Int
ern
at
i
onal
Journ
al of Inf
orm
at
ic
s
and
Co
m
munic
at
i
on
Tec
hn
olog
y (IJ
-
I
CT)
Vo
l.
7
,
No.
2
,
A
ugus
t
201
8
,
pp
.
77
~
81
IS
S
N:
22
52
-
8776
, DO
I: 10
.11
591/iji
ct
.
v7
i
2.p
p77
-
81
77
Journ
al h
om
e
page
:
http:
//
ia
escore.c
om/j
ourn
als/i
ndex.
ph
p/IJI
C
T
Eva
lu
ation o
f Wat
er
C
ontent in
Milk
Usin
g Microw
ave
Tra
nsmi
ssion Ap
proach with
Ho
rn Ant
en
na
Ashok
.
Ja
*
,
K
ir
an
P
at
il
Depa
rtment
Of
El
e
ct
roni
cs
and
Te
l
ec
om
m
unic
ation
Annasheb
Coll
eg
e
of
Engi
n
e
eri
ng
and Te
chn
olog
y
Maha
rasthr
a
india
,
Ind
ia.
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
A
pr
12
, 201
8
Re
vised Ju
n
15
, 2018
Accepte
d
J
un
20
, 201
8
As
the
d
ia
r
y
ind
ustrie
s
de
li
ve
r
y
t
he
m
il
k
to
cons
um
er
bott
l
es
in
pouche
s
th
e
thi
ckn
ess
of
the
m
il
k
is
var
i
ed
with
e
ac
h
compan
y
.
Th
e
m
ic
rowave
tra
nsm
ission
li
ne
appr
oa
ch
with
horn
ant
enn
a
is
int
roduc
ed
as
a
n
eff
ectiv
e
m
et
hod
for
the
a
ss
ess
m
ent
of
m
il
k
for
the
f
irst
t
i
m
e
with
diffe
r
en
t
fre
quen
c
y
.
The
el
e
ct
r
ic
a
l
p
e
rm
it
ti
vity
is
an
i
ntri
nsic
par
ame
t
er
of
a
m
a
te
ri
al
tha
t
ca
n
be
used
as
an
inde
x
of
d
eliver
ed
wa
te
r
content
.
For
t
he
p
ermitt
iv
ity
c
al
cu
la
t
ion
,
we
use
horn
an
t
enna
tra
nsm
it
t
er
and
recei
v
er
in
m
ic
rowave
fre
q
uency
.
Th
e
m
ic
rowave
tra
ve
l
le
d
via
th
e
m
il
k
accordi
ng
to
it
s
thi
ckn
ess.
A
des
ign
of
ho
r
n
ant
enn
a
and
i
ts minimum
dista
nce
tr
ansm
ission
l
engt
h
is
foll
owe
d
to
eva
lu
ate
the
thickness
of
m
il
k.
Th
e
d
issipat
ed
power
is
ca
l
cul
a
te
d
with
the
h
el
p
o
f
VS
W
R
m
et
er
in
dec
ib
el.
Th
e
r
e
sults
are
p
lo
tted
with
diff
ere
n
t
m
ic
rowave
fre
quency
wi
th
unique
wave
l
en
gth
wi
th
un
ique
conc
en
tra
t
ion
of
wat
er
with
m
il
k.
Mat
la
b
i
s
used
to
plo
t
t
he
gra
ph
.
Th
e
e
ffe
ct
s
exhi
bi
t
th
at
th
e
m
il
k
thi
ckn
ess
m
ake
s
the
w
ave
tra
v
el
of
m
ic
rowave
produc
es
th
e
de
ci
be
l
va
lue
cha
ng
es
at
the
o
utput
.
Th
e
m
ea
surem
ent
s
demon
strat
e
th
at
th
e
p
e
rm
it
ti
vity
of
m
il
k
inc
re
ase
s
with
increa
sed
ad
ded
wate
r
.
A
re
l
at
ionship
b
et
wee
n
the
add
ed
wate
r
cont
en
t
an
d
the
p
ermitt
iv
ity
of
m
il
k
-
wat
er
m
ixt
ure
is
est
ablished,
whi
ch
should
be a
pow
erf
ul tool
fo
r
d
etec
t
ing
m
il
k
thick
ness.
Ke
yw
or
d:
An
te
nn
a
Ele
ct
rical
p
erm
it
ti
vity
Mi
cro
wa
ve
t
ra
ns
m
issi
on
Copyright
©
201
8
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Asho
k.
Ja
,
Dep
a
rtm
ent o
f El
ect
ro
nics
and
Tel
ec
omm
un
ic
at
ion
,
Annas
he
b
Coll
eg
e
of Engine
e
rin
g
a
nd Tech
nolo
gy,
Ma
har
ast
hr
ai
ndia
,
India
.
Em
a
il
:
j
ahangir@s
outher
n.
e
du.
bd
1.
INTROD
U
CTION
The
a
ve
rag
e
co
m
po
sit
ion
of
m
il
k
is
a
bout
87.
7%
water
,
4.9
%
la
ct
os
e
(car
bohydrat
e),
3.4
%
fat,
3.3%
protei
n,
a
nd
0.7%
m
inerals.
I
t
has
bee
n
us
e
d
as
a
high
nu
t
riti
on
al
f
ood
or
as
a
rem
edy
fo
r
m
any
diseas
es.
F
or
these
be
nef
it
s
to
m
a
te
rial
ise
,
m
il
k
sh
ould
be
fr
ee
from
wat
er
an
d
oth
er
s
weetene
rs.
Howev
e
r,
for
ec
onom
ic
gain,
m
on
ey
ha
s
bec
om
e
a
ta
rg
et
of
diluti
on
of
m
il
k
with
water
w
or
l
dw
i
de.
For
e
xam
ple,
wate
r
is
one
of
th
e
com
m
on
ingre
dients
for
m
il
k.
T
he
thic
kn
e
s
s
is
both
f
ra
udulent
a
nd
unfa
ir
to
c
ons
um
er
s,
a
nd
it
co
uld
cause
healt
h
c
on
ce
r
ns.
T
her
e
fore
,
de
te
rm
inati
on
of
the
t
hick
ness
of
m
il
k
y
is
i
m
portant
for
m
a
nufactu
rer
s
,
retai
le
rs,
consum
ers
an
d re
gu
la
to
ry aut
horiti
es.
Tw
o
ty
pes
of
m
et
ho
ds
a
re
c
om
m
on
ly
us
ed
f
or
m
il
k
qu
al
it
y
assessm
ent:
se
ns
ory
an
d
c
he
m
ic
al
analy
ses
(e.
g.
,
li
qu
i
d
c
hrom
at
og
ra
ph
y,
inf
r
are
d
s
pect
ro
sc
opy
an
d
m
ass
sp
ect
r
ome
try
).
T
he
se
nsory
a
naly
sis
m
ai
nly
detect
s
m
i
lk
co
lour,
viscosity
,
sm
e
ll
,
flavour
and
cryst
al
li
sat
ion
.
H
oweve
r,
the
accu
racy
of
the
se
ns
ory
a
na
ly
sis
is
lim
it
ed
and
us
ua
ll
y
de
pend
s
on
the
ex
per
i
ence
of
the
se
ns
ory
pa
nelli
s
ts.
C
hem
ic
a
l
an
al
ysi
s
al
so
has
so
m
e
disad
va
ntages,
su
c
h
as
need
f
or
highly
sk
il
le
d
la
bo
ur
,
ultr
apure
pr
e
par
at
i
on,
ex
pe
ns
ive
i
ns
tr
um
ents
and
la
rge
instru
m
ental
analy
sis
[
1
-
5]. Thus,
the d
e
ve
lop
m
ent
of
a
re
la
ti
vely
fast,
ea
sy
-
to
-
us
e
an
d
l
ow
-
c
os
t
m
easur
em
ent
m
et
ho
d bec
ome
s att
racti
ve.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2252
-
8776
IJ
-
ICT
V
ol.
7
,
No.
2
,
Aug
us
t
20
1
8
:
77
–
81
78
A
m
ic
ro
wav
e
-
base
d
te
ch
niqu
e
can
be
a
n
al
te
rn
at
ive
m
et
ho
d
for
qual
it
y
assessm
ent
on
the
basis
of
the
el
ect
rical
per
m
it
ti
vity
,
wh
ic
h
is
an
i
ntrin
sic
par
am
et
er
of
a
m
a
te
rial
that
represe
nts
the
interact
ion
wit
h
th
e
el
ect
ro
m
agn
et
ic
f
ie
ld.
Esta
bl
ishing
e
xpli
ci
t
relat
ion
-
s
hips
betwee
n
th
e
per
m
it
t
ivit
y
and
t
he
ag
r
i
-
foo
d
const
it
uen
ts
ca
n
pro
vid
e
a
m
eans
of
ra
pid
inspe
ct
ion
of
pro
per
ti
es
(e.
g.
,
m
oistur
e,
fat
and
sal
t
co
nte
nt)
[6
]
.
Mi
cro
wa
ve
de
te
ct
ion
m
et
ho
ds
hav
e
var
i
ou
s
at
trib
utes,
s
uc
h
a
s
fast
(
few
m
inu
te
s
rat
he
r
t
han
hours)
,
non
-
hazar
dous,
cap
able
of
m
easur
ing
bu
l
k p
rope
rtie
s
and
le
ss
s
ensiti
ve t
o e
nv
i
ronm
ental
conditi
on
s
[7
]
.
A n
um
ber
of
m
ic
ro
wav
e
te
chn
iq
ues
ha
ve
bee
n
re
porte
d
i
n
the
li
te
ratur
e
an
d
us
e
d
in
pr
act
ic
e
f
or
li
qu
ids:
open
-
e
nd
e
d
rectan
gu
la
r
w
aveguide
[
8]/
coax
ia
l
pr
ob
e
[
9]
te
chn
i
qu
es
,
tim
e
do
m
ai
n
sp
ect
r
os
c
op
y/
ref
le
ct
om
et
ry
(TD
R
)
m
et
ho
d
[10],
re
so
na
nce
m
et
hods
(e
.g.,
res
on
a
nt
cavit
y
[11]
a
nd
c
ouple
d
s
plit
-
rin
g
res
onat
or
[
13
]
a
nd
fr
ee
sp
ace
m
et
ho
ds
[
14
]
.
Each
m
et
ho
d
has
it
s
ow
n
pa
rtic
ular
a
dv
a
nt
ages,
disa
dv
a
nt
ages
an
d
a
ppli
cat
ion
s.
S
peci
fical
ly
,
the
open
-
e
nd
e
d
m
et
ho
ds
a
re
pr
on
e
t
o
er
rors
in
tro
du
ce
d
by
i
m
pro
per
c
on
ta
ct
betwee
n
the
prob
e
a
nd
the
m
a
te
rial
su
r
face
(
e.
g.
,
a
ir
ga
ps
or
ai
r
bubble
s).
The
TDR
m
easur
in
g
instr
um
ents
are
ex
pe
n
sive
[12].
T
he
res
onan
c
e
appr
oach
is
i
nhere
ntly
na
rro
wb
a
nd
a
nd
requires
care
fu
l
s
a
m
ple
prepa
rat
ion
an
d
cal
ibra
ti
on
[
7].
F
or
th
e
f
ree
sp
ace
te
ch
niqu
e,
sp
eci
al
at
te
ntion
m
us
t
be
pa
id
to
the
ch
oice
of
horn
a
nten
na
e,
desig
n
of
th
e
sa
m
ple
ho
lde
r
an
d
the
sam
ple
geo
m
et
ry
and
loca
ti
on
.
On
th
e
ot
her
hand,
t
he
tr
ansm
issi
on
li
ne
te
chn
iq
ue
is
r
obus
t,
broa
db
a
nd
a
nd
su
it
ed
to
m
at
erial
s
with
a
wide
range
of
diel
ect
ric
loss.
F
or
the
li
qu
id
m
easur
em
ent,
two
wav
e
guide
cel
ls
with
window
can
be
m
ade
to
ho
l
d
t
he
sam
ple
on
eac
h
side.
How
e
ver,
few
stu
dies
ha
ve
been
m
ade
on
the
m
easur
em
ent of m
il
k
us
ing t
hi
s tech
nique.
In
t
his
w
ork,
the
tra
ns
m
issi
on
li
ne
te
ch
niqu
e
is
adopte
d
f
or
the
st
ud
y
of
m
il
ks
and
m
il
k
adu
lt
erati
on
for
the
fi
rst
tim
e.
First,
the
c
ol
e
-
cole
e
qu
at
io
n
for
the
descr
i
pt
ion
of
diele
ct
ric
pro
per
ti
es
of
li
qu
ids
is
int
rodu
ce
d.
The
n
the
li
qu
i
d
per
m
it
t
ivit
y
m
easur
em
ent
and
the
avail
a
ble
m
et
ho
ds
f
or
the
pe
rm
i
tt
i
vity
cal
culat
ion
a
re
addresse
d.
Fro
m
the
m
easur
e
m
ent
of
disti
ll
ed
wate
r
at
r
oo
m
tem
per
at
ur
e
ov
e
r
tw
o
fr
e
quency
ba
nd
s
(i.e
.,
6
-
8
GH
z
an
d
8
-
12
GH
z
),
t
he
e
ff
ic
ie
ncy
an
d
acc
uracy
of
eac
h
pe
rm
i
tt
ivity
cal
c
ulati
on
m
et
hod
an
d
fr
e
qu
e
ncy
ba
nd
are
e
valuated
by
com
par
is
on
with
the
em
pirical
data
pro
vid
e
d
by
the
c
ole
-
c
ole
eq
uati
on.
Af
te
r
wa
rds,
the
m
easur
em
ents
of
tw
o
ty
pe
s
of
m
il
k
s
are
c
onduct
ed
ove
r
6
-
8
GH
z
.
A
ge
netic
al
gorithm
,
an
op
ti
m
isa
t
ion
m
et
hod,
is
em
plo
ye
d
to
quic
kly
sea
rch
the
possible
pe
rm
i
tt
ivity
of
the
m
il
k
sam
pl
es,
wh
ic
h
will
be
a
sta
rtin
g
point
f
or
furthe
r a
cc
ur
at
e
cal
culat
ion.
I
t
is
al
so
use
d
to
obta
in
the
cole
-
cole
diele
ct
ric
pa
ram
et
ers fo
r
cl
ear
a
nd
set
m
il
ks
.
The
e
ff
ect
of
t
he
cal
ib
rati
on
error
s
on
t
he
a
ccur
acy
of
the
per
m
it
t
ivit
y
m
easur
em
ent
is
s
tud
ie
d.
In
ad
diti
on
,
it
s
accuracy
is
c
om
par
ed
with
t
ha
t
by
a
c
omm
e
rcial
ly
avail
able
diele
ct
ric
pro
be.
Finall
y,
the
eff
e
ct
of
the
a
dde
d
water
on the
p
e
rm
i
tt
ivity of
m
il
k
-
water
m
ixt
ur
e
is in
vestiga
te
d
in
detai
l an
d
a
purity
in
de
x
is i
den
ti
fie
d.
2.
DIA
EL
E
CTR
IC
P
ROP
ERT
IES OF
LIQ
U
IDS
2.1.
Def
ini
ti
on
of
Permi
tt
ivi
ty
Wh
e
n
m
ic
ro
wav
e
e
nergy
is
directed
to
wards
a
m
at
erial
,
pa
rt
of
t
he
e
nergy
is
ref
le
c
te
d,
par
t
i
s
transm
itted
an
d
pa
rt
is
ab
sorb
ed
by
t
he
sam
ple.
The
portio
n
of
e
nergy
that
f
al
ls
into
these
t
hr
ee
cat
e
gories
hav
e
been
de
fine
d
in
te
rm
s
of
the
scat
te
ring
par
a
m
et
ers
and
the
se
can
be
relat
ed
to
the
diele
ct
ric
prop
e
rtie
s
of
t
he
sam
ple
[6
]
.
T
he
f
undam
ental
el
ect
rical
property
thr
ough
w
hich
the
i
ntera
ct
ion
s
betwe
en
the
el
ect
r
om
a
gn
et
i
c
wav
e
and t
he m
at
erial
are
de
scribe
d
is t
he
c
om
plex
pe
rm
i
ttivit
y ε. It is m
at
he
-
m
at
ic
a
ll
y e
xpresse
d
as:
ε ¼ ε
0εr
(1)
Εris
the
relat
iv
e
pe
rm
i
tt
ivity.
The
real
pa
rt
ε
0rof
ε
r,
or
di
el
ect
ric
co
ns
ta
nt,
cha
racteri
ses
t
he
a
bili
ty
of
a
m
at
erial
or
a
su
bst
ance
to
st
or
e
the
el
ect
ric
fiel
d
e
nergy.
ε
00r
(
posit
ive),
or
diele
ct
ric
lo
ss
f
act
or,
ref
le
ct
s
the
abili
ty
o
f
a m
ater
ia
l t
o dissi
pat
e the elec
tric
e
nergy in
the
form
o
f
heat.
The
va
riat
ion
of
t
he
relat
ive
pe
rm
i
tt
ivity
of
w
at
er
with
res
pe
ct
to
f
reque
ncy
and
te
m
per
at
ure
ove
r
1e
20
GH
z
is
pl
otted
in
Fi
g
ure
1.
I
n
this
fr
e
que
nc
y
range,
ε0
r
a
nd
ε
00r
a
re
of
t
he
sam
e
order
of
m
agn
it
ude,
wh
ic
h
dem
on
strat
es t
he hig
h
-
l
os
s c
ha
racteri
sti
c of
water.
2.2
.
Penetr
at
i
on
D
epth
o
f Mic
ro
w
av
e
s in
t
o Diel
ectric
s
Con
si
der
i
ng
t
he
eff
ect
of
the
lossy
m
edium
,
the
power
of
the
m
ic
ro
w
aves
de
cay
s
expo
nen
ti
al
ly
thr
ough
the
thi
ckn
e
ss.
T
he
pe
netrati
on
de
pt
h
dp
is
a
pr
act
ic
al
par
am
et
er
us
ed
for
the
e
valua
-
ti
on
of
s
ign
al
pro
pag
at
io
n.
It
is
def
ine
d
as
the
de
pth
wher
e
the
am
pli
tude
of
the
si
gn
al
is
reduced
t
o
1/
e
(ab
out
37%
)
belo
w
the s
urface.
For a
diele
ct
ric m
at
erial
, dp
ca
n be calc
ulate
d b
y (Poza
r,
20
11):
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
ICT
IS
S
N:
22
52
-
8776
Evalu
atio
n of
Water C
on
te
nt
in Mil
k U
si
ng
Mi
crowave
Tr
an
s
mi
ssio
n Ap
pr
oac
h
wi
th …
(
Ashok.
Ja
)
79
3.
HORN
A
NTENNAS
Horn
a
nten
nas
are
ver
y
po
pu
l
ar
at
U
HF
(
300
MHz
-
3
G
Hz
)
and
higher
f
requen
ci
es
(
I'
ve
h
eard
of
hor
n
anten
nas
ope
ra
ti
ng
as
hi
gh
as
140
GH
z
).
Hor
n
ante
nn
as
oft
en
ha
ve
a
di
recti
on
al
ra
diati
on
patte
rn
with
a
high
anten
na
gain,
wh
ic
h
ca
n
ra
ng
e
up
to
25
dB
in
s
om
e
cases,
with
10
-
20
dB
bein
g
ty
pical
.
Horn
ante
nn
a
s
hav
e
a
wide i
m
ped
anc
e
band
width,
i
m
pl
yi
ng
that
th
e
input
i
m
ped
a
nce
is
slow
ly
va
ryi
ng
ov
e
r a
wide f
re
qu
e
nc
y
ran
ge
(which
al
so
im
plies
lo
w
value
s
f
or
S11
or
V
S
W
R)
.
T
he
ba
ndwidt
h
for
pract
ic
al
ho
r
n
a
nt
enn
as
can
be
on
the
order o
f 2
0:1
(for i
ns
ta
nce
, ope
rati
ng fro
m
1
GH
z
-
20
G
Hz),
w
it
h
a
10:1
b
a
ndwidt
h no
t
be
ing
uncom
m
on
.
The
gai
n
of
hor
n
ante
nn
as
of
te
n
i
ncr
ease
s
(a
nd
t
he
bea
m
width
decr
e
ases)
as
t
he
frequ
e
ncy
of
op
e
rati
on
is
i
nc
reased
.
T
his
i
s
beca
us
e
the
siz
e
of
the
ho
r
n
a
per
t
ur
e
is
a
lway
s
m
easur
ed
in
wav
el
e
ng
t
hs
;
at
higher
fr
e
quen
ci
es
the
hor
n
a
nten
na
is
"el
ect
rical
ly
la
rg
er";
this
is
becau
s
e
a
higher
fr
e
quency
has
a
s
m
al
le
r
wav
el
e
ng
t
h.
Since
the
hor
n
a
nten
na
has
a
f
ixed
ph
ysi
cal
siz
e
(say
a
square
a
pe
rtur
e
of
20
cm
acro
ss,
f
or
instance)
,
th
e
a
per
t
ur
e
is
m
or
e
wav
el
e
ng
t
hs
ac
ro
ss
at
hi
gher
frequ
e
ncies
a
nd
a
rec
urrin
g
the
m
e
in
anten
na
t
heor
y
is t
hat larg
er a
nten
nas (in te
r
m
s o
f
wa
velen
gth
s
in
siz
e
) ha
ve hig
her dire
c
ti
viti
es.
Horn
ante
nn
a
s
ha
ve
ve
ry
li
tt
l
e
loss,
so
t
he
directi
vity
of
a
hor
n
is
r
ough
ly
equ
al
to
it
s
gain.
H
orn
anten
nas
are
s
om
ewh
at
i
ntu
it
ive
a
nd
relat
iv
el
y
si
m
ple
to
m
anu
fact
ur
e
.
I
n
a
ddit
ion
,
ac
ousti
c
ho
rn
a
ntenn
a
s
a
re
al
so
us
e
d
in
tra
ns
m
itti
ng
sound
wa
ves
(
f
or
e
xam
ple,
with
a
m
egap
ho
ne).
Horn
a
nten
nas
are
al
so
o
ften
us
e
d
t
o
feed a
dish
a
nt
enn
a
, or
as a "
s
ta
nd
a
rd g
ai
n"
a
nten
na
in
m
eas
u
rem
ents.
4.
HORN
A
NTENNA
DESI
G
NING
Horns
are
am
on
g
t
he
sim
plest
and
m
os
t
wide
ly
us
ed
m
ic
ro
wav
e
a
nten
nas
and
they
fi
nd
app
li
cat
io
ns
in
the
a
reas
of
wireless
c
omm
un
ic
at
io
ns
,
el
ect
ro
m
agn
et
ic
sensing
RF
heat
ing
a
nd
bio
m
e
dicine
[15].
T
he
hor
n
anten
na
m
ay
be
co
ns
ide
re
d
as
an
RF
tra
nsfo
r
m
er
or
im
ped
ance
m
at
ch
bet
ween
the
wa
ve
gu
i
de
feed
e
r
a
nd
f
re
e
sp
ace
wh
ic
h
ha
s
an
im
ped
an
ce
of
37
7
oh
m
s
by
hav
i
ng
a
ta
pe
red
or
hav
i
ng
a
flare
d
e
nd
to
t
he
wav
e
guide.
Horn
anten
na
offe
rs
sever
al
ben
e
fits
wh
e
n
em
plo
ye
d
i
n
that
besi
d
es
m
at
ching
th
e
im
ped
ance
of
the
gu
i
de
t
o
th
at
of
fr
ee
s
pace o
r
vi
ce
ver
sa
,
it
hel
ps
s
uppress
si
gnal
s
tra
velli
ng
v
ia
un
wan
te
d
m
od
es
in
the wa
veguide
f
ro
m
bein
g
rad
ia
te
d
an
d
it
pro
vid
es
si
gn
i
f
ic
ant
le
vel
of
di
recti
vity
and
ga
in
[
16
]
.
Wh
il
e
it
serv
es
a
s
ent
r
y
m
ediu
m
fo
r
sign
al
intercepti
on
f
or
processi
ng
in
the
case
of
re
c
ei
vin
g
syst
em
s
,
it
ser
ves
in
th
e
case
of
tra
nsm
issi
on
to
il
lu
m
inate
dish
a
nten
na
f
r
om
i
ts
fo
cal
ar
ea
est
i
m
at
ed
fr
om
the
f/d
para
m
et
ers
of
the
par
a
boli
c
dish
[17].
D
ual
m
od
e
fee
d
hor
ns
oft
en
prov
i
de
excell
ent
perform
ance
ov
er
wide
r
an
ge
of
m
ic
ro
wa
ve
bands
[18].
Th
e
two
ba
sic
ty
pes
of
hor
n
a
nten
na
a
re
t
he
pyram
id
an
d
c
on
ic
al
hor
n
ante
nn
a
.
O
ther
m
od
ific
at
ion
s
i
nclu
de
s
e
ct
or
ia
l
(E
or
H
plane
),
expo
nen
ti
al
,
c
orrugated
,
ri
dged
a
nd
sept
um
ho
rns
.
Fi
g
ures
1
of
py
ram
idal
and
con
ic
al
hor
ns
are
sh
ow
n belo
w [
16
]
.
Figure
1.
Pyra
m
idal and
C
on
ic
al
Im
ped
ance
m
at
ching
is
ver
y
desira
ble
with
rad
i
o
f
re
qu
e
nc
y
transm
issi
on
li
nes.
Sta
nd
i
ng
wa
ves
le
a
d
to inc
reased
lo
sses and
freq
ue
ntly
cause
t
he
t
ran
sm
it
te
d
to
m
al
fu
nctio
n [19].
5.
E
X
PERI
MEN
TAL SET
UP
The
set
up
f
or
the
E
valuati
on
of
water
co
nte
nt
in
m
il
k
us
in
g
m
ic
ro
wa
ve
tr
ansm
issi
on
ap
proac
h
wit
h
hor
n
ante
nn
ai
s
sh
ow
n
in
F
i
gure
2.
The
dott
ed
li
ne
bl
ocks
a
t
the
transm
it
t
i
ng
e
nd
are
s
ho
wing
the
set
up
f
or
a
c
onve
ntion
al
m
ic
ro
wa
ve ben
ch wh
il
e the
sol
id b
loc
k
i
nd
ic
at
ing
the
setu
p pr
ese
ntly
u
se
d i
n
la
bs
.
F
or
rec
epti
on
of
the
RF
sig
na
l
an
oth
e
r
a
nte
nn
a
is
place
d
on
a
r
otati
ng
ta
ble
befor
e
the
detect
or.
We
c
an
us
e
ei
ther
of
th
ree
diff
e
re
nt
powe
r
detect
ors.
We
use
Po
we
r m
eter
for
detect
io
n
of
po
wer.
In
f
act
r
otati
ng
ta
bl
e
is
no
t
re
quir
ed
f
or
this
ex
per
im
ent.
W
e
a
re
putt
ing
m
i
lk
in
bet
ween
tra
ns
m
itt
ing
&
receivin
g a
nten
na
wh
ic
h y
ou
ca
n
see
f
ro
m
the
photo o
f
e
xperi
m
ental
setup
a
s sho
wn in
F
ig
ur
e
3.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2252
-
8776
IJ
-
ICT
V
ol.
7
,
No.
2
,
Aug
us
t
20
1
8
:
77
–
81
80
Figure
2.
Ex
pe
rim
ental
Setup
Figure
3. P
ho
t
o of Ex
pe
rim
e
nt Setu
p
6.
E
X
PERI
MEN
TAL RES
UL
TS
The res
ult of th
e ex
per
im
ental
is sh
own
in Ta
ble 1.
Table
1.
E
xper
i
m
ental
Re
su
lt
s
No
Rep
eller
v
o
ltag
e
Bea
m
cu
rr
en
t
Vswr
Po
wer
in
db
1
281
1
9
.7
1
1
.9
2
281
1
9
.7
2
0
.5
3
281
1
9
.7
2
1
.2
4
281
1
9
.7
2
1
.5
5
281
1
9
.7
1
2
.1
6
281
1
9
.7
2
0
.06
7
281
1
9
.7
2
1
.4
8
281
1
9
.7
2
1
.5
9
282
1
9
.7
1
1
.8
10
282
1
9
.7
1
2
.1
11
282
1
9
.7
2
1
.2
12
282
1
9
.7
21
7.
E
X
PERI
MEN
TAL SET
UP
WITH H
ORN
ANTE
N
NA
The
hor
n
a
nten
na
set
up
is
set
in
the
m
ini
m
u
m
distance
by
the
f
or
m
ula.
U
s
our
ho
rn
ante
nn
a
is
hav
i
ng
the
wi
dth
of
0.1 M
,
It
is
kep
t
with
the
m
ini
m
um
distance.
We
us
e
t
he
50
m
l
beak
er
in
w
hich t
he m
il
k i
s
te
ste
d.
At
First
t
he
th
e
bea
ker
is
ket
em
pty
and
ra
diate
d
powe
r
thr
ough
the
gla
ss
is
cal
culat
ed
w
it
h
V
S
WR
m
et
er.
The
n
w
at
er
is
po
ur
e
d
in
the
bea
ker
and
the
n
t
he
r
adiat
ed
t
hroug
h
the
wate
r
a
nd
receive
d
power
is
cal
culat
ed.
T
he
n 10
m
l, 2
0
m
l,
30
ml
, 4
0
m
l, 5
0
m
l .
The
m
e
asur
e
d value
s
of
deci
bel is
plo
tt
ed
in
the
T
a
ble
1.
8.
DISCU
SSI
ONS
The
te
sti
ng
of
m
il
k
thickness
with
water
pro
ves
that
the
int
ensity
of
the
m
il
k
increas
es
th
e
power
of
t
he
receive
d
decib
el
is
le
ss.
By
usi
ng
t
he
m
ic
ro
w
ave
hor
n
a
nten
na
th
e
m
il
k
thickne
ss
ca
n
be
ca
lc
ulate
d.
So
we
trai
n
a nural
netw
ork
syst
em
f
or
a
na
ly
sing
the
thic
kn
e
ss
of
t
he
m
i
lk.
9
.
NEU
RA
L
N
E
TWOR
K BA
S
ED S
YS
TE
M
Neural
netw
ork base
d
syst
em
is sh
own
in
Figure
4, 5, a
nd
6.
Figur
4
.
N
ur
al
Netw
ork
Evaluation Warning : The document was created with Spire.PDF for Python.
IJ
-
ICT
IS
S
N:
22
52
-
8776
Evalu
atio
n of
Water C
on
te
nt
in Mil
k U
si
ng
Mi
crowave
Tr
an
s
mi
ssio
n Ap
pr
oac
h
wi
th …
(
Ashok.
Ja
)
81
Figure
5
.
O
utp
ut of
N
ur
al
N
e
twork
Figure
6.
O
utp
ut
o
f N
ur
al
Network
10.
RESU
LT
The
po
te
ntial
of
the
m
ic
ro
wa
ve
transm
issi
on
li
ne
te
chn
i
qu
e
for
e
valuati
on
of
a
dded
water
con
te
nt
in
m
il
k
has
prese
nted.
T
he
pe
rm
it
ti
vity
can
be
us
e
d
as
a
m
il
k
purity
in
dicat
or.
The
per
m
it
t
i
vity
of
t
he
t
hic
k
m
il
k
is
sli
gh
tl
y
s
m
aller
than
that
of
the
diluted
m
ilk
.
T
he
pe
rm
ittiv
it
y
decr
eas
es
with
incr
easi
ng
fr
e
qu
e
ncy.
F
rom
the
stud
y
of
the
m
easur
em
ent
uncertai
nty,
it
ha
s
been
s
how
n
that
high
accu
ra
cy
of
pe
rm
i
tt
iv
it
y
cal
culat
ion
can
be
achieve
d.
By
usi
ng
nural
net
w
ork
E
valuati
on
of
water
c
on
te
nt
in
m
i
lk
us
in
g
m
ic
ro
wav
e
tr
ansm
issi
on
appro
ac
h
with
hor
n
a
ntenn
a
is easy
for us t
o
ac
hiv
e.
REFERE
NCE
S
[1]
Shiinoki
,
Y.
,
Motouri
,
Y.,
I
t
o
,
K
.
,
“
On
-
li
ne
Monitor
ing
of
Moisture
and
Salt
Conte
n
ts
b
y
th
e
Microwa
ve
‘lka
nsm
ission
Method
in
a
Conti
nuous
Salt
ed
Bu
tt
er
-
m
aki
ng
Proc
ess”,
Journal
o
f
Food
Eqh
ee
n’ng
38
(1998)
153
-
1670
1998
Ek
ev
ie
r
Sc
ie
nc
e Li
m
i
te
d.
[2]
Agranovi
ch
,
D
.
,
Renh
art
,
I.
,
Ishai
,
P.B
.
,
Ka
tz
,
G.
,
Bez
m
an
,
D.
,
Feldman
,
Y
.
,
“
A
m
ic
rowav
e
sensor
for
th
e
cha
ra
cteri
z
at
ion
of
bovine m
il
k
”,
Food
Con
trol
0
956
-
7135/©
201
5
El
sev
ie
r
Lt
d
.
[3]
The
eff
e
ct of
m
i
cro
wave
paste
ur
i
za
t
ion
onth
e co
m
positi
on
of
m
ilk
[4]
Micha
e
l
O
.
Fr
y
e
r,
Andr
ea
J.
Hil
l
s,
L
.
Morrison,”
Method
and
app
ara
tus
for
m
ea
su
ring
bu
tt
erf
at
a
n
d
prot
ei
n
con
te
n
t
using micr
owav
e
absorpt
ion te
c
hnique
s
”
U.S. P
ate
nt
Nov.
14
,
20
00
,
US
6147502
[5]
Baz
ar
,
G.,
Rom
v_ar
i,
R
.
,
Szab
_o,
A.
,
Som
og
y
i,
T.,
_Eles,
V.,
Tsenkov
a,
R.
,
2016.
NIR
“
detec
t
ion
o
f
hon
ey
adult
erati
on
rev
eal
s di
ff
ere
n
ce
s
i
n
wate
r spe
ct
ral
patt
ern.
”
[6]
Gibson,
A.A.P.
,
Ng,
S.K.,
Noh,
B.
B.
M.
,
Chua
,
H.S.,
Haigh
,
A.
D.,
Parkinson,
G
.
,
Ainsw
orth,
P.
,
Plunket
t,
A.,
20
08.
“A
n
over
vie
w of
m
ic
rowave
tech
nique
s for the
eff
ic
i
ent
m
e
asure
m
ent
of
food
m
ate
ria
ls”
.
Food
Ma
nuf.
Ef
f
ic
.
2
,
1
e9
.
[7]
Krasze
ws
ki,
A.
W
.
,
N
el
son,
S.
O.,
1992.
Obs
er
vat
ions
on
reso
nant
c
avi
t
y
p
erturbat
ionb
y
di
ele
ct
ri
c
objects
.
I
E
EE
Tr
ans.
Mic
row. T
heory
Tech
.
40
,
151
e155. http://
dx.
doi.
org
/10.
11
09/22.
108334.
[8]
Karpi
sz,
T
.
,
Ko
p
y
t
,
P.,
Sa
lski,
B.
,
Krupka
,
J.,
2016.
Op
en
-
end
ed
wa
ve
gu
ide
m
easurement
of
li
q
uids
at
mi
ll
ime
t
er
wave
l
engt
hs
.
I
n:
2016
I
EEE
MT
T
-
S
In
t
ern
ational
Mi
c
rowave
S
y
m
posium
(IMS
).
IEE
E
,
pp.
1
e
4.
htt
p://dx.doi
.
org/
10.
1109/MW
SYM.2016
.
753995
4.
[9]
Guo,
W
.
,
Zhu
,
X.,
Li
u
,
Y.
,
Zhu
ang,
H.,
2010.
“
Sugar
and
wat
er
contents
of
hon
e
y
withd
ie
l
ectric
prope
rt
y
sensin
g
”
.
J.
Food Eng
.
97,
275e281. http://
dx.
doi.
org
/10.
10
16/j
.
j
foode
ng.
20
09.
10.
024
.
[10]
Puranik,
S.
,
Ku
m
bhar
khane
,
A
.
,
Mehr
otra,
S.
,
19
91.
“
Diel
e
ct
r
ic
p
rope
rties
of
hone
y
wat
ermixture
s be
twee
n
10
MH
z
to
10
GH
z
using
ti
m
e
dom
ai
n
te
chn
ique
”
.
J
.
Mi
crow.
Powe
r
Elec
tr
omagn.
Ene
rgy
26,
196e201
.
htt
p://dx.doi.org/10.1080/083278
23.
1991.
116881
57.
[11]
Li
ao
,
X.
,
R
agha
v
an,
V
.
G.,
Meda
,
V.,
Ya
y
l
a
y
a
n,
V.
A.,
2001
.
“
Diele
ct
ri
c
prop
ert
i
es
o
f
supersat
ur
at
ed
Alpha
-
D
-
gluc
os
e
aque
ous solut
ion
s
at
2450
MH
z
”
.
J. M
i
crow.
Pow
er
[12]
Venka
te
sh,
M.S
.
,
Ragh
ava
n,
G.
S.V.,
2
005.
“
An
over
v
ie
w
of
die
l
ec
tr
ic
prope
r
ti
esm
ea
suring
technique
s”
.
Ca
n
.
Bi
osyst.
Eng
.
/Le. Geni
e
Des.
Bi
os
yst.
au
Can
.
47,
1
5e30.
ht
tp:
/
/dx.
d
oi.
org/10
.
1109/
URS
IGA
SS
.
201
1.
6050287.
[13]
W
at
ts,
C
.
,
Hanh
am,
S.M.
,
Ahm
ad,
M.M.,
Adab
i,
M.,
Kle
in,
N.
,
2016
.
“
Coupled
di
el
e
ct
ricspl
itr
ing
microwav
e
resonator
for
li
q
uid
measur
eme
n
ts
in
microf
lui
di
c
chann
el
sat
na
noli
te
r
”
v
olume
s.
In:
46
th
Euro
pea
n
Mi
cro
wav
e
Confer
ence,
pp
.
257e260.
[14]
Jos
e,
K.A.
,
Var
ada
n,
V.K.
,
Var
ada
n,
V.V.
,
200
1.
“
W
ide
band
a
nd
noncontac
t
c
har
acte
ri
zation
of
the
compl
e
x
per
m
it
ti
vi
t
y
of
liquids
”
.
M
ic
row.
Opt.
Techno
l. Le
tt
.
30,
75
e79. ht
tp
:/
/dx.
do
i.
org
/10.1002/m
op
.
1225.
[15]
Kons
ta
nti
nos
B.
Baltzi
s
“
Pol
y
no
m
ia
l
–
Based
E
val
ua
ti
on
of
the
Im
pac
t
of
Aper
ture
Phas
e
Ta
pe
r
on
the
Gain
o
f
Rec
t
angul
ar
Hor
ns
”,
Journal
of
El
e
ct
rom
agnet
i
c
Analysis
&
Appl
ic
ati
ons
,
2010
,
Vol.
2
,
pp424
-
430
[16]
W
iki
pedi
a
,
ww
w.e
nwikipe
d
ia.o
rg/wiki
/Horn_a
n
te
nna
.
[17]
Dick
Com
l
y
,
“
Paraboli
c Ant
enn
as and
their F
eeds
”.
N3A
OG
.
ww
w.
Packr
a
tvhf.com
/Arti
cle
9/Di
sh_N
ot.
pdf
[18]
Paul
W
ade,
“
Opt
imum Dual
–
mo
de
F
ee
dhorn
”
.
W
1Ghz,
2006
[19]
Sam
uel
Y. Lia
o
,
“
Mic
rowave
De
v
ic
es
and
c
ircui
ts,
3rd E
di
ti
on
”
,
Pr
ent
i
ce
H
al
l
,
New
Jerse
y
,
199
0.
[20]
Ghane
m
,
T
.
H
“
die
l
ec
tr
ic
prop
erti
es
of
li
qu
id
food
s
aff
ecte
d
b
y
m
oi
sture
contents
an
d
te
m
per
a
ture
s”
,
Misr
J.
Ag.
Eng
.
,
27
(2):
688
–
69
8,
Proce
ss
Engi
n
ee
ring
.
Evaluation Warning : The document was created with Spire.PDF for Python.