T
E
L
K
O
M
NIKA
T
elec
o
mm
un
ica
t
io
n Co
m
pu
t
i
ng
E
lect
ro
nics
a
nd
Co
ntr
o
l
Vo
l.
24
,
No
.
5
,
Octo
b
er
20
26
,
p
p
.
1
6
3
2
~
1
6
4
3
I
SS
N:
1
6
9
3
-
6
9
3
0
,
DOI
:
1
0
.
1
2
9
2
8
/
T
E
L
KOM
NI
K
A
.
v
24
i
5
.
26012
1632
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//telko
mn
ika
.
u
a
d
.
a
c.
i
d
Ra
infall det
ection
on a
t
ro
pica
l isla
nd using
a co
mm
unica
tion
sa
tellit
e dow
nlink sig
na
l and a
rtif
ic
ia
l int
ellig
ence
mo
dels
Ra
den Yud
ha
M
a
rdy
a
ns
y
a
h
1,
2
,
B
ud
hy
K
urnia
w
a
n
2
,
Sa
nt
o
s
o
So
ek
irno
2
,
Da
na
ng
E
k
o
Nury
a
nto
4
,
F
er
ry
B
ud
hi Su
s
et
y
o
3
1
C
o
mm
u
n
i
c
a
t
i
o
n
N
e
t
w
o
r
k
C
e
n
t
e
r
,
B
a
d
a
n
M
e
t
e
o
r
o
l
o
g
i
K
l
i
mat
o
l
o
g
i
d
a
n
G
e
o
f
i
si
k
a
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
2
D
e
p
a
r
t
me
n
t
o
f
P
h
y
si
c
s,
F
a
c
u
l
t
y
o
f
M
a
t
h
e
mat
i
c
s
a
n
d
N
a
t
u
r
a
l
S
c
i
e
n
c
e
s
,
U
n
i
v
e
r
si
t
a
s
I
n
d
o
n
e
si
a
,
D
e
p
o
k
,
I
n
d
o
n
e
si
a
3
D
e
p
a
r
t
me
n
t
o
f
M
e
c
h
a
n
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
F
a
c
u
l
t
y
o
f
En
g
i
n
e
e
r
i
n
g
,
U
n
i
v
e
r
si
t
a
s Ne
g
e
r
i
Jak
a
r
t
a
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
4
R
e
se
a
r
c
h
a
n
d
D
e
v
e
l
o
p
me
n
t
C
e
n
t
e
r
,
B
a
d
a
n
M
e
t
e
o
r
o
l
o
g
i
K
l
i
mat
o
l
o
g
i
d
a
n
G
e
o
f
i
si
k
a
,
Ja
k
a
r
t
a
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
an
26
,
2
0
2
6
R
ev
i
s
ed
J
u
n
25
,
2026
A
cc
ep
ted
J
u
n
30
,
2026
T
h
e
d
e
v
e
lo
p
m
e
n
t
o
f
ra
in
f
a
ll
e
stim
a
tes
b
a
se
d
o
n
sa
telli
te
l
in
k
s
i
n
s
e
v
e
ra
l
h
ig
h
latit
u
d
e
re
g
io
n
s
sh
o
w
s
g
o
o
d
re
su
lt
s.
Ho
w
e
v
e
r,
it
s
d
e
v
e
lo
p
m
e
n
t
f
o
r
tro
p
ica
l
re
g
io
n
s
is
stil
l
v
e
r
y
les
s
p
ro
m
isin
g
,
lea
v
in
g
th
e
q
u
e
stio
n
o
f
w
h
ich
m
e
th
o
d
is
w
o
rth
u
si
n
g
.
In
t
h
is
w
o
rk
,
w
e
in
v
e
stig
a
t
e
a
rti
f
icia
l
in
telli
g
e
n
c
e
(A
I)
m
o
d
e
ls
b
y
u
ti
li
z
in
g
th
e
Ku
-
b
a
n
d
sa
telli
te
d
o
w
n
li
n
k
sig
n
a
l
f
o
r
ra
in
f
a
ll
d
e
te
c
ti
o
n
o
n
a
tro
p
ica
l
islan
d
.
L
o
n
g
-
term
e
v
a
lu
a
ti
o
n
o
f
a
ll
t
h
e
m
o
d
e
ls
is
p
e
rf
o
rm
e
d
u
sin
g
o
n
e
y
e
a
r
o
f
ra
in
f
a
ll
in
ten
sity
d
a
ta.
T
h
e
f
in
d
in
g
s
o
f
th
is
st
u
d
y
d
e
m
o
n
stra
te
th
a
t
o
n
e
-
d
im
e
n
sio
n
a
l
c
o
n
v
o
l
u
ti
o
n
n
e
u
ra
l
n
e
tw
o
rk
(1
DCN
N),
lo
n
g
sh
o
rt
-
term
m
e
m
o
r
y
(L
S
T
M
),
a
n
d
ra
n
d
o
m
f
o
re
st
(RF
)
m
o
d
e
ls
c
a
n
e
ff
e
c
ti
v
e
l
y
id
e
n
ti
fy
ra
in
f
a
ll
o
c
c
u
rre
n
c
e
s
o
n
tro
p
ica
l
islan
d
s,
w
it
h
t
h
e
a
re
a
u
n
d
e
r
t
h
e
re
c
e
iv
e
r
o
p
e
ra
ti
n
g
c
h
a
ra
c
teristic
c
u
rv
e
(
RO
C)
v
a
lu
e
o
f
0
.
8
4
,
0
.
8
4
,
a
n
d
0
.
8
3
,
re
sp
e
c
ti
v
e
l
y
.
T
h
e
se
th
re
e
m
o
d
e
ls
c
o
n
siste
n
tl
y
p
ro
v
id
e
b
e
tt
e
r
a
c
c
u
ra
c
y
in
d
e
tec
ti
n
g
th
e
re
g
io
n
’
s
d
iv
e
rse
ra
in
f
a
ll
p
a
tt
e
rn
s
th
r
o
u
g
h
o
u
t
t
h
e
y
e
a
r.
T
h
e
1
DCN
N
h
a
s
b
e
tt
e
r
p
e
rf
o
r
m
a
n
c
e
c
o
m
p
a
re
d
to
th
e
o
th
e
r
m
o
d
e
ls,
a
c
h
iev
in
g
a
c
c
u
ra
c
y
v
a
lu
e
s
o
f
0
.
7
6
a
n
d
0
.
8
2
d
u
rin
g
t
h
e
p
e
ri
o
d
s
o
f
lo
w
e
st
a
n
d
h
ig
h
e
st
ra
in
f
a
ll
,
re
sp
e
c
ti
v
e
l
y
.
Co
n
sid
e
ri
n
g
th
e
p
re
se
n
c
e
o
f
n
u
m
e
ro
u
s
p
o
ten
t
ial
m
isl
a
b
e
led
ra
in
sa
m
p
les
in
t
h
e
d
a
tas
e
t,
th
is
re
su
lt
is
e
x
c
e
ll
e
n
t.
I
n
a
d
d
it
io
n
,
th
e
u
ti
li
z
a
ti
o
n
o
f
h
ig
h
e
r
re
so
lu
ti
o
n
ra
in
f
a
ll
in
ten
sity
d
a
ta
c
a
n
b
e
a
g
r
e
a
t
p
o
ten
ti
a
l
f
o
r
re
a
l
-
ti
m
e
ra
in
fa
ll
m
o
n
it
o
ri
n
g
in
th
e
tr
o
p
ica
l
re
g
io
n
.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
Ku
-
b
an
d
s
atelli
te
lin
k
lo
n
g
-
ter
m
p
er
f
o
r
m
a
n
ce
Ma
ch
i
n
e
lear
n
i
n
g
Satellite c
o
m
m
u
n
icatio
n
S
ea
s
o
n
-
b
ased
ev
al
u
atio
n
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
:
B
u
d
h
y
K
u
r
n
ia
w
an
Dep
ar
t
m
en
t o
f
P
h
y
s
ics,
Fac
u
lt
y
o
f
Ma
t
h
e
m
a
tics
a
n
d
Natu
r
al
Scien
ce
s
,
Un
i
v
er
s
ita
s
I
n
d
o
n
e
s
i
a
B
u
ild
in
g
F,
Un
iv
er
s
itas
I
n
d
o
n
esia
C
a
m
p
u
s
,
Dep
o
k
1
6
4
2
4
,
W
est J
av
a,
I
n
d
o
n
esia
E
m
ail:
b
u
d
h
y
.
k
u
r
n
ia
w
a
n
@
s
ci.
u
i.a
c.
id
1.
I
NT
RO
D
UCT
I
O
N
Sev
er
al
s
t
u
d
ies
h
av
e
b
ee
n
c
o
n
d
u
cted
o
v
er
th
e
p
ast
1
0
y
ea
r
s
o
n
th
e
u
s
e
o
f
telec
o
m
m
u
n
ica
tio
n
s
s
atellite
s
’
c
h
a
n
n
el
s
ig
n
al
s
o
p
er
atin
g
i
n
t
h
e
m
icr
o
w
a
v
e
f
r
eq
u
en
c
y
r
an
g
e
[
1
]
in
th
e
K
u
(
1
2
–
1
8
GHz
)
an
d
Ka
(
2
7
–
4
0
GHz
)
f
r
eq
u
en
c
y
b
a
n
d
s
[
2
]
as
alter
n
ativ
es
to
r
ain
f
all
m
ea
s
u
r
e
m
e
n
t
s
.
T
h
e
tr
an
s
m
is
s
io
n
o
f
elec
tr
o
m
ag
n
etic
w
a
v
es
in
t
h
e
s
e
f
r
eq
u
en
c
y
r
a
n
g
e
s
ca
n
b
e
co
n
s
id
er
ab
ly
a
f
f
ec
ted
b
y
r
ai
n
atte
n
u
at
io
n
,
as
w
e
ll
as
o
th
er
ef
f
ec
t
s
ca
u
s
ed
b
y
air
g
a
s
es a
n
d
c
lo
u
d
s
,
alt
h
o
u
g
h
t
h
ese
ef
f
ec
ts
ar
e
less
s
i
g
n
i
f
ican
t b
u
t
m
o
r
e
f
r
eq
u
e
n
t.
T
h
e
r
ed
u
ctio
n
in
th
e
s
tr
e
n
g
th
o
f
th
e
r
ec
eiv
ed
s
ig
n
al
is
d
ir
ec
tly
p
r
o
p
o
r
tio
n
al
to
th
e
q
u
an
tit
y
o
f
p
r
ec
ip
itatio
n
en
co
u
n
ter
ed
alo
n
g
t
h
e
p
ath
o
f
th
e
w
av
e
[
3
]
.
Nu
m
er
o
u
s
r
esear
c
h
er
s
h
a
v
e
i
n
v
esti
g
ated
r
ain
f
all
m
ea
s
u
r
e
m
e
n
t
f
r
o
m
s
atelli
te
lin
k
s
i
g
n
als
u
s
in
g
b
o
th
th
e
p
o
w
er
-
la
w
ca
lc
u
latio
n
[
4
]
an
d
ar
tif
icial
in
tel
lig
e
n
ce
(
A
I
)
m
o
d
els
[
5
]
.
A
d
ir
o
s
i
et
a
l.
[
6
]
d
em
o
n
s
tr
ated
th
at
r
ea
l
-
ti
m
e
r
ain
f
all
in
te
n
s
it
y
ca
n
b
e
esti
m
ated
f
r
o
m
th
e
atte
n
u
at
io
n
o
f
Ku
-
an
d
Ka
-
b
an
d
s
atellite
r
ec
ep
tio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
R
a
in
fa
ll d
etec
tio
n
o
n
a
tr
o
p
ica
l isl
a
n
d
u
s
in
g
a
c
o
mmu
n
ica
tio
n
s
a
tellite
… (
R
a
d
en
Yu
d
h
a
Ma
r
d
ya
n
s
ya
h
)
1633
s
ig
n
al
s
b
y
u
s
i
n
g
t
h
e
p
o
w
er
-
la
w
m
o
d
el
an
d
r
ai
n
clo
u
d
h
ei
g
h
t
d
ata.
Se
v
er
al
ad
d
itio
n
al
s
a
tellite
f
r
eq
u
en
c
ies,
in
cl
u
d
in
g
K
-
b
an
d
[
7
]
,
[
8
]
,
X
-
b
an
d
[
9
]
,
an
d
Q/V
-
b
an
d
[
1
0
]
h
a
v
e
al
s
o
b
ee
n
u
s
ed
.
R
esear
ch
er
s
h
a
v
e
co
n
s
tr
u
cted
a
s
y
s
te
m
f
o
r
th
e
d
etec
tio
n
an
d
esti
m
atio
n
o
f
r
ain
f
all
f
ield
s
f
r
o
m
d
ig
ita
l
v
id
eo
b
r
o
a
d
ca
s
tin
g
s
atellite
r
ec
eiv
er
s
as
p
ar
t
o
f
t
h
e
n
e
t
w
o
r
k
f
o
r
r
ain
f
o
r
ec
asti
n
g
u
s
i
n
g
b
r
o
ad
b
an
d
s
atellite
telec
o
m
m
u
n
icat
io
n
s
(
NE
FO
C
A
ST
)
an
d
in
n
o
v
ati
v
e
s
atell
ite
-
b
ased
r
ain
f
all
d
etec
tio
n
a
n
d
esti
m
atio
n
f
o
r
r
ain
(
I
NSI
DE
R
A
I
N
)
P
r
o
j
ec
t
in
I
tal
y
,
ai
m
ed
at
ac
h
iev
in
g
r
eg
io
n
al
co
v
er
a
g
e
[
1
1
]
–
[
1
4
]
.
R
ain
f
a
ll i
n
ten
s
it
y
p
er
m
in
u
te
i
n
an
i
n
v
er
s
io
n
m
o
d
el
f
o
r
lo
n
g
-
ter
m
(
i.e
.
,
o
n
e
y
ea
r
)
p
er
f
o
r
m
a
n
ce
h
a
s
als
o
b
ee
n
ev
alu
at
ed
;
th
i
s
m
ea
s
u
r
e
m
en
t
w
a
s
tak
e
n
w
it
h
a
Ku
-
b
a
n
d
s
atellite
r
ec
eiv
er
b
u
ilt
i
n
Nan
j
in
g
,
C
h
i
n
a
[
1
5
]
.
Gh
ar
an
j
ik
et
a
l.
[
7
]
u
s
ed
an
ar
tif
icia
l
n
e
u
r
al
n
et
w
o
r
k
(
ANN)
to
d
is
ce
r
n
b
et
w
ee
n
ra
in
an
d
d
r
y
e
v
en
t
s
b
y
an
al
y
zi
n
g
t
h
e
ca
r
r
ier
-
to
-
n
o
is
e
p
o
w
er
r
atio
d
ata
f
r
o
m
t
h
e
f
o
r
w
ar
d
lin
k
an
d
r
etu
r
n
li
n
k
o
f
a
b
r
o
ad
b
an
d
s
atellite
co
m
m
u
n
icatio
n
n
et
w
o
r
k
in
B
etzd
o
r
f
,
L
u
x
e
m
b
o
u
r
g
.
R
ec
e
n
t r
e
s
ea
r
ch
o
n
r
ai
n
f
al
l d
etec
tio
n
in
Na
n
j
in
g
,
C
h
i
n
a,
h
a
s
u
s
ed
a
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
i
n
e
(
S
VM
)
to
id
en
tify
r
ain
y
an
d
n
o
n
-
r
ai
n
y
p
er
io
d
s
w
it
h
Ku
-
b
an
d
s
atelli
te
lin
k
s
ig
n
als
[
1
6
]
.
T
h
e
d
ev
elo
p
m
e
n
t
o
f
r
ai
n
f
a
ll
e
s
ti
m
ates
b
ased
o
n
s
a
tellite
li
n
k
s
i
n
s
e
v
er
al
h
i
g
h
latit
u
d
e
r
eg
i
o
n
s
s
h
o
w
s
g
o
o
d
r
es
u
lts
.
Ho
w
e
v
er
,
its
d
ev
elo
p
m
e
n
t
f
o
r
tr
o
p
ical
r
eg
io
n
s
is
s
till
v
er
y
le
s
s
p
r
o
m
is
i
n
g
,
leav
in
g
th
e
q
u
e
s
tio
n
o
f
w
h
ic
h
m
eth
o
d
is
w
o
r
t
h
u
s
in
g
.
I
n
th
i
s
r
esear
ch
,
w
e
w
i
ll
s
tu
d
y
A
I
m
o
d
els
t
h
at
h
a
v
e
b
ee
n
u
s
ed
at
h
ig
h
latitu
d
es
p
lu
s
th
r
ee
o
th
er
m
o
d
els
w
h
ich
ar
e
ex
p
ec
te
d
to
p
r
o
v
id
e
b
etter
p
er
f
o
r
m
a
n
c
es.
T
h
e
lo
n
g
-
ter
m
ev
alu
a
tio
n
o
f
al
l th
e
m
o
d
el
s
is
also
p
er
f
o
r
m
ed
b
ased
o
n
o
n
e
y
ea
r
o
f
r
ain
f
all
i
n
te
n
s
it
y
d
ata.
2.
M
E
T
H
O
D
I
n
th
i
s
w
o
r
k
,
w
e
e
x
a
m
in
e
o
n
e
-
d
i
m
e
n
s
io
n
al
co
n
v
o
lu
tio
n
n
e
u
r
al
n
et
w
o
r
k
(
1
DC
NN)
,
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
ST
M)
,
an
d
r
an
d
o
m
f
o
r
est
(
R
F)
b
in
ar
y
class
i
f
icati
o
n
m
o
d
els
th
at
h
a
v
e
n
e
v
er
b
ef
o
r
e
b
ee
n
a
p
p
lied
t
o
th
e
d
etec
tio
n
o
f
r
ain
f
all
f
r
o
m
s
atellite
d
o
w
n
l
in
k
s
ig
n
als.
T
h
e
SVM
a
n
d
a
m
u
lti
la
y
er
p
er
ce
p
tr
o
n
(
m
u
lt
ila
y
e
r
p
er
ce
p
tr
o
n
/
ML
P
,
a
mu
lti
-
la
y
er
A
NN)
ar
e
ev
al
u
ated
as
co
m
p
ar
is
o
n
m
o
d
els.
F
ig
u
r
e
s
1
(
a)
an
d
(
b
)
illu
s
tr
ates
th
e
ar
ch
itect
u
r
al
co
n
ce
p
t o
f
th
e
1
DC
NN
a
n
d
L
ST
M
m
o
d
els.
(
a)
(
b
)
Fig
u
r
e
1
.
T
h
e
ar
ch
itectu
r
e
o
f
(
a)
1
DC
NN
an
d
(
b
)
an
d
L
ST
M
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
5
,
Octo
b
e
r
20
26
:
1
6
3
2
-
1643
1634
T
h
e
1
DC
NN
is
a
s
p
ec
ialized
t
y
p
e
o
f
co
n
v
o
l
u
tio
n
n
e
u
r
al
n
e
t
w
o
r
k
(
C
NN
)
[
1
7
]
u
s
ed
f
o
r
p
r
o
ce
s
s
in
g
o
n
e
-
d
i
m
en
s
io
n
al
d
ata,
lik
e
s
e
q
u
en
ce
s
o
r
ti
m
e
s
er
ies.
I
t
e
m
p
lo
y
s
co
n
v
o
l
u
tio
n
al
la
y
er
s
w
i
th
f
il
ter
s
to
ca
p
tu
r
e
lo
ca
l
p
atter
n
s
an
d
f
ea
tu
r
es
i
n
th
e
in
p
u
t
d
ata.
A
s
d
ep
icted
in
Fig
u
r
e
1
(
a
)
,
1
DC
NN
co
n
s
is
t
s
o
f
co
n
v
o
lu
tio
n
a
l
la
y
er
,
p
o
o
lin
g
la
y
er
,
d
en
s
e
lay
er
an
d
o
u
tp
u
t
la
y
er
w
h
ic
h
a
cts
as
th
e
p
r
ed
ictio
n
o
u
tp
u
t
o
f
th
e
m
o
d
el.
T
h
e
L
ST
M
m
o
d
el,
p
r
o
p
o
s
ed
b
y
H
o
ch
r
eiter
an
d
Sch
m
id
h
u
b
er
[
1
8
]
,
is
a
r
o
b
u
s
t
r
ec
u
r
r
en
t
n
eu
r
al
s
y
s
te
m
s
p
ec
i
f
icall
y
en
g
i
n
ee
r
ed
to
ad
d
r
e
s
s
th
e
is
s
u
es
o
f
ex
p
lo
d
in
g
o
r
v
an
i
s
h
in
g
g
r
ad
ien
ts
t
h
at
co
m
m
o
n
l
y
o
cc
u
r
d
u
r
in
g
t
h
e
lear
n
in
g
o
f
lo
n
g
-
ter
m
d
ep
en
d
e
n
cies.
L
ik
e
t
h
e
1
DC
NN,
t
h
e
L
ST
M
m
o
d
el
is
co
m
p
o
s
ed
o
f
a
la
y
e
r
o
f
L
ST
M
u
n
its
,
a
d
en
s
e
la
y
er
,
an
d
an
o
u
tp
u
t
la
y
er
th
at
s
er
v
es
as
th
e
m
o
d
el
’
s
p
r
ed
icti
o
n
o
u
tp
u
t,
as
d
e
p
icted
in
F
ig
u
r
e
1
(
b
)
.
T
h
e
L
ST
M
ar
ch
itect
u
r
e
co
m
p
r
is
es
a
co
llectio
n
o
f
i
n
ter
co
n
n
ec
te
d
s
u
b
-
n
et
w
o
r
k
s
,
r
ef
er
r
ed
to
as
m
e
m
o
r
y
b
lo
ck
s
o
r
L
ST
M
u
n
it
s
.
T
h
e
p
u
r
p
o
s
e
o
f
th
e
m
e
m
o
r
y
b
lo
ck
is
to
p
r
eser
v
e
its
s
tate
o
v
er
ti
m
e
an
d
co
n
tr
o
l
th
e
f
lo
w
o
f
in
f
o
r
m
atio
n
t
h
r
o
u
g
h
n
o
n
-
li
n
ea
r
g
atin
g
u
n
it
s
.
T
h
e
tr
an
s
p
o
r
t
o
f
in
f
o
r
m
at
io
n
to
th
e
n
e
x
t
ce
ll
i
n
v
o
l
v
es
t
w
o
s
tate
s
:
th
e
ce
ll
s
tate
an
d
t
h
e
h
id
d
en
s
t
ate.
T
h
e
h
id
d
en
s
tate
at
t
i
m
e
s
tep
t
en
co
m
p
as
s
es
t
h
e
o
u
tp
u
t
o
f
th
e
L
ST
M
la
y
er
f
o
r
th
i
s
s
p
ec
i
f
ic
ti
m
e
s
tep
.
T
h
e
ce
ll
s
tate
e
n
co
m
p
as
s
es
k
n
o
w
l
ed
g
e
ac
q
u
ir
ed
f
r
o
m
p
r
ec
ed
in
g
ti
m
e
s
tep
s
.
Du
r
i
n
g
ea
ch
ti
m
e
s
tep
,
th
e
la
y
er
eith
er
in
co
r
p
o
r
ates
n
e
w
i
n
f
o
r
m
atio
n
in
to
th
e
ce
ll
s
tate
o
r
eli
m
i
n
ate
s
ex
is
t
in
g
in
f
o
r
m
atio
n
f
r
o
m
it.
T
h
e
R
F
tech
n
iq
u
e
i
s
an
en
s
e
m
b
le
alg
o
r
ith
m
f
ir
s
t
in
tr
o
d
u
ce
d
b
y
B
r
eim
a
n
[
1
9
]
.
I
t
h
as
b
ee
n
d
esig
n
ed
to
h
an
d
le
v
ar
io
u
s
n
o
n
li
n
ea
r
cl
ass
i
f
icatio
n
an
d
r
e
g
r
ess
io
n
ta
s
k
s
.
T
h
e
b
o
o
ts
tr
ap
s
a
m
p
lin
g
m
eth
o
d
an
d
r
an
d
o
m
f
ea
t
u
r
e
s
elec
tio
n
w
er
e
u
s
ed
to
cr
ea
te
a
lar
g
e
an
d
d
iv
er
s
e
f
o
r
est
co
n
s
i
s
ti
n
g
o
f
n
u
m
er
o
u
s
d
ec
is
io
n
tr
ee
s
[
2
0
]
.
R
F
o
u
tp
er
f
o
r
m
s
a
s
i
n
g
le
d
ec
is
io
n
tr
ee
an
d
m
iti
g
ates
o
v
er
-
f
itt
in
g
b
y
e
m
p
lo
y
i
n
g
an
av
er
a
g
i
n
g
tec
h
n
iq
u
e.
T
h
e
in
s
tab
il
it
y
o
f
d
ec
is
io
n
tr
ee
s
is
attr
ib
u
ted
to
th
eir
h
ig
h
v
ar
iati
o
n
an
d
lo
w
b
ias.
T
h
e
f
in
al
r
esu
lt
is
d
eter
m
i
n
ed
b
y
eith
er
s
elec
ti
n
g
th
e
ca
te
g
o
r
y
w
it
h
t
h
e
h
i
g
h
est
n
u
m
b
er
o
f
v
o
tes
in
t
h
e
class
if
ica
tio
n
m
o
d
el
o
r
ca
lcu
latin
g
th
e
av
er
ag
e
p
r
ed
ictio
n
f
r
o
m
all
d
e
cisi
o
n
tr
ee
s
i
n
t
h
e
r
eg
r
e
s
s
io
n
m
o
d
el.
Si
g
n
i
f
ica
n
tl
y
,
it
f
ac
i
lita
tes r
ap
id
p
r
o
ce
s
s
in
g
f
o
r
ex
te
n
s
i
v
el
y
p
ar
allel
co
m
p
u
ti
n
g
o
n
s
u
b
s
ta
n
tial
d
ataset
s
o
r
tr
ain
in
g
m
u
ltid
i
m
e
n
s
io
n
al
v
ec
to
r
s
w
it
h
a
h
i
g
h
lev
el
o
f
p
r
ed
ictiv
e
ac
c
u
r
ac
y
.
2
.
1
.
Da
t
a
s
et
T
h
e
m
o
d
els
e
m
p
lo
y
ed
in
t
h
i
s
s
tu
d
y
u
ti
lize
d
ata
o
n
th
e
e
n
e
r
g
y
p
er
s
y
m
b
o
l
to
n
o
is
e
p
o
wer
s
p
ec
tr
al
d
en
s
it
y
r
atio
(
E
s
/No
)
ac
q
u
ir
ed
o
v
er
a
p
e
r
io
d
o
f
o
n
e
h
o
u
r
.
T
h
is
d
ata
r
ep
r
esen
ts
th
e
q
u
a
lit
y
o
f
th
e
s
atell
ite
s
ig
n
al
r
ec
eiv
ed
b
y
t
h
e
u
s
er
ter
m
i
n
al
(
UT
)
in
a
m
u
lti
-
s
p
o
t
-
b
e
a
m
s
atel
lite
n
et
w
o
r
k
,
as
d
ep
icted
in
Fig
u
r
e
2
[
2
1
]
.
Nu
m
er
o
u
s
UT
s
ar
e
d
is
p
er
s
ed
ac
r
o
s
s
s
p
e
cif
ic
s
atell
ite
b
ea
m
p
o
in
ts
an
d
ar
e
co
n
n
ec
ted
to
o
n
e
o
r
m
o
r
e
n
et
w
o
r
k
g
ate
w
a
y
s
,
w
h
ic
h
s
er
v
e
a
s
t
h
e
n
et
w
o
r
k
’
s
co
m
m
u
n
icatio
n
h
u
b
.
E
ac
h
UT
r
ec
eiv
es
K
u
-
b
an
d
d
o
w
n
li
n
k
s
i
g
n
al
s
f
r
o
m
A
p
s
tar
-
5C
’
s
1
3
8
ea
s
t
o
r
b
ital
p
o
s
itio
n
.
Sin
ce
t
h
e
Gat
e
w
a
y
s
atel
lite
n
et
w
o
r
k
e
m
p
lo
y
s
a
s
ite
d
iv
er
s
i
t
y
s
y
s
te
m
[
2
2
]
,
as
p
o
s
tu
lated
b
y
Gh
ar
an
j
ik
et
a
l.
[
7
]
,
th
e
E
s
/No
d
ata
r
ec
eiv
ed
b
y
th
e
UT
is
d
o
m
i
n
ate
d
b
y
r
ain
o
n
th
e
d
o
w
n
li
n
k
s
id
e
f
r
o
m
t
h
e
s
at
ellite
to
th
e
UT
.
E
ac
h
UT
h
as
a
p
at
h
to
th
e
in
t
er
n
et
at
t
h
e
n
e
t
w
o
r
k
g
ate
w
a
y
f
o
r
clien
t
ac
ce
s
s
.
I
n
ad
d
itio
n
,
a
n
et
w
o
r
k
m
an
a
g
e
m
e
n
t
s
y
s
te
m
(
NM
S)
s
er
v
es
as
a
d
ata
tr
af
f
ic
co
n
tr
o
ll
er
an
d
co
llects
th
e
E
s
/No
d
at
a
r
ec
eiv
e
d
b
y
ea
c
h
UT
at
th
e
g
ate
w
a
y
lo
ca
tio
n
.
T
h
e
E
s
/No
d
ata
f
o
r
a
s
p
ec
if
ic
p
er
io
d
o
f
ti
m
e
is
s
to
r
ed
in
a
d
atab
ase.
B
y
u
s
in
g
h
o
u
r
l
y
r
ain
f
all
i
n
te
n
s
i
t
y
d
at
a
f
r
o
m
t
h
e
n
ea
r
e
s
t
Hell
m
a
n
n
r
ai
n
g
au
g
e
(
R
G)
[
2
3
]
,
[
2
4
]
at
th
e
R
an
ai
Me
teo
r
o
lo
g
ical
Statio
n
,
ea
c
h
h
o
u
r
o
f
E
s
/
No
d
ata
is
class
if
ied
as
eith
er
r
ain
o
r
d
r
y
co
n
d
itio
n
s
.
T
h
e
R
an
a
i
Me
teo
r
o
lo
g
ical
Statio
n
is
a
wea
th
er
o
b
s
er
v
atio
n
f
ac
ilit
y
t
h
at
is
u
n
d
er
t
h
e
o
w
n
er
s
h
ip
o
f
th
e
Me
teo
r
o
lo
g
ical
C
li
m
ato
lo
g
ical
a
n
d
Geo
p
h
y
s
i
ca
l
Ag
en
c
y
o
f
I
n
d
o
n
esia.
T
h
e
r
ain
f
all
in
te
n
s
it
y
d
ata
f
r
o
m
Hell
m
an
n
’
s
h
a
s
u
n
d
er
g
o
n
e
q
u
a
lit
y
co
n
tr
o
l
d
u
e
to
th
e
m
e
m
b
er
s
h
ip
o
f
th
e
R
an
ai
Me
teo
r
o
lo
g
ical
S
tatio
n
in
th
e
Glo
b
al
B
asic
Ob
s
er
v
i
n
g
Net
w
o
r
k
o
f
t
h
e
W
o
r
ld
Me
teo
r
o
l
o
g
ical
Or
g
a
n
izatio
n
(
W
MO
)
.
T
h
e
W
MO
I
n
te
g
r
a
ted
Glo
b
al
Ob
s
er
v
i
n
g
S
y
s
te
m
(
W
I
GOS
)
Statio
n
I
d
en
tifie
r
o
f
R
an
ai
Me
teo
r
o
lo
g
ical
Statio
n
i
s
0
-
20000
-
0
-
9
6
1
4
7
(
h
ttp
s
://o
s
ca
r
.
wm
o
.
in
t
/s
u
r
f
ac
e/
#
/
).
Fig
u
r
e
2
.
Ku
-
b
a
n
d
s
atellite
co
m
m
u
n
icatio
n
n
et
w
o
r
k
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
R
a
in
fa
ll d
etec
tio
n
o
n
a
tr
o
p
ica
l isl
a
n
d
u
s
in
g
a
c
o
mmu
n
ica
tio
n
s
a
tellite
… (
R
a
d
en
Yu
d
h
a
Ma
r
d
ya
n
s
ya
h
)
1635
I
n
Fi
g
u
r
e
3
,
t
h
e
UT
an
d
R
G
ar
e
lo
ca
ted
o
n
Natu
n
a
I
s
la
n
d
,
w
h
ic
h
i
s
a
tr
o
p
ical
is
la
n
d
lo
ca
ted
in
t
h
e
So
u
t
h
C
h
in
a
Sea
i
n
s
id
e
th
e
I
n
d
o
n
esian
ter
r
ito
r
y
.
T
h
e
r
ed
lin
e
in
d
icate
s
th
e
d
ir
ec
tio
n
o
f
th
e
lin
k
b
et
w
ee
n
t
h
e
UT
an
d
th
e
s
atellite.
T
h
e
UT
is
p
ar
t
o
f
th
e
h
ig
h
-
t
h
r
o
u
g
h
p
u
t
-
s
atellite
b
r
o
ad
b
an
d
n
et
w
o
r
k
o
p
er
ated
b
y
T
elk
o
m
s
at,
a
s
u
b
s
id
iar
y
o
f
T
elk
o
m
t
h
at
p
r
o
v
id
es
s
atellite
s
e
r
v
ices
in
I
n
d
o
n
e
s
ia.
T
h
e
co
m
p
an
y
s
p
ec
ializes
i
n
p
r
o
v
id
in
g
h
ig
h
-
q
u
ali
t
y
a
n
d
in
t
er
n
atio
n
al
-
s
tan
d
ar
d
u
p
s
tr
ea
m
t
o
d
o
w
n
s
tr
ea
m
s
er
v
ices.
Fig
u
r
e
3
.
L
o
ca
tio
n
o
f
t
h
e
UT
an
d
R
G
o
n
Nat
u
n
a
I
s
lan
d
E
v
er
y
t
w
o
m
in
u
te
s
,
th
e
n
et
w
o
r
k
g
ate
w
a
y
s
to
r
es
t
h
e
E
s
/No
d
ata
s
en
t
b
y
t
h
e
UT
in
t
h
e
d
atab
ase.
T
h
e
R
G
(
3
.
9
1
2
0
3
4
°N,
1
0
8
.
3
9
3
5
3
0
°E)
,
lo
ca
ted
3
.
3
k
ilo
m
e
ter
s
a
w
a
y
f
r
o
m
t
h
e
UT
(
3
.
9
4
1
1
1
0
°N,
1
0
8
.
3
8
7
6
6
6
°E)
,
m
ea
s
u
r
es
r
ai
n
f
all
i
n
te
n
s
i
t
y
o
r
r
ain
r
ate
(
R
R
)
i
n
m
ill
i
m
e
ter
s
p
er
h
o
u
r
.
B
ased
o
n
s
e
v
er
al
m
eso
s
ca
le
clo
u
d
s
tu
d
ie
s
in
I
n
d
o
n
esia
[
2
5
]
,
[
2
6
]
,
th
e
lo
ca
tio
n
o
f
t
h
e
R
G
is
co
n
s
id
er
ed
to
r
ep
r
esen
t
r
ain
f
all
at
t
h
e
UT
s
ite
b
ec
au
s
e
th
e
f
o
r
m
er
is
f
o
u
n
d
in
th
e
s
a
m
e
w
e
t
r
eg
io
n
d
u
r
in
g
ex
tr
e
m
e
w
ea
t
h
er
ev
en
ts
.
T
h
e
UT
an
d
th
e
R
G
ar
e
also
s
itu
ated
i
n
th
e
s
a
m
e
c
o
astal
r
eg
io
n
,
w
h
ich
e
n
s
u
r
es
th
at
th
e
W
o
r
ld
Me
teo
r
o
lo
g
ical
Or
g
an
iza
tio
n
’
s
r
ec
o
m
m
e
n
d
ed
n
et
w
o
r
k
co
v
er
a
g
e
f
o
r
r
ain
g
au
g
es i
s
m
a
in
ta
in
ed
[
2
7
]
.
W
e
d
ef
in
e
th
e
E
s
/No
p
ar
am
et
er
w
ith
d
r
y
o
r
r
ain
lab
els
b
ase
d
o
n
th
e
r
ain
f
all
in
te
n
s
it
y
d
ata,
w
h
er
e
R
R
>
0
d
en
o
tes
a
r
ain
lab
el.
Fig
u
r
e
4
(
a
)
s
h
o
w
s
th
e
E
s
/
No
an
d
R
R
d
ata
f
r
o
m
Dec
e
m
b
er
2
0
1
9
t
o
Sep
tem
b
er
2
0
2
2
.
E
ac
h
h
o
u
r
o
f
E
s
/No
d
ata
th
at
co
r
r
esp
o
n
d
s
to
R
R
>
0
is
h
ig
h
li
g
h
ted
i
n
g
r
ee
n
i
n
Fig
u
r
e
4
(
b
)
an
d
r
e
p
r
esen
ts
a
r
ain
lab
el;
o
th
er
w
is
e,
it
r
ep
r
es
en
ts
a
d
r
y
lab
el.
T
h
e
o
n
e
-
h
o
u
r
s
a
m
p
le
o
f
E
s
/No
d
ata,
d
en
o
t
ed
as
=
[
̅
n
-
1
,
̅
n
-
2
,
̅
n
-
3
,
.
.
.
̅
n
-
3
0
]
,
is
f
o
r
m
ed
b
y
co
n
ca
te
n
ati
n
g
t
w
o
-
m
in
u
te
E
s
/No
d
ata
in
to
a
s
in
g
le
v
ec
to
r
r
ep
r
esen
tin
g
a
ti
m
e
f
r
a
m
e
o
f
o
n
e
h
o
u
r
,
ac
co
r
d
in
g
to
th
e
m
et
h
o
d
u
s
ed
b
y
Ha
b
i a
n
d
Me
s
s
er
[
2
8
]
.
(
a)
(
b
)
Fig
u
r
e
4
.
Data
s
et
o
f
t
h
e
E
s
/No
in
d
ec
ib
els (
d
B
)
an
d
R
R
in
m
i
lli
m
eter
s
p
er
h
o
u
r
(
m
m
/
h
)
; (
a)
f
r
o
m
Dec
e
m
b
er
2
0
1
9
th
r
o
u
g
h
Sep
te
m
b
er
2
0
2
2
an
d
(
b
)
s
am
p
le
d
ata
ill
u
s
tr
ati
n
g
E
s
/No
m
ar
k
i
n
g
s
f
o
r
b
o
th
d
r
y
an
d
r
ain
co
n
d
itio
n
s
2
.
2
.
P
re
pro
ce
s
s
ing
F
ig
u
r
e
5
p
r
e
s
en
ts
th
e
f
l
o
w
ch
ar
t
o
f
th
e
r
ai
n
f
al
l
d
e
t
ec
t
i
o
n
m
o
d
e
l
,
w
h
i
ch
c
o
m
p
r
is
e
s
t
h
r
ee
s
ta
g
e
s
:
(
i
)
d
a
ta
p
r
e
p
r
o
c
e
s
s
i
n
g
,
(
i
i
)
m
o
d
e
l
c
o
n
s
tr
u
c
ti
o
n
,
an
d
(
i
i
i
)
m
o
d
e
l
ev
al
u
a
ti
o
n
.
T
h
e
in
i
ti
a
l
s
t
e
p
in
d
a
t
a
p
r
ep
r
o
c
e
s
s
in
g
in
v
o
lv
es
t
h
e
u
ti
l
iz
a
t
i
o
n
o
f
li
n
e
a
r
in
t
e
r
p
o
la
t
i
o
n
o
n
E
s
/N
o
d
a
t
a
t
o
a
d
d
r
es
s
t
h
e
g
a
p
s
th
a
t
a
r
i
s
e
d
u
e
t
o
th
e
ab
s
e
n
c
e
o
f
a
s
at
e
ll
i
te
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
5
,
Octo
b
e
r
20
26
:
1
6
3
2
-
1643
1636
c
o
n
n
e
c
ti
o
n
l
in
k
.
Su
b
s
eq
u
en
t
l
y
,
d
ata
n
o
r
m
aliza
tio
n
is
ca
r
r
ied
o
u
t,
a
p
r
o
ce
d
u
r
e
th
at
h
a
s
b
ee
n
e
x
p
lo
r
ed
b
y
s
ev
er
al
a
u
th
o
r
s
[
2
9
]
,
[
3
0
]
.
I
n
ter
p
o
latio
n
is
li
m
ited
to
d
ata
g
ap
s
o
f
u
p
to
s
ix
m
i
n
u
te
s
.
As
p
ar
t
o
f
th
e
n
o
r
m
aliza
t
io
n
p
r
o
ce
s
s
,
th
e
s
u
b
tr
ac
tio
n
o
f
th
e
m
ed
ia
n
v
al
u
e
d
er
iv
ed
f
r
o
m
th
e
en
tire
t
y
o
f
th
e
p
r
ec
ed
in
g
an
d
f
o
llo
w
in
g
12
-
h
o
u
r
d
ata
is
co
n
d
u
cted
f
o
r
ev
er
y
ti
m
e
s
tep
.
T
h
is
m
eth
o
d
i
s
i
m
p
le
m
en
te
d
to
o
p
ti
m
ize
t
h
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
m
o
d
el
th
a
t
w
i
ll
b
e
e
m
p
lo
y
ed
.
Af
ter
i
n
t
er
p
o
latin
g
a
n
d
ex
cl
u
d
in
g
all
t
h
e
m
is
s
in
g
v
alu
e
s
,
th
er
e
ar
e
a
to
tal
o
f
2
2
,
4
5
6
s
am
p
les.
Fig
u
r
e
5
.
Flo
w
c
h
ar
t o
f
th
e
r
ai
n
f
a
ll d
etec
tio
n
m
o
d
el
Af
ter
in
ter
p
o
latio
n
a
n
d
n
o
r
m
al
izatio
n
,
th
e
d
ataset
is
d
iv
id
ed
i
n
to
tr
ain
i
n
g
a
n
d
test
in
g
s
et
s
.
T
h
e
m
o
d
el
is
tr
ain
ed
w
it
h
d
atasets
r
a
n
g
in
g
f
r
o
m
Dec
e
m
b
er
2
0
1
9
to
A
p
r
il
2
0
2
2
.
Fo
r
lo
n
g
-
ter
m
p
er
f
o
r
m
an
ce
e
v
al
u
atio
n
,
we
e
m
p
lo
y
d
ata
f
r
o
m
J
u
n
e
2
0
2
2
t
o
Ma
y
2
0
2
3
in
m
o
d
el
test
in
g
.
Dr
a
w
in
g
o
n
Hab
i
an
d
Me
s
s
er
[
2
8
]
,
w
e
b
alan
ce
t
h
e
r
ai
n
an
d
d
r
y
clas
s
es
i
n
b
o
th
th
e
tr
ain
i
n
g
a
n
d
tes
tin
g
d
ataset
s
to
e
n
h
an
ce
m
o
d
e
l
p
er
f
o
r
m
a
n
ce
.
B
y
r
an
d
o
m
l
y
r
ed
u
ci
n
g
th
e
n
u
m
b
er
o
f
d
r
y
s
a
m
p
le
s
to
m
atc
h
th
e
n
u
m
b
er
o
f
r
ain
s
a
m
p
les,
t
h
e
u
n
d
e
r
s
a
m
p
li
n
g
ap
p
r
o
ac
h
is
u
s
ed
to
ac
q
u
ir
e
a
n
eq
u
alize
d
(
5
0
:5
0
)
class
r
atio
.
I
n
th
e
e
n
d
,
th
e
d
ataset
co
n
s
is
t
s
o
f
3
,
0
4
6
tr
ain
in
g
s
a
m
p
les
an
d
4
8
0
test
i
n
g
s
a
m
p
les.
Du
r
i
n
g
th
e
tr
ai
n
i
n
g
o
f
t
h
e
1
DC
N
N,
M
L
P
,
an
d
L
ST
M
m
o
d
els,
a
r
an
d
o
m
s
elec
tio
n
o
f
2
0
%
o
f
t
h
e
tr
ai
n
in
g
d
a
ta
w
as
al
lo
ca
ted
f
o
r
d
ata
v
alid
atio
n
p
u
r
p
o
s
es.
T
h
is
s
u
b
s
et
o
f
d
ata
w
a
s
u
tili
ze
d
to
ass
e
s
s
t
h
e
p
er
f
o
r
m
a
n
ce
an
d
p
r
o
g
r
ess
o
f
t
h
e
m
o
d
el
s
th
r
o
u
g
h
o
u
t t
h
e
tr
ain
i
n
g
p
r
o
ce
s
s
.
2
.
3
.
B
ina
ry
cla
s
s
if
ica
t
io
n
m
o
del
I
n
th
is
s
tu
d
y
,
w
e
e
x
a
m
in
e
th
e
p
er
f
o
r
m
a
n
ce
o
f
f
iv
e
A
I
m
o
d
el
s
f
o
r
b
in
ar
y
cla
s
s
i
f
icatio
n
.
T
h
e
1
DC
NN
e
m
p
lo
y
s
P
o
lz
et
a
l.
’
s
co
n
f
i
g
u
r
atio
n
[
2
9
]
f
o
r
m
icr
o
w
a
v
e
-
li
n
k
r
ain
d
etec
tio
n
r
esear
c
h
.
T
h
e
ML
P
is
e
m
p
lo
y
ed
b
y
u
s
in
g
d
e
n
s
e
la
y
er
s
d
er
i
v
ed
f
r
o
m
t
h
e
1
D
C
NN.
T
h
e
L
ST
M
m
o
d
el
in
co
r
p
o
r
ates
t
h
e
la
y
er
i
n
tr
o
d
u
ce
d
b
y
Hab
i
an
d
Me
s
s
er
[
2
8
]
,
em
p
lo
y
in
g
th
e
s
a
m
e
a
m
o
u
n
t
o
f
d
en
s
e
la
y
er
s
a
s
t
h
e
M
L
P
.
T
ab
le
1
p
r
esen
ts
th
e
la
y
er
co
n
f
i
g
u
r
atio
n
f
o
r
th
e
1
DC
NN,
L
ST
M,
an
d
ML
P
m
o
d
els.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
R
a
in
fa
ll d
etec
tio
n
o
n
a
tr
o
p
ica
l isl
a
n
d
u
s
in
g
a
c
o
mmu
n
ica
tio
n
s
a
tellite
… (
R
a
d
en
Yu
d
h
a
Ma
r
d
ya
n
s
ya
h
)
1637
T
ab
le
1
.
L
ay
er
co
n
f
i
g
u
r
atio
n
f
o
r
th
e
1
DC
NN,
L
ST
M,
an
d
ML
P
m
o
d
els
1
D
C
N
N
M
L
P
L
S
T
M
L
a
y
e
r
A
c
t
i
v
a
t
i
o
n
L
a
y
e
r
A
c
t
i
v
a
t
i
o
n
L
a
y
e
r
A
c
t
i
v
a
t
i
o
n
L
a
y
e
r
c
o
n
f
i
g
u
r
a
t
i
o
n
C
o
n
v
1
D
(
2
4
,
3
)
R
e
l
u
-
-
-
-
C
o
n
v
1
D
(
2
4
,
3
)
R
e
l
u
-
-
-
-
M
a
x
P
o
o
l
1
D
-
-
-
-
-
C
o
n
v
1
D
(
4
8
,
3
)
R
e
l
u
-
-
-
-
C
o
n
v
1
D
(
4
8
,
3
)
R
e
l
u
-
-
-
-
M
a
x
P
o
o
l
1
D
-
-
-
-
-
C
o
n
v
1
D
(
4
8
,
3
)
R
e
l
u
-
-
-
-
C
o
n
v
1
D
(
4
8
,
3
)
R
e
l
u
-
-
-
-
M
a
x
P
o
o
l
1
D
-
-
-
-
-
F
l
a
t
t
e
n
-
-
-
L
S
T
M
(
1
2
8
)
R
e
l
u
D
e
n
se
(
6
4
)
R
e
l
u
D
e
n
se
(
6
4
)
R
e
l
u
D
e
n
se
(
6
4
)
R
e
l
u
D
r
o
p
o
u
t
(
0
.
4
)
-
D
r
o
p
o
u
t
(
0
.
4
)
-
D
r
o
p
o
u
t
(
0
.
4
)
-
D
e
n
se
(
6
4
)
R
e
l
u
D
e
n
se
(
6
4
)
R
e
l
u
D
e
n
se
(
6
4
)
R
e
l
u
D
r
o
p
o
u
t
(
0
.
4
)
-
D
r
o
p
o
u
t
(
0
.
4
)
-
D
r
o
p
o
u
t
(
0
.
4
)
-
D
e
n
se
(
1
)
S
i
g
mo
i
d
D
e
n
se
(
1
)
S
i
g
mo
i
d
D
e
n
se
(
1
)
S
i
g
mo
i
d
O
p
t
i
mi
z
e
r
S
G
D
(
l
e
a
r
n
i
n
g
_
r
a
t
e=
0
.
0
0
6
)
S
G
D
(
l
e
a
r
n
i
n
g
_
r
a
t
e
=
0
.
0
0
5
)
A
d
a
m
(
l
e
a
r
n
i
n
g
_
r
a
t
e
=
0
.
0
0
1
)
Fo
r
th
e
SVM
an
d
R
F
m
o
d
els
,
th
e
d
ef
au
lt
p
ar
a
m
eter
s
ar
e
ap
p
lied
.
A
lth
o
u
g
h
SV
M
an
d
R
F
ar
e
n
o
t
class
i
f
ied
as
d
ee
p
n
eu
r
al
n
e
t
w
o
r
k
tec
h
n
iq
u
es,
t
h
e
y
ar
e
co
m
m
o
n
l
y
u
s
ed
f
o
r
h
an
d
li
n
g
ti
m
e
s
er
ies
d
ata
an
d
b
in
ar
y
cla
s
s
i
f
icatio
n
m
o
d
el
s
[
3
1
]
–
[
3
4
]
.
T
h
e
p
r
o
ce
s
s
f
o
r
ac
h
i
ev
in
g
t
h
e
b
est
r
ain
f
all
d
etec
tio
n
f
o
r
ea
ch
m
o
d
el
i
s
illu
s
tr
ated
in
s
ta
g
e
2
o
f
F
i
g
u
r
e
5
.
W
e
tr
ain
th
e
1
DC
NN,
M
L
P
,
an
d
L
ST
M
m
o
d
els b
y
u
s
i
n
g
a
b
atch
s
ize
o
f
f
o
u
r
f
o
r
a
m
a
x
i
m
u
m
o
f
1
,
0
0
0
e
p
o
ch
s
,
w
i
th
v
a
lid
atio
n
lo
s
s
m
o
n
ito
r
in
g
b
ased
o
n
ea
r
ly
s
to
p
p
in
g
.
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
f
o
r
ea
ch
m
o
d
el
in
v
o
lv
es
n
u
m
er
o
u
s
iter
atio
n
s
,
w
h
er
ein
th
e
tr
ain
in
g
d
a
ta
(
s
p
ec
if
ical
l
y
,
th
e
s
elec
tio
n
o
f
d
r
y
s
a
m
p
le
s
)
is
r
a
n
d
o
m
ized
i
n
v
ar
io
u
s
w
a
y
s
.
T
h
is
r
a
n
d
o
m
iza
tio
n
is
p
er
f
o
r
m
ed
r
ep
ea
ted
l
y
u
n
til t
h
e
b
es
t r
es
u
lts
ar
e
o
b
tain
ed
.
A
ll th
e
m
o
d
els a
r
e
ev
alu
ated
w
it
h
t
h
e
s
a
m
e
s
et
o
f
test
d
ata.
2
.
4
.
M
o
del
ev
a
lua
t
i
o
n
T
h
e
f
in
a
l
s
ta
g
e
o
f
Fi
g
u
r
e
5
i
n
v
o
l
v
ed
t
h
e
ev
a
lu
at
io
n
o
f
ea
ch
m
o
d
el
in
th
i
s
s
tu
d
y
.
T
h
e
lo
n
g
-
ter
m
ev
alu
a
tio
n
o
f
ea
c
h
clas
s
i
f
ica
tio
n
m
o
d
el
y
ie
ld
s
p
r
ed
ictio
n
s
th
at
ar
e
s
u
b
s
eq
u
e
n
tl
y
co
m
p
ar
ed
w
it
h
ac
t
u
al
r
ef
er
en
ce
p
o
in
t
s
.
T
h
e
o
u
tco
m
e
is
a
co
n
f
u
s
io
n
m
atr
i
x
co
m
p
r
is
in
g
tr
u
e
p
o
s
it
iv
e
s
(
T
P
)
,
f
alse p
o
s
itiv
e
s
(
FP
)
,
f
alse
n
eg
at
iv
e
s
(
FN)
,
an
d
tr
u
e
n
eg
ativ
es
(
T
N)
,
as
illu
s
tr
ated
in
T
ab
le
2
.
Mo
r
e
o
v
er
,
th
e
o
u
tco
m
es
o
f
th
e
co
n
f
u
s
io
n
m
atr
i
x
ar
e
u
tili
ze
d
to
ca
lc
u
lat
e
ev
al
u
atio
n
m
e
tr
ics
s
u
c
h
a
s
t
h
e
tr
u
e
p
o
s
iti
v
e
r
ate
(
T
P
R
)
,
th
e
tr
u
e
n
eg
a
tiv
e
r
ate
(
T
NR
)
,
th
e
o
v
er
all
ac
cu
r
ac
y
,
an
d
th
e
F1
s
co
r
e
f
o
r
p
r
ed
ictin
g
r
ain
,
as d
ef
i
n
ed
in
T
ab
le
3
.
T
ab
le
2
.
C
o
n
f
u
s
io
n
m
atr
i
x
o
f
b
in
ar
y
clas
s
i
f
icatio
n
m
o
d
els
P
r
e
d
i
c
t
i
o
n
D
r
y
R
a
i
n
A
c
t
u
a
l
D
r
y
T
N
:
p
r
e
d
i
c
t
e
d
d
r
y
,
a
c
t
u
a
l
d
r
y
F
P
:
p
r
e
d
i
c
t
e
d
r
a
i
n
,
a
c
t
u
a
l
d
r
y
R
a
i
n
F
N
:
p
r
e
d
i
c
t
e
d
d
r
y
,
a
c
t
u
a
l
r
a
i
n
T
P
:
p
r
e
d
i
c
t
e
d
r
a
i
n
,
a
c
t
u
a
l
r
a
i
n
T
ab
le
3
.
E
v
alu
atio
n
m
etr
ics eq
u
atio
n
s
M
e
t
r
i
c
s
Eq
u
a
t
i
o
n
s
T
P
R
TP
T
P
+
F
N
T
N
R
TN
T
N
+
F
P
A
c
c
u
r
a
c
y
TP +
T
N
T
P
+
T
N
+
F
P
+
F
N
F
1
sco
r
e
2
T
P
2
T
P
+
F
P
+
F
N
I
n
ad
d
itio
n
to
th
e
af
o
r
e
m
e
n
ti
o
n
ed
ev
alu
at
io
n
m
etr
ics,
th
e
r
ec
eiv
er
o
p
er
atin
g
ch
ar
ac
ter
is
tic
(
R
OC
)
cu
r
v
e
an
d
th
e
ar
ea
u
n
d
er
th
e
cu
r
v
e
(
AUC)
g
r
ap
h
s
ar
e
also
u
tili
ze
d
to
d
ep
ict
th
e
p
er
f
o
r
m
an
ce
o
f
in
d
iv
id
u
al
b
in
ar
y
cla
s
s
i
f
icat
io
n
m
o
d
els
[
3
5
]
.
T
h
e
ev
al
u
atio
n
en
co
m
p
a
s
s
es
all
o
n
e
-
y
ea
r
te
s
t
d
ata
a
s
th
e
R
O
C
c
u
r
v
e
is
u
n
a
f
f
ec
ted
b
y
th
e
r
atio
o
f
d
r
y
a
n
d
r
ain
p
er
io
d
s
[
2
9
]
.
A
p
er
f
ec
t
class
i
f
ier
w
ill
b
e
lo
ca
ted
at
th
e
u
p
p
er
-
lef
t
co
r
n
er
o
f
th
e
R
O
C
s
p
ac
e,
w
h
i
ch
is
at
t
h
e
co
o
r
d
in
ates
(
0
,
1
)
.
I
n
t
h
e
co
n
te
x
t
o
f
clas
s
i
f
icatio
n
,
th
e
A
U
C
m
etr
ic
in
d
icate
s
th
e
li
k
eli
h
o
o
d
o
f
a
class
if
ier
co
r
r
ec
tl
y
r
an
k
i
n
g
a
r
an
d
o
m
l
y
ch
o
s
e
n
p
o
s
itiv
e
in
s
tan
ce
ab
o
v
e
a
r
an
d
o
m
l
y
c
h
o
s
e
n
n
e
g
at
i
v
e
in
s
t
an
ce
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
5
,
Octo
b
e
r
20
26
:
1
6
3
2
-
1643
1638
As
a
f
i
n
al
ev
al
u
atio
n
,
t
h
e
m
o
d
els
’
co
n
s
is
te
n
c
y
o
f
p
er
f
o
r
m
a
n
ce
in
d
etec
ti
n
g
d
iv
er
s
e
r
ain
f
all
p
atter
n
s
is
ev
al
u
ated
o
n
a
s
ea
s
o
n
al
b
as
is
,
in
ac
co
r
d
an
ce
w
it
h
th
e
cli
m
ato
lo
g
y
o
f
Natu
n
a
I
s
la
n
d
[
3
6
]
,
[
3
7
]
.
T
h
e
p
er
io
d
s
ex
a
m
in
ed
ar
e
Dec
e
m
b
er
to
Fe
b
r
u
ar
y
(
DJ
F),
Ma
r
ch
to
Ma
y
(
MA
M)
,
J
u
n
e
to
Au
g
u
s
t
(
J
J
A
)
,
an
d
Sep
tem
b
er
to
No
v
e
m
b
er
(
SON)
[
3
8
]
.
T
h
is
ev
alu
a
tio
n
ai
m
s
to
ex
a
m
i
n
e
t
h
e
co
n
s
is
ten
c
y
o
f
t
h
e
m
o
d
els
’
p
er
f
o
r
m
an
ce
s
ac
r
o
s
s
a
v
ar
iet
y
o
f
r
ai
n
f
all
p
atter
n
s
in
o
n
e
y
ea
r
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
W
e
ev
alu
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
all
th
e
m
o
d
els
b
y
u
s
i
n
g
R
R
d
ata
s
a
m
p
les
f
r
o
m
J
u
n
e
2
0
2
2
to
Ma
y
2
0
2
3
,
as
s
h
o
w
n
in
Fi
g
u
r
e
6
.
T
h
ese
s
a
m
p
les
h
a
v
e
v
ar
io
u
s
a
m
o
u
n
ts
a
n
d
in
ten
s
itie
s
o
f
r
ain
ev
e
n
ts
f
o
r
ea
ch
s
ea
s
o
n
.
Fi
g
u
r
e
6
(
a
)
p
r
esen
ts
2
,
0
2
9
s
am
p
les
o
f
1
3
9
r
ain
ev
en
ts
w
ith
a
m
ax
i
m
u
m
in
te
n
s
it
y
o
f
3
9
m
m
/
h
,
w
h
ic
h
w
er
e
co
llected
b
et
w
ee
n
J
u
n
e
an
d
Au
g
u
s
t
2
0
2
2
.
Fig
u
r
e
6
(
b
)
d
ep
icts
2
,
0
4
3
s
am
p
les
o
f
2
1
8
r
ain
ev
en
ts
w
it
h
a
m
ax
i
m
u
m
i
n
te
n
s
it
y
o
f
3
3
.
7
m
m
/
h
f
r
o
m
Sep
te
m
b
er
to
No
v
e
m
b
er
2
0
2
2
.
Fig
u
r
e
6
(
c
)
s
h
o
ws
2
,
0
7
0
s
am
p
les
o
f
3
1
4
r
ain
ev
en
ts
w
i
th
a
m
ax
i
m
u
m
i
n
te
n
s
it
y
o
f
3
0
m
m
/
h
f
r
o
m
Dec
e
m
b
er
2
0
2
2
to
Feb
r
u
ar
y
2
0
2
3
.
Fig
u
r
e
6
(
d
)
p
r
esen
ts
2
,
0
8
0
s
a
m
p
le
s
o
f
1
0
9
r
ain
ev
e
n
t
s
w
it
h
a
m
a
x
i
m
u
m
i
n
te
n
s
it
y
o
f
3
5
m
m
/
h
f
r
o
m
Ma
r
ch
to
Ma
y
2
0
2
3
.
T
h
er
e
ar
e
a
t
o
tal
o
f
7
8
0
r
ain
ev
en
t
s
i
n
th
e
8
,
2
2
2
s
a
m
p
les u
s
e
d
to
ev
alu
ate
th
e
m
o
d
els.
(
a)
(
b
)
(
c)
(
d
)
Fig
u
r
e
6
.
Sa
m
p
les o
f
r
ai
n
f
all
i
n
ten
s
it
y
f
o
r
lo
n
g
-
ter
m
m
o
d
el
ev
alu
a
tio
n
;
(
a)
J
u
n
e
–
Au
g
u
s
t
2
0
2
2
,
(
b
)
Sep
tem
b
er
–
No
v
e
m
b
er
2
0
2
2
,
(
c)
Dec
em
b
er
2022
–
Feb
r
u
ar
y
2
0
2
3
,
an
d
(
d
)
Ma
r
ch
–
Ma
y
2
0
2
3
T
h
e
ev
alu
atio
n
o
f
th
e
b
in
ar
y
class
if
icatio
n
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
i
s
in
itial
l
y
as
s
ess
ed
th
r
o
u
g
h
t
h
e
u
tili
za
t
io
n
o
f
th
e
R
OC
c
u
r
v
e,
w
h
ich
i
n
co
r
p
o
r
ates
v
ar
io
u
s
th
r
es
h
o
ld
v
alu
e
s
.
T
h
is
is
illu
s
t
r
ated
in
Fig
u
r
e
7
.
R
OC
g
r
ap
h
s
ar
e
g
r
ap
h
ical
r
ep
r
esen
tatio
n
s
th
a
t
d
ep
ict
th
e
r
el
atio
n
s
h
ip
b
et
w
ee
n
th
e
T
P
R
an
d
th
e
f
alse
p
o
s
itiv
e
r
ate
(
FP
R
)
.
T
h
ese
g
r
ap
h
s
ar
e
t
w
o
-
d
i
m
e
n
s
io
n
al,
w
it
h
t
h
e
T
PR
r
ep
r
esen
ted
o
n
t
h
e
y
-
a
x
is
a
n
d
th
e
FP
R
p
lo
tted
o
n
th
e
x
-
a
x
i
s
.
T
h
e
T
P
R
,
also
k
n
o
w
n
a
s
t
h
e
h
it
r
ate
o
r
r
ec
all
r
ate,
r
ep
r
esen
ts
th
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
tl
y
p
r
ed
icted
o
cc
u
r
r
en
ce
s
o
f
r
ain
in
r
elatio
n
to
th
e
o
v
er
all
n
u
m
b
er
o
f
r
ain
y
e
v
e
n
ts
.
C
o
n
v
er
s
e
l
y
,
t
h
e
FP
R
,
o
f
te
n
r
ef
er
r
ed
to
as
th
e
f
al
s
e
alar
m
r
ate,
in
d
icate
s
th
e
r
atio
o
f
i
n
co
r
r
ec
tl
y
p
r
ed
icted
d
r
y
ev
e
n
ts
to
th
e
to
tal
d
r
y
a
m
o
u
n
t.
T
h
e
A
U
C
d
escr
ib
es
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
ac
r
o
s
s
a
ll
th
e
p
o
s
s
ib
le
clas
s
i
f
icatio
n
t
h
r
esh
o
ld
s
.
W
ith
t
h
e
ex
ce
p
tio
n
o
f
th
e
SVM
m
o
d
el,
all
th
e
m
o
d
els
d
e
m
o
n
s
tr
ate
e
x
ce
lle
n
t
p
er
f
o
r
m
a
n
ce
a
s
b
in
ar
y
clas
s
i
f
ier
s
i
n
r
ai
n
d
etec
tio
n
.
T
h
is
is
ev
id
en
t
f
r
o
m
t
h
e
co
n
v
e
x
s
h
ap
e
o
f
th
e
R
OC
g
r
ap
h
s
an
d
th
e
A
U
C
v
alu
es,
w
h
ich
ex
ce
ed
0
.
8
[
3
9
]
,
[
4
0
]
.
T
h
e
A
UC
v
al
u
e
s
f
o
r
th
e
1
DC
NN,
L
ST
M,
R
F,
ML
P
,
an
d
SVM
ar
e
0
.
8
4
,
0
.
8
4
,
0
.
8
3
,
0
.
8
2
,
an
d
0
.
7
9
,
r
esp
ec
tiv
el
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
R
a
in
fa
ll d
etec
tio
n
o
n
a
tr
o
p
ica
l isl
a
n
d
u
s
in
g
a
c
o
mmu
n
ica
tio
n
s
a
tellite
… (
R
a
d
en
Yu
d
h
a
Ma
r
d
ya
n
s
ya
h
)
1639
Fig
u
r
e
7
.
T
h
e
R
OC
-
AUC
g
r
ap
h
o
f
m
o
d
el
p
er
f
o
r
m
a
n
ce
T
h
e
ev
alu
atio
n
o
f
th
e
b
in
ar
y
class
if
icatio
n
m
o
d
el
’
s
p
er
f
o
r
m
an
ce
i
s
in
itial
l
y
as
s
ess
ed
th
r
o
u
g
h
t
h
e
u
tili
za
t
io
n
o
f
th
e
R
OC
c
u
r
v
e,
w
h
ich
i
n
co
r
p
o
r
ates
v
ar
io
u
s
th
r
es
h
o
ld
v
alu
e
s
.
T
h
is
is
illu
s
t
r
ated
in
Fig
u
r
e
7
.
R
OC
g
r
ap
h
s
ar
e
g
r
ap
h
ical
r
ep
r
esen
tatio
n
s
t
h
at
d
ep
ict
t
h
e
r
elatio
n
s
h
ip
b
et
w
ee
n
th
e
T
P
R
an
d
th
e
FP
R
.
T
h
ese
g
r
ap
h
s
ar
e
t
w
o
-
d
i
m
e
n
s
io
n
al,
w
it
h
t
h
e
T
P
R
r
ep
r
esen
ted
o
n
th
e
y
-
a
x
is
a
n
d
th
e
FP
R
p
lo
tted
o
n
th
e
x
-
a
x
i
s
.
T
h
e
T
P
R
,
also
k
n
o
w
n
as
t
h
e
h
it
r
a
te
o
r
r
ec
all
r
ate,
r
ep
r
esen
ts
th
e
p
r
o
p
o
r
tio
n
o
f
co
r
r
ec
tly
p
r
ed
ic
ted
o
cc
u
r
r
en
ce
s
o
f
r
ain
in
r
elatio
n
to
th
e
o
v
er
all
n
u
m
b
er
o
f
r
ai
n
y
e
v
e
n
ts
.
C
o
n
v
er
s
el
y
,
t
h
e
FP
R
,
o
f
te
n
r
ef
er
r
ed
to
as th
e
f
al
s
e
alar
m
r
at
e,
in
d
icate
s
th
e
r
atio
o
f
i
n
co
r
r
ec
tly
p
r
ed
icted
d
r
y
ev
e
n
t
s
to
th
e
to
tal
d
r
y
a
m
o
u
n
t.
T
h
e
A
UC
d
escr
ib
es
p
r
ed
ictiv
e
p
er
f
o
r
m
an
ce
ac
r
o
s
s
all
t
h
e
p
o
s
s
ib
le
clas
s
i
f
icati
o
n
th
r
es
h
o
ld
s
.
W
ith
th
e
e
x
ce
p
tio
n
o
f
t
h
e
SVM
m
o
d
el,
all
th
e
m
o
d
els
d
e
m
o
n
s
tr
ate
ex
ce
l
len
t
p
er
f
o
r
m
a
n
ce
as
b
in
ar
y
clas
s
if
ier
s
in
r
ain
d
etec
tio
n
.
T
h
is
is
ev
id
en
t
f
r
o
m
t
h
e
co
n
v
ex
s
h
ap
e
o
f
th
e
R
OC
g
r
ap
h
s
a
n
d
th
e
A
U
C
v
a
lu
e
s
,
w
h
ic
h
e
x
ce
ed
0
.
8
[
3
9
]
,
[
4
0
]
.
T
h
e
A
U
C
v
a
lu
e
s
f
o
r
th
e
1
D
C
NN,
L
ST
M,
R
F,
M
L
P
,
an
d
SVM
ar
e
0
.
8
4
,
0
.
8
4
,
0
.
8
3
,
0
.
8
2
,
an
d
0
.
7
9
,
r
esp
ec
tiv
ely
.
I
n
o
r
d
er
to
ev
alu
ate
th
e
T
P
R
,
T
NR
,
ac
cu
r
ac
y
,
a
n
d
F1
s
co
r
e,
it
is
r
eq
u
ir
ed
to
b
alan
ce
th
e
test
d
ata
b
y
r
an
d
o
m
l
y
r
ed
u
ci
n
g
th
e
n
u
m
b
er
o
f
d
r
y
s
a
m
p
le
s
to
m
atc
h
t
h
e
n
u
m
b
er
o
f
r
ain
s
a
m
p
les.
Fi
g
u
r
e
8
p
r
ese
n
ts
t
h
e
lo
n
g
-
ter
m
e
v
alu
at
io
n
m
etr
ic
s
f
o
r
all
t
h
e
m
o
d
el
s
w
it
h
a
th
r
e
s
h
o
ld
o
f
0
.
5
.
Ov
er
all,
al
l
t
h
e
m
o
d
els
d
eliv
er
m
o
r
e
ac
cu
r
ate
p
r
ed
ictio
n
s
o
f
d
r
y
c
o
n
d
itio
n
s
t
h
an
r
ai
n
co
n
d
itio
n
s
,
as
in
d
icate
d
b
y
t
h
e
f
ac
t
t
h
at
th
e
T
NR
v
alu
es
ex
ce
ed
th
e
T
P
R
o
n
es.
T
h
is
r
e
s
u
lt
is
i
n
lin
e
w
it
h
r
esear
ch
co
n
d
u
cted
b
y
P
o
lz
et
a
l.
[
2
9
]
.
T
h
r
ee
m
o
d
els
;
th
e
1
DC
NN,
L
ST
M,
an
d
R
F
ca
n
d
etec
t
r
ain
b
etter
th
an
th
e
r
est
ac
co
r
d
in
g
to
th
e
T
P
R
v
alu
es.
On
av
er
a
g
e,
2
9
%
o
f
r
ain
f
al
l
ev
e
n
ts
ar
e
i
n
co
r
r
ec
tl
y
d
etec
ted
b
y
th
e
L
ST
M,
3
2
%
b
y
t
h
e
1
DC
NN,
an
d
3
3
%
b
y
t
h
e
R
F,
w
h
er
ea
s
t
h
e
ML
P
an
d
SVM
i
n
co
r
r
ec
tl
y
d
etec
t 4
7
% a
n
d
5
2
% o
f
r
ain
ev
en
ts
,
r
esp
ec
tiv
el
y
.
Fig
u
r
e
8
.
E
v
alu
atio
n
m
e
tr
ics
f
o
r
all
th
e
m
o
d
els
A
cc
u
r
ac
y
a
n
d
F1
s
co
r
e
ar
e
f
u
r
th
er
asp
ec
t
s
o
f
ev
a
lu
atio
n
.
I
n
ter
m
s
o
f
ac
cu
r
ac
y
p
er
f
o
r
m
an
c
e,
t
h
e
1
DC
NN,
R
F,
a
n
d
L
ST
M
m
o
d
els
o
u
tp
er
f
o
r
m
t
h
e
M
L
P
a
n
d
SVM
w
it
h
a
n
ac
c
u
r
ac
y
r
an
g
e
o
f
0
.
7
7
–
0
.
7
8
,
w
h
ic
h
is
h
ig
h
er
t
h
a
n
t
h
e
v
al
u
es
f
o
r
th
e
M
L
P
(
0
.
7
3
)
an
d
th
e
SV
M
(
0
.
7
1
)
.
C
alcu
latin
g
th
e
w
e
ig
h
ted
av
er
a
g
e
o
f
a
m
o
d
el
’
s
ac
cu
r
ac
y
(
also
k
n
o
w
n
as
p
o
s
itiv
e
p
r
ed
icti
v
e
v
al
u
e)
a
n
d
r
ec
all
(
i.e
.
,
s
en
s
iti
v
it
y
)
y
iel
d
s
th
e
F1
s
co
r
e.
I
n
o
th
er
w
o
r
d
s
,
th
e
F1
s
co
r
e
tak
es
i
n
to
ac
co
u
n
t
FP
(
ac
tu
al
d
r
y
b
u
t
p
r
ed
icted
r
ain
)
an
d
FN
(
ac
tu
al
r
ai
n
b
u
t
p
r
ed
icted
d
r
y
)
an
d
m
o
r
e
ac
cu
r
atel
y
r
ef
lects
th
e
p
er
f
o
r
m
a
n
ce
o
f
a
b
in
ar
y
clas
s
i
f
icatio
n
m
o
d
el.
Ag
ai
n
,
th
e
1
DC
NN,
L
ST
M,
an
d
R
F
m
o
d
els
o
u
tp
er
f
o
r
m
t
h
e
o
th
er
m
o
d
els
i
n
d
etec
ti
n
g
r
ai
n
,
w
it
h
F1
s
co
r
es
r
an
g
i
n
g
f
r
o
m
0
.
7
5
to
0
.
7
6
,
w
h
er
ea
s
t
h
e
F1
s
co
r
es o
f
th
e
M
L
P
an
d
th
e
SV
M
ar
e
0
.
6
6
an
d
0
.
6
2
,
r
esp
ec
tiv
el
y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
1
6
9
3
-
6930
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
,
Vo
l.
24
,
No
.
5
,
Octo
b
e
r
20
26
:
1
6
3
2
-
1643
1640
A
s
a
f
in
a
l
p
e
r
f
o
r
m
an
c
e
ev
a
lu
a
t
i
o
n
,
t
h
e
a
c
cu
r
a
cy
o
f
ea
ch
m
o
d
e
l
i
s
t
es
t
e
d
o
n
a
s
e
as
o
n
a
l
b
a
s
is
.
T
h
is
e
v
a
lu
at
i
o
n
i
s
c
o
n
d
u
ct
e
d
t
o
ex
am
in
e
th
e
c
o
n
s
is
t
en
cy
o
f
th
e
m
o
d
e
l
s
’
p
e
r
f
o
r
m
an
c
es
in
d
e
t
e
ct
in
g
d
iv
e
r
s
e
r
a
in
f
a
l
l
p
a
t
t
e
r
n
s
.
F
ig
u
r
e
9
s
h
o
w
s
th
e
p
e
r
f
o
r
m
an
c
e
s
o
f
th
e
ex
am
in
ed
m
o
d
e
ls
d
u
r
i
n
g
th
e
J
J
A
,
S
ON
,
D
J
F
,
an
d
MA
M
p
e
r
i
o
d
s
.
A
ll
th
e
m
o
d
e
ls
h
av
e
h
ig
h
e
r
p
e
r
f
o
r
m
an
c
e
s
in
MA
M
b
e
c
a
u
s
e
th
e
n
u
m
b
e
r
o
f
r
a
in
ev
en
ts
in
th
e
t
es
t
d
a
ta
f
o
r
t
h
i
s
p
e
r
i
o
d
i
s
th
e
l
o
w
es
t
co
m
p
a
r
e
d
t
o
t
h
e
o
th
e
r
p
e
r
i
o
d
s
.
A
c
c
o
r
d
in
g
t
o
t
h
e
r
es
u
l
ts
o
f
th
is
e
v
alu
a
t
i
o
n
,
th
e
1
D
C
NN
,
L
S
T
M
,
a
n
d
R
F
c
o
n
s
i
s
t
en
t
ly
o
u
t
p
e
r
f
o
r
m
th
e
ML
P
a
n
d
S
V
M
,
w
h
ich
h
av
e
b
e
en
em
p
l
o
y
e
d
in
p
r
e
v
i
o
u
s
s
tu
d
i
es
[
7
]
,
[
1
6
]
.
I
n
th
e
D
J
F
a
n
d
MA
M
p
e
r
i
o
d
s
,
w
h
e
r
e
th
e
n
u
m
b
e
r
o
f
r
ain
ev
en
ts
is
h
ig
h
es
t
an
d
l
o
w
es
t
,
r
e
s
p
e
c
t
i
v
e
l
y
,
t
h
e
1
D
C
N
N
o
u
t
p
e
r
f
o
r
m
s
t
h
e
L
S
T
M
a
n
d
R
F
w
i
t
h
a
c
c
u
r
a
c
y
v
a
l
u
e
s
o
f
0
.
7
6
i
n
D
J
F
a
n
d
0
.
8
2
i
n
M
A
M
.
I
n
g
e
n
er
al,
th
e
p
er
f
o
r
m
a
n
ce
o
f
1
DC
NN,
L
ST
M,
an
d
R
F,
wh
ich
w
e
p
r
o
p
o
s
e
as
b
in
ar
y
cla
s
s
i
f
icatio
n
m
o
d
el
s
f
o
r
r
ain
f
all
d
etec
tio
n
,
o
u
tp
er
f
o
r
m
t
h
e
M
L
P
an
d
S
VM
as
d
ep
icted
in
T
ab
le
4
.
T
h
e
f
in
d
i
n
g
s
o
f
th
is
s
tu
d
y
d
e
m
o
n
s
tr
ate
th
at
1
D
C
NN,
L
ST
M,
an
d
R
F
m
o
d
els
ca
n
ef
f
ec
ti
v
el
y
id
e
n
ti
f
y
r
ai
n
f
all
o
cc
u
r
r
en
ce
s
o
n
tr
o
p
ical
is
lan
d
s
.
T
h
e
R
F
m
o
d
el
h
as
a
d
etec
tio
n
p
r
o
b
ab
i
lit
y
o
f
8
3
%,
w
h
er
ea
s
b
o
th
1
D
C
NN
an
d
L
ST
M
m
o
d
els
h
av
e
a
d
etec
tio
n
p
r
o
b
ab
ilit
y
o
f
8
4
%
ea
ch
.
C
o
n
s
id
er
in
g
t
h
e
p
r
esen
ce
o
f
n
u
m
er
o
u
s
p
o
ten
tial
m
i
s
lab
eled
r
ain
s
a
m
p
les
i
n
th
e
d
ataset,
t
h
is
r
esu
lt
is
e
x
ce
lle
n
t.
T
h
is
d
is
cr
ep
an
c
y
is
attr
ib
u
ted
to
th
e
s
ig
n
i
f
ica
n
t
d
is
tan
c
e
b
et
w
ee
n
UT
an
d
R
G,
r
esu
ltin
g
in
a
lik
eli
h
o
o
d
o
f
s
ev
er
al
r
ain
o
cc
u
r
r
en
ce
s
t
h
at
d
if
f
er
b
etw
ee
n
th
e
t
w
o
s
ites
.
Fu
r
t
h
er
m
o
r
e,
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
R
F
is
r
eliab
le
a
n
d
co
m
p
e
titi
v
e
w
h
en
co
m
p
ar
ed
to
th
e
1
DC
N
N
an
d
t
h
e
L
ST
M,
d
esp
ite
th
e
f
ac
t t
h
at
d
e
f
au
lt p
ar
a
m
eter
s
ar
e
u
s
ed
.
Fig
u
r
e
9
.
Seaso
n
al
ac
cu
r
ac
y
p
er
f
o
r
m
an
ce
o
f
th
e
m
o
d
els
T
ab
le
4
.
C
o
m
p
ar
is
o
n
o
f
m
o
d
el
s
’
p
er
f
o
r
m
a
n
ce
M
o
d
e
l
A
U
C
A
c
c
u
r
a
c
y
F1
-
s
c
o
r
e
L
o
n
g
-
t
e
r
m
a
c
c
u
r
a
c
y
JJA
S
O
N
D
JF
M
A
M
1
D
C
N
N
0
.
8
4
0
.
7
8
0
.
7
6
0
.
7
8
0
.
7
8
0
.
7
6
0
.
8
2
L
S
T
M
0
.
8
4
0
.
7
7
0
.
7
6
0
.
7
8
0
.
7
8
0
.
7
5
0
.
8
1
RF
0
.
8
3
0
.
7
8
0
.
7
5
0
.
7
8
0
.
7
7
0
.
7
2
0
.
8
1
M
L
P
0
.
8
2
0
.
7
3
0
.
6
6
0
.
7
4
0
.
7
3
0
.
6
8
0
.
7
6
S
V
M
0
.
7
9
0
.
7
1
0
.
6
2
0
.
7
3
0
.
7
0
0
.
6
5
0
.
7
4
4.
CO
NCLU
SI
O
N
I
n
th
i
s
ar
ticle,
w
e
i
n
v
e
s
ti
g
ate
A
I
tec
h
n
iq
u
es
b
ased
o
n
b
i
n
ar
y
clas
s
i
f
icatio
n
to
d
etec
t
r
ai
n
f
r
o
m
K
u
-
b
an
d
co
m
m
u
n
ica
tio
n
s
atel
lite
d
o
w
n
l
in
k
s
i
g
n
a
ls
.
T
h
e
s
tu
d
y
a
r
ea
is
lo
ca
ted
o
n
th
e
I
n
d
o
n
esian
I
s
lan
d
o
f
Nat
u
n
a,
w
h
ic
h
is
in
th
e
tr
o
p
ical
eq
u
ato
r
ial
r
eg
io
n
.
Ou
r
r
esear
ch
d
em
o
n
s
tr
ate
s
th
at
s
atell
ite
d
o
w
n
lin
k
s
ig
n
a
ls
ca
n
d
etec
t
r
ain
f
al
l
o
n
a
tr
o
p
ical
is
lan
d
.
T
h
e
r
esu
lts
o
f
t
h
is
s
t
u
d
y
i
n
d
icate
th
at
1
D
C
NN,
L
ST
M,
an
d
R
F
m
o
d
els
ar
e
ca
p
ab
le
o
f
ac
cu
r
atel
y
d
etec
tin
g
r
ain
f
all
ev
e
n
t
s
,
w
i
th
a
d
etec
tio
n
r
ate
o
f
u
p
to
8
4
%.
Fu
r
th
er
m
o
r
e,
ea
ch
m
o
d
el
p
r
o
v
id
es
co
n
s
is
te
n
c
y
o
f
p
er
f
o
r
m
an
ce
w
it
h
b
etter
ac
cu
r
ac
y
i
n
d
etec
tin
g
th
e
r
eg
io
n
’
s
d
iv
er
s
e
r
ain
f
al
l
p
atter
n
s
th
r
o
u
g
h
o
u
t
t
h
e
y
ea
r
.
I
n
t
h
e
D
J
F
an
d
MA
M
p
er
io
d
s
,
w
h
e
n
t
h
e
n
u
m
b
er
o
f
r
ain
e
v
en
ts
is
h
ig
h
e
s
t
a
n
d
lo
w
e
s
t,
r
esp
ec
tiv
el
y
,
t
h
e
1
DC
NN
o
u
t
p
er
f
o
r
m
s
all
t
h
e
o
th
er
m
o
d
els
w
ith
ac
cu
r
ac
y
v
al
u
es
o
f
0
.
7
6
in
DJ
F
an
d
0
.
8
2
in
MA
M.
T
h
ese
f
i
n
d
i
n
g
s
co
n
f
ir
m
t
h
at
t
h
e
u
s
e
o
f
AI
m
o
d
el
s
f
o
r
r
ain
f
all
d
etec
tio
n
u
s
i
n
g
s
at
ellite
li
n
k
s
i
g
n
a
ls
i
s
p
r
o
m
i
s
in
g
f
o
r
tr
o
p
ical
r
eg
io
n
s
.
C
o
n
s
id
er
i
n
g
t
h
e
er
r
o
r
s
in
r
ai
n
-
dr
y
lab
elin
g
i
n
b
o
th
t
h
e
tr
ai
n
in
g
a
n
d
test
d
ata,
th
is
r
e
s
u
l
t
is
e
x
ce
lle
n
t
.
F
u
t
u
r
e
r
esear
ch
is
e
x
p
ec
ted
to
u
s
e
th
e
UT
an
d
R
G
at
t
h
e
s
a
m
e
lo
ca
t
io
n
to
i
m
p
r
o
v
e
t
h
e
m
o
d
el
s
’
ac
c
u
r
ac
y
.
I
n
ad
d
itio
n
,
th
e
u
ti
lizatio
n
o
f
h
ig
h
er
r
eso
lu
tio
n
r
ain
f
a
ll
in
te
n
s
it
y
d
ata
ca
n
b
e
a
g
r
ea
t
p
o
ten
tial f
o
r
r
ea
l
-
ti
m
e
r
ai
n
f
a
ll
m
o
n
ito
r
i
n
g
i
n
t
h
e
tr
o
p
ical
r
eg
io
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KOM
NI
K
A
T
elec
o
m
m
u
n
C
o
m
p
u
t E
l
C
o
n
tr
o
l
R
a
in
fa
ll d
etec
tio
n
o
n
a
tr
o
p
ica
l isl
a
n
d
u
s
in
g
a
c
o
mmu
n
ica
tio
n
s
a
tellite
… (
R
a
d
en
Yu
d
h
a
Ma
r
d
ya
n
s
ya
h
)
1641
ACK
NO
WL
E
D
G
E
M
E
NT
S
T
h
e
r
esear
ch
is
s
u
p
p
o
r
ted
b
y
B
ad
an
R
is
et
d
an
I
n
o
v
asi
Na
s
io
n
al
’
s
De
g
r
ee
b
y
R
e
s
ea
r
ch
p
r
o
g
r
am
.
B
ad
an
Me
teo
r
o
lo
g
i
Kli
m
ato
lo
g
i
d
an
Geo
f
is
i
k
a
d
an
P
T
.
T
el
k
o
m
s
at
I
n
d
o
n
e
s
ia
ar
e
also
th
a
n
k
ed
f
o
r
s
u
p
p
l
y
i
n
g
th
e
d
ata.
F
UNDIN
G
I
NF
O
RM
AT
I
O
N
T
h
e
au
th
o
r
s
w
is
h
to
ex
p
r
ess
th
eir
ap
p
r
ec
iatio
n
to
Kem
e
n
d
i
k
b
u
d
r
is
te
k
o
f
R
ep
u
b
li
k
I
n
d
o
n
esia
f
o
r
th
e
f
i
n
an
cia
l su
p
p
o
r
t o
f
Hib
ah
P
en
elitian
Di
s
er
tasi
Do
k
to
r
n
o
.
NKB
-
1
1
3
1
/UN2
.
R
ST
/HKP
.
0
5
.
0
0
/2
0
2
3
.
AUTHO
R
CO
NT
RIB
UT
I
O
NS ST
A
T
E
M
E
NT
T
h
is
j
o
u
r
n
al
u
s
e
s
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT
)
to
r
ec
o
g
n
ize
in
d
i
v
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
t
h
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
lla
b
o
r
atio
n
.
Na
m
e
o
f
Aut
ho
r
C
M
S
o
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
R
ad
en
Y
u
d
h
a
Ma
r
d
y
a
n
s
y
a
h
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
B
u
d
h
y
K
u
r
n
ia
w
an
✓
✓
✓
✓
✓
San
to
s
o
So
ek
ir
n
o
✓
✓
✓
✓
Dan
a
n
g
E
k
o
Nu
r
y
a
n
to
✓
✓
✓
✓
✓
Fer
r
y
B
u
d
h
i S
u
s
e
t
y
o
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
si
s
I
:
I
n
v
e
st
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
si
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
ter
est
.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
i
n
d
i
n
g
s
o
f
t
h
is
s
tu
d
y
ar
e
av
ailab
le
f
r
o
m
t
h
e
co
r
r
esp
o
n
d
in
g
a
u
t
h
o
r
,
R
.
Y.
M.
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est
.
RE
F
E
R
E
NC
E
S
[
1
]
R
.
T
.
H
i
t
c
h
c
o
c
k
,
“
R
a
d
i
o
‐
F
r
e
q
u
e
n
c
y
a
n
d
M
i
c
r
o
w
a
v
e
R
a
d
i
a
t
i
o
n
,
”
Pa
t
t
y
’
s
T
o
x
i
c
o
l
o
g
y
.
W
i
l
e
y
,
p
p
.
1
3
3
–
1
6
8
,
A
u
g
.
1
7
,
2
0
1
2
.
d
o
i
:
1
0
.
1
0
0
2
/
0
4
7
1
4
3
5
1
3
9
.
t
o
x
1
0
1
.
p
u
b
2
.
[
2
]
I
EEE,
“
I
EEE
S
t
a
n
d
a
r
d
L
e
t
t
e
r
D
e
si
g
n
a
t
i
o
n
s
f
o
r
R
a
d
a
r
-
F
r
e
q
u
e
n
c
y
B
a
n
d
s,
”
2
0
0
2
.
d
o
i
:
1
0
.
1
1
0
9
/
I
EEEST
D
.
2
0
0
3
.
9
4
2
2
4
.
[
3
]
G
.
J.
G
.
U
p
t
o
n
,
A
.
R
.
H
o
l
t
,
R
.
J.
C
u
m
mi
n
g
s,
A
.
R
.
R
a
h
i
m
i
,
a
n
d
J.
W
.
F
.
G
o
d
d
a
r
d
,
“
M
i
c
r
o
w
a
v
e
l
i
n
k
s:
T
h
e
f
u
t
u
r
e
f
o
r
u
r
b
a
n
r
a
i
n
f
a
l
l
me
a
su
r
e
me
n
t
?
,
”
A
t
m
o
s
p
h
e
r
i
c
Re
s
e
a
r
c
h
,
v
o
l
.
7
7
,
n
o
.
1
-
4
S
P
E
C
.
I
S
S
.
,
p
p
.
3
0
0
–
3
1
2
,
2
0
0
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
t
mo
sr
e
s.2
0
0
4
.
1
0
.
0
0
9
.
[
4
]
R
.
O
l
se
n
,
D
.
R
o
g
e
r
s,
a
n
d
D
.
H
o
d
g
e
,
“
T
h
e
a
R
b
r
e
l
a
t
i
o
n
i
n
t
h
e
c
a
l
c
u
l
a
t
i
o
n
o
f
r
a
i
n
a
t
t
e
n
u
a
t
i
o
n
,
”
I
E
EE
T
ra
n
s
a
c
t
i
o
n
s
o
n
A
n
t
e
n
n
a
s
a
n
d
Pro
p
a
g
a
t
i
o
n
,
v
o
l
.
2
6
,
n
o
.
2
,
p
p
.
3
1
8
–
3
2
9
,
1
9
7
8
,
d
o
i
:
1
0
.
1
1
0
9
/
T
A
P
.
1
9
7
8
.
1
1
4
1
8
4
5
.
[
5
]
A
.
H
.
M
a
r
b
l
e
st
o
n
e
,
G
.
W
a
y
n
e
,
a
n
d
K
.
P
.
K
o
r
d
i
n
g
,
“
T
o
w
a
r
d
a
n
I
n
t
e
g
r
a
t
i
o
n
o
f
D
e
e
p
L
e
a
r
n
i
n
g
a
n
d
N
e
u
r
o
sci
e
n
c
e
,
”
Fr
o
n
t
i
e
rs
i
n
C
o
m
p
u
t
a
t
i
o
n
a
l
N
e
u
r
o
sc
i
e
n
c
e
,
v
o
l
.
1
0
,
2
0
1
6
,
d
o
i
:
1
0
.
3
3
8
9
/
f
n
c
o
m.
2
0
1
6
.
0
0
0
9
4
.
[
6
]
E.
A
d
i
r
o
si
e
t
a
l
.
,
“
Ex
p
l
o
i
t
i
n
g
sa
t
e
l
l
i
t
e
K
a
a
n
d
K
u
l
i
n
k
s
f
o
r
t
h
e
r
e
a
l
-
t
i
me
e
st
i
mat
i
o
n
o
f
r
a
i
n
i
n
t
e
n
si
t
y
,
”
i
n
2
0
1
7
3
2
n
d
G
e
n
e
r
a
l
Assemb
l
y
a
n
d
S
c
i
e
n
t
i
f
i
c
S
y
m
p
o
si
u
m
o
f
t
h
e
I
n
t
e
rn
a
t
i
o
n
a
l
U
n
i
o
n
o
f
R
a
d
i
o
S
c
i
e
n
c
e
,
U
R
S
I
G
A
S
S
2
0
1
7
,
2
0
1
7
,
v
o
l
.
2
0
1
7
-
J
a
n
u
a
,
p
p
.
1
–
4
,
d
o
i
:
1
0
.
2
3
9
1
9
/
U
R
S
I
G
A
S
S
.
2
0
1
7
.
8
1
0
5
1
2
8
.
[
7
]
A
.
G
h
a
r
a
n
j
i
k
,
M
.
R
.
B
h
a
v
a
n
i
S
h
a
n
k
a
r
,
F
.
Z
i
mm
e
r
,
a
n
d
B
.
O
t
t
e
r
st
e
n
,
“
C
e
n
t
r
a
l
i
z
e
d
r
a
i
n
f
a
l
l
e
st
i
mat
i
o
n
u
s
i
n
g
c
a
r
r
i
e
r
t
o
n
o
i
se
o
f
sat
e
l
l
i
t
e
c
o
mm
u
n
i
c
a
t
i
o
n
l
i
n
k
s
,
”
I
EEE
J
o
u
r
n
a
l
o
n
S
e
l
e
c
t
e
d
A
re
a
s
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
3
6
,
n
o
.
5
,
p
p
.
1
0
6
5
–
1
0
7
3
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
JS
A
C
.
2
0
1
8
.
2
8
3
2
7
9
8
.
[
8
]
A
.
G
h
a
r
a
n
j
i
k
,
K
.
V
.
M
i
s
h
r
a
,
B
.
S
h
a
n
k
a
r
,
a
n
d
B
.
O
t
t
e
r
st
e
n
,
“
L
e
a
r
n
i
n
g
-
b
a
se
d
r
a
i
n
f
a
l
l
e
st
i
ma
t
i
o
n
v
i
a
c
o
mm
u
n
i
c
a
t
i
o
n
sa
t
e
l
l
i
t
e
l
i
n
k
s
,
”
i
n
2
0
1
8
I
EEE
S
t
a
t
i
s
t
i
c
a
l
S
i
g
n
a
l
Pr
o
c
e
ssi
n
g
Wo
r
k
s
h
o
p
,
S
S
P
2
0
1
8
,
2
0
1
8
,
p
p
.
1
1
5
–
1
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
S
S
P
.
2
0
1
8
.
8
4
5
0
7
2
6
.
[
9
]
L
.
X
u
,
X
.
S
h
e
n
,
D
.
D
.
H
u
a
n
g
,
X
.
F
e
n
g
,
a
n
d
W
.
W
a
n
g
,
“
T
o
mo
g
r
a
p
h
i
c
r
e
c
o
n
s
t
r
u
c
t
i
o
n
o
f
r
a
i
n
f
a
l
l
f
i
e
l
d
s
u
si
n
g
s
a
t
e
l
l
i
t
e
c
o
mm
u
n
i
c
a
t
i
o
n
l
i
n
k
s
,
”
i
n
2
0
1
7
2
3
r
d
As
i
a
-
P
a
c
i
f
i
c
C
o
n
f
e
re
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
(
AP
C
C
)
,
2
0
1
7
,
p
p
.
1
–
5
,
d
o
i
:
1
0
.
2
3
9
1
9
/
A
P
C
C
.
2
0
1
7
.
8
3
0
3
9
8
6
.
[
1
0
]
T
.
R
o
ssi
,
M
.
D
e
S
a
n
c
t
i
s,
S
.
D
i
D
o
me
n
i
c
o
,
M
.
R
u
g
g
i
e
r
i
,
a
n
d
E.
C
i
a
n
c
a
,
“
Ex
p
e
r
i
me
n
t
a
l
M
o
d
e
l
o
f
R
a
i
n
f
a
l
l
R
a
t
e
Es
t
i
mat
i
o
n
T
h
r
o
u
g
h
t
h
e
O
p
p
o
r
t
u
n
i
s
t
i
c
U
se
o
f
Q
-
/V
-
B
a
n
d
S
a
t
e
l
l
i
t
e
L
i
n
k
s
,
”
I
EEE
T
r
a
n
s
a
c
t
i
o
n
s o
n
Ae
r
o
sp
a
c
e
a
n
d
E
l
e
c
t
r
o
n
i
c
S
y
s
t
e
m
s
,
v
o
l
.
5
8
,
n
o
.
5
,
p
p
.
3
8
0
2
–
3
8
0
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
T
A
ES.2
0
2
2
.
3
1
6
3
1
2
8
.
[
1
1
]
E.
A
d
i
r
o
si
e
t
a
l
.
,
“
Ev
a
l
u
a
t
i
o
n
o
f
r
a
i
n
f
a
l
l
Ee
st
i
mat
i
o
n
d
e
r
i
v
e
d
f
r
o
m
c
o
mm
e
r
c
i
a
l
i
n
t
e
r
a
c
t
i
v
e
D
V
B
r
e
c
e
i
v
e
r
s
u
si
n
g
d
i
s
d
r
o
me
t
e
r
,
r
a
i
n
Evaluation Warning : The document was created with Spire.PDF for Python.