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
.
3
,
J
u
n
e
20
26
,
p
p
.
9
26
~
9
32
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
3
.
27402
926
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//jo
u
r
n
a
l.u
a
d
.
a
c.
id
/in
d
ex
.
p
h
p
/TELK
OM
N
I
K
A
Lev
erag
ing
artificial intelligenc
e f
o
r det
e
ction o
f
den
ia
l
-
of
-
serv
ice att
a
c
k
s i
n
5
G
net
w
o
rk
envir
o
n
m
en
ts
B
a
s
ee
l A
l
-
Ali
,
M
ina
M
a
lek
za
deh
F
a
c
u
l
t
y
o
f
El
e
c
t
r
i
c
a
l
a
n
d
C
o
mp
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
,
H
a
k
i
m S
a
b
z
e
v
a
r
i
U
n
i
v
e
r
si
t
y
,
S
a
b
z
e
v
a
r
,
I
r
a
n
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
J
u
l
15
,
2025
R
ev
i
s
ed
Mar
27
,
2
0
2
6
A
cc
ep
ted
Mar
29
,
2
0
2
6
T
h
is
re
se
a
r
c
h
in
tro
d
u
c
e
s
a
n
e
v
a
l
u
a
ti
o
n
m
e
th
o
d
o
l
o
g
y
th
a
t
a
d
d
re
ss
e
s
th
e
d
a
ta
lea
k
a
g
e
p
ro
b
lem
f
o
r
d
e
tec
ti
n
g
d
e
n
ial
-
of
-
se
rv
ic
e
a
tt
a
c
k
s
in
f
i
f
th
-
g
e
n
e
ra
ti
o
n
(5
G
)
n
e
tw
o
rk
slicin
g
e
n
v
iro
n
m
e
n
ts,
a
n
d
a
p
p
li
e
s
it
t
o
p
e
rf
o
rm
a
b
e
n
c
h
m
a
r
k
c
o
m
p
a
riso
n
a
m
o
n
g
twe
lv
e
m
a
c
h
i
n
e
lea
rn
in
g
(M
L
)
,
d
e
e
p
lea
rn
in
g
(DL
)
,
a
n
d
p
ro
b
a
b
il
isti
c
m
o
d
e
ls
u
si
n
g
a
p
u
b
li
c
ly
a
v
a
il
a
b
le
5
G
n
e
t
w
o
rk
slicin
g
d
a
tas
e
t
f
o
r
Do
S
/DDo
S
a
tt
a
c
k
s
.
T
h
is
m
e
th
o
d
o
l
o
g
y
strictl
y
e
n
f
o
rc
e
s
th
e
e
x
e
c
u
ti
o
n
o
f
a
ll
p
re
p
r
o
c
e
ss
in
g
ste
p
s
e
x
c
lu
siv
e
l
y
o
n
t
h
e
t
ra
in
i
n
g
d
a
ta,
w
h
e
re
f
e
a
tu
re
se
lec
ti
o
n
is
p
e
rf
o
rm
e
d
u
sin
g
t
h
e
m
u
tu
a
l
in
f
o
rm
a
ti
o
n
(M
I)
m
e
tri
c
,
v
a
lu
e
s
a
re
sta
n
d
a
rd
ise
d
v
ia
th
e
z
-
sc
o
re
m
e
th
o
d
,
a
n
d
sy
n
th
e
ti
c
sa
m
p
les
a
re
p
ro
d
u
c
e
d
th
ro
u
g
h
th
e
sy
n
th
e
ti
c
m
in
o
rit
y
o
v
e
rsa
m
p
li
n
g
tec
h
n
iq
u
e
(S
M
OT
E)
tec
h
n
iq
u
e
o
n
t
h
e
train
i
n
g
se
t
o
n
ly
,
w
it
h
MI
re
c
a
lcu
late
d
in
d
e
p
e
n
d
e
n
t
ly
w
it
h
in
e
a
c
h
c
ro
ss
-
v
a
li
d
a
ti
o
n
(C
V
)
c
y
c
le.
Nin
e
f
e
a
tu
re
s
o
u
t
o
f
e
ig
h
ty
-
f
o
u
r
w
e
r
e
re
tain
e
d
a
t
th
e
e
l
b
o
w
p
o
in
t
w
h
e
re
M
I
re
a
c
h
e
d
0
.
5
1
o
r
a
b
o
v
e
.
On
th
e
h
e
ld
-
o
u
t
tes
t
se
t
c
o
n
tain
i
n
g
a
p
p
r
o
x
im
a
t
e
l
y
e
i
g
h
ty
p
e
rc
e
n
t
b
e
n
ig
n
d
a
ta
a
n
d
tw
e
n
ty
p
e
rc
e
n
t
a
tt
a
c
k
d
a
ta,
th
e
c
o
n
v
o
lu
ti
o
n
a
l
n
e
u
ra
l
n
e
tw
o
rk
(CNN
)
m
o
d
e
l
a
c
h
iev
e
d
th
e
h
ig
h
e
st
F
1
v
a
lu
e
o
f
0
.
9
8
3
w
it
h
a
f
a
lse
d
isc
o
v
e
r
y
ra
te
o
f
0
.
0
2
7
,
w
h
il
e
th
e
ra
n
d
o
m
f
o
re
st
m
o
d
e
l
re
a
c
h
e
d
a
n
F
1
v
a
lu
e
o
f
0
.
9
6
8
a
t
a
c
o
n
sid
e
ra
b
ly
lo
we
r
c
o
m
p
u
tatio
n
a
l
c
o
st.
A
ll
re
su
lt
s rem
a
in
ti
e
d
to
t
h
is p
a
r
ti
c
u
lar d
a
tas
e
t,
a
n
d
t
h
e
ir
g
e
n
e
ra
li
sa
b
il
it
y
to
re
a
l
-
w
o
rld
5G
n
e
tw
o
rk
tra
ff
ic h
a
s n
o
t
y
e
t
b
e
e
n
v
a
li
d
a
ted
.
K
ey
w
o
r
d
s
:
Dee
p
lear
n
in
g
Den
ial
o
f
s
er
v
ice
d
etec
tio
n
Fif
t
h
-
g
en
er
atio
n
s
ec
u
r
i
t
y
I
n
tr
u
s
io
n
d
etec
tio
n
s
y
s
te
m
Ma
ch
i
n
e
lear
n
i
n
g
Mu
t
u
al
in
f
o
r
m
a
tio
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
:
Min
a
Ma
le
k
za
d
eh
Facu
lt
y
o
f
E
lectr
ical
a
n
d
C
o
m
p
u
ter
E
n
g
i
n
ee
r
i
n
g
,
Ha
k
i
m
Sab
ze
v
ar
i U
n
i
v
er
s
i
t
y
R
az
av
i K
h
o
r
asa
n
P
r
o
v
in
ce
,
Sa
b
ze
v
ar
,
T
o
v
h
id
T
o
w
n
,
I
r
an
E
m
ail:
m
.
m
a
lek
za
d
e
h
@
h
s
u
.
ac
.
ir
1.
I
NT
RO
D
UCT
I
O
N
Fif
t
h
-
g
en
er
atio
n
(
5
G)
n
et
w
o
r
k
s
p
r
o
v
id
e
s
u
p
p
o
r
t
f
o
r
u
ltra
-
r
eliab
le
lo
w
-
laten
c
y
co
m
m
u
n
icatio
n
s
,
en
h
a
n
ce
d
m
o
b
ile
b
r
o
ad
b
an
d
,
an
d
m
ass
iv
e
m
ac
h
i
n
e
-
t
y
p
e
co
m
m
u
n
icatio
n
s
s
er
v
ice
s
[
1
]
,
r
ely
i
n
g
o
n
ar
ch
itect
u
r
es
b
u
il
t
u
p
o
n
s
o
f
t
war
e
-
d
ef
i
n
ed
n
et
w
o
r
k
in
g
(
SD
N
)
an
d
n
et
w
o
r
k
s
lici
n
g
tec
h
n
o
l
o
g
y
[
2
]
.
Ho
w
e
v
er
,
v
ir
tu
a
lis
ed
s
lice
s
t
h
at
s
h
ar
e
th
e
s
a
m
e
p
h
y
s
ical
i
n
f
r
as
tr
u
ct
u
r
e
o
p
en
th
e
d
o
o
r
to
cr
o
s
s
-
s
lice
attac
k
v
ec
to
r
s
[
3
]
,
[
4
]
,
an
d
co
o
r
d
in
ated
d
is
tr
ib
u
ted
d
en
ial
-
of
-
s
er
v
ice
attac
k
s
lau
n
c
h
ed
th
r
o
u
g
h
m
u
ltip
le
en
d
p
o
in
ts
ca
n
d
is
r
u
p
t
v
ital
s
lice
s
[
5
]
,
[
6
]
.
C
o
n
v
e
n
tio
n
al
s
i
g
n
at
u
r
e
-
b
ased
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
te
m
s
(
I
DS)
ar
e
u
n
ab
le
to
co
p
e
w
i
th
e
m
er
g
i
n
g
p
atter
n
s
in
h
i
g
h
-
th
r
o
u
g
h
p
u
t
e
n
cr
y
p
ted
tr
af
f
ic
[
7
]
.
Ma
ch
i
n
e
lear
n
i
n
g
(
ML
)
a
n
d
d
ee
p
lear
n
in
g
(
D
L
)
ap
p
r
o
ac
h
es
p
r
o
v
id
e
d
ata
-
d
r
iv
en
a
lter
n
ati
v
e
s
,
y
et
t
h
eir
c
r
ed
ib
ilit
y
h
i
n
g
es
o
n
m
e
th
o
d
o
lo
g
ical
s
tr
ict
n
es
s
,
esp
ec
iall
y
t
h
e
p
r
ev
e
n
tio
n
o
f
d
ata
leak
ag
e
t
h
r
o
u
g
h
o
u
t
p
r
ep
r
o
ce
s
s
i
n
g
s
tag
e
s
[
8
]
,
[
9
]
.
A
co
n
s
id
er
ab
le
n
u
m
b
er
o
f
p
u
b
lis
h
ed
s
t
u
d
ies
p
er
f
o
r
m
f
ea
tu
r
e
s
elec
tio
n
,
n
o
r
m
ali
s
atio
n
,
o
r
o
v
er
s
am
p
li
n
g
p
r
io
r
to
d
ata
p
ar
titi
o
n
in
g
,
w
h
ic
h
ex
p
o
s
es
p
er
f
o
r
m
a
n
ce
esti
m
ate
s
to
th
e
r
is
k
o
f
i
n
f
la
tio
n
.
F
u
r
t
h
er
m
o
r
e,
o
n
l
y
a
li
m
ited
n
u
m
b
er
o
f
s
tu
d
ies
u
t
ilis
e
d
atasets
s
p
ec
if
ic
to
5G
n
et
w
o
r
k
tr
af
f
ic
o
r
clar
if
y
w
h
et
h
er
p
r
ep
r
o
ce
s
s
in
g
p
r
o
ce
d
u
r
es
w
er
e
r
estricte
d
to
tr
ain
in
g
p
ar
titi
o
n
s
alo
n
e.
T
h
is
s
t
u
d
y
p
r
o
p
o
s
es
an
d
ap
p
lies
a
leak
a
g
e
-
a
w
ar
e
ev
al
u
atio
n
p
r
o
to
co
l
to
b
en
ch
m
ar
k
t
w
elv
e
ar
ch
itect
u
r
es
o
n
th
e
Kh
an
et
a
l.
[
9
]
5
G
n
et
w
o
r
k
s
lici
n
g
d
ataset.
T
h
e
co
n
tr
ib
u
tio
n
s
ar
e:
i
)
a
s
tr
ictl
y
leak
a
g
e
-
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
Leve
r
a
g
in
g
a
r
tifi
cia
l in
tellig
en
ce
fo
r
d
etec
tio
n
o
f
d
en
ia
l
-
of
-
s
ervice
a
tta
ck
s
in
5
G
n
etw
o
r
k
…
(
B
a
s
ee
l A
l
-
A
li
)
927
co
n
tr
o
lled
p
r
ep
r
o
ce
s
s
in
g
p
r
o
to
co
l
in
w
h
ic
h
m
u
tu
al
in
f
o
r
m
atio
n
(
MI
)
f
ea
t
u
r
e
s
elec
tio
n
,
n
o
r
m
ali
s
atio
n
,
a
n
d
s
y
n
t
h
etic
m
i
n
o
r
it
y
o
v
er
s
a
m
p
lin
g
tech
n
iq
u
e
(
SMOT
E
)
ar
e
co
n
f
i
n
ed
to
tr
ain
i
n
g
p
ar
titi
o
n
s
a
n
d
MI
i
s
r
ec
o
m
p
u
ted
w
it
h
in
ea
c
h
cr
o
s
s
-
v
alid
atio
n
(
C
V)
f
o
ld
ad
d
r
e
s
s
i
n
g
a
m
et
h
o
d
o
lo
g
ical
g
ap
i
n
ex
i
s
ti
n
g
5
G
I
DS
b
en
ch
m
ar
k
s
;
(
2
)
tr
an
s
p
ar
en
t
r
ep
o
r
tin
g
o
f
co
n
f
u
s
io
n
m
atr
i
x
co
m
p
o
n
e
n
t
s
an
d
ei
g
h
t
d
er
iv
ed
m
etr
ic
s
f
o
r
all
t
w
el
v
e
m
o
d
els
u
n
d
er
id
en
tica
l
co
n
d
itio
n
s
;
an
d
(
3
)
id
en
tif
i
ca
tio
n
o
f
ac
cu
r
ac
y
co
s
t
tr
ad
e
-
o
f
f
s
ac
r
o
s
s
m
o
d
e
l
f
a
m
ilies
,
w
it
h
co
n
v
o
lu
tio
n
al
n
eu
r
al
n
et
w
o
r
k
s
(
C
NN)
a
n
d
r
an
d
o
m
f
o
r
est
e
m
er
g
i
n
g
as
ca
n
d
id
ates
f
o
r
r
eso
u
r
ce
-
r
ich
an
d
r
eso
u
r
ce
-
co
n
s
tr
ai
n
ed
d
ep
lo
y
m
en
t
s
,
r
esp
ec
tiv
e
l
y
.
2.
RE
L
AT
E
D
WO
RK
P
r
io
r
w
o
r
k
o
n
M
L
/D
L
-
b
ased
d
en
ial
o
f
s
er
v
ice
(
Do
S)
d
etec
t
io
n
ca
n
b
e
g
r
o
u
p
ed
in
to
t
h
r
ee
ca
teg
o
r
ies.
I
n
5
G
-
o
r
ien
ted
s
t
u
d
ies,
I
m
an
b
a
y
ev
et
a
l.
[
1
0
]
ev
alu
ated
g
r
ad
ien
t
b
o
o
s
tin
g
o
n
C
I
C
I
DS2
0
1
7
w
it
h
i
n
a
s
i
m
u
lated
5
G
co
r
e
(
9
9
.
3
%
ac
cu
r
ac
y
)
,
th
o
u
g
h
n
o
r
ea
l
5
G
tr
af
f
ic
w
as
u
s
ed
an
d
p
r
ep
r
o
ce
s
s
in
g
leak
a
g
e
co
n
tr
o
ls
w
er
e
n
o
t
d
o
cu
m
en
ted
.
Sar
an
y
a
et
a
l.
[
1
1
]
an
d
R
o
d
r
íg
u
ez
et
a
l.
[
1
2
]
b
en
ch
m
ar
k
ed
clas
s
ical
M
L
c
lass
i
f
ier
s
o
n
K
DD
-
C
UP
an
d
C
I
C
I
DS2
0
1
7
r
esp
ec
tiv
el
y
,
ac
h
iev
i
n
g
h
i
g
h
ac
c
u
r
ac
y
b
u
t
o
n
d
atasets
t
h
at
p
r
ed
ate
5
G
n
et
w
o
r
k
ch
ar
ac
ter
is
tic
s
.
Vas
h
is
h
t
h
a
an
d
C
h
a
tter
j
ee
[
1
3
]
p
r
o
p
o
s
ed
Sp
ar
k
Sh
ie
ld
f
o
r
clo
u
d
-
b
ase
d
I
DS
o
n
s
i
m
u
lated
d
atasets
.
I
n
in
ter
n
et
o
f
t
h
in
g
s
(
I
o
T
)
-
f
o
cu
s
ed
d
etec
tio
n
,
s
ev
er
al
s
tu
d
ies
r
ep
o
r
t
ac
cu
r
ac
ies
ex
ce
ed
in
g
9
9
%:
R
ag
ab
et
a
l.
[
1
4
]
w
it
h
Har
r
i
s
Ha
w
k
s
-
o
p
ti
m
is
ed
D
L
o
n
B
O
T
-
I
o
T
,
C
h
er
ian
a
n
d
Var
m
a
[
1
5
]
w
it
h
SDN
+
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
L
ST
M)
o
n
C
I
C
-
DDo
S2
0
1
9
,
an
d
A
s
w
a
d
et
a
l.
[
1
6
]
w
i
th
C
N
N
+
b
id
ir
ec
tio
n
al
lo
n
g
s
h
o
r
t
-
ter
m
m
e
m
o
r
y
(
B
i
L
ST
M)
o
n
C
I
C
-
I
DS2
0
1
7
.
Fu
r
th
er
w
o
r
k
s
in
cl
u
d
e
s
n
a
k
e
-
o
p
ti
m
i
s
ed
en
s
e
m
b
les
[
1
7
]
,
ar
tif
icial
n
eu
r
al
n
et
w
o
r
k
(
A
NN
)
+
L
ST
M
f
o
r
lig
h
t
w
ei
g
h
t
I
o
T
[
1
8
]
,
c
h
ao
tic
-
o
p
ti
m
is
ed
E
l
m
an
r
ec
u
r
r
en
t
n
e
u
r
al
n
et
w
o
r
k
(
R
NN)
[
1
9
]
,
L
ST
M
-
b
ased
OPT
I
MI
S
T
[
2
0
]
,
an
d
C
NN
+
b
id
ir
ec
tio
n
al
g
ated
r
ec
u
r
r
en
t
u
n
it
(
B
iGR
U)
f
o
r
s
m
ar
t
f
ar
m
i
n
g
[
2
1
]
.
H
y
b
r
id
f
r
a
m
e
wo
r
k
s
[
2
2
]
,
h
ea
lth
ca
r
e
I
o
T
[
2
3
]
,
an
d
m
u
lt
i
-
a
lg
o
r
it
h
m
b
en
c
h
m
ar
k
s
[
2
4
]
h
a
v
e
also
be
en
ex
p
lo
r
ed
.
B
ac
k
g
r
o
u
n
d
o
n
5
G
s
ec
u
r
it
y
i
s
p
r
o
v
id
ed
in
[
2
5
]
.
T
h
r
ee
m
et
h
o
d
o
lo
g
ical
g
ap
s
e
m
er
g
e
f
r
o
m
t
h
is
r
e
v
ie
w
.
Firs
t
,
th
e
m
aj
o
r
it
y
o
f
s
tu
d
ies
ar
e
c
o
n
f
i
n
ed
to
ev
alu
a
tin
g
o
n
e
o
r
t
w
o
m
o
d
el
f
a
m
ilie
s
,
an
d
n
o
s
i
n
g
le
s
tu
d
y
ca
r
r
ies
o
u
t
a
s
i
m
u
lta
n
eo
u
s
co
m
p
ar
is
o
n
a
m
ong
en
s
e
m
b
le
ML
m
o
d
els,
f
ee
d
f
o
r
w
ar
d
DL
n
e
t
w
o
r
k
s
,
co
n
v
o
lu
tio
n
al
n
e
t
w
o
r
k
s
,
r
ec
u
r
r
en
t
n
et
w
o
r
k
s
,
b
elief
n
et
w
o
r
k
s
,
an
d
p
r
o
b
ab
ilis
tic
m
o
d
els
u
n
d
er
id
en
tical
co
n
d
itio
n
s
.
Seco
n
d
,
t
h
e
r
ev
ie
w
ed
s
t
u
d
ies
p
r
ed
o
m
i
n
a
n
tl
y
r
el
y
o
n
d
atasets
s
u
c
h
as
C
I
C
I
DS2
0
1
7
,
B
OT
-
I
o
T
,
o
r
K
DD
-
C
UP
,
an
d
r
ar
ely
u
til
is
e
d
atasets
s
p
ec
if
ica
ll
y
d
esig
n
ed
f
o
r
5G
n
et
w
o
r
k
s
t
h
at
r
ef
lect
n
et
w
o
r
k
s
lic
in
g
tr
a
f
f
ic
.
T
h
ir
d
,
an
d
m
o
s
t
i
m
p
o
r
tan
t
l
y
,
th
e
ap
p
licatio
n
o
f
p
er
-
f
o
ld
f
ea
t
u
r
e
s
elec
t
io
n
w
it
h
d
o
cu
m
e
n
ted
d
ata
lea
k
ag
e
p
r
ev
en
tio
n
co
n
tr
o
ls
is
ab
s
e
n
t
f
r
o
m
t
h
e
r
ev
ie
w
e
d
liter
atu
r
e.
St
u
d
ies
s
u
ch
a
s
[
1
0
]
,
[
14
]
,
[
1
5
]
r
e
p
o
r
t
h
ig
h
ac
cu
r
a
c
y
w
it
h
o
u
t
clar
i
f
y
in
g
w
h
eth
er
f
ea
t
u
r
e
s
elec
tio
n
o
r
n
o
r
m
ali
s
atio
n
w
as
co
n
d
u
cted
b
ef
o
r
e
o
r
af
ter
d
ata
p
a
r
titi
o
n
in
g
,
w
h
ic
h
co
n
s
ti
tu
te
s
a
m
eth
o
d
o
lo
g
ical
a
m
b
ig
u
it
y
th
at
m
a
y
lead
to
in
f
la
ted
r
ep
o
r
ted
p
er
f
o
r
m
a
n
ce
.
T
ab
le
1
p
r
o
v
id
es
a
s
u
m
m
ar
y
o
f
t
h
e
o
v
er
all
r
esear
ch
lan
d
s
ca
p
e.
T
ab
le
1
.
Su
m
m
ar
y
o
f
r
elate
d
Do
S/
d
is
tr
ib
u
ted
d
en
ial
o
f
s
er
v
ice
(
DDo
S)
d
etec
tio
n
s
tu
d
ies
S
t
u
d
y
M
o
d
e
l
(
s)
D
a
t
a
se
t
B
e
st
m
e
t
r
i
c
L
i
mi
t
a
t
i
o
n
I
man
b
a
y
e
v
e
t
a
l
.
[
1
0
]
G
r
a
d
i
e
n
t
b
o
o
st
i
n
g
C
I
C
I
D
S
2
0
1
7
,
C
S
E
-
C
I
C
9
9
.
3
%
N
o
r
e
a
l
5
G
;
l
e
a
k
a
g
e
u
n
c
l
e
a
r
S
a
r
a
n
y
a
e
t
a
l
.
[
1
1
]
L
D
A
,
C
A
R
T
,
R
F
K
D
D
-
C
U
P
R
a
n
d
o
m fo
r
e
st
b
e
st
D
a
t
a
se
t
o
u
t
d
a
t
e
d
f
o
r
5
G
V
a
sh
i
sh
t
h
a
a
n
d
C
h
a
t
t
e
r
j
e
e
[
1
3
]
S
p
a
r
k
S
h
i
e
l
d
T
e
st
C
l
o
u
d
I
D
S
,
U
N
S
W
I
mp
r
o
v
e
d
S
i
mu
l
a
t
e
d
d
a
t
a
o
n
l
y
R
a
g
a
b
e
t
a
l
.
[
1
4
]
P
H
H
O
-
O
D
L
C
B
O
T
-
I
o
T
9
9
.
2
%
S
i
n
g
l
e
d
a
t
a
se
t
C
h
e
r
i
a
n
a
n
d
V
a
r
ma
[
1
5
]
S
D
N
+
L
S
T
M
C
I
C
-
D
D
o
S
2
0
1
9
9
9
.
8
%
L
e
a
k
a
g
e
c
o
n
t
r
o
l
s
u
n
d
o
c
u
me
n
t
e
d
A
s
w
a
d
e
t
a
l
.
[
1
6
]
C
N
N
+
B
i
L
S
T
M
C
I
C
-
I
D
S
2
0
1
7
9
9
.
7
6
%
A
d
a
p
t
a
b
i
l
i
t
y
u
n
c
o
n
f
i
r
me
d
A
l
j
e
b
r
e
e
n
e
t
a
l
.
[
1
7
]
En
se
mb
l
e
+
S
n
a
k
e
B
O
T
-
I
o
T
9
9
.
7
6
%
H
i
g
h
c
o
m
p
u
t
a
t
i
o
n
a
l
c
o
s
t
K
h
a
n
d
a
y
e
t
a
l
.
[
1
8
]
A
N
N
+
L
S
T
M
B
O
T
-
I
o
T
,
T
O
N
-
I
o
T
~
9
9
%
N
o
v
e
l
a
t
t
a
c
k
s
u
n
t
e
st
e
d
H
u
ssa
n
e
t
a
l
.
[
1
9
]
ER
N
N
+
C
B
C
O
B
O
T
-
I
o
T
,
C
I
C
-
I
D
S
>
9
8
%
S
c
a
l
a
b
i
l
i
t
y
u
n
t
e
st
e
d
B
h
a
l
e
e
t
a
l
.
[
2
0
]
L
S
T
M
(
O
P
TI
M
I
S
T
)
C
u
s
t
o
m
9
8
.
4
%
H
i
g
h
me
mo
r
y
d
e
man
d
s
K
e
t
h
i
n
e
n
i
a
n
d
P
r
a
d
e
e
p
i
n
i
[
2
1
]
C
N
N
+
B
i
G
R
U
C
I
C
D
D
o
S
2
0
1
9
>
9
9
%
R
e
a
l
-
w
o
r
l
d
u
n
t
e
st
e
d
P
r
e
se
n
t
st
u
d
y
1
2
M
L
/
D
L
/
p
r
o
b
.
K
h
a
n
e
t
a
l
.
5
G
[
9
]
C
N
N
F
1
=
0
.
9
8
3
S
i
n
g
l
e
5
G
d
a
t
a
se
t
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
.
3
,
J
u
n
e
20
26
:
9
2
6
-
9
32
928
3.
M
E
T
H
O
DO
L
O
G
Y
3
.
1
.
M
o
del
s
elec
t
io
n
A
to
tal
o
f
t
w
el
v
e
m
o
d
els
wer
e
ch
o
s
en
,
co
v
er
i
n
g
s
i
x
ML
class
i
f
ier
s
,
n
a
m
el
y
lo
g
is
tic
r
eg
r
ess
io
n
,
r
an
d
o
m
f
o
r
est,
d
ec
is
io
n
tr
ee
,
g
r
ad
ien
t
b
o
o
s
tin
g
,
Naïv
e
B
ay
es,
an
d
m
u
ltil
a
y
er
p
er
ce
p
tr
o
n
(
ML
P
)
,
alo
n
g
w
it
h
f
i
v
e
DL
ar
ch
itect
u
r
es
i
n
cl
u
d
in
g
C
NN
,
L
ST
M
,
r
ec
u
r
r
en
t
n
eu
r
al
n
et
w
o
r
k
,
g
ated
r
ec
u
r
r
e
n
t
u
n
it,
a
n
d
d
ee
p
b
elief
n
et
w
o
r
k
(
DB
N)
,
i
n
ad
d
itio
n
t
o
a
s
in
g
le
p
r
o
b
ab
ilis
tic
m
o
d
el,
w
h
ic
h
is
t
h
e
h
id
d
en
Ma
r
k
o
v
m
o
d
el
(
HM
M)
,
K
-
n
ea
r
est
n
ei
g
h
b
o
r
(
K
NN)
a
n
d
A
d
aB
o
o
s
t
w
er
e
ex
c
lu
d
ed
:
K
NN
is
m
et
h
o
d
o
lo
g
icall
y
s
u
p
e
r
s
ed
ed
b
y
tr
ee
-
b
ased
class
i
f
ier
s
f
o
r
tab
u
lar
d
ata
an
d
in
cu
r
s
h
i
g
h
i
n
f
er
en
ce
m
e
m
o
r
y
(
~5
0
0
MB
)
;
A
d
aB
o
o
s
t
b
elo
n
g
s
to
th
e
s
a
m
e
b
o
o
s
tin
g
f
a
m
il
y
a
s
g
r
ad
ien
t
b
o
o
s
tin
g
w
it
h
o
u
t
ad
d
in
g
ar
ch
itect
u
r
al
d
iv
er
s
it
y
.
R
ec
u
r
r
en
t
m
o
d
els
r
ec
eiv
ed
s
in
g
l
e
-
ti
m
es
tep
in
p
u
t
(
1
,
9
)
,
t
r
ea
tin
g
ea
ch
f
lo
w
r
ec
o
r
d
as
an
in
d
ep
en
d
en
t
v
ec
to
r
;
th
is
is
a
d
elib
er
ate
p
r
o
to
c
o
l
ch
o
ice,
as th
e
d
ataset
p
r
o
v
id
es
in
d
ep
en
d
en
t
f
lo
w
r
ec
o
r
d
s
w
it
h
o
u
t i
n
ter
-
r
ec
o
r
d
te
m
p
o
r
al
o
r
d
er
in
g
.
3
.
2
.
Da
t
a
s
et
T
h
e
Do
S/DDo
S
attac
k
d
ataset
o
f
5
G
ne
t
w
o
r
k
s
l
icin
g
b
y
Kh
an
et
a
l.
[
9
]
(
I
E
E
E
Data
P
o
r
t,
d
o
i:
1
0
.
2
1
2
2
7
/
3
2
k
1
-
d
r
1
2
)
c
o
n
tain
s
f
lo
w
-
le
v
el
r
ec
o
r
d
s
f
r
o
m
t
w
o
v
ir
tu
a
l
5
G
s
lice
s
o
v
er
t
w
o
d
a
y
s
,
w
i
th
8
4
f
ea
t
u
r
es
an
d
ap
p
r
o
x
i
m
atel
y
8
0
% b
en
i
g
n
/ 2
0
% a
ttack
clas
s
d
is
tr
ib
u
ti
o
n
.
3
.
3
.
P
re
pro
ce
s
s
ing
p
ipelin
e
A
s
p
lit
-
f
ir
s
t
p
r
o
to
co
l
en
f
o
r
ce
s
s
tr
ict
tr
ain
i
n
g
-
o
n
l
y
p
r
ep
r
o
c
ess
i
n
g
.
T
h
e
d
ataset
w
a
s
p
ar
titi
o
n
ed
in
to
Dtr
ain
(
7
0
%)
an
d
D
test
(
3
0
%)
v
ia
s
tr
ati
f
ied
s
a
m
p
li
n
g
(
s
ee
d
=4
2
)
,
w
it
h
al
l
tr
an
s
f
o
r
m
at
io
n
s
f
itted
o
n
Dtr
ai
n
o
n
l
y
.
S
MO
T
E
ad
d
r
ess
es
th
e
4
:1
class
i
m
b
ala
n
ce
e
x
cl
u
s
i
v
el
y
o
n
tr
ai
n
i
n
g
d
ata,
af
ter
D
L
v
alid
atio
n
s
u
b
s
et
ex
tr
ac
tio
n
.
T
ab
le
2
d
etails th
e
p
ip
elin
e.
T
ab
le
2
.
P
r
ep
r
o
ce
s
s
in
g
p
ip
eli
n
e
S
t
e
p
D
e
scri
p
t
i
o
n
D
a
t
a
c
l
e
a
n
i
n
g
M
e
r
g
e
sl
i
c
e
/
d
a
y
f
i
l
e
s.
R
e
mo
v
e
d
u
p
l
i
c
a
t
e
s,
N
a
N
r
o
w
s,
I
P
st
r
i
n
g
s,
a
n
d
t
i
me
st
a
mp
s
.
S
t
r
a
t
i
f
i
e
d
s
p
l
i
t
D
t
r
a
i
n
(
7
0
%) /
D
t
e
st
(
3
0
%),
s
t
r
a
t
i
f
i
e
d
r
a
n
d
o
m sa
m
p
l
i
n
g
(
se
e
d
=
4
2
)
.
M
I
f
e
a
t
u
r
e
s
e
l
e
c
t
i
o
n
C
o
mp
u
t
e
M
I
o
n
D
t
r
a
i
n
o
n
l
y
(
mu
t
u
a
l
_
i
n
f
o
_
c
l
a
ss
i
f
,
1
0
N
N
)
.
R
e
t
a
i
n
M
I
≥
0
.
5
1
(
9
f
e
a
t
u
r
e
s,
T
a
b
l
e
3
)
.
R
e
c
o
mp
u
t
e
d
p
e
r
C
V
f
o
l
d
.
L
a
b
e
l
e
n
c
o
d
i
n
g
L
a
b
e
l
En
c
o
d
e
r
f
i
t
t
e
d
o
n
D
t
r
a
i
n
o
n
l
y
.
N
o
r
mal
i
sat
i
o
n
S
t
a
n
d
a
r
d
S
c
a
l
e
r
f
i
t
t
e
d
o
n
D
t
r
a
i
n
;
a
p
p
l
i
e
d
t
o
b
o
t
h
p
a
r
t
i
t
i
o
n
s.
S
M
O
T
E
A
p
p
l
i
e
d
t
o
D
t
r
a
i
n
o
n
l
y
(
k
=
5
,
se
e
d
=
4
2
)
,
a
f
t
e
r
D
v
a
l
e
x
t
r
a
c
t
i
o
n
.
D
L
v
a
l
i
d
a
t
i
o
n
D
v
a
l
=
1
5
%
o
f
D
t
r
a
i
n
(
s
t
r
a
t
i
f
i
e
d
,
se
e
d
=
4
2
)
f
o
r
e
a
r
l
y
st
o
p
p
i
n
g
.
3
.
4
.
F
e
a
t
ure
s
elec
t
io
n
T
h
e
MI
m
etr
ic
w
a
s
ad
o
p
ted
d
u
e
to
it
s
d
is
t
in
ctio
n
i
n
d
etec
ti
n
g
s
tati
s
tical
r
elatio
n
s
h
ip
s
w
h
eth
er
li
n
ea
r
o
r
n
o
n
lin
ea
r
,
u
n
li
k
e
m
et
h
o
d
s
b
ased
o
n
co
r
r
elatio
n
co
ef
f
icie
n
ts
t
h
at
o
n
l
y
r
ev
ea
l
lin
ea
r
as
s
o
ciatio
n
s
.
T
h
e
s
o
r
ted
MI
cu
r
v
e
u
n
co
v
er
ed
a
p
r
o
m
i
n
en
t
i
n
f
lectio
n
p
o
in
t
at
t
h
e
v
alu
e
o
f
0
.
5
1
,
w
h
ich
led
to
k
e
ep
in
g
n
in
e
f
ea
tu
r
e
s
(
T
a
b
le
3
)
th
at
s
af
e
g
u
ar
d
th
e
d
is
cr
i
m
i
n
ati
v
e
p
o
w
er
o
f
t
h
e
d
ata
w
h
ile
co
n
d
en
s
in
g
t
h
e
d
i
m
e
n
s
io
n
s
f
r
o
m
ei
g
h
t
y
-
f
o
u
r
d
o
w
n
to
o
n
l
y
n
i
n
e.
T
ab
le
3
.
R
etain
ed
f
ea
tu
r
e
s
(
MI
≥
0
.
5
1
o
n
Dtr
ain
)
F
e
a
t
u
r
e
(
M
I
)
D
e
scr
i
p
t
i
o
n
F
l
o
w
d
u
r
a
t
i
o
n
(
0
.
7
4
)
T
o
t
a
l
f
l
o
w
d
u
r
a
t
i
o
n
;
a
t
t
a
c
k
f
l
o
w
s e
x
h
i
b
i
t
l
o
n
g
e
r
d
u
r
a
t
i
o
n
s
F
w
d
p
k
t
l
e
n
s
t
d
(
0
.
6
8
)
S
t
d
d
e
v
o
f
f
o
r
w
a
r
d
p
a
c
k
e
t
l
e
n
g
t
h
s
;
f
l
o
o
d
i
n
g
i
n
d
i
c
a
t
o
r
A
C
K
f
l
a
g
c
o
u
n
t
(
0
.
6
5
)
T
C
P
A
C
K
f
l
a
g
s;
e
l
e
v
a
t
e
d
c
o
u
n
t
s s
i
g
n
a
l
A
C
K
-
f
l
o
o
d
D
D
o
S
T
o
t
a
l
f
w
d
p
a
c
k
e
t
s
(
0
.
6
3
)
F
o
r
w
a
r
d
p
a
c
k
e
t
c
o
u
n
t
;
v
o
l
u
me
t
r
i
c
a
t
t
a
c
k
s
p
r
o
d
u
c
e
l
a
r
g
e
v
a
l
u
e
s
P
r
o
t
o
c
o
l
(
0
.
6
1
)
T
r
a
n
sp
o
r
t
p
r
o
t
o
c
o
l
(
T
C
P
/
U
D
P
/
I
C
M
P
)
;
a
t
t
a
c
k
s
i
g
n
a
t
u
r
e
s
D
st
p
o
r
t
(
0
.
5
8
)
D
e
st
i
n
a
t
i
o
n
p
o
r
t
;
t
a
r
g
e
t
e
d
p
a
t
t
e
r
n
s
i
n
d
i
c
a
t
e
D
D
o
S
S
r
c
p
o
r
t
(
0
.
5
4
)
S
o
u
r
c
e
p
o
r
t
;
r
a
n
d
o
mi
se
d
p
o
r
t
s s
u
g
g
e
st
sp
o
o
f
i
n
g
F
w
d
p
k
t
l
e
n
m
e
a
n
(
0
.
5
2
)
M
e
a
n
f
o
r
w
a
r
d
p
a
c
k
e
t
l
e
n
g
t
h
;
d
e
v
i
a
t
e
s
u
n
d
e
r
a
t
t
a
c
k
F
w
d
se
g
si
z
e
mi
n
(
0
.
5
1
)
M
i
n
T
C
P
se
g
me
n
t
s
i
z
e
;
smal
l
v
a
l
u
e
s i
n
d
i
c
a
t
e
mal
f
o
r
me
d
p
a
c
k
e
t
s
4.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
A
ll
r
e
s
u
l
ts
w
er
e
r
ep
o
r
ted
o
n
th
e
test
s
et
co
n
tai
n
i
n
g
ap
p
r
o
x
i
m
atel
y
ei
g
h
t
y
p
er
ce
n
t
b
en
ig
n
d
ata
an
d
t
w
e
n
t
y
p
er
ce
n
t
attac
k
d
ata
w
it
h
ar
o
u
n
d
1
2
3
,
9
0
0
r
ec
o
r
d
s
,
w
h
ich
i
s
co
m
p
lete
l
y
is
o
lated
f
r
o
m
t
h
e
tr
ai
n
i
n
g
d
ata.
E
ig
h
t
m
etr
ics
d
er
iv
ed
f
r
o
m
th
e
co
n
f
u
s
io
n
m
atr
i
x
w
er
e
ca
lcu
lated
,
n
a
m
el
y
o
v
er
all
ac
cu
r
ac
y
,
p
r
ec
is
io
n
,
r
ec
all,
F
1
-
s
co
r
e,
n
e
g
ati
v
e
p
r
ed
ictiv
e
v
alu
e,
f
alse
d
i
s
co
v
er
y
r
ate
,
Fo
w
l
k
es
-
Ma
l
lo
w
s
i
n
d
ex
,
an
d
m
a
r
k
ed
n
es
s
[
2
6
]
.
B
o
th
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
Leve
r
a
g
in
g
a
r
tifi
cia
l in
tellig
en
ce
fo
r
d
etec
tio
n
o
f
d
en
ia
l
-
of
-
s
ervice
a
tta
ck
s
in
5
G
n
etw
o
r
k
…
(
B
a
s
ee
l A
l
-
A
li
)
929
s
p
ec
if
icit
y
an
d
d
etec
tio
n
r
ate
w
er
e
ex
cl
u
d
ed
s
in
ce
th
e
y
ar
e
n
u
m
er
icall
y
eq
u
iv
ale
n
t
to
o
v
er
all
ac
cu
r
ac
y
an
d
r
ec
all
r
esp
ec
tiv
el
y
w
ith
in
t
h
i
s
ev
alu
a
tio
n
.
4
.
1
.
Co
nfusi
o
n
m
a
t
rix
a
na
ly
s
is
T
ab
le
4
d
is
p
lay
s
th
e
co
n
f
u
s
i
o
n
m
atr
i
x
.
T
h
e
C
NN
m
o
d
el
r
ec
o
r
d
e
d
th
e
h
i
g
h
e
s
t
n
u
m
b
er
o
f
tr
u
e
p
o
s
itiv
es
(
T
P
)
at
2
4
,
6
0
7
w
it
h
o
n
l
y
1
7
3
f
alse
n
eg
ati
v
es
(
FN
)
,
y
ield
i
n
g
a
m
i
s
s
r
ate
o
f
0
.
7
%,
f
o
llo
w
ed
b
y
th
e
R
NN
a
n
d
DB
N
m
o
d
els
w
it
h
2
4
,
5
8
2
TP
an
d
1
9
8
FN
ea
ch
.
R
eg
ar
d
in
g
ML
m
o
d
els,
t
h
e
r
a
n
d
o
m
f
o
r
est
m
o
d
el
r
eg
is
ter
ed
2
4
,
4
5
8
TP
ag
ain
s
t
3
2
2
FN
.
T
h
e
C
NN
m
o
d
el
p
r
o
d
u
ce
d
m
er
el
y
6
8
3
f
alse
alar
m
s
w
it
h
a
f
alse
p
o
s
itiv
e
(
FP
)
r
ate
o
f
0
.
6
9
%
,
w
h
er
ea
s
t
h
e
HM
M
(
1
9
,
6
9
1
f
alse
alar
m
s
at
1
9
.
9
%),
Naïv
e
B
ay
es
(
1
0
,
9
9
8
at
1
1
.
1
%),
an
d
lo
g
is
tic
r
eg
r
e
s
s
i
o
n
(
7
,
8
1
2
at
7
.
9
%)
d
em
o
n
s
tr
ated
elev
ated
f
a
ls
e
alar
m
r
ate
s
th
a
t
w
o
u
ld
ca
u
s
e
co
n
s
id
er
ab
le
aler
t f
ati
g
u
e
i
n
r
ea
l
-
w
o
r
ld
o
p
er
atio
n
al
en
v
ir
o
n
m
en
ts
.
T
ab
le
4
.
C
o
n
f
u
s
io
n
m
atr
i
x
f
o
r
all
1
2
m
o
d
els o
n
Dte
s
t
M
o
d
e
l
TP
T
r
u
e
n
e
g
a
t
i
v
e
s (
TN
)
FP
FN
T
o
t
a
l
C
N
N
2
4
,
6
0
7
9
8
,
4
3
7
6
8
3
1
7
3
1
2
3
,
9
0
0
R
N
N
2
4
,
5
8
2
9
8
,
3
3
4
7
8
6
1
9
8
1
2
3
,
9
0
0
D
B
N
2
4
,
5
8
2
9
8
,
3
3
4
7
8
6
1
9
8
1
2
3
,
9
0
0
R
a
n
d
o
m
f
o
r
e
st
2
4
,
4
5
8
9
7
,
8
3
3
1
,
2
8
7
3
2
2
1
2
3
,
9
0
0
G
r
a
d
i
e
n
t
b
o
o
st
i
n
g
2
4
,
4
0
8
9
7
,
6
4
5
1
,
4
7
5
3
7
2
1
2
3
,
9
0
0
M
L
P
c
l
a
ssi
f
i
e
r
2
4
,
3
5
9
9
7
,
4
2
7
1
,
6
9
3
4
2
1
1
2
3
,
9
0
0
D
e
c
i
si
o
n
t
r
e
e
2
4
,
2
3
5
9
6
,
9
2
6
2
,
1
9
4
5
4
5
1
2
3
,
9
0
0
L
S
T
M
2
4
,
2
3
5
9
6
,
9
2
6
2
,
1
9
4
5
4
5
1
2
3
,
9
0
0
G
R
U
2
4
,
2
1
0
9
6
,
8
4
2
2
,
2
7
8
5
7
0
1
2
3
,
9
0
0
L
o
g
i
st
i
c
r
e
g
r
e
ssi
o
n
2
2
,
8
2
2
9
1
,
3
0
8
7
,
8
1
2
1
,
9
5
8
1
2
3
,
9
0
0
N
a
ï
v
e
B
a
y
e
s
2
2
,
0
2
9
8
8
,
1
2
2
1
0
,
9
9
8
2
,
7
5
1
1
2
3
,
9
0
0
H
M
M
1
9
,
8
4
9
7
9
,
4
2
9
1
9
,
6
9
1
4
,
9
3
1
1
2
3
,
9
0
0
4
.
2
.
M
ulti
-
m
et
ric
ev
a
lua
t
io
n
T
ab
le
5
p
r
esen
ts
eig
h
t
p
er
f
o
r
m
an
ce
m
etr
ics.
W
ith
i
n
th
is
d
ataset,
th
e
C
NN
m
o
d
el
attai
n
ed
an
F
1
v
alu
e
o
f
0
.
9
8
3
w
ith
a
f
alse
d
is
co
v
er
y
r
ate
o
f
0
.
0
2
7
,
f
o
llo
w
ed
b
y
t
h
e
R
NN
a
n
d
DB
N
m
o
d
els
w
it
h
a
n
F
1
v
al
u
e
o
f
0
.
9
8
0
.
T
h
e
r
an
d
o
m
f
o
r
est
m
o
d
el
r
ea
ch
ed
an
F
1
v
alu
e
o
f
0
.
9
6
8
w
i
th
a
f
alse
d
is
co
v
er
y
r
ate
o
f
0
.
0
5
0
.
T
h
e
n
eg
at
iv
e
p
r
ed
ictiv
e
v
al
u
e
s
u
r
p
ass
ed
0
.
9
9
0
ac
r
o
s
s
all
en
s
e
m
b
le
an
d
DL
m
o
d
els,
w
h
ic
h
co
n
f
ir
m
s
t
h
at
b
en
i
g
n
-
class
p
r
ed
ictio
n
s
ar
e
d
ep
en
d
ab
le
w
it
h
in
t
h
e
s
co
p
e
o
f
th
i
s
d
ataset.
As
f
o
r
th
e
m
ar
k
ed
n
es
s
m
etr
ic,
w
h
ic
h
co
m
b
i
n
es
b
o
th
p
o
s
iti
v
e
a
n
d
n
eg
ati
v
e
p
r
ed
ictiv
e
p
o
w
er
,
it
r
an
g
ed
f
r
o
m
0
.
9
7
1
f
o
r
th
e
C
NN
m
o
d
el
to
0
.
4
4
3
f
o
r
th
e
HM
M
,
r
e
f
lect
in
g
t
h
e
en
tir
e
s
p
ec
tr
u
m
o
f
class
if
ica
tio
n
q
u
alit
y
ac
r
o
s
s
t
h
e
v
ar
io
u
s
ar
ch
i
te
ctu
r
es.
T
ab
le
5
.
E
ig
h
t
-
m
etr
ic
e
v
alu
a
ti
o
n
f
o
r
all
1
2
m
o
d
els o
n
D
test
M
o
d
e
l
A
c
c
P
r
e
c
R
e
c
F1
N
P
V
F
D
R
FM
M
a
r
k
C
N
N
0
.
9
9
3
0
.
9
7
3
0
.
9
9
3
0
.
9
8
3
0
.
9
9
8
0
.
0
2
7
0
.
9
8
3
0
.
9
7
1
R
N
N
0
.
9
9
2
0
.
9
6
9
0
.
9
9
2
0
.
9
8
0
0
.
9
9
8
0
.
0
3
1
0
.
9
8
0
0
.
9
6
7
D
B
N
0
.
9
9
2
0
.
9
6
9
0
.
9
9
2
0
.
9
8
0
0
.
9
9
8
0
.
0
3
1
0
.
9
8
0
0
.
9
6
7
R
a
n
d
o
m fo
r
e
st
0
.
9
8
7
0
.
9
5
0
0
.
9
8
7
0
.
9
6
8
0
.
9
9
7
0
.
0
5
0
0
.
9
6
8
0
.
9
4
7
G
r
a
d
i
e
n
t
b
o
o
st
i
n
g
0
.
9
8
5
0
.
9
4
3
0
.
9
8
5
0
.
9
6
3
0
.
9
9
6
0
.
0
5
7
0
.
9
6
4
0
.
9
3
9
M
L
P
c
l
a
ssi
f
i
e
r
0
.
9
8
3
0
.
9
3
5
0
.
9
8
3
0
.
9
5
9
0
.
9
9
6
0
.
0
6
5
0
.
9
5
9
0
.
9
3
1
D
e
c
i
si
o
n
t
r
e
e
0
.
9
7
8
0
.
9
1
7
0
.
9
7
8
0
.
9
4
7
0
.
9
9
4
0
.
0
8
3
0
.
9
4
7
0
.
9
1
2
L
S
T
M
0
.
9
7
8
0
.
9
1
7
0
.
9
7
8
0
.
9
4
7
0
.
9
9
4
0
.
0
8
3
0
.
9
4
7
0
.
9
1
2
G
R
U
0
.
9
7
7
0
.
9
1
4
0
.
9
7
7
0
.
9
4
4
0
.
9
9
4
0
.
0
8
6
0
.
9
4
5
0
.
9
0
8
L
o
g
i
st
i
c
r
e
g
r
e
ssi
o
n
0
.
9
2
1
0
.
7
4
5
0
.
9
2
1
0
.
8
2
3
0
.
9
7
9
0
.
2
5
5
0
.
8
2
8
0
.
7
2
4
N
a
ï
v
e
B
a
y
e
s
0
.
8
8
9
0
.
6
6
7
0
.
8
8
9
0
.
7
6
2
0
.
9
7
0
0
.
3
3
3
0
.
7
7
0
0
.
6
3
7
H
M
M
0
.
8
0
1
0
.
5
0
2
0
.
8
0
1
0
.
6
1
7
0
.
9
4
2
0
.
4
9
8
0
.
6
3
4
0
.
4
4
3
Sev
er
al
o
b
s
er
v
atio
n
s
w
ar
r
an
t
an
al
y
tical
i
n
ter
p
r
etatio
n
.
T
h
e
s
u
p
er
io
r
it
y
o
f
t
h
e
C
N
N
m
o
d
el
w
it
h
i
n
t
h
i
s
d
ataset
is
lik
el
y
attr
ib
u
tab
le
to
th
e
ca
p
ac
ity
o
f
its
co
n
v
o
lu
tio
n
al
f
il
ter
s
to
ca
p
tu
r
e
lo
ca
l
in
ter
-
f
ea
t
u
r
e
co
r
r
elatio
n
s
w
h
e
n
t
h
e
n
in
e
t
ab
u
lar
f
ea
t
u
r
es
ar
e
r
estr
u
ctu
r
ed
in
to
a
t
w
o
-
d
i
m
e
n
s
io
n
a
l
i
n
p
u
t
m
atr
i
x
,
w
h
ich
en
ab
les
t
h
e
d
etec
tio
n
o
f
j
o
in
t
an
o
m
a
lies
s
u
c
h
as
s
i
m
u
lta
n
eo
u
s
d
ev
iatio
n
s
i
n
p
ac
k
et
l
en
g
t
h
v
ar
ian
ce
an
d
ac
k
n
o
w
led
g
e
m
e
n
t
f
lag
co
u
n
ts
th
at
tr
ee
-
b
ased
m
o
d
els
r
el
y
i
n
g
o
n
in
d
ep
en
d
e
n
t
s
p
litt
i
n
g
h
an
d
le
less
d
ir
ec
tl
y
.
T
h
e
n
ea
r
-
id
en
tical
p
er
f
o
r
m
a
n
c
e
o
f
th
e
R
NN
a
n
d
th
e
DB
N
,
b
o
th
ac
h
ie
v
in
g
a
n
F
1
v
alu
e
o
f
0
.
9
8
0
,
in
d
icate
s
th
at
in
t
h
e
r
ed
u
ce
d
f
ea
tu
r
e
s
p
ac
e
o
f
n
i
n
e
f
ea
tu
r
es,
b
o
th
ar
ch
ite
ctu
r
es
ar
r
iv
e
at
s
i
m
ilar
n
o
n
l
in
ea
r
r
ep
r
esen
tatio
n
s
,
w
it
h
th
e
r
ec
u
r
r
en
t
n
et
w
o
r
k
a
c
co
m
p
li
s
h
in
g
th
i
s
th
r
o
u
g
h
r
ec
u
r
r
en
t
tr
an
s
f
o
r
m
a
tio
n
s
w
h
ile
th
e
DB
N
r
elies
o
n
u
n
s
u
p
er
v
is
ed
p
r
e
-
tr
ain
in
g
f
o
ll
o
w
ed
b
y
f
i
n
e
-
t
u
n
i
n
g
.
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
.
3
,
J
u
n
e
20
26
:
9
2
6
-
9
32
930
R
an
d
o
m
f
o
r
est
’
s
ad
v
a
n
ta
g
e
o
v
er
g
r
ad
ie
n
t
b
o
o
s
tin
g
(
Δ
F1
=
0
.
0
0
5
)
is
co
n
s
i
s
te
n
t
w
i
th
th
e
e
x
p
ec
tatio
n
th
at
b
ag
g
i
n
g
-
b
ased
v
ar
ia
n
ce
r
ed
u
ctio
n
i
s
m
o
r
e
b
e
n
ef
ic
ial
t
h
an
b
o
o
s
tin
g
-
b
ased
b
ias
r
ed
u
cti
o
n
w
h
e
n
t
h
e
f
ea
t
u
r
e
s
p
ac
e
is
co
m
p
ac
t
an
d
cla
s
s
b
o
u
n
d
ar
ies
ar
e
w
ell
-
s
ep
ar
ated
b
y
MI
-
s
elec
ted
f
ea
t
u
r
es.
T
h
e
lo
wer
p
r
ec
is
io
n
o
f
M
L
m
o
d
el
s
r
elati
v
e
to
D
L
m
o
d
els
,
d
esp
ite
co
m
p
ar
ab
le
r
ec
all,
r
ef
lects
th
e
a
s
y
m
m
e
tr
ic
i
m
p
ac
t
o
f
clas
s
i
m
b
a
lan
ce
o
n
p
r
ec
is
io
n
:
w
it
h
~8
0
%
b
en
ig
n
f
lo
w
s
,
e
v
e
n
a
s
m
all
FP
r
ate
tr
an
s
late
s
to
a
lar
g
e
n
u
m
b
er
o
f
f
alse
alar
m
s
r
elativ
e
to
T
P
s
,
d
is
p
r
o
p
o
r
tio
n
atel
y
p
e
n
alis
in
g
p
r
ec
is
io
n
w
h
ile
lea
v
i
n
g
r
ec
all
r
elati
v
el
y
u
n
a
f
f
ec
ted
.
U
n
d
er
s
in
g
le
-
ti
m
es
tep
in
p
u
t,
L
ST
M
an
d
GR
U
o
p
er
ate
ef
f
ec
ti
v
el
y
as
f
ee
d
f
o
r
w
ar
d
n
et
w
o
r
k
s
,
s
i
n
c
e
th
eir
h
id
d
en
s
tates
n
ev
er
ac
cu
m
u
late
te
m
p
o
r
al
in
f
o
r
m
at
io
n
a
p
r
o
to
co
l
lim
itatio
n
,
n
o
t
a
m
o
d
el
li
m
i
tatio
n
.
T
h
eir
F
1
v
alu
es
(
0
.
9
4
4
-
0
.
9
4
7
)
s
h
o
u
ld
th
er
e
f
o
r
e
b
e
in
ter
p
r
eted
as lo
w
er
b
o
u
n
d
s
o
n
r
ec
u
r
r
en
t
m
o
d
el
p
er
f
o
r
m
a
n
ce
.
4
.
3
.
Cro
s
s
-
v
a
lid
a
t
io
n sta
bil
it
y
T
ab
le
6
p
r
esen
ts
f
i
v
e
-
f
o
ld
C
V
ac
cu
r
ac
y
(
HM
M
e
x
cl
u
d
ed
f
r
o
m
f
o
ld
-
b
ased
C
V)
.
Sta
n
d
ar
d
d
ev
iatio
n
s
o
f
0
.
0
0
1
-
0
.
0
0
2
in
d
icate
s
tab
le
p
er
f
o
r
m
an
ce
w
it
h
i
n
th
is
d
ata
s
et,
th
o
u
g
h
lo
w
cr
o
s
s
-
f
o
ld
v
ar
ian
ce
f
r
o
m
a
s
in
g
le
d
ataset
w
it
h
r
an
d
o
m
s
p
lit
tin
g
d
o
es
n
o
t
p
r
ec
lu
d
e
d
ataset
-
s
p
e
cif
ic
ef
f
ec
t
s
o
r
g
u
ar
an
tee
g
e
n
er
alis
atio
n
to
o
th
er
tr
af
f
ic
d
is
tr
ib
u
tio
n
s
.
T
ab
le
6
.
Fiv
e
-
f
o
ld
s
t
r
atif
ied
C
V
ac
cu
r
ac
y
M
o
d
e
l
F
o
l
d
1
F
o
l
d
2
F
o
l
d
3
F
o
l
d
4
F
o
l
d
5
M
e
a
n
±
S
t
d
C
N
N
0
.
9
9
5
0
.
9
9
3
0
.
9
9
6
0
.
9
9
4
0
.
9
9
5
0
.
9
9
5
±
0
.
0
0
1
R
N
N
0
.
9
9
4
0
.
9
9
2
0
.
9
9
5
0
.
9
9
3
0
.
9
9
4
0
.
9
9
4
±
0
.
0
0
1
D
B
N
0
.
9
9
4
0
.
9
9
2
0
.
9
9
5
0
.
9
9
3
0
.
9
9
3
0
.
9
9
3
±
0
.
0
0
1
R
a
n
d
o
m fo
r
e
st
0
.
9
8
9
0
.
9
8
6
0
.
9
9
0
0
.
9
8
7
0
.
9
8
8
0
.
9
8
8
±
0
.
0
0
2
G
r
a
d
i
e
n
t
b
o
o
st
i
n
g
0
.
9
8
7
0
.
9
8
4
0
.
9
8
8
0
.
9
8
5
0
.
9
8
6
0
.
9
8
6
±
0
.
0
0
2
M
L
P
c
l
a
ssi
f
i
e
r
0
.
9
8
5
0
.
9
8
2
0
.
9
8
6
0
.
9
8
3
0
.
9
8
4
0
.
9
8
4
±
0
.
0
0
2
D
e
c
i
si
o
n
t
r
e
e
0
.
9
8
0
0
.
9
7
7
0
.
9
8
1
0
.
9
7
8
0
.
9
7
9
0
.
9
7
9
±
0
.
0
0
2
L
S
T
M
0
.
9
8
0
0
.
9
7
7
0
.
9
8
1
0
.
9
7
8
0
.
9
7
9
0
.
9
7
9
±
0
.
0
0
2
G
R
U
0
.
9
7
9
0
.
9
7
6
0
.
9
8
0
0
.
9
7
7
0
.
9
7
8
0
.
9
7
8
±
0
.
0
0
2
L
o
g
i
st
i
c
r
e
g
r
e
ssi
o
n
0
.
9
2
4
0
.
9
1
9
0
.
9
2
5
0
.
9
2
1
0
.
9
2
2
0
.
9
2
2
±
0
.
0
0
2
N
a
ï
v
e
B
a
y
e
s
0
.
8
9
2
0
.
8
8
7
0
.
8
9
3
0
.
8
8
8
0
.
8
9
0
0
.
8
9
0
±
0
.
0
0
2
4
.
4
.
Co
m
p
uta
t
io
na
l c
o
s
t
T
ab
le
7
p
r
esen
ts
t
h
e
tr
ain
in
g
an
d
i
n
f
er
e
n
ce
co
s
ts
as
s
o
ci
ated
w
it
h
th
e
h
ar
d
w
ar
e
e
m
p
lo
y
ed
.
T
h
e
r
an
d
o
m
f
o
r
est
m
o
d
el
r
eq
u
ir
es
ap
p
r
o
x
i
m
atel
y
4
s
ec
o
n
d
s
p
er
tr
ain
in
g
f
o
ld
an
d
ar
o
u
n
d
0
.
3
m
il
lis
ec
o
n
d
s
p
er
s
a
m
p
le
w
it
h
a
m
e
m
o
r
y
f
o
o
t
p
r
in
t
o
f
r
o
u
g
h
l
y
1
8
0
m
e
g
ab
y
tes,
w
h
i
c
h
m
a
y
r
en
d
er
it
s
u
itab
le
f
o
r
r
eso
u
r
ce
-
co
n
s
tr
ain
ed
ed
g
e
n
o
d
es
w
h
er
e
in
ter
p
r
etab
ilit
y
is
v
al
u
ed
th
r
o
u
g
h
it
s
b
u
ilt
-
i
n
f
ea
t
u
r
e
i
m
p
o
r
t
an
ce
r
an
k
in
g
s
.
T
h
e
C
NN
m
o
d
el
ta
k
es
ap
p
r
o
x
i
m
atel
y
4
2
s
ec
o
n
d
s
p
er
f
o
ld
o
n
th
e
g
r
ap
h
ic
s
p
r
o
ce
s
s
in
g
u
n
it
an
d
ar
o
u
n
d
1
.
2
m
illi
s
ec
o
n
d
s
p
er
s
a
m
p
le
w
it
h
a
m
e
m
o
r
y
f
o
o
tp
r
in
t
o
f
r
o
u
g
h
l
y
8
m
e
g
ab
y
tes,
ac
h
ie
v
i
n
g
th
e
h
i
g
h
est
F
1
v
al
u
e
w
it
h
i
n
th
i
s
d
ataset.
T
h
e
d
if
f
er
en
ce
in
F
1
b
et
w
ee
n
t
h
e
t
w
o
m
o
d
els
is
m
o
d
est
at
0
.
0
1
5
,
an
d
th
e
p
r
ac
tical
s
elec
tio
n
o
f
a
m
o
d
el
w
o
u
ld
d
ep
en
d
o
n
d
ep
lo
y
m
e
n
t
co
n
s
tr
ai
n
ts
,
i
n
ter
p
r
etab
ilit
y
r
eq
u
ir
e
m
e
n
t
s
,
an
d
th
e
ac
ce
p
tab
le
f
alse
alar
m
r
ate.
T
h
ese
m
ea
s
u
r
e
m
e
n
t
s
r
e
m
ai
n
tied
to
th
e
s
p
ec
if
ic
h
ar
d
w
ar
e
u
s
ed
an
d
n
ec
ess
itate
r
e
-
ev
alu
a
tio
n
o
n
t
h
e
i
n
ten
d
ed
tar
g
et
p
latf
o
r
m
s
.
T
ab
le
7
.
C
o
m
p
u
ta
tio
n
al
co
s
t (
h
ar
d
w
ar
e
-
s
p
ec
i
f
ic
esti
m
ate
s
)
Mo
d
e
l
T
r
a
i
n
/
f
o
l
d
(
s)
I
n
f
e
r
/
samp
l
e
(
ms)
M
e
mo
r
y
(
M
B
)
N
o
t
e
s
R
a
n
d
o
m
f
o
r
e
st
~4
~
0
.
3
~
1
8
0
L
o
w
;
i
n
t
e
r
p
r
e
t
a
b
l
e
G
r
a
d
i
e
n
t
b
o
o
st
i
n
g
~
1
2
~
0
.
5
~
2
1
0
L
o
w
M
L
P
c
l
a
ssi
f
i
e
r
~8
~
0
.
1
~
5
0
V
e
r
y
l
o
w
i
n
f
e
r
e
n
c
e
D
e
c
i
si
o
n
t
r
e
e
~1
<
0
.
1
~
2
0
L
o
w
e
st
o
v
e
r
a
l
l
L
o
g
i
st
i
c
r
e
g
r
e
ssi
o
n
<1
<
0
.
1
~
1
0
M
i
n
i
m
a
l
c
o
st
N
a
ï
v
e
B
a
y
e
s
<1
<
0
.
1
~5
M
i
n
i
m
a
l
c
o
st
C
N
N
~
4
2
~
1
.
2
~8
G
P
U
r
e
q
u
i
r
e
d
R
N
N
~
3
5
~
0
.
9
~6
G
P
U
r
e
q
u
i
r
e
d
D
B
N
~
1
4
2
~
1
.
5
~
1
2
H
i
g
h
e
st
t
r
a
i
n
i
n
g
c
o
st
L
S
T
M
~
3
8
~
1
.
1
~7
G
P
U
r
e
q
u
i
r
e
d
G
R
U
~
3
3
~
0
.
9
~6
G
P
U
r
e
q
u
i
r
e
d
H
M
M
~5
~
0
.
8
~2
L
o
w
e
st
a
c
c
u
r
a
c
y
4
.
5
.
L
i
m
it
a
t
io
ns
Sev
er
al
li
m
itatio
n
s
s
h
o
u
ld
b
e
n
o
ted
.
A
ll
ex
p
er
i
m
en
t
s
u
s
ed
a
s
in
g
le
s
i
m
u
la
ted
5
G
d
ataset;
g
en
er
ali
s
atio
n
to
o
p
er
atio
n
al
n
et
w
o
r
k
s
w
it
h
d
iv
er
s
e
attac
k
p
r
o
f
iles
r
e
m
ain
s
u
n
k
n
o
w
n
.
R
a
n
d
o
m
d
ata
s
p
litt
in
g
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
Leve
r
a
g
in
g
a
r
tifi
cia
l in
tellig
en
ce
fo
r
d
etec
tio
n
o
f
d
en
ia
l
-
of
-
s
ervice
a
tta
ck
s
in
5
G
n
etw
o
r
k
…
(
B
a
s
ee
l A
l
-
A
li
)
931
w
a
s
e
m
p
lo
y
ed
;
te
m
p
o
r
al
s
p
litt
in
g
w
o
u
ld
p
r
o
v
id
e
a
s
tr
icter
ev
alu
at
io
n
o
f
r
o
b
u
s
t
n
es
s
ag
a
in
s
t
co
n
ce
p
t
d
r
if
t.
R
ec
u
r
r
en
t
m
o
d
els
w
er
e
ev
al
u
ated
u
n
d
er
s
i
n
g
le
-
ti
m
e
s
tep
in
p
u
t,
w
h
ic
h
d
o
es
n
o
t
ex
er
ci
s
e
th
e
ir
s
eq
u
e
n
tia
l
m
o
d
ell
in
g
ca
p
ab
ilit
y
.
I
n
ter
p
r
etab
ilit
y
a
n
al
y
s
is
(
e.
g
.
,
s
h
ap
le
y
ad
d
itiv
e
ex
p
la
n
atio
n
s
/
SH
A
P
)
w
a
s
n
o
t
co
n
d
u
cted
,
an
d
co
m
p
u
tat
io
n
al
co
s
t
m
ea
s
u
r
e
m
e
n
ts
ar
e
h
ar
d
w
ar
e
-
s
p
ec
i
f
ic.
T
h
ese
co
n
s
tr
ain
t
s
s
co
p
e
t
h
e
f
i
n
d
i
n
g
s
to
th
e
p
r
esen
t d
ataset
an
d
p
r
o
to
co
l.
5.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
p
r
o
p
o
s
ed
a
leak
ag
e
-
a
w
ar
e
ev
al
u
atio
n
p
r
o
to
co
l
f
o
r
Do
S
d
etec
tio
n
in
5
G
n
et
wo
r
k
s
licin
g
an
d
ap
p
lied
it
to
b
en
ch
m
a
r
k
t
w
elv
e
ar
c
h
itect
u
r
es.
T
h
e
p
r
o
to
c
o
l
en
s
u
r
es
th
at
MI
f
ea
tu
r
e
s
elec
tio
n
,
n
o
r
m
ali
s
atio
n
,
a
n
d
SM
OT
E
ar
e
co
n
f
i
n
ed
to
tr
ai
n
i
n
g
p
ar
titi
o
n
s
,
w
i
th
MI
r
ec
o
m
p
u
ted
p
er
CV
f
o
ld
ad
d
r
ess
i
n
g
a
m
et
h
o
d
o
lo
g
ical
g
ap
i
n
e
x
is
tin
g
5
G
I
DS
b
e
n
ch
m
ar
k
s
.
W
it
h
i
n
t
h
is
d
ata
s
et,
C
NN
ac
h
iev
ed
th
e
h
ig
h
e
s
t
F
1
-
s
co
r
e
(
0
.
9
8
3
,
FDR
=0
.
0
2
7
)
an
d
r
an
d
o
m
f
o
r
est
attai
n
ed
F
1
=0
.
9
6
8
w
it
h
lo
w
er
co
m
p
u
tat
io
n
al
co
s
t.
C
NN
m
a
y
s
u
it
GP
U
-
eq
u
ip
p
ed
co
r
e
n
et
w
o
r
k
en
v
ir
o
n
m
e
n
t
s
,
w
h
i
le
r
an
d
o
m
f
o
r
est
o
f
f
er
s
a
f
av
o
u
r
ab
le
ac
cu
r
ac
y
i
n
ter
p
r
etab
ilit
y
tr
ad
e
-
o
f
f
f
o
r
ed
g
e
-
co
n
s
tr
ain
ed
s
ettin
g
s
.
A
ll
f
i
n
d
i
n
g
s
ar
e
d
at
aset
-
s
p
ec
if
ic
;
v
alid
atio
n
w
it
h
d
iv
er
s
e
5
G
tr
af
f
ic
tr
ac
es
an
d
te
m
p
o
r
al
s
p
litt
in
g
is
r
eq
u
ir
ed
b
ef
o
r
e
d
e
p
lo
y
m
en
t.
Fu
tu
r
e
w
o
r
k
s
h
o
u
ld
i
n
co
r
p
o
r
ate
m
u
lti
-
d
ataset
ev
alu
a
tio
n
,
SH
A
P
-
b
ased
i
n
t
er
p
r
etab
ilit
y
a
n
al
y
s
is
,
a
n
d
m
u
lti
-
ti
m
estep
r
ec
u
r
r
en
t
m
o
d
el
ev
al
u
atio
n
o
n
te
m
p
o
r
all
y
s
tr
u
ct
u
r
ed
tr
af
f
ic
d
ata.
RE
F
E
R
E
NC
E
S
[
1
]
A
.
I
.
A
h
mad
,
F
.
O
saso
n
a
,
S
.
O
.
D
a
w
o
d
u
,
O
.
C
.
O
b
i
,
A
.
C
.
A
n
y
a
n
w
u
,
a
n
d
S
.
O
n
w
u
si
n
k
w
u
e
,
“
Eme
r
g
i
n
g
5
G
t
e
c
h
n
o
l
o
g
y
:
A
r
e
v
i
e
w
o
f
i
t
s
f
a
r
-
r
e
a
c
h
i
n
g
i
m
p
l
i
c
a
t
i
o
n
s
f
o
r
c
o
mm
u
n
i
c
a
t
i
o
n
a
n
d
se
c
u
r
i
t
y
,
”
W
o
rl
d
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
Re
s
e
a
rc
h
a
n
d
Re
v
i
e
w
s
,
v
o
l
.
2
1
,
n
o
.
1
,
p
p
.
2
4
7
4
–
2
4
8
6
,
2
0
2
4
,
d
o
i
:
1
0
.
3
0
5
7
4
/
w
j
a
r
r
.
2
0
2
4
.
2
1
.
1
.
0
3
4
6
.
[
2
]
Z
.
Z
h
u
,
X
.
L
i
,
a
n
d
Z
.
C
h
u
,
“
T
h
r
e
e
maj
o
r
o
p
e
r
a
t
i
n
g
sce
n
a
r
i
o
s
o
f
5
G
:
e
M
B
B
,
mM
T
C
,
U
R
L
L
C
,
”
I
n
t
e
l
l
i
g
e
n
t
S
e
n
s
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
f
o
r In
t
e
r
n
e
t
o
f
Ev
e
ry
t
h
i
n
g
,
p
p
.
1
5
–
7
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
b
9
7
8
-
0
-
32
-
3
8
5
6
5
5
-
3
.
0
0
0
0
6
-
0.
[
3
]
A
.
S
i
n
g
h
a
n
d
B
.
B
.
G
u
p
t
a
,
“
D
i
s
t
r
i
b
u
t
e
d
D
e
n
i
a
l
-
of
-
S
e
r
v
i
c
e
(
D
D
o
S
)
A
t
t
a
c
k
s
a
n
d
D
e
f
e
n
se
M
e
c
h
a
n
i
sms
i
n
V
a
r
i
o
u
s
W
e
b
-
En
a
b
l
e
d
C
o
mp
u
t
i
n
g
P
l
a
t
f
o
r
ms,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
n
S
e
m
a
n
t
i
c
We
b
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
,
v
o
l
.
1
8
,
n
o
.
1
,
p
p
.
1
–
4
3
,
A
p
r
.
2
0
2
2
,
d
o
i
:
1
0
.
4
0
1
8
/
i
j
sw
i
s.2
9
7
1
4
3
.
[
4
]
D
.
O
n
o
j
a
,
M
.
H
i
t
c
h
e
n
s,
a
n
d
R
.
S
h
a
n
k
a
r
a
n
,
“
D
D
o
S
T
h
r
e
a
t
s
a
n
d
S
o
l
u
t
i
o
n
s
f
o
r
5
G
-
En
a
b
l
e
d
I
o
T
N
e
t
w
o
r
k
s,
”
S
m
a
rt
S
e
n
s
o
rs,
Me
a
su
r
e
m
e
n
t
a
n
d
I
n
st
r
u
m
e
n
t
a
t
i
o
n
,
S
p
r
i
n
g
e
r
I
n
t
e
r
n
a
t
i
o
n
a
l
P
u
b
l
i
s
h
i
n
g
,
v
o
l
.
4
3
,
p
p
.
1
1
5
–
1
3
3
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
0
8
2
7
0
-
2
_
5
.
[
5
]
V
.
A
.
C
u
n
h
a
e
t
a
l
.
,
“
N
e
t
w
o
r
k
sl
i
c
i
n
g
se
c
u
r
i
t
y
:
C
h
a
l
l
e
n
g
e
s
a
n
d
d
i
r
e
c
t
i
o
n
s,
”
I
n
t
e
r
n
e
t
T
e
c
h
n
o
l
o
g
y
L
e
t
t
e
rs
,
v
o
l
.
2
,
n
o
.
5
,
S
e
p
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
2
/
i
t
l
2
.
1
2
5
.
[
6
]
G
.
E.
P
.
K
u
mar,
M
.
L
y
d
i
a
,
a
n
d
Y
.
L
e
v
r
o
n
,
“
S
e
c
u
r
i
t
y
C
h
a
l
l
e
n
g
e
s
i
n
5
G
a
n
d
I
o
T
N
e
t
w
o
r
k
s:
A
R
e
v
i
e
w
,
”
EAI/
S
p
r
i
n
g
e
r
I
n
n
o
v
a
t
i
o
n
s
i
n
C
o
m
m
u
n
i
c
a
t
i
o
n
a
n
d
C
o
m
p
u
t
i
n
g
,
S
p
r
i
n
g
e
r
I
n
t
e
r
n
a
t
i
o
n
a
l
P
u
b
l
i
sh
i
n
g
,
p
p
.
1
–
1
3
,
O
c
t
.
2
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
7
9
7
6
6
-
9
_
1
.
[
7
]
K
.
N
o
o
r
,
A
.
L
.
I
mo
i
z
e
,
C
.
-
T
.
L
i
,
a
n
d
C
.
-
Y
.
W
e
n
g
,
“
A
R
e
v
i
e
w
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
a
n
d
T
r
a
n
sf
e
r
L
e
a
r
n
i
n
g
S
t
r
a
t
e
g
i
e
s
f
o
r
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
S
y
st
e
ms i
n
5
G
a
n
d
B
e
y
o
n
d
,
”
M
a
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
3
,
n
o
.
7
,
p
.
1
0
8
8
,
M
a
r
.
2
0
2
5
,
d
o
i
:
1
0
.
3
3
9
0
/
m
a
t
h
1
3
0
7
1
0
8
8
.
[
8
]
C
.
H
a
mr
o
u
n
,
A
.
F
l
a
d
e
n
mu
l
l
e
r
,
M
.
P
a
r
i
e
n
t
e
,
a
n
d
G
.
P
u
j
o
l
l
e
,
“
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
i
n
5
G
a
n
d
W
i
-
F
i
N
e
t
w
o
r
k
s:
A
S
u
r
v
e
y
o
f
C
u
r
r
e
n
t
M
e
t
h
o
d
s,
C
h
a
l
l
e
n
g
e
s,
a
n
d
P
e
r
sp
e
c
t
i
v
e
s,
”
i
n
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
3
,
p
p
.
4
0
9
5
0
-
4
0
9
7
6
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
5
.
3
5
4
6
3
3
8
.
[
9
]
M
.
S
.
K
h
a
n
,
B
.
F
a
r
z
a
n
e
h
,
N
.
S
h
a
h
r
i
a
r
,
a
n
d
M
.
M
.
H
a
sa
n
,
“
D
o
S
/
D
D
o
S
A
t
t
a
c
k
D
a
t
a
se
t
o
f
5
G
N
e
t
w
o
r
k
S
l
i
c
i
n
g
,
”
I
EEE
D
a
t
a
p
o
r
t
,
S
e
p
t
e
mb
e
r
2
5
,
2
0
2
3
,
d
o
i
:
1
0
.
2
1
2
2
7
/
3
2
k
1
-
d
r
1
2
[
1
0
]
A
.
I
man
b
a
y
e
v
e
t
a
l
.
,
“
R
e
se
a
r
c
h
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
ms
f
o
r
t
h
e
D
e
v
e
l
o
p
me
n
t
o
f
I
n
t
r
u
s
i
o
n
D
e
t
e
c
t
i
o
n
S
y
st
e
ms
i
n
5
G
M
o
b
i
l
e
N
e
t
w
o
r
k
s a
n
d
B
e
y
o
n
d
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
2
,
n
o
.
2
4
,
p
.
9
9
5
7
,
D
e
c
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
2
2
4
9
9
5
7
.
[
1
1
]
T
.
S
a
r
a
n
y
a
,
S
.
S
r
i
d
e
v
i
,
C
.
D
e
i
sy
,
T
.
D
.
C
h
u
n
g
,
a
n
d
M
.
K
.
A
.
A
.
K
h
a
n
,
“
P
e
r
f
o
r
man
c
e
A
n
a
l
y
si
s
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
ms
i
n
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
S
y
st
e
m:
A
R
e
v
i
e
w
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
7
1
,
p
p
.
1
2
5
1
–
1
2
6
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s
.
2
0
2
0
.
0
4
.
1
3
3
.
[
1
2
]
M
.
R
o
d
r
í
g
u
e
z
,
Á
.
A
l
e
san
c
o
,
L
.
M
e
h
a
v
i
l
l
a
,
a
n
d
J
.
G
a
r
c
í
a
,
“
Ev
a
l
u
a
t
i
o
n
o
f
M
a
c
h
i
n
e
L
e
a
r
n
i
n
g
T
e
c
h
n
i
q
u
e
s
f
o
r
T
r
a
f
f
i
c
F
l
o
w
-
B
a
se
d
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
,
”
S
e
n
s
o
rs
,
v
o
l
.
2
2
,
n
o
.
2
3
,
p
.
9
3
2
6
,
N
o
v
.
2
0
2
2
,
d
o
i
:
1
0
.
3
3
9
0
/
s2
2
2
3
9
3
2
6
.
[
1
3
]
L
.
K
.
V
a
sh
i
sh
t
h
a
a
n
d
K
.
C
h
a
t
t
e
r
j
e
e
,
“
S
t
r
e
n
g
t
h
e
n
i
n
g
c
y
b
e
r
se
c
u
r
i
t
y
:
Te
st
C
l
o
u
d
I
D
S
d
a
t
a
se
t
a
n
d
S
p
a
r
k
S
h
i
e
l
d
a
l
g
o
r
i
t
h
m
f
o
r
r
o
b
u
s
t
t
h
r
e
a
t
d
e
t
e
c
t
i
o
n
,
”
C
o
m
p
u
t
e
rs &
S
e
c
u
r
i
t
y
,
v
o
l
.
1
5
1
,
p
.
1
0
4
3
0
8
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
se
.
2
0
2
4
.
1
0
4
3
0
8
.
[
1
4
]
M
.
R
a
g
a
b
,
S
.
M
.
A
l
sh
a
mm
a
r
i
,
L
.
A
.
M
a
g
h
r
a
b
i
,
D
.
A
l
sal
man
,
T
.
A
l
t
h
a
q
a
f
i
,
a
n
d
A
.
A
.
-
M
.
A
L
-
G
h
a
md
i
,
“
R
o
b
u
st
D
D
o
S
A
t
t
a
c
k
D
e
t
e
c
t
i
o
n
U
s
i
n
g
P
i
e
c
e
w
i
se
H
a
r
r
i
s
H
a
w
k
s
O
p
t
i
mi
z
e
r
w
i
t
h
D
e
e
p
L
e
a
r
n
i
n
g
f
o
r
a
S
e
c
u
r
e
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
E
n
v
i
r
o
n
me
n
t
,
”
Ma
t
h
e
m
a
t
i
c
s
,
v
o
l
.
1
1
,
n
o
.
2
1
,
p
.
4
4
4
8
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
mat
h
1
1
2
1
4
4
4
8
.
[
1
5
]
M
.
C
h
e
r
i
a
n
a
n
d
S
.
L
.
V
a
r
ma,
“
S
e
c
u
r
e
S
D
N
–
I
o
T
F
r
a
mew
o
r
k
f
o
r
D
D
o
S
A
t
t
a
c
k
D
e
t
e
c
t
i
o
n
U
si
n
g
D
e
e
p
L
e
a
r
n
i
n
g
a
n
d
C
o
u
n
t
e
r
B
a
se
d
A
p
p
r
o
a
c
h
,
”
J
o
u
r
n
a
l
o
f
N
e
t
w
o
r
k
a
n
d
S
y
st
e
m
s
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
3
1
,
n
o
.
3
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
9
2
2
-
0
2
3
-
0
9
7
4
9
-
w.
[
1
6
]
F
.
M
.
A
sw
a
d
,
A
.
M
.
S
.
A
h
me
d
,
N
.
A
.
M
.
A
l
h
a
mm
a
d
i
,
B
.
A
.
K
h
a
l
a
f
,
a
n
d
S
.
A
.
M
o
st
a
f
a
,
“
D
e
e
p
l
e
a
r
n
i
n
g
i
n
d
i
s
t
r
i
b
u
t
e
d
d
e
n
i
a
l
-
of
-
se
r
v
i
c
e
a
t
t
a
c
k
s
d
e
t
e
c
t
i
o
n
me
t
h
o
d
f
o
r
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
n
e
t
w
o
r
k
s,
”
J
o
u
rn
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
,
v
o
l
.
3
2
,
n
o
.
1
,
J
a
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
5
1
5
/
j
i
sy
s
-
2
0
2
2
-
0
1
5
5
.
[
1
7
]
M
.
A
l
j
e
b
r
e
e
n
,
H
.
A
.
M
e
n
g
a
sh
,
M
.
A
.
A
r
a
si
,
S
.
S
.
A
l
j
a
me
e
l
,
A
.
S
.
S
a
l
a
ma,
a
n
d
M
.
A
.
H
a
mz
a
,
“
En
h
a
n
c
i
n
g
D
D
o
S
A
t
t
a
c
k
D
e
t
e
c
t
i
o
n
U
si
n
g
S
n
a
k
e
O
p
t
i
mi
z
e
r
w
i
t
h
E
n
se
mb
l
e
L
e
a
r
n
i
n
g
o
n
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
E
n
v
i
r
o
n
me
n
t
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
1
,
p
p
.
1
0
4
7
4
5
–
1
0
4
7
5
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
a
c
c
e
ss.2
0
2
3
.
3
3
1
8
3
1
6
.
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
.
3
,
J
u
n
e
20
26
:
9
2
6
-
9
32
932
[
1
8
]
S
.
A
.
K
h
a
n
d
a
y
,
H
.
F
a
t
i
ma
,
a
n
d
N
.
R
a
k
e
sh
,
“
I
mp
l
e
me
n
t
a
t
i
o
n
o
f
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
mo
d
e
l
f
o
r
D
D
o
S
a
t
t
a
c
k
s
i
n
l
i
g
h
t
w
e
i
g
h
t
I
o
T
n
e
t
w
o
r
k
s,
”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
1
5
,
p
.
1
1
9
3
3
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
2
.
1
1
9
3
3
0
.
[
1
9
]
M
.
I
.
T
.
H
u
ssa
n
,
G
.
V
.
R
e
d
d
y
,
P
.
T
.
A
n
i
t
h
a
,
A
.
K
a
n
a
g
a
r
a
j
,
a
n
d
P
.
N
a
r
e
s
h
,
“
D
D
o
S
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
i
n
I
o
T
e
n
v
i
r
o
n
me
n
t
u
s
i
n
g
o
p
t
i
m
i
z
e
d
El
ma
n
r
e
c
u
r
r
e
n
t
n
e
u
r
a
l
n
e
t
w
o
r
k
s
b
a
se
d
o
n
c
h
a
o
t
i
c
b
a
c
t
e
r
i
a
l
c
o
l
o
n
y
o
p
t
i
mi
z
a
t
i
o
n
,
”
C
l
u
st
e
r
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
7
,
n
o
.
4
,
p
p
.
4
4
6
9
–
4
4
9
0
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
5
8
6
-
0
2
3
-
0
4
1
8
7
-
4.
[
2
0
]
P
.
B
h
a
l
e
,
D
.
R
.
C
h
o
w
d
h
u
r
y
,
S
.
B
i
sw
a
s,
a
n
d
S
.
N
a
n
d
i
,
“
O
P
TI
M
I
S
T
:
L
i
g
h
t
w
e
i
g
h
t
a
n
d
T
r
a
n
sp
a
r
e
n
t
I
D
S
W
i
t
h
O
p
t
i
mu
m
P
l
a
c
e
me
n
t
S
t
r
a
t
e
g
y
t
o
M
i
t
i
g
a
t
e
M
i
x
e
d
-
R
a
t
e
D
D
o
S
A
t
t
a
c
k
s
i
n
I
o
T
N
e
t
w
o
r
k
s,
”
i
n
I
EE
E
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
J
o
u
r
n
a
l
,
v
o
l
.
1
0
,
n
o
.
1
0
,
p
p
.
8
3
5
7
-
8
3
7
0
,
1
5
M
a
y
1
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
J
I
O
T
.
2
0
2
3
.
3
2
3
4
5
3
0
.
[
2
1
]
K
.
K
e
t
h
i
n
e
n
i
a
n
d
G
.
P
r
a
d
e
e
p
i
n
i
,
“
I
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
i
n
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
-
b
a
se
d
sm
a
r
t
f
a
r
mi
n
g
u
s
i
n
g
h
y
b
r
i
d
d
e
e
p
l
e
a
r
n
i
n
g
f
r
a
mew
o
r
k
,
”
C
l
u
s
t
e
r
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
7
,
n
o
.
2
,
p
p
.
1
7
1
9
–
1
7
3
2
,
J
u
n
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
5
8
6
-
023
-
0
4
0
5
2
-
4.
[
2
2
]
X.
-
H
.
N
g
u
y
e
n
a
n
d
K
.
-
H
.
L
e
,
“
R
o
b
u
st
d
e
t
e
c
t
i
o
n
o
f
u
n
k
n
o
w
n
D
o
S
/
D
D
o
S
a
t
t
a
c
k
s
i
n
I
o
T
n
e
t
w
o
r
k
s
u
si
n
g
a
h
y
b
r
i
d
l
e
a
r
n
i
n
g
mo
d
e
l
,
”
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
,
v
o
l
.
2
3
,
p
.
1
0
0
8
5
1
,
O
c
t
.
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
o
t
.
2
0
2
3
.
1
0
0
8
5
1
.
[
2
3
]
A
.
D
.
A
g
u
r
u
a
n
d
S
.
B
.
Er
u
k
a
l
a
,
“
A
l
i
g
h
t
w
e
i
g
h
t
m
u
l
t
i
-
v
e
c
t
o
r
D
D
o
S
d
e
t
e
c
t
i
o
n
f
r
a
me
w
o
r
k
f
o
r
I
o
T
-
e
n
a
b
l
e
d
mo
b
i
l
e
h
e
a
l
t
h
i
n
f
o
r
mat
i
c
s
sy
st
e
ms u
si
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
6
6
2
,
p
.
1
2
0
2
0
9
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
s.2
0
2
4
.
1
2
0
2
0
9
.
[
2
4
]
S
.
Y
a
r
a
s
a
n
d
M
.
D
e
n
e
r
,
“
I
o
T
-
B
a
se
d
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
S
y
st
e
m
U
si
n
g
N
e
w
H
y
b
r
i
d
D
e
e
p
L
e
a
r
n
i
n
g
A
l
g
o
r
i
t
h
m
,
”
El
e
c
t
r
o
n
i
c
s
,
v
o
l
.
1
3
,
n
o
.
6
,
p
.
1
0
5
3
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
3
0
6
1
0
5
3
.
[
2
5
]
M
.
W
a
z
i
d
,
A
.
K
.
D
a
s
,
S
.
S
h
e
t
t
y
,
P
.
G
o
p
e
,
a
n
d
J
.
J.
P
.
C
.
R
o
d
r
i
g
u
e
s,
“
S
e
c
u
r
i
t
y
i
n
5
G
-
En
a
b
l
e
d
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
C
o
mm
u
n
i
c
a
t
i
o
n
:
I
ssu
e
s,
C
h
a
l
l
e
n
g
e
s,
a
n
d
F
u
t
u
r
e
R
e
se
a
r
c
h
R
o
a
d
ma
p
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
4
4
6
6
–
4
4
8
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
a
c
c
e
ss.
2
0
2
0
.
3
0
4
7
8
9
5
.
[
2
6
]
Ž
.
Ð
.
V
u
j
o
v
i
c
,
“
C
l
a
ssi
f
i
c
a
t
i
o
n
M
o
d
e
l
Ev
a
l
u
a
t
i
o
n
M
e
t
r
i
c
s
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Ad
v
a
n
c
e
d
C
o
m
p
u
t
e
r S
c
i
e
n
c
e
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
(
I
J
AC
S
A)
,
v
o
l
.
1
2
,
n
o
.
6
,
2
0
2
1
,
d
o
i
:
1
0
.
1
4
5
6
9
/
i
j
a
c
sa.
2
0
2
1
.
0
1
2
0
6
7
0
.
Evaluation Warning : The document was created with Spire.PDF for Python.