T
E
L
K
O
M
N
I
K
A
T
elec
o
m
m
un
ica
t
io
n,
Co
m
pu
t
ing
,
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
,
p
p
.
1
~
8
I
SS
N:
1
6
9
3
-
6
9
3
0
,
ac
cr
ed
ited
First Gr
ad
e
b
y
Kem
en
r
is
tek
d
i
k
ti,
Dec
r
ee
No
: 2
1
/E/KPT
/2
0
1
8
DOI
: 1
0
.
1
2
9
2
8
/TE
L
KOM
NI
K
A.
v
1
9
i1
.
1
5
9
1
2
1
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Predic
ting th
e not
ch band
freque
nc
y
of an
ultra
-
wide
ba
nd
a
ntenna usi
ng
art
ificial neura
l
net
wo
rks
L
a
hcen
Ag
un
i
1
,
Sa
m
ira
Cha
ba
a
2
,
Sa
ida
I
bn
y
a
ich
3
,
Abdel
o
uh
a
b Z
er
o
ua
l
4
1
,
3,
4
De
p
a
rtme
n
t
o
f
p
h
y
sic
s,
F
a
c
u
lt
y
o
f
S
c
ien
c
e
s,
Ca
d
i
A
y
y
a
d
U
n
iv
e
rsity
,
M
o
ro
c
c
o
2
In
d
u
strial
En
g
i
n
e
e
rin
g
De
p
a
rtme
n
t,
Na
ti
o
n
a
l
S
c
h
o
o
l
o
f
Ap
p
li
e
d
S
c
ien
c
e
s,
Ib
n
Zo
h
r
Un
i
v
e
rsity
,
M
o
r
o
c
c
o
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Feb
2
3
,
2
0
2
0
R
ev
is
ed
J
u
n
1
6
,
2
0
2
0
Acc
ep
ted
Au
g
2
9
,
2
0
2
0
In
t
h
is
p
a
p
e
r
we
p
ro
p
o
se
to
p
re
d
i
c
t
th
e
n
o
tc
h
fre
q
u
e
n
c
y
o
f
a
n
u
l
tra
-
wid
e
b
a
n
d
(UWB)
a
n
ten
n
a
wh
ich
o
p
e
ra
tes
in
th
e
fre
q
u
e
n
c
y
b
a
n
d
fr
o
m
3
.
8
5
G
Hz
to
1
2
.
3
8
G
H
z
.
Th
e
p
re
d
icti
o
n
o
f
t
h
e
n
o
tc
h
fre
q
u
e
n
c
y
i
n
o
rd
e
r
to
a
v
o
id
in
t
e
rfe
re
n
c
e
s
b
e
twe
e
n
(W
LAN)
IEE
E8
0
2
.
1
1
a
a
n
d
HIPE
R
LAN/2
WL
AN
a
p
p
li
c
a
ti
o
n
s
a
n
d
UWB
tec
h
n
o
l
o
g
y
is
a
c
h
iev
e
d
u
sin
g
t
h
e
a
rti
ficia
l
n
e
u
ra
l
n
e
tw
o
r
k
s
(AN
N)
tec
h
n
iq
u
e
.
T
h
e
d
e
v
e
lo
p
e
d
AN
N
is
o
p
ti
m
ize
d
wit
h
t
h
e
h
e
lp
o
f
K
-
fo
ld
c
ro
ss
v
a
li
d
a
ti
o
n
m
e
th
o
d
wh
ich
a
ll
o
ws
u
s
to
d
i
v
id
e
th
e
d
a
tas
e
ts
in
t
o
1
0
su
b
se
ts
in
t
h
e
train
in
g
p
h
a
se
.
Th
e
sim
u
late
d
d
a
tas
e
ts
a
r
e
g
e
n
e
ra
ted
b
y
c
o
n
tro
l
li
n
g
h
ig
h
fre
q
u
e
n
c
y
stru
c
t
u
ra
l
sim
u
lato
r
(
H
F
S
S
)
fro
m
M
ATLAB
u
si
n
g
a
VB
sc
rip
t.
Th
e
p
e
rfo
rm
a
n
c
e
o
f
th
e
AN
N
tec
h
n
i
q
u
e
is
a
ss
e
ss
e
d
u
sin
g
s
o
m
e
sta
ti
sti
c
a
l
c
rit
e
ria.
Du
rin
g
t
h
e
trai
n
in
g
p
r
o
c
e
ss
,
th
e
m
e
a
n
a
b
so
lu
te
p
e
rc
e
n
tag
e
e
rro
r
(M
APE
)
b
e
twe
e
n
th
e
sim
u
late
d
a
n
d
th
e
p
r
e
d
icte
d
AN
N
n
o
tch
fre
q
u
e
n
c
ies
i
s
0
,
1
2
5
.
A
c
o
m
p
a
riso
n
b
e
twe
e
n
sim
u
late
d
,
th
e
o
re
ti
c
a
l,
a
n
d
AN
N
re
su
lt
s
h
a
s
b
e
e
n
a
c
h
iev
e
d
d
u
rin
g
t
h
e
tes
t
a
n
d
v
a
li
d
a
ti
o
n
p
ro
c
e
ss
,
g
o
o
d
a
c
c
u
ra
c
y
i
s
o
b
tai
n
e
d
b
e
twe
e
n
th
e
sim
u
late
d
a
n
d
th
e
A
NN
p
re
d
icti
o
n
s.
Th
e
p
r
o
p
o
se
d
U
WB
a
n
ten
n
a
e
x
h
ib
it
s
a
n
o
tc
h
b
a
n
d
fro
m
5
.
1
G
H
z
to
6
.
0
G
Hz
with
a
n
o
tch
fr
e
q
u
e
n
c
y
o
f
a
p
p
ro
x
ima
tely
5
.
5
1
G
Hz
.
K
ey
w
o
r
d
s
:
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
K
-
f
o
ld
cr
o
s
s
v
alid
atio
n
Ultr
a
-
wid
eb
an
d
a
n
ten
n
a
T
h
is i
s
a
n
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC B
Y
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
L
ah
ce
n
Ag
u
n
i
Dep
ar
tm
en
t o
f
p
h
y
s
ics
,
Facu
lty
o
f
Scien
ce
s
C
ad
i A
y
y
ad
Un
iv
er
s
ity
Ma
r
r
ak
esh
,
Mo
r
o
cc
o
E
m
ail:
ag
u
n
ilah
ce
n
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
I
n
wir
eless
co
m
m
u
n
icatio
n
s
s
y
s
tem
s
,
th
e
an
ten
n
a
is
o
n
e
o
f
th
e
m
o
s
t
im
p
o
r
tan
t
elem
en
ts
.
T
ec
h
n
o
lo
g
ies,
s
u
ch
as:
g
lo
b
al
p
o
s
itio
n
in
g
s
y
s
tem
(
GPS
)
,
wir
eless
lo
ca
l
-
ar
ea
n
etwo
r
k
(
W
L
AN)
/
wo
r
ld
wid
e
in
ter
o
p
er
a
b
ilit
y
f
o
r
m
icr
o
wav
e
ac
ce
s
s
(
W
I
MA
X
)
,
u
ltra
-
wi
d
eb
an
d
(
UW
B
)
,
n
ee
d
an
in
d
is
p
en
s
ab
le
p
ar
t
to
tr
an
s
m
it
an
d
r
ec
ei
v
e
th
e
i
n
f
o
r
m
atio
n
,
wh
ich
is
th
e
m
icr
o
-
s
tr
ip
p
atch
a
n
ten
n
a.
T
h
e
p
r
in
cip
al
co
m
p
o
n
en
ts
o
f
a
s
im
p
lest
p
atch
an
ten
n
a
ar
e
a
co
n
d
u
ctiv
e
lay
e
r
,
a
s
u
b
s
tr
ate
lay
er
,
an
d
a
g
r
o
u
n
d
p
lan
e
[
1
,
2
]
.
C
o
n
s
eq
u
en
tly
,
m
u
ltip
le
p
atch
an
ten
n
as
h
av
e
b
ee
n
f
ab
r
icate
d
with
d
if
f
er
e
n
t
s
h
ap
es,
d
if
f
er
e
n
t
d
im
en
s
io
n
s
,
a
n
d
v
ar
io
u
s
d
ielec
tr
ic
s
u
b
s
tr
ates
in
o
r
d
er
t
o
s
atis
f
y
th
e
d
esire
d
s
p
ec
if
icatio
n
s
[
3
,
4
]
.
UW
B
is
ch
ar
ac
ter
ized
b
y
m
an
y
ad
v
an
tag
es
lik
e:
h
ig
h
b
an
d
wid
th
,
h
u
g
e
d
ata
r
at
e,
v
er
y
lo
w
p
o
wer
co
n
s
u
m
p
tio
n
,
an
d
l
o
w
co
s
t
[
5
,
6
]
.
T
h
e
s
m
all
s
ize,
th
e
lig
h
t
weig
h
t,
th
e
ea
s
y
f
a
b
r
icatio
n
,
an
d
th
e
th
i
n
p
r
o
f
ile
co
n
f
ig
u
r
a
tio
n
s
[
7
]
m
a
k
e
th
e
m
in
iat
u
r
iz
ed
m
icr
o
s
tr
ip
p
atc
h
an
ten
n
a
a
b
etter
ca
n
d
id
ate
to
b
e
u
s
ed
in
UW
B
ap
p
licatio
n
s
.
T
h
ese
an
ten
n
as
h
av
e
t
o
s
atis
f
y
p
er
f
o
r
m
an
ce
s
s
u
ch
as;
g
o
o
d
im
p
ed
an
ce
b
an
d
wid
th
,
co
n
s
tan
t
g
ain
,
g
o
o
d
r
ad
iatio
n
p
atter
n
,
an
d
s
m
all
d
esig
n
.
T
h
e
Fed
er
al
C
o
m
m
u
n
icatio
n
s
C
o
m
m
is
s
io
n
(
FC
C
)
ap
p
r
o
v
ed
th
e
u
s
e
o
f
th
e
s
p
ec
tr
u
m
3
.
1
-
1
0
.
6
GHz
f
o
r
UW
B
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
-
8
2
ap
p
licatio
n
s
[
8
]
.
B
u
t
th
er
e
ar
e
alr
ea
d
y
l
icen
s
ed
ap
p
licatio
n
s
u
s
in
g
a
p
ar
t
o
f
th
is
f
r
eq
u
en
cy
b
a
n
d
s
u
ch
as
W
L
AN
I
E
E
E
8
0
2
.
1
1
a
an
d
HI
PERLAN/2
W
L
AN
o
p
er
atin
g
i
n
th
e
5
.
1
5
–
5
.
8
2
5
GHz
b
an
d
.
Fo
r
t
h
is
r
ea
s
o
n
,
d
esig
n
er
s
a
n
d
r
esear
ch
er
s
h
av
e
p
r
o
p
o
s
ed
m
a
n
y
s
h
ap
es a
n
d
m
an
y
tech
n
iq
u
e
s
to
n
o
tch
t
h
e
f
r
e
q
u
en
c
y
b
a
n
d
5
.
1
5
–
5
.
8
2
5
GHz
i
n
o
r
d
er
t
o
o
v
e
r
co
m
e
in
te
r
f
er
en
c
es with
UW
B
.
I
n
th
e
liter
atu
r
e,
th
e
m
o
s
t
u
s
ed
tech
n
iq
u
e
to
n
o
tch
a
b
an
d
is
cu
ttin
g
a
s
lo
t,
eith
er
o
n
th
e
r
ad
i
atin
g
p
atch
o
r
o
n
th
e
g
r
o
u
n
d
p
lan
e
[
9
-
1
1
]
.
I
n
[
9
]
,
th
e
n
o
tch
b
an
d
is
r
ea
lized
b
y
in
s
er
tin
g
a
U
-
s
h
ap
ed
s
lo
t
o
n
th
e
r
a
d
iatin
g
p
atch
to
d
esig
n
a
p
atch
an
ten
n
a
with
th
e
r
ejec
tio
n
o
f
W
L
AN
b
an
d
.
A
co
p
lan
a
r
wav
eg
u
id
e
f
ed
co
m
p
ac
t
UW
B
an
ten
n
a
is
p
r
o
p
o
s
ed
in
[
1
1
]
.
T
h
e
n
o
tch
o
f
W
L
AN
b
an
d
is
ac
h
iev
ed
b
y
etc
h
in
g
h
alf
wav
elen
g
th
C
-
s
h
ap
e
d
an
n
u
lar
r
i
n
g
s
lo
t
in
th
e
r
ad
i
at
in
g
p
atch
.
I
n
an
o
th
er
r
esear
ch
,
m
u
ltip
le
b
a
n
d
s
ar
e
ac
h
ie
v
e
d
.
I
n
[
1
2
]
,
au
th
o
r
s
p
r
o
p
o
s
e
an
UW
B
an
ten
n
a
with
two
s
lits
to
av
o
id
th
e
in
t
er
f
er
en
ce
with
W
L
AN
an
d
5
G
b
an
d
s
.
I
n
[
1
3
]
,
an
in
s
et
-
f
ed
UW
B
an
ten
n
a
with
th
r
ee
n
o
tch
es
is
g
en
e
r
ated
b
y
in
tr
o
d
u
cin
g
U
-
s
h
a
p
e
d
,
cr
escen
t
-
s
h
ap
ed
,
a
n
d
L
-
s
h
ap
ed
s
lo
ts
.
A
wid
e
b
a
n
d
wid
th
f
o
r
UW
B
ap
p
licatio
n
s
is
ac
h
iev
ed
u
s
in
g
a
m
o
n
o
p
o
le
an
ten
n
a
with
s
in
g
le
an
d
d
u
al
b
a
n
d
-
n
o
tch
ed
ch
ar
ac
t
er
is
tics
[
1
4
]
.
A
C
-
s
h
ap
ed
s
lo
t
i
n
th
e
p
atch
to
im
p
lem
en
t
a
p
la
n
ar
UW
B
an
ten
n
a
with
3
.
4
/5
.
5
GHz
d
u
al
b
a
n
d
-
n
o
tch
ed
is
p
r
esen
ted
i
n
[
1
0
]
.
Ov
er
th
ese
d
esig
n
s
,
th
eo
r
etica
l
m
eth
o
d
is
u
s
ed
to
d
eter
m
in
e
th
e
n
o
tch
b
a
n
d
f
r
eq
u
e
n
cy
.
Oth
e
r
tech
n
iq
u
es
ar
e
u
s
ed
to
n
o
tch
t
h
e
UW
B
b
an
d
.
T
h
e
n
o
tch
f
r
e
q
u
en
cy
o
f
s
lo
t
-
lo
a
d
ed
p
r
in
ted
UW
B
an
ten
n
as
ca
n
b
e
p
r
e
d
icted
b
y
a
p
p
ly
in
g
t
h
e
s
lo
t
-
lin
e
th
eo
r
y
[
1
5
]
.
Use
o
f
g
e
n
etic
alg
o
r
ith
m
s
to
p
lace
f
r
eq
u
e
n
cy
n
o
tch
es
with
in
UW
B
b
an
d
is
p
r
esen
te
d
[
1
6
]
.
Sin
ce
,
a
r
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
ANN
)
ar
e
wid
el
y
u
s
ed
in
th
e
d
esig
n
o
f
m
icr
o
s
tr
ip
p
atch
an
ten
n
a
[
1
7
-
2
3
]
.
T
h
e
o
r
ig
in
ality
o
f
th
is
w
o
r
k
is
th
e
im
p
lem
en
tatio
n
o
f
t
h
e
ANN
tech
n
iq
u
e
u
s
in
g
k
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
m
eth
o
d
in
o
r
d
er
to
en
h
a
n
ce
th
e
p
er
f
o
r
m
an
c
e
o
f
th
e
p
r
o
p
o
s
ed
ANN
m
o
d
el
an
d
t
o
ac
cu
r
ately
p
r
ed
ict
t
h
e
n
o
tch
f
r
eq
u
en
cy
o
f
th
e
p
r
o
p
o
s
ed
UW
B
an
ten
n
a.
R
elate
d
to
th
e
r
ese
ar
ch
es
cited
a
b
o
v
e,
th
e
an
ten
n
a
p
r
esen
ted
in
th
is
p
ap
er
,
is
p
h
y
s
ically
s
m
all
wit
h
a
p
ar
tial
g
r
o
u
n
d
p
la
n
e,
a
n
d
h
as
th
e
n
o
tch
b
an
d
ch
ar
ac
ter
is
tic
b
esid
es g
o
o
d
im
p
ed
an
ce
b
an
d
wid
th
.
2.
ANN
M
O
D
E
L
2
.
1
.
P
r
o
po
s
ed
a
nte
nn
a
I
n
th
is
wo
r
k
,
we
ar
e
b
ased
o
n
t
h
e
ANN
tech
n
iq
u
e
to
p
r
ed
ict
t
h
e
n
o
tch
f
r
e
q
u
en
c
y
o
f
an
UW
B
an
ten
n
a.
T
h
e
p
r
o
p
o
s
ed
UW
B
an
ten
n
a
co
n
s
is
ts
o
f
a
d
ielec
tr
ic
s
u
b
s
tr
ate
FR
4
-
ep
o
x
y
with
d
i
elec
tr
ic
p
er
m
itti
v
ity
(
_
=
4
.
3
)
,
lo
s
s
tan
g
en
t
(
=
0
.
025
)
,
len
g
th
(
=
2
4
)
,
wid
th
(
=
14
)
,
an
d
th
ick
n
ess
(
ℎ
=
0
.
8
)
.
I
n
th
e
b
o
tto
m
s
id
e
o
f
th
e
s
u
b
s
tr
ate
we
h
av
e
a
p
a
r
tial
g
r
o
u
n
d
p
lan
e
(
l
g
=
9
)
to
e
n
s
u
r
e
a
g
o
o
d
im
p
ed
an
ce
b
an
d
wid
th
.
I
n
th
e
o
th
er
s
id
e
o
f
th
e
s
u
b
s
tr
ate,
w
e
h
av
e
a
cir
cu
lar
p
atch
o
f
c
o
p
p
er
(
=
7
)
i
n
wh
ich
we
in
s
er
ted
a
s
p
lit
r
in
g
to
ac
h
iev
e
t
h
e
n
o
tc
h
b
an
d
ch
ar
ac
ter
is
tic.
T
h
e
co
n
f
ig
u
r
ati
o
n
o
f
t
h
e
p
r
o
p
o
s
ed
an
ten
n
a
is
illu
s
tr
ated
in
Fig
u
r
e
1
.
T
h
e
s
p
lit
r
in
g
is
ch
ar
ac
ter
ize
d
b
y
th
e
d
im
e
n
s
io
n
s
a,
b
,
an
d
g
,
th
ese
d
im
en
s
io
n
s
ca
n
b
e
u
s
ed
t
o
ca
lcu
late
th
e
n
o
tch
b
an
d
f
r
eq
u
en
cy
b
y
a
p
p
ly
in
g
th
e
g
iv
e
n
f
o
r
m
u
la
[
15
]
as sh
o
wn
in
(
1
)
:
f
c
=
c
2
L
NB
√
ε
ef
f
(
1
)
wh
er
e
c
is
th
e
s
p
ee
d
o
f
lig
h
t,
L
NB
is
th
e
len
g
th
o
f
th
e
etc
h
ed
s
p
lit
r
in
g
i
n
th
e
cir
cu
la
r
p
atch
wh
ich
is
(
2
+
2
(
−
)
−
)
,
an
d
ε
ef
f
is
th
e
ef
f
ec
tiv
e
d
ielec
tr
i
c
co
n
s
tan
t (
2
)
.
ε
ef
f
=
ε
r
+
1
2
(
2
)
(
a)
(
b
)
Fig
u
r
e
1
.
C
o
n
f
ig
u
r
atio
n
o
f
th
e
p
r
o
p
o
s
ed
an
ten
n
a
; (
a)
f
r
o
n
t
v
i
ew
an
d
(
b
)
b
ac
k
v
iew
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
P
r
ed
ictin
g
th
e
n
o
tc
h
b
a
n
d
fr
eq
u
en
cy
o
f
a
n
u
ltr
a
-
w
id
eb
a
n
d
a
n
ten
n
a
u
s
in
g
…
(
La
h
ce
n
A
g
u
n
i
)
3
2.
2
.
P
a
ra
m
et
ric
a
na
ly
s
is
As
d
em
o
n
s
tr
ated
in
(
1
)
,
th
e
p
h
y
s
ical
d
im
en
s
io
n
s
o
f
th
e
s
p
li
t
r
in
g
a,
b
,
g
an
d
th
e
elec
tr
ical
p
ar
am
eter
ε
r
af
f
ec
t
d
ir
ec
tly
t
h
e
n
o
tc
h
b
a
n
d
f
r
eq
u
e
n
cy
.
Fo
r
th
is
r
ea
s
o
n
,
w
e
h
av
e
d
o
n
e
a
p
ar
am
etr
ic
a
n
al
y
s
is
in
wh
ich
we
s
tu
d
y
th
e
ef
f
ec
t
o
f
ch
an
g
in
g
th
e
p
ar
a
m
eter
s
a,
b
,
g
,
a
n
d
ε
r
o
n
th
e
VSW
R
an
d
co
n
s
eq
u
e
n
tly
th
e
n
o
tch
b
a
n
d
f
r
eq
u
e
n
cy
as
s
h
o
wn
in
Fig
u
r
e
2
.
W
e
ca
n
n
o
tice
f
r
o
m
Fig
u
r
e
2
(
a
)
,
th
e
in
cr
ea
s
e
o
f
th
e
p
ar
a
m
eter
a
(
in
n
er
r
a
d
iu
s
o
f
th
e
s
p
lit
r
in
g
)
f
r
o
m
1
.
4
1
to
3
.
8
1
m
m
,
th
e
ce
n
tr
al
f
r
eq
u
en
cy
s
h
if
ted
t
o
th
e
lo
wer
f
r
eq
u
en
cies
u
n
til
its
d
is
ap
p
ea
r
an
ce
.
W
h
ile,
b
y
i
n
cr
ea
s
in
g
th
e
o
u
ter
r
a
d
iu
s
o
f
th
e
s
p
lit
r
in
g
(
p
ar
a
m
eter
b
)
,
th
e
V
SW
R
in
cr
ea
s
es
an
d
th
e
n
o
tch
f
r
eq
u
en
cy
s
h
if
te
d
to
lo
wer
f
r
eq
u
en
cies a
s
s
h
o
wn
in
Fig
u
r
e
2
(
b
)
.
T
h
e
g
ap
o
f
th
e
s
p
lit
r
in
g
(
p
ar
am
eter
g
)
af
f
ec
ts
also
th
e
VSW
R
an
d
th
e
n
o
tch
b
an
d
f
r
eq
u
e
n
cy
.
B
y
v
ar
y
in
g
th
e
g
ap
g
f
r
o
m
1
.
4
2
to
3
.
8
2
m
m
,
th
e
ce
n
tr
al
f
r
eq
u
en
c
y
m
o
v
es
s
lig
h
tly
to
h
ig
h
f
r
eq
u
e
n
cies
as
illu
s
tr
ated
in
Fig
u
r
e
2
(
c
)
.
T
h
e
last
p
ar
am
eter
wh
ich
af
f
ec
ts
t
h
e
VSW
R
is
th
e
elec
tr
ical
p
ar
am
eter
ε
r
.
W
e
h
av
e
tak
en
th
r
ee
d
if
f
e
r
en
t
d
ielec
tr
ic
s
u
b
s
tr
ates
(
2
.
2
,
4
.
3
,
an
d
6
.
1
5
)
,
as
g
iv
en
in
Fig
u
r
e
2
(
d
)
,
th
e
n
o
tch
b
an
d
f
r
e
q
u
en
c
y
to
o
k
th
r
ee
d
if
f
er
en
t
v
alu
es
4
.
8
GHz
,
5
.
5
GHz
,
an
d
6
.
5
GHz
.
I
t is cle
ar
f
r
o
m
th
ese
r
esu
lts
th
a
t th
e
p
a
r
am
eter
s
a,
b
,
g
,
a
n
d
ε
r
af
f
ec
t d
ir
ec
tly
t
h
e
V
SW
R
an
d
th
e
n
o
tc
h
b
an
d
f
r
e
q
u
en
cy
.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
2
.
Par
am
etr
ic
a
n
aly
s
is
;
(
a)
p
ar
am
ete
r
a,
(
b
)
p
a
r
am
eter
b
,
(
c)
p
ar
am
eter
g
,
(
d
)
p
a
r
am
et
er
ε
r
2.
3
.
Dev
el
o
ped AN
N
m
o
del
T
o
estab
lis
h
th
e
ANN
m
o
d
el,
we
f
o
llo
we
d
th
e
s
tep
s
d
escr
i
b
ed
b
elo
w
.
As
a
f
ir
s
t
s
tep
in
cr
ea
tin
g
an
ANN
n
etwo
r
k
,
we
g
en
e
r
ated
t
h
e
d
atab
ase
b
y
v
ar
y
i
n
g
th
e
p
a
r
am
eter
s
a,
b
,
g
,
an
d
ε
r
with
in
a
s
p
ec
if
ied
r
an
g
e
an
d
r
ec
o
r
d
in
g
th
e
c
o
r
r
esp
o
n
d
i
n
g
f
r
e
q
u
en
c
y
f
o
r
VSW
R
>
2
wh
ich
d
ef
in
es
th
e
n
o
tc
h
f
r
e
q
u
en
cy
.
T
h
e
co
llectin
g
o
f
d
atasets
wa
s
r
ea
lized
u
s
in
g
a
v
is
u
al
b
asic
(
VB
)
s
cr
ip
t
wh
i
ch
co
n
tr
o
ls
th
e
HFSS
s
im
u
lat
o
r
f
r
o
m
MA
T
L
AB
.
Hen
ce
,
3
6
0
s
im
u
latio
n
s
h
av
e
b
ee
n
ca
r
r
ie
d
o
u
t
with
v
ar
ian
t
d
i
m
en
s
io
n
s
a,
b
,
g
,
an
d
d
if
f
e
r
en
t
d
ielec
tr
ic
s
u
b
s
tr
ate
ch
ar
ac
ter
ized
b
y
ε
r
as
s
h
o
wn
in
T
ab
le
1
an
d
Fig
u
r
e
3
.
T
h
e
n
o
tch
f
r
e
q
u
en
c
y
f
c
was
u
s
ed
as
th
e
o
u
tp
u
t
o
f
th
e
m
o
d
el,
w
h
ile
th
e
p
ar
a
m
eter
s
a,
b
,
g
,
an
d
ε
r
ar
e
u
s
ed
as th
e
in
p
u
t o
f
th
e
m
o
d
el.
An
o
th
er
s
tep
to
cr
ea
te
t
h
e
AN
N
m
o
d
el
is
to
d
eter
m
in
e
th
e
n
u
m
b
er
o
f
h
id
d
en
lay
er
s
a
n
d
th
e
n
u
m
b
er
o
f
n
eu
r
o
n
s
in
ea
c
h
h
id
d
e
n
lay
e
r
,
t
h
ese
p
ar
am
eter
s
ar
e
d
eter
m
in
e
d
ac
cu
r
ately
b
y
o
b
s
er
v
in
g
s
o
m
e
s
tatis
tical
cr
iter
ia
lik
e
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
MSE
)
an
d
m
ea
n
ab
s
o
lu
te
p
e
r
ce
n
t
ag
e
er
r
o
r
(
MA
PE)
.
I
n
th
is
wo
r
k
,
we
ad
o
p
ted
two
h
id
d
en
lay
er
s
with
2
5
n
eu
r
o
n
s
ea
ch
.
Af
ter
war
d
s
,
we
ch
o
o
s
e
th
e
ap
p
r
o
p
r
iate
tr
ain
in
g
alg
o
r
ith
m
am
o
n
g
r
esil
ien
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
-
8
4
b
ac
k
p
r
o
p
ag
atio
n
(
R
P),
Po
lak
-
R
ib
ier
e
co
n
ju
g
ate
g
r
ad
ien
t
(
C
GP)
,
o
n
e
s
tep
s
ec
an
t
(
OSS
)
,
s
ca
led
co
n
ju
g
ate
g
r
ad
ien
t (
SC
G)
,
an
d
L
ev
en
b
er
g
-
Ma
r
q
u
a
r
d
t (
L
M)
.
I
t seem
s
th
at
th
e
L
M
alg
o
r
ith
m
g
iv
es
a
b
etter
ac
cu
r
ac
y
.
T
h
e
p
ar
am
eter
s
o
f
th
e
ANN
m
o
d
e
l
u
s
ed
in
th
is
s
tu
d
y
ar
e
s
u
m
m
ar
ized
in
T
ab
le
2
.
T
h
e
f
lo
wch
ar
t
m
eth
o
d
o
lo
g
y
o
f
th
is
r
esear
ch
is
s
h
o
wn
in
Fig
u
r
e
4
.
T
ab
le
1
.
3
6
0
g
en
er
ated
d
atasets
N
u
mb
e
r
o
f
sam
p
l
e
s
a
b
g
ε
r
1
2
0
2
,
3
,
4
,
5
2
.
5
,
3
.
5
,
4
.
5
,
5
.
5
1
,
2
,
3
,
4
2
.
2
,
4
.
3
,
6
.
1
5
1
2
0
2
.
5
,
3
.
5
,
4
.
5
,
5
.
5
3
,
4
,
5
,
6
1
,
2
,
3
,
4
2
.
2
,
4
.
3
,
6
.
1
5
1
2
0
2
,
3
,
4
,
5
2
.
3
,
3
.
3
,
4
.
3
,
5
.
3
1
.
5
,
2
.
5
,
3
.
5
,
4
.
5
2
.
2
,
4
.
3
,
6
.
1
5
Fig
u
r
e
3
.
Simu
lated
n
o
tch
b
an
d
f
r
eq
u
en
cy
T
ab
le
2
.
ANN
m
o
d
el
p
ar
am
et
er
s
A
N
N
p
a
r
a
m
e
t
e
r
s
A
t
t
r
i
b
u
t
e
s
D
a
t
a
b
a
se
3
6
0
I
n
p
u
t
a
,
b
,
g
,
ε
r
O
u
t
p
u
t
f
c
Tr
a
i
n
i
n
g
a
l
g
o
r
i
t
h
m
LM
A
c
t
i
v
a
t
i
o
n
f
u
n
c
t
i
o
n
‘
Ta
n
s
i
g
’
(
h
i
d
d
e
n
l
a
y
e
r
)
,
‘
p
u
r
e
l
i
n
’
(
o
u
t
p
u
t
l
a
y
e
r
)
S
t
r
u
c
t
u
r
e
4
-
25
-
25
-
1
Fig
u
r
e
4
.
Flo
wch
ar
t
o
f
th
e
s
tu
d
y
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
P
r
ed
ictin
g
th
e
n
o
tc
h
b
a
n
d
fr
eq
u
en
cy
o
f
a
n
u
ltr
a
-
w
id
eb
a
n
d
a
n
ten
n
a
u
s
in
g
…
(
La
h
ce
n
A
g
u
n
i
)
5
3.
K
-
F
O
L
D
CO
RSS VA
L
I
DA
T
I
O
N
M
E
T
H
O
D
I
n
th
is
s
tu
d
y
,
we
ap
p
lied
th
e
k
-
f
o
ld
c
r
o
s
s
v
alid
atio
n
m
eth
o
d
to
d
iv
id
e
th
e
d
atasets
.
I
n
th
e
tr
ain
in
g
p
h
ase,
we
u
s
ed
1
0
-
f
o
ld
cr
o
s
s
v
alid
atio
n
m
eth
o
d
.
T
h
e
d
atasets
ar
e
d
iv
id
ed
in
1
0
s
u
b
s
ets;
w
h
er
e
ea
ch
tim
e,
o
n
e
o
f
th
e
1
0
s
u
b
s
ets is
u
s
ed
as th
e
test
s
et
an
d
th
e
r
em
a
in
i
n
g
9
s
u
b
s
ets ar
e
jo
in
e
d
to
g
eth
er
t
o
f
o
r
m
th
e
tr
ain
in
g
s
et.
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
is
th
en
r
e
p
ea
ted
1
0
tim
es.
T
h
e
a
d
v
an
tag
e
o
f
th
is
m
eth
o
d
is
th
at
ev
er
y
s
u
b
s
et
h
as
o
n
e
c
h
an
ce
to
b
e
in
a
test
s
et
ex
ac
tly
o
n
ce
,
an
d
h
as
th
e
ch
a
n
ce
to
b
e
in
a
tr
ain
in
g
s
et
9
ti
m
es.
T
h
e
1
0
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
m
eth
o
d
is
illu
s
tr
ated
in
Fig
u
r
e
5
.
Fig
u
r
e
5
.
10
-
f
o
ld
cr
o
s
s
-
v
alid
atio
n
m
eth
o
d
Fo
r
ea
ch
tim
e,
we
ca
lc
u
late
t
h
e
m
ea
n
s
q
u
ar
e
d
er
r
o
r
(
MSE
)
an
d
th
e
m
ea
n
a
b
s
o
lu
te
p
e
r
ce
n
tag
e
er
r
o
r
(
MA
PE)
af
ter
ea
ch
iter
atio
n
d
u
r
in
g
t
h
e
test
p
h
ase.
T
h
e
MSE
an
d
th
e
MA
PE
ar
e
g
iv
en
b
y
th
e
f
o
llo
win
g
eq
u
atio
n
s
[
2
4
,
2
5
]
:
Me
an
s
q
u
ar
ed
e
r
r
o
r
:
M
SE
=
(
1
n
)
∑
(
t
i
−
α
i
)
2
n
i
=
1
(
3
)
Me
an
ab
s
o
lu
te
p
er
ce
n
tag
e
er
r
o
r
:
M
A
PE
=
(
100
n
∑
|
t
i
−
α
i
t
i
|
n
i
=
1
)
(
4
)
wh
er
e
n
is
th
e
to
tal
n
u
m
b
er
o
f
s
am
p
les,
an
d
t
i
an
d
α
i
r
ep
r
esen
t,
r
esp
ec
tiv
ely
,
th
e
tar
g
et
an
d
o
u
tp
u
t
d
ata.
T
h
e
MSE
an
d
MA
PE
af
ter
1
0
tu
r
n
s
ar
e
g
iv
en
in
T
ab
le
3
.
T
h
e
MSE
(
T
o
tal)
an
d
MA
PE
(
T
o
tal)
ar
e
r
esp
ec
tiv
ely
0
.
0
3
9
an
d
0
.
1
2
5
.
M
SE
(
To
ta
l
)
=
1
10
∑
M
SE
(
5
)
M
A
PE
(
To
ta
l
)
=
1
10
∑
M
A
PE
(
6
)
T
ab
le
3
.
MSE
an
d
MA
PE
f
o
r
1
0
iter
atio
n
s
I
t
e
r
a
t
i
o
n
M
S
E
M
A
P
E
1
f
o
l
d
0
.
3
1
4
0
.
4
9
9
2
f
o
l
d
0
.
0
4
2
0
.
2
5
6
3
f
o
l
d
0
.
0
2
0
0
.
1
7
0
4
f
o
l
d
0
.
0
1
5
0
.
1
3
3
5
f
o
l
d
0
.
0
0
3
0
.
0
7
8
6
f
o
l
d
0
.
0
0
1
0
.
0
5
6
7
f
o
l
d
7
.
4
7
E
-
04
0
.
0
3
0
8
f
o
l
d
6
.
6
4
E
-
05
0
.
0
1
2
9
f
o
l
d
3
.
7
0
E
-
05
0
.
0
1
0
1
0
f
o
l
d
1
.
4
5
E
-
05
0
.
0
0
5
360
D
atas
et
s
T
e
s
t d
a
t
a
3
6
T
r
a
in
i
n
g
d
a
t
a
3
2
4
I
te
ra
tio
n
1
I
te
ra
tio
n
2
I
te
ra
tio
n
3
I
te
ra
tio
n
4
I
te
ra
tio
n
5
I
te
ra
tio
n
6
I
te
ra
tio
n
7
I
te
ra
tio
n
8
I
te
ra
tio
n
9
I
te
ra
tio
n
1
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
-
8
6
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
NS
I
n
o
r
d
e
r
to
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
d
e
v
elo
p
ed
m
o
d
e
l,
we
p
r
esen
t
in
Fig
u
r
e
6
th
e
r
eg
r
ess
io
n
co
ef
f
icien
t R o
f
o
u
r
ANN
n
et
wo
r
k
wh
ich
d
escr
ib
es th
e
r
elat
io
n
s
h
ip
b
etwe
en
th
e
p
r
ed
icted
v
alu
es (
o
u
tp
u
t
)
an
d
th
e
s
im
u
lated
v
alu
es
(
tar
g
et
)
.
T
h
e
d
ata
s
h
o
u
ld
f
all
alo
n
g
a
4
5
-
d
eg
r
ee
lin
e,
f
o
r
a
p
er
f
ec
t
f
it,
wh
er
e
th
e
n
etwo
r
k
o
u
tp
u
ts
ar
e
eq
u
al
to
th
e
ta
r
g
et
s
.
Fo
r
th
is
p
r
o
b
lem
,
th
e
f
it
is
g
o
o
d
f
o
r
all
d
ata
s
ets,
with
R
v
a
lu
es
in
ea
ch
ca
s
e
is
R
=1
.
I
n
ter
m
o
f
th
e
v
alu
e
o
f
th
is
co
ef
f
icien
t,
th
e
b
u
ilt
ANN
n
etwo
r
k
with
th
e
s
tr
u
ctu
r
e
(
4
-
25
-
25
-
1
)
is
ef
f
icien
t
to
p
r
ed
ict
th
e
n
o
tch
b
an
d
f
r
eq
u
en
cy
o
f
th
e
p
r
esen
ted
UW
B
an
ten
n
a.
Fig
u
r
e
6
.
R
eg
r
ess
io
n
cu
r
v
es o
f
tr
ain
ed
ANN
T
h
e
d
if
f
e
r
en
ce
b
etwe
en
th
e
n
o
tch
f
r
e
q
u
en
c
y
o
b
tain
ed
b
y
ANN,
s
im
u
lated
,
an
d
ca
lcu
la
ted
r
esu
lts
d
u
r
in
g
t
h
e
tr
ain
in
g
p
r
o
ce
s
s
is
p
lo
tted
in
Fig
u
r
e
7
.
As
s
h
o
wn
in
Fig
u
r
e
7
,
we
ca
n
n
o
tice
a
g
o
o
d
m
atch
in
g
b
etwe
en
th
e
s
im
u
lated
an
d
th
e
p
r
e
d
icted
ANN
n
o
tch
f
r
e
q
u
en
c
y
.
I
n
th
e
t
esti
n
g
p
r
o
ce
s
s
,
3
6
d
atasets
wh
ich
ar
e
n
o
t
in
v
o
l
v
ed
in
th
e
tr
ai
n
in
g
p
h
ase
ar
e
u
s
ed
to
ev
alu
ate
th
e
p
er
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
ANN
m
o
d
el.
Du
r
in
g
th
is
ev
alu
ati
o
n
,
th
e
ANN
n
o
tch
b
an
d
f
r
e
q
u
en
cy
o
u
tp
u
t
is
co
m
p
ar
ed
with
t
h
e
th
eo
r
etica
l
an
d
s
im
u
lated
f
i
n
d
i
n
g
s
as
s
h
o
wn
in
T
ab
le
4
.
I
n
o
r
d
er
to
v
alid
ate
th
e
p
r
o
p
o
s
ed
ANN
m
o
d
el
p
er
f
o
r
m
ed
b
y
th
e
k
-
f
o
ld
cr
o
s
s
v
alid
atio
n
m
eth
o
d
,
we
s
im
u
late
th
e
p
r
o
p
o
s
ed
UW
B
an
ten
n
a
u
s
in
g
th
e
h
ig
h
f
r
eq
u
en
cy
s
tr
u
ctu
r
al
s
im
u
lato
r
(
HFSS
)
s
im
u
lato
r
.
T
h
e
v
alid
atio
n
co
n
s
is
ts
o
f
d
eter
m
in
i
n
g
th
e
c
o
r
r
esp
o
n
d
in
g
n
o
tch
f
r
eq
u
e
n
cy
b
y
ex
p
lo
itin
g
t
h
e
VSW
R
r
esu
lt
an
d
co
m
p
ar
in
g
it
with
th
e
th
eo
r
etic
al
o
n
e,
an
d
th
e
ANN
f
in
d
in
g
s
as
s
h
o
wn
in
T
ab
le
5
.
Du
r
in
g
th
e
test
an
d
v
alid
atio
n
p
r
o
ce
s
s
,
g
o
o
d
ac
cu
r
ac
y
is
atta
in
ed
b
etwe
en
th
e
s
im
u
lated
a
n
d
th
e
ANN
r
esu
lts
.
I
n
Fig
u
r
e
8
,
we
p
lo
tted
th
e
s
im
u
lated
S
11
r
esu
lt
u
s
in
g
b
o
t
h
elec
tr
o
m
ag
n
etic
s
im
u
lato
r
s
H
FS
S
an
d
C
ST.
Fro
m
th
e
-
1
0
d
B
b
an
d
wid
th
s
im
u
lated
r
esu
lts
illu
s
tr
ated
in
Fig
u
r
e
8
,
it
is
clea
r
th
at
th
e
an
t
en
n
a
co
v
e
r
s
th
e
en
tire
b
a
n
d
f
r
o
m
3
.
8
5
-
1
2
.
3
8
GHz
ex
ce
p
t
th
e
n
o
tc
h
ed
b
a
n
d
5
.
1
-
6
GHz
,
th
is
r
esu
lt
in
d
icate
s
th
at
th
e
p
r
o
p
o
s
ed
an
ten
n
a
ca
n
b
e
a
b
etter
ca
n
d
id
ate
f
o
r
UW
B
ap
p
licatio
n
s
.
T
h
e
d
if
f
er
en
ce
b
etwe
en
th
e
HFSS
an
d
C
ST
r
esu
lts
is
d
u
e
to
th
e
f
ac
t
th
at
th
e
u
s
ed
m
eth
o
d
s
b
y
th
e
two
s
im
u
lato
r
s
ar
e
d
if
f
er
e
n
t;
HFSS
is
b
ase
d
o
n
f
in
ite
elem
en
t
m
et
h
o
d
(
FEM
)
wh
ile
C
ST
is
b
ased
u
p
o
n
f
in
ite
in
te
g
r
atio
n
t
ec
h
n
iq
u
e
(
FIT
)
.
Fig
u
r
e
7
.
C
o
m
p
a
r
ativ
e
r
esu
lts
o
f
th
e
n
o
tch
f
r
eq
u
e
n
cy
: simu
la
ted
,
p
r
ed
icted
,
an
d
ca
lcu
lated
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
P
r
ed
ictin
g
th
e
n
o
tc
h
b
a
n
d
fr
eq
u
en
cy
o
f
a
n
u
ltr
a
-
w
id
eb
a
n
d
a
n
ten
n
a
u
s
in
g
…
(
La
h
ce
n
A
g
u
n
i
)
7
T
ab
le
4
.
T
esti
n
g
p
r
o
ce
s
s
A
n
t
e
n
n
a
p
a
r
a
me
t
e
r
s
f
c
[
G
H
z
]
A
n
t
e
n
n
a
p
a
r
a
me
t
e
r
s
f
c
[
G
H
z
]
a
b
g
S
I
M
ANN
C
A
L
a
B
g
S
I
M
ANN
C
A
L
2
4
.
5
1
2
.
2
6
,
3
9
2
6
,
3
9
2
6
7
,
1
6
1
3
.
5
5
4
22
5
,
6
2
4
5
,
6
1
3
3
5
,
6
5
2
3
2
5
.
5
1
2
.
2
5
,
6
2
4
5
,
6
2
0
6
6
,
3
8
9
3
3
.
5
6
4
2
.
2
5
,
0
3
6
5
,
0
3
4
4
5
,
1
6
0
4
2
5
.
5
4
2
.
2
6
,
5
8
4
6
,
5
8
4
8
7
,
6
2
1
2
3
.
5
4
1
4
.
3
4
,
4
3
6
4
,
4
4
1
9
4
,
1
9
2
2
4
4
.
5
4
2
.
2
5
,
4
4
4
5
,
4
4
6
4
5
,
3
6
1
4
.
5
5
3
4
.
3
3
,
7
8
8
3
,
7
8
0
1
3
,
5
0
8
9
5
5
.
5
1
22
3
.
92
3
.
92
3
,
7
7
6
6
2
.
5
3
4
6
.
15
6
,
7
7
6
6
,
7
8
4
9
6
,
2
4
6
7
2
3
.
5
2
4
.
3
6
.
74
6
,
7
4
1
1
6
,
7
9
5
3
2
53
15
2
.
2
6
,
0
3
2
6
,
0
2
8
7
6
,
7
1
4
9
2
3
.
5
4
4
.
3
7
.
82
7
,
8
2
3
4
7
,
9
7
1
2
5
.
3
3
.
5
2
.
2
6
,
6
6
8
6
,
6
6
8
6
7
,
5
7
2
5
2
5
.
5
4
4
.
3
5
,
7
0
8
5
,
7
0
7
1
5
,
9
2
1
9
3
4
.
3
1
.
5
2
.
2
5
.
72
5
.
72
5
,
9
4
7
1
4
4
.
5
3
4
.
3
4
,
1
7
2
4
,
1
6
8
7
3
,
9
8
5
5
3
5
.
3
25
2
.
2
5
,
3
9
6
5
,
3
9
7
2
5
,
6
6
3
1
4
4
.
5
4
4
.
3
4
,
4
1
2
4
,
4
1
0
1
4
,
1
6
5
7
4
4
.
3
15
2
.
2
4
.
76
4
,
7
6
1
4
,
8
9
6
2
5
5
.
5
4
4
.
3
3
,
6
0
8
3
,
6
0
6
3
,
2
4
4
5
2
5
.
3
1
.
5
4
.
3
5
,
1
6
8
5
,
1
6
8
6
5
,
2
1
7
7
2
2
.
5
4
6
.
15
8
,
8
6
4
8
,
8
7
1
3
8
,
2
9
8
4
2
5
.
3
2
.
5
4
.
3
5
,
5
0
4
5
,
5
0
3
3
5
,
5
3
0
9
3
4
.
5
2
6
.
15
4
.
4
4
,
3
9
8
9
3
,
9
9
8
6
2
5
.
3
4
.
5
4
.
3
6
,
0
3
2
6
,
0
2
5
6
6
,
2
8
5
4
5
5
.
5
2
6
.
15
2
,
9
1
2
2
,
9
1
6
2
,
6
0
9
6
3
4
.
3
3
.
5
4
.
3
5
.
59
5
,
5
9
2
6
5
,
1
3
6
2
2
.
5
4
1
2
.
2
6
.
2
6
,
2
0
0
4
6
,
6
9
9
7
4
5
.
3
3
.
5
4
.
3
4
,
1
7
2
4
,
1
7
1
6
3
,
8
0
4
5
3
.
5
4
1
2
.
2
5
,
3
3
6
5
,
3
3
3
4
5
,
3
9
5
2
3
5
.
3
3
.
5
6
.
15
4
,
3
8
8
4
,
3
8
7
7
3
,
9
7
8
6
3
.
5
4
2
2
.
2
5
,
7
9
2
5
,
7
8
8
5
,
6
5
2
3
5
5
.
3
2
.
5
6
.
15
2
.
96
2
,
9
5
5
5
2
,
6
8
9
3
3
.
5
5
2
2
.
2
5
,
0
2
4
5
,
0
2
4
1
5
,
1
6
0
4
5
5
.
3
3
.
5
6
.
15
3
.
08
3
,
0
7
8
1
2
,
7
8
3
6
T
ab
le
5
.
C
o
m
p
a
r
ativ
e
r
esu
lts
o
f
th
e
n
o
tch
f
r
eq
u
e
n
cy
o
f
th
e
p
r
o
p
o
s
ed
an
ten
n
a
M
e
t
h
o
d
S
i
mu
l
a
t
e
d
ANN
C
a
l
c
u
l
a
t
e
d
N
o
t
c
h
f
r
e
q
u
e
n
c
y
[
G
H
z
]
5
,
5
1
7
5
,
5
1
6
5
,
1
7
0
Fig
u
r
e
8
.
R
ef
lectio
n
c
o
ef
f
icien
t o
f
th
e
p
r
esen
ted
an
ten
n
a
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
we
u
s
ed
th
e
AN
N
tech
n
iq
u
e
to
p
r
ed
ict
with
h
ig
h
ac
cu
r
ac
y
th
e
n
o
tch
f
r
e
q
u
e
n
cy
o
f
a
n
UW
B
an
ten
n
a.
T
h
e
p
r
ec
is
io
n
o
f
th
e
p
r
o
p
o
s
ed
tech
n
i
q
u
e
h
as
b
ee
n
im
p
r
o
v
ed
b
y
em
p
lo
y
i
n
g
th
e
k
-
f
o
l
d
cr
o
s
s
v
alid
atio
n
m
eth
o
d
.
T
h
e
d
e
v
e
lo
p
ed
ANN
tech
n
iq
u
e
was
ev
alu
ate
d
u
s
in
g
d
if
f
er
en
t
cr
ite
r
ia.
Hen
ce
,
d
u
r
i
n
g
th
e
tr
ain
in
g
an
d
test
in
g
p
h
ases
,
we
h
av
e
d
eter
m
in
ed
t
h
e
m
ea
n
ab
s
o
lu
te
p
er
ce
n
ta
g
e
er
r
o
r
(
M
APE)
an
d
th
e
m
ea
n
s
q
u
ar
ed
er
r
o
r
(
MSE
)
b
etwe
en
th
e
s
im
u
lated
an
d
t
h
e
p
r
e
d
icted
ANN
n
o
tch
f
r
eq
u
en
cy
,
g
o
o
d
f
in
d
in
g
s
ar
e
attain
e
d
.
Fu
r
th
er
m
o
r
e,
we
d
et
er
m
in
ed
th
e
r
eg
r
ess
io
n
co
ef
f
i
cien
t
wh
ich
in
d
icate
s
a
v
alu
e
o
f
R
=1
v
er
if
y
in
g
a
g
o
o
d
c
o
n
s
is
ten
cy
b
etwe
en
t
h
e
s
im
u
lated
an
d
th
e
p
r
ed
icte
d
n
o
tch
f
r
eq
u
e
n
cy
.
T
h
e
p
r
o
p
o
s
ed
UW
B
an
ten
n
a
d
esig
n
ed
in
HFSS
an
d
C
ST
s
i
m
u
lato
r
s
ex
h
ib
its
a
n
o
tc
h
b
an
d
f
r
o
m
5
.
1
t
o
6
.
0
GHz
ce
n
ter
e
d
at
5
.
5
1
GHz
,
th
is
ch
ar
ac
ter
is
tic
allo
ws
to
av
o
id
in
ter
f
er
en
ce
s
b
etwe
en
(
W
L
AN)
I
E
E
E
8
0
2
.
1
1
a
an
d
HI
PERLAN/2
W
L
AN
ap
p
licatio
n
s
an
d
UW
B
tech
n
o
lo
g
y
.
RE
F
E
R
E
NC
E
S
[1
]
G
a
rg
R
.
,
Bh
a
rti
a
P
.
,
Ba
h
l
I
.
,
Iti
p
ib
o
o
n
A.
,
“
M
icro
stri
p
a
n
ten
n
a
d
e
sig
n
h
a
n
d
b
o
o
k
,”
Artec
h
Ho
u
se
,
Ch
a
p
ter
I:
2
,
Bo
sto
n
–
L
o
n
d
o
n
.
2
0
0
0
.
[2
]
P
o
z
a
r
D
.
M.
,
“
M
icro
stri
p
An
ten
n
a
s
,”
Pro
c
.
IEE
E
,
v
o
l.
80
,
p
p
.
79
-
81
,
1
9
9
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6
9
3
0
T
E
L
KOM
NI
KA
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l Co
n
tr
o
l
,
Vo
l.
19
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
1
:
1
-
8
8
[3
]
Ik
ra
m
M
.
,
S
h
a
ra
wi
M
.
S
.
,
S
h
a
m
im
A.
,
“
A n
o
v
e
l
v
e
ry
wid
e
b
a
n
d
in
teg
ra
ted
a
n
te
n
n
a
s
y
ste
m
fo
r
4
G
a
n
d
5
G
mm
-
wa
v
e
a
p
p
li
c
a
ti
o
n
s
,”
M
icr
o
w
Op
t
T
e
c
h
n
o
l
L
e
tt
.
,
v
o
l.
59
,
p
p
.
3
0
8
2
-
3
0
8
8
,
2
0
1
7
.
[4
]
Ka
u
sh
ik
M
.
,
S
a
rk
a
r
P
.
P.
,
“
A
c
o
m
p
a
c
t
lo
w
-
p
r
o
fil
e
wi
d
e
b
a
n
d
U
-
s
h
a
p
e
a
n
te
n
n
a
wit
h
sl
o
tt
e
d
c
ircu
la
r
g
r
o
u
n
d
p
lan
e
,”
In
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
E
lec
tro
n
i
c
s a
n
d
Co
mm
u
n
ica
ti
o
n
s
,
v
o
l.
70
,
n
o
.
3
,
p
p
.
3
3
6
-
3
4
0
,
2
0
1
6
.
[5
]
Ch
o
n
g
C
.
C
.
,
Wata
n
a
b
e
F
.
,
In
a
m
u
ra
H.
,
“
P
o
ten
ti
a
l
o
f
UWB
tec
h
n
o
lo
g
y
fo
r
t
h
e
n
e
x
t
g
e
n
e
ra
ti
o
n
wire
les
s
c
o
m
m
u
n
ica
ti
o
n
s
,”
Pro
c
e
e
d
in
g
s
o
f
9
t
h
IEE
E
I
n
ter
n
a
ti
o
n
a
l
S
y
mp
o
siu
m
S
p
re
a
d
S
p
e
c
tru
m
T
e
c
h
n
i
q
u
e
s
a
n
d
Ap
p
li
c
a
t
io
n
s
,
M
a
n
a
u
s
-
Am
a
z
o
n
,
Bra
z
il
,
pp.
4
2
2
-
4
2
9
,
2
0
0
6
.
[6
]
Ya
n
g
L
.
,
G
ian
n
a
k
is
G
.
B.
,
“
Ultra
-
wid
e
b
a
n
d
c
o
m
m
u
n
ica
ti
o
n
s
—
a
n
i
d
e
a
wh
o
se
t
ime
h
a
s
c
o
m
e
,”
IE
EE
S
ig
n
a
l
Pro
c
e
ss
M
a
g
.
,
v
o
l
.
21
,
n
o
.
6
,
p
p
.
26
-
54
,
2
0
1
4
.
[7
]
Ba
lan
is C
.
A.
,
“
An
ten
n
a
T
h
e
o
r
y
An
a
ly
sis a
n
d
De
sig
n
,”
J
o
h
n
W
il
e
y
,
Ne
w Yo
rk
,
USA
,
2
0
0
5
.
[8
]
F
e
d
e
ra
l
Co
m
m
u
n
ica
ti
o
n
s
Co
m
m
i
ss
io
n
,
“
Re
v
isi
o
n
o
f
P
a
rt
1
5
o
f
th
e
Co
m
m
issio
n
’s
Ru
le
Re
g
a
rd
i
n
g
Ultra
-
Wi
d
e
b
a
n
d
Tran
sm
issio
n
S
y
ste
m
,”
Fed
e
ra
l
C
o
mm
u
n
ic
a
ti
o
n
s Co
mm
issio
n
,
2
0
0
2
.
[9
]
Hu
ss
a
in
N
.
,
Je
o
n
g
M
.
,
P
a
r
k
J
.
,
R
h
e
e
S
.
,
Kim
P
.
,
Kim
N.
,
“
A
c
o
m
p
a
c
t
siz
e
2
.
9
-
2
3
.
5
G
Hz
m
icro
strip
p
a
tch
a
n
ten
n
a
with
WL
AN
b
a
n
d
-
re
jec
ti
o
n
,”
M
ic
ro
w
.
Op
t
.
T
e
c
h
n
o
l
.
L
e
tt
.
,
v
o
l.
61
,
n
o
.
5
,
p
p
.
1
3
0
7
-
1
3
1
3
,
2
0
1
9
.
[
1
0
]
Qin
g
-
Xi
n
C
.
,
Yi
n
g
-
Yi
n
g
Y.
,
“
A
Co
m
p
a
c
t
Ult
ra
wid
e
b
a
n
d
A
n
ten
n
a
wit
h
3
.
4
/5
.
5
G
Hz
Du
a
l
Ba
n
d
-
N
o
tch
e
d
Ch
a
ra
c
teristics
,”
IEE
E
T
ra
n
s
a
c
ti
o
n
s
o
n
An
ten
n
a
s a
n
d
Pro
p
a
g
a
ti
o
n
,
v
o
l
.
56
,
n
o
.
12
,
2
0
0
8
.
[
1
1
]
S
y
e
d
A
.
,
Al
d
h
a
h
e
ri
R
.
W.
,
“
A v
e
r
y
c
o
m
p
a
c
t
a
n
d
l
o
w p
ro
fi
le UWB
p
lan
a
r
a
n
ten
n
a
with
WL
AN
b
a
n
d
re
jec
ti
o
n
,”
T
h
e
S
c
i
e
n
ti
fi
o
c
W
o
rld
J
o
u
rn
a
l,
v
o
l.
2
0
1
,
p
p
.
1
-
7
,
2
0
1
6
.
[
1
2
]
Bo
n
g
H
.
U
.
,
Hu
ss
a
in
N
.
,
R
h
e
e
S
.
Y
.
,
G
il
S
.
K
.
,
Kim
N.
,
“
De
sig
n
o
f
a
n
UWB
a
n
ten
n
a
wit
h
two
slit
s
f
o
r
5
G
/W
LAN
-
n
o
tch
e
d
b
a
n
d
s
,”
M
icr
o
w
Op
t
T
e
c
h
n
o
l
L
e
tt
.
,
v
o
l.
61
,
n
o
.
5
,
p
p
.
1
2
9
5
-
1
3
0
0
,
2
0
1
9
.
[
1
3
]
S
y
e
d
A
.
,
Ald
h
a
h
e
ri
R
.
W.
,
“
A
n
e
w
in
se
t
-
fe
d
UWB
p
ri
n
ted
a
n
t
e
n
n
a
wit
h
tri
p
le
3
.
5
/
5
.
5
/7
.
5
-
G
H
z
b
a
n
d
-
n
o
tch
e
d
c
h
a
ra
c
teristics
,”
T
u
rk
J
El
e
c
En
g
&
Co
mp
S
c
i
.
,
v
o
l.
26
,
n
o
.
3
,
p
p
.
1
1
9
0
-
1
2
0
1
,
2
0
1
8
.
[
1
4
]
Oja
ro
u
d
i
N
.
,
Oja
ro
u
d
i
M
.
,
“
No
v
e
l
De
sig
n
o
f
Du
a
l
Ba
n
d
-
No
tch
e
d
M
o
n
o
p
o
le An
ten
n
a
with
Ba
n
d
wi
d
t
h
En
h
a
n
c
e
m
e
n
t
fo
r
UWB
A
p
p
li
c
a
ti
o
n
s
,”
IEE
E
An
ten
n
a
s
a
n
d
W
ire
les
s P
ro
p
a
g
a
ti
o
n
L
e
tt
e
rs
,
v
o
l
.
12
,
p
p
.
6
9
8
-
7
0
1
,
2
0
1
3
.
[
1
5
]
Diss
a
n
a
y
a
k
e
T
.
,
Esse
ll
e
K
.
P.
,
“
P
re
d
ictio
n
o
f
th
e
No
tch
F
re
q
u
e
n
c
y
o
f
S
l
o
t
Lo
a
d
e
d
P
ri
n
ted
UWB
A
n
ten
n
a
s
,”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
An
ten
n
a
s a
n
d
Pr
o
p
a
g
a
ti
o
n
,
v
o
l
.
55
,
n
o
.
11
,
p
p
.
3
3
2
0
-
3
3
2
5
,
2
0
0
7
.
[
1
6
]
Ke
rk
h
o
ff
A
.
,
Li
n
g
H.
,
“
De
sig
n
o
f
a
p
lan
a
r
a
n
ten
n
a
f
o
r
u
se
with
u
lt
ra
-
wid
e
b
a
n
d
(UWB)
h
a
v
i
n
g
a
b
a
n
d
-
n
o
tch
e
d
c
h
a
ra
c
teristic
,
”
Pro
c
.
IEE
E
In
t.
S
y
mp
.
An
te
n
n
a
s P
ro
p
a
g
a
t
.
,
v
o
l.
1
,
p
p
.
8
3
0
-
833
,
2
0
0
3
.
[
1
7
]
An
e
e
sh
M
.
,
S
in
g
h
A
.,
An
sa
ri
J
.
A
.
,
“
Ka
m
a
k
sh
i,
S
a
y
e
e
d
S
S
.
I
n
v
e
stig
a
ti
o
n
s
f
o
r
p
e
rf
o
rm
a
n
c
e
imp
ro
v
e
m
e
n
t
o
f
X
-
sh
a
p
e
d
RM
S
A
u
si
n
g
a
rti
ficia
l
n
e
u
ra
l
n
e
t
wo
rk
b
y
p
re
d
ictin
g
sl
o
t
siz
e
,”
Pro
g
re
ss
in
El
e
c
tro
ma
g
n
e
ti
c
s
Res
e
a
rc
h
C.
,
v
o
l.
47
,
pp.
55
-
63
,
2
0
1
4
.
[
1
8
]
Tu
rk
e
r
N
.
,
G
u
n
e
s
F
.
,
Yild
ir
im
T.
,
“
Artifi
c
ial
n
e
u
ra
l
d
e
sig
n
o
f
m
icro
strip
a
n
ten
n
a
s
,”
T
u
rk
J
El
e
c
E
n
g
&
C
o
mp
S
c
i
.
,
v
o
l.
14
,
n
o
.
3
,
p
p
.
4
4
5
-
4
5
3
,
2
0
0
6
.
[
1
9
]
Ak
d
a
g
li
A
.
,
T
o
k
tas
A
.
,
Ka
y
a
b
a
si
A
.
,
De
v
e
li
I.
,
“
An
a
p
p
li
c
a
ti
o
n
o
f
a
rti
ficia
l
n
e
u
ra
l
n
e
two
rk
to
c
o
m
p
u
te
th
e
re
so
n
a
n
t
fre
q
u
e
n
c
y
o
f
Es
h
a
p
e
d
c
o
m
p
a
c
t
m
icro
strip
a
n
ten
n
a
s
,”
J
.
E
lec
tr.
En
g
.
,
v
o
l.
64
,
n
o
.
5
,
p
p
.
3
1
7
-
3
2
2
,
2
0
1
3
.
[
2
0
]
An
e
e
sh
M
.
,
A
n
sa
ri
J
.
A
.
,
S
i
n
g
h
A
.
,
Ka
m
a
k
sh
i,
S
a
y
e
e
d
S
.
S.
,
“
An
a
ly
sis
o
f
m
icro
strip
l
i
n
e
fe
e
d
sl
o
t
lo
a
d
e
d
p
a
tch
a
n
ten
n
a
u
sin
g
a
rti
ficia
l
n
e
u
ra
l
n
e
t
wo
rk
,”
Pr
o
g
re
ss
in
El
e
c
tro
ma
g
n
e
ti
c
s R
e
se
a
rc
h
B
,
v
o
l.
58
,
p
p
.
35
-
46
,
2
0
1
4
.
[
2
1
]
Ve
g
n
i
L
.
,
To
sc
a
n
o
A
.
,
“
An
a
ly
sis o
f
m
icro
strip
a
n
ten
n
a
s
u
si
n
g
n
e
u
r
a
l
n
e
two
rk
s
,”
IEE
E
T
ra
n
s.
M
a
g
n
.,
v
o
l
.
33
,
n
o
.
2
,
p
p
.
1
4
1
4
-
1
4
1
9
,
1
9
9
7
.
[
2
2
]
M
ish
ra
R
.
K
.
,
P
a
tn
a
i
k
A.
,
“
Ne
u
ra
l
n
e
two
rk
-
b
a
se
d
CAD
m
o
d
e
l
fo
r
th
e
d
e
sig
n
o
f
sq
u
a
re
p
a
tch
a
n
ten
n
a
s
,”
IEE
E
T
ra
n
sa
c
ti
o
n
s
o
n
An
ten
n
a
s a
n
d
Pr
o
p
a
g
a
ti
o
n
,
v
o
l
.
46
,
n
o
.
12
,
p
p
.
1
8
9
0
-
1
8
9
1
,
1
9
9
8
.
[
2
3
]
P
a
tn
a
ik
A
.
,
M
ish
ra
R
.
K
.
,
P
a
tra
G
.
K
.
,
e
t
a
l
.
,
“
An
a
rti
flcia
l
n
e
u
ra
l
n
e
two
rk
m
o
d
e
l
fo
r
e
ffe
c
ti
v
e
d
iele
c
tri
c
c
o
n
sta
n
t
o
f
m
icro
strip
li
n
e
,”
IEE
E
T
ra
n
sa
c
ti
o
n
s o
n
An
ten
n
a
s a
n
d
Pro
p
a
g
a
ti
o
n
,
v
o
l
.
45
,
n
o
.
11
,
1
9
9
7
.
[
2
4
]
Hy
n
d
m
a
n
R
.
J
.
,
Ko
e
h
ler
A
.
B.
,
“
An
o
th
e
r
l
o
o
k
a
t
m
e
a
su
re
s
o
f
fo
re
c
a
st
a
c
c
u
ra
c
y
,”
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
Fo
re
c
a
stin
g
,
v
o
l.
22
,
n
o
.
4
,
p
p
.
6
7
9
-
6
8
8
,
2
0
0
6
.
[
2
5
]
Bo
u
t
h
e
v
il
lai
n
K
.
,
M
a
th
is
A.
,
“
P
ré
v
isio
n
s
:
m
e
su
re
s,
e
rre
u
rs
e
t
p
rin
c
i
p
a
u
x
ré
su
lt
a
ts
,
”
Eco
n
o
mie
e
t
S
t
a
ti
stiq
u
e
,
p
p
.
89
-
1
0
0
,
1
9
9
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.