I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
p
u
t
er
Science
Vo
l.
9
,
No
.
1
,
J
an
u
ar
y
201
8
,
p
p
.
89
~
92
I
SS
N:
2502
-
4752
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ee
cs
.
v
9
.
i1
.
p
p
89
-
92
89
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e.
co
m/jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
ijeec
s
Audio
Denois
ing
Ba
sed o
n Sho
rt
T
i
m
e
F
o
urie
r Tra
n
sfo
r
m
J
.
S.
Ash
w
in
1
,
N.
M
a
no
ha
ra
n
2
1
Re
se
a
rc
h
S
c
h
o
lar,
De
p
a
rtm
e
n
t
o
f
El
e
c
tri
c
a
l
a
n
d
El
e
c
tr
o
n
ics
E
n
g
in
e
e
rin
g
(M
a
rin
e
),
A
M
ET
Un
iv
e
rsity
,
Ch
e
n
n
a
i
2
Re
c
to
r,
A
M
ET
Un
iv
e
rsit
y
,
Ch
e
n
n
a
i
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Ju
l 8
,
2
0
1
7
R
ev
i
s
ed
No
v
2
0
,
2
0
1
7
A
cc
ep
ted
Dec
1
1
,
2
0
1
7
T
h
is
p
a
p
e
r
p
re
se
n
ts
a
n
o
v
e
l
a
u
d
io
d
e
-
n
o
isin
g
sc
h
e
m
e
in
a
g
iv
e
n
sp
e
e
c
h
sig
n
a
l.
T
h
e
re
c
o
v
e
r
y
o
f
o
rig
in
a
l
f
ro
m
th
e
c
o
m
m
u
n
ica
ti
o
n
c
h
a
n
n
e
l
w
it
h
o
u
t
a
n
y
n
o
ise
is
a
d
iff
icu
lt
tas
k
.
M
a
n
y
d
e
-
n
o
isin
g
tec
h
n
iq
u
e
s
h
a
v
e
b
e
e
n
p
r
o
p
o
se
d
f
o
r
th
e
re
m
o
v
a
l
o
f
n
o
ise
s
f
ro
m
a
d
ig
it
a
l
sig
n
a
l.
In
th
is
p
a
p
e
r,
a
n
a
u
d
io
d
e
-
n
o
isi
n
g
tec
h
n
iq
u
e
b
a
se
d
o
n
S
h
o
rt
T
im
e
F
o
u
rier
T
ra
n
sf
o
r
m
(S
T
F
T
)
is
im
p
le
m
e
n
ted
.
T
h
e
p
ro
p
o
se
d
a
rc
h
it
e
c
tu
re
u
se
s
a
n
o
v
e
l
a
p
p
ro
a
c
h
to
e
stim
a
te
e
n
v
iro
n
m
e
n
tal
n
o
ise
f
ro
m
sp
e
e
c
h
a
d
a
p
ti
v
e
ly
.
He
re
o
rig
in
a
l
sp
e
e
c
h
sig
n
a
ls
a
re
g
iv
e
n
a
s
in
p
u
t
sig
n
a
l.
Us
in
g
A
WG
N,
n
o
ise
s
a
re
a
d
d
e
d
to
t
h
e
sig
n
a
l.
T
h
e
n
n
o
ise
d
sig
n
a
ls
a
re
d
e
-
n
o
ise
d
u
si
n
g
S
T
F
T
tec
h
n
iq
u
e
s.
F
i
n
a
ll
y
S
ig
n
a
l
to
No
ise
Ra
ti
o
(S
NR),
P
e
a
k
S
ig
n
a
l
to
No
i
se
Ra
ti
o
(P
S
NR)
v
a
lu
e
s
f
o
r
n
o
ise
d
a
n
d
d
e
-
n
o
ise
d
sig
n
a
ls are
o
b
tai
n
e
d
.
K
ey
w
o
r
d
s
:
Au
d
io
d
e
-
n
o
i
s
in
g
A
W
GN
P
SNR
SNR
ST
FT
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
J
.
S.
A
s
h
w
i
n
,
R
esear
ch
Sch
o
lar
,
Dep
ar
t
m
e
n
t
o
f
E
lectr
ical
an
d
E
lectr
o
n
ics
E
n
g
i
n
ee
r
i
n
g
(
Ma
r
i
n
e)
,
A
ME
T
Un
i
v
er
s
it
y
,
C
h
en
n
ai
1.
I
NT
RO
D
UCT
I
O
N
Gen
er
all
y
i
n
t
h
e
s
i
g
n
al
p
r
o
ce
s
s
in
g
ap
p
licatio
n
s
th
e
n
o
i
s
e
d
is
tr
ib
u
tio
n
s
ee
m
s
to
b
e
a
m
aj
o
r
p
r
o
b
lem
.
T
h
e
n
o
n
ess
e
n
tial
s
i
g
n
a
ls
ar
e
s
u
p
er
i
m
p
o
s
ed
o
v
er
an
u
n
d
is
t
u
r
b
ed
s
ig
n
al.
W
h
en
t
h
e
r
eg
ar
it
y
o
f
t
h
e
n
o
is
e
r
ed
u
ce
s
th
en
t
h
e
d
en
o
i
s
in
g
m
et
h
o
d
s
g
e
t
m
o
r
e
d
if
f
icu
l
t.
2.
B
ACK
G
RO
UND
Sp
ee
ch
d
e
-
r
ev
er
b
er
atio
n
a
n
d
d
e
-
n
o
is
i
n
g
u
s
i
n
g
lear
n
i
n
g
s
p
ec
tr
al
m
ap
p
in
g
is
p
r
ese
n
ted
in
[
1
]
.
Sp
ec
tr
al
m
ap
p
in
g
is
lear
n
ed
d
ir
ec
tl
y
u
s
in
g
t
h
e
tr
ai
n
in
g
o
f
d
ee
p
n
eu
r
al
n
et
w
o
r
k
s
f
r
o
m
t
h
e
m
ag
n
it
u
d
e
s
p
ec
tr
o
g
r
a
m
o
f
co
n
ta
m
i
n
ated
s
p
ee
c
h
to
th
at
o
f
h
y
g
ie
n
ic
s
p
ee
ch
.
T
h
is
ap
p
r
o
ac
h
co
n
s
id
er
ab
ly
a
tten
u
ate
s
t
h
e
t
w
i
s
t
ca
u
s
ed
b
y
r
ev
er
b
er
atio
n
,
as
w
el
l a
s
b
ac
k
g
r
o
u
n
d
n
o
is
e.
Sp
ec
tr
al
s
ca
r
cit
y
b
ased
m
u
lti
ch
an
n
el
a
u
d
io
d
e
-
n
o
is
i
n
g
f
o
r
m
u
latio
n
i
s
d
escr
ib
ed
in
[
2
]
.
T
w
o
s
ta
g
e
m
et
h
o
d
s
ar
e
u
s
ed
f
o
r
th
is
e
v
al
u
atio
n
p
r
o
b
le
m
.
I
t
d
o
es
n
o
t
n
e
ed
an
y
d
etail
s
ab
o
u
t
n
o
i
s
e.
T
h
er
e
ar
e
t
w
o
s
ta
g
es
ar
e
in
v
o
l
v
ed
:
f
ir
s
t
s
tag
e
is
u
s
e
d
to
o
b
tain
lin
ea
r
co
m
b
i
n
atio
n
u
s
i
n
g
t
h
is
a
s
s
u
m
p
tio
n
a
n
d
s
ec
o
n
d
s
tag
e
esti
m
ate
s
th
e
n
u
m
b
er
o
f
r
e
m
ain
in
g
n
o
is
e.
Gr
ee
d
y
t
i
m
e
-
f
r
eq
u
e
n
c
y
s
h
r
i
n
k
a
g
e
b
ased
s
p
ar
s
e
a
u
d
io
d
e
-
n
o
is
i
n
g
is
d
is
c
u
s
s
ed
in
[
3
]
.
Ma
tch
in
g
p
u
r
s
u
it
i
n
th
e
b
a
ck
g
r
o
u
n
d
o
f
au
d
io
d
e
-
n
o
i
s
in
g
is
an
al
y
s
ed
.
Facto
r
s
cr
iti
ca
l
to
its
s
u
cc
ess
is
id
en
ti
f
ied
u
s
i
n
g
in
ter
p
r
etin
g
t
h
e
alg
o
r
ith
m
li
k
e
ea
s
y
s
h
r
in
k
ag
e
ap
p
r
o
ac
h
.
Sp
ec
tr
al
s
u
b
tr
ac
tio
n
tec
h
n
iq
u
e
b
ased
au
d
io
d
e
-
n
o
is
in
g
is
e
x
p
lain
ed
in
[
4
]
.
A
n
e
f
f
icien
t
ar
ch
itect
u
r
e
in
h
ar
d
w
ar
e
f
o
r
th
e
al
g
o
r
it
h
m
o
f
s
p
ec
tr
al
s
u
b
tr
ac
tio
n
is
ap
p
lied
to
s
p
ee
ch
i
m
p
r
o
v
e
m
e
n
t
is
u
s
ed
.
E
n
v
ir
o
n
m
e
n
tal
n
o
is
e
is
est
i
m
ated
f
r
o
m
s
p
ee
ch
ad
ap
tiv
el
y
.
No
is
e
s
a
m
p
le
s
ar
e
s
u
b
tr
ac
ted
in
th
e
i
n
p
u
t
s
p
ee
c
h
af
ter
t
h
e
n
o
i
s
e
est
i
m
a
tio
n
.
T
h
er
e
ar
e
t
w
o
p
r
in
cip
al
b
lo
ck
s
ar
e
f
o
llo
w
ed
li
k
e
p
h
ase
b
lo
ck
a
n
d
n
o
is
e
est
i
m
a
tio
n
-
s
u
b
tr
ac
tio
n
b
lo
ck
w
h
ic
h
ar
e
ex
ec
u
ted
s
i
m
u
lta
n
eo
u
s
l
y
ex
p
lo
itin
g
th
e
p
ar
allel
lo
g
i
c
b
lo
ck
s
o
f
f
ield
p
r
o
g
r
am
m
ab
le
g
a
te
ar
r
a
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2502
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l
.
9
,
No
.
1
,
J
an
u
ar
y
201
8
:
89
–
92
90
De
-
n
o
is
i
n
g
ap
p
r
o
ac
h
b
ased
o
n
s
tatis
t
ical
E
m
p
ir
ical
Mo
d
e
De
co
m
p
o
s
i
tio
n
(
E
MD
)
f
o
r
m
u
l
ti
r
ate
h
ig
h
-
r
eso
lu
tio
n
s
i
g
n
al
r
ec
o
n
s
tr
u
c
tio
n
is
p
r
ese
n
ted
in
[
5
]
.
De
-
n
o
is
i
n
g
p
r
o
ce
d
u
r
e
is
ap
p
lied
b
ased
o
n
E
MD
in
ter
v
al
-
th
r
es
h
o
ld
to
ev
er
y
n
o
is
y
lo
w
r
eso
lu
tio
n
m
ea
s
u
r
e
m
e
n
t.
T
h
en
o
n
l
y
ca
n
f
i
lter
th
e
A
W
GN.
ST
FT
an
d
w
a
v
elet
d
e
-
n
o
is
i
n
g
b
ased
L
in
ea
r
Fre
q
u
en
c
y
Mo
d
u
lated
(
L
FM)
Sig
n
al
s
Dete
c
t
io
n
i
n
lo
w
SNR
i
s
e
x
p
lain
ed
i
n
[
6
]
.
I
n
p
u
t
s
i
g
n
als
ar
e
s
h
o
r
t
-
ti
m
e
F
o
u
r
ier
tr
an
s
f
o
r
m
ed
i
n
to
co
h
e
r
en
t
i
n
teg
r
atio
n
o
f
f
r
eq
u
en
c
y
-
s
h
i
f
t
m
o
d
el
s
eq
u
en
ce
s
w
i
th
co
m
p
o
u
n
d
en
v
elo
p
es
,
ac
h
iev
i
n
g
a
ti
m
e
-
f
r
eq
u
en
c
y
c
u
r
v
e
w
h
ich
i
s
to
b
e
p
r
o
c
ess
ed
b
y
w
a
v
elet,
i
n
o
r
d
er
to
w
ea
k
e
n
th
e
n
o
i
s
e
a
n
d
to
d
etec
t
L
FM
s
i
g
n
a
ls
ef
f
ec
ti
v
el
y
.
T
h
e
i
m
p
u
ls
e
n
o
is
e
r
ed
u
ctio
n
i
s
p
r
esen
ted
a
n
e
f
f
ic
ien
t
ap
p
r
o
ac
h
f
o
r
th
e
r
e
m
o
v
al
o
f
b
ip
o
lar
i
m
p
u
l
s
e
n
o
is
e
u
s
i
n
g
m
ed
ia
n
f
ilter
[
7
]
.
T
h
e
d
en
o
is
in
g
p
r
o
ce
s
s
ca
n
al
s
o
b
e
d
o
n
e
f
o
r
d
en
o
is
i
n
g
t
h
e
U
ltra
s
o
n
ic
E
c
h
o
s
i
g
n
als
a
s
i
n
[
9
]
.
E
m
p
ir
ical
m
o
d
e
d
ec
o
m
p
o
s
itio
n
b
ased
d
en
o
is
in
g
m
et
h
o
d
f
o
r
h
ea
r
t
s
o
u
n
d
s
i
g
n
al
an
d
its
p
er
f
o
r
m
an
ce
a
n
al
y
s
is
is
d
is
c
u
s
s
ed
in
[
1
0
]
.
A
ls
o
th
e
d
en
o
is
i
n
g
p
r
o
ce
s
s
ca
n
al
s
o
b
e
u
s
ed
f
o
r
ap
p
licatio
n
s
li
k
e
S
AR
I
m
a
g
e
De
n
o
is
i
n
g
a
s
in
[
1
1
]
.
3.
P
RO
B
L
E
M
S
Usu
al
l
y
w
h
en
t
h
e
s
i
g
n
als
ar
e
tr
an
s
m
itted
o
v
er
a
d
is
tan
ce
b
y
an
y
m
ea
n
s
th
er
e
m
a
y
b
e
s
o
m
e
p
r
o
b
lem
s
o
cc
u
r
r
ed
d
u
e
to
th
e
n
o
i
s
es
th
at
ar
e
af
f
ec
ted
b
y
t
h
e
m
ea
n
s
o
f
t
h
e
e
n
v
ir
o
n
m
e
n
t.
So
d
u
e
t
o
th
is
t
h
er
e
m
a
y
b
e
n
o
is
es
ad
d
ed
to
th
e
s
i
g
n
al
w
h
i
ch
a
f
f
ec
t
s
t
h
e
i
n
f
o
r
m
at
io
n
t
h
at
ar
e
s
to
r
ed
in
it.
So
in
o
r
d
er
to
less
e
n
t
h
e
n
o
is
e
i
n
an
ef
f
icie
n
t
w
a
y
o
u
r
p
r
o
p
o
s
ed
m
et
h
o
d
s
is
in
tr
o
d
u
ce
d
.
4.
P
RO
P
O
SE
D
SO
L
UT
I
O
N
T
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
u
s
es
th
e
ST
FT
b
lo
ck
th
r
es
h
o
ld
m
et
h
o
d
.
I
t
is
u
s
ed
to
d
e
-
n
o
i
s
i
n
g
th
e
a
u
d
io
s
ig
n
al
e
f
f
ec
t
iv
el
y
.
First
t
h
e
g
i
v
en
i
n
p
u
t
s
ig
n
al
s
ar
e
r
ea
d
th
en
A
W
GN
n
o
is
e
i
s
u
s
ed
to
ad
d
s
o
m
e
n
o
is
e
to
th
e
in
p
u
t
a
u
d
io
s
i
g
n
a
ls
.
T
h
e
r
ed
u
ctio
n
o
f
n
o
is
e
i
n
t
h
e
d
iesel
e
n
g
in
e
s
i
s
ex
p
lai
n
ed
in
th
e
E
x
h
au
s
t
n
o
is
e
r
ed
u
ct
io
n
tech
n
iq
u
es
i
n
d
ir
ec
t
in
j
ec
tio
n
(
D.
I
.
)
d
iesel
en
g
in
e
s
[
8
]
.
T
h
en
th
e
n
o
i
s
ed
s
ig
n
al
s
ar
e
d
e
-
n
o
i
s
ed
u
s
in
g
t
h
is
n
o
v
el
ST
FT
tech
n
iq
u
e.
F
in
all
y
S
N
R
an
d
P
SNR
o
f
th
e
o
r
ig
in
al
a
n
d
d
e
-
n
o
i
s
ed
s
ig
n
al
s
v
al
u
e
ar
e
ca
lcu
lated
.
Fi
g
u
r
e
1
s
h
o
w
s
t
h
e
b
lo
ck
d
iag
r
a
m
o
f
t
h
e
p
r
o
p
o
s
ed
au
d
io
d
e
-
n
o
is
in
g
te
ch
n
iq
u
e.
5.
M
E
T
H
O
DO
L
O
G
Y
5
.
1
.
Sh
o
rt
T
i
m
e
F
o
urier
T
ra
ns
f
o
r
m
T
h
e
ST
F
T
is
also
ca
lled
as
s
h
o
r
t
-
ter
m
Fo
u
r
ier
tr
an
s
f
o
r
m
b
ec
au
s
e
it
i
s
a
Fo
u
r
ier
-
r
elate
d
tr
an
s
f
o
r
m
to
d
eter
m
in
e
th
e
s
i
n
u
s
o
id
al
f
r
eq
u
en
c
y
a
n
d
p
h
ase
co
n
te
n
t
o
f
l
o
ca
l
s
ec
tio
n
s
o
f
a
s
i
g
n
al.
L
o
n
g
er
t
i
m
e
s
i
g
n
a
l
is
d
iv
id
ed
in
to
s
h
o
r
te
r
s
e
g
m
e
n
t
s
w
h
i
le
co
m
p
u
ti
n
g
ST
FT
in
p
r
ac
tice
an
d
th
e
n
ca
lcu
la
tes
th
e
Fo
u
r
ier
tr
a
n
s
f
o
r
m
in
d
iv
id
u
all
y
o
n
ev
er
y
s
h
o
r
ter
s
eg
m
e
n
t.
Fig
u
r
e
1
.
B
lo
ck
Diag
r
a
m
o
f
t
h
e
P
r
o
p
o
s
ed
A
u
d
io
De
-
No
is
in
g
S
y
s
te
m
Au
d
io
I
n
p
u
t
Sig
n
al
A
W
GN
No
is
e
Den
o
is
ed
Si
g
n
a
l
B
lo
ck
T
h
r
esh
o
ld
in
g
u
s
i
n
g
ST
FT
C
o
m
p
u
te
SN
R
&
P
SNR
Au
d
io
Ou
tp
u
t
Sig
n
al
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4752
A
u
d
io
Den
o
is
in
g
B
a
s
ed
o
n
S
h
o
r
t Time
F
o
u
r
ier Tr
a
n
s
fo
r
m
(
J
.
S
.
A
s
h
w
in
)
91
T
h
is
r
ev
ea
ls
th
e
Fo
u
r
ier
s
p
ec
tr
u
m
o
n
e
v
er
y
s
h
o
r
ter
s
eg
m
e
n
t.
On
e
t
h
en
u
s
u
all
y
p
lo
ts
t
h
e
v
ar
y
i
n
g
s
p
ec
tr
a
as a
ti
m
e
f
u
n
ctio
n
.
Fig
u
r
e
2
s
h
o
w
s
t
h
e
ex
a
m
p
le
o
f
S
T
F
T
.
Fig
u
r
e
2
.
ST
F
T
Diag
r
a
m
6.
RE
SU
L
T
S & D
I
SCU
SS
I
O
N
S
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
p
r
o
p
o
s
ed
s
y
s
te
m
f
o
r
d
ig
ital
a
u
d
io
d
e
-
n
o
i
s
in
g
u
s
i
n
g
ST
FT
m
et
h
o
d
h
as
d
o
n
e.
T
h
e
p
r
o
p
o
s
ed
ar
ch
itectu
r
e
u
s
es
a
n
o
v
el
ap
p
r
o
ac
h
to
es
ti
m
ate
en
v
ir
o
n
m
en
tal
n
o
is
e
f
r
o
m
s
p
ee
ch
ad
ap
ti
v
el
y
.
Her
e
o
r
ig
in
al
s
p
ee
ch
s
ig
n
al
s
ar
e
g
iv
e
n
as
in
p
u
t
s
i
g
n
al.
Usi
n
g
A
W
GN,
n
o
is
es
ar
e
ad
d
ed
to
th
e
s
ig
n
al.
T
h
e
n
n
o
is
ed
s
i
g
n
als
ar
e
d
e
-
n
o
is
ed
u
s
i
n
g
ST
FT
tech
n
iq
u
e
s
.
Fi
n
al
l
y
SN
R
,
P
SNR
v
al
u
es
f
o
r
n
o
is
ed
an
d
d
e
-
n
o
i
s
ed
s
ig
n
al
s
ar
e
o
b
tain
ed
.
Fig
u
r
e
3
.
No
is
y
Si
g
n
al
u
s
in
g
A
W
GN
Fig
u
r
e
4
.
De
-
No
is
ed
A
u
d
io
Si
g
n
al
u
s
in
g
ST
FT
Fig
u
r
e
2
s
h
o
w
s
t
h
e
n
o
is
ed
s
ig
n
a
l
w
h
ic
h
w
e
u
s
ed
.
Her
e
t
h
e
n
o
is
es
w
er
e
ad
d
ed
u
s
in
g
ad
d
w
h
it
e
Gau
s
s
ia
n
n
o
is
e
m
eth
o
d
w
ith
SNR
r
ate.
Fig
u
r
e
3
s
h
o
w
s
th
e
d
e
-
n
o
is
ed
o
u
tp
u
t
a
u
d
io
s
ig
n
a
l.
T
h
e
d
e
-
n
o
is
i
n
g
is
b
ased
o
n
ST
F
T
w
i
th
b
lo
ck
t
h
r
esh
o
ld
m
et
h
o
d
an
d
h
a
n
n
in
g
w
i
n
d
o
w
.
7.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
ST
FT
h
as
b
ee
n
p
r
o
p
o
s
ed
to
d
e
-
n
o
is
e
a
n
a
u
d
io
s
ig
n
al
f
r
o
m
t
h
e
g
i
v
en
in
p
u
t
s
i
g
n
al.
Fig
u
r
e
4
s
h
o
w
s
t
h
e
d
e
-
n
o
is
ed
au
d
io
s
ig
n
al
u
s
i
n
g
ST
FT
m
e
th
o
d
.
Fin
all
y
P
SNR
o
f
t
h
i
s
p
r
o
p
o
s
ed
m
eth
o
d
is
b
etter
w
h
i
le
co
m
p
ar
ed
to
ex
is
ti
n
g
m
et
h
o
d
.
T
h
e
r
esu
lts
i
n
d
icate
th
at
t
h
is
p
er
f
o
r
m
s
b
etter
th
an
t
h
e
o
th
er
d
is
tr
ib
u
tio
n
s
.
T
h
e
f
u
t
u
r
e
s
co
p
e
o
f
th
i
s
p
ap
er
is
d
esig
n
i
n
g
a
b
etter
alg
o
r
ith
m
w
it
h
b
etter
f
ea
t
u
r
es.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2502
-
4752
I
n
d
o
n
esia
n
J
E
lec
E
n
g
&
C
o
m
p
Sci,
Vo
l
.
9
,
No
.
1
,
J
an
u
ar
y
201
8
:
89
–
92
92
RE
F
E
R
E
NC
E
S
[1
]
Ha
n
K,
W
a
n
g
Y,
W
a
n
g
D,
W
o
o
d
s
W
S
,
M
e
rk
s
I,
Zh
a
n
g
T
.
L
e
a
rn
in
g
sp
e
c
tral
m
a
p
p
in
g
f
o
r
sp
e
e
c
h
d
e
re
v
e
rb
e
ra
ti
o
n
a
n
d
d
e
n
o
isi
n
g
.
IEE
E
T
ra
n
sa
c
ti
o
n
s o
n
A
u
d
io
,
S
p
e
e
c
h
,
a
n
d
L
a
n
g
u
a
g
e
Pro
c
e
ss
in
g
.
2
0
1
5
;
2
3
(6
)
:
9
8
2
-
9
9
2
.
[2
]
Ba
y
ra
m
I
.
A
m
u
lt
ich
a
n
n
e
l
a
u
d
i
o
d
e
n
o
isi
n
g
f
o
r
m
u
latio
n
b
a
se
d
o
n
sp
e
c
tral
sp
a
rsit
y
.
IEE
E/
ACM
T
r
a
n
sa
c
ti
o
n
s
o
n
Au
d
i
o
,
S
p
e
e
c
h
a
n
d
L
a
n
g
u
a
g
e
Pr
o
c
e
ss
in
g
.
2
0
1
5
;
2
3
(
1
2
)
:
2
2
7
2
-
2
2
8
5
.
[3
]
Bh
a
tt
a
c
h
a
ry
a
G
,
De
p
a
ll
e
P
.
S
p
a
rs
e
d
e
n
o
isin
g
o
f
a
u
d
io
b
y
g
re
e
d
y
ti
me
-
fre
q
u
e
n
c
y
sh
rin
k
a
g
e
.
IE
E
E
In
tern
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
A
c
o
u
stics
,
S
p
e
e
c
h
a
n
d
S
ig
n
a
l
P
r
o
c
e
ss
in
g
.
2
0
1
4
:
2
8
9
8
-
2
9
0
2
.
[4
]
Biswa
s
T
,
P
a
l
C,
M
a
n
d
a
l
S
B,
C
h
a
k
ra
b
a
rt
i
A
.
Au
d
io
d
e
-
n
o
isi
n
g
b
y
sp
e
c
tra
l
su
b
tra
c
ti
o
n
tec
h
n
i
q
u
e
imp
lem
e
n
ted
o
n
re
c
o
n
fi
g
u
ra
b
le h
a
r
d
wa
re
.
IE
EE
S
e
v
e
n
th
In
tern
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Co
n
tem
p
o
ra
r
y
Co
m
p
u
ti
n
g
.
2
0
1
4
:
2
3
6
-
2
4
1
.
[5
]
Uk
te
A
,
Kiz
il
k
a
y
a
A
,
El
b
i
M
D
.
S
ta
ti
stica
l
mu
l
ti
ra
te h
ig
h
-
re
so
lu
ti
o
n
sig
n
a
l
re
c
o
n
stru
c
ti
o
n
u
si
n
g
th
e
e
mp
irica
l
mo
d
e
d
e
c
o
mp
o
sit
io
n
b
a
se
d
d
e
n
o
isi
n
g
a
p
p
ro
a
c
h
.
IEE
E
I
n
tern
a
ti
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
A
p
p
li
e
d
El
e
c
tro
n
ics
.
2
0
1
4
:
3
0
3
-
3
0
6
.
[6
]
Yu
D,
Jin
z
h
e
n
W
,
S
h
a
o
y
in
g
S
,
Zen
g
p
in
g
C.
De
tec
ti
o
n
o
f
L
FM
sig
n
a
ls
in
l
o
w
S
N
R
b
a
se
d
o
n
S
T
FT
a
n
d
w
a
v
e
let
d
e
n
o
isi
n
g
.
IEE
E
In
ter
n
a
ti
o
n
a
l
Co
n
f
e
re
n
c
e
o
n
A
u
d
io
,
L
a
n
g
u
a
g
e
a
n
d
Im
a
g
e
P
ro
c
e
ss
in
g
.
2
0
1
4
:
9
2
1
-
9
2
5
.
[7
]
Ka
d
a
li
KS,
Ra
jaji
L
.
A
n
e
ff
ici
e
n
t
a
p
p
ro
a
c
h
f
o
r
th
e
re
m
o
v
a
l
o
f
b
i
p
o
lar
im
p
u
lse
n
o
ise
u
sin
g
m
e
d
ia
n
f
il
ter.
In
d
i
a
n
J
o
u
rn
a
l
o
f
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
.
2
0
1
5
;
8
(1
3
).
[8
]
S
u
n
d
a
ra
Ra
m
a
n
R,
S
a
n
k
a
ra
N
a
ra
y
a
n
a
n
,
G
,
M
a
n
o
h
a
ra
n
N.
Ex
h
a
u
st
n
o
ise
re
d
u
c
ti
o
n
tec
h
n
i
q
u
e
s
in
d
irec
t
in
jec
ti
o
n
(D.I.
)
d
ies
e
l
e
n
g
in
e
s,
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
A
p
p
li
e
d
E
n
g
in
e
e
rin
g
Res
e
a
rc
h
.
2
0
1
4
;
9
(
1
8
)
:
3
9
4
9
-
3
9
5
4
.
[9
]
M
ish
ra
S
.
Ca
sc
a
d
e
c
o
m
b
in
a
ti
o
n
o
f
w
a
v
e
let
a
n
d
a
d
a
p
t
iv
e
f
il
ter
f
o
r
n
o
ise
c
a
n
c
e
ll
a
ti
o
n
.
I
n
ter
n
a
ti
o
n
a
l
jo
u
rn
a
l
o
f
a
d
v
a
n
c
e
s in
si
g
n
a
l
a
n
d
ima
g
e
sc
ien
c
e
s.
2
0
1
6
;
2
(
2
):
2
1
-
2
6
.
[1
0
]
M
o
h
a
m
m
a
d
i
M
HD
.
Im
p
ro
v
e
d
De
n
o
isi
n
g
M
e
th
o
d
f
o
r
Ultras
o
n
ic
Ec
h
o
w
it
h
M
o
th
e
r
W
a
v
e
let
Op
ti
m
iz
a
ti
o
n
a
n
d
Be
st
-
Ba
sis S
e
lec
ti
o
n
.
I
n
ter
n
a
t
io
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
C
o
mp
u
ter
En
g
i
n
e
e
rin
g
.
2
0
1
6
;
6
(
6
):
2
7
4
2
-
2
7
5
4
.
[1
1
]
S
a
lm
a
n
A
H,
A
h
m
a
d
i
N,
M
e
n
g
k
o
R,
L
a
n
g
i
A
Z,
M
e
n
g
k
o
T
L
.
Em
p
iri
c
a
l
M
o
d
e
De
c
o
m
p
o
siti
o
n
(EM
D)
Ba
se
d
De
n
o
isin
g
M
e
th
o
d
f
o
r
He
a
rt
S
o
u
n
d
S
ig
n
a
l
a
n
d
Its
P
e
rf
o
rm
a
n
c
e
An
a
ly
sis.
In
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
E
n
g
in
e
e
rin
g
.
2
0
1
6
;
6
(
5
):
2
1
9
7
-
2
2
0
4
.
[1
2
]
S
u
b
ra
m
a
n
y
a
m
M
V
,
P
ra
sa
d
G
.
A
Ne
w
A
p
p
ro
a
c
h
f
o
r
S
A
R
I
m
a
g
e
D
e
n
o
isin
g
.
I
n
ter
n
a
ti
o
n
a
l
J
o
u
rn
a
l
o
f
El
e
c
trica
l
a
n
d
Co
mp
u
ter
E
n
g
in
e
e
rin
g
.
2
0
1
5
;
5
(5
):
9
8
4
-
9
9
1
.
Evaluation Warning : The document was created with Spire.PDF for Python.