I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
38
,
No
.
1
,
A
p
r
il
20
25
,
p
p
.
1
33
~1
44
I
SS
N:
2
502
-
4
7
52
,
DOI
: 1
0
.
1
1
5
9
1
/ijee
cs
.v
38.
i
1
.
p
p
1
33
-
1
44
133
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs
.
ia
esco
r
e.
co
m
SM
O
TE
t
ree
-
ba
s
ed auto
enco
der
m
ulti
-
sta
g
e d
etec
tion for ma
n
-
in
-
the
-
middle
in S
CADA
F
re
s
k
a
Ro
la
ns
a
1
,2
,
J
a
zi
E
k
o
I
s
t
iy
a
nto
1
,
Af
ia
ha
y
a
t
i
1
,
Auf
a
cla
v
Z
a
t
u K
us
um
a
F
risk
y
1
1
D
e
p
a
r
t
me
n
t
o
f
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
E
l
e
c
t
r
o
n
i
c
s,
F
a
c
u
l
t
y
o
f
M
a
t
h
e
ma
t
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
s
i
t
a
s Ga
d
j
a
h
M
a
d
a
,
Y
o
g
y
a
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
El
e
c
t
r
i
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
P
o
n
t
i
a
n
a
k
S
t
a
t
e
P
o
l
y
t
e
c
h
n
i
c
,
P
o
n
t
i
a
n
a
k
,
I
n
d
o
n
e
si
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
May
29
,
2
0
2
4
R
ev
is
ed
Oct
22
,
202
4
Acc
ep
ted
Oct
30
,
2
0
2
4
S
e
c
u
rit
y
i
n
c
id
e
n
ts
targ
e
ti
n
g
s
u
p
e
rv
is
o
ry
c
o
n
tr
o
l
a
n
d
d
a
ta
a
c
q
u
isit
io
n
(S
CAD
A)
in
fra
stru
c
tu
re
a
re
in
c
re
a
sin
g
,
wh
ich
c
a
n
lea
d
t
o
d
isa
ste
rs
su
c
h
a
s
p
ip
e
li
n
e
fires
o
r
e
v
e
n
l
o
st
of
l
iv
e
s.
M
a
n
-
in
-
t
h
e
-
m
id
d
le
(M
IT
M
)
a
tt
a
c
k
s
re
p
re
se
n
t
a
sig
n
if
ica
n
t
th
re
a
t
t
o
th
e
se
c
u
rit
y
a
n
d
re
li
a
b
il
it
y
o
f
S
CAD
A.
De
tec
ti
n
g
M
I
TM
a
tt
a
c
k
s
o
n
th
e
M
o
d
b
u
s
S
CAD
A
n
e
two
rk
s
is
t
h
e
o
b
jec
ti
v
e
o
f
t
h
is
w
o
rk
.
In
a
d
d
it
i
o
n
,
th
is
wo
rk
i
n
tro
d
u
c
e
s
S
M
OTE
t
re
e
-
b
a
se
d
a
u
to
e
n
c
o
d
e
r
m
u
lt
i
-
sta
g
e
d
e
tec
ti
o
n
(S
TA
M
)
u
sin
g
t
h
e
El
e
c
tra
d
a
t
a
se
t.
Th
is
wo
rk
p
ro
p
o
se
s
a
fo
u
r
-
sta
g
e
a
p
p
ro
a
c
h
in
v
o
l
v
i
n
g
d
a
ta
p
re
p
r
o
c
e
ss
in
g
,
d
a
ta
b
a
lan
c
in
g
,
a
n
a
u
to
e
n
c
o
d
e
r,
a
n
d
tr
e
e
c
las
sifica
ti
o
n
fo
r
a
n
o
m
a
ly
d
e
t
e
c
ti
o
n
a
n
d
m
u
lt
i
-
c
las
s
c
las
sifica
ti
o
n
.
I
n
ter
m
s
o
f
a
tt
a
c
k
i
d
e
n
ti
fica
ti
o
n
,
th
e
p
ro
p
o
se
d
m
o
d
e
l
p
e
rfo
rm
s
wit
h
h
ig
h
e
st
p
re
c
isio
n
,
d
e
tec
ti
o
n
ra
te
/rec
a
ll
,
a
n
d
F
1
sc
o
re
.
I
n
p
a
rti
c
u
lar,
t
h
e
m
o
d
e
l
a
c
h
iev
e
s
a
n
F
1
sc
o
re
o
f
1
0
0
%
f
o
r
a
n
o
m
a
ly
d
e
te
c
ti
o
n
a
n
d
a
n
F
1
sc
o
re
o
f
9
9
.
3
7
%
fo
r
m
u
lt
i
-
c
las
s
c
las
sifica
ti
o
n
,
wh
ich
is
p
re
e
m
in
e
n
c
e
to
o
t
h
e
r
m
o
d
e
ls.
M
o
re
o
v
e
r,
th
e
e
n
h
a
n
c
e
d
p
e
rf
o
rm
a
n
c
e
o
f
m
u
lt
i
-
c
las
s
c
las
sific
a
ti
o
n
wit
h
S
TAM
o
n
m
in
o
rit
y
a
tt
a
c
k
c
las
s
e
s
(re
p
lay
a
n
d
re
a
d
)
h
a
s
sh
o
wn
sim
il
a
r
c
h
a
ra
c
teristics
in
fe
a
tu
re
s
a
n
d
a
re
d
u
c
e
d
n
u
m
b
e
r
o
f
m
isc
las
sifica
ti
o
n
s in
t
h
e
se
c
las
se
s
.
K
ey
w
o
r
d
s
:
An
o
m
aly
d
etec
tio
n
Au
to
en
co
d
e
r
Mu
lti
-
class
c
las
s
if
icatio
n
Mu
lti
-
s
tag
e
SC
ADA
SMOT
E
T
r
ee
c
lass
if
icatio
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
:
J
az
i E
k
o
I
s
tiy
an
to
Dep
ar
tm
en
t
o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
lectr
o
n
ics,
Facu
lty
o
f
Ma
th
em
atics a
n
d
Natu
r
al
Sci
en
ce
s
Un
iv
er
s
itas
Gad
jah
Ma
d
a
Yo
g
y
ak
ar
ta,
I
n
d
o
n
esia
E
m
ail:
jazi@
u
g
m
.
ac
.
id
1.
I
NT
RO
D
UCT
I
O
N
A
n
etwo
r
k
co
m
p
r
is
in
g
o
f
v
ar
i
o
u
s
co
m
p
o
n
en
ts
r
esp
o
n
s
ib
le
f
o
r
s
u
p
er
v
is
in
g
a
n
d
c
o
n
tr
o
llin
g
in
d
u
s
tr
ial
p
r
o
ce
s
s
es
is
r
ef
er
r
ed
to
as
s
u
p
er
v
is
o
r
y
c
o
n
tr
o
l
a
n
d
d
ata
ac
q
u
is
itio
n
(
SC
ADA)
.
Usi
n
g
m
o
d
er
n
tech
n
o
lo
g
y
s
u
ch
as
co
m
p
u
ter
s
,
elec
tr
ical
,
m
ec
h
an
ical
s
y
s
tem
s
,
an
d
n
etwo
r
k
in
g
d
e
v
ices,
SC
ADA
is
u
s
ed
in
cr
itical
in
f
r
astru
ctu
r
e
to
m
o
n
ito
r
p
h
y
s
ical
p
r
o
ce
s
s
es
[
1
]
.
I
t
en
c
o
m
p
ass
es
a
r
an
g
e
o
f
h
eter
o
g
e
n
eo
u
s
co
m
p
o
n
en
ts
,
in
clu
d
in
g
r
em
o
te
ter
m
i
n
al
u
n
it
s
(
R
T
U
s
)
,
m
aster
ter
m
in
al
u
n
it
s
(
MT
U
s
)
,
h
u
m
a
n
m
ac
h
i
n
e
in
ter
f
ac
e
(
HM
I
)
,
h
is
to
r
ian
,
p
r
o
g
r
am
m
a
b
le
lo
g
ic
co
n
tr
o
ller
s
(
PLC
s
)
,
s
en
s
o
r
s
,
a
n
d
ac
tu
ato
r
s
.
T
h
e
d
i
v
er
s
ity
o
f
d
ev
ices
em
p
lo
y
ed
b
y
SC
ADA
r
en
d
er
s
s
y
s
tem
s
ec
u
r
ity
m
ain
ten
an
ce
ch
allen
g
in
g
[
2
]
.
I
n
cid
en
ts
lik
e
p
ip
elin
e
f
ir
es,
p
r
o
d
u
ctio
n
p
r
o
ce
s
s
s
h
u
td
o
wn
s
,
a
n
d
n
u
c
lear
r
ea
cto
r
o
u
tag
es
r
esu
ltin
g
f
r
o
m
SC
ADA
m
alf
u
n
ctio
n
s
u
n
d
e
r
s
co
r
e
its
in
cr
ea
s
in
g
ly
p
iv
o
tal
r
o
le
in
cr
i
tical
in
f
r
astru
ctu
r
e
o
p
er
atio
n
s
[
3
]
.
Hea
lth
ca
r
e,
en
er
g
y
s
ec
to
r
,
n
u
clea
r
r
ea
cto
r
s
,
ag
r
icu
ltu
r
e,
tr
a
n
s
p
o
r
tatio
n
,
civ
il,
ch
em
ical
e
n
g
in
ee
r
in
g
,
wate
r
p
lan
ts
,
an
d
r
esear
c
h
h
a
v
e
wid
ely
a
d
o
p
ted
SC
ADA
[
3
]
.
C
o
m
p
ar
e
d
to
o
th
er
s
ec
to
r
s
,
t
h
e
en
er
g
y
s
ec
to
r
is
th
e
m
o
s
t
tar
g
eted
f
o
r
SC
AD
A
cy
b
er
attac
k
s
[
4
]
.
Stu
x
n
et
is
a
wo
r
m
th
at
was
d
is
co
v
er
ed
in
2
0
1
0
th
at
tar
g
ets
PLC,
w
h
ich
ar
e
u
s
ed
in
p
o
wer
p
lan
ts
an
d
g
as
p
ip
elin
es.
T
h
e
co
m
p
u
ter
wo
r
m
Stu
x
n
et
is
a
m
alicio
u
s
p
r
o
g
r
a
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
1
,
Ap
r
il
20
25
:
1
33
-
1
44
134
I
t
h
as
th
e
ab
ilit
y
t
o
d
estro
y
it
s
elf
in
ce
n
tr
i
f
u
g
es
at
an
I
r
an
i
an
u
r
a
n
iu
m
en
r
ic
h
m
en
t
f
ac
ilit
y
[
5
]
.
I
n
th
e
Un
ite
d
States
in
2
0
2
1
,
r
a
n
s
o
m
war
e
cy
b
er
attac
k
s
tar
g
eted
n
etwo
r
k
ed
d
e
v
ices
m
an
ag
in
g
o
il
p
i
p
elin
e
s
y
s
tem
s
.
T
h
e
s
u
s
p
en
s
io
n
o
f
all
p
ip
elin
e
o
p
er
atio
n
s
led
to
a
p
an
d
em
ic
in
th
e
o
il
s
u
p
p
ly
.
T
h
e
r
esto
r
a
tio
n
o
f
th
e
s
y
s
tem
s
r
eq
u
ir
ed
th
e
p
a
y
m
en
t o
f
a
r
a
n
s
o
m
o
f
4
.
4
m
ill
io
n
US d
o
llar
s
[
6
]
.
I
n
ter
n
et
-
co
n
n
ec
ted
SC
ADA
ex
h
ib
its
n
u
m
e
r
o
u
s
v
u
ln
er
a
b
ilit
ies,
r
en
d
er
i
n
g
it
a
n
in
c
r
ea
s
in
g
l
y
attr
ac
tiv
e
tar
g
et
f
o
r
c
y
b
er
attac
k
s
[
3
]
.
V
u
ln
er
ab
ilit
y
attr
ac
ts
attac
k
e
r
s
to
d
is
r
u
p
t
SC
ADA
b
ec
au
s
e
o
f
th
e
d
an
g
er
it
ca
n
ev
en
co
s
t
liv
es
[
7
]
,
[
8
]
.
Netw
o
r
k
v
u
ln
er
a
b
ilit
ies
m
u
s
t
also
b
e
co
n
s
id
er
e
d
,
as
th
ey
ca
n
h
a
v
e
n
eg
ati
v
e
im
p
ac
ts
o
n
b
u
s
in
ess
es
an
d
u
s
er
p
o
p
u
la
tio
n
s
,
p
ar
ticu
lar
ly
if
th
e
attac
k
tar
g
ets
cr
itical
in
f
r
astru
ctu
r
e
u
s
ed
b
y
m
an
y
,
lik
e
p
o
wer
s
y
s
tem
[
9
]
.
SC
ADA
h
an
d
les
s
en
s
itiv
e
in
f
o
r
m
atio
n
,
m
ak
in
g
th
e
co
m
p
r
o
m
is
e
o
r
m
an
ip
u
latio
n
o
f
s
u
c
h
d
ata
a
th
r
ea
t
to
s
y
s
tem
in
teg
r
ity
an
d
u
s
er
p
r
iv
ac
y
.
T
h
e
th
r
ee
m
o
s
t
d
an
g
er
o
u
s
th
r
ea
t
v
ec
to
r
s
in
SC
ADA
ar
e
r
an
s
o
m
war
e,
ex
to
r
tio
n
,
o
r
o
t
h
er
f
in
an
cially
m
o
tiv
ated
cr
i
m
es,
f
o
llo
wed
b
y
n
atio
n
-
s
tat
e
cy
b
er
-
attac
k
s
,
an
d
f
in
ally
d
e
v
ices
an
d
th
in
g
s
a
d
d
ed
to
th
e
n
etwo
r
k
[
4
]
.
Mo
d
b
u
s
an
d
DNP3
ar
e
wid
ely
u
s
ed
p
r
o
to
co
ls
in
th
e
in
d
u
s
tr
y
,
b
u
t
th
ey
p
o
s
s
ess
s
e
cu
r
ity
v
u
ln
er
a
b
ilit
ies
an
d
r
is
k
s
.
T
h
is
v
u
ln
er
ab
ilit
y
is
f
u
r
th
er
co
m
p
o
u
n
d
ed
b
y
leg
ac
y
co
n
tr
o
l
elem
en
ts
lik
e
R
T
U
o
r
PLC
[
1
]
.
Data
co
m
m
u
n
icatio
n
in
th
e
Mo
d
b
u
s
p
r
o
t
o
co
l
ad
h
er
es
to
th
e
s
tr
u
ctu
r
e
o
f
t
h
e
p
r
o
to
co
l d
ata
u
n
it
(
PDU)
with
f
u
n
ctio
n
co
d
e
s
ex
ch
an
g
ed
b
etwe
en
th
e
clien
t a
n
d
s
er
v
er
[
1
0
]
.
I
n
cid
en
ts
ca
u
s
ed
b
y
attac
k
er
s
ca
n
r
esu
lt
in
p
h
y
s
ical
d
a
m
ag
e
an
d
ev
en
ca
s
u
alties
.
T
h
i
s
s
tu
d
y
is
d
esig
n
ed
f
o
r
c
y
b
er
s
ec
u
r
ity
,
a
d
h
er
in
g
to
th
e
I
E
C
6
1
8
5
0
p
r
o
to
co
l,
p
r
im
ar
ily
d
e
p
lo
y
ed
in
s
u
b
s
tatio
n
s
.
I
n
o
r
d
er
to
ef
f
ec
tiv
ely
d
ef
en
d
ag
ain
s
t
v
ar
io
u
s
attac
k
s
,
t
h
e
p
r
o
to
co
l
s
p
ec
if
icatio
n
s
,
p
h
y
s
ical
k
n
o
wled
g
e,
an
d
lo
g
ical
b
eh
av
io
u
r
h
as
b
ee
n
em
p
lo
y
e
d
to
co
n
s
tr
u
ct
th
e
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
(
I
DS
)
[
1
1
]
.
Oth
er
wo
r
k
s
also
ex
am
in
ed
th
r
ee
ty
p
es
o
f
attac
k
s
in
I
C
S,
n
am
ely
r
ec
o
n
n
aiss
an
ce
,
f
alse
d
ata
in
jectio
n
,
an
d
r
ep
lay
attac
k
s
o
n
th
e
Mo
d
b
u
s
a
n
d
S7
p
r
o
to
co
ls
[
7
]
.
Ad
d
itio
n
ally
,
attac
k
e
x
p
lo
itat
io
n
o
n
th
e
test
b
ed
u
tili
ze
d
th
e
Mo
d
b
u
s
/TCP
with
d
e
c
i
s
i
o
n
t
r
e
e
(
D
T
)
m
o
d
e
l
,
e
n
c
o
m
p
a
s
s
e
s
r
e
p
l
a
y
a
t
t
a
c
k
,
M
I
T
M
,
d
e
n
i
a
l
o
f
s
e
r
v
i
c
e
(D
o
S
)
,
a
n
d
r
e
c
o
n
n
a
i
s
s
a
n
c
e
[
1
2
]
.
Fu
r
th
er
m
o
r
e
,
a
v
ir
tu
al
test
b
ed
an
d
d
o
cu
m
e
n
tatio
n
h
as
b
e
en
d
e
v
elo
p
ed
to
in
v
esti
g
ate
wea
k
n
ess
es
in
th
e
M
o
d
b
u
s
p
r
o
to
c
o
l
a
n
d
Do
S a
tta
ck
s
[
1
3
]
.
Man
-
in
-
th
e
-
m
id
d
l
e
(
MI
T
M)
a
ttack
s
r
ep
r
esen
t
th
e
m
o
s
t
s
ig
n
if
ican
t
th
r
ea
t
to
SC
ADA
n
et
wo
r
k
s
an
d
ca
n
h
av
e
an
im
p
ac
t
o
n
n
etwo
r
k
r
eliab
ilit
y
an
d
s
ec
u
r
ity
,
esp
ec
ially
in
SC
ADA
n
etwo
r
k
s
em
p
lo
y
in
g
th
e
Mo
d
b
u
s
p
r
o
to
c
o
l,
o
win
g
to
t
h
e
p
r
o
to
c
o
l'
s
in
h
er
en
t
s
ec
u
r
ity
l
im
itatio
n
s
[
1
]
.
I
n
MI
T
M
,
th
e
a
ttack
er
p
o
s
es
as
an
au
th
en
tic
u
s
er
b
etwe
en
th
e
en
d
s
o
f
th
e
co
m
m
u
n
icatio
n
.
T
h
ese
attac
k
s
ca
n
d
is
r
u
p
t
Mo
d
b
u
s
co
m
m
u
n
icatio
n
p
r
o
to
co
ls
,
p
er
m
itti
n
g
a
m
alici
o
u
s
to
p
o
s
e
as
a
co
n
tr
o
ller
an
d
tr
an
s
m
it
d
am
a
g
in
g
s
ig
n
als
to
f
ield
d
ev
ices
[
1
4
]
.
Fo
r
in
s
tan
ce
,
MI
T
M
attac
k
s
o
n
s
m
ar
t
g
r
id
s
[
9
]
.
MI
T
M
e
x
to
r
ts
v
ictim
s
b
y
u
s
in
g
a
r
an
s
o
m
war
e
p
atter
n
.
I
t
f
o
r
g
es
m
ess
ag
es
f
r
o
m
r
ea
l
cr
im
in
als
in
o
r
d
er
to
p
u
t
m
o
r
e
p
r
ess
u
r
e
o
n
th
eir
m
an
a
g
er
s
t
o
m
ak
e
r
esti
tu
tio
n
.
Fu
r
th
er
m
o
r
e
,
p
er
p
r
etr
at
o
r
m
o
d
i
f
ied
th
e
b
itco
in
ad
d
r
ess
lin
k
e
d
to
th
e
ex
to
r
tio
n
p
ay
m
en
t
an
d
ch
an
g
ed
th
e
e
m
ail
m
ess
ag
e
[
1
5
]
.
MI
T
M
attac
k
l
ea
d
s
to
u
n
au
th
o
r
ized
c
o
n
tr
o
l,
m
o
d
if
icatio
n
s
,
o
r
in
jectio
n
s
p
r
io
r
to
th
e
p
ac
k
et
r
ea
ch
in
g
i
ts
in
ten
d
e
d
d
esti
n
ati
o
n
,
th
er
e
b
y
d
is
r
u
p
tin
g
i
n
d
u
s
tr
i
al
o
p
er
atio
n
s
.
So
m
e
p
r
e
v
io
u
s
s
tu
d
y
h
as
f
o
cu
s
ed
o
n
t
h
e
p
r
o
tectio
n
o
f
SC
ADA
s
y
s
tem
s
.
Fo
r
in
s
tan
ce
,
th
e
d
etec
tio
n
o
f
ad
v
er
s
ar
ial
ex
am
p
les
b
y
i
d
en
tify
in
g
in
c
o
n
s
is
ten
cies
b
et
wee
n
m
an
if
o
ld
ev
alu
atio
n
s
an
d
th
e
I
DS
m
o
d
el
in
f
er
en
ce
[
1
6
]
.
Usi
n
g
a
f
ilter
-
b
ased
ap
p
r
o
ac
h
[
1
7
]
an
d
o
n
e
class
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
es
(
OC
SVM)
[
1
8
]
ca
n
ef
f
ec
tiv
ely
d
etec
t
cy
b
e
r
attac
k
s
in
in
d
u
s
tr
ial
co
n
t
r
o
l
s
y
s
tem
(
I
C
S
)
.
I
n
a
d
d
itio
n
,
t
h
e
n
etwo
r
k
tr
af
f
ic
wa
s
class
if
ied
u
s
in
g
n
eu
r
al
n
etwo
r
k
s
(
NN)
an
d
d
ec
is
io
n
t
r
ee
(
D
T
)
with
in
th
e
co
n
s
t
r
u
cted
s
im
u
latio
n
en
v
ir
o
n
m
e
n
t.
D
iv
er
s
e
m
ac
h
in
e
lear
n
in
g
(
M
L
)
class
if
icatio
n
alg
o
r
ith
m
s
wer
e
em
p
lo
y
ed
an
d
ev
alu
ate
d
to
d
etec
t
Mo
d
b
u
s
-
r
elate
d
th
r
ea
ts
[
1
9
]
.
Fu
r
th
er
m
o
r
e
,
a
co
n
v
o
lu
tio
n
al
n
e
u
r
al
n
e
two
r
k
(
C
NN)
ar
c
h
itectu
r
e
f
o
r
SC
ADA
n
etwo
r
k
s
[
2
0
]
,
[
2
1
]
h
as b
ee
n
s
h
o
wn
to
im
p
r
o
v
e
d
th
e
e
f
f
ec
tiv
en
ess
o
f
t
h
e
d
etec
tio
n
.
T
h
e
r
e
a
r
e
t
w
o
t
y
p
e
s
o
f
I
D
S
:
s
i
g
n
a
t
u
r
es
(
S
I
DS
)
a
n
d
a
n
o
m
a
li
e
s
(
A
I
D
S
)
.
A
n
e
x
t
e
n
s
i
v
e
m
eth
o
d
o
l
o
g
y
c
o
m
p
a
r
i
s
o
n
b
e
t
we
e
n
A
I
DS
an
d
S
I
D
S
i
s
c
a
r
r
i
e
d
o
u
t
[
2
2
]
.
T
h
e
I
D
S
w
as
d
e
v
e
l
o
p
e
d
u
s
i
n
g
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
[
2
3
]
–
[
2
8
]
,
D
e
e
p
l
ea
r
n
i
n
g
[
2
9
]
,
[
3
0
]
.
C
o
m
b
i
n
i
n
g
s
e
v
e
r
a
l
M
L
[
3
1
]
–
[
3
3
]
s
u
c
h
a
s
r
a
n
d
o
m
f
o
r
e
s
t
(
R
F
)
,
b
o
o
s
t
i
n
g
w
it
h
e
x
t
r
e
m
e
g
r
a
d
i
e
n
t (
X
g
b
o
o
s
t
)
a
n
d
a
d
a
p
t
i
v
e
(
A
d
aB
o
o
s
t
)
,
h
as
p
r
o
v
e
n
t
o
b
e
a
b
l
e
to
d
e
t
e
c
t
r
a
n
s
o
m
w
a
r
e
a
n
d
o
t
h
e
r
m
a
l
i
c
i
o
u
s
s
o
f
t
w
a
r
e
[
3
4
]
.
F
u
r
t
h
e
r
m
o
r
e
,
t
h
e
i
n
t
e
g
r
a
t
i
o
n
o
f
D
T
a
n
d
A
d
a
B
o
o
s
t
i
n
cr
e
a
s
es
a
c
c
u
r
a
c
y
i
n
d
e
t
e
c
t
i
n
g
f
r
a
u
d
[
3
5
]
.
A
n
a
l
te
r
n
a
t
i
v
e
a
p
p
r
o
a
c
h
is
t
o
i
m
p
l
e
m
e
n
t
a
d
i
m
e
n
s
i
o
n
a
l
r
e
d
u
ct
i
o
n
s
t
r
a
t
eg
y
,
w
h
i
c
h
e
n
h
a
n
c
es
t
h
e
a
c
c
u
r
a
c
y
[
2
5
]
.
I
n
a
d
d
i
ti
o
n
,
h
y
b
r
i
d
d
e
e
p
l
e
a
r
n
i
n
g
t
ec
h
n
i
q
u
e
s
[
3
6
]
,
i
n
c
l
u
d
i
n
g
p
r
i
n
ci
p
a
l
co
m
p
o
n
e
n
t
a
n
a
l
y
s
i
s
(
P
C
A
)
,
s
p
at
i
a
l
c
l
u
s
t
e
r
i
n
g
u
s
i
n
g
d
e
n
s
i
t
y
w
i
t
h
n
o
is
e
,
p
a
r
ti
c
l
e
s
w
a
r
m
o
p
t
i
m
i
z
at
i
o
n
(
PS
O
)
,
a
n
d
a
u
t
o
e
n
c
o
d
e
r
(
A
E
)
,
h
a
v
e
b
e
e
n
d
e
m
o
n
s
t
r
a
t
e
d
t
o
a
c
h
i
e
v
e
n
e
a
r
-
p
e
r
f
e
c
t
a
c
c
u
r
a
c
y
i
n
th
e
d
e
v
e
l
o
p
m
e
n
t
o
f
I
D
S
[
3
7
]
.
Attack
ty
p
es
ar
e
ch
a
n
g
in
g
q
u
i
ck
ly
.
T
h
is
m
ak
es
th
e
p
u
b
lic
d
atasets
u
s
ed
to
tr
ain
ML
m
o
d
els
o
u
t
o
f
d
ate
an
d
i
n
ef
f
ec
tiv
e
a
g
ain
s
t
n
ew
ty
p
es
o
f
attac
k
s
.
A
f
u
r
t
h
e
r
s
tu
d
y
s
p
ec
if
ically
d
etec
ts
an
o
m
alies
in
I
C
S
by
an
aly
zin
g
n
etwo
r
k
p
ac
k
ets
u
s
in
g
th
e
Mo
d
b
u
s
p
r
o
to
co
l
with
th
e
latest
E
lectr
a
d
ata
s
et.
T
h
er
e
ar
e
two
m
eth
o
d
s
to
u
s
e
th
e
ML
ap
p
r
o
ac
h
:
s
u
p
er
v
is
ed
an
d
u
n
s
u
p
er
v
is
ed
.
Su
p
er
v
is
ed
tech
n
iq
u
es
in
clu
d
e
R
F,
SV
M,
an
d
NN
.
Un
s
u
p
er
v
is
ed
lear
n
in
g
tech
n
i
q
u
es
in
clu
d
e
th
e
is
o
latio
n
f
o
r
e
s
t
(
I
F)
an
d
th
e
OC
SVM.
B
ase
d
o
n
th
e
r
esu
lts
,
th
e
R
F
d
em
o
n
s
tr
ated
th
e
h
ig
h
est
p
r
ec
is
io
n
,
wh
ile
th
e
SVM
ac
h
ie
v
ed
t
h
e
h
ig
h
est
r
ec
all
a
n
d
F1
s
co
r
es
[
7
]
.
An
o
th
er
s
tu
d
y
p
r
o
p
o
s
es
to
id
e
n
tify
a
n
o
m
alies
in
I
C
S
u
s
in
g
a
co
m
b
in
ed
DNN
a
n
d
g
en
e
r
ativ
e
a
d
v
er
s
ar
ial
n
etwo
r
k
(
GAN)
m
o
d
el.
As
a
r
esu
lt,
th
e
r
ec
all
m
etr
ic
was
0
.
9
8
[
3
8
]
.
B
in
ar
y
class
c
lass
if
icatio
n
is
ap
p
lied
to
an
o
m
al
y
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
S
MOTE
tr
ee
-
b
a
s
ed
a
u
to
en
c
o
d
er mu
lti
-
s
ta
g
e
d
etec
tio
n
fo
r
ma
n
-
in
-
th
e
-
mid
d
le
…
(
F
r
eska
R
o
la
n
s
a
)
135
d
etec
tio
n
in
s
ev
er
al
s
tu
d
ies,
b
u
t
m
u
lti
-
class
class
if
icat
io
n
is
o
n
ly
u
s
ed
i
n
r
esear
ch
[
3
9
]
.
Fu
r
th
er
m
o
r
e
,
th
er
e
a
r
e
s
till
m
an
y
d
etec
tio
n
er
r
o
r
s
[
3
9
]
,
esp
ec
ially
f
o
r
m
in
o
r
ities
.
Nev
er
th
eless
,
th
er
e
is
a
g
ap
in
th
e
ex
is
tin
g
r
esear
ch
o
n
m
u
lti
-
class
clas
s
if
icatio
n
,
wh
ich
is
o
n
ly
co
n
d
u
cte
d
b
y
r
esear
ch
o
n
an
aly
zin
g
n
etwo
r
k
p
ac
k
ets
u
s
in
g
th
e
Mo
d
b
u
s
p
r
o
t
o
co
l
with
th
e
E
lectr
a
d
ataset.
Fu
r
th
er
m
o
r
e
,
th
e
n
u
m
b
er
o
f
m
in
o
r
ity
class
if
icatio
n
er
r
o
r
s
r
em
ain
s
h
ig
h
.
T
h
er
ef
o
r
e,
th
is
r
esear
ch
p
r
o
p
o
s
es
SMOT
E
T
r
ee
-
b
ased
a
u
t
o
en
c
o
d
er
m
u
lti
-
s
tag
e
d
etec
tio
n
f
o
r
m
an
-
in
-
th
e
-
m
i
d
d
le
in
SC
ADA.
Ou
r
p
r
o
p
o
s
ed
m
o
d
el
h
as
f
o
u
r
m
ain
s
tag
es:
p
r
ep
r
o
ce
s
s
in
g
,
b
alan
cin
g
,
au
t
o
en
co
d
e
r
,
an
d
tr
ee
class
if
icati
o
n
,
wh
ich
r
eq
u
i
r
e
s
s
eq
u
en
tial
ex
ec
u
tio
n
to
d
etec
t
an
o
m
alies
an
d
class
if
y
m
u
lti
-
class
es
with
p
r
ee
m
in
en
ce
.
A
tr
ee
class
if
icatio
n
m
o
d
el
was
d
e
v
elo
p
e
d
u
s
in
g
o
p
tim
ized
h
y
p
er
p
a
r
am
eter
s
an
d
SMOT
E
-
b
ased
tech
n
iq
u
es
to
h
a
n
d
le
u
n
b
ala
n
ce
d
d
ata,
s
p
ec
if
ically
to
im
p
r
o
v
e
th
e
d
etec
tio
n
an
d
class
if
icatio
n
o
f
m
in
o
r
ity
attac
k
class
es.
I
n
a
d
d
itio
n
,
b
y
in
clu
d
in
g
an
a
u
to
en
c
o
d
er
ar
ch
itectu
r
e
f
o
r
th
e
a
d
ju
s
tm
en
t
o
f
th
e
v
ar
iatio
n
in
t
h
e
d
ata
p
r
io
r
to
th
e
r
e
d
u
ctio
n
o
f
th
e
d
im
en
s
io
n
ality
.
2.
M
E
T
H
O
D
SC
ADA
s
y
s
tem
s
ar
e
u
s
ed
to
co
n
tr
o
l
lar
g
e
an
d
co
m
p
lex
f
ac
ilit
ies
with
in
d
u
s
tr
ial
co
n
tr
o
l
p
r
o
ce
s
s
es.
T
h
e
f
ac
to
r
y
co
m
p
r
is
es
SC
A
D
A
en
d
p
o
in
ts
,
wh
ich
a
r
e
s
en
s
o
r
s
an
d
ac
tu
ato
r
s
.
T
h
e
p
r
o
p
o
s
e
d
d
etec
tio
n
m
o
d
el
(
STAM
)
is
u
s
ed
t
o
d
etec
t
attac
k
s
d
u
r
i
n
g
M
o
d
b
u
s
T
C
P
co
m
m
u
n
icatio
n
b
etwe
en
clien
t
an
d
s
er
v
er
.
A
d
etec
tio
n
m
o
d
el
is
th
en
d
e
v
elo
p
ed
u
s
in
g
th
e
E
lectr
a
p
u
b
lic
d
ataset,
wh
ich
r
ep
r
esen
ts
th
e
r
ea
l
wo
r
l
d
o
f
in
d
u
s
tr
ial
co
n
tr
o
l
in
SC
ADA.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
n
ee
d
s
to
b
e
r
u
n
in
a
s
eq
u
e
n
tial
m
an
n
e
r
,
with
ea
ch
s
tag
e
f
o
llo
w
s
th
e
p
r
ev
i
o
u
s
o
n
e.
T
h
e
s
tag
es
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
is
s
h
o
wn
i
n
Fig
u
r
e
1
.
T
h
e
p
r
ep
r
o
ce
s
s
in
g
s
tag
e
im
p
o
r
t
s
th
e
E
lectr
a
d
ataset,
an
d
r
e
m
o
v
es
r
ed
u
n
d
an
t
d
ata
.
T
h
en
th
e
ca
teg
o
r
y
d
ata
is
co
n
v
e
r
ted
to
n
u
m
er
ical
d
ata
u
s
in
g
b
o
t
h
o
n
e
h
o
t
en
c
o
d
in
g
(
OHE
)
an
d
lab
el
e
n
co
d
i
n
g
.
T
h
e
n
ex
t
s
tep
is
to
n
o
r
m
alize
th
e
d
ata
u
s
in
g
s
tan
d
ar
d
s
ca
ler
n
o
r
m
aliza
tio
n
.
T
h
e
d
ata
is
th
en
b
alan
ce
d
u
s
in
g
th
e
s
y
n
t
h
etic
m
in
o
r
ity
o
v
er
s
am
p
lin
g
tech
n
i
q
u
e
(
SMOT
E
)
.
T
h
e
a
u
to
en
co
d
er
is
th
en
u
s
ed
to
ad
ju
s
t
v
ar
i
atio
n
an
d
r
ed
u
ce
th
e
d
im
en
s
io
n
o
f
th
e
d
ata
.
T
h
e
E
l
ec
tr
a
d
ataset
co
n
s
is
ts
o
f
th
e
tr
ain
in
g
s
et
an
d
th
e
test
in
g
s
et.
T
h
ese
n
ee
d
s
to
b
e
s
p
lit
in
to
8
0
%
tr
ain
i
n
g
s
et
an
d
2
0
%
tes
tin
g
s
et.
I
n
th
e
tr
ain
i
n
g
s
et,
f
iv
e
class
if
ier
s
(
SVM,
KNN,
L
R
,
R
F,
an
d
DT
)
ar
e
ev
alu
ated
,
an
d
t
h
e
b
es
t
is
s
elec
ted
.
A
tr
ee
m
o
d
el
is
m
ad
e
u
s
in
g
th
e
DT
class
if
ier
with
h
y
p
e
r
p
ar
am
ete
r
o
p
tim
izatio
n
.
T
h
e
test
in
g
s
et
is
ca
r
r
ied
o
u
t
b
y
e
v
alu
atin
g
th
e
m
o
d
el
an
d
m
e
asu
r
in
g
t
h
e
p
er
f
o
r
m
an
c
e
o
f
an
o
m
aly
d
etec
tio
n
an
d
m
u
lticl
ass
clas
s
if
icatio
n
.
Fig
u
r
e
1.
T
h
e
p
r
o
p
o
s
ed
d
etec
tio
n
m
o
d
els
2
.
1
.
T
estbed
a
nd
d
a
t
a
s
et
T
h
e
liq
u
id
h
an
d
lin
g
s
y
s
tem
(
L
HS)
is
an
I
C
S
test
b
ed
ap
p
li
ed
to
th
e
b
ev
er
ag
e
in
d
u
s
tr
y
a
n
d
en
s
u
r
es
h
ig
h
-
q
u
alit
y
p
r
o
d
u
cts
ar
e
s
af
e
f
o
r
co
n
s
u
m
p
tio
n
.
L
HS
u
tili
zi
n
g
PLC
co
n
tr
o
ller
u
s
in
g
C
PX
-
E
-
C
E
C
-
M1
ty
p
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
1
,
Ap
r
il
20
25
:
1
33
-
1
44
136
T
h
is
s
y
s
tem
h
as
3
a
x
is
X,
Y,
an
d
Z
m
o
v
e
m
en
ts
u
s
in
g
a
s
tep
p
er
m
o
to
r
d
r
iv
e.
I
t
u
s
es
a
co
n
v
ey
o
r
th
at
f
u
n
ctio
n
s
to
r
ec
eiv
e
em
p
ty
b
o
ttles
an
d
ca
p
s
an
d
to
d
eliv
e
r
f
u
lly
f
i
lle
d
b
o
ttles
.
T
h
e
wo
r
k
in
g
s
y
s
tem
is
to
f
ill
th
e
liq
u
id
in
to
th
e
b
o
ttle
an
d
th
en
clo
s
e
it
to
b
e
s
en
t
to
th
e
r
o
b
o
t
ass
em
b
ly
.
PLC
is
co
n
n
ec
ted
to
s
en
s
o
r
s
(
p
r
o
x
im
ity
s
witch
)
an
d
ac
tu
ato
r
s
(
to
o
th
ed
b
elt,
s
tep
p
er
m
o
to
r
,
s
er
v
o
d
r
i
v
e,
m
in
i sli
d
e
u
n
it,
r
o
tar
y
g
r
i
p
p
er
m
o
d
u
le,
p
ar
allel
g
r
ip
p
er
,
p
r
ess
u
r
e
v
ac
u
u
m
g
en
er
ato
r
(
PVGA)
,
p
ip
ette
h
ea
d
)
.
C
o
n
tr
o
l
p
r
o
g
r
am
m
i
n
g
o
n
L
H
S
u
s
in
g
C
ODE
SY
S
V3
.
5
with
co
m
m
u
n
icatio
n
u
s
in
g
th
e
Mo
d
b
u
s
/TCP
p
r
o
to
c
o
l.
T
h
e
p
h
y
s
ical
test
b
ed
is
s
h
o
wn
in
Fig
u
r
e
2
.
T
h
e
d
ataset
u
s
ed
in
th
is
r
esea
r
ch
is
E
lectr
a,
wh
ich
is
a
r
ec
e
n
t,
r
ea
lis
tic,
an
d
cu
s
to
m
ized
d
ataset
f
o
r
tr
ain
in
g
m
ac
h
in
e
lear
n
i
n
g
-
b
as
ed
I
DS
m
o
d
els
b
ased
o
n
n
etwo
r
k
tr
af
f
ic
d
ata.
I
t
is
g
en
er
ated
f
r
o
m
n
etwo
r
k
tr
af
f
ic
in
elec
tr
i
ca
l
tr
ac
tio
n
s
u
b
s
tatio
n
s
o
p
er
atin
g
u
n
d
er
atta
ck
an
d
n
o
r
m
al
co
n
d
itio
n
s
.
T
h
e
E
lectr
a
d
ataset
is
co
n
s
tr
u
cted
f
r
o
m
SC
ADA
an
d
PLC
s
y
s
tem
d
ev
ices,
an
d
it
is
co
n
tr
o
lled
u
s
in
g
th
e
Mo
d
b
u
s
an
d
S7
co
m
m
p
r
o
to
co
ls
,
m
ir
r
o
r
in
g
r
ea
l
-
wo
r
l
d
s
ce
n
ar
io
s
[
4
0
]
.
T
h
e
1
0
attr
ib
u
t
es
o
f
th
is
d
ataset
ar
e
ca
teg
o
r
iz
ed
in
to
o
n
e
lab
el,
n
am
ely
MI
T
M
u
n
alter
ed
,
r
ec
o
g
n
itio
n
,
r
ea
d
,
wr
ite,
r
esp
o
n
s
e,
f
o
r
ce
attac
k
,
a
n
d
n
o
r
m
al
.
A
f
u
ll
d
escr
ip
tio
n
o
f
th
e
E
le
ctr
a
d
ataset
ca
n
b
e
f
o
u
n
d
in
T
ab
le
1
.
An
attac
k
er
m
u
s
t
p
er
f
o
r
m
a
r
ec
o
n
n
aiss
an
ce
attac
k
u
s
in
g
th
e
''
f
u
n
ctio
n
co
d
e
r
ec
o
g
n
itio
n
attac
k
''
t
o
o
b
tain
in
f
o
r
m
atio
n
ab
o
u
t
th
e
t
ar
g
et
an
d
attac
k
th
e
PLC.
Fals
e
d
ata
in
jectio
n
attac
k
s
attem
p
t
to
g
ain
c
o
n
tr
o
l
o
f
co
n
tr
o
l d
ev
ices in
an
I
C
S
u
s
in
g
co
n
tr
o
l p
r
o
to
co
ls
to
tr
an
s
m
it m
o
d
if
ied
d
ata
.
T
h
ese
attac
k
s
ar
e
class
if
ied
b
ased
o
n
th
e
m
o
d
if
ie
d
d
ata.
Sp
o
o
f
e
d
p
ac
k
ets
attem
p
t
'
R
ea
d
'
o
r
'
W
r
ite
'
o
n
th
e
PLC
's
m
em
o
r
y
ad
d
r
ess
.
‘
R
esp
o
n
s
e
m
o
d
if
icatio
n
attac
k
’
o
r
‘
f
o
r
ce
er
r
o
r
’
v
ia
f
o
r
g
ed
s
lav
e
d
ev
ice
p
ac
k
ets.
‘
C
o
m
m
an
d
m
o
d
if
icat
io
n
attac
k
’
th
r
o
u
g
h
m
an
ip
u
lated
m
aster
d
ev
ice
p
a
ck
ets.
Pack
ets
d
eliv
er
ed
b
y
s
l
av
e
o
r
m
aster
d
ev
ices
m
a
y
h
av
e
th
eir
r
ec
ep
tio
n
r
ate
alter
ed
b
y
‘r
e
p
lay
attac
k
s
’
.
W
ith
in
th
e
E
lectr
a
d
ataset,
t
h
er
e
ar
e
1
6
.
2
8
9
.
2
7
7
r
ec
o
r
d
s
o
f
n
etwo
r
k
tr
a
f
f
ic
i
n
th
e
Mo
d
b
u
s
p
r
o
t
o
co
l,
wh
ich
e
n
co
m
p
ass
d
ata
v
ar
iatio
n
s
co
n
s
is
tin
g
o
f
1
5
.
4
4
4
.
9
4
0
d
ata
r
ec
o
r
d
s
u
n
d
e
r
No
r
m
al
co
n
d
itio
n
s
(
n
o
r
m
al
,
MI
T
M
u
n
alter
ed
)
an
d
8
4
4
.
3
3
7
d
ata
r
e
co
r
d
s
u
n
d
e
r
attac
k
co
n
d
itio
n
s
(
r
ec
o
g
n
itio
n
,
r
ea
d
,
wr
ite,
r
esp
o
n
s
e,
f
o
r
ce
attac
k
a
n
d
r
ep
la
y
attac
k
)
.
Fig
u
r
e
2
.
T
h
e
liq
u
id
h
an
d
lin
g
s
tatio
n
an
d
s
er
v
er
s
T
ab
le
1
.
Ov
e
r
v
iew
o
f
t
h
e
E
lec
tr
a
No
F
e
a
t
u
r
e
D
e
scri
p
t
i
o
n
D
a
t
a
t
y
p
e
1
Ti
me
Ti
me
t
r
a
f
f
i
c
n
e
t
w
o
r
k
st
r
i
n
g
2
S
mac
O
r
i
g
i
n
a
t
i
n
g
M
a
c
a
d
d
r
e
ss
st
r
i
n
g
3
d
ma
c
Ta
r
g
e
t
M
a
c
a
d
d
r
e
ss
st
r
i
n
g
4
si
p
O
r
i
g
i
n
a
t
i
n
g
I
P
a
d
d
r
e
ss
st
r
i
n
g
5
d
i
p
Ta
r
g
e
t
I
P
a
d
d
r
e
ss
st
r
i
n
g
6
r
e
q
u
e
st
w
h
e
t
h
e
r
o
r
n
o
t
t
h
e
r
e
q
u
e
s
t
st
r
i
n
g
7
fc
F
u
n
c
t
i
o
n
C
o
d
e
i
n
M
o
d
b
u
s
i
n
t
e
g
e
r
8
e
r
r
o
r
D
i
sp
l
a
y
s wh
e
t
h
e
r
a
n
e
r
r
o
r
b
o
o
l
e
a
n
9
mad
d
M
e
m
o
r
y
a
d
d
r
e
ss re
a
d
/
w
r
i
t
e
o
p
e
r
a
t
i
o
n
s
i
n
t
e
g
e
r
10
d
a
t
a
D
i
sp
l
a
y
s
d
a
t
a
s
e
n
t
o
r
r
e
c
e
i
v
e
d
i
n
t
e
g
e
r
11
La
b
e
l
C
l
a
s
s fo
r
t
y
p
e
a
t
t
a
c
k
o
r
N
o
r
m
a
l
st
r
i
n
g
2
.
2
.
P
re
pro
ce
s
s
ing
s
t
a
g
e
T
h
e
p
r
e
p
r
o
ce
s
s
in
g
s
tep
s
a
p
p
lied
to
th
e
r
esear
ch
in
clu
d
e
elim
in
atin
g
r
ed
u
n
d
an
t
d
ata,
en
co
d
i
n
g
ca
teg
o
r
ical
d
ata,
a
n
d
n
o
r
m
al
izin
g
d
ata.
T
h
e
in
d
u
s
tr
ial
co
n
tr
o
l
s
y
s
tem
d
ataset,
n
u
m
e
r
o
u
s
r
ed
u
n
d
an
t
d
ata
p
ac
k
ag
es
wer
e
id
en
tifie
d
d
u
e
to
r
ep
ea
ted
ex
ec
u
tio
n
s
in
m
u
ltip
le
co
n
tr
o
l
p
r
o
ce
s
s
es.
T
h
is
elim
in
atio
n
o
f
r
ed
u
n
d
an
t
d
ata
is
ac
h
iev
e
d
b
y
d
is
r
eg
ar
d
i
n
g
t
h
e
tim
e
f
ea
tu
r
e
to
id
en
tify
id
en
tical
d
ata.
Fo
r
id
en
tical
d
ata,
o
n
ly
th
e
in
itial
d
ata
is
r
etain
ed
,
an
d
th
e
r
est
is
co
n
s
id
er
ed
r
e
d
u
n
d
an
t
an
d
m
u
s
t
b
e
r
em
o
v
ed
.
T
h
er
e
ar
e
n
u
m
er
o
u
s
d
u
p
licate
d
ata
r
ec
o
r
d
s
th
r
o
u
g
h
o
u
t th
e
d
ataset.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
S
MOTE
tr
ee
-
b
a
s
ed
a
u
to
en
c
o
d
er mu
lti
-
s
ta
g
e
d
etec
tio
n
fo
r
ma
n
-
in
-
th
e
-
mid
d
le
…
(
F
r
eska
R
o
la
n
s
a
)
137
T
h
e
s
ec
o
n
d
s
tep
in
v
o
l
v
es
en
co
d
in
g
.
So
m
e
f
ield
s
h
av
e
b
e
en
m
o
d
if
ied
to
p
er
f
o
r
m
ca
teg
o
r
ical
d
ata
co
n
v
er
s
io
n
u
s
in
g
b
o
th
OHE
an
d
L
a
b
el
E
n
c
o
d
in
g
.
OHE
is
em
p
lo
y
ed
f
o
r
ca
teg
o
r
ical
d
ata
th
at
lack
s
a
s
eq
u
en
tial
r
elatio
n
s
h
ip
,
s
u
ch
a
s
th
e
s
m
ac
,
d
m
ac
,
s
ip
an
d
d
ip
.
Ad
d
itio
n
ally
,
OHE
c
o
n
v
er
ts
ca
teg
o
r
ical
d
ata
to
in
teg
er
s
,
with
v
alu
es
r
an
g
in
g
f
r
o
m
0
t
o
1
,
u
tili
zin
g
a
f
ix
ed
n
u
m
b
er
o
f
d
im
en
s
io
n
s
.
O
n
th
e
o
th
er
h
an
d
,
ca
teg
o
r
ical
d
ata,
wh
ich
ex
h
ib
its
lit
tle
o
r
n
o
s
eq
u
en
tial
r
elatio
n
s
h
ip
,
is
en
co
d
ed
u
s
in
g
L
a
b
el
E
n
co
d
in
g
.
Fo
r
in
s
tan
ce
,
th
is
ap
p
lies
to
attr
ib
u
tes s
u
ch
as f
c,
m
ad
d
,
an
d
d
at
a.
No
r
m
aliz
atio
n
is
ap
p
lied
to
e
n
s
u
r
e
th
at
all
r
em
ain
i
n
g
f
ea
t
u
r
es
in
th
e
d
ataset
f
all
with
i
n
th
e
s
am
e
r
an
g
e
as
th
e
last
s
tep
o
f
th
e
f
ir
s
t
p
h
ase.
T
h
e
s
tan
d
ar
d
s
c
aler
n
o
r
m
aliza
tio
n
m
eth
o
d
is
em
p
lo
y
ed
f
o
r
th
is
p
u
r
p
o
s
e.
R
escales
th
e
d
is
tr
ib
u
tio
n
o
f
v
alu
es
s
o
th
at
th
e
m
ea
n
o
f
th
e
o
b
s
er
v
e
d
v
alu
es
is
0
a
n
d
t
h
e
s
tan
d
ar
d
d
ev
iatio
n
is
1
,
t
h
u
s
r
ed
u
cin
g
t
h
e
d
if
f
er
e
n
ce
s
in
th
e
f
ea
tu
r
es.
T
h
is
p
r
o
ce
s
s
is
ap
p
lied
to
t
r
a
in
in
g
an
d
test
d
ata
d
u
r
in
g
th
e
d
e
v
elo
p
m
e
n
t o
f
t
h
e
class
if
icatio
n
m
o
d
el.
T
h
e
s
tan
d
ar
d
s
ca
ler
n
o
r
m
aliza
tio
n
is
g
i
v
en
in
(
1
)
.
X
=
X
i
−
X
m
ea
n
X
std
(
1
)
W
h
er
e,
X
-
n
o
r
m
alize
d
d
ata,
X
i
-
i
n
p
u
t v
alu
e,
X
mean
-
f
ea
tu
r
e
m
ea
n
,
an
d
X
std
-
f
ea
tu
r
e
s
tan
d
ar
d
d
ev
iatio
n
.
2
.
3
.
B
a
la
ncing
s
t
a
g
e
T
h
e
s
ec
o
n
d
p
h
ase
f
o
c
u
s
es
o
n
cr
ea
tin
g
b
alan
ce
d
d
ata.
T
h
e
E
lectr
a
is
an
u
n
b
alan
ce
d
d
a
ta
s
et
th
at
s
h
o
ws
a
s
m
all
n
u
m
b
e
r
o
f
attac
k
s
co
m
p
ar
e
d
to
th
e
lar
g
e
n
u
m
b
er
o
f
n
o
r
m
al
class
es.
Ov
er
s
am
p
lin
g
th
e
m
in
o
r
ity
class
is
o
n
e
ap
p
r
o
ac
h
to
d
ea
li
n
g
with
u
n
b
alan
ce
d
d
atase
ts
.
T
h
e
s
im
p
lest
ap
p
r
o
ac
h
is
to
d
u
p
licate
ex
am
p
les
in
th
e
m
in
o
r
ity
class
.
T
h
e
SMO
T
E
tech
n
iq
u
e
[
4
1
]
is
ap
p
lied
in
th
is
wo
r
k
.
L
et
m
b
e
th
e
o
v
er
s
am
p
lin
g
r
ate,
m
ea
n
in
g
t
h
at
ea
ch
m
in
o
r
ity
s
am
p
le
will
b
e
o
v
e
r
s
am
p
led
m
tim
es,
n
b
ein
g
t
h
e
to
tal
n
u
m
b
er
o
f
m
in
o
r
ity
s
am
p
les.
W
h
en
X
i
is
a
m
in
o
r
it
y
s
am
p
le,
1
≤
i
≤
n
T
h
ese
ar
e
t
h
e
s
tep
s
th
at
SMOT
E
will
tak
e
in
o
r
d
e
r
to
cr
ea
te
m
n
ew
s
am
p
les b
ased
o
n
X
i
.
-
First s
tep
:
a
p
p
ly
th
e
(K
-
Nea
r
e
s
t N
eig
h
b
o
u
r
)
KNN
to
X
i
(
b
el
o
n
g
in
g
to
th
e
m
in
o
r
ity
s
am
p
le)
to
f
in
d
th
e
s
et
R
i o
f
k
m
in
o
r
ity
s
am
p
les clo
s
est to
X
i
.
-
Seco
n
d
s
tep
:
a
m
in
o
r
ity
s
am
p
l
e
X
j
is
ar
b
itra
r
ily
s
elec
ted
f
r
o
m
R
i
,
an
d
a
n
ew
s
y
n
th
esized
s
am
p
le
X
new
is
g
en
er
ated
b
ased
o
n
(
2
)
.
X
n
ew
=
X
i
+
w
(
X
j
−
X
i
)
(
2
)
wh
er
e
w
is
a
v
alu
e
b
etwe
en
0
an
d
1
t
h
at
ca
n
b
e
ch
o
s
en
at
r
a
n
d
o
m
.
-
T
h
ir
d
s
tep
:
i
f
th
e
n
u
m
b
e
r
o
f
n
ew
s
am
p
les
s
y
n
th
esis
ed
b
ased
o
n
X
i
is
less
th
an
th
e
o
v
e
r
s
am
p
lin
g
r
ate,
p
r
o
ce
ed
t
o
s
tep
2
.
I
n
im
b
alan
ce
d
class
if
icatio
n
task
s
,
th
e
m
in
o
r
ity
class
is
u
s
u
ally
th
e
m
o
s
t
im
p
o
r
tan
t.
B
y
s
y
n
th
esizin
g
n
ew
ex
am
p
les
f
r
o
m
t
h
e
m
in
o
r
ity
class
,
th
e
SMOT
E
tech
n
iq
u
e
is
u
s
ed
to
in
cr
ea
s
e
th
e
n
u
m
b
er
o
f
ex
am
p
les
f
r
o
m
th
e
m
in
o
r
ity
.
T
h
is
allo
ws
th
e
m
o
d
el
to
o
u
tp
er
f
o
r
m
th
e
m
ajo
r
ity
class
in
p
r
ed
i
ctin
g
th
e
class
o
r
p
r
o
b
a
b
ilit
y
o
f
th
e
m
in
o
r
ity
class
.
Fig
u
r
e
3
s
h
o
ws
a
c
o
m
p
a
r
is
o
n
o
f
th
e
n
u
m
b
er
o
f
attac
k
cla
s
s
d
is
tr
ib
u
tio
n
s
o
n
th
e
E
lectr
a
d
ataset
b
ef
o
r
e
an
d
af
ter
ap
p
ly
in
g
th
e
S
MO
T
E
m
eth
o
d
.
Fig
u
r
e
3
(
a)
s
h
o
ws
th
e
im
b
alan
ce
d
class
d
is
tr
ib
u
tio
n
b
ef
o
r
e
SMOT
E
,
a
n
d
Fig
u
r
e
3
(
b
)
th
e
b
ala
n
ce
d
cl
ass
d
is
tr
ib
u
tio
n
af
ter
SMOT
E
.
(
a)
(
b
)
Fig
u
r
e
3
.
T
h
e
n
u
m
b
er
o
f
attac
k
class
d
is
tr
ib
u
tio
n
s
(
a)
b
ef
o
r
e
SMOT
E
an
d
(
b
)
af
ter
SMOT
E
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
1
,
Ap
r
il
20
25
:
1
33
-
1
44
138
2
.
4
.
Aut
o
enco
der
An
au
to
en
c
o
d
er
is
a
t
y
p
e
o
f
n
eu
r
al
n
etwo
r
k
ar
c
h
itectu
r
e
th
a
t
co
n
s
is
ts
o
f
an
en
c
o
d
er
an
d
a
d
ec
o
d
er
to
en
co
d
e
in
p
u
t
d
ata
to
ess
en
tial
f
ea
tu
r
es.
T
h
e
o
r
ig
in
al
i
n
p
u
t
is
r
eb
u
ilt
f
r
o
m
th
e
c
o
m
p
r
ess
ed
r
ep
r
esen
tatio
n
.
T
h
e
en
co
d
er
d
er
iv
es
f
ea
tu
r
es
f
r
o
m
r
aw
d
ata,
an
d
th
e
d
ec
o
d
e
r
r
eb
u
ild
s
th
e
d
ata
u
s
in
g
t
h
ese
f
ea
t
u
r
es.
T
h
e
ex
t
r
ac
ted
f
ea
tu
r
es
allo
w
th
e
d
ec
o
d
er
to
r
ec
o
n
s
tr
u
ct
th
e
d
ata.
T
h
e
AE
ar
ch
ite
ctu
r
e
co
n
s
is
ts
o
f
in
p
u
t,
laten
t,
an
d
o
u
tp
u
t
lay
er
s
co
n
n
ec
ted
b
etwe
en
n
e
u
r
o
n
s
.
y
=
α
(
n
.
x
T
+
b
)
(
3
)
W
h
er
e
y
is
v
ec
to
r
o
u
tp
u
t
,
x
is
v
ec
to
r
in
p
u
t,
b
is
a
b
ias v
alu
e,
n
is
th
e
v
ec
to
r
o
f
n
e
u
r
o
n
co
n
n
ec
tio
n
weig
h
ts
,
α
is
ac
tiv
atio
n
f
u
n
cti
o
n
,
a
n
d
x
T
is
t
h
e
tr
an
s
p
o
s
e
o
f
th
e
in
p
u
t v
ec
to
r
x
.
T
h
e
s
tr
u
ctu
r
e
o
f
AE
is
s
h
o
wn
in
Fig
u
r
e
4
.
T
h
e
au
to
e
n
co
d
e
r
s
tag
e
is
p
er
f
o
r
m
e
d
b
y
ad
ju
s
tin
g
th
e
v
ar
iatio
n
o
f
th
e
d
ata
b
ef
o
r
e
d
im
en
s
io
n
r
ed
u
ctio
n
.
Var
iatio
n
is
cr
u
cial
to
th
e
class
if
icatio
n
p
r
o
ce
s
s
.
T
ab
le
2
d
ep
ict
t
h
e
p
ar
a
m
eter
s
in
an
o
m
al
y
d
etec
tio
n
an
d
m
u
lti
-
class
class
if
icatio
n
.
Fig
u
r
e
4
.
T
h
e
s
tr
u
ctu
r
e
o
f
a
u
to
en
co
d
er
s
T
ab
le
2
.
Mo
d
el
p
ar
am
ete
r
s
f
o
r
a
u
to
en
co
d
er
M
o
d
e
l
a
r
c
h
i
t
e
c
t
u
r
e
s
A
n
o
m
a
l
y
M
u
l
t
i
-
C
l
a
ss
En
c
o
d
e
r
D
e
n
se
1
64
1
2
8
D
e
n
se
2
32
64
D
e
c
o
d
e
r
D
e
n
se
1
32
64
D
e
n
se
2
64
1
2
8
Ep
o
c
h
20
50
B
a
t
c
h
s
i
z
e
50
1
0
0
Th
e
o
p
t
i
m
i
z
e
r
A
d
a
d
e
l
t
a
A
d
a
d
e
l
t
a
A
c
t
i
v
a
t
i
o
n
f
u
n
c
t
i
o
n
o
f
e
a
c
h
l
a
y
e
r
Re
LU
Re
LU
A
c
t
i
v
a
t
i
o
n
f
u
n
c
t
i
o
n
o
f
o
u
t
p
u
t
l
a
y
e
r
S
o
f
t
M
a
x
S
o
f
t
M
a
x
2
.
5
.
T
re
e
c
la
s
s
if
ica
t
io
n
T
h
e
tr
ain
in
g
p
r
o
ce
s
s
s
tar
ts
in
th
e
th
ir
d
p
h
ase
wh
ich
is
d
o
n
e
b
y
s
p
litt
in
g
th
e
d
ata
in
to
tr
ain
in
g
an
d
test
in
g
.
Sp
ec
if
ically
,
8
0
%
o
f
th
e
d
ata
is
u
s
ed
to
tr
ain
,
an
d
th
e
r
em
ain
in
g
2
0
%
is
u
s
ed
to
test
,
r
an
d
o
m
ly
ass
ig
n
ed
.
I
n
th
is
s
tu
d
y
,
2
0
%
o
f
r
ec
o
r
d
s
f
r
o
m
th
e
b
e
n
ig
n
cl
ass
wer
e
r
an
d
o
m
ly
s
elec
ted
f
o
r
test
in
g
,
wh
ile
th
e
r
em
ain
in
g
8
0
%
wer
e
u
tili
ze
d
f
o
r
tr
ain
in
g
.
Ad
d
itio
n
ally
,
to
m
ain
tain
d
ataset
b
alan
ce
in
ter
m
s
o
f
attac
k
class
es,
2
0
%
o
f
th
e
s
am
p
les
f
r
o
m
ea
c
h
attac
k
class
ar
e
s
et
asid
e
f
o
r
test
in
g
,
an
d
th
e
r
em
ain
in
g
8
0
%
ar
e
em
p
lo
y
ed
f
o
r
tr
ain
in
g
.
Su
b
s
eq
u
e
n
tly
,
all
th
e
2
0
%
s
eg
m
en
ts
ar
e
co
n
s
o
lid
ated
to
co
n
s
tr
u
ct
th
e
test
s
et,
an
d
th
e
s
am
e
p
r
o
ce
d
u
r
e
is
ap
p
lied
to
th
e
tr
ain
in
g
s
et.
T
h
e
s
elec
tio
n
o
f
class
if
icatio
n
alg
o
r
ith
m
s
is
b
as
ed
o
n
th
e
r
esu
lts
o
f
ex
p
er
im
en
tal
wo
r
k
o
n
s
ev
e
r
al
class
if
ier
s
,
n
am
ely
DT
,
KN
N,
SVM,
L
R
,
an
d
R
F.
T
h
e
r
esu
lts
o
f
m
u
lti
-
class
class
if
icatio
n
ex
p
er
im
en
ts
s
h
o
w
th
at
R
F
an
d
DT
alg
o
r
ith
m
s
h
av
e
th
e
h
i
g
h
est
ac
cu
r
ac
y
r
esu
lts
.
On
th
e
b
asis
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
S
MOTE
tr
ee
-
b
a
s
ed
a
u
to
en
c
o
d
er mu
lti
-
s
ta
g
e
d
etec
tio
n
fo
r
ma
n
-
in
-
th
e
-
mid
d
le
…
(
F
r
eska
R
o
la
n
s
a
)
139
th
e
r
esu
lts
o
f
th
e
ex
p
er
im
e
n
t
al
test
s
o
n
th
e
E
lectr
a
d
atas
et,
it
was
f
o
u
n
d
th
at
th
e
R
F
alg
o
r
ith
m
h
as
th
e
d
is
ad
v
an
tag
e
o
f
s
h
o
win
g
m
o
r
e
d
etec
tio
n
f
au
lts
th
an
th
e
DT
.
I
n
ad
d
itio
n
,
R
F
alg
o
r
ith
m
s
h
av
e
d
if
f
icu
lt
y
in
ter
p
r
etin
g
d
ata,
wh
ich
is
m
o
r
e
am
b
ig
u
o
u
s
f
o
r
th
e
cla
s
s
if
icatio
n
p
r
o
ce
s
s
.
T
h
er
ef
o
r
e
,
in
th
is
wo
r
k
,
DT
alg
o
r
ith
m
w
ith
h
y
p
er
p
a
r
am
ete
r
o
p
tim
izatio
n
is
ad
o
p
te
d
f
o
r
m
o
d
els.
Par
am
eter
tu
n
in
g
is
co
n
d
u
cte
d
o
n
th
e
DT
an
o
m
aly
d
etec
tio
n
a
n
d
m
u
lti
-
class
clas
s
if
icatio
n
.
R
an
d
o
m
a
n
d
g
r
id
s
ea
r
ch
es
ar
e
co
m
b
i
n
e
d
an
d
u
s
ed
to
s
elec
t
th
e
b
est
h
y
p
er
p
ar
am
eter
v
alu
es.
I
n
T
ab
le
3
,
th
e
v
alu
es
s
elec
ted
f
o
r
class
if
icatio
n
ar
e
m
ar
k
ed
with
aster
is
k
s
an
d
in
b
o
ld
.
B
ased
o
n
e
x
p
er
i
m
en
ts
co
m
p
ar
in
g
th
e
r
esu
lts
o
f
th
e
b
est
class
if
ier
s
,
we
s
elec
t
th
e
p
ar
am
eter
s
o
f
th
e
g
r
id
s
ea
r
ch
.
T
h
e
p
r
o
ce
s
s
o
f
m
u
lti
-
class
class
if
icatio
n
f
o
r
d
ataset
is
d
escr
ib
ed
in
Alg
o
r
it
h
m
1
.
T
ab
le
3
.
Hy
p
er
p
ar
a
m
eter
t
u
n
i
n
g
H
y
p
e
r
p
a
r
a
me
t
e
r
s
R
a
n
d
o
m
i
z
e
s
e
a
r
c
h
G
r
i
d
s
e
a
r
c
h
M
a
x
i
m
u
m
d
e
p
t
h
o
f
t
r
e
e
N
o
n
e
,
2
,
4
,
6
,
8
*
,
1
0
3
,
5
,
7
,
8*
,
1
0
M
i
n
i
m
u
m
n
u
m
b
e
r
o
f
s
a
mp
l
e
s
t
o
a
s
p
l
i
t
2*
,
5
,
1
0
2*
,
4
,
5
,
7
M
i
n
i
m
u
m
n
u
m
b
e
r
o
f
s
a
mp
l
e
s
t
o
b
e
a
t
a
l
e
a
f
n
o
d
e
1
,
2
,
4*
1
,
2
,
3*
,4
Alg
o
r
ith
m
1
.
T
h
e
p
s
eu
d
o
co
d
e
o
f
tr
ee
d
etec
tio
n
with
h
y
p
er
p
a
r
am
eter
o
p
tim
izatio
n
Input X: time, smac, dmac, sip, dip, request, fc,error, madd, data
Output O: Normal , Recognition attack, Read attack, Write attack,
Responses attack,
Force error Attack.
Function TreeDetection(Sample D, Input X, Output O, Hyperparameters H):
If stopping_condition(D, X) is true then
Leaf = createNode()
leafLabel = classify(D, O)
Return
Leaf
Root = createNode()
Root.test_condition = findBestSplit(D, X, H)
Z = {z | z is a potential outcome of Root.test_condition}
For each value z in Z:
Subclass = {d | Root.test_condition(d) = z and d is in D}
Child = TreeDetection(Subclass, X, O, H)
Add Child as a child of Root and label the edge
{Root → Child} as z
Return Root
2
.
6
.
E
v
a
lua
t
i
o
n
m
et
rics
T
h
e
q
u
ality
o
f
a
m
ac
h
in
e
le
ar
n
in
g
m
o
d
el
o
r
alg
o
r
ith
m
is
d
eter
m
in
ed
b
y
a
p
ar
am
et
er
ca
lled
th
e
ev
alu
atio
n
m
atr
ix
.
Sin
ce
E
lect
r
a
d
ataset
u
s
ed
h
as
im
b
alan
ce
d
d
ata,
p
r
ec
is
io
n
,
r
ec
all/d
etec
t
io
n
r
ate
(
DR
)
,
an
d
F1
s
co
r
e
wer
e
s
elec
ted
a
s
m
etr
ics in
th
e
p
er
f
o
r
m
an
ce
ev
alu
at
io
n
.
T
h
ese
m
etr
ics ar
e
d
ef
in
ed
in
(
4
)
,
(
5
)
an
d
(
6
)
.
T
h
e
ac
cu
r
ac
y
m
etr
ic
was n
o
t a
d
o
p
ted
i
n
th
e
m
o
d
el
ev
alu
atio
n
.
Pr
e
c
isio
n
=
TP
TP
+
FP
(4
)
R
e
c
a
l
l
/
DR
=
TP
TP
+
FN
(
5
)
F1
Score
=
2
∗
(
Pr
ecis
i
o
n
∗
Recal
l
)
Pr
ec
i
s
i
o
n
+
Recal
l
(
6
)
W
h
er
e
T
P
is
tr
u
e
p
o
s
itiv
e
,
TN
is
tr
u
e
n
eg
ativ
e
,
FP
is
f
alse p
o
s
itiv
e
,
an
d
FN
is
f
alse n
eg
ativ
e
.
3.
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
in
clu
d
es
th
e
p
er
f
o
r
m
an
ce
a
n
d
c
o
m
p
ar
is
o
n
o
f
t
h
e
p
r
o
p
o
s
ed
m
o
d
els
,
a
d
etailed
an
aly
s
is
an
d
d
is
cu
s
s
io
n
.
3
.
1
.
P
er
f
o
r
m
a
nce
ev
a
lua
t
io
n
C
o
m
m
u
n
icatio
n
u
s
in
g
M
o
d
b
u
s
p
r
o
to
c
o
l
is
u
s
ed
b
y
PLC
to
co
n
tr
o
l
p
r
ess
u
r
e
(
o
u
tp
u
t)
an
d
v
alv
e
(
PGVA
s
ettin
g
)
p
ar
am
eter
s
with
s
in
g
le
r
eg
is
ter
wr
itin
g
a
p
p
ly
in
g
f
u
n
ctio
n
co
d
e
(
0
6
)
.
I
n
ad
d
itio
n
,
f
u
n
ctio
n
co
d
e
(
0
4
)
is
ap
p
lied
to
r
ea
d
in
p
u
t
r
eg
is
ter
s
o
n
th
e
ac
tp
r
ess
p
ar
am
eter
f
o
r
t
h
e
ac
tu
al
o
u
t
p
u
t
p
r
ess
u
r
e
o
f
PGVA
an
d
th
e
ac
tp
r
ess
tan
k
t
o
k
n
o
w
th
e
ac
tu
al
p
r
ess
u
r
e
o
f
th
e
PGVA
tan
k
.
PLC
is
co
n
n
ec
ted
to
T
P
-
L
I
NK
s
witch
v
ia
E
th
er
n
et
with
I
P
ad
d
r
ess
1
9
2
.
1
6
8
.
0
.
1
0
2
/
2
4
.
I
n
ad
d
itio
n
,
th
er
e
ar
e
SC
ADA
an
d
I
DS
s
y
s
tem
s
th
at
u
s
e
Dell
Po
wer
E
d
g
e
R
2
5
0
s
er
v
e
r
s
w
ith
I
P
a
d
d
r
ess
r
an
g
es
1
9
2
.
1
6
8
.
0
.
1
0
0
/2
4
an
d
1
9
2
.
1
6
8
.
0
.
9
9
/
2
4
.
C
y
b
er
s
ec
u
r
ity
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
1
,
Ap
r
il
20
25
:
1
33
-
1
44
140
attac
k
s
ar
e
m
o
s
tly
ca
r
r
ied
o
u
t
o
n
SC
ADA.
Attack
er
PC
is
u
s
ed
as
an
attac
k
er
to
p
er
f
o
r
m
MI
T
M
attac
k
s
with
I
P
R
an
g
e
1
9
2
.
1
6
8
.
0
.
1
0
3
/2
4
.
I
DS
is
co
n
n
ec
ted
to
th
e
s
witch
to
ca
p
tu
r
e
all
n
etwo
r
k
ac
tiv
ities
th
at
o
cc
u
r
in
th
e
test
b
ed
.
Fu
r
th
e
r
m
o
r
e
,
T
P
lin
k
r
o
u
ter
s
ar
e
u
s
ed
to
co
n
n
ec
t
th
e
test
b
ed
t
o
an
o
p
en
n
etwo
r
k
o
r
t
h
e
in
ter
n
et.
Fin
ally
,
th
e
attac
k
er
'
s
d
ev
ice
ca
n
co
n
n
ec
t th
r
o
u
g
h
t
h
e
s
witch
v
ia
E
th
er
n
et
ca
b
le
o
r
r
o
u
ter
v
i
a
th
e
in
ter
n
et.
T
h
e
m
o
d
el
ac
h
ie
v
es
p
er
f
ec
t
an
o
m
al
y
d
etec
tio
n
with
1
0
0
%
p
r
ec
is
io
n
,
r
ec
all/DR
an
d
f
1
s
co
r
e.
Ad
d
itio
n
ally
,
th
e
m
u
lti
-
class
c
lass
if
icatio
n
o
f
MI
T
M
with
th
e
m
o
d
el
d
etec
tio
n
f
r
am
ewo
r
k
ac
h
iev
es
an
f
1
s
co
r
e
o
f
9
9
%.
T
h
e
a
n
o
m
aly
d
e
tectio
n
an
d
m
u
lti
-
class
class
if
i
ca
tio
n
r
ep
o
r
t is d
is
p
lay
ed
in
T
ab
le
4
.
T
ab
le
4
.
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
p
r
o
p
o
s
ed
m
o
d
el
B
i
n
a
r
y
C
l
a
ss
P
r
e
c
i
s
i
o
n
DR
F
1
S
c
o
r
e
S
u
p
p
M
u
l
t
i
-
C
l
a
ss
e
s
P
r
e
c
i
s
i
o
n
DR
F
1
S
c
o
r
e
S
u
p
p
N
o
r
mal
1
.
0
0
1
.
0
0
1
.
0
0
4
9
0
0
N
o
r
mal
1
.
0
0
0
.
9
8
0
.
9
9
4
9
3
7
F
o
r
c
e
Er
r
o
r
A
t
t
a
c
k
1
.
0
0
1
.
0
0
1
.
0
0
5
0
6
6
R
e
a
d
A
t
t
a
c
k
1
.
0
0
0
.
9
8
0
.
9
9
4
835
A
n
o
m
a
l
y
1
.
0
0
1
.
0
0
1
.
0
0
4
8
6
4
R
e
c
o
g
n
i
t
i
o
n
A
t
t
a
c
k
1
.
0
0
1
.
0
0
1
.
0
0
5
0
4
9
R
e
p
l
a
y
A
t
t
a
c
k
0
.
9
8
1
.
0
0
0
.
9
9
4
9
7
3
R
e
s
p
o
n
se
A
t
t
a
c
k
0
.9
8
1
.
0
0
0
.
9
9
49
87
W
r
i
t
e
A
t
t
a
c
k
1
.
0
0
1
.
0
0
1
.
0
0
4
9
7
8
A
c
c
u
r
a
c
y
1
.
0
0
9
7
6
4
A
c
c
u
r
a
c
y
0
.
9
9
3
5
0
4
4
Fo
r
m
u
ltip
le
class
es,
in
clu
d
in
g
n
o
r
m
al,
f
o
r
ce
e
r
r
o
r
,
r
ea
d
,
r
ec
o
g
n
itio
n
,
r
ep
lay
,
r
esp
o
n
s
e,
an
d
wr
ite
.
W
ith
o
u
t
u
s
in
g
an
y
ex
ter
n
al
tr
ain
in
g
d
ata,
th
e
r
esu
lts
s
h
o
w
th
at
th
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
c
an
p
r
o
v
id
e
s
u
p
e
r
io
r
class
if
icatio
n
r
esu
lts
f
o
r
MI
T
M
m
u
lti
-
class
clas
s
if
icatio
n
.
Fig
u
r
e
5
d
e
m
o
n
s
tr
ates
th
e
co
n
f
u
s
io
n
m
atr
i
x
(
C
M)
r
esu
lts
o
f
th
e
p
r
o
p
o
s
ed
STA
M
m
o
d
el
.
Fig
u
r
e
5
(
a
)
an
o
m
aly
d
etec
tio
n
'
s
C
M
an
d
Fig
u
r
e
5
(
b
)
m
u
lti
-
class
class
if
icatio
n
'
s
C
M.
(
a)
(
b
)
Fig
u
r
e
5
.
C
M
p
r
o
p
o
s
ed
m
o
d
el
s
(
a)
an
o
m
aly
d
etec
tio
n
a
n
d
(
b
)
m
u
lti
-
class
class
if
icatio
n
3
.
2
.
Co
m
pa
ra
t
iv
e
a
na
ly
s
is
C
o
m
p
ar
ativ
e
an
aly
s
is
was
co
n
d
u
cted
o
n
an
o
m
aly
d
etec
tio
n
an
d
m
u
lti
-
class
class
if
icati
o
n
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
with
o
th
er
m
o
d
els.
T
ab
le
5
c
o
m
p
a
r
es
th
e
ev
alu
atio
n
r
esu
lts
o
f
d
if
f
er
e
n
t
m
o
d
els,
in
clu
d
in
g
R
F,
SVM,
NN,
O
C
SVM,
I
F,
GAN
+
DNN,
DAE
+
SMOT
E
+T
-
L
in
k
+X
Gb
o
o
s
t
an
d
th
e
p
r
o
p
o
s
ed
m
o
d
el.
T
ab
le
5
.
C
o
m
p
a
r
is
o
n
m
eth
o
d
s
f
o
r
an
o
m
aly
d
etec
tio
n
M
e
t
h
o
d
P
r
e
c
i
s
i
o
n
DR
F
1
S
c
o
r
e
RF
[
7
]
9
8
,
7
7
9
8
,
7
1
9
8
,
7
4
S
V
M
[
7
]
9
7
,
5
6
1
0
0
9
8
,
7
6
NN
[
7
]
9
6
,
9
2
1
0
0
9
8
,
4
3
O
C
S
V
M
[
7
]
9
8
,
6
2
9
8
,
5
6
9
8
,
5
9
IF
[
7
]
8
7
,
3
9
1
0
0
9
3
,
2
7
G
A
N
+
D
N
N
[
3
8
]
-
98
-
D
A
E
+
S
M
O
TE
+
T
-
Li
n
k
+
X
G
B
o
o
s
t
[
3
9
]
1
0
0
1
0
0
1
0
0
Pr
o
p
o
sed
m
o
d
e
l
1
0
0
1
0
0
1
0
0
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2
5
0
2
-
4
7
52
S
MOTE
tr
ee
-
b
a
s
ed
a
u
to
en
c
o
d
er mu
lti
-
s
ta
g
e
d
etec
tio
n
fo
r
ma
n
-
in
-
th
e
-
mid
d
le
…
(
F
r
eska
R
o
la
n
s
a
)
141
T
h
e
E
lectr
a
d
ataset
is
u
s
ed
f
o
r
th
is
co
m
p
ar
is
o
n
,
an
d
th
e
an
o
m
aly
d
etec
tio
n
r
esu
lts
ar
e
ev
alu
ated
u
s
in
g
p
r
ec
is
io
n
,
DR
,
an
d
F1
s
co
r
es.
I
t
ca
n
b
e
s
ee
n
th
at
th
e
p
e
r
f
o
r
m
an
ce
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
is
eq
u
al
to
m
o
d
el
DAE
+
SMOT
E
+T
-
L
in
k
+
XGb
o
o
s
t
an
d
h
ig
h
e
r
th
an
th
e
o
th
er
m
eth
o
d
s
.
T
h
e
p
r
o
p
o
s
ed
an
o
m
aly
d
etec
tio
n
m
o
d
el
ac
h
iev
es p
r
ec
is
io
n
,
d
ete
ctio
n
r
ate,
an
d
an
F1
s
co
r
e
o
f
1
0
0
%.
T
ab
le
6
s
h
o
ws
th
e
co
m
p
ar
is
o
n
o
f
th
e
ev
alu
atio
n
r
esu
lts
b
etwe
en
DAE
+SM
OT
E
+T
-
L
in
k
+X
GB
o
o
s
t
an
d
th
e
p
r
o
p
o
s
ed
m
o
d
el,
as
well
as
th
e
ev
alu
atio
n
o
f
th
e
m
u
lti
-
class
cla
s
s
if
icat
io
n
r
esu
lt
u
s
in
g
p
r
ec
is
io
n
,
DR
,
an
d
F1
s
co
r
es.
T
o
th
e
b
est
o
f
o
u
r
k
n
o
wled
g
e,
th
e
r
e
is
o
n
ly
o
n
e
s
tu
d
y
[
3
9
]
t
h
at
h
as
r
ep
o
r
ted
an
a
n
o
m
alo
u
s
m
u
lti
-
class
clas
s
if
icatio
n
.
T
h
e
r
esu
lts
in
d
icate
d
th
at
th
e
p
r
o
p
o
s
ed
m
o
d
el
e
x
h
ib
ited
p
r
e
em
in
en
ce
p
er
f
o
r
m
an
ce
co
m
p
a
r
ed
to
th
e
alter
n
ativ
e
m
o
d
els,
ac
h
iev
in
g
im
p
r
o
v
em
e
n
ts
o
f
u
p
t
o
9
9
,
3
7
%
in
DR
,
an
d
9
9
.
3
7
%
in
F1
-
s
co
r
e
.
I
n
a
d
d
itio
n
,
t
h
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
el
i
n
p
er
f
o
r
m
in
g
a
m
u
lti
-
class
c
lass
if
icatio
n
o
n
u
n
b
alan
ce
d
d
ata
h
as
also
b
ee
n
ev
alu
ated
u
s
in
g
p
r
ec
is
io
n
r
ec
a
ll c
u
r
v
es.
Fig
u
r
e
6
s
h
o
ws th
e
c
u
r
v
e
with
n
ea
r
ly
p
e
r
f
ec
t c
lass
if
icatio
n
r
esu
lts
.
I
n
Mo
d
b
u
s
tr
a
f
f
ic,
s
o
m
e
an
o
m
alies
d
u
e
to
r
ea
d
attac
k
s
ar
e
in
co
r
r
ec
tly
class
if
ied
as
r
ep
lay
attac
k
s
.
An
o
m
alies
f
r
o
m
r
ea
d
an
d
r
e
p
l
ay
attac
k
s
d
is
p
lay
s
im
ilar
n
et
wo
r
k
tr
a
f
f
ic
p
atter
n
s
,
as
o
u
tlin
ed
in
th
e
r
e
p
o
r
t
[
7
]
.
T
h
e
r
ea
d
a
ttac
k
an
d
r
ep
lay
a
ttack
class
es
r
ep
r
esen
t
a
m
i
n
o
r
ity
o
f
t
h
e
d
ata
in
th
e
im
b
a
lan
ce
d
ataset.
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
s
h
o
we
d
im
p
r
o
v
ed
class
if
icatio
n
p
er
f
o
r
m
an
ce
f
o
r
h
an
d
lin
g
im
b
alan
ce
s
ag
ain
s
t
r
ep
lay
attac
k
s
,
with
o
n
ly
1
m
is
class
if
icatio
n
co
m
p
ar
ed
to
th
e
o
th
er
m
o
d
el
[
3
9
]
,
wh
ic
h
h
a
d
9
m
is
class
if
icati
o
n
s
.
Fu
r
th
e
r
m
o
r
e
,
STAM
m
o
d
el
o
n
ly
m
is
p
r
ed
i
cted
119
r
ea
d
attac
k
s
co
m
p
ar
ed
to
6
8
5
8
[
3
9
]
.
Fig
u
r
e
7
illu
s
tr
ates
th
e
m
o
d
el
co
m
p
ar
is
o
n
o
f
th
e
r
esu
lts
o
f
m
is
class
if
icatio
n
ag
ain
s
t
m
in
o
r
ity
class
es,
n
am
ely
th
e
r
ea
d
attac
k
an
d
r
e
p
lay
attac
k
.
T
ab
le
6
.
C
o
m
p
a
r
is
o
n
m
eth
o
d
s
f
o
r
m
u
lti
-
class
class
if
icatio
n
M
e
t
h
o
d
P
r
e
c
i
s
i
o
n
DR
F1
-
S
c
o
r
e
D
A
E
+
S
M
O
TE
+
T
-
Li
n
k
+
X
G
B
o
o
s
t
[
3
9
]
9
9
,
9
9
9
7
,
6
7
9
8
,
5
0
Pr
o
p
o
sed
M
o
d
e
l
9
9
,
3
8
9
9
,
3
7
9
9
,
3
7
Fig
u
r
e
6
.
Pre
cisi
o
n
r
e
ca
ll c
u
r
v
e
f
o
r
m
u
lti
-
class
class
if
icatio
n
Fig
u
r
e
7
.
C
o
m
p
a
r
is
o
n
o
f
m
is
class
if
icatio
n
f
o
r
m
in
o
r
ity
attac
k
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
52
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
38
,
No
.
1
,
Ap
r
il
20
25
:
1
33
-
1
44
142
I
n
f
u
tu
r
e
wo
r
k
s
,
th
e
ef
f
ec
tiv
en
ess
o
f
d
etec
tio
n
will
b
e
d
e
v
elo
p
ed
with
d
if
f
e
r
en
t
t
y
p
es
o
f
att
ac
k
s
,
s
u
ch
as
b
o
tn
et,
DD
o
S
,
an
d
ze
r
o
-
d
a
y
with
o
u
t
u
s
in
g
t
h
e
r
ea
l
en
v
ir
o
n
m
en
t
o
f
SC
ADA.
Mo
d
el
d
e
v
elo
p
m
en
t
ca
n
also
u
s
e
o
th
er
in
d
u
s
tr
ial
co
m
m
u
n
icatio
n
p
r
o
to
co
ls
s
u
ch
as
et
h
er
ca
t,
p
r
o
f
i
n
et,
an
d
o
th
er
s
th
at
h
av
e
d
if
f
er
en
t
ch
ar
ac
ter
is
tics
o
r
f
ea
tu
r
es.
M
o
r
eo
v
e
r
,
th
e
p
o
ten
tial
to
c
o
m
p
ar
e
th
e
p
er
f
o
r
m
an
ce
o
f
t
h
e
m
o
d
el
with
o
th
er
d
etec
tio
n
m
eth
o
d
s
r
elate
d
to
t
h
e
d
ev
elo
p
m
en
t
o
f
f
u
tu
r
e
attac
k
ty
p
es is
p
o
ten
tiall
y
p
o
s
s
ib
le.
4.
CO
NCLU
SI
O
N
T
h
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
p
r
o
p
o
s
ed
m
o
d
el
s
h
o
ws
a
n
ea
r
l
y
p
er
f
ec
t
class
if
icatio
n
.
T
h
e
p
r
o
p
o
s
ed
an
o
m
aly
d
etec
tio
n
m
o
d
el
d
e
m
o
n
s
tr
ates
an
o
p
tim
al
lev
el
o
f
p
r
ec
is
io
n
,
d
etec
tio
n
r
ate,
a
n
d
F1
s
co
r
e,
with
a
v
alu
e
o
f
1
0
0
%.
I
n
th
e
m
u
lti
-
class
c
las
s
if
icatio
n
,
th
e
p
r
o
p
o
s
ed
m
o
d
el
h
as
th
e
h
ig
h
est
d
e
tectio
n
r
ate
an
d
F1
s
co
r
es
co
m
p
ar
e
d
t
o
o
th
er
m
et
h
o
d
s
.
Fu
r
th
er
m
o
r
e,
t
h
e
m
u
lti
-
class
class
if
icatio
n
p
er
f
o
r
m
a
n
ce
u
s
in
g
STAM
is
b
etter
in
th
e
m
in
o
r
ity
attac
k
class
es
(
r
ea
d
an
d
r
ep
lay
att
ac
k
)
,
wh
ich
h
a
v
e
f
ewe
r
m
is
class
if
icatio
n
s
ev
en
th
o
u
g
h
th
e
attr
ib
u
tes o
r
f
ea
tu
r
es h
av
e
s
im
ilar
p
atter
n
s
.
ACK
NO
WL
E
DG
E
M
E
NT
S
T
h
e
r
esear
ch
h
as
b
ee
n
f
u
n
d
ed
b
y
Un
iv
er
s
itas
Gad
jah
Ma
d
a
v
ia
th
e
R
T
A
s
ch
e
m
e
n
u
m
b
er
5
0
7
5
/UN1
.
P.I
I
/Dit
-
L
it/P
T
.
0
1
.
0
1
/2
0
2
3
.
T
h
e
au
th
o
r
s
ar
e
g
r
atef
u
l f
o
r
th
e
f
in
a
n
cial
s
u
p
p
o
r
t.
RE
F
E
R
E
NC
E
S
[
1
]
M
.
C
o
n
t
i
,
D
.
D
o
n
a
d
e
l
,
a
n
d
F
.
Tu
r
r
i
n
,
“
A
S
u
r
v
e
y
o
n
I
n
d
u
s
t
r
i
a
l
C
o
n
t
r
o
l
S
y
st
e
m
T
e
st
b
e
d
s
a
n
d
D
a
t
a
se
t
s
f
o
r
S
e
c
u
r
i
t
y
R
e
s
e
a
r
c
h
,
”
I
EEE
C
o
m
m
u
n
.
S
u
r
v
.
T
u
t
o
r
i
a
l
s
,
v
o
l
.
2
3
,
n
o
.
4
,
p
p
.
2
2
48
–
2
2
9
4
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
S
T.
2
0
2
1
.
3
0
9
4
3
6
0
.
[
2
]
A
.
B
.
A
j
ma
l
,
M
.
A
l
a
m,
A
.
A
.
K
h
a
l
i
q
,
S
.
K
h
a
n
,
Z.
Q
a
d
i
r
,
a
n
d
M
.
A
.
P
.
M
a
h
m
u
d
,
“
L
a
st
Li
n
e
o
f
D
e
f
e
n
se
:
R
e
l
i
a
b
i
l
i
t
y
t
h
r
o
u
g
h
I
n
d
u
c
i
n
g
C
y
b
e
r
T
h
r
e
a
t
H
u
n
t
i
n
g
w
i
t
h
D
e
c
e
p
t
i
o
n
i
n
S
C
A
D
A
N
e
t
w
o
r
k
s
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
9
,
p
p
.
1
2
6
7
8
9
–
1
2
6
8
0
0
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
1
1
1
4
2
0
.
[
3
]
G
.
Y
a
d
a
v
a
n
d
K
.
P
a
u
l
,
“
A
r
c
h
i
t
e
c
t
u
r
e
a
n
d
se
c
u
r
i
t
y
o
f
S
C
A
D
A
sy
st
e
ms:
A
r
e
v
i
e
w
,
”
I
n
t
.
J
.
C
r
i
t
.
I
n
f
ra
s
t
ru
c
t
.
Pro
t
.
,
v
o
l
.
3
4
,
p
.
1
0
0
4
3
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
c
i
p
.
2
0
2
1
.
1
0
0
4
3
3
.
[
4
]
M
.
B
r
i
s
t
o
w
,
“
A
S
A
N
S
2
0
2
1
S
u
r
v
e
y
:
O
T
/
I
C
S
C
y
b
e
r
sec
u
r
i
t
y
,
”
n
o
.
A
u
g
u
s
t
,
p
p
.
1
–
2
3
,
2
0
2
1
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
w
w
w
.
c
i
sa
.
g
o
v
/
c
r
i
t
i
c
a
l
-
i
n
f
r
a
st
r
u
c
t
u
r
e
-
s
e
c
t
o
r
s.
[
5
]
W
.
J.
B
r
o
a
d
,
J.
M
a
r
k
o
f
f
,
a
n
d
D
.
E.
S
a
n
g
e
r
,
“
I
sr
a
e
l
i
Te
s
t
o
n
W
o
r
m
C
a
l
l
e
d
C
r
u
c
i
a
l
i
n
I
r
a
n
N
u
c
l
e
a
r
D
e
l
a
y
.
”
h
t
t
p
s
:
/
/
w
w
w
.
n
y
t
i
me
s.c
o
m
/
2
0
1
1
/
0
1
/
1
6
/
w
o
r
l
d
/
mi
d
d
l
e
e
a
s
t
/
1
6
s
t
u
x
n
e
t
.
h
t
m
l
(
a
c
c
e
sse
d
F
e
b
.
1
9
,
2
0
2
4
)
.
[
6
]
C
.
Ea
t
o
n
a
n
d
D
.
V
o
l
z
,
“
U
.
S
.
P
i
p
e
l
i
n
e
c
y
b
e
r
a
t
t
a
c
k
f
o
r
c
e
s
c
l
o
s
u
r
e
,
”
2
0
2
1
.
h
t
t
p
s
:
/
/
w
w
w
.
w
sj
.
c
o
m
/
a
r
t
i
c
l
e
s
/
c
o
l
o
n
i
a
l
-
p
i
p
e
l
i
n
e
-
c
e
o
-
t
e
l
l
s
-
w
h
y
-
he
-
p
a
i
d
-
h
a
c
k
e
r
s
-
a
-
4
-
4
-
m
i
l
l
i
o
n
-
r
a
n
so
m
-
1
1
6
2
1
4
3
5
6
3
6
(
a
c
c
e
ss
e
d
D
e
c
.
2
0
,
2
0
2
4
)
.
[
7
]
Á
.
L.
P
.
G
ó
me
z
e
t
a
l
.
,
“
O
n
t
h
e
G
e
n
e
r
a
t
i
o
n
o
f
A
n
o
m
a
l
y
D
e
t
e
c
t
i
o
n
D
a
t
a
set
s
i
n
I
n
d
u
st
r
i
a
l
C
o
n
t
r
o
l
S
y
s
t
e
ms
,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
7
,
p
p
.
1
7
7
4
6
0
–
1
7
7
4
7
3
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
1
9
.
2
9
5
8
2
8
4
.
[
8
]
R
.
L
e
szc
z
y
n
a
,
“
R
e
v
i
e
w
o
f
c
y
b
e
r
s
e
c
u
r
i
t
y
a
ssess
me
n
t
m
e
t
h
o
d
s
:
A
p
p
l
i
c
a
b
i
l
i
t
y
p
e
r
s
p
e
c
t
i
v
e
,
”
C
o
m
p
u
t
.
S
e
c
u
r.
,
v
o
l
.
1
0
8
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
se.
2
0
2
1
.
1
0
2
3
7
6
.
[
9
]
P
.
W
l
a
z
l
o
e
t
a
l
.
,
“
M
a
n
-
in
-
t
h
e
-
m
i
d
d
l
e
a
t
t
a
c
k
s a
n
d
d
e
f
e
n
c
e
i
n
a
p
o
w
e
r
s
y
st
e
m
c
y
b
e
r
-
p
h
y
si
c
a
l
t
e
st
b
e
d
,
”
I
ET C
y
b
e
r
-
P
h
y
s
i
c
a
l
S
y
s
t
e
m
s:
T
h
e
o
ry
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
6
,
n
o
.
3
.
p
p
.
1
6
4
–
1
7
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
4
9
/
c
p
s
2
.
1
2
0
1
4
.
[
1
0
]
V
.
R
a
n
a
d
e
,
“
A
l
a
b
o
r
a
t
o
r
y
f
o
r
c
y
b
e
r
-
a
t
t
a
c
k
g
e
n
e
r
a
t
i
o
n
a
n
d
t
e
st
i
n
g
i
n
I
n
d
u
s
t
r
i
a
l
C
o
n
t
r
o
l
S
y
st
e
ms:
D
e
s
i
g
n
a
n
d
S
i
mu
l
a
t
i
o
n
,
”
2
0
2
1
.
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s:
/
/
r
e
s
o
l
v
e
r
.
t
u
d
e
l
f
t
.
n
l
/
u
u
i
d
:
a
d
5
5
4
d
6
8
-
4
5
0
3
-
4
5
4
4
-
b
5
1
b
-
e
4
8
3
7
9
f
c
7
2
1
6
.
[
1
1
]
Y
.
Y
a
n
g
,
H
.
-
Q
.
X
u
,
L
.
G
a
o
,
Y
.
-
B
.
Y
u
a
n
,
K
.
M
c
L
u
g
h
l
i
n
,
a
n
d
S
.
S
e
z
e
r
,
“
M
u
l
t
i
d
i
me
n
si
o
n
a
l
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
S
y
s
t
e
m
f
o
r
I
EC
6
1
8
5
0
-
B
a
s
e
d
S
C
A
D
A
N
e
t
w
o
r
k
s,”
I
E
EE
T
r
a
n
s
.
P
o
w
e
r De
l
i
v
.
,
v
o
l
.
3
2
,
n
o
.
2
,
p
p
.
1
0
6
8
–
1
0
7
8
,
2
0
1
7
.
[
1
2
]
M
.
B
a
s
h
e
n
d
y
,
S
.
El
t
a
n
b
o
u
l
y
,
A
.
T
a
n
t
a
w
y
,
a
n
d
A
.
Er
r
a
d
i
,
“
D
e
si
g
n
a
n
d
I
mp
l
e
men
t
a
t
i
o
n
o
f
C
y
b
e
r
-
P
h
y
si
c
a
l
A
t
t
a
c
k
s
o
n
M
o
d
b
u
s/
T
C
P
P
r
o
t
o
c
o
l
,
”
Wo
rl
d
C
o
n
g
ress
o
n
I
n
d
u
st
ri
a
l
C
o
n
t
r
o
l
S
y
st
e
m
s
S
e
c
u
r
i
t
y
(
W
C
I
C
S
S
-
2
0
2
0
)
,
n
o
.
D
e
c
e
mb
e
r
,
p
p
.
3
8
–
4
5
,
2
0
2
1
,
d
o
i
:
1
0
.
2
0
5
3
3
/
w
c
i
c
ss.
2
0
2
0
.
0
0
0
5
.
[
1
3
]
A
.
R
a
h
ma
n
,
G
.
M
u
s
t
a
f
a
,
A
.
Q
.
K
h
a
n
,
M
.
A
b
i
d
,
a
n
d
M
.
H
.
D
u
r
a
d
,
“
La
u
n
c
h
o
f
d
e
n
i
a
l
o
f
ser
v
i
c
e
a
t
t
a
c
k
s
o
n
t
h
e
m
o
d
b
u
s/
TC
P
p
r
o
t
o
c
o
l
a
n
d
d
e
v
e
l
o
p
m
e
n
t
o
f
i
t
s
p
r
o
t
e
c
t
i
o
n
m
e
c
h
a
n
i
sms
,
”
I
n
t
.
J
.
C
ri
t
.
I
n
f
r
a
st
ru
c
t
.
Pr
o
t
.
,
v
o
l
.
3
9
,
n
o
.
S
e
p
t
e
m
b
e
r
,
p
.
1
0
0
5
6
8
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
j
c
i
p
.
2
0
2
2
.
1
0
0
5
6
8
.
[
1
4
]
D
.
P
l
i
a
t
s
i
o
s,
P
.
S
a
r
i
g
i
a
n
n
i
d
i
s
,
T
.
La
g
k
a
s,
a
n
d
A
.
G
.
S
a
r
i
g
i
a
n
n
i
d
i
s,
“
A
S
u
r
v
e
y
o
n
S
C
A
D
A
S
y
s
t
e
ms
:
S
e
c
u
r
e
P
r
o
t
o
c
o
l
s,
I
n
c
i
d
e
n
t
s
,
Th
r
e
a
t
s
a
n
d
T
a
c
t
i
c
s,”
I
EE
E
C
o
m
m
u
n
i
c
a
t
i
o
n
s
S
u
rv
e
y
s
&
T
u
t
o
r
i
a
l
s
,
v
o
l
.
2
2
,
n
o
.
3
,
p
p
.
1
9
4
2
–
1
9
7
6
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
S
T.
2
0
2
0
.
2
9
8
7
6
8
8
.
[
1
5
]
P
.
D
u
c
l
i
n
,
“
R
a
n
s
o
mw
a
r
e
t
a
l
e
s:
T
h
e
M
i
t
M
a
t
t
a
c
k
t
h
a
t
r
e
a
l
l
y
h
a
d
a
M
a
n
i
n
t
h
e
M
i
d
d
l
e
,
”
2
0
2
3
.
h
t
t
p
s
:
/
/
n
e
w
s.so
p
h
o
s.
c
o
m/
e
n
-
u
s/
2
0
2
3
/
0
5
/
2
4
/
r
a
n
s
o
mw
a
r
e
-
t
a
l
e
s
-
t
h
e
-
mi
t
m
-
a
t
t
a
c
k
-
t
h
a
t
-
r
e
a
l
l
y
-
h
a
d
-
a
-
m
a
n
-
in
-
t
h
e
-
mi
d
d
l
e
/
(
a
c
c
e
s
se
d
F
e
b
.
2
1
,
2
0
2
4
)
.
[
1
6
]
N
.
W
a
n
g
,
Y
.
C
h
e
n
,
Y
.
H
u
,
W
.
Lo
u
,
a
n
d
Y
.
T
.
H
o
u
,
“
M
A
N
D
A
:
O
n
a
d
v
e
r
s
a
r
i
a
l
e
x
a
mp
l
e
d
e
t
e
c
t
i
o
n
f
o
r
n
e
t
w
o
r
k
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
st
e
m,
”
Pr
o
c
.
-
I
EE
E
I
N
FO
C
O
M
,
v
o
l
.
2
0
2
1
-
M
a
y
,
p
p
.
1
–
1
0
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
N
F
O
C
O
M
4
2
9
8
1
.
2
0
2
1
.
9
4
8
8
8
7
4
.
[
1
7
]
F
.
S
I
C
A
R
D
,
É.
ZA
M
A
I
,
a
n
d
J
.
M
.
F
LA
U
S
,
“
A
n
a
p
p
r
o
a
c
h
b
a
s
e
d
o
n
b
e
h
a
v
i
o
r
a
l
m
o
d
e
l
s
a
n
d
c
r
i
t
i
c
a
l
s
t
a
t
e
s
d
i
s
t
a
n
c
e
n
o
t
i
o
n
f
o
r
i
mp
r
o
v
i
n
g
c
y
b
e
r
s
e
c
u
r
i
t
y
o
f
i
n
d
u
s
t
r
i
a
l
c
o
n
t
r
o
l
s
y
st
e
ms,
”
Re
l
i
a
b
.
E
n
g
.
S
y
s
t
.
S
a
f
.
,
v
o
l
.
1
8
8
,
n
o
.
M
a
r
c
h
2
0
1
8
,
p
p
.
5
8
4
–
6
0
3
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
r
e
ss.
2
0
1
9
.
0
3
.
0
2
0
.
[
1
8
]
M
.
W
a
n
,
W
.
S
h
a
n
g
,
a
n
d
P
.
Z
e
n
g
,
“
D
o
u
b
l
e
B
e
h
a
v
i
o
r
C
h
a
r
a
c
t
e
r
i
st
i
c
s
f
o
r
O
n
e
-
C
l
a
ss
C
l
a
ss
i
f
i
c
a
t
i
o
n
A
n
o
m
a
l
y
D
e
t
e
c
t
i
o
n
i
n
N
e
t
w
o
r
k
e
d
C
o
n
t
r
o
l
S
y
st
e
ms,
”
I
E
EE
T
ra
n
s.
I
n
f
.
Fo
re
n
si
c
s
S
e
c
u
r.
,
v
o
l
.
1
2
,
n
o
.
1
2
,
p
p
.
3
0
1
1
–
3
0
2
3
,
2
0
1
7
,
d
o
i
:
1
0
.
1
1
0
9
/
TI
F
S
.
2
0
1
7
.
2
7
3
0
5
8
1
.
[
1
9
]
S
.
D
.
A
n
t
o
n
,
S
.
K
a
n
o
o
r
,
D
.
F
r
a
u
n
h
o
l
z
,
a
n
d
H
.
D
.
S
c
h
o
t
t
e
n
,
“
Ev
a
l
u
a
t
i
o
n
o
f
mac
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
a
l
g
o
r
i
t
h
ms
o
n
a
n
i
n
d
u
st
r
i
a
l
mo
d
b
u
s
/
T
C
P
d
a
t
a
s
e
t
,
”
A
C
M
I
n
t
.
C
o
n
f
.
Pr
o
c
e
e
d
i
n
g
S
e
r.
,
2
0
1
8
,
d
o
i
:
1
0
.
1
1
4
5
/
3
2
3
0
8
3
3
.
3
2
3
2
8
1
8
.
[
2
0
]
H
.
Y
a
n
g
,
L
.
C
h
e
n
g
,
a
n
d
M
.
C
.
C
h
u
a
h
,
“
D
e
e
p
-
Le
a
r
n
i
n
g
-
B
a
se
d
N
e
t
w
o
r
k
I
n
t
r
u
si
o
n
D
e
t
e
c
t
i
o
n
f
o
r
S
C
A
D
A
S
y
st
e
ms,”
2
0
1
9
I
EE
E
C
o
n
f
.
C
o
m
m
u
n
.
N
e
t
w
.
S
e
c
u
r.
C
N
S
2
0
1
9
,
2
0
1
9
,
d
o
i
:
1
0
.
1
1
0
9
/
C
N
S
.
2
0
1
9
.
8
8
0
2
7
8
5
.
Evaluation Warning : The document was created with Spire.PDF for Python.