEEE
Antennas & Propagation Magazine, Vol. 54, No.
4, pp. 255-271, 2012.
[2] T. Y. Peng, “Study on the Application of RFID
Technology in Retail Logistics Industry,” IEEE
3rd International Conference on Communication
Software and Networks (ICCSN), pp. 396-400,
2011.
[3] M. G. Linquist, “RFID in Libraries – Introduction
to the Issues,” World Library and Information
Congress: 69th IFLA General Conference and
Council, Berlin, Aug. 2003.
“RFID Reader Antennas for Tag Detection in
Self-Confined Volumes at UHF,” IEEE Antennas
and Propagation Magazine, Vol. 53, No. 2, pp.
39-50, April 2011.
[5] A. Christ, W. Kainz, E. Hahn, K. Honegger, M.
Zefferer, E. Neufeld, W. Rascher, R. Janka, W.
Bautz, J. Chen, B. Kiefer, P. Schmitt, H-P.
Hollenbach, J. Shen, M. Oberle, D. Szczerba, A.
Kam, J. Guag, and N. Kuster, “The Virtual
Family – development of surface-based
anatomical models of two adults and two children
for dosimetric simulations,” Physics in Medicine
and Biology, Vol. 55, No. 2, pp. 23-38, Jan. 2010.
[6] S. Gabriel, R. W. Lau, and C. Gabriel, “The
dielectric properties of biological tissues: III.
Parametric models for the dielectric spectrum of
tissues,” Physics in Medicine and Biology, Vol.
41, No. 11, pp. 2271-2293, Nov. 1996.
[7] P. Dimbylow, “SAR in the mother and foetus for
RF plane wave irradiation,” Physics in Medicine
and Biology, Vol. 52, No. 13, pp. 3791-3802, July
2007.

230
Authorized licensed use limited to: INSTITUTO FEDERAL DE SAO PAULO. Downloaded on April 25,2023 at 23:06:57 UTC from IEEE Xplore. Restrictions apply.