Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
VOLUME XX, XXXX 1
Received XX Month XXXX; revised X Month XXXX; accepted XX Month XXXX. Date of publication XX Month XXXX; date of current version XX Month XXXX.
Digital Object Identifier 10.1109/OAJPE.2020.2976889
Preliminary Design and Test of a Microwave Inline
Moisture Sensor for the Carasau Bread Industry
GIACOMO MUNTONI1, MEMBER, IEEE, MATTEO B. LODI1, MEMBER, IEEE, ALESSANDRO FEDELI2,
MEMBER, IEEE, ANDREA MELIS1, CLAUDIA MACCIÒ1, MATTEO PASTORINO2, FELLOW, IEEE, ANDREA
RANDAZZO2, SENIOR MEMBER, IEEE, GIUSEPPE MAZZARELLA1, SENIOR MEMBER, IEEE,
ALESSANDRO FANTI1, SENIOR MEMBER, IEEE
1Department of Electric and Electronic Engineering, University of Cagliari, Cagliari, 09123, Italy
2Department of Electrical, Electronic, Telecommunications Engineering and Naval Architecture, University of Genoa, Genoa, 16145, Italy
CORRESPONDING AUTHOR: Alessandro Fanti (e-mail: alessandro.fanti@unica.it).
This work was partially supported in part by tMinistero dello Sviluppo Economico, in AGRIFOOD Programma Operativo Nazionale (PON) Imprese e Competitività
(I&C) 2014-2020, through the Project ‘‘Ingegnerizzazione e Automazione del Processo di Produzione Tradizionale del Pane Carasau mediante l’utilizzo di tecnologie
IoT (IAPC)’’ under Grant CUP: B21B19000640008 COR: 1406652. This work was partially supported in part by Italian Ministry of Enterprises and Made in Italy
(MIMIT), within the “ACCORDI PER L’INNOVAZIONE” (2021–2026), through the Project AISAC under Grant CUP: B29J23001120005 – COR: 1607797.
ABSTRACT Within the framework of the recent agri-food technological advancement, the design and
validation of a methodology for the water content estimation in the Carasau bread manufacturing process is
herein presented. Following a thorough evaluation of the dough dielectric properties, a suitable antenna layout
has been selected, pointing out the advantages in the choice of a contactless narrow-band antenna in
comparison to wide-band and dual-band ones. The presented simulated results are then validated using a
prototype sensor and an ad hoc measurement system to confirm the antenna ability to discriminate among
doughs with different water content. In addition, an accurate analysis of possible sources of misinterpretation
of the results is presented.
INDEX TERMS dielectric properties, moisture sensor, Carasau bread, patch antenna, food engineering
I. INTRODUCTION
Within the branch of microwave non-communication
applications, the agri-food sector is witnessing a high-paced
technological advancement. Through microwave technology,
wireless sensors are proficiently employed for food dielectric
characterization [1], grains and fruit quality inspection [2]–[4],
adulteration and contamination detection [5]–[7], and
traceability [8], [9]. The convenience of automating such
processes also affects productions that are intrinsically
antithetical to large-scale industrial advancements, like
traditional bakery. Traditional food is the backbone of Italian
cuisine, and it is so rooted in its history that each region boasts
countless typical products. One example is the Carasau bread,
a thin, crunchy traditional Italian bread made in Sardinia [10].
The Carasau bread industry has already started a technological
advancement. Indeed, recently, a novel wireless sensor
network able to monitor several parameters (i.e., ambient
temperature, relative air humidity, CO and CO2 concentration,
speed of the conveyor belt, morphology and texture of the
bread) at every production step [11], and a blockchain-based
traceability system [12] have been developed. Nonetheless, a
dedicated sensor or a disruptive methodology for quality
inspection in early production stages is still missing.
Carasau bread is made from re-milled semolina of durum
wheat, sea salt, natural yeast, and de-chlorinated drinking
water [13]. The production chain includes kneading,
leavening, sheeting, and baking stages, which will be better
explained in Section II. It is worth noting that one of the most
important passages in the bread production is between the
leavening and baking stages, wherein the dough must have the
proper water content () to secure a successful baking [14],
[15]. Otherwise, the final product could be defected, rising the
overall production cost to compensate the waste. This analysis
is the result of a survey in a real Carasau bakery located in
Fonni (NU), central Sardinia, Italy. As an example, to justify
the urge of a reliable sensing system, the latter bakery reports
an amount of wasted product, due to inaccuracy of water
content in the dough, equal to 300 kg per day, which is equal
to about 12% of the total production. With this industrial
stakeholder, the following system requirements have been set:
i) cost-effectiveness, ii) easy deployment, iii) be able to sense
water content variation between 46% and 54%, in a robust and
sensitive way. To date, in the literature, rheological [16],
nuclear magnetic resonance [17] and Fourier transform
infrared (FTIR) measurements [18] were proposed as
methodologies for assessing the quality and composition of
Carasau bread doughs. However, these quantitative
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
approaches require expensive and cumbersome equipment,
with a non-trivial result interpretation (calling for expert
users), and demand a specific sample preparation, thus being
not suitable for in-line inspection. This is why, recently,
broadband dielectric spectroscopy in the range 1 Hz-10 MHz,
performed at cryogenic temperatures, was investigated as a
potential alternative methodology [19]. However, common
devices and tools used for dielectric spectroscopy (e.g., open-
ended coaxial probe, resonant cavities, etc.) cannot be used in
industrial settings, requiring contact with sample, being
invasive or determining a potential contamination of the food
product. Actually, despite being somehow a niche topic,
already published literature dealt with general improvement of
the Carasau production chain but with no dielectric
characterization of the product [10], [11], quantitative
methodologies for quality and dielectric composition
assessment of the dough but very expensive and more suitable
for a laboratory than an industrial production line [16]–[18],
dielectric characterization of the dough that, however, cannot
be applied in an inline scenario [19], and dielectric
characterization of the sheeted dough based on the amount of
water that cannot rely on real measurement to support its
conclusions [13]. This latter aspect needs further
investigations.
In this perspective, it is clear the necessity to develop a
robust, cost-effective and electromagnetic-based (contactless)
methodology for measuring the water content in the
production chain between the leavening and baking steps, so a
suitable device that can be easily integrated in the proximity
of the conveyor belt. Such methodology, unprecedented in the
open scientific literature due to its specific product-related
nature, should consist of an antenna sensor that, in a
contactless way, is able to sense a difference in permittivity
among doughs with different water content. The antenna is
assumed to be located in a fixed position, taking advantage of
the movement of the conveyor belt, to analyze every unbaked
bread sheet being transported. This system takes advantage of
the modification of the antenna resonance due to the different
permittivity of the dough, which depends on the water content.
Microwave sensors based on microstrip technology are
amongst the most employed for permittivity [20], and
humidity/moisture detection, preferable, for instance, to
destructive and invasive methods such as the coaxial probe
[21]. They can be employed in several fields of application,
such as humidity evaluation in soils [22]–[26], and in food and
food-related products [2], [3], [27]–[29]. Many examples are
reported, in the open literature, of water content sensors
relying on microstrip patch antennas for their low cost and low
profile [25], [30]–[32]. However, most of these sensors
operate in direct contact with the Material Under Test (MUT),
a solution that must be avoided in the presented scenario. In
the case of the Carasau bread, a suitable inline microwave
sensor should be taken into account to be fully integrated into
the production chain, maintaining a proper clearance between
the antenna and the bread sheet, and acquiring the moisture
data in a non-destructive way.
In this work, the design, numerical simulations, and
validation of a new methodology for a contactless inline
microwave sensing of Carasau bread sheet moisture is
presented. Aim of the paper is to assess what type of antenna
layout is the best choice for the presented measurement
scenario and satisfy the industrial requirements. In particular,
we focused on the sensor capability of discriminating small
percentages variation (~4%) of water content inside the
Carasau dough. A thorough numerical investigation had to be
carried out to understand if the existing difference in
permittivity was even measurable.
We selected some representative test cases, agreed with our
industrial stakeholders and the specific needs of this
production. The first phase of the study has been focused on
the measurement of the complex dielectric permittivity of the
dough with different (46%, 50%, and 54% w.r.t. the total
weight of the semolina). A commercial open-ended coaxial
probe has been used for the measurements. However, since the
probe cannot be used for inline measurement and is an
invasive and destructive method, a suitable microwave sensor
has to be placed between the second leavening and baking
steps (see Fig. 1 for reference and consult Section 3 of the
supplementary material). Thus, the second phase included the
setup of the simulation environment and the choice of the
antenna layout. Based on a direct comparison between wide-
band, dual-band, and narrow-band antennas, it has been found
that only the latter offer a fair discrimination of the moisture
variation. A simple, but effective, coaxial-fed microstrip patch
antenna has been chosen, accordingly. The selection of the
operating frequency has been dictated by the small-scale
industrial environment (which limited the available
bandwidths to the ISM frequencies, according to European
directive 2006/42/CE) and by the intrinsic dielectric
characteristics of the dough. Therefore the 5.8 GHz ISM band
has been chosen.
The simulations results highlight that a simple patch
antenna close enough to the bread sheet can discriminate the
based on the magnitude difference of the S11 minima. To
endorse this statement, a set of measurements has been
performed using an ad hoc setup. Despite the discrepancy
between the simplistic numerical environment (preferred for a
light computational load) and the dedicated measurement
setup, a good agreement has been found in the comparison. In
addition to the validation of the methodology, a further study
has been conducted to assess the robustness of the presented
system to factors unrelated to the variation of water content.
These factors include geometrical variation to the envisioned
scenario, such as changes in the antenna-sheet distance and
changes in the sheet thickness.
The paper is organized as follows:
- Section II is devoted to the explanation of the Carasau
bread manufacturing process, highlighting its key aspects and
the dough ingredients.
- Section III describes the characterization of the Carasau
dough using a commercial coaxial probe to derive the general
dielectric profile to be used in the simulations, as well as
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
retrieving how the dielectric properties vary with water
content.
- Section IV delves into the choice of the antenna layout
suitable for the contactless characterization of the sheeted
dough and presents the electromagnetic simulations of the
coupling between the chosen antenna and the sheeted dough
with different water content.
- Section V outlines the fabrication of the prototype used for
the measurements, describes the measurement setup used, and
shows the comparison between simulated and measured
results.
- Section VI warns about potential causes of
misinterpretation of the results and how to recognize them
with further comparisons between simulations and
measurements.
- Section VII draws the conclusion of the work and suggests
a possible employment of the data retrieved in this study.
FIGURE 1. Schematic representation of the Carasau bread baking
process.
II. CARASAU BREAD MANUFACTURING PROCESS
The production process of the Carasau bread starts from raw
materials: re-milled semolina of durum wheat, sea salt, natural
yeast, and de-chlorinated tap water. The ingredients are mixed
inside a kneading machine and converted into dough that has
to leaven at a temperature of 28-32°C [10]–[18]. Next, the
dough is sheeted by a machine and disks of 36 cm in diameter
are obtained (actually, the disks are not perfectly circular, as it
is pointed out in Section IV). Then, the disks are transported
in a dedicated leavening room by the conveyor belt, wherein
they undergo a second session of leavening. It follows a first
baking at 570 °C in an oven, after which the disks are left to
cool for around 5 minutes and then separated manually by
trained operators to obtain two sheets. Lastly, the latter are
baked again at 400 °C to obtain the final product. A schematic
representation of the Carasau bread baking process is reported
in Fig. 1.
a)
b)
FIGURE 2. Preliminary measurements of the real (a) and
imaginary (b) part of the complex dielectric permittivity for the
Carasau bread dough with different water content.
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
III. COMPLEX DIELECTRIC PERMITTIVITY OF
CARASAU BREAD DOUGH AND SHEET
The methodology for obtaining the relative complex
dielectric permittivity, , of the Carasau dough
was re-adapted from [19]. These measurements have been
performed primarily in order to model the sheeted dough
within the CST simulation environment. The measurement
setup included the commercial DAK 3.5 Speag open-ended
dielectric probe connected to the Rhode & Schwarz vector
network analyzer (VNA) VN8. The frequency range has been
set between 0.5 and 8 GHz, using a 50 MHz span. The
measurements results are presented as the average value of 10
measurements, considering the standard deviation and the
combined standard deviation. They have been performed on
small dough samples inserted in suitable sample holders.
The standard recipe for the Carasau bread dough consists
of 4.5 g of salt, 4.5 g of yeast, 300 g of semolina, and 150 g of
water (50% of w.r.t. the total weight of the semolina). In
order to mimic the excess or defect of water typically
encountered during the bakery operations [10]–[18], the recipe
has been modified by altering the percentage of water to 54%
and 46%, respectively. After the required steps of kneading
(20 min) and first leavening (40 min), part of the dough has
been used to create the small samples for the probe
measurements, whereas the remaining part of the dough has
been manually sheeted using a pasta machine, reaching the
desired thickness of about 1 mm, to match the real industrial
sheets. According to the preliminary measurements on the
small dough samples with 46%, 50%, and 54% , a relevant
difference in both real and imaginary part of the complex
dielectric permittivity is found between 4 and 6 GHz, as shown
in Fig. 2. Consequently, the ISM operating frequency of 5.8
GHz has been selected. The values reported in Fig. 2 have
been used to model the dielectric profile of the sheeted dough
within the CST simulation environment.
After having identified a suitable working frequency in
which the dielectric contrast is relevant, the sheeted dough has
been cut into 21 cm x 12 cm rectangular samples. This
experiment has been performed in order to understand if there
was a large disparity of in the same sheet. The effective
dielectric properties of each sample have been evaluated in
different positions of the sheeted dough by considering a 3 x 3
grid, in order to map the spatial distribution of the relative
effective
1
complex permittivity (i.e., ). In this way, we
aimed at evaluating the spatial diversity and level of
homogeneity of the sheets. In fact, the different would
affect the dough texture and impact on the drying process, with
effects on the final product quality [33]. Each rectangular
element of the considered grid has size 4 cm x 7 cm, as
reported in the schematic in Fig. 3. A picture of the
measurement process is reported in Fig. 4. The measurement
1
Note that the term effective has been used to take into account also possible
contribution from the materials below the sheeted samples.
of has been made by placing the probe in the center
of each grid rectangle.
The results of the effective complex permittivity
measurements of the samples for the considered at 5.8
GHz are shown in Fig. 5a, 5b, and 5c. The Carasau dough
sheet with 46% presents a more spread distribution of both
and (± 0.1 unit of permittivity per cm2) with respect to
the other 2 cases, though it is still contained. Indeed, as the
water content increases, the deviation of the complex
permittivity decreases, and the dough sheets at 50% and 54%
becomes very homogeneous (~ ± 0.01 unit of permittivity per
cm2). The average effective complex permittivity for the
variation at 5.8 GHz and the relative standard deviation are
also reported in Fig. 5d and 5e. Based on these findings, we
can conclude that the permittivity profile, within the same
sheet, is pretty much uniform. This allows us to hugely
simplify the simulations, as we could model the dough sheets
with a dielectric profile that changes with frequency but not
with space.
FIGURE 3. Schematic of the grid considered for acquiring
dielectric measurements over the Carasau dough samples.
FIGURE 4. Picture of the measurement process using the coaxial
probe connected to the VNA (Rhode & Schwarz ZNB 8).
IV. CHOICE OF THE ANTENNA DESIGN AND
SIMULATIONS
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
In order to select the proper layout for the envisioned
measurement methodology and satisfy the system
requirements given by the industrial stakeholder, different
types of antennas have been simulated, testing the sensitivity,
with respect to the variation, of wide-band, dual-band,
and narrow-band antennas.
To choose which type of antenna could be suitable for this
application, the authors decided to divide the antennas to be
tested into classes based on a bandwidth criterion, rather than
show countless layouts with the same outcome, and highlight
the simplest ones (easily reproducible). Basically, more than
one antenna for each class has been tested but for the sake of
simplicity, since the conclusion is arguably the same for each
antenna in the same class, we opted to show one representative
case. The wide-band antenna is a WR 187 pyramidal horn
antenna whose dimensional drawing is reported in Fig. 6. The
waveguide section of the pyramidal horn is a WR 187, thus
with size 47.55 mm x 22.15 mm. The a1 x b1 aperture (with a1
= 73.4 mm and b1 = 53.85 mm) is spaced out from the feeding
waveguide by pE = 139.7 mm. A summary of the horn
dimensions is reported in the caption of Fig. 6. The
dimensional drawing of the dual-band patch antenna is
reported in Fig. 7. The dual-band antenna is a series fed
stacked patch with both substrates made of Polylactic Acid
(PLA, εr = 2.54, tanδ = 0.015) having thickness h = 1 mm and
size S x S (with S = 50 mm). The substrates are separated by a
small gap g = 0.5 mm. The bottom parasitic patch size is W1
x L1 (with W1 = 22 mm and L1 = 18.25 mm), whereas the top
radiating patch size is W x L (with W = 24 mm and L = 18.5
mm). Both patches are made of copper. The coaxial feed is
located at xf = 4.8 mm from the center of the patch. A summary
of the series fed stacked patch dimensions is reported in the
caption of Fig. 7. Finally, the narrow-band antenna (resonating
at 5.8 GHz) is a simple coaxial-fed patch, having a PLA Sp x
Sp x hp substrate (with Sp = 35 mm and hp = 2 mm). The PLA
has been chosen to speed up the prototyping process with 3D-
printing, in preparation for the validation measurements. The
copper patch has size Wp x Lp (with Wp = 19.54 mm and Lp =
15.15 mm) and the feeding point is placed at xfp = 3 mm from
the center of the patch (for a summary of the dimensions of the
5.8 GHz patch, please refer to the caption of Fig. 12, where the
simulation environment is reported).
For comparison, the magnitude of the input reflection
coefficient (|S11|) in free space for the three layout is displayed
in Fig. 8, highlighting the difference in bandwidth between the
three antennas.
The simulation environment for the moisture sensitivity
has been built in the CST Studio Suite workspace and consists
of the antenna for the estimation, placed above the bread
sheet, which is in direct contact with a rubber conveyor belt
(see Fig. 12 for reference). The default distance between the
antenna and the bread sheet has been chosen equal to d = 10
mm. In this respect, shorter distances improve the overall
sensitivity of the sensor and the chosen value is the lowest one
that does not hinder the baking process. The shape and size of
the Carasau bread sheet has been derived from several
measurements on samples from a real traditional bakery. It has
been chosen of elliptical shape with axis equal to 370 mm and
FIGURE 5. Distribution of the relative complex permittivity for the rectangular sample with 46% (a), 50% (b), and 54% (c), and
average complex permittivity (real – d, and imaginary – e) for the rectangular sample with respect to variation at 5.8 GHz.
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
360 mm, and thickness equal to 1 mm, since the unbaked bread
sheets do not have a perfect circular shape. The selected
simulation environment is rather simple but convenient,
without additional components that would burden the
computational load of the simulation. Aim of these
simulations is simply to verify if the sensor is capable of
discriminating the difference in , whereas the real
validation of the presented point is entrusted to the
measurements. A total of three simulation have been carried
out, accounting, respectively, for the horn, the stacked patch
and the 5.8 GHz patch antennas as the moisture sensor for a
Carasau dough sheet with different (46%, 50%, and 54%).
In Figs. 9 and 10, the frequency response of the first two
cases is studied. From the images, it is clear that these layouts
are pretty much unsusceptible to the variation of water inside
the dough. The shift of the |S11| curve is barely visible in both
cases. To be more precise, the maximum difference in
magnitude between the highest and the lowest
W%
in the case
of the horn antenna is < 0.5 dB, whereas for the dual-frequency
patch there is a difference < 1 dB between 46% and 50%, and
< 0.5 dB between 50% and 54%, for both peaks.
FIGURE 6. Dimensional drawing of the WR 187 pyramidal horn
antenna. Parameters: a = 47.55 mm, b = 22.15 mm, a1 = 73.4 mm,
b1 = 53.85 mm, pE = 139.7 mm.
FIGURE 7. Dimensional drawing of the series fed stacked patch
antenna. Parameters: S = 50 mm, h = 1 mm, a1 = 73.4 mm, W1 = 22
mm, L1 = 18.25 mm, xf = 4.8 mm, g = 0.5 mm, W = 24 mm, L = 18.5
mm.
Fig. 11a, on the other hand, shows the frequency response
using the simple narrow-band coaxial-fed patch antenna. In
this case, the difference in is evident in the shift of the S11
magnitude minima, with lower minima corresponding to
higher water content. Upon further inspection it is clear that
the difference between the minima corresponding to each
W%
is about 5 dB. The different sensitivity of the three layouts has
been summarized in Table 1. This difference can be regarded
as a matching difference due to the change in complex
permittivity (which is obviously tied to the different W%).
Indeed, the 5.8 GHz rectangular patch antenna used as a sensor
is a simplistic layout but very well matched with the 50 Ω
feeding in free space, as shown in Fig. 8. Very good matching
is maintained also for high water contents (i.e., 54%, see Fig.
11a), whereas it worsens slightly as the water content lowers
(see also Section 5 in the supplementary material).
FIGURE 8. Comparison between the simulated S11 magnitude of
the wide-band horn, the dual-band patch, and the 5.8 GHz patch
antennas in free space.
FIGURE 9. Simulated magnitude of the S11 for the wide-band horn
antenna placed at d = 10 mm, for different water contents.
An additional investigation on the S11 phase shows that its
trend is barely affected by the water percentage when using
the 5.8 GHz patch (see Fig. 11b), a piece of information that
is going to be significant in the next steps of the study. This
proves, at least numerically, that a simple patch antenna could
be used for the discrimination of the inside the dough. A
representation of the simulation environment is reported in
Fig. 12 (the model is the same used also for the other antenna
layouts).
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
FIGURE 10. Simulated magnitude of the S11 for the dual-band
patch antenna placed at d = 10 mm, for different water contents.
a)
b)
FIGURE 11. Simulated magnitude (a) and phase (b) of the S11 for
the 5.8 GHz patch antenna placed at d = 10 mm, for different water
contents.
TABLE 1. Sensitivity of the presented layouts with respect to the
W% variation, evaluated as difference in the |S11| minima.
min |S11|
from 46 % to
50%
from 50% to
54%
WR 187 Horn
< 0.25 dB
< 0.25 dB
Series-fed stacked patch
< 0.5 dB
< 1 dB
5.8 GHz patch
~ 5 dB
~ 5 dB
V. EARLY PROTOTYPE FABRICATION AND
EXPERIMENTAL MEASUREMENTS
The simulations provided a hint that a simple patch antenna
working at 5.8 GHz might be capable of discriminating small
percentage variations of water content inside the sheeted
Carasau bread dough. This assumption is, nonetheless, based
mainly on the observed difference of about 5 dB in S11
magnitude minima within a numerical simulation. Actually,
this information could be lost in a real measurement, as this
matching difference might not be relevant in absence of the
“ideal” context provided by the simulation. Indeed, to validate
this finding a real case scenario must be tested, hence, an ad
hoc measurement setup has been designed. Three different
doughs with the targeted (46%, 50%, and 54%) have been
prepared and, after kneading and a first leavening, have been
sheeted to obtain small samples for the measurements. The
samples have been sheeted using a pasta machine, to reach a
thickness of about 1 mm. It must be noticed that such small
samples have not the same size of the disks used in the
numerical simulations, for practical reasons. However, this
does not affect the frequency response of the antenna, since
the edges of the sheeted samples are still far enough from the
sensor. The measurement setup consists of:
- a laser sensor (optoNCDT ILD 1900-2 from Micro-
epsilon) employed to determine the thickness of the
sheeted samples and the distance of the antenna from
the sheet;
- the 5.8 GHz patch antenna described in Section IV,
employed as a water content sensor;
- a custom 3D printed PLA arm with interchangeable
housing for the laser sensor and the antenna;
- a metallic support to adjust the plastic arm height;
- a 3D printed PLA plate to lay down the sheeted sample;
- a commercial VNA (Hewlett-Packard 8720C) with
frequency range from 5 to 7 GHz, number of points
equal to 801 (which is equivalent to a frequency
resolution of about 2.5 MHz), system dynamic range
equal to 103 dBm and input power equal to 10 dBm.
A picture of the described setup is shown in Fig. 13. The
measurements have been performed in a dedicated laboratory
at room temperature (~26°C). The precise thickness of the
sheeted dough measured with the laser was 1.165 mm,
whereas the default measured distance of the antenna from the
sheet was 34 mm. This value is not the same one used in the
simulation, but the two environments greatly differ from each
other, in the first place, and, secondly, the measurement
antenna-sheet distance value has been chosen in order to
obtain the best impedance matching (RL > 20 dB) for the
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
standard dough (= 50%). The rationale is that spotting the
predicted 5 dB difference between the different water content
would be much easier starting from a well matched frequency
response, as in the case of the simulations.
FIGURE 12. Representation of the simulation environment with
the simple 5.8 GHz patch antenna (the same model is used for all
the layouts). Parameters: Sp = 35 mm, Wp = 19.54 mm, Lp = 15.15
mm, xfp = 3 mm, hp = 2 mm (substrate thickness), d = 10 mm.
The measured frequency response, both in magnitude and
phase, of the described environment, is reported in Fig. 14. As
it can be seen from the image, the shift in |S11| minima due to
the variation of can be appreciated also from these results,
and quite surprisingly, there is a difference of about 5 dB, in
agreement with what the simulations have produced. The
similarity does not stop to the magnitude analysis, as also the
phase shows a common trend in the measurements, just like
the numerical simulations, thus demonstrating the validity and
suitability of the proposed moisture content measurement
method. The presented antenna, despite being simplistic in
nature and probably not perfectly suitable as a definitive
layout employed in the real industrial application, is a rather
promising prototype that successfully demonstrate the proof-
of-concept of a methodology tied to the Carasau industry and
the Carasau dielectric profile. It is not intended as a test plan
for the industrial integration, lacking the operating structure
effectively employed for the bread production. Still, a general
agreement between simulations and measurements is a right
step forward for a real industrial application, which may
employ a layout not drastically different from the one herein
described.
It would be useful to define a Figure of Merit (FoM) for the
presented problem, able to quantify the performance of the
sensor antenna. The most useful information, as already
discussed previously, is the |S11| minima variation with respect
to the variation. A difference of about 5 dB can be seen
from both simulations and measurements. Thus, we can define
a FoM as it follows:
wherein is the difference, in dB, between the |S11|
minima for different and is the difference in water
percentage of the dough. Using this formulation, for the
presented case, we obtain F = 1.25 dBppp (decibel per
percentage point). Additionally, if we consider a difference of
about 3 dB (half power) as the minimum acceptable
difference, we can also define a lower limit of F, in order to
establish the suitability of a sensor for the same purpose as the
one herein presented. In this case, we have:
On a note of a possible configuration that could be used in
a real operating industrial setting, the presented system could
be easily integrated in the production chain, taking advantage
of a mounting structure, above the conveyor belt, in which the
sensor could be located. To help the visualization of this
system, a conceptual drawing has been reported in Fig. 15. The
sensor should consist of more than one antenna in single mode
operation (not an array), in order to map the entirety of the
dough sheet (each antenna “illuminates” a different portion of
the sample), at an optimal distance d from the sheeted dough,
in order to have a good matching. The antennas should be
simply displaced in a linear fashion along the -direction, as
we can take advantage of the conveyor belt movement in the
-direction, with reference to the drawing of Fig. 15. The
sensor should be connected to a small VNA (e.g., a Nano-
VNA) or an electronic board able to fulfill the same functions
(e.g., like an FPGA), which, in turn, should be connected to a
system for the information processing. Further insights on the
latter topic can be found in Section VII. We would like to
stress out that the Hewlett-Packard 8720C VNA used in the
presented measurements is an old and bulky model and it was
employed only for validation purposes, but it would result
unsuitable for a real-case scenario. In fact, within the described
system of Fig. 15, given the speed of the conveyor belt (from
few to tens of m/s) and the distance from two consecutive
dough sheets, at steady state, an arbitrary section of the
conveyor belt can be crossed by 1 dough sheet every 2
seconds. The sweep time of the employed VNA (with the
provided frequency span and number of points) is equal to
2.408 seconds, thus it would be unable to perform even one
full acquisition in a real scenario. However, today more
compact and performing VNAs are available commercially
that are more than capable to handle multiple data acquisitions
in a time frame of 2 seconds.
VI. RECOGNITION OF EXTERNAL FACTORS
In addition to the presented analysis, a further study has
been carried out to investigate possible sources of
misinterpretation of the results. In particular, two effects that
could be detrimental for the main purpose of the water content
sensor are some changes in the geometry of the setup herein
used for the measurements. It must be absolutely clear that
these events are highly unlike to happen to an extent that could
be troubling, but take them into due account is a dutiful part of
a comprehensive uncertainty analysis, to clearly set the
reliable operative conditions of the proposed sensor. The two
effects that we are going to consider in the following
discussion are the change of the antenna-sheet distance d and
the change of the sheet thickness hs, and if the variation of
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
these quantities could be a possible source of confusion when
compared to the water content variation.
a)
b)
FIGURE 13. Picture of the employed measurement setup with the
laser sensor attachment (a) and the water content sensor
attachment (b).
As in the case of water content variation, to address these
points, a preliminary simulation (with the same environment
described in Section IV) is carried out, followed by a
measurement validation. In this respect, in Fig. 16, the
simulated magnitude and phase of the S11 when accounting for
an antenna-sheet distance d varying between 5 and 15 mm (for
the sake of clarity the has been kept equal to 50% and the
sheet thickness hs equal to 1 mm) is shown. The translation of
the |S11| minima is visible also in this case but it is paired with
a large frequency shift in the order 60-70 MHz (compare
Fig.16a with Fig.11a). Since the |S11| minima frequency shift,
as we already pointed out, is not ascribable to the variation of
the , this is a clear effect of the variation of the antenna-
sheet distance. Additionally, the S11 phase appears highly
dependent from the distance d (see Fig.16b) and there is no
common trend in the phase curves. This does not happen in the
case of varying , wherein the phase trend remains stable,
so it can be a good indicator of the real cause behind the
min(|S11|) shift. In the same fashion, in Fig. 17, the frequency
response for the sheet thickness (hs) variation between 1 and
2.5 mm is reported (with fixed to 50%). In this case, there
is a change in |S11 | minima but not a frequency shift, picturing
a situation similar to the variation.
a)
b)
FIGURE 14. Measured magnitude (a) and phase (b) of the S11 for
the 5.8 GHz patch antenna placed at d = 10 mm, for different water
contents.
FIGURE 15. Conceptualization of a possible implementation of the
moisture sensing system in a real industrial scenario.
However, the phase curves do not match (compare Fig. 17b
with Fig. 11b), and, as in the case of the distance variation, this
discrepancy can be used to discriminate the two effects (
variation from hs variation).
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
a)
b)
FIGURE 16. Simulated magnitude (a) and phase (b) of the S11 for
the 5.8 GHz patch antenna for different values of d (water content
fixed to 50%, hs fixed to 1 mm).
The next step is to validate the presented simulations by real
measurements, using the same setup described in Section V.
As in the case of the numerical simulations, the distance d and
thickness hs are progressively increased by 2.5-mm and 0.5-
mm steps, respectively (distances and thicknesses have been
measured using the laser sensor). The measured S11 for the
increment of the distance d is shown in Fig. 18. Again, it can
be noticed the agreement between simulations and
measurements, since the shift in both frequency and
magnitude is still visible (in this case the frequency shift is in
the order of 20-25 MHz but it is, nonetheless, important
especially with respect to the unnoticeable shift experienced
for the variation). The same can be inferred for the phase,
wherein no common trend with the default case is observed.
Finally, in Fig. 19, the measured S11 for the increment of the
thickness hs is shown. While the |S11| pictures a similar
situation to the simulation (change in magnitude but not in
frequency), the information that allowed to discriminate this
effect is lost in the phase, since a common trend is observed.
This inconvenience might be due to the employment of non-
conformal devices for the sheeting of the dough (we could
achieve a 0.5 mm control on the dough thickness with the
available pasta machine that is different from industrial
sheeters) or to the elastic nature of the dough, whose relaxation
and leavening cycles are still a subject for further study. Still,
it must be remarked that these effects are taken to an extreme
and while the proposed system is already capable of
discriminating between water content and distance variation,
the varying thickness problem could be solve with additional
knowledge on the dough characteristics. In order to help the
reader to navigate with ease through the presented results and
findings, a summary table for quick comparison between
simulated and measured results is reported in Fig. 20. The
latter allows also to compare the frequency response of the
antenna due to different effects. The summary is organized as
follows: the rows (A, B, and C) are useful for a straight
comparison of simulated and measured S11 (magnitude and
phase) referred to a single factor, A for water content variation,
B for antenna-sheet distance variation, and C for sheet
thickness variation; the columns I, II, III, and IV allow a
prompt comparison of the effect that each factor has on the
frequency response of the antenna.
a)
b)
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
FIGURE 17. Simulated magnitude (a) and phase (b) of the S11 for
the 5.8 GHz patch antenna for different values hs (water content
fixed to 50%).
VII. CONCLUSION AND FUTURE PERSPECTIVES
To comply with the technological advancements that the
food industry is facing, involving also small-scale activities,
the design and early validation of a sensor for the water content
estimation inside a typical Italian bread dough is herein
presented. Carasau bread differs from classic bread in its shape
and texture, thus needing an ad hoc system and a specific
measurement methodology. Based on a survey on the
operating conditions of a real bakery, the most important step
within the production process lies between the second
leavening and baking stages, wherein the water content inside
the sheeted dough must be properly evaluated. An excessive
or defective amount of water results in a flawed product that
has to be discarded, affecting the economic returns of the
production line. This raises the need of a contactless
measurement and sensing procedure, with a device to be
integrated in a fixed position above the conveyor belt, able to
sense the different permittivity profile of every unbaked sheet
that slides underneath.
a)
b)
FIGURE 18. Measured magnitude (a) and phase (b) of the S11 for
the 5.8 GHz patch antenna for different values of d (water content
fixed to 50%, hs fixed to 1 mm).
a)
b)
FIGURE 19. Measured magnitude (a) and phase (b) of the S11 for
the 5.8 GHz patch antenna for different values hs (water content
fixed to 50%).
Simulation results show that a narrow-band patch antenna
is preferable to wide-band and dual-band counterparts. A
simple contactless inline sensor is capable of discriminating
small water percentage variations (~ 4%) inside the bread
dough through the matching of the antenna response. This
feature allows to monitor the water content inside the dough,
avoiding defected products and wastes. The validation
measurements confirm that the chosen antenna is capable of
discriminating the water content. Additionally, a further study
for the recognition of external factors that could undermine the
discrimination of the water content variation has been
presented. Based on the simulations, the chosen sensor is able
to differentiate the effects unrelated to the water content, such
as the variation of the antenna-sheet distance and the variation
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
of the sheet thickness. These finding are confirmed also by the
measurements, with regard to the antenna-sheet distance,
whereas for the variation of sheet thickness further studies are
needed.
Future work will deal with the specific usage of the
information provided in the presented study. It could involve
inversion algorithms of the scattering matrix, thus opting for
the implementation of a microwave imaging system to map
the moisture of the bread sheet with more accuracy, or a large
database construction for a machine learning algorithm
training. The presented contactless measurement
methodology could also be applied to other food products with
FIGURE 20. Summary of the simulated and measured results taking account for shift of magnitude and phase of the S11 due to the
variation of water content W%, antenna-sheet distance d, and sheet thickness hs.
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
characteristic akin to those of the Carasau bread (thus, that
undergo a baking step such as common bread, pastry, and so
on) and other industrial processing scenarios calling for
innovation.
ACKNOWLEGMENT
The authors would like to express their thanks Massimo
Bauco and Luigi Lorusso from the Rohde & Schwarz Italia for
the VNA freely provided.
REFERENCES
[1] S. O. Nelson, and P. G. Bartley, “Measuring frequency- and
temperature-dependent permittivities of food materials,” in IEEE
Trans. Instrum. Meas., vol. 51, no. 4, pp. 589–592, Aug. 2002.
[2] N. Javanbakht, G. Xiao, and R. E. Amaya, “Portable Microwave
Sensor Based on Frequency-Selective Surface for Grain Moisture
Content Monitoring,” in IEEE Sens. Lett., vol. 5, no. 11, pp. 1–4, Sep.
2021.
[3] T. Limpiti, and M. Krairiksh, “In Situ Moisture Content Monitoring
Sensor Detecting Mutual Coupling Magnitude between Parallel and
Perpendicular Dipole Antennas,” in IEEE Trans. Instrum. Meas., vol.
61, no. 8, pp. 2230–2241, Mar. 2012.
[4] T. Tantisopharak, H. Moon, P. Youryon, K. Bunya-Athichart, M.
Krairiksh, T. K. Sarkar, “Nondestructive determination of the maturity
of the durian fruit in the frequency domain using the change in the
natural frequency,” in IEEE Trans. Antennas Propag., vol. 64, no. 5,
pp. 1779-1787, 2016.
[5] S. Karuppuswami, A. Kaur, H. Arangali, and P. Chahal, “A hybrid
magnetoelastic wireless sensor for detection of food adulteration,” in
IEEE Sens. J., vol. 17, no. 6, pp. 1706–1714, Mar. 2017.
[6] J.A. Tobón Vásquez, et al., “Noninvasive inline food inspection via
microwave imaging technology: an application example in the food
industry,” in IEEE Antennas Propag. Mag., vol. 62, no. 5, pp. 18–32,
Aug. 2020.
[7] M. Ricci, J. A. T. Vasquez, R. Scapaticci, L. Crocco, F. Vipiana,
“Multi-Antenna System for In-Line Food Imaging at Microwave
Frequencies,” in IEEE Trans. Antennas Propag., Early Access, 2022.
[8] F. Gandino, B. Montrucchio, M. Rebaudengo, and E. R. Sanchez, “On
improving automation by integrating RFID in the traceability
management of the agri-food sector,” in IEEE Trans. Ind. Electron.,
vol. 56, no. 7, pp. 2357–2365, Jul. 2009.
[9] I. Cuiñas, R. Newman, M. Trebar, L. Catarinucci, and A. A. Melcon,
“Rfid-based traceability along the food-production chain [Wireless
Corner], ” in IEEE Antennas Propag. Mag., vol. 56, no. 2, pp. 196–
207, Apr. 2014.
[10] F. Paschino, F. Gabella, F. Giubellino, F. Clemente, “The level of
automation of “carasau” bread production plants,” in J. Agric. Eng.,
vol. 38, pp. 61–64, 2007.
[11] M. Baire, A. Melis, M. B. Lodi, L. Lodi, A. Fanti, and G. Mazzarella
“Empowering traditional carasau bread production using wireless
sensor network,” in Proc. 2021 IEEE Int. Symp. Circuits Syst.
(ISCAS), pp. 1–4, 22–28 May 2021, Daegu, Korea.
[12] L. Cocco, et al., “A blockchain-based traceability system in agri-food
SME: case study of a traditional bakery,” in IEEE Access, vol. 9, pp.
62899–62915, Apr. 2021.
[13] G. Muntoni, et al., “Designing a Microwave Moisture Content Sensor
for Carasau Bread: A Feasibility Study,” in Proc. 2022 16th Eur. Conf.
Antennas Propag. (EuCAP), pp. 1–4, Madrid, Spain, 27 March – 01
April 2022.
[14] F. Fanari, I. Frau, F. Desogus, E. A. Scano, G. Carboni, M. Grosso,
“Influence of wheat varieties, mixing time and water content on the
rheological properties of semolina doughs,” in Chemical Engineering
Transactions, vol. 75, pp. 529 - 534, 2019.
[15] F. Fanari, F. Desogus, E. A. Scano, G. Carboni, M. Grosso, “The effect
of the relative amount of ingredients on the rheological properties of
semolina doughs,” in Sustainability, vol. 12, no. 7, 2705, 2020.
[16] F. Fanari, I. F. Naue, F. Desogus, M. Grosso, M. Wilhelm, “Durum
wheat dough torque measurements: Characterization and study of the
mixing process parameters as a function of water and salt amounts,”
in Chemical Engineering Transactions, vol .87, pp. 205-210, 2021.
[17] F. Fanari, G. Carboni, F. Desogus, M. Grosso, M. Wilhelm, “A
Chemometric Approach to Assess the Rheological Properties of
Durum Wheat Dough by Indirect FTIR Measurements,” in Food and
Bioprocess Technology, vol. 15, no. 5, pp. 1040-1054, 2022.
[18] F. Fanari, C. Iacob, G. Carboni, F. Desogus, M. Grosso, M. Wilhelm,
“Broadband Dielectric Spectroscopy (BDS) investigation of
molecular relaxations in durum wheat dough at low temperatures and
their relationship with rheological properties,” in LWT, vol. 161,
113345, 2022.
[19] M. B. Lodi, N. Curreli, A. Melis, E. Garau, F. Fanari, A. Fedeli, A.
Randazzo, G. Mazzarella, A. Fanti, “Microwave characterization and
modeling of the Carasau bread doughs during leavening,” in IEEE
Access, vol. 9, pp. 159833-159847, 2021.
[20] M. Bogosanovich, “Microstrip patch sensor for measurement of the
permittivity of homogeneous dielectric materials,” in IEEE Trans.
Instrum. Meas., vol. 49, no. 5, pp. 1144–1148, Oct. 2000.
[21] N. Javanbakht, G. Xiao and R. E. Amaya, “A Comprehensive Review
of Portable Microwave Sensors for Grains and Mineral Materials
Moisture Content Monitoring,” in IEEE Access, vol. 9, pp. 120176–
120184, 2021.
[22] K. Sarabanid and E. Li, “Microstrip ring resonator for soil moisture
measurements,” in IEEE Trans. Geosci. Remote Sens., vol. 35, no. 5,
pp. 1223–1232, Sep. 1997.
[23] M. J. Tiusanen, “Wideband Antenna for Underground Soil Scout
Transmission,” in IEEE Antennas and Wirel. Propag. Lett., vol. 5, pp.
517–519, 2006.
[24] A. Cataldo, G. Monti, E. De Benedetto, G. Cannazza and L. Tarricone,
“A Noninvasive Resonance-Based Method for Moisture Content
Evaluation Through Microstrip Antennas,” in IEEE Trans. Instrum.
Meas., vol. 58, no. 5, pp. 1420–1426, May 2009.
[25] P. Leekul, B. Mgawe, T. Kazema, H. N. Dao, P. Sirisuk and M.
Krairiksh, “Simple and Effective Design Concept for Constructing In-
Situ Soil Dielectric Property Sensor With Dual Low-Cost COTS
Microwave Modules,” in IEEE Access, vol. 10, pp. 54516–54524,
2022.
[26] R. Keshavarz, J. Lipman, D. M. M.-P. Schreurs and N. Shariati,
“Highly Sensitive Differential Microwave Sensor for Soil Moisture
Measurement,” in IEEE Sens. J., vol. 21, no. 24, pp. 27458–27464, 15
Dec.15, 2021
[27] N. Hosseini and M. Baghelani, “Selective real-time non-contact
multivariable water-alcohol-sugar concentration analysis during
fermentation process using microwave split-ring resonator based
sensor,” in Sens. Actuators A, Phys., vol. 325, Jul. 2021.
[28] S. Julrat and S. Trabelsi, “Influence of Peanut Orientation on
Microwave Sensing of Moisture Content in Cleaned Unshelled
Peanuts,” in IEEE Sens. J., vol. 22, no. 11, pp. 10515–10523, Jun.
2022.
[29] S. Jiarasuwan, K. Chamnongthai and N. Kittiamornkul, “A Design
Method for a Microwave-Based Moisture Sensing System for
Granular Materials in Arbitrarily Shaped Containers,” in IEEE Sens.
J., vol. 21, no. 17, pp. 19436–19452, Sept. 2021.
[30] M. M. Ghretli, K. Khalid, I. V. Grozescu, M. H. Sahri and Z. Abbas,
“Dual-Frequency Microwave Moisture Sensor Based on Circular
Microstrip Antenna,” in IEEE Sens. J., vol. 7, no. 12, pp. 1749–1756,
Dec. 2007.
[31] N. Khalid, R. Mirzavand, H. Saghlatoon, M. M. Honari, and P.
Mousavi, “A three-port zero-power RFID sensor architecture for IoT
applications,” in IEEE Access, vol. 8, pp. 66888–66897, 2020.
[32] G. Gugliandolo, K. Naishadham, G. Neri, V. C. Fernicola and N.
Donato, “A Novel Sensor-Integrated Aperture Coupled Microwave
Patch Resonator for Humidity Detection, ” in IEEE Trans. Instrum.
Meas., vol. 70, pp. 1–11, 2021.
[33] L. Tebben, Y. Shen, and Y. Li, “Improvers and functional ingredients
in whole wheat bread: A review of their effects on dough properties
and bread quality,” Trends Food Sci. Technol., vol. 81, pp. 10–24.
Nov. 2018.
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
GIACOMO MUNTONI (Member IEEE)
graduated in Electronic Engineering and
Telecommunication Engineering at the
University of Cagliari in 2010 and 2015,
respectively. In 2019 he received, the PhD in
Electronic Engineering and Computer
Science, from the University of Cagliari. He is
currently working as technologist in the
Applied Electromagnetics Group at the
University of Cagliari.
His research activity involves: design and characterization of
antennas for biomedical and aerospace applications, microwave–
based dielectric characterization of materials, 3D printing of RF
components, and monitoring of the space debris environment in Low
Earth Orbit with the Sardinia Radio Telescope, in collaboration with
the Cagliari Astronomical Observatory.
MATTEO B. LODI (Member, IEEE) received
the Bachelor's degree in Biomedical
Engineering from the University of Cagliari
(2016) and the Master's degree in Biomedical
Engineering from Politecnico di Torino
(2018). In 2022 he received, with honor, the
PhD in Electronic Engineering and Computer
Science from the University of Cagliari. He is
currently working as research assistant
(RTD-A) in the Applied Electromagnetics
Group. His research activity deals with the modelling of
bioelectromagnetic phenomena, especially hyperthermia treatment;
the study, manufacturing, and synthesis of magnetic biomaterials for
tissue engineering applications; and the use of microwaves for
biotechnology and environmental applications, while working in the
design and characterization of antennas for space and wearable
applications.
He has been awarded with the Roberto Sorrentino Young Scientist
award at the 2022 Italian URSI Assembly. He has been awarded as
Young Scientists at the General Assembly and Scientific Symposium
of URSI in 2020 and 2021. He is a member of the WG2: "Better
thermal-based EM therapeutics" of the COST Action 17115
"MyWave". He has been appointed as the Chair of IEEE
Nanotechnology Council Young Professionals.
ALESSANDRO FEDELI (Member, IEEE)
received the B.Sc. and M.Sc. degrees in
electronic engineering and the Ph.D. degree in
science and technology for electronic and
telecommunications engineering from the
University of Genoa, Genoa, Italy, in 2011, 2013,
and 2017, respectively, where he is currently an
Assistant Professor with the Department of
Electrical, Electronic, Telecommunications
Engineering, and Naval Architecture. His research activities, carried
out at the Applied Electromagnetics Laboratory, are mainly focused
on the development and the application of computational methods for
the solution of forward and inverse scattering problems, and
electromagnetic imaging. He has coauthored more than 130
scientific contributions published in international journals and
conference proceedings. He is a member of the IEEE Antennas and
Propagation Society, the Italian Society of Electromagnetism, and
the Interuniversity Center for the Interaction between
Electromagnetic Fields and Biosystems.
ANDREA MELIS received the bachelor’s
degree in biomedical engineering from the
University of Cagliari, Italy, in 2017. He
worked as an Assistant Researcher with the
University of Cagliari. His research interests
include EM modeling and development of RF
coils at low and high frequencies, especially
for MRI at high field, the design and realization
of WSN systems for the monitoring of
industrial processes, such as bread
manufacturing, and intelligent transportation systems.
CLAUDIA MACCIÒ received the bachelor’s
and master’s degrees in physics in 2015
and 2018, respectively, from the University
of Cagliari, Cagliari, Italy, where she is
currently working toward the Ph.D. degree
in electronic engineering and computer
science. Following graduation, she also
attended a oneyear training course for
mechatronics experts with POEMA srl, a
spin-off company of the Italian National
Institute Astrophysics (INAF), receiving research grants from INAF
and the University of Cagliari. Her research interests include the use
of microwave for material characterization, in particular for food
industry applications, design of microwave devices, structures and
sensors, and innovative manufacturing process for microwave
devices, such as 3D printing and electroplating.
ANDREA RANDAZZO (Senior Member, IEEE)
received the Laurea degree in
telecommunication engineering and the Ph.D.
degree in information and communication
technologies from the University of Genoa,
Genoa, Italy, in 2001 and 2006, respectively. He
is currently a Full Professor of electromagnetic
fields with the Department of Electrical,
Electronic, Telecommunication Engineering, and
Naval Architecture, University of Genoa. He has coauthored the book
Microwave Imaging Methods and Applications (Artech House, 2018)
and more than 270 articles published in journals and conference
proceedings. His primary research interests are in the field of
microwave imaging, inverse-scattering techniques, numerical
methods for electromagnetic scattering and propagation, electrical
tomography, and smart antennas.
MATTEO PASTORINO (Fellow, IEEE) was
a Full Professor of electromagnetic fields
with the University of Genoa, Genoa, Italy,
where he was the Director of the Department
of Electrical, Electronic,
Telecommunications Engineering, and
Naval Architecture (DITEN). He had
coauthored about 500 articles in international
journals and conference proceedings. His
research interests include microwave and
millimeter wave imaging, direct and inverse scattering problems,
industrial and medical applications, smart antennas, and analytical
and numerical methods in electromagnetism. Prof. Pastorino was the
Chair of the National URSI Commission B (Fields and Waves), Vice
Director of the Interuniversity Center for the Interaction between
Electromagnetic Fields and Biosystems. He was an Associate Editor
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
VOLUME XX, XXXX 1
for the IEEE Antennas and Propagation Magazine and IEEE Open
Journal of Antennas and Propagation.
GIUSEPPE MAZZARELLA (Senior Member,
IEEE) received the degree (summa cum laude)
in electronic engineering from the Università
Federico II of Naples, in 1984, and the Ph.D.
degree in electronic engineering and computer
science, in 1989. In 1990, he became an
Assistant Professor at the Dipartimento di
Ingegneria Elettronica, Università Federico II of
Naples. Since 1992, he has been with the Dipartimento di Ingegneria
Elettrica ed Elettronica, Università di Cagliari, first as an Associate
Professor and then, since 2000, as a Full Professor, teaching
courses in electromagnetics, microwave, antennas and remote
sensing. He is the author (or coauthor) of over 100 articles in
international journals and a reviewer for many EM journals. His
research interests include the efficient design of large arrays of slots,
power synthesis of array factor, with emphasis on inclusion of
constraints, microwave holography techniques for the diagnosis of
large reflector antennas, use of evolutionary programming for the
solution of inverse problems, in particular problems of synthesis of
antennas and periodic structures, agrifood and therapeutic
applications of microwaves.
ALESSANDRO FANTI Alessandro Fanti (Senior
Member, IEEE) received the Laurea degree in
electronic engineering and the Ph.D. degree in
electronic engineering and computer science from
the University of Cagliari, Cagliari, Italy, in 2006
and 2012, respectively. From 2013 to 2016, he
was a Postdoctoral Fellow of the Electromagnetic
Group, University of Cagliari, where he was an
Assistant Professor in electromagnetic fields with the Department of
Electrical and Electronic Engineering, from March 2017 to March 2024.
He is currently an Associate Professor with the University of Cagliari.
He has authored or coauthored 70 articles in international journals. His
research interests include the use of numerical techniques for modes
computation of guiding structures, optimization techniques, analysis
and design of waveguide slot arrays, analysis and design of patch
antennas, radio propagation in urban environment, modeling of
bioelectromagnetic phenomena, and microwave exposure systems for
biotechnology and bioagriculture. He is a member of the IEEE
Antennas and Propagation Society, Italian Society of
Electromagnetism, National Inter-University Consortium for
Telecommunications, and Interuniversity Center for the Interaction
Between Electromagnetic Fields and Biosystems. From 2020 to 2023,
he had been acting as a Principal Investigator of the IAPC Project,
which was funded with five million euros by Italian Ministry of Economic
Development (MISE), within the AGRIFOOD PON I&C (2014–2020).
Since 2024, he has been acting as a Principal Investigator of the AISAC
Project funded with 15 million euros by Italian Ministry of Enterprises
and Made in Italy (MIMIT), within the “ACCORDI PER
L’INNOVAZIONE” (2021–2026).
He is also an Associate Editor of the IEEE JOURNAL OF
ELECTROMAGNETICS, RF AND MICROWAVES IN MEDICINE AND
BIOLOGY.
This article has been accepted for publication in IEEE Open Journal of Antennas and Propagation. This is the author's version which has not been fully edited and
content may change prior to final publication. Citation information: DOI 10.1109/OJAP.2024.3431092
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License. For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
