I
AE
S In
t
er
na
t
io
na
l J
o
urna
l o
f
Art
if
icia
l In
t
ellig
ence
(
I
J
-
AI
)
Vo
l.
14
,
No
.
3
,
J
u
n
e
2
0
2
5
,
p
p
.
2479
~
2
4
8
9
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
14
.i
3
.
p
p
2
4
7
9
-
2
4
8
9
2479
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
Em
po
wering
SD
N wit
h DDoS
atta
ck det
ection:
leve
ra
g
ing
h
y
brid ma
chine
l
ea
rning
bas
ed ID
PS
co
nt
ro
ller f
o
r
ro
bust
security
F
lo
ra
nce
G
.
1
,
R.
J
.
Ana
nd
hi
2
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
n
g
i
n
e
e
r
i
n
g
,
N
e
w
H
o
r
i
z
o
n
C
o
l
l
e
g
e
o
f
E
n
g
i
n
e
e
r
i
n
g
,
A
f
f
i
l
i
a
t
e
d
t
o
V
i
sv
e
sv
a
r
a
y
a
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
2
D
e
p
a
r
t
me
n
t
o
f
I
n
f
o
r
mat
i
o
n
S
c
i
e
n
c
e
a
n
d
E
n
g
i
n
e
e
r
i
n
g
,
N
e
w
H
o
r
i
z
o
n
C
o
l
l
e
g
e
o
f
En
g
i
n
e
e
r
i
n
g
,
A
f
f
i
l
i
a
t
e
d
t
o
V
i
s
v
e
s
v
a
r
a
y
a
Te
c
h
n
o
l
o
g
i
c
a
l
U
n
i
v
e
r
si
t
y
,
B
a
n
g
a
l
o
r
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Sep
3
,
2
0
2
4
R
ev
is
ed
Feb
2
6
,
2
0
2
5
Acc
ep
ted
Ma
r
1
5
,
2
0
2
5
S
o
ftwa
re
-
d
e
fin
e
d
n
e
tw
o
rk
(S
DN
)
is
a
n
in
n
o
v
a
ti
v
e
n
e
two
r
k
in
g
f
ra
m
e
wo
rk
wh
e
re
a
c
e
n
tralize
d
c
o
n
tro
ll
e
r
m
a
n
a
g
e
s
n
e
two
rk
i
n
g
a
d
m
in
istrati
o
n
a
n
d
s
o
rts
o
u
t
n
e
two
r
k
traffic
issu
e
s.
It
b
e
c
o
m
e
s
d
iffi
c
u
l
t
fo
r
t
h
e
c
o
n
tro
ll
e
r
t
o
i
d
e
n
ti
f
y
th
e
m
a
li
c
io
u
s
u
se
r
wh
o
is
se
n
d
i
n
g
a
larg
e
n
u
m
b
e
r
o
f
sp
o
o
fe
d
p
a
c
k
e
ts,
su
c
h
a
s
in
a
d
istri
b
u
ted
d
e
n
ial
o
f
se
rv
ice
(DD
o
S
)
a
tt
a
c
k
.
To
p
re
v
e
n
t
DD
o
S
a
t
tac
k
s
fro
m
d
a
m
a
g
i
n
g
leg
it
ima
te u
se
rs,
i
t
is
im
p
o
rtan
t
t
o
tak
e
ste
p
s
to
p
re
v
e
n
t
t
h
e
m
.
Th
e
issu
e
o
f
p
re
v
e
n
ti
n
g
DD
o
S
a
tt
a
c
k
s
in
S
DN
re
m
a
in
s
u
n
re
so
l
v
e
d
d
e
sp
it
e
m
a
n
y
a
lg
o
rit
h
m
s
p
r
o
p
o
se
d
.
M
e
th
o
d
s
p
re
se
n
ted
i
n
th
is
p
a
p
e
r
e
m
p
lo
y
b
a
n
d
wi
d
th
t
h
re
sh
o
l
d
e
stim
a
ti
o
n
,
wh
ich
tr
ig
g
e
rs
th
e
in
tru
si
o
n
d
e
te
c
ti
o
n
a
n
d
p
re
v
e
n
ti
o
n
sy
ste
m
(IDP
S
)
c
o
n
tr
o
ll
e
r
if
th
e
th
re
sh
o
ld
is
e
x
c
e
e
d
e
d
.
Wh
e
n
e
v
e
r
th
e
th
re
sh
o
ld
is
e
x
c
e
e
d
e
d
d
u
e
to
n
e
two
rk
c
o
n
g
e
stio
n
,
tran
sfe
rre
d
p
a
c
k
e
ts
a
re
fil
tere
d
a
t
th
e
se
rv
e
r
lev
e
l
b
y
i
d
e
n
ti
f
y
in
g
th
e
u
ti
li
z
a
ti
o
n
o
f
b
a
n
d
wid
t
h
in
Op
e
n
Da
y
li
g
h
t
(OD
L)
a
n
d
P
OX
.
K
-
n
e
a
re
st
n
e
ig
h
b
o
r
(K
-
NN
)
a
n
d
su
p
p
o
rt
v
e
c
to
r
m
a
c
h
in
e
(S
VM)
a
re
u
se
d
b
y
t
h
e
IDPS
c
o
n
tro
ll
e
r
to
d
e
tec
t
a
n
d
th
wa
r
t
DD
o
S
a
tt
a
c
k
s.
Us
in
g
M
in
in
e
t,
tw
o
S
DN
c
e
n
tralize
d
c
o
n
tro
ll
e
rs are
sim
u
late
d
to
imp
r
o
v
e
p
e
rfo
rm
a
n
c
e
sig
n
ifi
c
a
n
tl
y
.
Ba
se
d
o
n
S
VM
in
th
e
OD
L
c
o
n
tro
ll
e
r,
th
is
wo
rk
h
a
s
p
ro
v
id
e
d
m
it
i
g
a
ti
o
n
tec
h
n
i
q
u
e
s
fo
r
p
re
v
e
n
ti
n
g
DD
o
S
a
tt
a
c
k
s
with
a
n
a
c
c
u
ra
c
y
o
f
9
6
.
7
5
%
c
o
m
p
a
re
d
to
p
re
v
i
o
u
sl
y
p
u
b
li
sh
e
d
a
c
c
u
ra
c
y
.
K
ey
w
o
r
d
s
:
DDo
S
attac
k
I
DPS
co
n
tr
o
ller
Op
en
Day
lig
h
t c
o
n
tr
o
ller
POX
co
n
tr
o
ller
So
f
twar
e
d
ef
in
ed
n
etwo
r
k
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
:
Flo
r
an
ce
G.
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
Scie
n
ce
an
d
E
n
g
in
ee
r
in
g
,
New
Ho
r
izo
n
C
o
lleg
e
o
f
E
n
g
i
n
ee
r
in
g
Af
f
iliated
to
Vis
v
esv
ar
ay
a
T
e
ch
n
o
lo
g
ical
Un
iv
er
s
ity
B
an
g
alo
r
e,
I
n
d
ia
E
m
ail: v
ijiflo
r
an
ce
5
9
@
g
m
ail.
co
m
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
en
o
r
m
o
u
s
ch
an
g
e
in
n
etwo
r
k
m
ig
r
atio
n
s
an
d
th
e
i
n
cr
ea
s
e
in
m
an
y
d
is
tr
ib
u
ted
a
p
p
l
icatio
n
s
o
n
th
e
in
ter
n
et
h
as
b
ec
o
m
e
m
o
r
e
co
m
p
lex
i
n
co
n
f
ig
u
r
atio
n
an
d
m
an
a
g
em
en
t.
T
h
e
co
n
v
en
tio
n
al
n
etwo
r
k
in
f
r
astru
ctu
r
e
an
d
o
t
h
er
r
elate
d
ev
o
lv
in
g
tech
n
o
lo
g
ies
m
an
ag
e
th
e
en
o
r
m
o
u
s
am
o
u
n
t
o
f
d
ata.
I
t
is
d
if
f
icu
lt
wh
en
th
e
u
s
er
n
ee
d
s
to
ch
a
n
g
e
th
e
r
e
q
u
ir
em
e
n
ts
an
d
d
y
n
am
ic
co
n
f
ig
u
r
atio
n
o
f
th
e
n
etwo
r
k
in
f
r
astru
ctu
r
e.
W
h
en
it
is
n
ec
es
s
ar
y
to
im
p
lem
en
t
a
n
ew
p
r
o
to
c
o
l,
u
p
d
ati
n
g
r
u
les
an
d
p
o
licies
f
o
r
all
n
etwo
r
k
in
g
d
ev
ices
b
ec
o
m
es
co
s
tly
an
d
tim
e
-
co
n
s
u
m
in
g
.
T
o
a
d
d
r
ess
th
ese,
s
o
f
twar
e
s
o
f
twar
e
-
d
ef
in
ed
n
etwo
r
k
(
SDN)
is
p
r
ef
er
ab
le.
SDN
p
r
o
v
i
d
es
s
im
p
lifie
d
m
an
a
g
em
en
t,
a
ce
n
t
r
al
co
n
tr
o
ller
,
p
r
o
g
r
am
m
ab
ili
ty
,
au
to
m
ated
,
an
d
d
y
n
am
ic
m
a
n
ag
em
e
n
t
[
1
]
.
SDN
is
a
m
o
r
e
p
o
p
u
lar
in
n
o
v
ativ
e
n
etwo
r
k
p
ar
a
d
ig
m
t
h
at
p
r
o
v
id
es
ef
f
icien
t
m
an
a
g
em
en
t
b
y
u
n
co
u
p
lin
g
th
e
d
ata
an
d
co
n
tr
o
l
p
lan
es
wh
er
e
th
e
ce
n
tr
al
ized
co
n
tr
o
ller
is
p
lace
d
[
2
]
.
T
h
e
d
ata
tr
an
s
f
er
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
3
,
J
u
n
e
2
0
2
5
:
2
4
7
9
-
2
4
8
9
2480
b
etwe
en
th
e
f
o
r
war
d
i
n
g
d
ev
ices
an
d
co
n
tr
o
ller
b
y
a
s
tan
d
ar
d
p
r
o
to
co
l
ca
lled
Op
e
n
Flo
w
(
OF)
,
wh
ich
s
eg
r
eg
ates
th
e
p
ac
k
ets
b
ased
o
n
OF
i
n
s
tr
u
ctio
n
s
.
I
n
SDN,
if
a
n
y
u
p
d
atin
g
o
f
p
o
licies
i
s
n
ee
d
ed
,
th
en
th
e
m
o
d
if
icatio
n
h
a
p
p
en
s
in
th
e
c
o
n
tr
o
ller
,
lik
e
m
i
n
im
izin
g
th
e
tim
e
co
n
s
u
m
p
tio
n
a
n
d
co
s
t
o
f
th
e
en
tire
p
r
o
ce
s
s
.
Du
r
in
g
d
ata
tr
an
s
f
er
,
SDN
s
w
itch
es
s
to
r
e
th
e
in
co
m
in
g
p
ac
k
ets
in
th
e
f
lo
w
tab
le
(
FT)
[
3
]
.
E
ac
h
en
tr
y
in
th
e
FT
co
n
s
is
ts
o
f
th
r
ee
f
ield
s
:
ac
tio
n
,
r
u
le,
an
d
co
u
n
ter
.
B
ased
o
n
FT
in
f
o
r
m
atio
n
,
p
ac
k
ets
ar
e
g
ettin
g
d
r
o
p
p
e
d
[
4
]
,
[
5
]
.
T
h
e
d
ata
s
ec
u
r
ity
p
r
o
v
id
ed
b
y
th
e
SDN
n
etwo
r
k
'
s
co
n
tr
o
ller
m
o
n
ito
r
s
m
alicio
u
s
tr
af
f
ic.
SDN
en
h
an
ce
s
n
etwo
r
k
i
n
f
r
astru
ctu
r
e
m
an
ag
em
e
n
t
an
d
ad
d
r
ess
es
ch
allen
g
es
in
ter
m
s
o
f
s
ec
u
r
it
y
,
p
r
o
g
r
am
m
a
b
ilit
y
an
d
lo
a
d
b
ala
n
cin
g
.
SDN
s
ec
u
r
ity
is
a
s
ig
n
if
ica
n
t
ch
allen
g
e
th
an
d
en
y
i
n
g
n
etwo
r
k
s
er
v
i
ce
d
u
e
to
m
alicio
u
s
tr
af
f
ic
[
6
]
,
[
7
]
.
Dis
tr
ib
u
ted
d
en
ial
o
f
s
er
v
ice
(
DDo
S)
atta
ck
s
ar
e
s
er
v
ice
-
d
en
y
in
g
attac
k
s
th
at
ex
h
au
s
t
th
e
b
en
ef
its
o
f
t
h
e
en
tire
n
etwo
r
k
an
d
s
atu
r
ate
th
e
b
an
d
wid
th
,
w
h
ich
s
to
p
s
s
er
v
icin
g
leg
itima
te
u
s
er
r
eq
u
ests
.
Du
e
to
m
an
y
s
p
o
o
f
ed
p
ac
k
ag
es,
th
e
en
tire
n
etwo
r
k
g
ets
f
lo
o
d
ed
,
r
esu
ltin
g
in
b
an
d
wid
th
d
ep
let
io
n
.
T
h
is
f
lo
o
d
in
g
lead
s
to
d
eg
r
a
d
in
g
th
e
s
y
s
tem
's p
er
f
o
r
m
a
n
ce
[
8
]
,
[
9
]
.
Ma
ch
in
e
lear
n
in
g
(
ML
)
is
an
ap
p
licatio
n
th
at
class
if
ies
p
ac
k
ets
u
s
in
g
in
tr
u
s
io
n
d
ete
ctio
n
an
d
p
r
ev
en
tio
n
s
y
s
tem
(
I
DPS)
co
n
tr
o
ller
s
in
SDN
[
1
0
]
,
[
1
1
]
.
T
ec
h
n
iq
u
es
s
u
ch
as
k
-
n
ea
r
est
n
eig
h
b
o
r
(K
-
NN)
,
n
aiv
e
B
ay
es,
k
-
m
ea
n
s
,
an
d
s
u
p
p
o
r
t
v
ec
to
r
m
ac
h
in
e
(
SVM)
ar
e
em
p
l
o
y
ed
to
d
etec
t
a
n
o
m
alo
u
s
p
ac
k
ets
in
I
DPS.
An
o
m
aly
-
b
ased
I
DPS
s
y
s
tem
s
u
s
e
th
e
co
n
ce
p
t
o
f
n
etwo
r
k
b
eh
av
io
r
.
ML
alg
o
r
ith
m
s
ass
is
t
s
et
o
f
tr
af
f
ics
g
en
er
ated
b
y
th
e
W
aik
ato
en
v
ir
o
n
m
en
t
f
o
r
k
n
o
wled
g
e
an
aly
s
is
(
W
E
K
A
)
to
o
l
an
d
p
r
o
v
id
e
p
r
e
d
ictio
n
s
f
o
r
tr
ai
n
in
g
d
ata
d
er
iv
e
d
f
r
o
m
tr
af
f
ic
p
atter
n
s
[
1
2
]
,
[
1
3
]
.
C
o
n
s
eq
u
en
tly
,
ML
-
b
ased
d
ete
ctio
n
tech
n
i
q
u
es
ar
e
u
s
ed
to
id
en
tif
y
s
p
o
o
f
ed
p
ac
k
ets
b
ased
o
n
ab
n
o
r
m
al
c
h
an
g
e
s
in
th
e
n
etwo
r
k
,
an
d
th
eir
p
e
r
f
o
r
m
a
n
ce
d
ep
e
n
d
s
o
n
ch
ar
ac
te
r
is
tics
co
n
s
id
er
ed
f
o
r
tr
ain
in
g
[
1
4
]
–
[
1
6
]
.
J
awa
h
a
r
et
a
l
.
[
1
7
]
p
r
o
p
o
s
e
d
M
L
m
o
d
els
t
h
a
t
u
s
e
t
h
e
E
t
h
e
r
e
u
m
b
l
o
ck
c
h
ai
n
t
o
lis
t
m
alic
io
u
s
I
P
ad
d
r
ess
es
.
J
e
b
r
il
e
t
a
l.
[
1
8
]
p
r
es
en
te
d
a
c
o
n
v
o
lu
ti
o
n
al
n
e
u
r
al
n
etw
o
r
k
-
b
i
d
i
r
e
cti
o
n
al
l
o
n
g
s
h
o
r
t
-
t
er
m
m
em
o
r
y
(
C
NN
-
B
iLS
T
M
)
m
o
d
el,
u
s
i
n
g
th
r
e
e
d
e
e
p
-
l
ea
r
n
i
n
g
a
lg
o
r
it
h
m
s
.
T
h
e
p
e
r
f
o
r
m
a
n
ce
o
f
t
h
is
m
o
d
el
is
e
v
al
u
a
te
d
b
y
p
e
r
f
o
r
m
a
n
c
e
cr
ite
r
i
a.
S
o
n
g
a
a
n
d
K
ar
r
i
[
1
9
]
in
tr
o
d
u
ce
d
,
f
e
at
u
r
es
s
el
ec
ti
o
n
is
c
o
m
b
i
n
e
d
as
a
c
lu
s
te
r
to
a
n
al
y
z
e
th
e
t
r
a
f
f
ic
.
Z
h
ao
a
n
d
Ya
n
g
[
2
0
]
d
em
o
n
s
t
r
a
te
d
t
h
e
u
n
f
o
r
t
u
n
a
te
f
l
o
o
d
i
n
g
t
h
at
s
p
o
ils
t
h
e
n
et
wo
r
k
t
h
at
a
r
e
ta
k
e
n
ca
r
e
b
y
t
h
e
a
d
a
p
t
iv
e
r
esil
i
e
n
t
c
o
n
tr
o
l
d
es
ig
n
.
Z
h
a
o
et
a
l.
[
2
1
]
f
o
c
u
s
e
d
o
n
r
esi
lie
n
t
e
v
e
n
t
-
d
r
i
v
en
d
y
n
am
i
c
f
e
ed
b
a
c
k
c
o
n
t
r
o
l
s
y
s
te
m
s
,
t
h
at
is
u
s
ed
f
o
r
t
h
e
im
p
a
ct
o
f
DD
o
S
att
ac
k
a
n
d
r
e
g
u
la
te
t
h
e
r
e
q
u
i
r
e
d
a
n
t
ici
p
ate
d
d
a
ta
tr
a
n
s
m
is
s
i
o
n
.
R
e
v
at
h
i
a
n
d
D
ev
i
[
2
2
]
f
o
c
u
s
e
d
o
n
h
y
b
r
id
i
n
t
r
u
s
io
n
d
et
ec
ti
o
n
m
et
h
o
d
u
s
in
g
th
e
n
o
v
el
d
ee
p
lea
r
n
i
n
g
m
et
h
o
d
m
o
d
i
f
i
e
d
h
y
b
r
i
d
d
e
e
p
b
e
lie
f
n
etw
o
r
k
w
it
h
wei
g
h
ts
(
M
HDBN‑W
)
.
R
a
ja
n
an
d
Ar
a
v
i
n
d
h
ar
[
2
3
]
p
r
o
p
o
s
ed
C
NN
-
L
STM
tech
n
i
q
u
es
s
h
o
win
g
th
eir
s
u
p
er
io
r
i
ty
in
s
p
o
ttin
g
m
alicio
u
s
.
Ma
n
s
o
o
r
et
a
l.
[
2
4
]
p
r
o
p
o
s
ed
an
a
p
p
r
o
ac
h
th
at
u
s
es,
d
ata
p
r
ep
r
o
ce
s
s
in
g
b
y
tr
a
n
s
f
o
r
m
atio
n
,
f
ea
tu
r
e
s
elec
tio
n
b
y
ch
i
-
s
q
u
ar
e,
a
n
d
d
etec
tin
g
attac
k
s
u
s
in
g
r
ec
u
r
r
e
n
t n
eu
r
al
n
etwo
r
k
s
(
R
NNs)
m
o
d
el.
Yao
et
a
l.
[
2
5
]
f
o
cu
s
ed
o
n
d
ec
ep
tio
n
attac
k
s
o
n
s
witch
ed
T
ak
a
g
i
-
Su
g
en
o
(
T
-
S)
f
u
zz
y
s
y
s
tem
s
.
Z
h
o
u
et
a
l.
[
2
6
]
p
r
o
p
o
s
ed
a
f
r
am
ewo
r
k
to
h
a
n
d
le
m
u
lti
-
ty
p
e
DDo
S
attac
k
s
u
s
in
g
a
f
ea
t
u
r
e
-
b
ased
m
eth
o
d
.
Kr
is
h
n
an
et
a
l.
[
2
7
]
f
o
cu
s
ed
o
n
a
m
o
r
e
p
r
e
d
ictab
le,
s
ec
u
r
ed
,
an
d
r
eliab
le
n
etw
o
r
k
u
s
in
g
t
h
e
SDN,
n
etwo
r
k
f
u
n
ctio
n
v
ir
tu
aliza
tio
n
,
a
n
d
m
ac
h
in
e
lear
n
i
n
g
.
Mu
s
u
m
ec
i
et
a
l.
[
2
8
]
d
esig
n
ed
f
o
r
DDo
S
attac
k
d
etec
tio
n
f
r
am
ewo
r
k
with
ML
ass
is
tan
ce
b
y
au
to
m
atica
lly
r
etr
iev
in
g
p
ac
k
et
in
f
o
r
m
atio
n
.
Swati
et
a
l.
[
2
9
]
d
escr
ib
ed
th
e
ca
p
ab
ilit
y
-
b
ased
ar
ch
itectu
r
e
o
f
a
tr
an
s
ien
t e
f
f
e
ct
r
in
g
o
s
cillato
r
PUF.
Ye
et
a
l
.
[
3
0
]
p
r
o
p
o
s
e
d
a
n
SV
M
m
o
d
el
t
o
d
ete
ct
D
D
o
S
a
tta
ck
s
b
y
d
et
er
m
i
n
i
n
g
f
lo
w
s
tat
u
s
in
f
o
r
m
at
io
n
p
er
io
d
i
ca
l
ly
a
n
d
it
e
x
t
r
a
cte
d
f
r
o
m
t
h
e
s
w
itc
h
FT.
S
u
m
at
h
i
et
a
l
.
[
1
1
]
d
is
cu
s
s
ed
h
y
b
r
i
d
M
L
tec
h
n
i
q
u
es
u
s
e
d
to
d
esi
g
n
t
h
e
in
tr
u
s
i
o
n
d
et
ec
ti
o
n
s
y
s
te
m
(
I
DS
)
.
B
h
ar
o
t
et
a
l
.
[
3
1
]
d
es
cr
i
b
e
th
e
He
lli
n
g
e
r
d
is
t
an
ce
f
u
n
ct
io
n
t
h
at
is
u
s
ed
to
co
m
p
a
r
e
th
e
i
n
c
o
m
i
n
g
a
n
d
b
ase
li
n
e
tr
af
f
i
c
wi
th
p
r
o
b
a
b
ilit
y
.
DD
o
S
a
tta
ck
d
et
ec
ti
o
n
is
b
ase
d
o
n
f
in
d
i
n
g
e
n
t
r
o
p
y
a
n
d
s
t
an
d
ar
d
d
e
v
i
ati
o
n
f
o
r
m
o
n
it
o
r
i
n
g
i
n
c
o
m
i
n
g
p
ac
k
ets
wi
th
tw
o
le
v
els
o
f
t
h
e
t
h
r
es
h
o
l
d
v
a
lu
e
in
[
3
2
]
.
T
h
e
m
ain
c
o
n
tr
ib
u
tio
n
o
f
th
is
p
ap
er
is
to
a
d
d
r
ess
th
e
ch
all
en
g
e
o
f
d
etec
tin
g
a
n
d
p
r
ev
e
n
tin
g
DDo
S
attac
k
s
with
in
SDN.
I
n
SDN,
a
ce
n
tr
alize
d
c
o
n
tr
o
ller
m
a
n
ag
es
n
etwo
r
k
ad
m
in
is
tr
atio
n
an
d
tr
af
f
ic,
wh
ich
m
ak
es
it
d
if
f
icu
lt
to
id
en
tify
m
alicio
u
s
u
s
er
s
,
p
ar
ticu
lar
ly
in
th
e
ca
s
e
o
f
DDo
S
attac
k
s
in
v
o
lv
in
g
lar
g
e
v
o
lu
m
es
o
f
s
p
o
o
f
ed
p
ac
k
ets.
T
o
m
itig
ate
th
is
is
s
u
e,
th
e
p
ap
er
p
r
o
p
o
s
es
a
s
o
lu
tio
n
th
at
u
s
es
b
an
d
wid
th
th
r
esh
o
ld
esti
m
atio
n
in
co
n
ju
n
ctio
n
with
an
I
DPS
co
n
tr
o
ller
.
T
h
e
I
DPS
co
n
tr
o
ller
,
in
t
u
r
n
e
x
tr
ac
ts
f
ea
tu
r
e
an
d
class
if
ies p
ac
k
ets as n
o
r
m
al
o
r
m
alicio
u
s
u
s
in
g
SVM
an
d
K
-
NN
class
if
icat
io
n
alg
o
r
ith
m
s
.
Sectio
n
2
illu
s
tr
ates
th
e
p
r
o
p
o
s
ed
s
k
eleto
n
f
o
r
m
itig
atin
g
D
Do
S
attac
k
d
etec
tio
n
s
tr
ateg
ie
s
.
Sectio
n
3
p
r
esen
ts
th
e
s
im
u
latio
n
an
d
p
er
f
o
r
m
an
ce
ass
ess
m
en
t
o
f
t
h
e
DDo
S
attac
k
d
etec
tio
n
f
r
am
ewo
r
k
.
Sectio
n
4
co
n
clu
d
es th
e
p
ap
er
a
n
d
d
is
cu
s
s
es th
e
r
elate
d
f
u
tu
r
e
en
h
an
ce
m
en
t o
f
th
e
r
esear
ch
s
co
p
e
.
2.
RE
S
E
ARCH
M
E
T
H
O
D
2
.
1
.
Det
er
m
ina
t
io
n o
f
ba
nd
width
t
hresh
o
l
d
T
h
e
b
an
d
wid
th
th
r
esh
o
ld
esti
m
atio
n
f
ilter
s
th
e
tr
af
f
ic
wh
en
th
e
tr
af
f
ic
o
cc
u
p
ies
th
e
b
an
d
wid
th
th
at
ex
ce
ed
s
a
ce
r
tain
lim
it.
T
h
is
esti
m
atio
n
u
s
es
th
e
to
tal
b
an
d
wid
th
C
,
th
e
b
an
d
wid
th
r
atio
is
B
0
,
th
e
am
o
u
n
t
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
mp
o
w
erin
g
S
DN
w
ith
DDo
S
a
tta
ck
d
etec
tio
n
:
leve
r
a
g
in
g
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
…
(
F
lo
r
a
n
ce
G
.
)
2481
tr
af
f
ic
f
o
r
leg
itima
te
an
d
m
ali
cio
u
s
p
ac
k
ets,
an
d
ev
alu
ates
th
e
p
er
f
o
r
m
an
ce
in
ter
m
s
o
f
c
o
n
g
esti
o
n
wh
en
th
er
e
is
a
f
lu
ctu
atio
n
o
f
th
e
cu
r
r
en
t
b
an
d
wid
th
r
atio
B
0
.
Firstl
y
,
th
e
am
o
u
n
t
o
f
tr
af
f
ic
ca
lcu
lated
u
s
in
g
(
1
)
.
(
,
)
=
∑
0
=
1
(
1
)
T
h
e
av
er
a
g
e
lin
k
d
elay
is
d
en
o
ted
in
(
2
)
,
=
=
(
Pkt
r
ect
i
m
e
−
p
kt
s
en
t
t
i
m
e
)
t
r
an
s
m
i
s
s
i
o
n
_
s
p
eed
(
2
)
w
h
er
e,
Pk
tr
ec
tim
e
is
p
ac
k
et
r
ec
eiv
e
tim
e
an
d
Pk
ts
en
ttime
is
p
ac
k
et
s
en
t
tim
e.
T
h
e
p
r
o
b
ab
ilit
y
o
f
d
ela
y
is
d
en
o
ted
as,
y
i.
=
(
>
)
T
h
e
m
ain
g
o
al
is
to
d
ef
in
e
th
e
b
an
d
wid
th
th
r
esh
o
ld
,
wh
ich
is
7
5
%
o
f
th
e
as
s
ig
n
ed
b
an
d
wid
th
.
B
y
f
in
e
-
tu
n
in
g
C
an
d
B
0
,
p
e
r
f
o
r
m
an
ce
is
in
cr
ea
s
ed
.
No
w
th
e
r
e
ar
e
two
ca
s
es
as,
L
is
th
e
s
et
o
f
co
m
p
letely
s
u
cc
ess
f
u
l
r
eq
u
ests
(
leg
itima
t
e
p
ac
k
ets)
an
d
M
is
th
e
s
et
o
f
r
eq
u
ests
th
at
ar
e
n
o
t
co
m
p
letely
s
u
cc
ess
f
u
l
(
m
alicio
u
s
p
ac
k
ets).
C
o
n
s
id
er
,
th
e
n
etwo
r
k
wh
ich
is
n
o
t
co
n
g
ested
with
a
s
et
o
f
r
eq
u
ests
,
an
d
ch
an
g
in
g
cu
r
r
en
t
b
an
d
wid
th
r
atio
in
(
3
)
an
d
(
4
)
.
∆
=
+
(
∑
∈
0
)
(
3
)
T
h
e
n
etwo
r
k
wh
ich
is
co
n
g
ested
with
r
eq
u
ests
an
d
c
h
an
g
i
n
g
cu
r
r
en
t
b
an
d
wid
t
h
r
atio
.
∆
=
−
(
∑
∈
0
)
(
4
)
T
h
e
n
u
m
b
er
o
f
p
ac
k
ets r
ec
eiv
ed
p
er
s
o
u
r
ce
I
P a
d
d
r
ess
is
d
ef
in
ed
u
s
in
g
e
n
tr
o
p
y
ca
lcu
latio
n
in
(
5
)
.
(
)
=
∑
(
)
l
og
2
(
)
=
1
(
5
)
(
)
=
(
1
)
+
(
2
)
+
⋯
+
(
)
T
h
e
s
u
p
p
o
r
tin
g
v
alu
e
o
f
b
an
d
wid
th
f
o
r
ea
c
h
s
o
u
r
ce
I
P a
d
d
r
e
s
s
is
f
o
u
n
d
in
(
6
)
.
(
_
)
=
(
_
)
∆
+
∆
(
6
)
T
h
e
u
t
i
l
it
y
f
u
n
c
t
i
o
n
i
n
t
h
e
b
a
n
d
w
i
d
t
h
t
h
r
e
s
h
o
l
d
v
al
u
e
o
f
7
5
%
f
o
r
b
a
n
d
w
i
d
t
h
o
f
e
a
c
h
s
o
u
r
c
e
I
P
i
s
d
e
f
i
n
e
d
i
n
(
7
)
.
(
_
)
=
(
_
)
∗
∑
=
1
(
7
)
(
)
=
∆
,
0
.
01
<
=
(
_
)
>
=
0
.
75
(
_
)
=
∆
,
(
_
)
>
=
0
.
76
T
h
e
av
er
a
g
e
u
tili
ty
o
f
ea
ch
s
o
u
r
ce
I
P a
d
d
r
ess
[
1
6
]
is
d
e
f
in
ed
in
(
8
)
.
=
(
_
)
|
|
(
8
)
T
h
e
m
in
im
u
m
b
an
d
wid
th
th
r
e
s
h
o
ld
is
f
o
u
n
d
in
(
9
)
.
=
=
(
_
)
|
|
(
9
)
I
f
th
e
b
a
n
d
wid
th
r
atio
o
f
m
ali
cio
u
s
r
eq
u
ests
in
cr
ea
s
es,
it
wi
ll
co
n
s
u
m
e
m
o
r
e
b
a
n
d
wid
th
o
f
p
ac
k
ets
r
ec
eiv
ed
f
r
o
m
ea
ch
s
o
u
r
ce
I
P
a
d
d
r
ess
;
h
en
ce
,
th
e
r
estrictio
n
o
f
m
alic
io
u
s
p
ac
k
ets
is
d
o
n
e
u
s
in
g
th
r
esh
o
ld
b
an
d
wid
th
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
3
,
J
u
n
e
2
0
2
5
:
2
4
7
9
-
2
4
8
9
2482
T
h
e
n
o
tatio
n
u
s
ed
in
t
h
e
p
r
o
p
o
s
ed
b
an
d
wid
th
th
r
esh
o
ld
es
tim
atio
n
is
s
h
o
wn
in
T
ab
le
1
.
Alg
o
r
ith
m
s
1
to
4
d
ep
ict
th
e
p
s
eu
d
o
co
d
e
o
f
t
h
e
p
r
o
p
o
s
ed
wo
r
k
f
lo
w.
T
ab
le
1
.
No
tatio
n
s
u
s
ed
in
th
e
p
r
o
p
o
s
ed
d
er
iv
atio
n
o
f
b
an
d
wid
th
th
r
esh
o
l
d
esti
m
atio
n
N
o
t
a
t
i
o
n
D
e
scri
p
t
i
o
n
(
,
)
A
mo
u
n
t
o
f
t
r
a
f
f
i
c
N
N
u
mb
e
r
o
f
r
e
q
u
e
st
s
BR
i
B
a
n
d
w
i
d
t
h
r
e
s
o
u
r
c
e
o
f
i
th
r
e
q
u
e
s
t
B
0
C
u
r
r
e
n
t
b
a
n
d
w
i
d
t
h
r
a
t
i
o
A
v
e
r
a
g
e
d
e
l
a
y
D
Li
n
k
d
i
st
a
n
c
e
P
P
r
o
c
e
ss
i
n
g
d
e
l
a
y
y
i
P
r
o
b
a
b
i
l
i
t
y
o
f
d
e
l
a
y
T
i
D
e
l
a
y
t
h
r
e
sh
o
l
d
C
To
t
a
l
b
a
n
d
w
i
d
t
h
∆B
L
B
a
n
d
w
i
d
t
h
r
a
t
i
o
f
o
r
l
e
g
i
t
i
mat
e
r
e
q
u
e
s
t
B
i
B
a
n
d
w
i
d
t
h
o
f
i
th
r
e
q
u
e
st
∆B
M
B
a
n
d
w
i
d
t
h
r
a
t
i
o
f
o
r
ma
l
i
c
i
o
u
s
r
e
q
u
e
st
(
_
)
En
t
r
o
p
y
v
a
l
u
e
o
f
b
a
n
d
w
i
d
t
h
o
f
p
a
c
k
e
t
s
(
_
)
P
r
o
b
a
b
i
l
i
t
y
o
f
b
a
n
d
w
i
d
t
h
u
t
i
l
i
z
e
d
b
y
t
h
e
I
P
a
d
d
r
e
s
s
(
_
)
U
t
i
l
i
t
y
f
u
n
c
t
i
o
n
d
e
f
i
n
e
d
i
n
b
a
n
d
w
i
d
t
h
t
h
r
e
s
h
o
l
d
A
v
e
r
a
g
e
u
t
i
l
i
t
y
o
f
e
a
c
h
so
u
r
c
e
I
P
a
d
d
r
e
ss
_
M
i
n
i
m
u
m
b
a
n
d
w
i
d
t
h
t
h
r
e
s
h
o
l
d
Alg
o
r
ith
m
1
.
B
an
d
wid
th
th
r
es
h
o
ld
esti
m
atio
n
Est_band_thres ()
1: Initially set
(
,
)
=
∑
0
=
1
2: Calculate
∆
,
∆
,
(
)
,
(
_
)
using (3), (4), (5) and (6)
3: Find
(
_
)
=
(
_
)
∗
∑
=
1
4: If (
0
.
01
≤
_
(
_
)
≥
0
.
75
) then
(
_
)
=
∆
5: Else if (100
≥
(
_
)
≥
0
.
76
) then
(
_
)
=
∆
6: Determine
=
(
_
)
|
|
7: Return
=
=
(
_
)
|
|
8: Monitoring the server that exceeded the bandwidth threshold i.e.,
If (100
≥
(
_
)
≥
0
.
76
) then invoke Algorithm 2
9: Exit
Alg
o
r
ith
m
2
.
I
DPS co
n
tr
o
ller
IDPS_controller
()
1:
Generated
traffic
using
Tor’s
hammer
tool
and
it’s
captured
by
the
Wireshark
tool
is
depicted in Table 2.
2:
Se
le
ct
th
e
fe
at
ur
es
b
y
ca
lc
ul
at
in
g
th
e
va
ri
ab
le
s
,
,
,
,
sh
ow
n
in
Table 3
3: Invoke the Algorithm 3
4: Exit
Alg
o
r
ith
m
3
.
C
lass
if
icatio
n
al
g
o
r
ith
m
s
Classify Algo ()
1: SVM ()
Construct
the
hyperplane
by
1
:
.
+
=
+
1
and
2
:
.
+
=
−
1
th
en
re
ad
th
e
da
ta
se
t
mentioned in Table 3. Invoke Perform_metrics ();
2: K
-
NN ()
-
Read the dataset and define K value.
-
Obtained Euclidian distance
ED=
√
(
2
−
1
)
2
+
(
2
−
1
)
2
from the
dataset.
-
Decide the most frequently matching with the least K value.
3: Invoke Algorithm 4
4: Exit
Alg
o
r
ith
m
4
.
Per
f
o
r
m
an
ce
m
et
r
ics
Perform_metrics ()
1: Calculate Accuracy, FAR, Recall, Precision, and F
-
measure.
2: Exit.
2
.
2
.
P
a
ck
e
t
f
iltr
a
t
io
n us
ing
hy
brid
m
a
chine le
a
rning
m
o
del in serv
er
T
h
e
p
r
o
p
o
s
ed
I
DPS
co
n
tr
o
ller
is
an
ap
p
licatio
n
t
h
at
o
r
ig
i
n
ated
to
in
v
esti
g
ate
m
alicio
u
s
b
e
h
av
io
r
s
.
I
t
is
in
s
talled
in
a
s
er
v
er
th
at
d
es
cr
ib
es th
e
f
u
n
ctio
n
s
to
d
etec
t
DDo
S a
ttack
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
mp
o
w
erin
g
S
DN
w
ith
DDo
S
a
tta
ck
d
etec
tio
n
:
leve
r
a
g
in
g
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
…
(
F
lo
r
a
n
ce
G
.
)
2483
2
.
2
.
1
.
T
ra
f
f
ic
o
rig
ina
t
io
n a
nd
pa
ck
et
ca
pturing
I
n
th
e
SDN
en
v
ir
o
n
m
en
t,
t
h
e
f
o
r
em
o
s
t
s
tep
to
d
etec
t
a
D
Do
S
attac
k
is
to
g
en
er
ate
n
et
wo
r
k
tr
af
f
ic
u
s
in
g
T
o
r
’
s
h
am
m
e
r
n
etwo
r
k
to
o
l
as
s
h
o
wn
in
Fig
u
r
e
1
.
I
t
ca
n
ca
u
s
e
h
u
g
e
tr
af
f
ic
i
n
n
et
wo
r
k
f
l
o
w,
m
ak
i
n
g
s
er
v
ice
u
n
av
ailab
le
to
th
e
leg
itima
te
u
s
er
.
On
e
o
f
th
e
m
o
s
t
p
o
p
u
lar
t
o
o
ls
u
s
ed
to
ca
p
tu
r
e
th
e
tr
af
f
ic
in
th
e
n
etwo
r
k
is
th
e
W
ir
esh
ar
k
to
o
l.
T
h
is
to
o
l
m
o
n
ito
r
s
an
d
ca
p
t
u
r
es
n
etwo
r
k
tr
af
f
ic
in
th
e
FT
v
i
a
th
e
s
er
v
er
.
W
h
en
th
e
esti
m
ated
b
an
d
wid
t
h
th
r
es
h
o
ld
r
ea
c
h
es
a
s
p
ec
if
ic
p
e
r
ce
n
tag
e
o
f
b
an
d
wid
th
,
it
s
tar
ts
m
ain
tain
in
g
th
e
t
r
af
f
ic
lo
g
s
f
o
r
f
u
tu
r
e
an
al
y
s
is
.
T
h
is
lo
g
ex
tr
ac
ts
th
e
s
o
u
r
ce
a
n
d
d
e
s
tin
atio
n
I
P
ad
d
r
ess
,
s
o
u
r
ce
a
n
d
d
esti
n
atio
n
p
o
r
t,
p
r
o
to
co
l,
tim
e,
T
T
L
v
alu
e,
a
n
d
in
f
o
r
m
atio
n
c
o
llected
f
r
o
m
p
ac
k
et
h
ea
d
e
r
s
lik
e
SYN,
FIN
,
AC
K,
Seq
,
an
d
L
en
.
Fig
u
r
e
1
.
Pro
p
o
s
ed
I
DPS co
n
t
r
o
ller
in
SDN
T
ab
le
2
.
Gen
e
r
ated
d
ataset
serv
e
r
sr
c
d
st
p
kt
c
o
u
n
t
b
y
t
e
c
o
u
n
t
d
u
r
a
t
i
o
n
d
u
r
_
n
sec
t
o
t
_
d
u
r
f
l
o
w
s
p
k
t
r
a
t
e
p
r
o
t
o
c
o
l
p
o
r
t
_
n
o
t
x
_
b
y
t
e
s
r
x
_
b
y
t
e
s
1
1
0
.
0
.
1
.
1
1
0
.
0
.
0
.
1
5
5
4
7
0
2
5
6
4
7
4
1
7
1
1
0
0
6
4
3
2
0
0
0
0
0
1
.
0
7
E+
1
1
2
2
7
8
OF
1
3
6
1
9
9
4
1
2
1
0
.
0
.
7
.
1
1
0
.
0
.
4
.
1
3
1
1
7
2
3
4
1
3
9
1
3
6
0
7
8
2
5
0
7
3
9
0
0
0
0
0
0
3
.
0
1
E+
1
1
4
5
9
1
OF
3
4
8
7
1
4
2
5
7
1
1
0
.
0
.
5
.
2
1
0
.
0
.
3
.
2
5
8
5
7
3
4
8
9
1
9
4
9
9
1
1
0
0
6
8
9
0
0
0
0
0
0
1
.
7
1
E+
1
1
5
3
4
2
OF
5
3
6
1
4
1
5
2
1
3
1
0
.
0
.
9
.
2
1
0
.
0
.
0
.
1
5
9
2
3
4
1
9
5
4
1
3
1
2
9
4
1
8
0
7
1
2
0
0
0
0
0
0
1
.
9
9
E+
1
1
2
6
9
5
OF
3
4
0
1
5
1
0
2
7
T
ab
le
3
.
Six
-
tu
p
le
ch
ar
ac
ter
is
t
ic
v
alu
es with
its
d
escr
ip
tio
n
P
a
r
a
me
t
e
r
F
o
r
mu
l
a
D
e
scri
p
t
i
o
n
N
u
mb
e
r
o
f
so
u
r
c
e
I
P
a
d
d
r
e
sse
s
(
N
S
I
P
A
)
=
_
Th
e
t
o
t
a
l
n
u
mb
e
r
o
f
so
u
r
c
e
I
P
a
d
d
r
e
sses
p
e
r
sam
p
l
i
n
g
i
n
t
e
r
v
a
l
i
s
d
e
n
o
t
e
d
i
n
t
i
me
.
N
u
mb
e
r
o
f
so
u
r
c
e
p
o
r
t
(
N
S
P
)
=
_
Th
e
t
o
t
a
l
n
u
m
b
e
r
o
f
p
o
r
t
s
s
e
n
d
i
n
g
p
a
c
k
e
t
s
p
e
r
sam
p
l
i
n
g
i
n
t
e
r
v
a
l
i
s
d
e
n
o
t
e
d
i
n
t
i
me
.
S
p
e
e
d
o
f
d
a
t
a
r
a
t
e
(
S
D
R
)
=
_
Th
e
n
u
m
b
e
r
o
f
p
a
c
k
e
t
s
t
r
a
n
sf
e
r
r
e
d
p
e
r
sec
o
n
d
u
n
i
t
.
En
t
r
o
p
y
v
a
l
u
e
o
f
f
l
o
w
p
a
c
k
e
t
s
(
EFP
)
=
∑
(
)
lo
g
2
(
)
=
1
Th
e
s
i
z
e
o
f
p
a
c
k
e
t
s e
n
t
e
r
i
n
g
t
o
F
T
i
s
e
st
i
m
a
t
e
d
b
y
t
h
e
t
o
t
a
l
n
u
mb
e
r
o
f
p
a
c
k
e
t
s
i
n
a
f
l
o
w
(
N
P
F
)
p
e
r
f
l
o
w
n
u
mb
e
r
(
F
N
)
.
En
t
r
o
p
y
v
a
l
u
e
o
f
f
l
o
w
b
y
t
e
s
(
EFB)
=
∑
(
)
lo
g
2
(
)
=
1
Th
e
s
i
z
e
o
f
b
y
t
e
s
e
n
t
e
r
i
n
g
t
o
F
T
i
s
e
st
i
m
a
t
e
d
b
y
t
h
e
t
o
t
a
l
n
u
m
b
e
r
o
f
b
y
t
e
s
i
n
a
f
l
o
w
p
e
r
f
l
o
w
n
u
mb
e
r
(
F
N
)
.
F
T
ma
t
c
h
i
n
g
r
a
t
i
o
(
F
M
R
)
=
ℎ
_
Th
e
n
u
m
b
e
r
o
f
f
l
o
w
s
i
s
g
e
t
t
i
n
g
m
a
t
c
h
e
d
w
i
t
h
t
h
e
sw
i
t
c
h
F
T
e
n
t
r
y
.
2
.
2
.
2
.
Select
io
n o
f
cha
ra
c
t
er
i
s
t
ic
v
a
lues
O
n
c
e
n
e
t
w
o
r
k
t
r
a
f
f
i
c
i
s
c
a
p
t
u
r
ed
a
n
d
p
r
e
-
p
r
o
c
e
s
s
i
n
g
i
s
d
o
n
e
,
th
e
n
e
x
t
is
t
o
e
x
t
r
a
c
t
f
e
a
t
u
r
es
a
n
d
c
r
e
a
t
e
a
d
a
t
a
s
et
.
T
h
e
M
L
m
o
d
e
l
c
o
n
v
e
r
t
s
t
h
e
n
e
c
ess
a
r
y
f
i
e
l
d
s
f
r
o
m
t
h
e
p
a
c
k
e
t
h
e
a
d
e
r
i
n
t
o
a
r
e
a
d
a
b
le
C
SV
f
o
r
m
a
t
.
T
h
is
p
a
p
e
r
c
o
m
p
a
r
e
s
w
i
t
h
t
h
e
e
x
i
s
ti
n
g
a
n
a
l
y
s
i
s
[
1
2
]
o
f
c
h
a
r
a
c
t
e
r
i
s
t
i
c
v
a
l
u
es
n
e
e
d
e
d
f
o
r
c
l
a
s
s
i
f
y
i
n
g
l
e
g
i
t
i
m
a
t
e
a
n
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
3
,
J
u
n
e
2
0
2
5
:
2
4
7
9
-
2
4
8
9
2484
m
a
l
i
c
i
o
u
s
t
r
a
f
f
ic
.
F
o
r
t
h
is
p
u
r
p
o
s
e
,
s
i
x
d
is
t
i
n
ct
t
u
p
le
v
a
l
u
es
a
r
e
t
a
k
e
n
t
o
d
e
t
e
ct
DD
o
S
a
t
t
ac
k
s
.
SV
M
a
n
d
K
-
NN
u
s
e
t
h
e
f
e
a
t
u
r
es
m
e
n
ti
o
n
e
d
i
n
T
a
b
l
e
3
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
n
e
t
w
o
r
k
t
r
a
f
f
i
c
t
o
c
l
as
s
i
f
y
n
o
r
m
a
l
a
n
d
a
b
n
o
r
m
a
l
t
r
a
f
f
i
c
.
2
.
2
.
3
.
Cla
s
s
if
ica
t
io
n o
f
net
wo
rk
t
ra
f
f
ic
T
h
is
s
ec
tio
n
illu
s
tr
ates
two
cla
s
s
if
icatio
n
alg
o
r
ith
m
s
u
s
ed
to
class
if
y
leg
itima
te
an
d
m
alicio
u
s
tr
af
f
ic:
SVM
an
d
K
-
NN.
T
h
ese
two
al
g
o
r
ith
m
s
will c
lass
if
y
th
e
n
etwo
r
k
tr
af
f
ic
u
s
in
g
6
tu
p
les ch
a
r
ac
ter
is
tic
v
alu
es.
‒
Su
p
p
o
r
t
v
ec
t
o
r
m
ac
h
in
e
:
SV
M
is
th
e
m
o
s
t
p
o
p
u
lar
class
if
i
ca
tio
n
tech
n
iq
u
e.
I
t
co
llects
F
T
en
tr
ies
in
th
e
s
witch
v
ia
a
s
er
v
e
r
at
in
ter
v
als
an
d
d
eter
m
i
n
es
th
e
s
ix
tu
p
le
c
h
ar
ac
ter
is
tics
v
alu
e
to
m
ak
e
a
s
am
p
le
s
et
Z
.
T
h
e
s
am
p
le
s
et
Z
=(
P,
Q)
,
wh
e
r
e
P
d
en
o
tes
th
e
v
alu
e
o
b
tain
e
d
f
o
r
s
ix
tu
p
le
ch
ar
ac
ter
is
tics
ca
lled
f
ea
tu
r
e
v
ec
to
r
s
o
f
f
lo
w
en
tr
y
,
a
n
d
Q
d
en
o
tes
th
e
lab
el
m
a
r
k
er
.
Q=
0
d
en
o
tes
a
leg
itima
te
s
tate,
an
d
Q=
1
d
en
o
tes
a
m
alicio
u
s
s
tate.
T
h
e
SVM
t
ec
h
n
iq
u
e
u
s
es
h
y
p
er
p
la
n
e
c
o
n
s
tr
u
ctio
n
to
class
if
y
th
e
tr
af
f
ic
.
T
h
e
tr
ain
in
g
d
ata
s
et
D=
{(
x
1
,
y
1
)
,
(
x
2
,
y
2
)
,
…,
(
x
n
,
y
n
)
}
wh
er
e
Xi
r
ep
r
e
s
en
ts
th
e
in
p
u
t
v
ec
to
r
an
d
Yi
r
ep
r
esen
ts
th
e
lab
el,
s
u
ch
as
∈
(
+
1
,
−
1
)
.
T
h
e
h
y
p
er
p
la
n
e
is
co
n
s
tr
u
cted
b
y
u
s
in
g
th
e
e
q
u
atio
n
,
.
+
=
0
,
wh
er
e
∈
,
∈
an
d
two
p
lan
es
r
u
n
p
ar
allel
to
th
e
h
y
p
er
p
la
n
e
as,
1
:
.
+
=
+
1
an
d
2
:
.
+
=
−
1
.
T
h
e
m
a
x
i
m
u
m
d
i
s
t
a
n
c
e
b
et
we
e
n
P
1
a
n
d
P
2
i
s
2
‖
‖
.
T
o
m
a
x
i
m
i
ze
t
h
e
d
i
s
ta
n
c
e
o
f
t
h
e
h
y
p
e
r
p
l
a
n
e
,
n
e
e
d
t
o
m
i
n
i
m
i
z
e
t
h
e
‖
‖
b
y
u
s
i
n
g
1
2
‖
‖
2
.
A
n
y
t
r
a
i
n
e
d
d
ata
s
e
t
f
a
l
li
n
g
b
e
y
o
n
d
P
1
a
n
d
P
2
t
h
a
t
m
i
n
i
m
iz
e
s
t
h
e
e
q
u
a
t
i
o
n
1
2
‖
‖
2
w
h
i
c
h
s
a
ti
s
f
ie
s
t
h
e
c
o
n
d
i
t
i
o
n
,
(
.
+
)
≥
1
∀
=
1
,
2
,
…
.
.
‒
K
-
n
ea
r
est
n
eig
h
b
o
r
:
K
-
NN
al
g
o
r
ith
m
u
s
es
th
e
ch
ar
ac
ter
is
tics
v
alu
e
s
im
ilar
to
th
e
g
iv
en
d
ata
q
u
er
y
with
th
e
tr
ain
in
g
d
ataset
to
ev
alu
at
e
class
if
icatio
n
m
o
d
els.
I
t
d
es
cr
ib
es
th
e
f
o
llo
win
g
s
tep
s
,
i)
t
h
e
m
o
d
el
r
ea
d
s
th
e
in
p
u
t
as
a
tr
ain
in
g
an
d
t
esti
n
g
d
ataset
;
ii)
d
ef
i
n
es
th
e
K
v
alu
e
,
wh
ich
r
ep
r
esen
ts
th
e
n
u
m
b
er
o
f
n
ea
r
est
p
o
in
ts
th
at
ar
e
in
s
co
p
e
;
iii)
E
u
clid
ian
d
is
tan
ce
is
ca
l
cu
lated
b
y
E
D=
√
(
2
−
1
)
2
+
(
2
−
1
)
2
b
ased
o
n
d
atasets
;
iv
)
o
n
ce
th
e
E
u
clid
ian
d
is
tan
ce
is
ca
lcu
l
ated
,
th
e
least
K
p
o
in
t
is
s
et
;
an
d
v
)
a
f
in
al
d
ec
is
io
n
is
o
b
tain
ed
b
ased
o
n
t
h
e
n
ea
r
est K
p
o
i
n
t.
Usi
n
g
SVM
an
d
K
-
NN
alg
o
r
ith
m
s
,
a
h
y
b
r
id
ML
clas
s
if
icatio
n
alg
o
r
ith
m
was
m
a
d
e
to
co
llect
th
e
ch
ar
ac
ter
is
tic's
v
alu
e
f
r
o
m
th
e
s
er
v
er
to
tr
ai
n
th
e
d
ata
an
d
cl
ass
if
y
th
e
n
o
r
m
al
a
n
d
a
b
n
o
r
m
al
tr
af
f
ic,
th
e
n
u
s
e
th
e
test
d
ata
to
ch
ec
k
wh
et
h
er
th
e
d
esig
n
ed
m
o
d
el
is
class
if
y
in
g
p
r
o
p
er
ly
o
r
n
o
t.
3.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
Th
e
SDN
n
etwo
r
k
s
im
u
latio
n
u
s
in
g
Min
in
et
with
two
SDN
co
n
tr
o
ller
s
is
,
ODL
an
d
POX
co
n
tr
o
ller
s
ar
e
in
s
talled
o
n
th
e
6
4
-
b
it
Ub
u
n
tu
1
4
.
0
4
o
p
er
atin
g
s
y
s
t
em
an
d
d
eter
m
in
e
th
e
p
e
r
f
o
r
m
an
ce
o
f
t
h
e
two
co
n
tr
o
ller
s
s
ep
ar
ately
.
T
h
e
S
DN
n
etwo
r
k
to
p
o
lo
g
y
is
cr
e
ated
with
1
5
s
witch
es
an
d
1
6
n
o
d
es,
s
h
o
wn
i
n
Fig
u
r
e
2
.
ODL
co
n
tr
o
ller
u
s
es
th
e
ODL
u
s
er
ex
p
er
ien
ce
(
DL
UX)
ap
p
licatio
n
in
ter
f
ac
e.
T
h
i
s
in
ter
f
ac
e
en
ab
les
DL
UX
k
ar
af
f
ea
t
u
r
es
an
d
p
o
r
t
8
1
8
1
is
lin
k
e
d
with
th
e
lo
ca
l
s
y
s
tem
I
P
ad
d
r
ess
,
th
en
in
s
tall
l2
-
s
witch
f
ea
tu
r
es
to
in
v
o
k
e
d
ata
tr
a
n
s
f
er
b
etwe
en
th
e
n
o
d
es
is
s
h
o
wn
in
Fig
u
r
e
3
.
W
h
en
d
ata
is
tr
an
s
f
er
r
ed
f
r
o
m
n
o
d
es 1
,
2
,
a
n
d
3
to
n
o
d
e
1
5
DL
UX
m
o
d
u
le
is
g
ettin
g
u
p
d
ated
is
s
h
o
wn
i
n
Fig
u
r
e
4.
POX
co
n
tr
o
ller
is
a
Py
th
o
n
an
d
L
in
u
x
-
b
ased
co
n
tr
o
ller
.
I
t
is
in
s
talled
with
a
M
in
in
et
v
ir
tu
a
l
m
ac
h
in
e
(
VM
)
with
m
in
ied
it.p
y
.
First,
s
tar
t
Min
in
et
to
p
o
lo
g
y
f
o
r
t
h
e
POX
co
n
tr
o
ller
u
s
in
g
th
e
co
m
m
an
d
as,
s
u
d
o
mn
--
t
o
p
o
s
in
g
le,
3
--
m
ac
--
a
r
p
--
s
witch
o
v
s
k
--
co
n
tr
o
ller
=r
em
o
te
.
Nex
t
o
p
e
n
GUI
POX
co
n
tr
o
ller
with
cr
ea
ted
n
etwo
r
k
,
ODL
p
r
o
v
id
es
th
e
v
is
u
alize
d
r
ep
r
esen
tatio
n
o
f
d
ata
tr
an
s
f
er
b
etwe
en
th
e
n
o
d
es
an
d
tr
af
f
ic
is
u
p
d
ated
i
n
DL
UX
m
o
d
u
le.
T
o
d
etec
t
th
e
DDo
S
attac
k
,
tr
af
f
i
c
will
b
e
g
en
er
ated
a
n
d
ca
p
tu
r
es
th
is
tr
af
f
ic
u
s
in
g
th
e
W
ir
esh
ar
k
to
o
l a
s
th
e
g
e
n
e
r
ated
d
ataset
is
s
h
o
wn
in
T
ab
l
e
2.
Fig
u
r
e
2
.
SDN
n
etwo
r
k
with
1
5
s
witch
es a
n
d
1
6
n
o
d
es
Fig
u
r
e
3
.
DL
UX
m
o
d
u
le
in
O
DL
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
mp
o
w
erin
g
S
DN
w
ith
DDo
S
a
tta
ck
d
etec
tio
n
:
leve
r
a
g
in
g
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
…
(
F
lo
r
a
n
ce
G
.
)
2485
Fig
u
r
e
4
.
Flo
ws s
tatis
tics
in
th
e
DL
UX
m
o
d
u
le
3
.
1
.
E
s
t
im
a
t
e
ba
nd
width
t
hr
esh
o
ld
T
o
ev
al
u
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
th
e
p
r
o
p
o
s
ed
esti
m
ated
th
r
e
s
h
o
ld
in
two
SDN
co
n
tr
o
ller
s
(
Fig
u
r
e
5
)
,
k
ee
p
s
en
d
in
g
th
e
tr
af
f
ic
g
en
e
r
ated
b
y
th
e
T
o
r
’
s
h
am
m
er
to
o
l
f
r
o
m
n
o
d
es
1
,
2
,
an
d
3
to
n
o
d
e
1
5
.
Her
e
two
s
ce
n
ar
io
s
ar
e
o
b
s
er
v
ed
.
I
n
th
e
f
ir
s
t
s
ce
n
ar
io
,
th
e
tr
af
f
ic
was
s
en
t
b
etwe
en
th
e
s
am
p
lin
g
p
er
io
d
o
f
T
1
5
to
T
3
0
.
T
h
is
s
am
p
lin
g
d
u
r
atio
n
ca
lcu
l
ates
th
e
b
an
d
wid
th
u
tili
za
tio
n
o
f
th
e
n
etwo
r
k
u
s
in
g
NSI
PA,
SDR
,
E
FP
,
E
FB
,
an
d
FMR
b
y
I
DPS
co
n
tr
o
ller
.
I
f
th
e
b
a
n
d
wid
th
u
tili
za
tio
n
i
s
less
th
an
7
5
%
o
f
th
e
ac
tu
al
b
an
d
wid
th
ass
ig
n
ed
to
th
e
n
etwo
r
k
,
t
h
en
t
h
at
tr
af
f
ic
is
leg
itima
te,
is
s
h
o
wn
in
F
ig
u
r
e
5
(
a)
.
Similar
ly
,
in
th
e
s
ec
o
n
d
s
ce
n
a
r
io
,
th
e
s
am
e
th
in
g
will
b
e
ca
lcu
late
d
d
u
r
in
g
th
e
s
am
p
lin
g
p
er
io
d
o
f
T
1
5
to
T
3
0
,
an
d
th
e
b
a
n
d
wid
th
u
tili
za
tio
n
.
I
f
it'
s
g
r
ea
ter
th
an
7
5
% o
f
th
e
ac
tu
al
b
an
d
wid
th
th
en
s
u
s
p
ec
t
th
at
tr
af
f
ic
is
m
alicio
u
s
,
as sh
o
wn
in
Fig
u
r
e
5
(
b
)
.
(
a)
(
b
)
Fig
u
r
e
5
.
B
an
d
wid
th
th
r
esh
o
ld
esti
m
atio
n
of
(
a)
th
e
b
a
n
d
wid
th
u
s
ag
e
is
less
th
an
7
5
% a
n
d
(
b
)
th
e
b
an
d
wid
t
h
u
s
ag
e
is
g
r
ea
ter
th
an
7
5
%
3
.
2
.
P
a
ck
e
t
f
iltr
a
t
io
n us
ing
m
a
chine le
a
rning
m
o
del in serv
er
On
ce
th
e
p
ar
tial
f
iltra
tio
n
is
d
o
n
e,
n
ex
t
m
o
v
e
o
n
to
s
er
v
er
-
lev
el
f
iltra
tio
n
f
o
r
s
u
s
p
ec
tin
g
tr
af
f
ic
b
y
ca
lcu
latin
g
th
e
s
ix
-
tu
p
le
ch
ar
ac
ter
is
tics
v
alu
e
s
u
ch
as
N
S
I
PA,
NSP,
SD
R
,
E
FP
,
E
FB
,
an
d
FMR
in
th
e
s
am
p
lin
g
in
ter
v
al
o
f
T
1
5
to
T
3
0
p
e
r
io
d
s
.
Du
r
i
n
g
th
e
p
er
io
d
o
f
T
1
5
to
T
3
0
,
th
e
s
ix
-
tu
p
le
f
ea
tu
r
e
v
ec
to
r
v
alu
e
was
ca
lcu
lated
b
ased
o
n
th
e
g
en
er
ated
d
ataset,
as
s
h
o
wn
in
Fig
u
r
e
6
.
Fig
u
r
e
s
6
(
a)
an
d
6
(
b
)
s
h
o
w
th
e
n
u
m
b
er
o
f
I
P
ad
d
r
ess
es
an
d
n
u
m
b
er
o
f
p
o
r
ts
in
th
e
p
a
r
ticu
lar
i
n
ter
v
al
o
f
T
1
5
to
T
3
0
.
T
h
e
n
u
m
b
er
o
f
p
ac
k
ets
tr
an
s
f
er
r
ed
p
er
s
ec
o
n
d
in
th
e
in
ter
v
al
is
s
h
o
wn
in
Fig
u
r
e
6
(
c)
.
T
h
e
en
tr
o
p
y
v
al
u
e
o
f
th
e
f
lo
w
p
ac
k
et
is
r
ep
r
esen
ted
b
y
tak
i
n
g
th
e
co
u
n
t
o
f
th
e
p
r
o
b
ab
ilit
y
o
f
p
ac
k
e
ts
an
d
its
lo
g
ar
ith
m
ic
v
alu
e
p
er
f
lo
w
n
u
m
b
e
r
o
f
T
1
5
to
T
3
0
is
s
h
o
wn
in
Fig
u
r
e
6
(
d
)
,
th
e
co
u
n
t
o
f
th
e
p
r
o
b
ab
ilit
y
o
f
s
ize
o
f
p
ac
k
ets
e
n
ter
ed
in
to
th
e
FT
an
d
its
lo
g
ar
ith
m
ic
v
alu
e
p
er
f
lo
w
n
u
m
b
er
i
n
th
e
in
ter
v
al
o
f
T
1
5
t
o
T
3
0
is
i
n
Fig
u
r
e
6
(
e)
an
d
th
e
FT
m
atch
in
g
r
atio
is
d
eter
m
in
ed
b
y
f
in
d
in
g
n
u
m
b
e
r
o
f
f
lo
ws
th
at
ar
e
g
ettin
g
m
atch
ed
with
FT
en
tr
y
p
er
n
u
m
b
er
o
f
f
lo
ws
is
in
Fig
u
r
e
6
(
f
)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
3
,
J
u
n
e
2
0
2
5
:
2
4
7
9
-
2
4
8
9
2486
(
a)
(
b
)
(
c)
(
d
)
(
e)
(f)
Fig
u
r
e
6
.
Six
tu
p
le
ch
ar
ac
ter
is
tic
v
alu
es sh
o
wn
in
two
co
n
tr
o
ller
s
as OD
L
an
d
POX
co
n
tr
o
l
ler
of
(
a
)
t
h
e
n
u
m
b
er
o
f
s
o
u
r
ce
I
P a
d
d
r
ess
,
(
b
)
n
u
m
b
er
o
f
s
o
u
r
ce
p
o
r
t
,
(
c)
s
p
ee
d
o
f
d
ata
r
ate
,
(
d
)
e
n
tr
o
p
y
v
alu
e
o
f
Flo
w
Pack
ets
,
(
e)
e
n
tr
o
p
y
v
al
u
e
o
f
F
lo
w
B
y
tes
,
an
d
(
f
)
F
T
m
atc
h
in
g
r
atio
3
.3
.
P
er
f
o
r
m
a
nce
i
nd
ica
t
o
rs
T
h
e
SVM
an
d
K
-
NN
m
o
d
els
ar
e
u
s
ed
to
tr
ai
n
th
e
d
ata
u
s
in
g
th
e
ca
lcu
lated
c
h
ar
ac
ter
is
tic
v
alu
es
o
f
NSI
PA,
NSP,
E
F
P,
E
FB
,
a
n
d
FMR
an
d
test
th
e
d
ata
u
s
in
g
th
e
m
o
d
els
to
p
r
e
d
ict
th
e
test
d
ata.
T
h
e
p
er
f
o
r
m
an
ce
e
v
alu
atio
n
o
f
p
r
o
p
o
s
ed
I
DPS
tech
n
iq
u
es
in
ML
in
d
icate
s
ac
cu
r
ac
y
,
r
ec
all,
f
alse
alar
m
r
ate
(
FAR
)
,
p
r
ec
is
io
n
,
an
d
F
-
m
ea
s
u
r
e.
I
t
is
b
ased
o
n
d
eter
m
in
in
g
tr
u
e
p
o
s
itiv
e
(
T
P)
an
d
f
alse
p
o
s
itiv
e
(
FP
)
s
h
o
ws
h
o
w
m
an
y
attac
k
s
ar
e
p
r
ed
ict
ed
co
r
r
ec
tl
y
an
d
in
co
r
r
ec
tly
,
t
r
u
e
n
e
g
ativ
e
(
T
N)
an
d
f
alse
n
eg
ativ
e
(
FN)
g
iv
es
th
e
h
o
w
m
an
y
n
o
r
m
al
f
lo
ws
h
as
p
r
ed
icted
co
r
r
ec
tly
an
d
in
co
r
r
ec
tly
.
SVM
ac
h
iev
es
g
o
o
d
r
esu
lts
in
ter
m
s
o
f
ac
cu
r
ac
y
,
r
ec
all,
an
d
F
-
m
ea
s
u
r
e
in
th
e
ODL
c
o
n
tr
o
ller
c
o
m
p
ar
ed
to
th
e
K
-
NN
alg
o
r
ith
m
,
is
s
h
o
wn
in
T
ab
le
4
an
d
p
e
r
f
o
r
m
an
ce
ch
ar
t
f
o
r
SVM
an
d
K
-
NN
m
o
d
el
as
s
h
o
wn
in
Fig
u
r
e
7
,
its
p
er
f
o
r
m
a
n
ce
o
f
ODL
co
n
tr
o
ller
is
s
h
o
wn
in
Fig
u
r
e
7
(
a)
an
d
POX
co
n
tr
o
ller
is
s
h
o
wn
in
Fig
u
r
e
7
(
b
)
.
T
ab
le
5
d
ep
icts
th
e
ML
ch
ar
ac
ter
is
tics
v
alu
e
an
d
its
m
etr
ics.
T
ab
le
6
p
o
r
tr
ay
s
th
e
c
o
m
p
ar
ativ
e
in
v
esti
g
atio
n
tak
en
b
etwe
en
th
e
p
r
o
p
o
s
ed
m
o
d
el
a
n
d
th
e
ex
is
tin
g
m
o
d
el.
T
ab
le
4
.
Per
f
o
r
m
an
ce
in
d
icato
r
s
f
o
r
SVM
an
d
K
-
NN
class
if
ier
s
C
o
n
t
r
o
l
l
e
r
s
C
l
a
s
si
f
i
e
r
A
c
c
u
r
a
c
y
F
a
l
se
a
l
a
r
m
r
a
t
e
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
F
-
mea
su
r
e
O
p
e
n
D
a
y
l
i
g
h
t
S
V
M
9
6
.
9
5
0
.
0
9
6
.
9
1
1
.
0
9
8
.
4
4
K
-
NN
9
6
.
6
4
0
.
0
9
6
.
6
1
1
.
0
9
8
.
1
8
P
O
X
S
V
M
9
6
.
8
8
0
.
0
9
6
.
8
2
1
.
0
9
8
.
3
4
K
-
NN
9
6
.
5
8
0
.
0
9
6
.
5
1
.
0
9
8
.
1
3
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J Ar
tif
I
n
tell
I
SS
N:
2252
-
8
9
3
8
E
mp
o
w
erin
g
S
DN
w
ith
DDo
S
a
tta
ck
d
etec
tio
n
:
leve
r
a
g
in
g
h
yb
r
id
ma
ch
in
e
lea
r
n
in
g
b
a
s
ed
…
(
F
lo
r
a
n
ce
G
.
)
2487
(
a)
(
b
)
Fig
u
r
e
7
.
Per
f
o
r
m
an
c
e
ch
ar
t
f
o
r
SVM
an
d
K
-
NN
m
o
d
el
in
(
a)
ODL
co
n
tr
o
ller
an
d
(
b
)
PO
X
co
n
tr
o
ller
T
ab
le
5
.
ML
c
h
ar
ac
ter
is
tics
v
alu
es a
n
d
m
etr
ics
C
o
n
t
r
o
l
l
e
r
s
A
l
g
o
r
i
t
h
m
TP
FP
TN
FN
A
c
c
u
r
a
c
y
R
e
c
a
l
l
P
r
e
c
i
s
i
o
n
F
-
mea
su
r
e
ODL
S
V
M
4
6
1
15
4
1
2
18
0
.
9
6
9
5
0
.
9
6
9
1
0
.
9
4
9
0
.
9
8
4
4
K
N
N
4
5
8
17
4
1
1
16
0
.
9
6
6
4
0
.
9
6
6
1
0
.
9
4
3
0
.
9
8
1
8
P
O
X
S
V
M
4
5
9
15
4
1
5
18
0
.
9
6
8
8
0
.
9
6
8
2
0
.
9
4
9
0
.
9
8
3
4
K
N
N
4
5
1
17
4
0
9
14
0
.
9
6
5
8
0
.
9
6
5
0
.
9
4
3
0
.
9
8
1
3
T
ab
le
6
.
C
o
m
p
a
r
ativ
e
an
aly
s
is
o
f
th
e
p
r
o
p
o
s
ed
m
o
d
el
with
th
e
ex
is
tin
g
m
o
d
el
M
o
d
e
l
/
f
r
a
mew
o
r
k
P
a
r
a
me
t
e
r
u
se
d
S
i
mu
l
a
t
i
o
n
/
d
a
t
a
se
t
s
A
c
c
u
r
a
c
y
(
%)
S
N
O
R
T
I
D
S
a
n
d
t
r
a
i
n
e
d
M
L
I
D
S
[
1
7
]
i
p
.
l
e
n
g
t
h
,
i
p
.
f
l
a
g
s.
d
f
,
t
c
p
.
w
i
n
d
o
w
s
_
si
z
e
,
t
c
p
.
l
e
n
g
t
h
,
i
p
.
f
r
a
g
.
o
f
f
se
t
I
o
T
d
e
v
i
c
e
w
i
t
h
R
y
u
c
o
n
t
r
o
l
l
e
r
a
n
d
U
N
S
W
-
N
B
1
5
94
A
r
t
i
f
i
c
i
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
(
A
N
N
)
,
R
a
n
d
o
m
f
o
r
e
s
t
[
1
8
]
P
k
t
Le
n
M
a
x
,
F
w
d
P
k
t
Le
n
M
a
x
,
P
k
t
Le
n
M
i
n
,
F
w
d
S
e
g
S
i
z
e
M
i
n
,
F
w
d
P
k
t
L
e
n
S
t
d
M
i
n
i
n
e
t
w
i
t
h
P
O
X
c
o
n
t
r
o
l
l
e
r
a
n
d
C
I
C
D
D
o
S
2
0
1
9
7
2
.
4
9
M
o
d
i
f
i
e
d
h
y
b
r
i
d
d
e
e
p
b
e
l
i
e
f
n
e
t
w
o
r
k
w
i
t
h
w
e
i
g
h
t
s
[
2
3
]
A
v
a
i
l
a
b
l
e
b
a
n
d
w
i
d
t
h
,
C
P
U
u
t
i
l
i
z
a
t
i
o
n
,
P
a
c
k
e
t
I
n
a
n
d
P
a
c
k
e
t
O
u
t
C
I
C
I
D
S
2
0
1
8
9
6
.
6
2
D
e
e
p
l
e
a
r
n
i
n
g
-
R
N
N
[
2
5
]
P
k
t
c
o
u
n
t
,
p
k
t
r
a
t
e
,
t
x
_
b
y
t
e
s
,
p
a
c
k
e
t
i
n
s,
p
k
t
p
e
r
f
l
o
w
a
n
d
p
r
o
t
o
c
o
l
M
i
n
i
n
e
t
w
i
t
h
R
y
u
c
o
n
t
r
o
l
l
e
r
a
n
d
D
D
O
S
a
t
t
a
c
k
S
D
N
D
a
t
a
s
e
t
9
4
.
9
8
P
r
o
p
o
se
d
m
o
d
e
l
(
I
D
P
S
_
c
o
n
t
r
o
l
l
e
r
i
n
P
O
X
a
n
d
O
D
L
c
o
n
t
r
o
l
l
e
r
)
N
S
I
P
A
,
N
S
P
,
S
D
R
,
EFP
,
EF
B
a
n
d
F
M
R
M
i
n
i
n
e
t
w
i
t
h
P
O
X
C
o
n
t
r
o
l
l
e
r
a
n
d
O
D
L
c
o
n
t
r
o
l
l
e
r
-
G
e
n
e
r
a
t
e
d
d
a
t
a
s
e
t
9
6
.
7
5
4.
CO
NCLU
SI
O
N
T
h
is
p
ap
er
an
aly
s
es
m
itig
ati
o
n
tech
n
iq
u
es
f
o
r
t
h
e
p
r
ev
e
n
tio
n
o
f
DDo
S
attac
k
s
u
s
in
g
s
ix
-
tu
p
le
ch
ar
ac
ter
is
tic
v
alu
es
r
elate
d
t
o
DDo
S
attac
k
s
.
T
h
e
ac
c
u
r
ac
y
o
f
f
in
d
in
g
m
alicio
u
s
p
ac
k
et
s
is
9
6
.
75
%
in
t
h
is
s
tu
d
y
.
T
o
ac
h
ie
v
e
a
b
etter
ac
cu
r
ac
y
b
an
d
wid
t
h
th
r
esh
o
ld
e
s
tim
atio
n
is
d
er
iv
ed
.
T
h
e
tr
af
f
ic
is
s
eg
r
eg
ated
b
y
u
s
in
g
th
e
th
r
esh
o
ld
le
v
el
o
f
7
5
%
b
an
d
wid
th
b
y
th
e
ODL
a
n
d
POX
co
n
tr
o
ller
s
.
Af
ter
d
ef
in
in
g
th
e
th
r
esh
o
ld
,
tr
af
f
ic
is
g
en
er
ated
u
s
in
g
T
o
r
’
s
h
am
m
ed
n
etwo
r
k
to
o
l
an
d
a
n
aly
ze
d
u
s
in
g
th
e
W
ir
esh
ar
k
c
ap
tu
r
in
g
to
o
l
in
th
e
s
er
v
er
,
s
u
b
s
eq
u
e
n
tly
,
th
e
d
ata
s
et
is
cr
ea
ted
.
Fro
m
th
e
d
ata
s
et,
s
ix
-
tu
p
le
f
ea
tu
r
e
v
ec
to
r
s
v
alu
e
is
ca
lcu
lated
d
u
r
in
g
th
e
s
am
p
lin
g
i
n
ter
v
al
o
f
T
1
5
to
T
3
0
an
d
class
if
ied
th
e
leg
itima
te
tr
af
f
ic
an
d
m
al
icio
u
s
tr
af
f
ic
u
s
in
g
SVM
an
d
K
-
NN
alg
o
r
ith
m
s
in
ODL
a
n
d
POX
co
n
tr
o
ller
s
.
Am
o
n
g
th
ese
ca
lcu
lated
v
al
u
es
,
NI
SP
A,
NSP,
E
FP
,
an
d
FMR
wer
e
u
s
ed
f
o
r
class
if
y
in
g
attac
k
s
.
I
n
co
m
p
a
r
in
g
th
e
two
co
n
tr
o
ller
s
,
th
e
ac
c
u
r
ac
y
r
ate
o
f
SVM
is
r
elativ
ely
h
ig
h
ac
cu
r
ac
y
in
th
e
ODL
co
n
tr
o
ller
co
m
p
a
r
e
d
to
th
e
POX
co
n
tr
o
ller
an
d
ac
h
iev
in
g
a
9
6
.
75
%
ac
cu
r
ac
y
in
d
etec
tin
g
m
alicio
u
s
p
ac
k
ets
.
I
n
th
e
f
u
tu
r
e,
m
u
lti
p
le
I
DPS
co
n
tr
o
ller
s
will
b
e
p
r
o
p
o
s
ed
v
ir
tu
all
y
to
ex
ec
u
te
DDo
S
attac
k
s
in
m
u
ltip
le
p
lace
s
to
im
p
r
o
v
e
th
e
p
er
f
o
r
m
an
ce
o
f
v
ar
io
u
s
class
if
ier
s
in
th
e
SDN
n
etwo
r
k
.
T
h
e
d
y
n
am
ic
t
h
r
esh
o
ld
esti
m
atio
n
will
b
e
e
m
b
ed
d
ed
in
th
e
r
ea
l
-
tim
e
s
ce
n
a
r
io
.
E
x
p
lo
r
in
g
th
e
in
tellig
en
ce
o
f
c
o
llab
o
r
ativ
e
a
p
p
r
o
ac
h
to
lev
e
r
ag
e
th
e
n
etwo
r
k
in
r
ea
l
-
tim
e.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
is
wo
r
k
is
n
o
t f
u
n
d
e
d
b
y
an
y
ag
en
c
y
o
r
i
n
s
titu
te
AUTHO
R
CO
NT
RI
B
UT
I
O
NS ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
tr
ib
u
to
r
R
o
les
T
ax
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
id
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
9
3
8
I
n
t J Ar
tif
I
n
tell
,
Vo
l.
1
4
,
No
.
3
,
J
u
n
e
2
0
2
5
:
2
4
7
9
-
2
4
8
9
2488
Na
m
e
o
f
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Flo
r
an
ce
G
.
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
R
.
J
.
An
an
d
h
i
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
O
F
I
N
T
E
R
E
S
T
ST
A
T
E
M
E
NT
T
h
e
au
th
o
r
s
s
tate
n
o
co
n
f
lict o
f
in
ter
est.
DATA AV
AI
L
AB
I
L
I
T
Y
T
h
e
d
ata
th
at
s
u
p
p
o
r
t
th
e
f
in
d
in
g
s
o
f
th
is
s
tu
d
y
ar
e
av
ailab
l
e
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
au
t
h
o
r
,
[
F
G
]
,
u
p
o
n
r
ea
s
o
n
ab
le
r
eq
u
est.
RE
F
E
R
E
NC
E
S
[
1
]
P
.
K
a
u
r
,
M
.
K
u
m
a
r
,
a
n
d
A
.
B
h
a
n
d
a
r
i
,
“
A
r
e
v
i
e
w
o
f
d
e
t
e
c
t
i
o
n
a
p
p
r
o
a
c
h
e
s
f
o
r
d
i
s
t
r
i
b
u
t
e
d
d
e
n
i
a
l
o
f
ser
v
i
c
e
a
t
t
a
c
k
s,
”
S
y
s
t
e
m
s
S
c
i
e
n
c
e
&
C
o
n
t
ro
l
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
5
,
n
o
.
1
,
p
p
.
3
0
1
–
3
2
0
,
Ja
n
.
2
0
1
7
,
d
o
i
:
1
0
.
1
0
8
0
/
2
1
6
4
2
5
8
3
.
2
0
1
7
.
1
3
3
1
7
6
8
.
[
2
]
B
.
I
sy
a
k
u
,
M
.
S
.
M
o
h
d
Za
h
i
d
,
M
.
B
t
e
K
a
ma
t
,
K
.
A
b
u
B
a
k
a
r
,
a
n
d
F
.
A
.
G
h
a
l
e
b
,
“
S
o
f
t
w
a
r
e
d
e
f
i
n
e
d
n
e
t
w
o
r
k
i
n
g
f
l
o
w
t
a
b
l
e
man
a
g
e
me
n
t
o
f
O
p
e
n
F
l
o
w
sw
i
t
c
h
e
s
p
e
r
f
o
r
m
a
n
c
e
a
n
d
s
e
c
u
r
i
t
y
c
h
a
l
l
e
n
g
e
s
:
a
s
u
r
v
e
y
,
”
F
u
t
u
r
e
I
n
t
e
r
n
e
t
,
v
o
l
.
1
2
,
n
o
.
9
,
A
u
g
.
2
0
2
0
,
d
o
i
:
1
0
.
3
3
9
0
/
f
i
1
2
0
9
0
1
4
7
.
[
3
]
Y
.
G
u
o
,
F
.
M
i
a
o
,
L
.
Zh
a
n
g
,
a
n
d
Y
.
W
a
n
g
,
“
C
A
TH
:
a
n
e
f
f
e
c
t
i
v
e
me
t
h
o
d
f
o
r
d
e
t
e
c
t
i
n
g
d
e
n
i
a
l
-
of
-
serv
i
c
e
a
t
t
a
c
k
s
i
n
so
f
t
w
a
r
e
d
e
f
i
n
e
d
n
e
t
w
o
r
k
s,”
S
c
i
e
n
c
e
C
h
i
n
a
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
s
,
v
o
l
.
6
2
,
n
o
.
3
,
p
.
3
2
1
0
6
,
M
a
r
.
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
4
3
2
-
0
1
7
-
9
4
3
9
-
7.
[
4
]
C
.
S
.
K
h
i
n
,
A
.
T
.
K
y
a
w
,
M
.
M
.
M
a
w
,
a
n
d
M
.
Z
.
O
o
,
“
R
e
d
u
c
i
n
g
p
a
c
k
e
t
-
i
n
messa
g
e
s
i
n
O
p
e
n
F
l
o
w
n
e
t
w
o
r
k
s
,
”
E
C
T
I
T
r
a
n
s
a
c
t
i
o
n
s
o
n
El
e
c
t
ri
c
a
l
E
n
g
i
n
e
e
r
i
n
g
,
El
e
c
t
ro
n
i
c
s,
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
0
,
n
o
.
1
,
p
p
.
1
–
9
,
F
e
b
.
2
0
2
2
,
d
o
i
:
1
0
.
3
7
9
3
6
/
e
c
t
i
-
e
e
c
.
2
0
2
2
2
0
1
.
2
4
4
9
4
4
.
[
5
]
P
.
O
h
r
i
,
A
.
D
a
n
i
e
l
,
S
.
G
.
N
e
o
g
i
,
a
n
d
S
.
K
.
M
u
t
t
o
o
,
“
B
l
o
c
k
c
h
a
i
n
-
b
a
se
d
se
c
u
r
i
t
y
f
r
a
mew
o
r
k
f
o
r
m
i
t
i
g
a
t
i
n
g
n
e
t
w
o
r
k
a
t
t
a
c
k
s i
n
m
u
l
t
i
-
S
D
N
c
o
n
t
r
o
l
l
e
r
e
n
v
i
r
o
n
me
n
t
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
J
u
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
1
8
7
0
-
0
2
4
-
0
1
9
3
3
-
8.
[
6
]
R
.
S
w
a
mi
,
M
.
D
a
v
e
,
a
n
d
V
.
R
a
n
g
a
,
“
D
e
t
e
c
t
i
o
n
a
n
d
a
n
a
l
y
s
i
s
o
f
T
C
P
-
S
Y
N
D
D
o
S
at
t
a
c
k
i
n
so
f
t
w
a
r
e
-
d
e
f
i
n
e
d
n
e
t
w
o
r
k
i
n
g
,
”
W
i
re
l
e
ss
Pe
rso
n
a
l
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
1
1
8
,
n
o
.
4
,
p
p
.
2
2
9
5
–
2
3
1
7
,
J
u
n
.
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
2
7
7
-
0
2
1
-
0
8
1
2
7
-
6.
[
7
]
D
.
K
i
m
,
B
.
K
i
m
,
I
.
K
i
m,
J.
K
i
m
,
a
n
d
H
.
C
h
o
,
“
E
n
d
p
o
i
n
t
mi
t
i
g
a
t
i
o
n
o
f
D
D
o
S
a
t
t
a
c
k
s b
a
s
e
d
o
n
d
y
n
a
mi
c
t
h
r
e
sh
o
l
d
i
n
g
,
”
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
S
e
c
u
ri
t
y
,
p
p
.
3
8
1
–
3
9
1
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
6
4
2
-
3
4
1
2
9
-
8
_
3
6
.
[
8
]
H
.
A
.
A
l
a
mr
i
a
n
d
V
.
Th
a
y
a
n
a
n
t
h
a
n
,
“
B
a
n
d
w
i
d
t
h
c
o
n
t
r
o
l
m
e
c
h
a
n
i
sm
a
n
d
e
x
t
r
e
me
g
r
a
d
i
e
n
t
b
o
o
st
i
n
g
a
l
g
o
r
i
t
h
m
f
o
r
p
r
o
t
e
c
t
i
n
g
so
f
t
w
a
r
e
-
d
e
f
i
n
e
d
n
e
t
w
o
r
k
s
a
g
a
i
n
s
t
D
D
o
S
a
t
t
a
c
k
s
,”
I
E
EE
Ac
c
e
s
s
,
v
o
l
.
8
,
p
p
.
1
9
4
2
6
9
–
1
9
4
2
8
8
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
0
.
3
0
3
3
9
4
2
.
[
9
]
E.
R
.
Ji
mso
n
,
K
.
N
i
s
a
r
,
a
n
d
M
.
H
.
b
i
n
A
h
ma
d
H
i
j
a
z
i
,
“
B
a
n
d
w
i
d
t
h
ma
n
a
g
e
m
e
n
t
u
si
n
g
so
f
t
w
a
r
e
d
e
f
i
n
e
d
n
e
t
w
o
r
k
a
n
d
c
o
m
p
a
r
i
so
n
o
f
t
h
e
t
h
r
o
u
g
h
p
u
t
p
e
r
f
o
r
ma
n
c
e
w
i
t
h
t
r
a
d
i
t
i
o
n
a
l
n
e
t
w
o
r
k
,
”
i
n
2
0
1
7
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
p
u
t
e
r
a
n
d
D
r
o
n
e
A
p
p
l
i
c
a
t
i
o
n
s
(
I
C
o
n
D
A)
,
N
o
v
.
2
0
1
7
,
p
p
.
7
1
–
76
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
O
N
D
A
.
2
0
1
7
.
8
2
7
0
4
0
2
.
[
1
0
]
C.
-
H
.
Li
n
,
J.
-
C
.
L
i
u
,
H
.
-
C
.
H
u
a
n
g
,
a
n
d
T.
-
C
.
Y
a
n
g
,
“
U
s
i
n
g
a
d
a
p
t
i
v
e
b
a
n
d
w
i
d
t
h
a
l
l
o
c
a
t
i
o
n
a
p
p
r
o
a
c
h
t
o
d
e
f
e
n
d
D
D
o
S
a
t
t
a
c
k
s
,
”
i
n
2
0
0
8
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Mu
l
t
i
m
e
d
i
a
a
n
d
U
b
i
q
u
i
t
o
u
s
E
n
g
i
n
e
e
ri
n
g
(
m
u
e
2
0
0
8
)
,
2
0
0
8
,
p
p
.
1
7
6
–
1
8
1
,
d
o
i
:
1
0
.
1
1
0
9
/
M
U
E.
2
0
0
8
.
2
3
.
[
1
1
]
S
.
S
u
ma
t
h
i
,
R
.
R
a
j
e
s
h
,
a
n
d
K
.
N
a
g
a
r
a
j
a
n
,
“
D
D
o
S
a
t
t
a
c
k
d
e
t
e
c
t
i
o
n
u
si
n
g
h
y
b
r
i
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
b
a
se
d
I
D
S
mo
d
e
l
s
,”
J
o
u
rn
a
l
o
f
S
c
i
e
n
t
i
f
i
c
&
I
n
d
u
st
r
i
a
l
R
e
se
a
rc
h
,
v
o
l
.
8
1
,
n
o
.
0
3
,
M
a
r
.
2
0
2
2
,
d
o
i
:
1
0
.
5
6
0
4
2
/
j
si
r
.
v
8
1
i
0
3
.
5
8
4
5
1
.
[
1
2
]
K
.
A
l
e
m
e
r
i
e
n
,
S
.
A
l
-
su
h
e
m
a
t
,
a
n
d
M
.
A
l
ma
h
a
d
i
n
,
“
T
o
w
a
r
d
s
o
p
t
i
m
i
z
e
d
mac
h
i
n
e
-
l
e
a
r
n
i
n
g
-
d
r
i
v
e
n
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
f
o
r
i
n
t
e
r
n
e
t
o
f
t
h
i
n
g
s
a
p
p
l
i
c
a
t
i
o
n
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
6
,
n
o
.
8
,
p
p
.
4
9
8
1
–
4
9
9
4
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
4
1
8
7
0
-
0
2
4
-
0
1
8
5
2
-
8.
[
1
3
]
W
.
P
a
r
k
,
J.
L
e
e
,
a
n
d
D
.
S
u
n
g
,
“
B
a
n
d
w
i
d
t
h
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
m
b
a
s
e
d
o
n
b
a
n
d
w
i
d
t
h
r
a
t
i
o
a
d
j
u
st
m
e
n
t
i
n
g
e
n
e
r
a
l
i
z
e
d
p
r
o
c
e
ss
o
r
sh
a
r
i
n
g
s
e
r
v
e
r
s
,
”
i
n
2
0
0
6
I
E
EE
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
2
0
0
6
,
p
p
.
6
9
9
–
7
0
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
.
2
0
0
6
.
2
5
4
7
8
9
.
[
1
4
]
E.
H
i
k
maw
a
t
i
,
N
.
U
.
M
a
u
l
i
d
e
v
i
,
a
n
d
K
.
S
u
r
e
n
d
r
o
,
“
M
i
n
i
m
u
m
t
h
r
e
s
h
o
l
d
d
e
t
e
r
mi
n
a
t
i
o
n
me
t
h
o
d
b
a
se
d
o
n
d
a
t
a
s
e
t
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s
i
n
a
sso
c
i
a
t
i
o
n
r
u
l
e
mi
n
i
n
g
,
”
J
o
u
rn
a
l
o
f
B
i
g
D
a
t
a
,
v
o
l
.
8
,
n
o
.
1
,
p
.
1
4
6
,
D
e
c
.
2
0
2
1
,
d
o
i
:
1
0
.
1
1
8
6
/
s
4
0
5
3
7
-
021
-
0
0
5
3
8
-
3.
[
1
5
]
A
.
T
a
l
u
k
d
e
r
,
S
.
F
.
A
b
e
d
i
n
,
M
.
S
.
M
u
n
i
r
,
a
n
d
C
.
S
.
H
o
n
g
,
“
D
u
a
l
t
h
r
e
sh
o
l
d
l
o
a
d
b
a
l
a
n
c
i
n
g
i
n
S
D
N
e
n
v
i
r
o
n
m
e
n
t
u
si
n
g
p
r
o
c
e
ss
mi
g
r
a
t
i
o
n
,
”
i
n
2
0
1
8
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
I
n
f
o
rm
a
t
i
o
n
N
e
t
w
o
r
k
i
n
g
(
I
C
O
I
N
)
,
Ja
n
.
2
0
1
8
,
p
p
.
7
9
1
–
796
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
O
I
N
.
2
0
1
8
.
8
3
4
3
2
2
6
.
[
1
6
]
U
.
H
.
G
a
r
b
a
,
A
.
N
.
To
o
si
,
M
.
F
.
P
a
s
h
a
,
a
n
d
S
.
K
h
a
n
,
“
S
D
N
-
b
a
se
d
d
e
t
e
c
t
i
o
n
a
n
d
m
i
t
i
g
a
t
i
o
n
o
f
D
D
o
S
a
t
t
a
c
k
s
o
n
sm
a
r
t
h
o
m
e
s,”
C
o
m
p
u
t
e
r
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
v
o
l
.
2
2
1
,
p
p
.
2
9
–
4
1
,
M
a
y
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
o
m
c
o
m
.
2
0
2
4
.
0
4
.
0
0
1
.
[
1
7
]
A
.
Jaw
a
h
a
r
e
t
a
l
.
,
“
D
D
o
S
mi
t
i
g
a
t
i
o
n
u
s
i
n
g
b
l
o
c
k
c
h
a
i
n
a
n
d
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
M
u
l
t
i
m
e
d
i
a
T
o
o
l
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
8
3
,
n
o
.
2
1
,
p
p
.
6
0
2
6
5
–
6
0
2
7
8
,
J
a
n
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
0
4
2
-
023
-
1
8
0
2
8
-
4.
[
1
8
]
I
.
Jeb
r
i
l
e
t
a
l
.
,
“
D
e
e
p
l
e
a
r
n
i
n
g
b
a
se
d
D
D
o
S
a
t
t
a
c
k
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:
a
n
o
p
t
i
m
i
z
e
d
C
N
N
-
B
i
LSTM
a
r
c
h
i
t
e
c
t
u
r
e
w
i
t
h
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
a
n
d
r
e
g
u
l
a
r
i
z
a
t
i
o
n
t
e
c
h
n
i
q
u
e
s
,”
I
n
f
o
c
o
m
m
u
n
i
c
a
t
i
o
n
s
j
o
u
rn
a
l
,
v
o
l
.
1
6
,
n
o
.
1
,
p
p
.
2
–
1
1
,
2
0
2
4
,
d
o
i
:
1
0
.
3
6
2
4
4
/
I
C
J.
2
0
2
4
.
1
.
1
.
[
1
9
]
A
.
V
.
S
o
n
g
a
a
n
d
G
.
R
.
K
a
r
r
i
,
“
A
n
i
n
t
e
g
r
a
t
e
d
S
D
N
f
r
a
mew
o
r
k
f
o
r
e
a
r
l
y
d
e
t
e
c
t
i
o
n
o
f
D
D
o
S
a
t
t
a
c
k
s
i
n
c
l
o
u
d
c
o
m
p
u
t
i
n
g
,
”
J
o
u
r
n
a
l
o
f
C
l
o
u
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
3
,
n
o
.
1
,
M
a
r
.
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s
1
3
6
7
7
-
0
2
4
-
0
0
6
2
5
-
9.
Evaluation Warning : The document was created with Spire.PDF for Python.