I
nte
rna
t
io
na
l J
o
urna
l o
f
Adv
a
nces in Applie
d Science
s
(
I
J
AAS)
Vo
l.
9
,
No
.
4
,
Dec
em
b
e
r
2
0
2
0
,
p
p
.
276
~
283
I
SS
N:
2
2
5
2
-
8
8
1
4
,
DOI
: 1
0
.
1
1
5
9
1
/ijaas.v
9
.
i4
.
pp
2
7
6
-
283
276
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
a
s
.
ia
esco
r
e.
co
m
Desig
n of f
requen
cy
select
iv
e su
rfac
e comp
rising
of
di
po
les
using
artificial n
e
ura
l net
wo
rk
M
o
no
j
it
Rudra
1
,
P
So
ni Re
d
dy
2
,
Ra
j
a
t
s
ub
hra
Cha
k
ra
bo
rt
y
3
,
P
a
r
t
ha
P
ra
t
i
m
Sa
r
k
a
r
4
1,
2,
4
De
p
a
rtme
n
t
o
f
E
n
g
i
n
e
e
rin
g
&
Tec
h
n
o
l
o
g
ica
l
S
tu
d
ies
,
Un
i
v
e
rsity
o
f
Ka
ly
a
n
i,
I
n
d
ia
3
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
E
n
g
i
n
e
e
rin
g
,
F
u
tu
re
I
n
stit
u
t
e
o
f
En
g
i
n
e
e
rin
g
a
n
d
M
a
n
a
g
e
m
e
n
t,
In
d
ia
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ap
r
1
6
,
2
0
2
0
R
ev
is
ed
J
u
n
14
,
2
0
2
0
Acc
ep
ted
J
u
n
1
8
2
0
2
0
Th
is
p
a
p
e
r
d
e
p
icts
th
e
d
e
sig
n
o
f
F
re
q
u
e
n
c
y
S
e
lec
ti
v
e
S
u
rf
a
c
e
(F
S
S
)
c
o
m
p
risin
g
o
f
d
i
p
o
les
u
si
n
g
Arti
ficia
l
Ne
u
ra
l
Ne
two
r
k
(AN
N).
It
h
a
s
b
e
e
n
o
b
se
rv
e
d
t
h
a
t
w
it
h
t
h
e
c
h
a
n
g
e
o
f
th
e
d
ime
n
si
o
n
s
a
n
d
p
e
rio
d
ici
ty
o
f
F
S
S
,
th
e
re
so
n
a
ti
n
g
fre
q
u
e
n
c
y
o
f
t
h
e
F
S
S
c
h
a
n
g
e
s
.
T
h
is
c
h
a
n
g
e
i
n
re
so
n
a
ti
n
g
fre
q
u
e
n
c
y
h
a
s
b
e
e
n
stu
d
ied
a
n
d
i
n
v
e
sti
g
a
ted
u
sin
g
sim
u
latio
n
so
ftwa
re
.
Th
e
sim
u
late
d
d
a
ta we
re
u
se
d
to
train
th
e
p
ro
p
o
se
d
AN
N m
o
d
e
ls.
T
h
e
train
e
d
AN
N
m
o
d
e
ls
a
re
fo
u
n
d
to
p
re
d
ict
th
e
F
S
S
c
h
a
ra
c
teristics
p
re
c
ise
ly
with
n
e
g
li
g
ib
le
e
rr
o
r.
Co
m
p
a
re
d
to
t
ra
d
it
io
n
a
l
EM
sim
u
latio
n
so
f
twa
re
s
(li
k
e
AN
S
OFT
De
sig
n
e
r),
th
e
p
ro
p
o
se
d
tec
h
n
iq
u
e
u
sin
g
AN
N
m
o
d
e
ls
is
fo
u
n
d
to
sig
n
ifi
c
a
n
t
ly
re
d
u
c
e
th
e
F
S
S
d
e
sig
n
c
o
m
p
lex
it
y
a
n
d
c
o
m
p
u
tati
o
n
a
l
ti
m
e
.
Th
e
F
S
S
sim
u
latio
n
s
we
re
m
a
d
e
u
sin
g
AN
S
OFT
De
sig
n
e
r
v
2
s
o
ftwa
re
a
n
d
th
e
n
e
u
ra
l
n
e
two
r
k
wa
s
d
e
sig
n
e
d
u
sin
g
M
ATLAB
s
o
ftwa
re
.
K
ey
w
o
r
d
s
:
ANN
FSS
MLP
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
:
Par
th
a
Pra
tim
Sar
k
ar
,
Dep
ar
tm
en
t o
f
E
n
g
in
ee
r
in
g
&
T
ec
h
n
o
lo
g
ical
s
tu
d
ies,
Un
iv
er
s
ity
o
f
Kaly
an
i,
Nad
ia,
W
est B
en
g
al,
7
4
1
2
3
5
,
I
n
d
ia.
E
m
ail:
p
ar
th
ab
e9
1
@
y
ah
o
o
.
co
.
in
1.
I
NT
RO
D
UCT
I
O
N
I
n
m
icr
o
wav
e
en
g
in
ee
r
in
g
,
Fr
eq
u
en
cy
Selectiv
e
Su
r
f
ac
es
(
F
SS
s)
ar
e
p
lan
ar
p
er
io
d
ic
ar
r
ay
s
o
f
m
etal
p
atch
es o
n
a
s
u
b
s
tr
ate
o
r
s
lo
ts
o
n
a
co
n
d
u
ctin
g
s
h
ee
t th
at
f
u
n
ctio
n
as a
f
ilter
f
o
r
f
r
ee
s
p
ac
e
r
ad
iatio
n
[
1
]
.
Ma
n
y
au
th
o
r
s
h
av
e
m
a
d
e
an
aly
s
is
o
n
FS
S
th
r
o
u
g
h
E
M
n
u
m
er
ical
m
eth
o
d
s
,
s
u
ch
as
th
e
Me
th
o
d
o
f
Mo
m
en
t
[
2
]
.
B
u
t
th
ese
n
u
m
er
ical
m
eth
o
d
s
r
eq
u
ir
e
h
ig
h
c
o
m
p
u
tatio
n
al
co
s
t.
So
to
av
o
i
d
th
is
,
Ar
tific
ial
Neu
r
al
Netwo
r
k
(
ANN
)
wh
ich
ar
e
p
r
ev
io
u
s
ly
tr
ain
ed
with
r
esu
lts
o
b
tain
ed
b
y
M
eth
o
d
o
f
Mo
m
e
n
t
ca
n
b
e
u
s
ed
f
o
r
an
aly
s
is
o
f
FS
S
[
3
-
4
]
.
Als
o
,
o
th
e
r
alg
o
r
it
h
m
s
lik
e
Gen
etic
Alg
o
r
ith
m
(
GA)
an
d
Par
ticle
Swar
m
Op
tim
izatio
n
(
PS
O)
ca
n
b
e
b
len
d
e
d
with
ANN
f
o
r
f
aster
an
d
ac
c
u
r
ate
tr
ain
in
g
o
f
th
e
ANN
[
5
–
1
6
]
.
I
n
th
is
p
ap
e
r
,
p
atch
ty
p
e
FS
S
co
n
s
is
tin
g
o
f
d
i
p
o
les
(
as
s
h
o
w
n
in
Fig
u
r
e
1
)
is
u
s
ed
w
h
o
s
e
d
im
en
s
io
n
s
(
i.e
.
p
atch
len
g
th
‘
L
’
,
p
atc
h
w
id
th
‘
W
’
,
x
-
p
e
r
io
d
icity
‘
T
x
’
an
d
y
-
p
er
io
d
icity
‘
T
y
’
)
ar
e
v
a
r
ied
an
d
th
e
co
r
r
esp
o
n
d
in
g
r
eso
n
atin
g
f
r
eq
u
en
cies
ar
e
n
o
ted
.
T
h
en
u
s
in
g
th
ese
s
im
u
lated
r
esu
lts
,
s
o
m
e
n
eu
r
al
n
etwo
r
k
m
o
d
els
ar
e
d
esig
n
ed
an
d
tr
a
in
ed
,
wh
ich
ca
n
b
e
u
s
ed
f
o
r
f
aster
an
aly
s
is
an
d
d
esi
g
n
o
f
FS
S
co
m
p
r
is
in
g
o
f
d
ip
o
le
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
Desig
n
o
f freq
u
en
cy
s
elec
tive
s
u
r
fa
ce
co
mp
r
is
in
g
o
f d
ip
o
les
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
Mo
n
o
jit R
u
d
r
a
)
277
Fig
u
r
e
1
.
FS
S c
o
m
p
r
is
in
g
o
f
d
ip
o
le
s:
(
a)
g
eo
m
et
r
y
o
f
u
n
it c
e
ll,
(
b
)
g
e
o
m
etr
y
o
f
ar
r
ay
2.
B
ACK
P
RO
P
AG
AT
I
O
N
T
R
AINI
NG
A
L
G
O
RI
T
H
M
T
h
e
B
ac
k
p
r
o
p
ag
atio
n
is
an
alg
o
r
ith
m
f
o
r
s
u
p
er
v
is
ed
tr
a
in
in
g
o
f
ANN
in
wh
ich
th
e
er
r
o
r
is
p
r
o
p
a
g
ated
b
ac
k
war
d
f
o
r
u
p
d
ate
o
f
weig
h
ts
o
f
d
if
f
er
en
t
lay
er
s
o
f
th
e
n
eu
r
al
n
etwo
r
k
.
A
M
u
lti
-
L
ay
er
Per
ce
p
tr
o
n
(
ML
P)
is
s
h
o
wn
in
Fig
u
r
e
2
,
h
av
in
g
o
n
e
in
p
u
t la
y
er
,
o
n
e
h
id
d
en
lay
e
r
an
d
o
n
e
o
u
tp
u
t la
y
e
r
.
Fig
u
r
e
2
.
E
x
am
p
le
o
f
an
ANN
m
o
d
el
T
h
e
s
tep
s
f
o
r
B
ac
k
p
r
o
p
ag
atio
n
Alg
o
r
ith
m
a
r
e
as f
o
llo
ws:
1.
W
eig
h
ts
an
d
lear
n
in
g
r
ate
(
α
)
ar
e
in
itialized
Hid
d
en
L
ay
e
r
W
eig
h
ts
: w
1
,
w2
,
w3
,
w4
Ou
tp
u
t L
ay
er
W
eig
h
ts
: w
5
,
w6
,
w7
,
w8
B
ias
W
eig
h
ts
: a
1
,
a2
,
b
1
,
b
2
L
ea
r
n
in
g
R
ate
(
α
)
: 0
.
0
0
0
1
2.
Step
s
3
to
1
0
ar
e
p
er
f
o
r
m
e
d
w
h
en
s
to
p
p
in
g
co
n
d
itio
n
is
f
alse
.
3.
Step
s
4
to
9
ar
e
p
er
f
o
r
m
ed
f
o
r
ea
ch
tr
ain
in
g
p
air
.
4.
E
ac
h
in
p
u
t
u
n
it r
ec
eiv
es in
p
u
t
s
ig
n
al
(
i
i
)
an
d
s
en
d
s
it to
th
e
h
id
d
en
u
n
it.
5.
E
ac
h
h
id
d
e
n
u
n
it su
m
s
its
wei
g
h
ted
in
p
u
t sig
n
als to
ca
lcu
lat
e
th
e
n
et
in
p
u
t (
n
et
h
i
)
.
n
et
h
1
=
w1
i
1
+
w3
i
2
+ a1
n
et
h
2
=
w2
i1
+
w4
i
2
+
b
1
T
h
en
an
Activ
atio
n
f
u
n
ctio
n
is
ap
p
lied
to
t
h
e
n
et
in
p
u
t to
ca
l
cu
late
th
e
o
u
tp
u
t o
f
t
h
e
h
id
d
en
u
n
it (
h
i
).
T
h
e
o
u
tp
u
t o
f
th
e
h
id
d
en
lay
e
r
is
th
en
s
en
t to
th
e
o
u
tp
u
t la
y
e
r
u
n
its
.
h
i =
f
(
n
et
h
i)
Her
e
B
ip
o
lar
Sig
m
o
id
Activ
at
io
n
Fu
n
ctio
n
is
u
s
ed
: f
(
x
)
=
1
/
(
1
+
e
-
x)
6.
Similar
ly
,
f
o
r
ea
ch
o
u
tp
u
t
u
n
it
th
e
n
et
in
p
u
t
(
n
et
o
i
)
is
ca
lcu
l
ated
an
d
th
e
ac
tiv
atio
n
f
u
n
ctio
n
is
ap
p
lied
t
o
co
m
p
u
te
th
e
o
u
tp
u
t sig
n
als (
o
i
).
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
e
r
2
0
2
0
:
2
7
6
–
2
8
3
278
7.
E
ac
h
o
u
tp
u
t
u
n
it
r
ec
eiv
es
a
tar
g
et
p
atter
n
(
t
i
)
co
r
r
es
p
o
n
d
in
g
to
th
e
in
p
u
t
tr
ain
in
g
p
atter
n
(
i
i
)
an
d
co
m
p
u
tes th
e
er
r
o
r
c
o
r
r
ec
tio
n
ter
m
(
i
):
1
=
(
t
1
-
o
1
)
f
'
(
n
et
o
1
)
2
=
(
t
2
-
o
2
)
f
'
(
n
et
o
2
)
W
h
er
e,
f
'
(
x
)
is
th
e
d
er
iv
ativ
e
o
f
f
(
x
)
T
h
ese
er
r
o
r
co
r
r
ec
tio
n
ter
m
s
ar
e
u
s
ed
to
ca
lcu
late
th
e
ch
an
g
e
in
weig
h
ts
(
Δ
w
i
)
an
d
ch
an
g
e
in
b
ias
weig
h
ts
(
Δ
a
i
an
d
Δ
b
i
).
Δ
w
5
=
α
1
h
1
Δ
a
2
=
α
1
Similar
ly
Δ
w
6
,
Δ
w
7
,
Δ
w
8
an
d
Δ
b
2
ar
e
ca
lcu
lated
8.
Similar
ly
ea
ch
h
id
d
e
n
u
n
it c
al
cu
lates its
er
r
o
r
co
r
r
ec
tio
n
ter
m
(
i
j
)
:
56
=
1
w
5
f
'
(
n
et
h
1
)
+
2
w
6
f
'
(
n
et
h
1
)
78
=
1
w
7
f
'
(
n
et
h
2
)
+
2
w
8
f
'
(
n
et
h
2
)
T
h
ese
er
r
o
r
c
o
r
r
ec
tio
n
ter
m
s
a
r
e
u
s
ed
to
ca
lcu
late
th
e
ch
an
g
e
in
weig
h
ts
an
d
b
ias we
ig
h
ts
.
Δ
w
1
=
α
56
i
1
Δ
a
1
=
α
56
Similar
ly
Δ
w
2
,
Δ
w
3
,
Δ
w
4
an
d
Δ
b
1
ar
e
ca
lcu
lated
9.
E
ac
h
o
u
tp
u
t u
n
it a
n
d
ea
ch
h
id
d
en
u
n
it
u
p
d
ates its
b
ias we
ig
h
ts
an
d
weig
h
ts
.
w
i
(
n
ew)
=w
i
(
o
ld
)
+
Δ
w
i
a
i
(
n
ew)
=a
i
(
o
ld
)
+
Δ
a
i
b
i
(
n
ew)
=
b
i
(
o
ld
)
+
Δ
b
i
10.
T
h
e
s
to
p
p
in
g
co
n
d
itio
n
is
ch
e
ck
ed
.
T
h
e
s
to
p
p
i
n
g
c
o
n
d
itio
n
m
ay
b
e
ce
r
tain
n
u
m
b
e
r
o
f
ep
o
ch
s
r
ea
ch
ed
o
r
wh
en
th
er
e
is
p
r
e
v
io
u
s
ly
s
ettled
m
in
im
u
m
er
r
o
r
b
etwe
en
ac
t
u
al
o
u
tp
u
t a
n
d
tar
g
et
o
u
t
p
u
t.
3.
RE
S
E
ARCH
M
E
T
H
O
D
Her
e
5
ML
Ps
ar
e
u
s
ed
.
E
ac
h
o
f
th
em
h
av
e
4
in
p
u
ts
an
d
1
o
u
t
p
u
t
as
s
h
o
wn
i
n
Fig
u
r
e
3
,
b
u
t
t
h
e
in
p
u
ts
an
d
o
u
t
p
u
ts
ar
e
d
if
f
er
en
t a
s
s
h
o
wn
in
T
ab
le
1
.
Fig
u
r
e
3
.
C
o
m
m
o
n
ANN
m
o
d
el
f
o
r
all
5
n
etwo
r
k
s
T
ab
le
1
.
I
n
p
u
ts
an
d
o
u
t
p
u
ts
o
f
th
e
5
ML
Ps
N
e
u
r
a
l
N
e
t
w
o
r
k
s
I
n
p
u
t
s
O
u
t
p
u
t
N
e
t
w
o
r
k
1
I
n
N
e
t
w
o
r
k
1
,
t
h
e
4
i
n
p
u
t
s
a
r
e
t
h
e
l
e
n
g
t
h
,
w
i
d
t
h
,
x
-
p
e
r
i
o
d
i
c
i
t
y
a
n
d
y
-
p
e
r
i
o
d
i
c
i
t
y
o
f
t
h
e
p
r
o
p
o
se
d
F
S
S
I
n
N
e
t
w
o
r
k
1
.
T
h
e
o
u
t
p
u
t
i
s
t
h
e
r
e
so
n
a
n
t
f
r
e
q
u
e
n
c
y
o
f
t
h
e
p
r
o
p
o
se
d
F
S
S
N
e
t
w
o
r
k
2
I
n
N
e
t
w
o
r
k
2
,
t
h
e
4
i
n
p
u
t
s
a
r
e
t
h
e
l
e
n
g
t
h
,
w
i
d
t
h
,
x
-
p
e
r
i
o
d
i
c
i
t
y
a
n
d
r
e
so
n
a
n
t
f
r
e
q
u
e
n
c
y
o
f
t
h
e
p
r
o
p
o
s
e
d
F
S
S
I
n
N
e
t
w
o
r
k
2
.
T
h
e
o
u
t
p
u
t
i
s
t
h
e
y
-
p
e
r
i
o
d
i
c
i
t
y
o
f
t
h
e
p
r
o
p
o
s
e
d
F
S
S
N
e
t
w
o
r
k
3
I
n
N
e
t
w
o
r
k
3
,
t
h
e
4
i
n
p
u
t
s
a
r
e
t
h
e
l
e
n
g
t
h
,
w
i
d
t
h
,
r
e
so
n
a
n
t
f
r
e
q
u
e
n
c
y
a
n
d
y
-
p
e
r
i
o
d
i
c
i
t
y
o
f
t
h
e
p
r
o
p
o
s
e
d
F
S
S
I
n
N
e
t
w
o
r
k
3
.
T
h
e
o
u
t
p
u
t
i
s
t
h
e
x
-
p
e
r
i
o
d
i
c
i
t
y
o
f
t
h
e
p
r
o
p
o
s
e
d
F
S
S
N
e
t
w
o
r
k
4
I
n
N
e
t
w
o
r
k
4
,
t
h
e
4
i
n
p
u
t
s
a
r
e
t
h
e
l
e
n
g
t
h
,
r
e
so
n
a
n
t
f
r
e
q
u
e
n
c
y
,
x
-
p
e
r
i
o
d
i
c
i
t
y
a
n
d
y
-
p
e
r
i
o
d
i
c
i
t
y
o
f
t
h
e
p
r
o
p
o
se
d
FSS
I
n
N
e
t
w
o
r
k
4
.
T
h
e
o
u
t
p
u
t
i
s
t
h
e
w
i
d
t
h
o
f
t
h
e
p
r
o
p
o
se
d
F
S
S
N
e
t
w
o
r
k
5
I
n
N
e
t
w
o
r
k
5
,
t
h
e
4
i
n
p
u
t
s
a
r
e
t
h
e
r
e
s
o
n
a
n
t
f
r
e
q
u
e
n
c
y
,
w
i
d
t
h
,
x
-
p
e
r
i
o
d
i
c
i
t
y
a
n
d
y
-
p
e
r
i
o
d
i
c
i
t
y
o
f
t
h
e
p
r
o
p
o
se
d
F
S
S
I
n
N
e
t
w
o
r
k
5
.
T
h
e
o
u
t
p
u
t
i
s
t
h
e
l
e
n
g
t
h
o
f
t
h
e
p
r
o
p
o
se
d
F
S
S
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
Desig
n
o
f freq
u
en
cy
s
elec
tive
s
u
r
fa
ce
co
mp
r
is
in
g
o
f d
ip
o
les
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
Mo
n
o
jit R
u
d
r
a
)
279
4.
RE
SU
L
T
S
A
ND
AN
AL
Y
SI
S
I
n
itially
,
th
e
p
r
o
p
o
s
ed
ANN
m
o
d
els
ar
e
tr
ain
ed
u
s
in
g
th
e
s
im
u
latio
n
r
esu
lts
o
b
tain
ed
f
r
o
m
ANSOFT
Desig
n
er
v
2
s
o
f
twar
e
.
T
h
e
tr
ai
n
in
g
d
ata
s
et
is
p
r
o
v
id
ed
i
n
T
a
b
le
2
.
T
ab
le
2
.
T
r
ai
n
in
g
d
ata
s
et
f
o
r
ANN
m
o
d
els
P
a
t
c
h
L
e
n
g
t
h
P
a
t
c
h
W
i
d
t
h
x
-
p
e
r
i
o
d
i
c
i
t
y
y
-
p
e
r
i
o
d
i
c
i
t
y
S
t
e
p
S
i
z
e
D
a
t
a
S
e
t
1
1
5
mm
1
.
5
mm
1
6
.
5
mm
2
.
4
mm
–
1
5
mm
0
.
9
mm
D
a
t
a
S
e
t
2
1
5
mm
1
.
5
mm
1
8
mm
2
.
4
mm
–
1
5
mm
0
.
9
mm
D
a
t
a
S
e
t
3
1
5
mm
1
.
5
mm
1
9
.
5
mm
1
.
5
mm
–
1
5
mm
0
.
9
mm
D
a
t
a
S
e
t
4
1
5
mm
1
.
5
mm
2
1
mm
2
.
4
mm
–
1
5
mm
0
.
9
mm
D
a
t
a
S
e
t
5
1
5
mm
1
.
5
mm
2
2
.
5
mm
1
.
9
5
mm
–
1
5
m
m
0
.
4
5
mm
D
a
t
a
S
e
t
6
1
5
mm
1
.
5
mm
1
5
.
5
mm
–
2
2
.
5
m
m
4
.
2
mm
0
.
5
mm
D
a
t
a
S
e
t
7
1
5
mm
1
.
5
mm
1
5
.
5
mm
–
2
2
.
5
m
m
6
.
9
mm
0
.
5
mm
D
a
t
a
S
e
t
8
1
5
mm
1
.
5
mm
1
5
.
5
mm
–
2
2
.
5
m
m
9
.
6
mm
0
.
5
mm
D
a
t
a
S
e
t
9
1
5
mm
1
.
5
mm
1
5
.
5
mm
–
2
2
.
5
m
m
1
2
.
3
mm
0
.
5
mm
D
a
t
a
S
e
t
1
0
1
5
mm
1
.
5
mm
1
5
.
5
mm
–
2
2
.
5
m
m
1
5
mm
0
.
5
mm
D
a
t
a
S
e
t
1
1
9
mm
1
.
2
mm
–
2
mm
2
2
.
5
mm
1
5
mm
0
.
3
mm
D
a
t
a
S
e
t
1
2
1
2
mm
0
.
5
mm
–
2
mm
2
2
.
5
mm
1
5
mm
0
.
1
mm
D
a
t
a
S
e
t
1
3
1
5
mm
0
.
1
mm
–
1
.
5
mm
2
2
.
5
mm
1
5
mm
0
.
1
mm
D
a
t
a
S
e
t
1
4
6
mm
–
2
0
mm
0
.
7
mm
2
2
.
5
mm
1
5
mm
1
mm
D
a
t
a
S
e
t
1
5
6
mm
–
2
0
mm
0
.
9
mm
2
2
.
5
mm
1
5
mm
1
mm
D
a
t
a
S
e
t
1
6
6
mm
–
2
0
mm
1
.
1
mm
2
2
.
5
mm
1
5
mm
1
mm
D
a
t
a
S
e
t
1
7
7
mm
–
2
0
mm
1
.
3
mm
2
2
.
5
mm
1
5
mm
1
mm
D
a
t
a
S
e
t
1
8
7
mm
–
2
0
mm
1
.
5
mm
2
2
.
5
mm
1
5
mm
1
mm
Nex
t,
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
ANN
m
o
d
els
in
ac
c
u
r
ately
p
r
e
d
ictin
g
th
e
FS
S
ch
ar
ac
ter
is
tics
is
v
alid
ated
u
s
in
g
a
test
d
ata.
Her
e,
an
FSS
co
m
p
r
is
in
g
o
f
d
ip
o
les
h
av
in
g
d
im
e
n
s
io
n
s
L
=
1
5
m
m
,
W
=
1
.
5
m
m
,
T
x
=
2
2
m
m
,
T
y
=
8
m
m
(
as
s
h
o
wn
in
Fig
u
r
e.
1
)
is
u
s
ed
as
a
te
s
t
d
ata
f
o
r
th
e
p
r
o
p
o
s
ed
ANN
m
o
d
el
v
alid
atio
n
.
T
h
e
d
i
m
en
s
io
n
s
co
n
s
id
er
ed
h
e
r
e
f
o
r
test
in
g
th
e
ANN
m
o
d
els
is
n
o
t
in
cl
u
d
ed
in
th
e
d
ata
s
et
u
s
ed
f
o
r
tr
ain
in
g
th
e
ML
Ps
.
T
h
e
FS
S
is
s
im
u
lated
u
s
in
g
ANS
OFT
Desig
n
er
v
2
s
o
f
twar
e
an
d
its
r
eso
n
an
t f
r
eq
u
en
cy
o
b
tain
ed
is
8
.
1
5
GHz
.
N
ex
t,
t
h
is
d
ata
is
u
s
ed
f
o
r
ch
ec
k
in
g
t
h
e
er
r
o
r
o
f
all
th
e
5
ML
Ps
,
as
s
h
o
wn
in
T
ab
le
3
.
Fin
ally
,
th
e
1
st
ML
P
(
h
a
v
in
g
L
,
W
,
T
x
an
d
T
y
as
in
p
u
ts
an
d
r
eso
n
atin
g
f
r
eq
u
e
n
cy
(
R
F)
a
s
o
u
tp
u
t)
is
u
s
ed
to
p
er
f
o
r
m
p
ar
a
m
etr
ic
s
tu
d
y
as
g
i
v
en
in
Sectio
n
4
.
2
.
4
.
1
.
Sin
g
le
v
a
lue c
heck
(
inte
rpo
la
t
io
n)
o
f
t
he
pro
po
s
ed
ANN
T
h
e
v
alid
atio
n
r
esu
lt
o
f
t
h
e
5
ML
Ps
u
s
in
g
th
e
s
in
g
le
test
d
ata
is
g
iv
en
in
T
a
b
le
3
.
ML
P1
i
s
f
o
u
n
d
to
p
r
ed
ict
th
e
r
eso
n
atin
g
f
r
e
q
u
e
n
cy
(
R
F)
o
f
th
e
FS
S
f
o
r
th
e
g
iv
en
s
et
o
f
FS
S
d
esig
n
p
a
r
a
m
eter
s
with
0
.
0
7
6
%
er
r
o
r
.
ML
P
2
–
5
p
r
e
d
icts
th
e
FS
S
d
esig
n
p
ar
am
eter
(
y
-
p
e
r
io
d
icity
,
x
-
p
er
io
d
icity
,
Patch
W
id
th
,
an
d
Patch
L
en
g
th
,
r
esp
ec
tiv
el
y
)
with
0
.
9
4
%,
1
.
7
2
4
%,
0
.
0
2
9
8
%,
a
n
d
0
.
0
5
6
% e
r
r
o
r
,
r
esp
ec
tiv
ely
f
o
r
th
e
g
iv
en
i
n
p
u
t set a
s
s
h
o
wn
in
T
ab
le
3
.
T
ab
le
3.
R
esu
lt f
o
r
s
in
g
le
v
alu
e
ch
ec
k
(
in
te
r
p
o
latio
n
)
ANN
N
e
t
w
o
r
k
I
n
p
u
t
1
I
n
p
u
t
2
I
n
p
u
t
3
I
n
p
u
t
4
S
i
mu
l
a
t
e
d
O
u
t
p
u
t
A
N
N
O
u
t
p
u
t
%
Er
r
o
r
M
LP
1
P
a
t
c
h
L
e
n
g
t
h
1
5
mm
P
a
t
c
h
W
i
d
t
h
1
.
5
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
2
2
mm
y
-
p
e
r
i
o
d
i
c
i
t
y
8
mm
RF
8
.
1
5
G
H
z
RF
8
.
1
5
6
2
G
H
z
0
.
0
7
6
%
M
LP
2
P
a
t
c
h
L
e
n
g
t
h
1
5
mm
P
a
t
c
h
W
i
d
t
h
1
.
5
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
2
2
mm
RF
8
.
1
5
G
H
z
y
-
p
e
r
i
o
d
i
c
i
t
y
8
mm
y
-
p
e
r
i
o
d
i
c
i
t
y
8
.
0
7
5
4
mm
0
.
9
4
%
M
LP
3
P
a
t
c
h
L
e
n
g
t
h
1
5
mm
P
a
t
c
h
W
i
d
t
h
1
.
5
mm
y
-
p
e
r
i
o
d
i
c
i
t
y
8
mm
RF
8
.
1
5
G
H
z
x
-
p
e
r
i
o
d
i
c
i
t
y
2
2
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
2
1
.
6
2
0
7
mm
1
.
7
2
4
%
M
LP
4
P
a
t
c
h
L
e
n
g
t
h
1
5
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
2
2
mm
y
-
p
e
r
i
o
d
i
c
i
t
y
8
mm
RF
8
.
1
5
G
H
z
P
a
t
c
h
W
i
d
t
h
1
.
5
mm
P
a
t
c
h
W
i
d
t
h
1
.
4
9
9
6
mm
0
.
0
2
9
8
%
M
LP
5
P
a
t
c
h
W
i
d
t
h
1
.
5
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
2
2
mm
y
-
p
e
r
i
o
d
i
c
i
t
y
8
mm
RF
8
.
1
5
G
H
z
P
a
t
c
h
L
e
n
g
t
h
1
5
mm
P
a
t
c
h
L
e
n
g
t
h
1
5
.
0
0
8
4
mm
0
.
0
5
6
%
An
FS
S
is
f
ab
r
icate
d
u
s
in
g
p
ar
am
eter
s
co
r
r
esp
o
n
d
in
g
to
ML
P1
g
iv
en
in
T
ab
le
3
.
T
h
e
f
ab
r
icate
d
p
r
o
to
ty
p
e
is
s
h
o
wn
in
Fig
u
r
e
4
(
a)
.
T
h
e
tr
an
s
m
is
s
io
n
ch
ar
ac
ter
is
tic
o
f
th
e
f
ab
r
icate
d
FS
S
is
m
ea
s
u
r
ed
u
s
in
g
R
&
S
VN
A
(
Z
NB
2
0
)
an
d
T
x
/R
x
h
o
r
n
an
ten
n
as.
T
h
e
s
im
u
lated
a
n
d
m
ea
s
u
r
ed
tr
an
s
m
is
s
io
n
(
S2
1
)
ch
ar
ac
ter
is
tic
o
f
th
e
f
ab
r
icate
d
FS
S
is
g
iv
en
in
Fig
u
r
e
4
(
b
)
.
I
t
is
f
o
u
n
d
th
at
th
e
r
eso
n
an
t
f
r
e
q
u
en
c
y
o
f
th
e
p
r
o
p
o
s
ed
FS
S is
n
ea
r
ly
s
am
e
in
b
o
t
h
s
im
u
latio
n
an
d
m
e
asu
r
em
en
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
e
r
2
0
2
0
:
2
7
6
–
2
8
3
280
(
a)
(
b
)
Fig
u
r
e
4
.
(
a
)
Fab
r
icate
d
p
r
o
to
t
y
p
e
o
f
th
e
p
r
o
p
o
s
ed
FS
S,
(
b
)
Simu
lated
an
d
m
ea
s
u
r
e
d
tr
an
s
m
is
s
io
n
ch
ar
ac
ter
is
tics
o
f
th
e
p
r
o
p
o
s
ed
FS
S
4
.
2
.
P
a
ra
m
e
t
ric
s
t
ud
y
I
n
th
is
s
ec
tio
n
,
ex
ten
s
iv
e
s
tu
d
y
h
as
b
ee
n
c
o
n
d
u
cted
to
test
t
h
e
p
er
f
o
r
m
an
ce
o
f
ML
P
1
.
M
L
P
1
ca
n
o
n
ly
p
r
o
v
id
e
th
e
r
eso
n
atin
g
f
r
eq
u
en
c
y
(
R
F)
as
th
e
o
u
tp
u
t
f
o
r
th
e
ch
an
g
e
in
FS
S
d
esig
n
p
ar
am
eter
s
-
Patch
L
en
g
th
,
Patch
W
id
th
,
x
-
p
er
i
o
d
icity
an
d
y
-
p
er
i
o
d
icity
.
T
h
e
v
alu
es
o
f
FS
S
d
esig
n
p
ar
am
eter
s
f
o
r
th
is
p
ar
am
etr
ic
s
tu
d
y
ar
e
tab
u
lated
in
T
ab
le
4
.
Her
e,
c
o
m
p
ar
is
o
n
is
d
o
n
e
b
etwe
en
th
e
s
im
u
l
ated
r
esu
lt
o
b
tain
ed
f
r
o
m
An
s
o
f
t D
esig
n
e
r
an
d
ANN.
T
ab
le
4
.
Data
f
o
r
p
a
r
am
etr
ic
s
tu
d
y
C
o
n
st
a
n
t
P
a
r
a
me
t
e
r
s
V
a
r
i
a
b
l
e
P
a
r
a
m
e
t
e
r
s
P
a
r
a
me
t
r
i
c
S
t
u
d
y
1
P
a
t
c
h
W
i
d
t
h
=
1
.
5
mm
,
x
-
p
e
r
i
o
d
i
c
i
t
y
=
2
2
.
5
mm
,
y
-
p
e
r
i
o
d
i
c
i
t
y
=
1
5
mm
P
a
t
c
h
L
e
n
g
t
h
=
7
mm
t
o
2
0
mm
w
i
t
h
st
e
p
1
mm
P
a
r
a
me
t
r
i
c
S
t
u
d
y
2
P
a
t
c
h
L
e
n
g
t
h
=
1
5
mm
,
x
-
p
e
r
i
o
d
i
c
i
t
y
=
2
2
.
5
mm
,
y
-
p
e
r
i
o
d
i
c
i
t
y
=
1
5
mm
P
a
t
c
h
W
i
d
t
h
=
0
.
1
mm
t
o
1
.
5
mm
w
i
t
h
s
t
e
p
0
.
1
mm
P
a
r
a
me
t
r
i
c
S
t
u
d
y
3
P
a
t
c
h
L
e
n
g
t
h
=
1
5
mm
,
P
a
t
c
h
W
i
d
t
h
=
1
.
5
mm
,
y
-
p
e
r
i
o
d
i
c
i
t
y
=
1
2
.
3
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
=
1
5
.
5
m
m t
o
2
2
.
5
mm
w
i
t
h
st
e
p
0
.
5
mm
P
a
r
a
me
t
r
i
c
S
t
u
d
y
4
P
a
t
c
h
L
e
n
g
t
h
=
1
5
mm
,
P
a
t
c
h
W
i
d
t
h
=
1
.
5
mm
,
x
-
p
e
r
i
o
d
i
c
i
t
y
=
2
2
.
5
mm
y
-
p
e
r
i
o
d
i
c
i
t
y
=
1
.
9
5
m
m t
o
1
5
mm
w
i
t
h
st
e
p
0
.
4
5
mm
Fig
u
r
e
5
s
h
o
ws
th
e
r
esu
lts
o
f
Par
am
etr
ic
Stu
d
y
1
wh
er
e
Patch
W
id
th
,
x
-
p
er
io
d
icity
,
y
-
p
er
io
d
icity
ar
e
k
ep
t
co
n
s
tan
t
an
d
Patch
L
en
g
th
is
v
ar
ied
.
Fig
u
r
e
6
s
h
o
ws
th
e
r
esu
lts
o
f
Par
am
etr
ic
Stu
d
y
2
wh
er
e
Patch
L
en
g
th
,
x
-
p
er
io
d
icity
,
y
-
p
er
i
o
d
icity
ar
e
k
ep
t
co
n
s
tan
t
an
d
Patch
W
id
th
i
s
v
ar
ied
.
I
n
Fig
u
r
e
7
th
e
r
esu
lts
o
f
Par
am
etr
ic
Stu
d
y
3
is
g
iv
e
n
wh
er
e
Patch
L
en
g
th
,
Patch
W
id
th
,
y
-
p
er
io
d
icity
ar
e
k
e
p
t
co
n
s
tan
t
an
d
x
-
p
er
io
d
icity
is
v
ar
ied
.
Fig
u
r
e
8
s
h
o
ws
th
e
r
esu
lts
o
f
Par
am
et
r
ic
Stu
d
y
4
w
h
er
e
Patch
L
en
g
th
,
Patch
W
id
th
,
x
-
p
er
io
d
icity
ar
e
k
ep
t c
o
n
s
tan
t a
n
d
y
-
p
er
io
d
icity
is
v
ar
ied
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
Desig
n
o
f freq
u
en
cy
s
elec
tive
s
u
r
fa
ce
co
mp
r
is
in
g
o
f d
ip
o
les
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
Mo
n
o
jit R
u
d
r
a
)
281
Fig
u
r
e
5.
R
esu
lts
o
f
p
ar
am
etr
ic
s
tu
d
y
1
: V
ar
iatio
n
o
f
p
atch
l
en
g
th
f
r
o
m
7
m
m
to
2
0
m
m
w
ith
s
tep
1
m
m
Fig
u
r
e
6.
R
esu
lts
o
f
p
ar
am
etr
ic
s
tu
d
y
2
: V
ar
iatio
n
o
f
p
atch
wid
th
f
r
o
m
0
.
1
m
m
to
1
.
5
m
m
with
s
tep
0
.
1
m
m
Fig
u
r
e
7.
R
esu
lts
o
f
p
ar
am
etr
ic
s
tu
d
y
3
: v
a
r
iatio
n
o
f
X
-
p
er
i
o
d
icity
f
r
o
m
1
5
.
5
m
m
t
o
2
2
.
5
m
m
with
s
tep
0
.
5
m
m
Fig
u
r
e
8.
R
esu
lts
o
f
p
ar
am
etr
ic
s
tu
d
y
4
: v
a
r
iatio
n
o
f
Y
-
p
er
i
o
d
icity
f
r
o
m
1
.
9
5
m
m
t
o
1
5
m
m
with
s
tep
0
.
4
5
m
m
4
.
3
.
P
er
f
o
rma
nce
co
m
pa
riso
n
o
f
t
he
pro
po
s
ed
t
ec
hn
iq
ue
u
s
i
ng
ANN
wit
h
t
ra
ditio
na
l
E
M
s
im
ula
t
io
n
s
o
f
t
wa
re
(
lik
e
Ans
o
f
t
Desig
ner)
I
n
tr
ad
itio
n
al
E
M
s
im
u
latio
n
s
o
f
twar
e,
an
aly
s
is
o
n
FS
S
is
d
o
n
e
u
s
in
g
E
M
n
u
m
er
ical
m
e
th
o
d
s
,
s
u
ch
as
th
e
Me
th
o
d
o
f
M
o
m
en
t
(
M
OM
)
.
T
h
ese
n
u
m
er
ical
m
eth
o
d
s
r
eq
u
ir
e
h
ig
h
co
m
p
u
tatio
n
al
co
s
t,
s
o
ev
e
n
with
a
s
y
s
tem
wi
th
h
ig
h
co
n
f
ig
u
r
atio
n
,
th
e
s
im
u
latio
n
tim
e
r
em
ain
s
v
er
y
lo
n
g
.
B
u
t
if
in
s
tead
A
NN
is
u
s
ed
,
th
en
it
d
o
esn
’
t r
eq
u
ir
e
a
n
y
co
m
p
lex
n
u
m
er
ical
m
eth
o
d
s
,
s
o
n
atu
r
all
y
th
e
co
m
p
u
tatio
n
tim
e
is
r
ed
u
ce
d
as c
o
m
p
ar
ed
to
tr
ad
itio
n
al
E
M
s
im
u
latio
n
s
o
f
twar
e.
Her
e,
f
ir
s
tly
a
n
FS
S
with
p
atch
-
len
g
th
=
1
5
m
m
,
p
atc
h
-
wid
th
=
1
.
5
m
m
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
: 2
2
5
2
-
8
8
1
4
I
n
t J Ad
v
Ap
p
l Sci
,
Vo
l.
9
,
No
.
4
,
Dec
em
b
e
r
2
0
2
0
:
2
7
6
–
2
8
3
282
x
-
p
er
io
d
icity
=
2
2
.
5
m
m
a
n
d
y
-
p
er
io
d
icity
=
1
5
m
m
is
tak
e
n
.
|
S2
1
|
v
s
.
Fre
q
u
en
cy
p
l
o
t
is
d
o
n
e
f
r
o
m
2
GHz
to
1
2
GHz
with
lo
w
s
tep
c
o
u
n
t
an
d
f
o
r
th
is
th
e
tim
e
elap
s
e
d
is
6
m
in
u
tes.
Fro
m
th
is
p
l
o
t
th
e
ap
p
r
o
x
im
ate
r
eso
n
an
t
f
r
e
q
u
en
c
y
is
lo
ca
ted
s
o
m
ewh
er
e
b
etwe
en
6
GHz
a
n
d
7
.
5
GHz
.
T
h
en
a
g
ain
p
lo
t
i
s
d
o
n
e
f
r
o
m
6
GHz
to
7
.
5
GHz
with
h
i
g
h
s
tep
co
u
n
t
wh
ich
t
o
o
k
2
m
in
u
tes
tim
e
an
d
th
e
r
eso
n
a
n
t
f
r
e
q
u
en
c
y
o
b
tain
ed
is
6
.
8
3
GHz
.
So
,
th
e
to
tal
s
im
u
latio
n
tim
e
r
eq
u
ir
ed
is
8
m
in
u
tes.
T
h
en
an
FS
S
with
p
atch
-
len
g
th
=
8
m
m
,
p
atc
h
-
wid
th
=
0
.
9
m
m
,
x
-
p
er
io
d
icity
=
2
2
.
5
m
m
an
d
y
-
p
er
io
d
icity
=
1
5
m
m
is
tak
en
.
|
S2
1
|
v
s
.
Fre
q
u
en
cy
p
lo
t
is
d
o
n
e
f
r
o
m
2
GHz
to
1
2
GHz
with
lo
w
s
tep
co
u
n
t
an
d
f
o
r
th
is
th
e
tim
e
elap
s
ed
is
4
m
in
u
tes.
Fro
m
th
is
p
lo
t
th
e
ap
p
r
o
x
im
ate
r
eso
n
an
t
f
r
eq
u
en
cy
is
lo
ca
ted
s
o
m
ewh
er
e
b
etwe
en
1
0
GHz
an
d
1
1
.
5
GHz
.
T
h
en
a
g
ain
p
l
o
t is d
o
n
e
f
r
o
m
1
0
GHz
to
1
1
.
5
GH
z
with
h
ig
h
s
tep
co
u
n
t
wh
ic
h
to
o
k
2
m
in
u
tes
ti
m
e
an
d
th
e
r
eso
n
an
t
f
r
eq
u
e
n
c
y
o
b
tain
e
d
is
1
0
.
7
6
GHz
.
So
,
t
o
tal
s
im
u
latio
n
tim
e
r
eq
u
ir
ed
is
6
m
in
u
tes.
B
u
t
in
ca
s
e
o
f
ANN,
g
en
er
atio
n
o
f
|
S2
1
|
v
s
.
Fre
q
u
en
cy
p
l
o
t
is
n
o
t
r
eq
u
ir
e
d
,
as
it
d
ir
ec
tly
g
iv
es
r
eso
n
an
t
f
r
eq
u
e
n
cy
as
o
u
tp
u
t.
I
n
f
ir
s
t
ca
s
e
th
e
r
eso
n
an
t
f
r
eq
u
en
cy
o
b
tain
e
d
i
s
6
.
8
3
4
8
GHz
,
a
n
d
in
th
e
s
ec
o
n
d
ca
s
e
th
e
r
eso
n
a
n
t
f
r
eq
u
e
n
cy
o
b
tain
ed
is
1
0
.
7
4
0
4
GHz
.
Her
e
in
b
o
th
th
e
c
ases
th
e
ANN
g
av
e
r
eso
n
an
t
f
r
eq
u
e
n
cy
o
u
tp
u
t
in
less
th
an
1
s
ec
tim
e
(
i.e
.
i
n
t
h
e
f
ir
s
t
ca
s
e
0
.
3
6
0
3
6
s
ec
an
d
in
th
e
s
ec
o
n
d
ca
s
e
0
.
2
7
7
0
1
s
ec
)
.
T
h
e
p
e
r
ce
n
tag
e
r
ed
u
ctio
n
i
n
co
m
p
u
tatio
n
al
ti
m
e
is
9
9
.
9
2
4
9
% in
th
e
f
ir
s
t c
ase
an
d
9
9
.
9
2
3
0
% in
th
e
s
ec
o
n
d
ca
s
e,
as sh
o
wn
in
T
ab
le
5
.
T
ab
le
5
.
Per
f
o
r
m
an
ce
co
m
p
a
r
is
o
n
o
f
th
e
p
r
o
p
o
s
ed
tech
n
iq
u
e
u
s
in
g
ANN
with
tr
ad
itio
n
al
E
M
s
im
u
latio
n
s
o
f
twar
e
P
a
r
a
me
t
e
r
s
S
i
mu
l
a
t
i
o
n
Ti
me
A
N
N
O
u
t
p
u
t
T
i
me
%
R
e
d
u
c
t
i
o
n
p
a
t
c
h
-
l
e
n
g
t
h
=
1
5
mm
p
a
t
c
h
-
w
i
d
t
h
=
1
.
5
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
=
2
2
.
5
m
m
y
-
p
e
r
i
o
d
i
c
i
t
y
=
1
5
mm
8
m
i
n
u
t
e
s
0
.
3
6
0
3
6
se
c
9
9
.
9
2
4
9
%
p
a
t
c
h
-
l
e
n
g
t
h
=
8
mm
p
a
t
c
h
-
w
i
d
t
h
=
0
.
9
mm
x
-
p
e
r
i
o
d
i
c
i
t
y
=
2
2
.
5
m
m
y
-
p
e
r
i
o
d
i
c
i
t
y
=
1
5
mm
6
m
i
n
u
t
e
s
0
.
2
7
7
0
1
se
c
9
9
.
9
2
3
0
%
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
a
n
Ar
tific
ial
Neu
r
al
Netwo
r
k
m
o
d
el
is
tr
ain
ed
u
s
in
g
d
ata
o
b
tain
ed
f
r
o
m
ANSOFT
Desig
n
er
v
2
s
o
f
twar
e
w
h
ich
u
s
es
Me
th
o
d
o
f
M
o
m
en
t
f
o
r
an
aly
s
is
o
f
FS
S.
Hen
ce
,
th
e
ANN
is
p
r
o
p
er
l
y
tr
ain
ed
an
d
g
iv
es
n
eg
lig
ib
le
e
r
r
o
r
.
Usi
n
g
th
is
ANN,
th
e
r
es
u
lts
ar
e
o
b
tain
e
d
v
er
y
q
u
ick
ly
wh
ich
s
av
es
tim
e
an
d
also
r
ed
u
ce
s
th
e
co
m
p
u
tat
io
n
al
co
s
t a
n
d
c
o
m
p
lex
ity
.
RE
F
E
R
E
NC
E
S
[1
]
Be
n
A.
M
u
n
k
,
“
F
re
q
u
e
n
c
y
S
e
lec
ti
v
e
S
u
rfa
c
e
s
–
T
h
e
o
r
y
a
n
d
De
sig
n
,
”
W
il
e
y
,
n
o
.
1
,
p
p
.
1
-
5
,
2
0
0
0
.
[2
]
Jia
n
x
u
n
S
u
a
n
d
Xia
o
we
n
Xu
,
“
M
OM
a
n
a
ly
sis
o
f
th
e
p
lan
a
r
a
n
d
c
u
rv
e
d
F
S
S
b
a
se
d
o
n
d
i
p
o
le
e
lem
e
n
ts
,
”
In
ter
n
a
t
io
n
a
l
Co
n
fer
e
n
c
e
o
n
M
icr
o
wa
v
e
T
e
c
h
n
o
l
o
g
y
a
n
d
Co
m
p
u
t
a
ti
o
n
a
l
E
lec
tro
ma
g
n
e
ti
c
s
(IC
M
T
CE
2
0
0
9
),
Beiji
n
g
,
Ch
i
n
a
,
p
p
.
1
2
7
-
1
3
0
,
2
0
0
9
.
[3
]
P
.
L.
d
a
S
il
v
a
a
n
d
A.G
.
D
’As
su
n
c
a
o
,
“
Ne
u
ro
m
o
d
e
ll
i
n
g
o
f
F
re
q
u
e
n
c
y
S
e
lec
ti
v
e
S
u
rfa
c
e
s
a
n
d
E
-
S
h
a
p
e
d
M
icro
stri
p
An
ten
n
a
s
,
”
T
h
e
2
0
0
6
I
EE
E
I
n
t
e
rn
a
ti
o
n
a
l
J
o
in
t
Co
n
fer
e
n
c
e
o
n
Ne
u
ra
l
Ne
two
rk
Pro
c
e
e
d
in
g
s,
Va
n
c
o
u
v
e
r,
BC,
Ca
n
a
d
a
,
p
p
.
2
3
7
4
-
2
3
7
7
,
2
0
0
6
.
[4
]
P
.
H.F
.
S
il
v
a
a
n
d
A.L
.
P
.
S
.
Ca
m
p
o
s,
“
F
a
st
a
n
d
a
c
c
u
ra
te
m
o
d
e
ll
i
n
g
o
f
fre
q
u
e
n
c
y
se
lec
ti
v
e
su
rfa
c
e
s
u
sin
g
a
n
e
w
m
o
d
u
lar
n
e
u
ra
l
n
e
two
r
k
c
o
n
fi
g
u
r
a
ti
o
n
o
f
m
u
lt
il
a
y
e
r
p
e
rc
e
p
tro
n
s
,
”
IET
M
icr
o
wa
v
e
s,
A
n
ten
n
a
s
&
Pro
p
a
g
a
ti
o
n
,
v
o
l.
2,
n
o
.
5
,
p
p
.
5
0
3
-
5
1
1
,
2
0
0
8
.
[5
]
P
.
H.
d
a
F
.
S
il
v
a
,
P
.
Lac
o
u
th
,
G
.
F
o
n
t
g
a
ll
a
n
d
,
A.
L.
P
.
S
.
Ca
m
p
o
s
a
n
d
A.G
.
D
’As
su
n
c
a
o
,
“
De
sig
n
o
f
fre
q
u
e
n
c
y
se
lec
ti
v
e
su
rfa
c
e
s
u
si
n
g
a
n
o
v
e
l
M
o
M
-
ANN
-
G
A
tec
h
n
iq
u
e
,
”
2
0
0
7
S
B
M
O/IEE
E
M
T
T
-
S
I
n
ter
n
a
t
io
n
a
l
M
icr
o
w
a
v
e
a
n
d
Op
t
o
e
lec
tro
n
ics
Co
n
fer
e
n
c
e
,
Bra
zil
,
p
p
.
2
7
5
-
2
7
9
,
2
0
0
7
.
[6
]
R.
M
.
S
.
Cr
u
z
,
P
.
H.
d
a
F
.
S
il
v
a
a
n
d
A.G
.
D
’As
su
n
c
a
o
,
“
S
y
n
th
e
sis
o
f
c
ro
ss
e
d
d
i
p
o
le
fre
q
u
e
n
c
y
se
l
e
c
ti
v
e
su
rfa
c
e
s
u
sin
g
g
e
n
e
ti
c
a
lg
o
rit
h
m
s
a
n
d
a
rti
f
icia
l
n
e
u
ra
l
n
e
t
wo
r
k
s
,
”
2
0
0
9
I
n
ter
n
a
ti
o
n
a
l
J
o
in
t
Co
n
fer
e
n
c
e
o
n
Ne
u
ra
l
Ne
tw
o
rk
s
,
At
la
n
ta
,
GA,
US
A
,
p
p
.
6
2
7
-
6
3
3
,
2
0
0
9
.
[7
]
P
.
H.
d
a
F
.
S
il
v
a
,
R
.
M
.
S
.
Cr
u
z
a
n
d
A.G
.
D
’As
su
n
ç
ã
o
,
“
Blen
d
i
n
g
P
S
O
a
n
d
AN
N
fo
r
Op
ti
m
a
l
De
sig
n
o
f
F
S
S
F
il
ters
Wi
th
K
o
c
h
Isla
n
d
P
a
tch
El
e
m
e
n
ts
,
”
IEE
E
T
r
a
n
sa
c
ti
o
n
s
o
n
M
a
g
n
e
ti
c
s
,
v
o
l.
4
6
,
n
o
.
8
,
p
p
.
3
0
1
0
-
3
0
1
3
,
2
0
1
0
.
[8
]
An
u
ra
d
h
a
,
A.
P
a
tn
a
i
k
,
S
.
N
S
in
h
a
a
n
d
J.R.
M
o
si
g
,
“
D
e
sig
n
o
f
c
u
sto
m
ize
d
fra
c
tal
F
S
S
,
”
Pr
o
c
e
e
d
in
g
s
o
f
th
e
2
0
1
2
IEE
E
In
ter
n
a
ti
o
n
a
l
S
y
mp
o
si
u
m o
n
An
ten
n
a
s a
n
d
Pro
p
a
g
a
ti
o
n
,
Ch
i
c
a
g
o
,
IL
,
US
A
,
2
0
1
2
.
[9
]
M
.
R.
d
a
S
il
v
a
,
C.
d
e
L.
Nó
b
re
g
a
,
P
.
H.
d
a
F
.
S
il
v
a
a
n
d
A.G
.
D
’A
ss
u
n
ç
ã
o
,
“
Op
t
ima
l
d
e
sig
n
o
f
fre
q
u
e
n
c
y
se
lec
ti
v
e
su
rfa
c
e
s with
fra
c
tal
m
o
ti
fs
,
”
IET
M
icr
o
wa
v
e
s,
A
n
ten
n
a
s &
Pro
p
a
g
a
ti
o
n
,
v
o
l
.
8,
no
9
,
p
p
.
6
2
7
-
6
3
1
,
2
0
1
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ad
v
Ap
p
l Sci
I
SS
N:
2
2
5
2
-
8
8
1
4
Desig
n
o
f freq
u
en
cy
s
elec
tive
s
u
r
fa
ce
co
mp
r
is
in
g
o
f d
ip
o
les
u
s
in
g
a
r
tifi
cia
l n
eu
r
a
l n
etw
o
r
k
(
Mo
n
o
jit R
u
d
r
a
)
283
[1
0
]
W.
C.
Ara
u
jo
,
A.
G
.
D
’
As
su
n
ç
ã
o
Ju
n
io
r
,
E.
E.
C.
Ol
i
v
e
ira an
d
A.
G
.
D
’As
su
n
ç
ã
o
,
“
F
S
S
d
e
si
g
n
s u
si
n
g
a
p
o
p
u
lati
o
n
-
b
a
se
d
h
y
b
ri
d
a
lg
o
rit
h
m
in
s
p
i
re
d
o
n
t
h
e
e
c
h
o
lo
c
a
ti
o
n
o
f
b
a
ts
,
”
2
0
1
5
9
th
Eu
r
o
p
e
a
n
C
o
n
fer
e
n
c
e
o
n
An
ten
n
a
s
a
n
d
Pro
p
a
g
a
ti
o
n
(E
u
CAP
),
L
isb
o
n
,
P
o
rtu
g
a
l
,
2
0
1
5
.
[1
1
]
M
.
C.
Alc
â
n
tara
Ne
to
,
F
.
J.
B.
B
a
rro
s,
J.
P
.
L.
Ara
ú
jo
,
H.
S
.
G
o
m
e
s,
G
.
P
.
S
.
Ca
v
a
lca
n
te
a
n
d
A.
G
.
D
’As
su
n
ç
ã
o
,
“
A
m
e
tah
e
u
risti
c
h
y
b
ri
d
o
p
ti
m
i
z
a
ti
o
n
tec
h
n
i
q
u
e
f
o
r
d
e
sig
n
i
n
g
b
ro
a
d
b
a
n
d
F
S
S
,
”
2
0
1
5
S
BM
O/IEE
E
M
T
T
-
S
In
ter
n
a
t
io
n
a
l
M
icr
o
w
a
v
e
a
n
d
Op
t
o
e
lec
tro
n
ics
Co
n
fer
e
n
c
e
(IM
OC),
Po
rto
d
e
Ga
li
n
h
a
s,
Bra
zil
,
2
0
1
5
.
[1
2
]
M
.
P
a
n
d
a
a
n
d
P
.
P
.
S
a
rk
a
r,
“
P
r
e
d
ictio
n
o
f
P
e
rio
d
icity
o
f
F
S
S
S
tru
c
tu
re
Us
i
n
g
P
a
rt
icle
S
wa
rm
Op
ti
m
iza
ti
o
n
,
”
I
-
ma
n
a
g
e
r’s
J
o
u
rn
a
l
o
n
El
e
c
tro
n
i
c
s E
n
g
i
n
e
e
rin
g
,
v
o
l.
7,
n
o
.
3
,
p
p
.
2
5
-
3
1
,
2
0
1
7
.
[1
3
]
N.
Li
u
,
X.
S
h
e
n
g
,
C.
Zh
a
n
g
,
J.
F
a
n
a
n
d
D.
G
u
o
,
“
De
sig
n
o
f
F
S
S
ra
d
o
m
e
u
sin
g
b
in
a
ry
p
a
rti
c
le
sw
a
rm
a
lg
o
rit
h
m
c
o
m
b
in
e
d
with
p
ix
e
l
-
o
v
e
rlap
tec
h
n
i
q
u
e
,
”
J
o
u
rn
a
l
o
f
El
e
c
tro
m
a
g
n
e
ti
c
W
a
v
e
s
a
n
d
A
p
p
li
c
a
ti
o
n
s
,
v
o
l.
31,
n
o
.
5
,
p
p
.
5
2
2
-
5
3
1
,
2
0
1
7
.
[1
4
]
M
.
P
a
n
d
a
a
n
d
P
.
P
.
S
a
r
k
a
r,
“
Hp
n
n
a
Ba
se
d
F
ss
De
sig
n
in
g
:
A
C
a
se
S
tu
d
y
,
”
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
S
c
ien
c
e
s a
n
d
En
g
i
n
e
e
rin
g
,
v
o
l.
6,
no.
5
,
p
p
.
7
9
2
-
7
9
6
,
2
0
1
8
.
[1
5
]
S
.
A
g
h
a
re
z
a
e
i
a
n
d
M
.
F
a
lam
a
rz
i,
“
P
a
rti
c
le
S
wa
rm
Op
ti
m
iza
ti
o
n
Al
g
o
rit
h
m
f
o
r
t
h
e
P
re
p
a
c
k
Op
ti
m
iza
ti
o
n
P
r
o
b
lem
,
”
Eco
n
o
mic
C
o
mp
u
ta
t
io
n
a
n
d
Eco
n
o
mic
Cy
b
e
rn
e
ti
c
s S
t
u
d
ie
s
a
n
d
Res
e
a
rc
h
,
v
o
l.
5
3
,
n
o
.
2
,
p
p
.
2
8
9
-
3
0
7
,
2
0
1
9
.
[1
6
]
M
.
C.
A.
Ne
t
o
,
J.
Ara
ú
jo
,
R.
J.
S
.
M
o
ta,
F
.
J.
B.
Ba
rro
s,
F
.
H.
C
.
S
.
F
e
rre
ira,
G
.
P
.
d
o
s
S
.
Ca
v
a
lca
n
te
a
n
d
B
.
S
.
L.
Ca
stro
,
“
De
sig
n
a
n
d
S
y
n
th
e
sis
o
f
a
n
Ultra
W
id
e
Ba
n
d
F
S
S
f
o
r
m
m
-
Wav
e
Ap
p
li
c
a
ti
o
n
v
ia
G
e
n
e
ra
l
Re
g
re
ss
io
n
Ne
u
ra
l
Ne
two
rk
a
n
d
M
u
lt
i
o
b
jec
ti
v
e
Ba
t
Al
g
o
ri
th
m
,
”
J
o
u
rn
a
l
o
f
M
i
c
ro
wa
v
e
s,
Op
t
o
e
lec
tro
n
ics
a
n
d
E
lec
tro
ma
g
n
e
ti
c
Ap
p
li
c
a
ti
o
n
s
,
v
o
l
.
1
8
,
n
o
.
4
,
p
p
.
5
3
0
-
5
4
4
,
2
0
1
9
.
B
I
O
G
RAP
H
I
E
S O
F
AUTH
O
RS
M
o
n
o
ji
t
Ru
d
ra
o
b
tai
n
e
d
h
is
B.
Te
c
h
d
e
g
re
e
in
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
fr
o
m
F
u
tu
re
In
stit
u
te
o
f
E
n
g
i
n
e
e
rin
g
a
n
d
M
a
n
a
g
e
m
e
n
t,
K
o
lk
a
ta
in
2
0
1
6
.
He
is
p
re
se
n
tl
y
p
u
rsu
i
n
g
M
.
Tec
h
in
C
o
m
m
u
n
ica
ti
o
n
E
n
g
in
e
e
rin
g
a
t
t
h
e
De
p
t.
o
f
En
g
in
e
e
rin
g
&
Tec
h
n
o
lo
g
ica
l
S
t
u
d
ies
i
n
Un
iv
e
rsity
o
f
Ka
ly
a
n
i,
Na
d
ia.
Hi
s
a
re
a
s
o
f
in
tere
st
a
re
F
re
q
u
e
n
c
y
S
e
lec
ti
v
e
S
u
rfa
c
e
s,
Artifi
c
ial
Ne
u
ra
l
Ne
two
rk
.
P
S
o
n
i
Re
d
d
y
h
a
s
re
c
e
iv
e
d
th
e
B.
Tec
h
.
d
e
g
re
e
i
n
El
e
c
tro
n
ics
&
Co
m
m
u
n
ica
ti
o
n
E
n
g
i
n
e
e
rin
g
fro
m
Ac
a
d
e
m
y
o
f
Tec
h
n
o
lo
g
y
,
Wes
t
Be
n
g
a
l,
I
n
d
ia,
i
n
2
0
1
1
u
n
d
e
r
th
e
a
ffil
iati
o
n
o
f
fo
rm
e
r
Wes
t
Be
n
g
a
l
Un
i
v
e
rsity
o
f
Tec
h
n
o
lo
g
y
a
n
d
t
h
e
M
.
Tec
h
.
d
e
g
re
e
in
E
lec
tro
n
ics
&
Co
m
m
u
n
ica
ti
o
n
En
g
i
n
e
e
rin
g
fro
m
Ka
ly
a
n
i
G
o
v
t.
En
g
in
e
e
rin
g
Co
l
leg
e
,
Ka
ly
a
n
i
,
Wes
t
Be
n
g
a
l,
I
n
d
ia,
i
n
2
0
1
6
u
n
d
e
r
t
h
e
a
ffil
iatio
n
o
f
M
a
u
lan
a
Ab
u
l
Ka
lam
Az
a
d
Un
iv
e
rsity
o
f
Tec
h
n
o
l
o
g
y
,
Wes
t
Be
n
g
a
l.
S
h
e
wo
rk
e
d
a
s
a
G
u
e
st
Lec
tu
re
r
(N
o
v
e
m
b
e
r
2
0
1
7
ti
ll
Au
g
u
st
2
0
1
8
)
in
t
h
e
De
p
a
rtme
n
t
o
f
En
g
in
e
e
rin
g
&
Tec
h
n
o
l
o
g
ica
l
S
tu
d
ies
,
U
n
iv
e
r
sity
o
f
Ka
ly
a
n
i,
Wes
t
Be
n
g
a
l,
In
d
ia.
S
i
n
c
e
Ap
ril
2
0
1
9
,
s
h
e
h
a
s
b
e
e
n
wo
rk
in
g
a
s
a
CS
IR
S
e
n
io
r
Re
se
a
rc
h
F
e
ll
o
w
(u
n
d
e
r
G
ra
n
t
0
9
/1
0
6
(
0
1
8
2
)/
2
0
1
9
-
EM
R
-
I)
i
n
th
e
De
p
a
rtme
n
t
o
f
En
g
in
e
e
rin
g
&
Tec
h
n
o
lo
g
ica
l
S
tu
d
ies
,
Un
i
v
e
rsity
o
f
Ka
ly
a
n
i
,
Wes
t
Be
n
g
a
l,
In
d
ia.
He
r
re
se
a
rc
h
a
re
a
s
in
c
lu
d
e
e
n
h
a
n
c
e
m
e
n
t
o
f
ra
d
iatio
n
c
h
a
r
a
c
teristics
o
f
m
icro
strip
p
a
tch
a
n
d
d
iele
c
tri
c
re
so
n
a
to
r
a
n
ten
n
a
s,
a
n
d
c
o
m
p
a
c
t
m
u
l
ti
b
a
n
d
DG
S
in
te
g
ra
ted
m
icro
stri
p
a
n
ten
n
a
.
Ra
jatsu
b
h
ra
C
h
a
k
ra
b
o
rty
is
a
fi
n
a
l
y
e
a
r
B.
Tec
h
st
u
d
e
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
E
n
g
i
n
e
e
rin
g
fro
m
F
u
tu
re
I
n
stit
u
te o
f
En
g
in
e
e
ri
n
g
a
n
d
M
a
n
a
g
e
m
e
n
t
,
Ko
l
k
a
ta.
Dr.
P
a
rth
a
P
ra
ti
m
S
a
rk
a
r
o
b
tain
e
d
h
is
P
h
.
D
in
E
n
g
in
e
e
rin
g
fro
m
J
a
d
a
v
p
u
r
U
n
iv
e
rsit
y
in
t
h
e
y
e
a
r
2
0
0
2
.
He
h
a
s
o
b
tain
e
d
h
is
M
.
E.
f
ro
m
Ja
d
a
v
p
u
r
Un
i
v
e
rsity
in
th
e
y
e
a
r
1
9
9
4
.
He
e
a
rn
e
d
h
is
B.
E.
d
e
g
re
e
in
El
e
c
tro
n
ics
a
n
d
Tele
c
o
m
m
u
n
ica
t
io
n
En
g
i
n
e
e
rin
g
fro
m
Be
n
g
a
l
E
n
g
i
n
e
e
rin
g
C
o
ll
e
g
e
in
th
e
y
e
a
r
1
9
9
1
.
He
is p
re
se
n
tl
y
wo
rk
in
g
in
t
h
e
ra
n
k
o
f
P
r
o
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
o
f
En
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
lo
g
ica
l
S
tu
d
ies
,
U
n
i
v
e
rsity
o
f
Ka
ly
a
n
i
,
si
n
c
e
Ja
n
u
a
r
y
2
0
0
5
.
His
a
re
a
o
f
re
se
a
rc
h
in
c
lu
d
e
s
M
icro
str
i
p
An
te
n
n
a
,
F
re
q
u
e
n
c
y
S
e
lec
ti
v
e
S
u
rfa
c
e
s,
a
n
d
Artifi
c
ial
Ne
u
ra
l
Ne
two
rk
.
He
h
a
s
c
o
n
tri
b
u
ted
t
o
n
u
m
e
ro
u
s
(m
o
re
th
a
n
2
4
5
p
u
b
li
c
a
ti
o
n
s)
re
se
a
r
c
h
a
rti
c
les
in
v
a
rio
u
s
j
o
u
r
n
a
ls
a
n
d
c
o
n
fe
re
n
c
e
s o
f
re
p
u
te.
He
is a
li
fe
fe
ll
o
w
o
f
IET
E
a
n
d
fe
ll
o
w
o
f
IE
(In
d
ia).
Evaluation Warning : The document was created with Spire.PDF for Python.