This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JRFID.2018.2847241,
IEEE Journal of Radio Frequency Identification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 1

Improved traceability solution based on UHF

RFID for cheese production sector
A. Abdelnour, Student Member, IEEE, F. Buchin, D. Kaddour, and S. Tedjini, Senior Member, IEEE

main challenges of RFID tags design for food products

Abstract—The industrial sector of cheese production with
particular quality protection occupies an important position in
the industrial context of the European Union. Compared to
conventional casein labels, UHF RFID tags can be read at greater
distances and can store more information about the product
which represents a more reliable solution for traceability. This
work discusses the main challenges of designing UHF RFID tags
for the identification of dairy products such as cheese wheels in
an industrial environment. Two different realization techniques
are presented in this paper, and the performances of the realized
prototypes were measured and compared using a professional
RFID measurement system and a commercial handheld RFID
reader in a real industrial environment.
traceability and to introduce some design techniques that
could improve the tags performance.
The paper is organized as follows. Section II introduces the
primary difficulties of traceability in an industrial environment
and compares between different traceability systems. Section
III presents some tag antenna design methods that enhance
UHF tag performance. In section IV, tags read ranges were
measured and compared using VoyanticTagformance
measurement system and a handheld RFID reader in a real
industrial environment. Finally, section V draws some
conclusions.

Index Terms—Antenna design, Food traceability, Smart labels, II. MAIN DRAWBACKS OF CONVENTIONAL
UHF RFID TRACEABILITY TECHNIQUES

A. Casein labels and marks of origin

I. INTRODUCTION
Nowadays, in order to mark the production information on
F ood safety has been a growing concern over the last
decades. The industrial sector of cheese production with
specific quality protection occupies a relevant position in
cheese wheels, the sector of cheese production uses casein
plaques and particular types of stamps. This information is of
the industrial context of the European Union. Moreover, the great importance for the cheese making industry; it indicates
laws regulating the food industry in Europe establish that “the the product’s country of origin, region, cheese type and date of
food traceability in all the production, transformation and fabrication as well as quality and certification markers.
distribution stages must be ensured” [1][2]. Throughout the cheese ripening, these marks can be affected
due to the environmental conditions such as temperature,
In order to enhance the competitiveness in the industrial
cheese sector, technological innovations increasing the moisture, product handling (turning) and shape evolution
production process automation and quality control should be which can reduce the readability of these labels, as shown in
achieved. In this context, RFID (Radio Frequency figure 1. It is for this reason that the production and quality
Identification) technology represents such an efficient and control are for the most part performed by bunches and the
effective system that transmits accurate, complete, and information stockpiling is done manually by the company
consistent information about products through the supply personnel [5].
chain, reducing thus operating costs and increasing
productivity [3]. RFID is currently considered as the worthy
successor of the barcode with a foreseen expansion not only in
the agrifood sector but also in many industrial sectors for
environmental monitoring (temperature, relative humidity and
luminosity) through namely WSN (Wireless Sensor Network).
Therefore, RFID smart labeling can be an attractive alternative
with more features than traditional labels providing thus a
robust traceability solution [4]. This work aims to study the

AbanobAbdelnour, DarineKaddour, SmailTedjini (*Univ. Grenoble Alpes,

Grenoble INP , LCIS, 26000 Valence, France *)
Email:abanob-assaad-amin.abdelnour@lcis.grenoble-inp.fr
Fabrice Buchin (*Ecole Nationale d'Industrie Laitière et des Biotechnologies, Fig. 1. Casein labels used for the cheese identification where the production
39800, Poligny, France) information is unclear due to moisture and polishing.

2469-7281 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JRFID.2018.2847241,
IEEE Journal of Radio Frequency Identification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 2

B. RFID in the cheese industry A. Tag with reflective ground plane

Previous works discussed the benefits of using RFID tags
for the traceability of dairy products [3-5]. Compared to casein
labels, RFID systems present some potential benefits like
efficiency improvement with the increased speed of
operations, increased accuracy of information, minimized
labor cost and reduced inventory losses. However, the
majority of the existing traceability systems used HF tags
which are very limited in reading range (few centimeters) [6].
In an industrial environment, cheese wheels are kept in huge
caves for maturation as shown in figure 2. In such an
environment, tracking each item using HF tags is very time-
consuming making HF RFID not practical for cheese wheels
tracking.
Due to this high loss tangent, a conventional RFID tag will
have poor performance as the cheese surface will act like a
bad ground plane which shorts the tag antenna and reduces its
directivity to the minimum. Designing a tag with a reflective
ground plane, in this case, will be more efficient. Indeed, the
reflective ground plane will separate the antenna from the
cheese surface and on the same time will get the advantage of
the conductivity of cheese surface to act as an infinite ground
plane which will increase the directivity of the tag and hence
its read range.
In this configuration, the primary parameter that affects the
antenna’s electrical performances is the thickness of the
dielectric substrate (h) on which the tag is realized. The more
the tag antenna is separated from the cheese surface, the more
the read range increases. In order to study the thickness impact
on antenna response, a classical dipole meander tag with a
ground plane on an FR4 substrate (𝜀′ 𝑟 = 4.3) is simulated with
the electromagnetic simulation tool HFSS[8]. Three different
thicknesses (h) are considered (1.6mm, 3.2mm and 5 mm).
A conventional configuration [8] composed of a loop
connected to a meander dipole antenna is considered for the
tag antenna in this paper. The detailed dimensions of the
optimized RFID tags are shown in figure 3 where the
maximum tag area does not exceed 1.4 cm x 7.5 cm.

Fig. 2. Thousands of cheese wheels kept in a huge cave for maturation[7]

Therefore, using UHF RFID system, which has a relatively

long read range, is an interesting alternative that will mainly
reduce the time for identification and information storing
during the cheese production process.
Fig. 3. The layout of UHF tag with a ground plane on FR4
III. PASSIVE UHF RFID TAG DESIGNS FOR CHEESE
TRACEABILITY Figure 4 shows a comparison between the three simulated
antenna gains. The highest gain is achieved for h = 5mm, with
The objective is to design a tag for the traceability of cheese
an increase of ≈ 3dB compared to 3.2 mm and ≈ 6 dB
wheels in an industrial environment. The following
compared to 1.6 mm.
subsections present two different topologies of RFID UHF
tags. The first one is realized on a traditionally printed circuit
board, and the second one is directly integrated on the casein
plaque. For design purposes and optimization, RFID tag
antennas were simulated using the 3-D full-wave
electromagnetic software: ANSYS HFSS V-17. The proposed
designs were optimized to match Monza_R6 IC which has a
sensitivity of -20 dBm (10 μW) and is modeled in simulation
as a parallel connection of resistance of 1.80 kΩ and a
capacitance of 1.37 pF. The dielectric model of cheese used
for simulations is based on the average value of measurements
realized on several types of pressed cheese where the average
dielectric constant = 20 and loss tangent = 0.7. These high
losses are explained by the high level of water content inside Fig. 4. 3D model of simulated tag on cheese and a comparison between the
the cheese texture which makes the cheese surface relatively gains of the tag (at phi=90 deg) simulated on FR4 substrates of 3 different
thicknesses
very conductive.

2469-7281 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JRFID.2018.2847241,
IEEE Journal of Radio Frequency Identification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 3

B. Tag on a polypropylene substrate

A French company specialized in the fabrication of labels

for dairy products developed a traceability solution based on
barcodes [9] which are realized on casein and polypropylene
as shown in figure 5. A polypropylene label is easily
recyclable and offers excellent resistance to water and
chemicals which makes it compatible with food product
constraints. This configuration can also be used to realize a
UHF RFID tag where the tag antenna is separated from the
cheese surface by the casein and polypropylene layers, and
therefore the tag performance is increased.
A polypropylene substrate and a casein patch of 1 mm and
0.36 mm thicknesses respectively were considered for the
simulation. The dielectric properties of polypropylene and a b
casein were characterized using a resonant cavity where Fig. 6. a) prototypes realized on FR4 b) Tag on polypropylene and casein
substrate
dielectric constant and loss tangent of polypropylene are 2.74
Indeed, RFID tag read range is defined as the maximum
and 0.005, respectively whereas casein has a higher dielectric
distance at which the tag can be read. It depends on the chip
constant of 5.85 and presents a loss tangent of 0.24.
sensitivity, the tag antenna gain, polarization, and the
The layout of the RFID UHF tag on polypropylene is shown
matching between the antenna and the chip. Using Friis
with detailed dimensions in Figure 5.c where the tag has a
equation, it is possible to calculate the RFID tag’s read range
total area of 4.5 cm x 1.1 cm.
performance as:
𝜆 𝑃𝑒𝑖𝑟𝑝 𝐺𝑡𝑎𝑔 𝑝 1 − 𝑆 2
𝑅𝑒𝑎𝑑𝑅𝑎𝑛𝑔𝑒 = (1)
4𝜋 𝑃𝑡ℎ
where 𝜆 is the wavelength of the carrier emitted by the reader,
𝑃𝑒𝑖𝑟𝑝 is the regulated equivalent isotropic radiated power,𝑃𝑡ℎ
is the activation power of the chip, 𝐺𝑡𝑎𝑔 is the tag antenna
gain, 𝑝 is the polarization loss factor between reader and tag
a b antennas. 𝑆 2 is the power reflection coefficient given by:
2
𝑍𝑐ℎ𝑖𝑝 − 𝑍𝑎𝑛𝑡 ∗
|𝑆|2 = (2)
𝑍𝑐ℎ𝑖𝑝 + 𝑍𝑎𝑛𝑡
Where 𝑍𝑎𝑛𝑡 and𝑍𝑐ℎ𝑖𝑝 are respectively tag antennaand IC
impedances.

c
Fig. 5. a) Barcode on casein and patch of polypropylene [9] b) 3D model of
RFID tag on a substrate of polypropylene c) The layout of the RFID tag on
polypropylene.

IV. RESULTS & DISCUSSION

A. Read range characterization using VoyanticTagformance

The read range performance of the realized tags was

measured using the VoyanticTagformance measurement
system [10]. This system can provide the backscattered signal
strength from the tag under test and thus evaluates its read
range. Figure 6 shows the final prototypes realized on FR4 and
casein label.
Fig. 7. Comparison between the measured read ranges of tags realized on
FR4

2469-7281 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JRFID.2018.2847241,
IEEE Journal of Radio Frequency Identification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 4

Figure 7 shows a comparison between the measured read

ranges of tags realized on FR4. The measurements were The installation of RFID tags comes during the fabrication
achieved on Emmental cheese samples in a frequency range process of cheese after the curdling step where the tag is
between 835 MHz to 935 MHz with a frequency step of placed between the poured curd and the cheese mold. Then,
1 MHz. The maximal read range of the tag fabricated on a the curd is pressed to allow evacuating the excessive
standard 1.6 mm FR4 and having a size of ≈ 10 cm² was proportion of water content, as shown in figure 10.
around 2 meters. The read range increases to reach 3.5 meters
and 4.5 meters for substrates of 3.2 mm and 5 mm thickness
respectively. These read ranges allow automating the process
of identification and offer the possibility for more applications
of data monitoring.
The second configuration fabricated on polypropylene and
casein has a lower read range around 1.3 meters as presented
in figure 8. This smaller read range is explained by the fact
that this configuration has a lower directivity due to the
absence of the reflective ground plane and the lesser distance
between the tag antenna and the cheese surface. Better
performance can be achieved by increasing the thickness of
the polypropylene substrate which will provide better isolation
between the tag antenna and the lossy surface of the cheese
product. a b
Tag on Casein & Polypropylene Fig. 10. Tag installation procedure: a) placing the tag between curd and
1.4 cheese mold b) curd pressing

The realized tags were installed on different types of cheese

1.2 and measurements were achieved in a real industrial
environment using a commercial RFID portable reader as
Read range (m)

shown in figure 11. The reader (Alien H450) used for

1
measurements has a limited output power of 30 dBm, which is
lower than the max power allowed for RFID measurements.
0.8 Therefore the tags’ read ranges were lower by a factor of ≈
0.5 compared to Tagformance measurements. In such an
environment, measurements were also affected by multipath
0.6 due to signal reflections which makes the tags performances
even lower.

0.4
820 840 860 880 900 920 940
Frequency (MHz)
Fig. 8. Measured read range of tag realized on casein and polypropylene

B. Tag installation and measurements in a real environment

According to cheese production legislations and regulations
[11][12], all tags must be encapsulated before installation in
order to maintain a high level of hygiene and to avoid any
contact between the cheese and the tag electronics as shown in
figure 9.

Fig. 11. Measurement in an industrial environment using handheld RFID

reader

A comparison between the averages of measured read range

Fig. 9. Tag encapsulated before installation on cheese
for each tag configuration on three types of pressed cheeses is
presented in figure 11. As the read range increases, the
uncertainty of measurements increases due to multi-path
effect. Better performances can be achieved using a portable
reader with higher output power.

2469-7281 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
information.
This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI 10.1109/JRFID.2018.2847241,
IEEE Journal of Radio Frequency Identification
> REPLACE THIS LINE WITH YOUR PAPER IDENTIFICATION NUMBER (DOUBLE-CLICK HERE TO EDIT) < 5

products in cheese industries,” in RFID Eurasia, 2007 1st Annual, 2007,

pp. 1–5.
[6] P. Papetti, C. Costa, F. Antonucci, S. Figorilli, S. Solaini, and P.
Menesatti, “A RFID web-based infotracing system for the artisanal
Italian cheese quality traceability,” Food Control, vol. 27, no. 1, pp.
234–241, Sep. 2012.
[7] “Comté: How the Largest Small-Batch Cheese in France Is Made |
Serious Eats.” [Online]. Available:
http://www.seriouseats.com/2011/07/comte-how-largest-small-batch-
cheese-is-made-in-france.html. [Accessed: 08-Sep-2017].
[8] G. Marrocco, “The art of UHF RFID antenna design: Impedance-
matching and size-reduction techniques,” IEEE Antennas Propag. Mag.,
vol. 50, no. 1, 2008.
[9] J. P. Maitre, “Food product e.g. cheese, information medium, has label
having two slots that are positioned so that main part of slots constitutes
Fig. 12. Comparison between tags’ read ranges (in meters) in an industrial central part of label, where side holding flanges are off-set on both sides
environment using a portable reader of main part,” FR2884955 (A1), 27-Oct-2006.
[10] “Tagformance Pro,” Voyantic. [Online]. Available:
http://voyantic.com/products/tagformance-pro. [Accessed: 19-Oct-2017].
[11] “EUR-Lex - 32003R1830 - EN - EUR-Lex.” [Online]. Available:
V. CONCLUSION http://eur-lex.europa.eu/legal-
content/en/ALL/?uri=CELEX%3A32003R1830. [Accessed: 20-Feb-
2018].
This work discussed the main drawbacks of currently used [12] Council Regulation (EC) No 2991/94 of 5 December 1994 laying down
labels for the traceability in the cheese production sector. standards for spreadable fats, vol. 316. 1994.
Compared to standard casein labels and stamps, UHF RFID
tags represent an interesting alternative that can store and
update more data during all the production chain as well as
achieve longer read distances. Two different design
configurations of UHF RFID tags were presented in the paper.
The tag installation procedure during cheese fabrication was
explained. Comparison between tags read ranges was realized
using a professional RFID measurement system and with a
handheld RFID reader in the real industrial environment.
Measurement results proved the possibility of achieving read
ranges of several meters; which open a window for more
applications of the data monitoring and wireless sensing in
dairy products industry.

ACKNOWLEDGMENT

Part of this work has been supported by "Région Auvergne

Rhône-Alpes" within the partnership of project AFFINID.

REFERENCES

[1] A. Regattieri, M. Gamberi, and R. Manzini, “Traceability of food

products: General framework and experimental evidence,” J. Food Eng.,
vol. 81, no. 2, pp. 347–356, Jul. 2007.
[2] F. Dabbene, P. Gay, and C. Tortia, “Traceability issues in food supply
chain management: A review,” Biosyst. Eng., vol. 120, pp. 65–80, Apr.
2014.
[3] L. Kumari, K. Narsaiah, M. K. Grewal, and R. K. Anurag, “Application
of RFID in agri-food sector,” Trends Food Sci. Technol., vol. 43, no. 2,
pp. 144–161, Jun. 2015.
[4] F. Bibi, C. Guillaume, N. Gontard, and B. Sorli, “A review: RFID
technology having sensing aptitudes for food industry and their
contribution to tracking and monitoring of food products,” Trends Food
Sci. Technol., vol. 62, pp. 91–103, Apr. 2017.
[5] R. Perez-Aloe, J. M. Valverde, A. Lara, J. M. Carrillo, I. Roa, and J.
Gonzalez, “Application of RFID tags for the overall traceability of

2469-7281 (c) 2018 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more
information.