Academic Edi o s: Ped o Pinho and
Ch is ian Vollai e
Recei ed: 13 Janua y 2025
Re ised: 17 Feb ua y 2025
Accep ed: 20 Feb ua y 2025
Published: 22 Feb ua y 2025
Ci a ion: Abounas , J.; Gha bi, M.E.;
Ga cía, R.F.; Gil, I. A High-Sensi i i y
Inkje -P in ed Flexible Resona o o
Moni o ing Dielec ic Changes in
Mea . Senso s 2025,25, 1338. h ps://
doi.o g/10.3390/s25051338
Copy igh : © 2025 by he au ho s.
Licensee MDPI, Basel, Swi ze land.
This a icle is an open access a icle
dis ibu ed unde he e ms and
condi ions o he C ea i e Commons
A ibu ion (CC BY) license
(h ps://c ea i ecommons.o g/
licenses/by/4.0/).
senso s
A icle
A High-Sensi i i y Inkje -P in ed Flexible Resona o o
Moni o ing Dielec ic Changes in Mea
Jamal Abounas * , Ma iam El Gha bi , Raúl Fe nández Ga cía and Ignacio Gil
Depa men o Elec onic Enginee ing, Uni e si a Poli ècnica de Ca alunya, 08222 Te assa, Spain;
[email p o ec ed] (M.E.G.); [email p o ec ed] (R.F.G.); [email p o ec ed] (I.G.)
*Co espondence: [email p o ec ed]
Abs ac : This pape in oduces a lexible loop an enna-based senso op imized o eal-
ime moni o ing o mea quali y by de ec ing changes in dielec ic p ope ies o e a six-day
s o age pe iod. Ope a ing wi hin he 2.4 GHz ISM band, he senso is designed using CST
Mic owa e S udio 2024 o deli e high sensi i i y and accu acy. The sensing mechanism
le e ages esonance equency shi s caused by a ia ions in pe mi i i y as he mea de-
g ades. Expe imen al alida ion ac oss i e samples showed a consis en equency shi
om 2.14 GHz (Day 0) o 1.29 GHz (Day 5), wi h an a e age sensi i i y o 0.173
GHz/day
.
A s ong co ela ion was obse ed be ween measu ed and simula ed esul s, as e idenced
by linea eg ession (
R2=
0.984 and
R2=
0.974 o measu ed and simula ed da a, e-
spec i ely). The senso demons a ed high p ecision and epea abili y, alida ed by low
s anda d de ia ions and minimal equency de ia ions. Compac , p in able, and cos -
e ec i e, he p oposed senso o e s a scalable solu ion o ood quali y moni o ing. I s
obus pe o mance highligh s i s po en ial o in eg a ion in o IoT pla o ms and ex en-
sion o o he pe ishable ood p oduc s, ad ancing eal- ime, non-in asi e, RF-based ood
sa e y echnologies.
Keywo ds: lexible loop an enna; inkje -p in ed senso ; mic owa e esonance; mea
eshness moni o ing; dielec ic p ope y analysis; eal- ime ood quali y assessmen ;
pe mi i i y-based sensing; non-in asi e moni o ing; 2.4 GHz ISM band; p in ed
elec onics
1. In oduc ion
Food sa e y and quali y a e c i ical aspec s o global public heal h and ood secu i y.
The inc easing complexi y o ood supply chains, coupled wi h g owing consume demand
o esh, high-quali y p oduc s, equi es obus moni o ing sys ems [
1
]. Ensu ing ha
ood, especially pe ishable i ems such as mea , emains sa e and o high quali y h oughou
i s s o age and dis ibu ion is essen ial o p e en ing oodbo ne illness, educing economic
losses, and building consume con idence [
2
,
3
]. Con amina ed o subs anda d ood c ea es
se ious heal h isks o consume s leading o illnesses associa ed wi h inges ion o ood,
which include acu e bowel diso de s, ch onic illness, and mo ali y [
4
]. I is epo ed by
he Wo ld Heal h O ganiza ion (WHO) ha millions o people ac oss he globe su e
om oodbo ne diseases e e y yea wi h, con amina ion by pa hogenic o ganisms such
as Salmonella, Lis e ia, and E. coli as key ac o s. I is also impo an o ealize ha he
spoilage o mea leads o he gene a ion o poisonous by-p oduc s as well as he g ow h o
pa hogens, hus wa an ing he need o s ingen supe ision [
5
]. On a wo ldwide scale,
ood sa e y and ood quali y a e impo an issues o ood secu i y. Spoilage and was e
appea as a challenge o ood secu i y, especially in unde de eloped coun ies wi h e y
Senso s 2025,25, 1338 h ps://doi.o g/10.3390/s25051338
Senso s 2025,25, 1338 2 o 16
li le cold chain acili ies. A de eloped means o supe ision may help o manage ood
was e by acing he mos suscep ible p oduc s d essed o be los in he chain, which will
help o ake p ope measu es as ea ly as possible.
Technological ad ances in adio equency (RF) and mic owa e echnologies ans-
o m ood sa e y and quali y assu ance by p o iding eal- ime, non-in asi e moni o ing
solu ions [
6
]. These echniques le e age he in e ac ion be ween elec omagne ic wa es
and ood ma e ials o de ec ce ain key quali y pa ame e s such as spoilage. RF-based
sys ems, including an enna-based senso s and RFID senso s, allow o con inuous moni o -
ing o en i onmen al condi ions and dielec ic p ope y changes in ood [
7
]. By o e ing
a sophis ica ed and scalable app oach o ood moni o ing, RF and mic owa e echniques
pa e he way o sa e and mo e e icien ood quali y managemen . In he con ex o ood
sa e y and quali y assu ance, dielec ic p ope ies, pa icula ly pe mi i i y, se e as eliable
indica o s o s uc u al and composi ional changes in pe ishable oods [
8
]. As ood ipens
o spoils, i s chemical composi ion unde goes signi ican al e a ions, leading o measu able
changes in i s dielec ic p ope ies. These changes make pe mi i i y a co ne s one o
adio equency (RF)- and mic owa e-based sensing and moni o ing echnologies [
9
]. By
accu a ely measu ing hese p ope ies, ad anced sys ems can deli e eal- ime insigh s
in o ood eshness, enhancing sa e y, educing was e, and ensu ing be e quali y con ol
h oughou he supply chain.
A ained indi idual, such as a s o e manage , can assess he quali y o mea by sigh ,
smell, and ouch. Howe e , his app oach poses po en ial heal h isks o he inspec o .
In addi ion, elying solely on subjec i e judgmen leads o he possibili y o e o s ha
a e no suppo ed by objec i e da a. Fu he mo e, when ouch and smell a e used in
he inspec ion p ocess, i becomes impossible o assess he packaged p oduc s [
10
]. The
exis ing mea quali y moni o ing sys ems a e cos ly o implemen and no widely u ilized.
Despi e hei p esence, cases o heal h issues and a ali ies esul ing om he consump ion
o spoiled mea a e s ill occu .
To add ess hese challenges, se e al s udies ha e demons a ed he e ec i eness o
RF and mic owa e echniques o moni o ing pe ishable oods such as mea , dai y, and
sea ood [
11
–
13
]. These me hods o e ad an ages o e adi ional app oaches, including
as e p ocessing imes, non-des uc i e analysis, and imp o ed scalabili y. Fo example, a
UHF RFID sys em was used o moni o ozen mea using ecei ed signal s eng h indica-
o (RSSI) da a, as epo ed in [
14
]. This me hod demons a ed mono onic ela ionships
be ween RSSI alues, empe a u e, and ha dness du ing de os ing, enabling e ec i e cold
chain moni o ing and sa e y. Ano he echnique was in oduced in [
15
], whe e dielec ic
measu emen s we e used o assess mea eshness by moni o ing changes in pe mi i i y
and conduc i i y o e s o age in e als. Using capaci ance and conduc ance da a measu ed
wi h an LCZ me e ac oss equencies om 10 kHz o 1 MHz, he s udy demons a ed a
decline in hese dielec ic p ope ies wi h inc easing s o age ime. In addi ion, an IoT-based
sys em combining came as and ai quali y senso s, using deep lea ning models o analyze
colo changes and p edic mea eshness in eal ime, was epo ed in [
16
]. This sys em
equi es complex calib a ion o di e en ypes and condi ions o mea , and i s applicabili y
may be limi ed o speci ic use cases. Howe e , all o he s udies men ioned abo e ely
p ima ily on igid subs a es, and hese kinds o subs a es may no be ideal o applica ions
ha equi e lexibili y, such as in eg a ion in o sma packaging o wea able sys ems o
eal- ime moni o ing.
This pape p esen s a lexible an enna-based senso o moni o ing mea (bee ) esh-
ness o e six days using mic owa e signals. The senso , a ci cula loop an enna p in ed on
a lexible polyamide subs a e, de ec s shi s in esonance equency caused by changes in
he mea ’s dielec ic p ope ies as i deg ades. These shi s al e he an enna’s esonance
Senso s 2025,25, 1338 3 o 16
cha ac e is ics, enabling non-in asi e eshness de ec ion. Measu emen s we e conduc ed
using a Vec o Ne wo k Analyze (VNA) o obse e esonance equency shi s o e ime.
The analysis was limi ed o six days because he mea was isibly o en beyond
his pe iod, making u he in es iga ion unnecessa y. Addi ionally, esonance equency
esul s on subsequen days showed no signi ican changes compa ed o Day 6, con i ming
comple e spoilage, which is e iden h ough isual and senso y inspec ion. Ex ending he
analysis u he was, hus, unnecessa y o ood quali y assessmen .
2. Senso Design and Wo king P inciple
2.1. Senso Design and Manu ac u ing
The senso , based on a loop an enna design, was de eloped o ope a e wi hin he
2.4 GHz
ISM band and op imized using CST Mic owa e S udio 2024. The geome ical pa-
ame e s o he p oposed an enna a e de ailed in Table 1and illus a ed in
Figu es 1and 2
.
These pa ame e s we e uned o achie e op imal pe o mance and ensu e e icien
elec omagne ic coupling wi h he a ge samples. Polyimide (PI) ilms, commonly known
by he b and name Kap on, we e selec ed as he subs a e ma e ial due o hei excellen
dielec ic p ope ies (
ϵ =
3.5, loss angen = 0.0027), ligh weigh s uc u e, and du abili y
unde ex eme en i onmen al condi ions. To ensu e consis ency o high-pe o mance
mic owa e applica ions, he elec ical p ope ies o he subs a e we e e i ied using a
Q-me e , as shown in Figu e 1a.
Figu e 1. O e iew . o he expe imen al se up and ab ica ion p ocess: (a) Q-me e used o
ex ac ing elec ical p ope ies o he subs a e; (b) Vol e a NOVA inkje and ex usion p in e used
o an enna ab ica ion; (c) Memme o en o d ying and cu ing he p in ed ink o ensu e p ope
adhesion and dis ibu ion; (d) inal ab ica ed loop an enna; (e) 10 g o mea sample; ( ) ec o
ne wo k analyze (VNA) o an enna pe o mance alida ion.
The conduc i e aces o he loop an enna we e ab ica ed using an inkje and ex usion
p in e (Vol e a NOVA, Figu e 1b) wi h a 225
µ
m nozzle. This sys em simpli ies he
calib a ion p ocess by au oma ing pa ame e s such as dispensing heigh , ink p essu e, and
empe a u e, making i highly use - iendly and e icien . The p in ing p ocess ypically
equi es less han 15 min o comple e. Inkje p in ing was chosen o i s apid p o o yping
capabili ies, minimal ma e ial was e, and sui abili y o lexible subs a es, making i an
ideal echnique o disposable and scalable ood quali y senso s.
A e p in ing, he an ennas we e d ied in a Memme o en (Figu e 1c) a 50
◦
C o
15 min o ensu e p ope ink adhesion and uni o m dis ibu ion. This s ep is c i ical o
Senso s 2025,25, 1338 4 o 16
main aining eliable conduc i i y and achie ing s able senso pe o mance. The inal
ab ica ed loop an enna is shown in Figu e 1d, and i demons a ed excellen elec ical
ma ching o he s anda d 50
Ω
impedance. This pe o mance was alida ed using a ec o
ne wo k analyze (VNA, Figu e 1 ).
To assess i s unc ionali y, he senso was es ed wi h a mea sample (Figu e 1e) placed
di ec ly on he loop in he sensing a ea illus a ed in Figu e 2. Measu emen s we e collec ed
o e mul iple days o analyze he impac o pe mi i i y a ia ions on he e lec ion coe i-
cien (
S11
) and esonance equency beha io . This obus and sys ema ic design ensu es
he senso ’s eliabili y o dielec ic sensing applica ions in mea quali y e alua ion.
Figu e 2. The geome ical pa ame e s o he p oposed an enna.
Table 1. Geome ical pa ame e s o he p oposed senso .
Pa ame e Value (mm) Desc ip ion
WSL50 Wid h o he subs a e
LSL60 Leng h o he subs a e
WL3 Wid h o he loop
RL20 Radius o he loop
W 10 Wid h o he eedline
gL4 Gap be ween loop and eedline
L 13 Leng h o he eedline
hLoop 0.19 Subs a e hickness
0.008 Conduc o hickness
2.2. Wo king P inciple
The p oposed loop an enna senso ope a es by gene a ing a localized elec omagne ic
(EM) ield in i s nea - ield egion. When a ma e ial, such as mea , is placed wi hin his ield,
he elec omagne ic wa es in e ac wi h he dielec ic p ope ies o he ma e ial, such as
i s ela i e pe mi i i y (
ϵ
). This in e ac ion modi ies he an enna’s impedance by ei he
abso bing o e lec ing he EM ene gy, esul ing in measu able changes in he e lec ion
coe icien (S11) and a shi in he esonance equency ( ).
A he esonance equency, he an enna’s impedance is minimized, enabling s ong
ene gy coupling wi h he su ounding medium. This occu s when he induc i e eac ance
(
XL
) o he loop an enna ma ches he capaci i e eac ance (
XC
) o he su ounding sys em.
The wo eac ances cancel each o he ou , lea ing only he esis i e componen in he
o al impedance:
Z o =R+j(XL−XC)(1)
A esonance, since XL=XC, he impedance simpli ies o
Z o =R(2)
Senso s 2025,25, 1338 5 o 16
This pu ely esis i e impedance esul s in he lowes impedance alue, allowing maxi-
mum cu en low in he loop. The s ong cu en low enhances he in e ac ion be ween
he an enna and he su ounding ma e ial, ampli ying he senso ’s sensi i i y o dielec ic
changes. As he dielec ic p ope ies o he ma e ial a ec he EM ield dis ibu ion, he
esonance equency and e lec ion coe icien shi acco dingly.
The in e ac ion be ween he loop an enna and he ma e ial is illus a ed in Figu e 3,
which highligh s how he loop an enna gene a es a localized EM ield ha in e ac s wi h
he ma e ial. The highes in ensi y egions, shown in ed and o ange, a e concen a ed
nea he loop an enna edges, while he ield weakens adially ou wa d. The mea sample
pe u bs he ield, as e idenced by he edis ibu ion o ield lines, con i ming he an enna’s
sensi i i y o dielec ic a ia ions in he sample.
The speci ic composi ion o he mea , along wi h he e ec i e dielec ic p ope ies
used in his s udy, is de ailed in he ollowing subsec ion. These p ope ies we e ca e ully
selec ed o ensu e accu a e simula ion and analysis o he in e ac ion be ween he loop
an enna and he mea samples.
Figu e 3. Simula ed elec ic ield highligh ing he in e ac ions be ween he loop an enna and he
mea sample: (a) side iew and (b) 3D iew o he E ield.
3. Expe imen al Se up and Measu emen Wo k low
To in es iga e he dielec ic p ope ies o bee samples and hei in e ac ion wi h
he loop an enna, a sys ema ic expe imen al wo k low was implemen ed. Each sample
was ca e ully p epa ed o weigh exac ly 10 g, wi h a composi ion o 80% muscle and 20%
a , e lec ing a ypical p o ile o e alua ing mea quali y. The samples we e cylind ical
in shape, wi h a adius o 1.5 cm and a heigh o 1.5 cm, co esponding o a olume
o app oxima ely 10.61 cm
3
. These geome ic and composi ional pa ame e s we e used
o bo h expe imen al measu emen s and simula ion models o ensu e consis ency and
accu acy in he analysis. Fi e samples we e p epa ed and es ed daily o e a six-day pe iod,
p o iding a obus da ase o analysis. The ela i e pe mi i i y (
ϵ
) and loss angen (
an δ
)
o he mea we e calcula ed using a weigh ed a e age app oach based on he dielec ic
p ope ies o muscle and a , ensu ing accu a e modeling and analysis. These p ope ies
we e de i ed using he ollowing equa ions [17,18]:
ϵe = muscle ·ϵmuscle + a ·ϵ a (3)
an δ=σ
ωϵ0ϵ (4)
whe e he a iables a e as ollows:
•ϵe : e ec i e ela i e pe mi i i y o he mix u e.
• muscle =0.8 and a =0.2: weigh ac ions o muscle and a .
•ϵmuscle and ϵ a : ela i e pe mi i i ies o muscle and a .
Senso s 2025,25, 1338 6 o 16
•σ: conduc i i y o he mix u e.
•ω=2π : angula equency, whe e =2.4 GHz.
•ϵ0=8.854 ×10−12 F/m: pe mi i i y o ee space.
Using hese equa ions and he dielec ic p ope ies o muscle (
ϵmuscle ≈
49,
σmuscle ≈
1.3
S/m
) and a (
ϵ a ≈
5.2,
σ a ≈
0.05
S/m
), he e ec i e p ope ies o he mix u e we e
calcula ed as ollows:
ϵe = (0.8 ·49) + (0.2 ·5.2)≈40.96 (5)
an δ=1.048
2π·2.4 ×109·8.854 ×10−12 ·40.96 ≈0.020 (6)
These calcula ed alues we e used o simula e he elec omagne ic in e ac ion be ween
he loop an enna and he mea sample.
The s o age condi ions we e ca e ully con olled o simula e ealis ic handling sce-
na ios while minimizing en i onmen al dis u bances. Each sample was s o ed o 12 h a
7
◦
C ( e ige a ed) o slow deg ada ion, ollowed by 8 h a 25
◦
C ( oom empe a u e) o
mimic ypical exposu e condi ions. This cycle was epea ed daily o e lec p ac ical s o age
a ia ions. To mi iga e he in luence o ambien humidi y and empe a u e luc ua ions, all
measu emen s we e pe o med immedia ely a e emo ing he samples om e ige a ion,
ensu ing consis ency in dielec ic beha io and minimizing ex e nal in e e ence.
The expe imen al se up included a loop an enna designed o ope a e wi hin he
2.4 GHz
ISM band and a ec o ne wo k analyze (VNA) o measu e he e lec ion coe -
icien (
S11
) o e a equency ange o 1–3 GHz, as p esen ed in Figu es 4and 5. Each
sample was ca e ully posi ioned on he loop an enna o ensu e p ope coupling wi h he
elec omagne ic ield. The placemen o he sample on he loop an enna allowed di ec
in e ac ion be ween he localized elec omagne ic ield and he dielec ic p ope ies o
he mea .
Figu e 4. Th ee-dimensional model ep esen a ion o he expe imen al p o ocol o assessing mea
quali y using mic owa e sensing. The model depic s esh samples (Day 0) and aged samples (Day 5)
placed in con aine s, connec ed o a mic owa e sensing se up and a ec o ne wo k analyze (VNA).
Senso s 2025,25, 1338 7 o 16
Figu e 5. Comp ehensi e expe imen al se up o analyzing dielec ic p ope y a ia ions in mea
samples om Day 0 o Day 5.
Fo da a collec ion,
S11
measu emen s we e eco ded ac oss he equency band o
each sample. Fi e samples we e measu ed daily, s a ing om Day 0 ( esh mea ) o Day 5
(s o ed mea ). Measu emen s we e epea ed unde iden ical condi ions each day o ack
ends in esonance equency o e ime. Da a acquisi ion and analysis we e pe o med
using MATLAB 2024B, which acili a ed he compa ison o
S11
a ia ions ac oss equencies
and he s a is ical analysis o all samples.
Da a p ocessing in ol ed analyzing he shi s in esonance equency, iden i ying
changes in he e lec ion coe icien , and pe o ming s a is ical compa isons ac oss all days,
as desc ibed in Figu e 4. This comp ehensi e app oach enabled p ecise moni o ing o
changes in he dielec ic p ope ies o he mea o e ime, p o iding aluable insigh s in o
he ela ionship be ween mea quali y and i s elec omagne ic esponse.
4. Resul s and Discussion
The expe imen al esul s in his sec ion p esen he capabili y o he loop an enna o
de ec a ia ions in he dielec ic p ope ies o mea samples s o ed unde con olled condi-
ions, as shown in Figu e 5. By moni o ing changes in he e lec ion coe icien (
S11
) and
esonance equency o e ime, he an enna’s sensi i i y o changes in pe mi i i y caused
by mea aging, s o age, and en i onmen al a ia ions is assessed. The measu emen s we e
benchma ked agains simula ed da a o con i m consis ency and es ablish a co ela ion
be ween expe imen al and heo e ical obse a ions.
Addi ionally, s a is ical analyses we e pe o med o e alua e he epea abili y o he
measu emen s and quan i y he obse ed equency shi s esul ing om he e ec s o aging,
s o age, and en i onmen al ac o s on he mea samples. The esul s a e discussed wi h a
ocus on he senso ’s pe o mance, he in luence o ma e ial p ope ies on measu emen
accu acy, and po en ial applica ions in ood quali y assessmen .
4.1. An enna Senso E alua ion in F ee Space
The e lec ion coe icien (
S11
) o he loop an enna was e alua ed bo h expe imen ally
and h ough simula ion o benchma k he an enna’s pe o mance in ee-space condi ions.
Figu e 6p esen s a compa a i e analysis o he measu ed and he simula ed
S11
as a unc ion
o equency wi hin he 1 GHz o 3 GHz ange. The simula ed esponse exhibi s a sha p
esonance dip a 2.4 GHz, which co esponds o he an enna’s design equency in he ISM
Senso s 2025,25, 1338 8 o 16
band. The measu ed esponse closely ollows he simula ion, showing a esonance dip a
app oxima ely 2.14 GHz. This co esponds o a equency shi o app oxima ely 10.83%.
The measu ed esponse is sligh ly lowe in equency han he simula ed esponse,
which can be a ibu ed o ab ica ion ole ances and a ia ions in ma e ial p ope ies.
Fac o s such as he ac ual pe mi i i y and hickness o he subs a e, which may de ia e
sligh ly om simula ion inpu s, could lead o his shi . Addi ionally, en i onmen al
condi ions du ing measu emen migh con ibu e o u he de ia ions.
Despi e his shi , he close ag eemen be ween he simula ed and measu ed da a
alida es he design p ocess and highligh s he e ec i eness o he ab ica ion me hod.
The minimum e lec ion a esonance con i ms good impedance ma ching o he s anda d
50
Ω
sys em, ensu ing e icien ene gy coupling and minimal powe loss a he ope a ional
equency. This benchma k es ablishes a eliable e e ence o subsequen expe imen s
in ol ing a ying pe mi i i y condi ions caused by mea samples.
Figu e 6. Compa ison o measu ed and simula ed S11 in ee space.
4.2. Pe mi i i y Va ia ion O e Days
To alida e he hypo hesis ega ding he in luence o pe mi i i y a ia ion on es-
onance equency shi s, a pa ame ic sweep simula ion was conduc ed in CST S udio
Sui e. This simula ion u ilized a cylind ical mea model wi h a adius o 1.5 cm, a heigh
o 1.5 cm, and a olume o app oxima ely 10.61 cm
3
, consis en wi h he expe imen ally
p epa ed samples. The ela i e pe mi i i y (
ϵ
) o he modeled mea sample was a ied,
wi h ini ial and inal alues anging om
ϵ =
37 o
ϵ =
91, as de i ed and jus i ied in
ea lie calcula ions. The goal o his simula ion was o obse e and quan i y he impac o
changing dielec ic p ope ies on he esonance equency beha io o he an enna sys em.
Figu e 7illus a es he esul s o he simula ion, showing a clea end in he equency
esponse: as he ela i e pe mi i i y o he ma e ial inc eased, he esonance equency
consis en ly shi ed owa ds lowe alues. Fo he ini ial pe mi i i y alue (
ϵ =
37), he
simula ed esonance equency was app oxima ely 2.1 GHz, co esponding o a s a e whe e
elec omagne ic coupling wi h he ma e ial was ela i ely low.
Howe e , as he pe mi i i y inc eased o i s maximum alue o
ϵ =
91, he esonance
equency dec eased signi ican ly o app oxima ely 1.3 GHz. This ma ked shi e lec s
he inc eased ene gy s o age capaci y o highe -pe mi i i y ma e ials, which modi ies he
elec omagne ic ield dis ibu ion and esul s in a educed esonance equency.
Senso s 2025,25, 1338 9 o 16
Figu e 7. Simula ed
S11
esponse o e he equency ange o a ious pe mi i i y alues co e-
sponding o di e en days.
The obse ed end p esen ed in Figu e 8aligns closely wi h heo e ical p edic ions,
whe e ma e ials wi h highe pe mi i i y a e expec ed o exhibi s onge elec omagne ic
coupling due o inc eased pola iza ion. This e ec educes he an enna’s ope a ing e-
quency as he sys em’s induc i e and capaci i e eac ance componen s adjus o he new
ma e ial p ope ies. Such beha io is indica i e o he senso ’s sensi i i y o he dielec ic
p ope ies o he su ounding medium and highligh s i s po en ial o de ec ing sub le
a ia ions in ma e ial composi ion.
By pe o ming a comp ehensi e pa ame ic sweep simula ion, he pe mi i i y alues
we e s a egically selec ed o align wi h he equencies obse ed in he measu emen
esul s (discussed in la e sec ions). This app oach ensu es ha he simula ed equency
shi s co espond closely o he expe imen ally ob ained da a, he eby alida ing he elia-
bili y o he p oposed sensing mechanism. The consis ency be ween he simula ion and
measu emen ou comes highligh s he an enna’s capabili y o e ec i ely moni o dielec ic
changes o e ime.
Figu e 8. Dynamic ela ionship be ween simula ed pe mi i i y and equency ac oss days.
Senso s 2025,25, 1338 16 o 16
7.
Bibi, F.; Guillaume, C.; Gon a d, N.; So li, B. A e iew: RFID echnology ha ing sensing ap i udes o ood indus y and hei
con ibu ion o acking and moni o ing o ood p oduc s. T ends Food Sci. Technol. 2017,62, 91–103. [C ossRe ]
8.
El Khaled, D.; No as, N.; Gazquez, J.A.; Ga cia, R.M.; Manzano-Aguglia o, F. F ui and ege able quali y assessmen ia dielec ic
sensing. Senso s 2015,15, 15363–15397. [C ossRe ] [PubMed]
9. Kim, S. Inkje -p in ed elec onics on pape o RF iden i ica ion (RFID) and sensing. Elec onics 2020,9, 1636. [C ossRe ]
10. VanO e beke, D. Handbook o Bee Sa e y and Quali y; CRC P ess: Boca Ra on, FL, USA, 2024.
11.
Wang, M.; Yang, Y.; Mu, B.; Niki ina, M.A.; Xiao, X. Millime e wa e-based non-des uc i e biosenso sys em o li e ish
moni o ing. Biosenso s 2022,12, 541. [C ossRe ] [PubMed]
12.
Wang, D.; Zhang, M.; Jiang, Q.; Mujumda , A.S. In elligen sys em/equipmen o quali y de e io a ion de ec ion o esh ood:
Recen ad ances and applica ion. Foods 2024,13, 1662. [C ossRe ] [PubMed]
13.
Beshai, H.; Sa abha, G.K.; Ra hi, P.; Alam, A.U.; Deen, M.J. F eshness moni o ing o packaged ege ables. Appl. Sci. 2020,10, 7937.
[C ossRe ]
14.
Ma ind a, A.M.J.; P a ama, B.M.; Su oso, D.J. Non-in asi e ozen mea moni o ing sys em using UHF RFID ag an enna-based
sensing and RSSI. In . J. Ad . Sci. Eng. In . Technol. 2023,13, 1–7. [C ossRe ]
15.
Gha ass, Z.F.; Soliman, M.M.; Mohamed, M.M. Dielec ic echnique o quali y con ol o bee mea in he ange 10 kHz–1 MHz.
Am.-Eu asian J. Sci. Res. 2008,3, 62–69.
16.
Kim, D.-E.; Mai, N.-D.; Chung, W.-Y. AIoT-based mea quali y moni o ing using came a and gas senso wi h wi eless cha ging.
IEEE Sens. J. 2023,24, 7317–7324. [C ossRe ]
17. Sih ola, A.H. Elec omagne ic Mixing Fo mulas and Applica ions; IET: S e enage, UK, 1999; Volume 47.
18.
Poza , D.M. Mic owa e Enginee ing, 4 h ed.; Uni e si y o Massachuse s a Amhe s , John Wiley & Sons, Inc.: Hoboken, NJ, USA,
2012; pp. 26–30.
19.
Kim, D.-E.; Nando, Y.A.; Chung, W.-Y. Ba e y- ee po k eshness es ima ion based on colo ime ic senso s and machine lea ning.
Appl. Sci. 2023,13, 4896. [C ossRe ]
20.
Cal e -Chau a d, M.; Jaque Gonzalez, P.; Vé onèse, T.; Dubuc, D.; G enie , K. Mic owa e-based senso dedica ed o he
cha ac e iza ion o mea eshness. In P oceedings o he 2020 IEEE MTT-S In e na ional Mic owa e Biomedical Con e ence
(IMBioC), Toulouse, F ance, 14–17 Decembe 2020 ; pp. 1–4.
21. Li, Z.; Suslick, K.S. Po able op oelec onic nose o moni o ing mea eshness. ACS Sens. 2016,1, 1330–1335. [C ossRe ]
22.
Mu ado , M.; Cullen, J.; Mason, A. Real- ime moni o ing o mea d ying p ocess using elec omagne ic wa e senso s. In Nex
Gene a ion Senso s and Sys ems; Sp inge : Be lin/Heidelbe g, Ge many, 2016; pp. 221–233.
23.
Nguyen, N.H.; Lam, M.B.; Chung, W.-Y. Ba e y-less po k eshness moni o ing based on high-e iciency RF ene gy ha es ing.
J. Sens. Sci. Technol. 2020,29, 293–302. [C ossRe ]
Disclaime /Publishe ’s No e: The s a emen s, opinions and da a con ained in all publica ions a e solely hose o he indi idual
au ho (s) and con ibu o (s) and no o MDPI and/o he edi o (s). MDPI and/o he edi o (s) disclaim esponsibili y o any inju y o
people o p ope y esul ing om any ideas, me hods, ins uc ions o p oduc s e e ed o in he con en .
