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.
1
4
,
No
.
1
,
Feb
r
u
ar
y
2
0
2
5
,
p
p
.
492
~
499
I
SS
N:
2
2
5
2
-
8
9
3
8
,
DOI
: 1
0
.
1
1
5
9
1
/ijai.v
1
4
.i
1
.
p
p
4
9
2
-
4
9
9
492
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
a
i
.
ia
esco
r
e.
co
m
H
y
brid
in
trusio
n
detec
tion
mo
del
f
o
r
hiera
rchica
l
wi
reless
sens
o
r
network
u
sing
federate
d
lea
rning
Sa
t
his
h
k
u
ma
r
M
a
n
i
1
,
P
a
r
a
s
u
ra
m Ch
a
nd
r
a
s
e
k
a
r
a
n
K
is
h
o
re
r
a
j
a
2
,
Ch
ris
t
ee
n
a
J
o
s
eph
3
,
Rej
i M
a
no
ha
ra
n
4
,
P
ra
s
a
nn
a
v
enk
a
t
esa
n
T
heer
t
ha
g
iri
1
1
D
e
p
a
r
t
me
n
t
of
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
,
G
I
TA
M
S
c
h
o
o
l
of
Te
c
h
n
o
l
o
g
y
,
G
I
TA
M
U
n
i
v
e
r
si
t
y
,
B
e
n
g
a
l
u
r
u
,
I
n
d
i
a
2
S
c
h
o
o
l
of
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
I
n
f
o
r
mat
i
o
n
S
y
s
t
e
ms
,
V
e
l
l
o
r
e
I
n
st
i
t
u
t
e
of
Te
c
h
n
o
l
o
g
y
,
V
e
l
l
o
r
e
,
I
n
d
i
a
3
D
e
p
a
r
t
me
n
t
o
f
El
e
c
t
r
o
n
i
c
s
a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
S
RM
I
n
st
i
t
u
t
e
of
S
c
i
e
n
c
e
a
n
d
Te
c
h
n
o
l
o
g
y
,
C
h
e
n
n
a
i
,
I
n
d
i
a
4
D
e
p
a
r
t
me
n
t
of
El
e
c
t
r
o
n
i
c
s a
n
d
C
o
m
mu
n
i
c
a
t
i
o
n
En
g
i
n
e
e
r
i
n
g
,
R
o
h
i
n
i
C
o
l
l
e
g
e
of
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
K
a
n
y
a
k
u
mari
,
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
Ap
r
18
,
2
0
2
4
R
ev
is
ed
Au
g
12
,
2
0
2
4
Acc
ep
ted
Au
g
30
,
2
0
2
4
Th
e
a
p
p
li
c
a
ti
o
n
s
of
wire
les
s
se
n
so
r
n
e
two
rk
s
a
re
v
a
st
a
n
d
p
o
p
u
lar
in
to
d
a
y
’s
tec
h
n
o
l
o
g
y
w
o
rld
.
T
h
e
se
n
e
two
r
k
s
c
o
n
sist
of
sm
a
ll
,
i
n
d
e
p
e
n
d
e
n
t
se
n
so
rs
t
h
a
t
a
re
c
a
p
a
b
le
of
m
e
a
su
rin
g
v
a
ri
o
u
s
p
h
y
sic
a
l
q
u
a
n
t
it
ies
.
De
p
l
o
y
m
e
n
t
of
wire
les
s
se
n
so
r
n
e
two
r
k
s
i
n
c
re
a
se
d
due
to
imm
e
n
se
a
p
p
li
c
a
ti
o
n
s
w
h
ich
a
re
s
u
sc
e
p
ti
b
le
to
d
iffere
n
t
t
y
p
e
s
of
a
tt
a
c
k
s
in
an
u
n
p
ro
tec
ted
a
n
d
o
p
e
n
re
g
i
o
n
.
In
tru
si
o
n
d
e
tec
ti
o
n
s
y
ste
m
s
(IDS)
p
lay
a
v
i
tal
p
a
rt
in
a
n
y
se
c
u
re
d
e
n
v
ir
o
n
m
e
n
t
fo
r
a
n
y
n
e
two
rk
.
ID
S
u
sin
g
fe
d
e
ra
ted
le
a
rn
in
g
h
a
v
e
th
e
p
o
ten
ti
a
l
to
a
c
h
i
e
v
e
b
e
tt
e
r
c
las
sifica
ti
o
n
a
c
c
u
ra
c
y
.
Us
u
a
ll
y
,
a
ll
th
e
d
a
ta
is
sto
re
d
in
c
e
n
tralize
d
se
rv
e
r
in
o
rd
e
r
to
c
o
m
m
u
n
ica
te
b
e
twe
e
n
t
h
e
sy
ste
m
s.
On
t
h
e
o
t
h
e
r
h
a
n
d
,
fe
d
e
ra
ted
lea
rn
in
g
is
a
d
istr
ib
u
ted
lea
rn
in
g
tec
h
n
i
q
u
e
th
a
t
d
o
e
s
not
tran
sfe
r
d
a
ta
but
train
s
m
o
d
e
ls
lo
c
a
ll
y
a
n
d
tran
sfe
rs
th
e
p
a
ra
m
e
ters
to
th
e
c
e
n
traliz
e
d
se
rv
e
r.
Th
e
p
ro
p
o
se
d
re
se
a
rc
h
u
se
s
a
h
y
b
rid
ID
S
f
o
r
wire
les
s
se
n
so
r
n
e
two
rk
s
u
sin
g
fe
d
e
ra
ti
n
g
lea
rn
i
n
g
.
T
h
e
d
e
tec
ti
o
n
tak
e
s
p
lac
e
in
re
a
l
-
ti
m
e
th
ro
u
g
h
d
e
tailed
a
n
a
ly
sis
of
a
tt
a
c
k
s
at
d
iffere
n
t
le
v
e
ls
in
a
d
e
c
e
n
tralize
d
m
a
n
n
e
r.
H
y
b
ri
d
IDS
a
re
d
e
sig
n
e
d
f
o
r
n
o
d
e
le
v
e
l,
c
l
u
ste
r
lev
e
l
a
n
d
t
h
e
b
a
se
sta
ti
o
n
wh
e
re
fe
d
e
ra
ted
lea
rn
in
g
a
c
ts
as
a
c
li
e
n
t
a
n
d
a
g
g
re
g
a
ted
se
rv
e
r.
K
ey
w
o
r
d
s
:
Attack
s
Fed
er
ated
lear
n
in
g
G
lo
b
al
ag
g
r
e
g
ato
r
s
er
v
e
r
I
n
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
s
W
ir
eles
s
s
en
s
o
r
n
etwo
r
k
T
h
is
is
an
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC
BY
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Par
asu
r
am
C
h
an
d
r
asek
ar
a
n
Kis
h
o
r
er
aja
Sch
o
o
l
of
C
o
m
p
u
ter
Scien
ce
a
n
d
I
n
f
o
r
m
atio
n
Sy
s
tem
s
,
Vello
r
e
I
n
s
titu
te
of
T
ec
h
n
o
lo
g
y
Vello
r
e,
I
n
d
ia
E
m
ail:
k
is
h
o
r
er
aja.
pc
@
v
it.a
c.
i
n
1.
I
NT
RO
D
UCT
I
O
N
T
h
e
wir
eless
s
en
s
o
r
n
etwo
r
k
is
ex
ten
s
iv
ely
u
s
ed
in
m
an
y
ap
p
licatio
n
s
lik
e
m
ea
s
u
r
in
g
th
e
t
em
p
er
atu
r
e
of
v
o
lcan
o
es.
T
h
e
n
etwo
r
k
is
s
im
p
le,
lo
w
-
co
s
t,
en
er
g
y
-
ef
f
ici
en
t,
an
d
d
is
tr
ib
u
ted
s
en
s
in
g
an
d
p
r
o
ce
s
s
in
g
,
wh
ic
h
d
ep
icts
th
e
n
etwo
r
k
to
s
ec
u
r
ity
attac
k
s
[
1
]
.
T
r
ad
itio
n
al
m
eth
o
d
s
lik
e
cr
y
p
to
g
r
ap
h
y
m
eth
o
d
s
ar
e
no
lo
n
g
er
u
s
e
d
to
d
ef
en
d
t
h
e
n
etwo
r
k
.
I
n
s
tead
,
an
in
tr
u
s
io
n
d
etec
tio
n
s
y
s
tem
(
I
DS)
d
etec
ts
all
k
in
d
s
of
attac
k
s
.
C
o
n
s
id
er
in
g
all
lim
itatio
n
s
in
wir
eless
s
en
s
o
r
n
etwo
r
k
s
,
I
DS
ar
e
s
p
ec
if
ic
to
d
etec
t
p
ar
ticu
lar
ty
p
es
of
attac
k
s
[
2
]
.
Mo
s
t
attac
k
s
h
ap
p
en
e
d
in
s
en
s
o
r
n
etwo
r
k
s
due
to
m
is
b
eh
av
io
r
of
r
o
u
te
u
p
d
ates.
T
h
is
r
esear
ch
wo
r
k
u
s
es
d
if
f
er
en
t
lev
els
of
I
DS
u
s
in
g
f
ed
er
ated
lear
n
in
g
.
Mo
s
t
of
th
e
I
DS
u
s
e
a
d
is
tr
ib
u
ted
d
etec
tio
n
p
r
o
ce
s
s
in
o
r
d
er
to
less
en
th
e
co
m
p
u
tatio
n
al
lo
a
d
,
b
u
t
an
o
t
h
er
p
r
o
b
lem
ar
is
es:
co
m
m
u
n
i
ca
tio
n
o
v
er
h
ea
d
[
3
]
,
[
4
]
.
T
h
e
au
th
o
r
s
p
r
esen
ted
an
o
m
aly
d
etec
tio
n
an
d
co
m
m
u
n
icate
d
to
a
g
lo
b
al
m
o
d
el
s
y
s
tem
.
It
u
s
es
e
s
tim
ato
r
s
f
o
r
it
s
an
o
m
aly
I
DS
[
4
]
.
T
h
er
e
ar
e
v
ar
io
u
s
class
if
ier
s
,
co
-
v
ar
ian
ce
p
ar
a
m
eter
s
,
an
d
s
tatis
tical
to
o
ls
u
s
ed
to
[
5
]
–
[
9
]
d
etec
t
d
is
tr
ib
u
te
d
an
o
m
alies.
Hy
b
r
id
alg
o
r
ith
m
s
ar
e
p
r
o
p
o
s
ed
u
s
in
g
Qu
a
n
tu
m
p
ar
ticle
s
war
m
o
p
tim
izatio
n
(
PSO
)
an
d
r
ad
ial
b
asis
f
u
n
ctio
n
n
eu
r
al
n
etwo
r
k
(
R
B
F
NN
)
[
1
0
]
–
[
1
2
]
.
Nu
m
e
r
o
u
s
n
e
u
r
al
n
etwo
r
k
-
b
ased
I
DS
ar
e
p
r
o
p
o
s
ed
,
wh
ic
h
g
iv
e
b
etter
ap
p
r
o
x
im
atio
n
ab
ilit
y
,
g
o
o
d
class
if
icatio
n
an
d
f
ast
co
n
v
er
g
en
ce
[
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
Hyb
r
id
in
tr
u
s
io
n
d
etec
tio
n
mo
d
el
fo
r
h
iera
r
ch
ica
l wi
r
eles
s
s
en
s
o
r
n
etw
o
r
k
…
(
S
a
th
is
h
ku
m
a
r
M
ani
)
493
In
r
ec
en
t
y
ea
r
s
,
m
ac
h
in
e
lear
n
in
g
(
ML
)
a
n
d
d
ee
p
lear
n
i
n
g
(
DL
)
alg
o
r
ith
m
s
h
a
v
e
b
ee
n
u
s
e
d
in
m
an
y
d
o
m
ain
s
,
s
u
c
h
as
h
ea
lth
ca
r
e
a
n
d
im
a
g
e
p
r
o
ce
s
s
in
g
.
I
DS,
u
s
i
n
g
ML
tech
n
iq
u
es,
lear
n
s
all
k
in
d
s
of
tr
af
f
ic
[
1
4
]
,
[
1
5
]
.
T
h
e
d
etec
tio
n
p
r
o
c
ess
is
to
tally
b
ased
on
d
ata
co
llected
an
d
s
to
r
ed
ce
n
tr
ally
in
th
e
s
er
v
er
.
It
is
f
o
u
n
d
th
at
th
e
ac
cu
r
ac
y
d
ec
r
ea
s
es
due
to
lar
g
e
d
ata
s
ets
with
h
ig
h
p
ac
k
et
lo
s
s
r
ates
[
1
6
]
,
[
1
7
]
.
T
h
e
p
r
o
b
lem
s
can
be
ad
d
r
ess
ed
by
f
ed
e
r
ated
lear
n
in
g
.
T
h
is
alg
o
r
ith
m
lear
n
s
d
ata
g
en
er
ated
by
th
e
d
ev
ices
in
a
co
llab
o
r
ativ
e
f
ash
io
n
with
o
u
t
an
y
ce
n
tr
alize
d
s
er
v
er
.
It
wo
r
k
s
with
d
ec
e
n
tr
alize
d
d
a
ta
f
r
o
m
d
ev
ices
wh
ic
h
ar
e
co
m
m
u
n
icate
d
in
eith
e
r
d
ir
ec
tio
n
.
T
h
e
tr
a
d
itio
n
al
ce
n
tr
alize
d
lear
n
in
g
m
eth
o
d
s
ex
p
ec
t
to
f
o
llo
w
lo
ca
l
lear
n
in
g
an
d
attain
m
en
t
of
p
r
i
v
ac
y
p
r
eser
v
atio
n
a
n
d
co
s
t
r
ed
u
ctio
n
[
1
8
]
–
[
2
0
]
.
T
h
is
r
esear
ch
p
r
o
p
o
s
ed
h
y
b
r
id
in
tr
u
s
io
n
d
etec
tio
n
f
o
r
wir
eless
s
en
s
o
r
n
etwo
r
k
s
u
s
in
g
f
ed
e
r
ated
lear
n
in
g
.
T
h
e
a
u
th
o
r
s
p
r
o
p
o
s
ed
a
DL
-
b
ased
I
DS
with
f
o
u
r
d
if
f
e
r
en
t
s
tr
ateg
ies
[
2
1
]
.
Kwo
n
et
a
l.
[
2
0
]
r
ev
iewe
d
o
n
l
y
s
ev
en
DL
-
b
ased
s
o
l
u
tio
n
s
f
o
r
I
DS.
T
h
er
e
ar
e
d
if
f
er
e
n
t
ty
p
es
of
I
DS
av
ailab
le,
wh
ich
ev
alu
ate
v
ar
io
u
s
in
f
o
r
m
atio
n
av
ailab
le
on
s
in
g
le
or
m
u
ltip
le
h
o
s
ts
as
well
as
an
al
y
zin
g
f
r
o
m
ca
p
t
u
r
ed
p
ac
k
ets
d
u
r
in
g
tr
an
s
m
is
s
io
n
b
etwe
en
th
e
n
o
d
es.
Sig
n
atu
r
e
-
b
ased
in
tr
u
s
io
n
d
etec
tio
n
u
s
es
p
atter
n
s
in
th
e
d
etec
tio
n
m
o
d
e
l,
wh
er
ea
s
th
e
a
n
o
m
al
y
d
etec
ti
o
n
m
o
d
el
lo
o
k
s
f
o
r
ab
n
o
r
m
ality
in
n
etwo
r
k
tr
a
f
f
i
c.
An
o
m
al
y
d
etec
tio
n
tech
n
iq
u
es
u
s
e
s
tatis
tical
m
o
d
els,
n
e
u
r
al
n
etwo
r
k
s
,
d
ata
m
in
in
g
,
an
d
co
m
p
u
tatio
n
al
i
n
tellig
en
ce
in
t
h
e
lear
n
i
n
g
m
o
d
u
le.
T
o
d
ay
,
DL
m
o
d
els
an
d
ar
tific
ial
in
tellig
en
ce
tech
n
iq
u
es
g
ath
er
a
lo
t
of
in
te
r
est
in
d
esig
n
in
g
th
e
in
tr
u
s
io
n
d
etec
tio
n
m
o
d
el.
On
e
of
th
e
p
r
o
b
lem
s
is
f
ea
tu
r
e
s
elec
tio
n
,
wh
ich
af
f
ec
ts
th
e
en
tire
p
er
f
o
r
m
an
ce
of
th
e
s
y
s
tem
s
.
T
h
er
e
is
a
tr
ad
eo
f
f
b
e
twee
n
s
ec
u
r
ity
an
d
p
er
f
o
r
m
an
ce
m
etr
ics
wh
ile
ch
o
o
s
in
g
an
I
DS
tech
n
iq
u
e.
T
h
er
e
ar
e
d
if
f
er
en
t
t
y
p
es
of
I
DS
f
o
r
wir
eless
s
en
s
o
r
n
etwo
r
k
s
with
r
esp
ec
t
to
ar
ch
itectu
r
al
d
esig
n
.
On
e
is
ce
n
tr
alize
d
,
an
d
th
e
o
th
er
is
d
is
tr
ib
u
ted
[
1
4
]
.
T
o
d
ay
,
m
o
s
t
of
th
e
I
DS
ar
e
d
is
tr
ib
u
ted
wh
er
e
th
e
d
etec
tio
n
is
done
in
a
lo
ca
l
n
o
d
e.
T
h
e
p
r
o
b
lem
is
to
s
p
e
n
d
a
s
ig
n
if
ican
t
en
er
g
y
f
o
r
co
o
r
d
in
atio
n
am
o
n
g
all
n
o
d
es.
Fu
r
t
h
er
,
th
e
n
o
d
es
ar
e
u
n
ab
le
to
d
etec
t
ce
r
tain
attac
k
s
s
in
ce
it
h
as
k
n
o
wled
g
e
ab
o
u
t
its
n
eig
h
b
o
r
h
o
o
d
o
n
ly
.
Sin
g
le
p
o
in
t
f
ailu
r
e
o
cc
u
r
s
in
ce
n
tr
alize
d
I
DS
due
to
co
m
m
u
n
icatio
n
p
r
o
b
lem
s
b
etwe
en
th
e
n
o
d
e
s
th
at
cr
ea
te
lar
g
e
co
m
m
u
n
icatio
n
o
v
e
r
h
ea
d
.
T
h
e
th
ir
d
ty
p
e
of
I
DS
is
th
e
h
y
b
r
id
m
o
d
el,
wh
ic
h
is
a
co
m
b
i
n
atio
n
of
d
is
tr
ib
u
ted
an
d
ce
n
tr
alize
d
I
DS.
L
ea
r
n
in
g
alg
o
r
ith
m
s
lik
e
ML
an
d
DL
ar
e
u
s
ed
in
m
an
y
d
o
m
ain
s
,
s
u
ch
as
h
ea
lth
ca
r
e
an
d
im
ag
e
p
r
o
ce
s
s
in
g
.
I
DS,
u
s
in
g
ML
tech
n
iq
u
es,
lear
n
s
all
k
in
d
s
of
tr
a
f
f
ic.
T
h
e
d
etec
tio
n
p
r
o
ce
s
s
is
t
o
tally
b
ased
on
d
at
a
co
llected
an
d
s
to
r
ed
ce
n
tr
ally
in
th
e
s
er
v
er
.
It
is
f
o
u
n
d
th
at
t
h
e
ac
cu
r
ac
y
d
ec
r
ea
s
es
due
to
a
lar
g
e
d
ata
s
et
with
a
h
ig
h
p
ac
k
et
l
o
s
s
r
ate
[
2
0
]
.
T
h
e
p
r
o
b
lem
s
can
be
a
d
d
r
ess
ed
by
f
ed
er
ated
lear
n
in
g
.
T
h
is
alg
o
r
ith
m
lea
r
n
s
d
ata
g
en
er
ated
by
th
e
d
e
v
ices
in
a
co
llab
o
r
ativ
e
f
ash
io
n
with
o
u
t
an
y
ce
n
tr
alize
d
s
er
v
e
r
[
2
2
]
.
It
wo
r
k
s
with
d
ec
en
tr
alize
d
d
ata
f
r
o
m
d
e
v
i
ce
s
wh
ich
ar
e
co
m
m
u
n
icate
d
in
eith
er
d
ir
ec
tio
n
.
T
h
er
e
ar
e
two
s
tag
es:
lo
ca
l
lear
n
in
g
a
n
d
m
o
d
el
tr
an
s
m
is
s
io
n
,
wh
ic
h
p
e
r
m
it
th
e
ac
co
m
p
lis
h
m
en
t
of
p
r
i
v
ac
y
p
r
eser
v
atio
n
an
d
co
s
t
r
ed
u
ctio
n
.
In
th
e
tr
ad
itio
n
al
m
eth
o
d
of
in
tr
u
s
io
n
d
etec
tio
n
,
all
in
f
o
r
m
ati
o
n
is
m
ain
tain
e
d
in
a
ce
n
tr
aliz
ed
s
er
v
er
an
d
also
tr
an
s
f
er
r
ed
th
is
in
f
o
r
m
atio
n
b
e
twee
n
s
er
v
er
an
d
h
o
s
t,
wh
ich
ar
e
v
u
ln
er
a
b
le
to
m
an
-
in
-
t
h
e
-
m
id
d
le
attac
k
s
[
2
3
]
.
Fed
er
ated
lear
n
in
g
m
eth
o
d
s
wo
r
k
in
a
d
ec
en
t
r
alize
d
m
an
n
er
with
th
is
in
f
o
r
m
atio
n
[
2
4
]
.
S
o
,
it
is
ef
f
icien
t
an
d
en
f
o
r
ce
s
a
p
r
i
v
ac
y
p
o
licy
f
o
r
s
en
s
itiv
e
d
ata.
T
h
er
e
ar
e
n
u
m
er
o
u
s
ap
p
r
o
ac
h
es
to
in
tr
u
s
io
n
d
etec
tio
n
u
s
in
g
f
ed
er
ated
lear
n
in
g
.
Su
n
n
y
et
a
l.
[
2
5
]
p
r
o
p
o
s
es
u
s
in
g
m
im
ic
lear
n
in
g
in
co
m
b
i
n
atio
n
with
f
ed
er
ated
lea
r
n
in
g
to
p
r
o
tect
ag
ain
s
t
r
ev
er
s
e
en
g
i
n
ee
r
in
g
attac
k
s
.
Mo
s
t
ML
an
d
DL
m
o
d
els
s
u
f
f
er
ed
f
r
o
m
f
alse
n
eg
ativ
e
alar
m
s
[
2
6
]
,
[
2
7
]
.
T
h
is
r
esear
ch
p
r
o
p
o
s
ed
h
y
b
r
i
d
in
tr
u
s
io
n
d
etec
tio
n
f
o
r
wir
e
less
s
en
s
o
r
n
etwo
r
k
s
u
s
in
g
a
f
ed
er
ated
lear
n
in
g
alg
o
r
ith
m
.
A
ty
p
ical
ar
tific
ial
n
eu
r
al
n
etwo
r
k
h
as
d
if
f
er
en
t
p
h
ases
.
It
u
s
es
s
u
p
er
v
is
e
d
an
d
u
n
s
u
p
e
r
v
is
ed
tr
ain
in
g
alg
o
r
ith
m
s
.
T
h
e
p
at
ter
n
r
ec
o
g
n
itio
n
p
r
o
b
lem
s
c
an
be
s
o
lv
e
d
by
in
co
r
p
o
r
atin
g
th
e
s
u
p
er
v
is
ed
alg
o
r
ith
m
s
.
T
h
e
class
if
icatio
n
p
r
o
b
lem
s
can
be
s
o
lv
ed
by
i
n
co
r
p
o
r
atin
g
u
n
s
u
p
er
v
is
ed
alg
o
r
ith
m
s
wh
er
e
th
e
n
etwo
r
k
lear
n
s
with
o
u
t
th
e
k
n
o
wled
g
e
of
th
e
d
esire
d
o
u
tp
u
t
[
2
8
]
,
[
2
9
]
.
T
h
e
s
ig
n
if
ican
ce
of
th
e
n
eu
r
al
n
etwo
r
k
is
th
at
it
r
ep
ea
te
d
ly
lear
n
s
th
e
co
ef
f
icien
ts
.
T
h
e
co
e
f
f
icien
ts
ar
e
ad
ju
s
ted
to
n
o
r
m
al
d
ata
an
d
attac
k
d
ata
d
u
r
in
g
th
e
tr
ain
in
g
p
h
ase.
T
h
e
n
e
u
r
al
n
etwo
r
k
ap
p
r
o
ac
h
im
p
r
o
v
es
th
e
d
etec
tio
n
r
ate,
an
d
th
e
f
alse
alar
m
is
r
ed
u
ce
d
[
3
0
]
,
[
3
1
]
.
2.
M
E
T
H
O
D
T
h
e
p
r
o
p
o
s
ed
m
o
d
el
is
lay
e
r
e
d
an
d
clu
s
ter
ed
with
a
h
y
b
r
id
I
DS
m
o
d
el.
E
ac
h
I
DS
is
p
la
ce
d
on
th
e
clien
t
s
id
e
with
s
en
s
o
r
n
o
d
es.
All
ar
e
o
p
er
ated
in
a
d
is
tr
ib
u
te
d
m
an
n
e
r
.
A
h
y
b
r
i
d
Hier
ar
ch
ic
al
n
etwo
r
k
c
o
n
s
is
ts
of
a
s
en
s
o
r
n
o
d
e
an
d
a
clien
t
wh
ich
h
o
ld
s
l
o
ca
l
I
DS
s
y
s
tem
s
.
It
id
en
tifie
s
th
e
d
ata
an
o
m
aly
with
r
esp
ec
t
to
th
e
s
en
s
o
r
node
an
d
clien
t.
An
o
m
a
ly
is
d
etec
ted
b
ased
on
th
e
m
ea
n
v
ar
ian
c
e
an
d
d
ata
d
is
tr
ib
u
tio
n
is
co
r
r
elate
d
with
s
en
s
o
r
node
an
d
clien
t
lo
ca
lly
.
Af
ter
th
e
s
elec
tio
n
of
clu
s
ter
h
ea
d
s
am
o
n
g
n
o
d
es,
clu
s
ter
-
b
ased
I
DS
(
C
B
I
DS)
is
ac
tiv
ated
in
o
r
d
e
r
to
d
etec
t
d
if
f
er
en
t
t
y
p
es
of
attac
k
s
s
u
ch
as
s
elec
tiv
e
f
o
r
war
d
i
n
g
,
f
lo
o
d
in
g
,
s
elf
is
h
m
is
b
eh
av
io
u
r
,
node
r
e
p
licatio
n
attac
k
s
an
d
s
in
k
h
o
le
attac
k
s
.
All
b
eh
av
io
u
r
an
aly
s
is
is
done
u
s
in
g
f
ed
er
ated
lear
n
in
g
ar
ch
itectu
r
e
.
T
h
e
f
ed
er
ated
lear
n
in
g
-
b
ased
I
DS
f
o
r
wir
eless
s
en
s
o
r
n
etwo
r
k
ar
ch
itectu
r
e
is
g
iv
en
in
Fig
u
r
e
1.
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
.
1
,
Feb
r
u
ar
y
2
0
2
5
:
492
-
4
9
9
494
Fig
u
r
e
1.
Fed
er
ate
d
lear
n
in
g
-
b
ased
I
DS
in
wir
eless
s
en
s
o
r
n
etwo
r
k
T
h
e
p
r
o
p
o
s
ed
ar
ch
itectu
r
e
is
d
iv
id
ed
in
to
two
p
a
r
ts
—
th
e
clien
t
s
id
e
an
d
th
e
s
er
v
er
s
id
e.
T
h
e
clien
t
-
s
id
e
h
as
lo
ca
l
m
o
d
el
d
at
a
ag
g
r
e
g
atio
n
a
n
d
s
en
s
o
r
d
e
v
ices.
T
h
e
s
er
v
er
s
id
e
h
as
a
g
lo
b
a
l
ag
g
r
e
g
ato
r
s
er
v
er
an
d
a
lo
ca
l
m
o
d
el
a
g
g
r
e
g
ato
r
,
wh
ich
lead
s
to
th
e
g
lo
b
al
m
o
d
el.
In
th
is
ar
ch
itect,
ea
c
h
lo
ca
l
clien
t
m
o
d
el
tr
ain
s
th
e
d
ata
ac
q
u
ir
ed
f
r
o
m
s
en
s
o
r
d
ev
ices
with
th
e
lo
ca
l
m
o
d
els
s
h
ar
ed
by
th
e
s
er
v
e
r
.
Fu
r
th
er
,
th
e
I
DS
at
th
e
clien
t
en
d
d
etec
ts
an
y
u
n
wan
ted
attac
k
s
at
th
e
node
lev
el
an
d
clu
s
ter
lev
el.
An
an
al
y
ze
r
is
u
s
ed
to
m
o
n
ito
r
an
d
tr
ac
k
th
eir
n
etwo
r
k
tr
a
f
f
ic
d
ata
as
w
ell
as
node
d
ata
f
o
r
s
u
b
s
eq
u
en
t
an
aly
s
is
.
C
lien
ts
ar
e
tr
ain
e
d
l
o
ca
lly
an
d
g
lo
b
ally
f
o
r
d
ata
a
g
g
r
e
g
atio
n
.
T
h
e
d
et
ec
tio
n
m
o
d
u
le
ag
g
r
eg
ates
all
tr
ain
ed
d
ata
f
r
o
m
clien
ts
an
d
an
aly
s
e
th
e
d
ata
to
ch
ec
k
th
e
ab
n
o
r
m
al
b
eh
a
v
io
r
of
th
e
wir
eless
s
en
s
o
r
n
etwo
r
k
.
T
h
e
u
s
e
of
a
g
l
o
b
al
a
g
g
r
e
g
atio
n
s
er
v
e
r
is
to
tr
an
s
f
o
r
m
l
o
ca
l
lear
n
in
g
in
to
g
lo
b
al
lear
n
i
n
g
.
A
clien
t
is
a
b
le
to
d
etec
t
i
n
tr
u
s
io
n
s
by
co
m
p
ar
in
g
b
eh
a
v
io
r
s
o
b
tain
ed
f
r
o
m
g
lo
b
al
lear
n
in
g
an
d
im
p
r
o
v
es
th
e
d
etec
tio
n
.
2
.
1
.
I
m
ple
m
ent
a
t
io
n
T
h
e
p
r
o
p
o
s
ed
wo
r
k
f
ir
s
t
clien
t
tr
ain
s
a
lo
ca
l
d
ataset
an
d
th
en
s
h
ar
es
th
e
d
ata
with
a
g
lo
b
al
ag
g
r
eg
ato
r
r
ath
er
th
an
on
a
ce
n
tr
al
s
er
v
e
r
.
T
h
e
g
lo
b
al
ag
g
r
eg
atio
n
s
er
v
er
in
ter
ac
ts
with
all
clien
ts
an
d
lo
o
k
s
f
o
r
lo
ca
l
I
DS
m
o
d
els.
It
cr
e
ates
an
u
p
d
ated
g
lo
b
al
m
o
d
el
with
all
clien
t'
s
I
DS
m
o
d
els
with
o
p
tim
al
p
ar
a
m
eter
s
.
T
h
e
e
q
u
atio
n
u
s
es
th
e
s
tar
tin
g
weig
h
ts
(
w)
an
d
n
u
m
b
e
r
of
f
ed
er
ate
d
lea
r
n
in
g
r
o
u
n
d
s
(
R
)
;
th
e
co
n
v
er
g
en
ce
lev
el
ca
n
be
ac
h
iev
ed
by
c
h
an
g
in
g
weig
h
ts
an
d
n
u
m
b
er
of
f
ed
er
ated
lear
n
i
n
g
r
o
u
n
d
s
ag
ain
an
d
ag
ain
.
At
r
o
u
n
d
t,
each
lo
ca
l
clien
t’
s
weig
h
t
is
co
m
m
u
n
icat
ed
an
d
u
p
d
ated
to
th
e
a
g
g
r
eg
at
io
n
s
er
v
er
(
1
)
is
u
s
ed
f
r
o
m
th
e
Fed
Av
g
alg
o
r
it
h
m
[
2
8
]
to
u
p
d
ate
th
e
m
o
d
el
weig
h
ts
.
+1
=
∑
=1
/
t+
1
(
1
)
W
h
er
e
is
th
e
to
tal
s
ize
of
a
ll
clien
t
d
ataset
s
,
an
d
r
ep
r
esen
ts
th
e
s
ize
of
each
clien
t
d
ataset.
t
+
1
is
th
e
u
p
d
ated
g
lo
b
al
m
o
d
el
af
ter
th
e
iter
atio
n
.
2
.
2
.
Alg
o
rit
hm
f
o
r
lo
ca
l
intr
us
io
n det
ec
t
io
n sy
s
t
em
on
t
he
client
s
ide
T
h
e
alg
o
r
ith
m
f
o
r
th
e
lo
ca
l
I
DS
on
t
h
e
clien
t
s
id
e
is
s
u
m
m
ar
ized
in
th
is
s
ec
tio
n
.
T
h
e
f
o
llo
win
g
alg
o
r
ith
m
is
im
p
lem
en
te
d
on
e
ac
h
clien
t
s
id
e,
an
d
t
h
e
lo
ca
l
a
n
aly
ze
r
ev
alu
ates
s
en
s
e
d
ata
f
o
r
ab
n
o
r
m
alities
.
Step
1:
C
h
ec
k
th
e
s
en
s
o
r
d
ata.
cr
ea
te
a
tab
le
an
d
s
to
r
e
it
Step
2:
T
ak
e
th
e
tab
le
.
C
h
ec
k
th
e
s
ize
an
d
co
m
p
ar
e
it
with
a
th
r
esh
o
ld
.
C
o
m
p
u
te
v
ar
ia
n
ce
Step
3:
C
o
m
p
u
te
ab
n
o
r
m
alitie
s
in
th
e
tab
le.
C
h
ec
k
th
e
c
o
n
d
i
tio
n
of
d
ata
an
o
m
al
y
with
a
th
r
esh
o
ld
v
alu
e
Step
4:
Oth
er
wis
e,
d
r
o
p
it.
Fo
r
war
d
to
a
g
l
o
b
al
lead
er
T
h
e
g
lo
b
al
ag
g
r
eg
atio
n
s
er
v
e
r
in
itiates
a
n
eu
r
al
n
etwo
r
k
m
o
d
el
(
NNM
)
f
r
o
m
a
g
l
o
b
al
in
tr
u
s
io
n
d
etec
tio
n
m
o
d
el.
E
ac
h
u
s
es
th
e
g
lo
b
al
m
o
d
el.
E
ac
h
clien
t
cr
ea
tes
lo
ca
l
weig
h
ts
w
ith
th
eir
p
r
iv
ate
d
ata,
an
d
each
clien
t
ca
lcu
lates
a
f
r
esh
s
et
of
lo
ca
l
weig
h
ts
an
d
wo
r
k
s
p
ar
allelly
with
th
e
g
lo
b
al
m
o
d
el.
T
h
e
clien
ts
u
s
e
s
en
s
o
r
d
ata
co
llected
lo
ca
lly
an
d
an
aly
s
e
with
lo
ca
l
an
aly
s
er
.
T
h
e
lo
ca
l
clien
t
m
o
d
el
wo
r
k
s
with
th
e
g
lo
b
al
ag
g
r
eg
atio
n
m
o
d
el
in
o
r
d
er
to
im
p
r
o
v
e
th
e
I
DS
an
d
co
m
m
u
n
icate
with
th
e
g
lo
b
al
ag
g
r
eg
atio
n
s
er
v
e
r
.
T
h
e
g
lo
b
al
ag
g
r
eg
atio
n
s
er
v
er
ad
a
p
ts
ch
an
g
es
r
ec
eiv
e
d
f
r
o
m
lo
ca
l
clien
ts
a
n
d
ad
d
s
t
h
e
weig
h
ts
f
r
o
m
th
e
v
ar
i
o
u
s
l
o
ca
l
n
o
d
e
m
o
d
els
in
o
r
d
er
to
p
r
o
d
u
ce
a
n
ew,
im
p
r
o
v
ed
m
o
d
el
(
1
)
.
T
h
e
p
ar
am
eter
s
ar
e
ev
alu
ated
on
th
e
b
asis
of
d
ataset
s
ize
at
ev
er
y
n
o
d
e.
On
ce
ag
ain
,
u
p
d
ated
m
o
d
el
p
ar
am
ete
r
s
ar
e
co
m
m
u
n
i
ca
ted
with
clien
ts
f
o
r
th
e
c
h
an
g
es
th
at
o
cc
u
r
r
e
d
in
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
Hyb
r
id
in
tr
u
s
io
n
d
etec
tio
n
mo
d
el
fo
r
h
iera
r
ch
ica
l wi
r
eles
s
s
en
s
o
r
n
etw
o
r
k
…
(
S
a
th
is
h
ku
m
a
r
M
ani
)
495
th
e
ce
n
tr
alize
d
s
er
v
er
.
E
v
e
r
y
c
lien
t
r
o
u
tin
e
th
e
n
o
v
el
class
ica
l
p
ar
am
eter
s
an
d
m
ak
es
v
ar
iati
o
n
s
to
th
em
b
ased
on
th
e
n
o
v
el
d
ata.
T
h
e
p
r
o
ce
s
s
is
r
ep
ea
ted
f
o
r
th
e
im
p
r
o
v
em
e
n
t
of
lear
n
in
g
.
Data
p
ac
k
et
in
f
o
r
m
ati
o
n
is
g
i
v
en
in
th
e
in
p
u
t
of
a
n
eu
r
al
n
etwo
r
k
,
wh
ich
h
as
o
n
e
in
p
u
t
lay
er
,
one
h
id
d
en
lay
e
r
an
d
an
o
u
tp
u
t
lay
er
.
T
h
is
n
eu
r
al
n
etwo
r
k
-
b
ased
I
DS
is
im
p
lem
en
ted
in
clu
s
ter
h
ea
d
f
o
r
d
etec
tin
g
th
e
attac
k
s
.
So
,
f
o
u
r
n
e
u
r
o
n
s
h
av
e
b
ee
n
g
iv
en
to
th
e
i
n
p
u
t
of
th
e
n
e
u
r
al
n
etwo
r
k
.
No
r
m
al
an
d
ab
n
o
r
m
al
co
n
d
itio
n
is
cr
ea
ted
in
th
e
tr
ai
n
in
g
p
h
ase.
C
en
tr
e
of
ac
tiv
atio
n
f
u
n
ctio
n
an
d
s
p
r
ea
d
f
ac
to
r
is
in
itiated
an
d
t
h
e
s
p
r
ea
d
f
ac
to
r
.
Per
f
o
r
m
a
n
ce
is
to
tally
b
ased
on
a
n
u
m
b
er
of
p
ar
a
m
eter
s
in
a
n
e
u
r
al
n
etwo
r
k
.
Hid
d
en
lay
er
p
ar
am
eter
s
an
d
r
ad
iu
s
of
th
e
R
DF
f
u
n
ctio
n
p
la
y
cr
u
cial
f
u
n
ct
io
n
s
in
th
e
p
e
r
f
o
r
m
an
ce
of
th
e
I
DS
s
y
s
tem
.
2
.
3
.
Alg
o
rit
hm
f
o
r
g
lo
ba
l
int
rus
io
n det
ec
t
io
n sy
s
t
em
T
h
e
alg
o
r
ith
m
f
o
r
t
h
e
g
lo
b
al
I
DS
is
s
u
m
m
ar
ized
in
th
is
s
ec
ti
o
n
.
T
h
e
i
n
p
u
t
p
ar
am
eter
s
to
th
e
f
ed
er
ated
lear
n
in
g
s
tr
u
ctu
r
e
ar
e
g
i
v
en
.
D
if
f
er
en
t
ty
p
es
of
attac
k
s
can
be
d
etec
ted
.
Step
1:
I
n
itialize
s
elf
-
o
r
g
a
n
is
atio
n
m
ap
p
ar
am
eter
s
Step
2:
I
n
itialize
th
e
weig
h
ts
f
o
r
th
e
n
e
u
r
al
n
etwo
r
k
Step
3:
C
o
m
p
u
te
th
e
o
u
tp
u
t
of
ev
er
y
node
a
n
d
e
r
r
o
r
Fu
n
ctio
n
4
.
Step
4:
C
h
ec
k
th
e
co
n
d
itio
n
f
o
r
er
r
o
r
Step
5:
Oth
er
wis
e,
u
p
d
ate
th
e
s
elf
-
o
r
g
an
izatio
n
p
ar
am
eter
s
a
n
d
weig
h
ts
of
t
h
e
n
eu
r
al
n
etw
o
r
k
.
C
alcu
late
th
e
o
u
tp
u
t
of
ev
e
r
y
n
o
d
e.
T
h
e
d
etec
tio
n
of
attac
k
a
n
d
te
ch
n
iq
u
es
is
tab
u
lated
in
T
ab
le
1.
T
ab
le
1.
Dete
ctio
n
a
n
d
t
ec
h
n
i
q
u
es
I
D
S
Te
c
h
n
i
q
u
e
C
B
I
D
S
F
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
met
h
o
d
s
S
B
I
D
S
F
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
met
h
o
d
s
N
B
I
D
S
R
u
l
e
-
b
a
se
d
3.
RE
SU
L
T
S
AND
D
I
SCU
SS
I
O
N
T
en
s
o
r
f
lo
w
an
d
k
er
as
ar
e
u
s
ed
f
o
r
ML
an
d
DL
.
Simu
latio
n
is
ca
r
r
ied
out
in
n
etwo
r
k
s
im
u
lato
r
v
er
s
io
n
2
(
NS2
)
in
o
r
d
er
to
e
x
tr
ac
t
wir
eless
s
en
s
o
r
n
etwo
r
k
p
ar
am
ete
r
s
.
Simu
latio
n
p
a
r
a
m
eter
s
ar
e
lis
ted
in
T
ab
le
2.
Data
s
ets
ar
e
g
en
er
ated
f
r
o
m
NS2
an
d
s
en
s
o
r
s
to
tr
a
in
an
d
test
I
DS.
T
h
e
test
b
ed
was
cr
ea
ted
f
o
r
two
ty
p
es
of
attac
k
s
wh
ich
a
r
e
d
is
t
r
ib
u
ted
d
en
ial
-
of
-
s
er
v
ice
(
DDo
S
)
attac
k
s
an
d
m
an
-
in
-
th
e
-
m
id
d
le
(
MI
M
)
attac
k
s
.
I
n
itially
,
th
e
d
ata
is
p
r
ep
r
o
ce
s
s
ed
.
T
h
e
f
ea
tu
r
e
s
elec
tio
n
ap
p
r
o
ac
h
es
wer
e
ap
p
lied
to
r
e
d
u
ce
tr
ain
in
g
a
n
d
class
if
icatio
n
tim
e.
T
ab
le
3
d
is
p
lay
s
th
e
s
elec
ted
d
ata
f
o
r
ML
m
o
d
els.
T
h
e
test
s
wer
e
p
e
r
f
o
r
m
ed
u
s
in
g
f
e
d
er
ated
lear
n
in
g
.
T
h
e
ef
f
icac
y
of
f
ed
er
ated
lear
n
in
g
was
ev
al
u
ated
w
ith
2
or
3
clien
ts
.
T
ab
le
2.
Simu
latio
n
p
ar
am
eter
s
P
a
r
a
me
t
e
r
s
V
a
l
u
e
s
A
r
e
a
6
0
0
×
6
0
0
N
o
d
e
s
(
N
u
m
b
e
r
)
1
0
0
S
i
mu
l
a
t
i
o
n
t
i
me
in
se
c
o
n
d
s
1
0
0
P
r
o
t
o
c
o
l
f
o
r
r
o
u
t
i
n
g
H
I
D
S
En
e
r
g
y
(
Jo
u
l
e
s)
1
0
0
I
n
t
e
r
v
a
l
4
to
6
N
u
mb
e
r
of
a
t
t
a
c
k
e
r
s
-
4
P
a
c
k
e
t
S
i
z
e
(
b
y
t
e
s)
50
to
1
0
0
T
ab
le
3.
Data
s
et
f
o
r
ML
m
o
d
el
Ty
p
e
To
t
a
l
Tr
a
i
n
Te
st
N
o
r
mal
1
1
,
2
2
2
8
,
8
5
6
2
,
4
6
6
D
D
o
S
_
U
D
P
A
t
t
a
c
k
5
,
5
0
8
4
,
4
7
8
3
,
2
9
9
D
D
o
S
_
I
C
M
P
A
t
t
a
c
k
8
,
1
9
5
6
,
9
8
9
3
,
9
5
6
D
D
o
S
_
H
TTP
A
t
t
a
c
k
9
,
7
8
9
7
,
2
2
1
3
,
7
7
7
D
D
o
S
_
T
C
P
A
t
t
a
c
k
8
,
3
5
8
6
,
1
3
6
3
,
5
4
6
M
I
M
A
t
t
a
c
k
1
,
7
2
5
1
,
2
3
8
6
7
4
3
.
1
.
P
er
f
o
r
m
a
t
io
n
e
v
a
lua
t
io
n
Netwo
r
k
p
er
f
o
r
m
a
n
ce
p
ar
a
m
eter
s
wer
e
an
aly
ze
d
with
d
if
f
er
e
n
t
p
ac
k
et
s
ize
e
a
n
d
in
ter
v
al.
Per
f
o
r
m
an
ce
is
m
ea
s
u
r
ed
with
two
p
ar
a
m
eter
s
.
T
h
e
y
ar
e
th
e
d
etec
tio
n
r
atio
an
d
f
alse
p
o
s
iti
v
e
r
ate.
T
h
e
g
r
ap
h
s
ar
e
an
aly
ze
d
in
Fig
u
r
e
2.
Fig
u
r
e
2
(
a)
d
ep
icts
th
e
p
ac
k
et
s
ize
v
s
jitt
er
,
Fig
u
r
e
2
(
b
)
d
ep
ict
s
th
e
p
ac
k
et
s
ize
v
s
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
.
1
,
Feb
r
u
ar
y
2
0
2
5
:
492
-
4
9
9
496
p
ac
k
ets
d
r
o
p
p
ed
,
Fig
u
r
e
2
(
c)
d
ep
icts
th
e
in
ter
v
al
v
s
p
ac
k
ets
d
r
o
p
p
e
d
,
Fig
u
r
e
2
(
d
)
d
e
p
icts
th
e
in
ter
v
al
v
s
th
r
o
u
g
h
p
u
t,
Fig
u
r
e
2
(
e)
d
e
p
icts
th
e
attac
k
er
v
s
th
r
o
u
g
h
p
u
t,
an
d
Fig
u
r
e
2
(
f
)
d
ep
icts
th
e
jitt
er
v
s
attac
k
er
.
T
h
e
r
esu
lts
s
h
o
w
an
im
p
r
o
v
em
en
t
in
th
e
p
er
f
o
r
m
an
ce
of
th
e
an
o
m
aly
d
etec
tio
n
p
er
f
o
r
m
an
ce
s
y
s
t
em
with
a
r
ed
u
ctio
n
of
f
alse
p
o
s
itiv
e
r
ate
an
d
h
ig
h
d
etec
tio
n
r
atio
.
It
s
h
o
ws
th
e
p
e
r
f
o
r
m
a
n
ce
g
r
ap
h
f
o
r
h
y
b
r
id
I
D
S
ar
ch
itectu
r
e.
I
DS
p
er
f
o
r
m
an
ce
can
be
ev
alu
ate
d
by
th
e
f
o
llo
win
g
p
a
r
am
eter
s
.
C
o
r
r
ec
t
p
o
s
itiv
e
(
C
P):
th
e
n
u
m
b
er
of
attac
k
s
am
p
les
out
is
d
iv
id
e
d
by
ac
cu
r
ately
d
etec
ted
attac
k
s
in
th
e
to
tal
s
a
m
p
les.
Un
tr
u
e
p
o
s
itiv
e
(
UP)
:
t
h
e
n
u
m
b
er
of
n
o
r
m
al
s
am
p
les
is
d
iv
id
ed
by
i
n
co
r
r
ec
tly
id
en
tifie
d
as
attac
k
s
in
th
e
n
o
r
m
al
s
am
p
les.
C
o
r
r
ec
t
n
eg
ativ
e
(
C
N)
:
th
e
n
u
m
b
er
of
b
en
ig
n
s
am
p
les
is
d
iv
id
ed
ac
cu
r
ately
an
d
class
if
i
ed
as
n
o
r
m
al.
Un
tr
u
e
n
eg
ativ
e
(
UN)
:
th
e
n
u
m
b
er
of
attac
k
s
am
p
les
is
d
iv
id
ed
by
wr
o
n
g
ly
r
ec
o
g
n
ized
as
n
o
r
m
al.
(
a)
(
b
)
(
c)
(
d
)
(
e)
(f)
Fig
u
r
e
2
.
Per
f
o
r
m
an
c
e
g
r
a
p
h
f
o
r
d
ata
a
n
o
m
aly
in
NB
I
DS:
(
a)
p
ac
k
et
s
ize
v
s
jitt
e,
(
b
)
p
ac
k
e
t size
v
s
p
ac
k
ets
d
r
o
p
p
ed
,
(
c
)
in
ter
v
al
v
s
p
ac
k
et
s
d
r
o
p
p
e
d
,
(
d
)
in
ter
v
al
v
s
th
r
o
u
g
h
p
u
t,
(
e)
attac
k
er
v
s
th
r
o
u
g
h
p
u
t,
a
n
d
(
f
)
jitt
er
v
s
attac
k
er
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
Hyb
r
id
in
tr
u
s
io
n
d
etec
tio
n
mo
d
el
fo
r
h
iera
r
ch
ica
l wi
r
eles
s
s
en
s
o
r
n
etw
o
r
k
…
(
S
a
th
is
h
ku
m
a
r
M
ani
)
497
T
h
e
p
er
f
o
r
m
an
ce
m
et
r
ics
ar
e
ev
alu
ated
by
two
p
ar
am
eter
s
.
Dete
ctio
n
r
atio
:
th
e
r
atio
b
e
twee
n
th
e
n
u
m
b
er
of
c
o
r
r
ec
tly
i
d
en
tifie
d
attac
k
s
an
d
e
x
p
ec
ted
attac
k
s
.
Fals
e
alar
m
r
ate:
it
is
th
e
r
atio
b
etwe
en
th
e
id
en
tific
atio
n
of
n
o
r
m
al
s
am
p
l
es
as
attac
k
with
n
o
r
m
al
s
am
p
les.
Fig
u
r
es
3
(
a)
an
d
3
(
b
)
s
h
o
ws
th
e
p
er
f
o
r
m
an
ce
g
r
ap
h
f
o
r
h
y
b
r
i
d
I
DS
a
r
ch
itect
u
r
e.
T
ab
le
4
s
h
o
ws
th
e
r
esu
lts
of
ML
ap
p
r
o
ac
h
es
f
o
r
a
ce
n
tr
al
ized
m
o
d
el
in
ter
m
s
of
d
etec
tio
n
r
atio
.
T
h
is
tab
le
g
iv
es
in
f
o
r
m
atio
n
ab
o
u
t
how
I
DS
d
if
f
er
en
tiates
attac
k
s
an
d
b
en
ig
n
class
es
in
th
e
d
ataset.
T
h
e
d
etec
tio
n
r
atio
f
o
r
R
NN
an
d
C
NN
ap
p
r
o
ac
h
es
is
r
ea
ch
ed
at
p
ea
k
v
alu
es
of
93%
an
d
9
5
%,
r
esp
ec
tiv
ely
.
T
ab
le
5
s
h
o
ws
th
e
co
m
p
ar
is
o
n
of
th
e
d
etec
tio
n
r
atio
in
g
lo
b
al
m
o
d
els.
(
a)
(
b
)
Fig
u
r
e
3
.
Per
f
o
r
m
an
c
e
g
r
a
p
h
f
o
r
h
y
b
r
id
I
DS a
r
ch
itectu
r
e
(
a
)
attac
k
er
v
s
f
alse p
o
s
itiv
e
an
d
(
b
)
attac
k
er
v
s
d
etec
tio
n
r
atio
T
ab
le
4
.
I
n
tr
u
s
io
n
d
etec
tio
n
at
g
lo
b
al
ag
g
r
eg
atio
n
s
er
v
er
C
l
a
s
s
D
e
t
e
c
t
i
o
n
r
a
t
i
o
C
N
N
R
N
N
N
o
r
mal
0
.
9
3
0
.
9
5
D
D
o
S
_
U
D
P
A
t
t
a
c
k
0
.
8
8
0
.
8
9
D
D
o
S
_
I
C
M
P
A
t
t
a
c
k
0
.
8
0
0
.
8
1
D
D
o
S
_
H
TTP
A
t
t
a
c
k
0
.
6
0
0
.
5
5
D
D
o
S
_
T
C
P
A
t
t
a
c
k
0
.
9
3
0
.
9
4
M
I
M
A
t
t
a
c
k
0
.
9
3
0
.
9
5
T
ab
le
5
C
o
m
p
ar
is
o
n
o
f
d
etec
tio
n
r
atio
C
l
a
s
si
f
i
e
r
C
l
i
e
n
t
s
F
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
mo
d
e
l
d
e
t
e
c
t
i
o
n
r
a
t
i
o
C
N
N
2
6
4
.
2
3
3
6
1
.
8
8
R
N
N
2
6
0
.
3
9
3
6
1
.
4
7
4.
CO
NCLU
SI
O
N
T
h
e
I
DS
m
o
d
el
is
p
r
o
p
o
s
ed
in
th
is
r
esear
ch
wo
r
k
u
s
in
g
f
ed
er
ated
lear
n
in
g
f
o
r
wir
el
ess
s
en
s
o
r
n
etwo
r
k
s
.
T
h
r
ee
I
DS
m
o
d
els
h
av
e
b
ee
n
d
esig
n
ed
at
th
r
ee
le
v
els.
On
e
is
at
th
e
s
en
s
o
r
s
id
e,
th
e
s
ec
o
n
d
is
at
th
e
clien
t
b
ase,
an
d
th
e
last
is
at
th
e
g
lo
b
al
a
g
g
r
e
g
atio
n
s
er
v
e
r
s
id
e.
T
h
r
ee
I
DS
aim
s
to
d
etec
t
d
if
f
er
en
t
ty
p
es
of
attac
k
s
u
s
in
g
a
f
e
d
er
ated
lear
n
in
g
m
o
d
el.
It
is
o
b
s
er
v
e
d
th
at
th
e
h
y
b
r
id
I
DS
m
o
d
el
u
s
in
g
f
ed
er
ated
lear
n
i
n
g
g
iv
es
a
h
ig
h
d
etec
tio
n
r
atio
a
b
o
v
e
92
%
—
an
d
a
lo
w
f
alse
p
o
s
itiv
e
r
ate.
Fu
r
th
er
,
I
DS
can
ac
h
iev
e
a
v
er
y
lo
w
f
alse
p
o
s
itiv
e
r
ate
by
c
h
an
g
in
g
th
e
p
ar
am
eter
s
in
t
h
e
f
ed
e
r
ate
d
lear
n
in
g
m
o
d
el.
RE
F
E
R
E
NC
E
S
[
1
]
Y
.
M
a
l
e
h
a
n
d
A
.
Ez
z
a
t
i
,
“
A
r
e
v
i
e
w
o
f
sec
u
r
i
t
y
a
t
t
a
c
k
s
a
n
d
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
sc
h
e
m
e
s
i
n
w
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
Wi
r
e
l
e
ss &
M
o
b
i
l
e
N
e
t
w
o
r
k
s
,
v
o
l
.
5
,
n
o
.
6
,
p
p
.
7
9
–
9
0
,
2
0
1
3
,
d
o
i
:
1
0
.
5
1
2
1
/
i
j
w
m
n
.
2
0
1
3
.
5
6
0
6
.
[
2
]
H
.
S
e
d
j
e
l
ma
c
i
a
n
d
S
.
M
.
S
e
n
o
u
c
i
,
“
A
l
i
g
h
t
w
e
i
g
h
t
h
y
b
r
i
d
se
c
u
r
i
t
y
f
r
a
mew
o
r
k
f
o
r
w
i
r
e
l
e
ss
se
n
s
o
r
n
e
t
w
o
r
k
s,
”
2
0
1
4
I
E
EE
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
m
u
n
i
c
a
t
i
o
n
s (I
C
C
)
,
S
y
d
n
e
y
,
A
u
st
r
a
l
i
a
,
2
0
1
4
,
p
p
.
3
6
3
6
-
3
6
4
1
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
.
2
0
1
4
.
6
8
8
3
8
8
6
.
[
3
]
Z.
Y
a
n
g
,
N
.
M
e
r
a
t
n
i
a
,
a
n
d
P
.
H
a
v
i
n
g
a
,
“
A
n
o
n
l
i
n
e
o
u
t
l
i
e
r
d
e
t
e
c
t
i
o
n
t
e
c
h
n
i
q
u
e
f
o
r
w
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
s
u
si
n
g
u
n
s
u
p
e
r
v
i
s
e
d
q
u
a
r
t
e
r
-
s
p
h
e
r
e
su
p
p
o
r
t
v
e
c
t
o
r
ma
c
h
i
n
e
,
”
i
n
2
0
0
8
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
I
n
t
e
l
l
i
g
e
n
t
S
e
n
s
o
rs,
S
e
n
so
r
N
e
t
w
o
rks
a
n
d
I
n
f
o
rm
a
t
i
o
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
.
1
,
Feb
r
u
ar
y
2
0
2
5
:
492
-
4
9
9
498
Pro
c
e
ssi
n
g
,
2
0
0
8
,
p
p
.
1
5
1
–
1
5
6
,
d
o
i
:
1
0
.
1
1
0
9
/
I
S
S
N
I
P
.
2
0
0
8
.
4
7
6
1
9
7
8
.
[
4
]
S
.
R
a
j
a
s
e
g
a
r
a
r
,
C
.
L
e
c
k
i
e
,
M
.
P
a
l
a
n
i
sw
a
mi
,
a
n
d
J.
C
.
B
e
z
d
e
k
,
“
Q
u
a
r
t
e
r
s
p
h
e
r
e
b
a
s
e
d
d
i
s
t
r
i
b
u
t
e
d
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
i
n
w
i
r
e
l
e
ss
sen
s
o
r
n
e
t
w
o
r
k
s,”
2
0
0
7
I
EEE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
G
l
a
s
g
o
w
,
U
K
,
2
0
0
7
,
p
p
.
3
8
6
4
-
3
8
6
9
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
.
2
0
0
7
.
6
3
7
.
.
[
5
]
N
.
S
h
a
h
i
d
,
I
.
H
.
N
a
q
v
i
,
a
n
d
S
.
B
.
Q
a
i
sar,
“
Q
u
a
r
t
e
r
-
s
p
h
e
r
e
S
V
M
:
A
t
t
r
i
b
u
t
e
a
n
d
s
p
a
t
i
o
-
t
e
m
p
o
r
a
l
c
o
r
r
e
l
a
t
i
o
n
s
b
a
se
d
o
u
t
l
i
e
r
&
e
v
e
n
t
d
e
t
e
c
t
i
o
n
i
n
w
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
s,
”
I
EEE
W
i
rel
e
ss
C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
N
e
t
w
o
rk
i
n
g
C
o
n
f
e
r
e
n
c
e
,
W
C
N
C
,
p
p
.
2
0
4
8
–
2
0
5
3
,
2
0
1
2
,
d
o
i
:
1
0
.
1
1
0
9
/
W
C
N
C
.
2
0
1
2
.
6
2
1
4
1
2
7
.
[
6
]
Y
.
Zh
a
n
g
,
N
.
A
.
S
.
H
a
mm
,
N
.
M
e
r
a
t
n
i
a
,
A
.
S
t
e
i
n
,
M
.
V.
D.
V
o
o
r
t
,
a
n
d
P
.
J.
M
.
H
a
v
i
n
g
a
,
“
S
t
a
t
i
st
i
c
s
-
b
a
se
d
o
u
t
l
i
e
r
d
e
t
e
c
t
i
o
n
f
o
r
w
i
r
e
l
e
s
s
se
n
s
o
r
n
e
t
w
o
r
k
s,”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
G
e
o
g
ra
p
h
i
c
a
l
I
n
f
o
rm
a
t
i
o
n
S
c
i
e
n
c
e
,
v
o
l
.
2
6
,
n
o
.
8
,
p
p
.
1
3
7
3
–
1
3
9
2
,
2
0
1
2
,
d
o
i
:
1
0
.
1
0
8
0
/
1
3
6
5
8
8
1
6
.
2
0
1
2
.
6
5
4
4
9
3
.
[
7
]
M
.
A
.
R
a
ss
a
m,
A
.
Za
i
n
a
l
,
a
n
d
M
.
A
.
M
a
a
r
o
f
,
“
O
n
e
-
c
l
a
ss
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
c
l
a
ssi
f
i
e
r
f
o
r
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
i
n
w
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
,
”
i
n
2
0
1
2
Fo
u
rt
h
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
p
u
t
a
t
i
o
n
a
l
As
p
e
c
t
s
o
f
S
o
c
i
a
l
N
e
t
w
o
rks
(
C
A
S
o
N
)
,
2
0
1
2
,
p
p
.
2
7
1
–
276
,
d
o
i
:
1
0
.
1
1
0
9
/
C
A
S
o
N
.
2
0
1
2
.
6
4
1
2
4
1
4
.
[
8
]
J.
W
e
n
g
,
Y
.
Z
h
a
n
g
,
a
n
d
W
.
S
.
H
w
a
n
g
,
“
C
a
n
d
i
d
c
o
v
a
r
i
a
n
c
e
-
f
r
e
e
i
n
c
r
e
me
n
t
a
l
p
r
i
n
c
i
p
a
l
c
o
m
p
o
n
e
n
t
a
n
a
l
y
si
s
,
”
I
EE
E
T
ra
n
s
a
c
t
i
o
n
s
o
n
Pa
t
t
e
r
n
An
a
l
y
si
s
a
n
d
Ma
c
h
i
n
e
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
2
5
,
n
o
.
8
,
p
p
.
1
0
3
4
–
1
0
4
0
,
2
0
0
3
,
d
o
i
:
1
0
.
1
1
0
9
/
TPA
M
I
.
2
0
0
3
.
1
2
1
7
6
0
9
.
[
9
]
M
.
A
.
R
a
ssam
,
A
.
Z
a
i
n
a
l
,
a
n
d
M
.
A
.
M
a
a
r
o
f
,
“
A
n
e
f
f
i
c
i
e
n
t
d
i
st
r
i
b
u
t
e
d
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
m
o
d
e
l
f
o
r
w
i
r
e
l
e
ss
s
e
n
s
o
r
n
e
t
w
o
r
k
s,
”
AAS
RI
Pr
o
c
e
d
i
a
,
v
o
l
.
5
,
p
p
.
9
–
1
4
,
2
0
1
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
a
a
sr
i
.
2
0
1
3
.
1
0
.
0
5
2
.
[
1
0
]
Y
.
Li
u
,
“
Q
P
S
O
-
o
p
t
i
mi
z
e
d
R
B
F
n
e
u
r
a
l
n
e
t
w
o
r
k
f
o
r
n
e
t
w
o
r
k
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
,
”
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
a
n
d
C
o
m
p
u
t
a
t
i
o
n
a
l
S
c
i
e
n
c
e
,
v
o
l
.
8
,
n
o
.
9
,
p
p
.
1
4
7
9
–
1
4
8
5
,
2
0
1
1
.
[
1
1
]
Y
.
P
e
n
g
,
Y
.
W
a
n
g
,
Y
.
N
i
u
,
a
n
d
Q
.
H
u
,
“
A
p
p
l
i
c
a
t
i
o
n
st
u
d
y
o
n
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
u
si
n
g
I
R
B
F
,
”
J
o
u
r
n
a
l
o
f
S
o
f
t
w
a
re
,
v
o
l
.
9
,
n
o
.
1
,
p
p
.
1
7
7
–
1
8
3
,
2
0
1
4
,
d
o
i
:
1
0
.
4
3
0
4
/
j
sw
.
9
.
1
.
1
7
7
-
1
8
3
.
[
1
2
]
P
.
Y
i
c
h
u
n
,
N
.
Y
i
,
a
n
d
H
.
Q
i
w
e
i
,
“
R
e
s
e
a
r
c
h
o
n
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m
b
a
s
e
d
o
n
I
R
B
F
,
”
i
n
2
0
1
2
Ei
g
h
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
a
t
i
o
n
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
S
e
c
u
r
i
t
y
,
2
0
1
2
,
p
p
.
5
4
4
–
5
4
8
,
d
o
i
:
1
0
.
1
1
0
9
/
C
I
S
.
2
0
1
2
.
1
2
8
.
[
1
3
]
S
.
D
e
v
a
r
a
j
u
a
n
d
S
.
R
a
ma
k
r
i
s
h
n
a
n
,
“
P
e
r
f
o
r
ma
n
c
e
a
n
a
l
y
si
s
o
f
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
s
t
e
m
u
si
n
g
v
a
r
i
o
u
s
n
e
u
r
a
l
n
e
t
w
o
r
k
c
l
a
ss
i
f
i
e
r
s
,
”
i
n
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
R
e
c
e
n
t
T
re
n
d
s
i
n
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
I
C
RTI
T
2
0
1
1
,
2
0
1
1
,
p
p
.
1
0
3
3
–
1
0
3
8
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
R
TI
T.
2
0
1
1
.
5
9
7
2
2
8
9
.
[
1
4
]
M
.
R
i
e
c
k
e
r
,
S
.
B
i
e
d
e
r
m
a
n
n
,
R
.
El
B
a
n
sark
h
a
n
i
,
a
n
d
M
.
H
o
l
l
i
c
k
,
“
L
i
g
h
t
w
e
i
g
h
t
e
n
e
r
g
y
c
o
n
s
u
mp
t
i
o
n
-
b
a
se
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
sy
s
t
e
m
f
o
r
w
i
r
e
l
e
ss
se
n
s
o
r
n
e
t
w
o
r
k
s,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
I
n
f
o
rm
a
t
i
o
n
S
e
c
u
r
i
t
y
,
v
o
l
.
1
4
,
n
o
.
2
,
p
p
.
1
5
5
–
1
6
7
,
2
0
1
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
2
0
7
-
0
1
4
-
0
2
4
1
-
1.
[
1
5
]
U
.
H
a
l
i
c
i
,
“
R
a
d
i
a
l
b
a
si
s
f
u
n
c
t
i
o
n
n
e
t
w
o
r
k
,
”
i
n
Ar
t
i
f
i
c
i
a
l
N
e
u
ra
l
N
e
t
w
o
r
k
s
,
A
n
k
a
r
a
,
Tu
r
k
e
y
:
M
E
TU
El
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
s
En
g
i
n
e
e
r
i
n
g
,
2
0
0
5
,
p
p
.
1
3
9
–
1
4
7
.
[
1
6
]
E.
H
o
d
o
,
X
.
B
e
l
l
e
k
e
n
s,
A
.
H
a
m
i
l
t
o
n
,
C
.
T
a
c
h
t
a
t
z
i
s,
a
n
d
R
.
A
t
k
i
n
s
o
n
,
“
S
h
a
l
l
o
w
a
n
d
d
e
e
p
n
e
t
w
o
r
k
s
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
m:
a
t
a
x
o
n
o
my
a
n
d
su
r
v
e
y
,
”
a
rX
i
v
-
C
o
m
p
u
t
e
r S
c
i
e
n
c
e
,
p
p
.
1
–
4
3
,
2
0
1
7
.
[
1
7
]
F
.
Ta
n
g
,
B
.
M
a
o
,
Z
.
M
.
F
a
d
l
u
l
l
a
h
,
a
n
d
N
.
K
a
t
o
,
“
O
n
a
n
o
v
e
l
d
e
e
p
-
l
e
a
r
n
i
n
g
-
b
a
se
d
i
n
t
e
l
l
i
g
e
n
t
p
a
r
t
i
a
l
l
y
o
v
e
r
l
a
p
p
i
n
g
c
h
a
n
n
e
l
a
ssi
g
n
m
e
n
t
i
n
S
D
N
-
I
o
T,
”
I
EE
E
C
o
m
m
u
n
i
c
a
t
i
o
n
s
M
a
g
a
z
i
n
e
,
v
o
l
.
5
6
,
n
o
.
9
,
p
p
.
8
0
–
8
6
,
S
e
p
.
2
0
1
8
,
d
o
i
:
1
0
.
1
1
0
9
/
M
C
O
M
.
2
0
1
8
.
1
7
0
1
2
2
7
.
[
1
8
]
C
.
Zh
a
n
g
,
Y
.
X
i
e
,
H
.
B
a
i
,
B
.
Y
u
,
W
.
L
i
,
a
n
d
Y
.
G
a
o
,
“
A
s
u
r
v
e
y
o
n
f
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
,
”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
st
e
m
s
,
v
o
l
.
2
1
6
,
p
p
.
1
–
1
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s
.
2
0
2
1
.
1
0
6
7
7
5
.
[
1
9
]
V
.
M
o
t
h
u
k
u
r
i
,
P
.
K
h
a
r
e
,
R
.
M
.
P
a
r
i
z
i
,
S
.
P
o
u
r
i
y
e
h
,
A
.
D
e
h
g
h
a
n
t
a
n
h
a
,
a
n
d
G
.
S
r
i
v
a
st
a
v
a
,
“
F
e
d
e
r
a
t
e
d
-
l
e
a
r
n
i
n
g
-
b
a
se
d
a
n
o
mal
y
d
e
t
e
c
t
i
o
n
f
o
r
I
o
T
se
c
u
r
i
t
y
a
t
t
a
c
k
s
,
”
I
EEE
I
n
t
e
r
n
e
t
o
f
T
h
i
n
g
s
J
o
u
r
n
a
l
,
v
o
l
.
9
,
n
o
.
4
,
p
p
.
2
5
4
5
–
2
5
5
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
JI
O
T.
2
0
2
1
.
3
0
7
7
8
0
3
.
[
2
0
]
D
.
K
w
o
n
,
H
.
K
i
m,
J.
K
i
m
,
S
.
C
.
S
u
h
,
I
.
K
i
m,
a
n
d
K
.
J
.
K
i
m,
“
A
s
u
r
v
e
y
o
f
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
n
e
t
w
o
r
k
a
n
o
m
a
l
y
d
e
t
e
c
t
i
o
n
,
”
C
l
u
s
t
e
r
C
o
m
p
u
t
i
n
g
,
v
o
l
.
2
2
,
p
p
.
9
4
9
–
9
6
1
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
0
5
8
6
-
0
1
7
-
1
1
1
7
-
8.
[
2
1
]
A
.
A
l
d
w
e
e
sh
,
A
.
D
e
r
h
a
b
,
a
n
d
A
.
Z.
Emam,
“
D
e
e
p
l
e
a
r
n
i
n
g
a
p
p
r
o
a
c
h
e
s
f
o
r
a
n
o
m
a
l
y
-
b
a
se
d
i
n
t
r
u
si
o
n
d
e
t
e
c
t
i
o
n
s
y
st
e
ms:
a
s
u
r
v
e
y
,
t
a
x
o
n
o
my
,
a
n
d
o
p
e
n
i
ss
u
e
s,”
K
n
o
w
l
e
d
g
e
-
B
a
se
d
S
y
s
t
e
m
s
,
v
o
l
.
1
8
9
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
k
n
o
s
y
s
.
2
0
1
9
.
1
0
5
1
2
4
.
[
2
2
]
B
.
C
e
t
i
n
,
A
.
L
a
z
a
r
,
J.
K
i
m,
A
.
S
i
m,
a
n
d
K
.
W
u
,
“
F
e
d
e
r
a
t
e
d
w
i
r
e
l
e
ss
n
e
t
w
o
r
k
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
,
”
i
n
2
0
1
9
I
EEE
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
Bi
g
D
a
t
a
(
B
i
g
D
a
t
a
)
,
2
0
1
9
,
p
p
.
6
0
0
4
–
6
0
0
6
,
d
o
i
:
1
0
.
1
1
0
9
/
B
i
g
D
a
t
a
4
7
0
9
0
.
2
0
1
9
.
9
0
0
5
5
0
7
.
[
2
3
]
N
.
B
o
u
a
c
i
d
a
a
n
d
P
.
M
o
h
a
p
a
t
r
a
,
“
V
u
l
n
e
r
a
b
i
l
i
t
i
e
s
i
n
f
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
9
,
p
p
.
6
3
2
2
9
–
6
3
2
4
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
1
.
3
0
7
5
2
0
3
.
[
2
4
]
A
.
T
h
a
k
k
a
r
a
n
d
R
.
L
o
h
i
y
a
,
“
A
r
e
v
i
e
w
o
n
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
p
e
r
sp
e
c
t
i
v
e
s
o
f
I
D
S
f
o
r
I
o
T:
r
e
c
e
n
t
u
p
d
a
t
e
s
,
s
e
c
u
r
i
t
y
i
ssu
e
s,
a
n
d
c
h
a
l
l
e
n
g
e
s
,
”
Ar
c
h
i
v
e
s
o
f
C
o
m
p
u
t
a
t
i
o
n
a
l
Me
t
h
o
d
s
i
n
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
2
8
,
n
o
.
4
,
p
p
.
3
2
1
1
–
3
2
4
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
1
8
3
1
-
0
2
0
-
0
9
4
9
6
-
0.
[
2
5
]
A
.
I
.
S
u
n
n
y
,
A
.
Z
h
a
o
,
L
.
Li
,
a
n
d
S
.
K
.
S
a
k
i
l
i
b
a
,
“
L
o
w
-
c
o
st
I
o
T
-
b
a
se
d
s
e
n
s
o
r
sy
s
t
e
m
:
A
c
a
s
e
st
u
d
y
o
n
h
a
r
s
h
e
n
v
i
r
o
n
me
n
t
a
l
mo
n
i
t
o
r
i
n
g
,
”
S
e
n
so
rs
,
v
o
l
.
2
1
,
n
o
.
1
,
p
p
.
1
–
1
2
,
2
0
2
1
,
d
o
i
:
1
0
.
3
3
9
0
/
s
2
1
0
1
0
2
1
4
.
[
2
6
]
N
.
A
.
A
.
Al
-
M
a
r
r
i
,
B
.
S
.
C
i
f
t
l
e
r
,
a
n
d
M
.
M
.
A
b
d
a
l
l
a
h
,
“
F
e
d
e
r
a
t
e
d
mi
mi
c
l
e
a
r
n
i
n
g
f
o
r
p
r
i
v
a
c
y
p
r
e
serv
i
n
g
i
n
t
r
u
s
i
o
n
d
e
t
e
c
t
i
o
n
,
”
i
n
2
0
2
0
I
E
EE
I
n
t
e
r
n
a
t
i
o
n
a
l
Bl
a
c
k
S
e
a
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
a
n
d
N
e
t
w
o
rk
i
n
g
(
Bl
a
c
k
S
e
a
C
o
m
)
,
2
0
2
0
,
p
p
.
1
–
6
,
d
o
i
:
1
0
.
1
1
0
9
/
B
l
a
c
k
S
e
a
C
o
m
4
8
7
0
9
.
2
0
2
0
.
9
2
3
4
9
5
9
.
[
2
7
]
M
.
A
.
F
e
r
r
a
g
,
O
.
F
r
i
h
a
,
D
.
H
a
mo
u
d
a
,
L.
M
a
g
l
a
r
a
s,
a
n
d
H
.
J
a
n
i
c
k
e
,
“
Ed
g
e
-
I
I
o
Tse
t
:
a
n
e
w
c
o
mp
r
e
h
e
n
s
i
v
e
r
e
a
l
i
st
i
c
c
y
b
e
r
sec
u
r
i
t
y
d
a
t
a
se
t
o
f
I
o
T
a
n
d
I
I
o
T
a
p
p
l
i
c
a
t
i
o
n
s
f
o
r
c
e
n
t
r
a
l
i
z
e
d
a
n
d
f
e
d
e
r
a
t
e
d
l
e
a
r
n
i
n
g
,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
4
0
2
8
1
–
4
0
3
0
6
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
6
5
8
0
9
.
[
2
8
]
H
.
B
.
M
c
M
a
h
a
n
,
E.
M
o
o
r
e
,
D
.
R
a
ma
g
e
,
S
.
H
a
m
p
s
o
n
,
a
n
d
B
.
A
.
y
A
r
c
a
s,
“
C
o
mm
u
n
i
c
a
t
i
o
n
-
e
f
f
i
c
i
e
n
t
l
e
a
r
n
i
n
g
o
f
d
e
e
p
n
e
t
w
o
r
k
s
f
r
o
m
d
e
c
e
n
t
r
a
l
i
z
e
d
d
a
t
a
,
”
Pr
o
c
e
e
d
i
n
g
s o
f
t
h
e
2
0
t
h
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
A
rt
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
S
t
a
t
i
st
i
c
s
,
AI
S
T
ATS
2
0
1
7
,
v
o
l
.
5
4
,
p
p
.
1
-
1
0
,
2
0
1
7
.
[
2
9
]
T.
D
.
N
g
u
y
e
n
,
S
.
M
a
r
c
h
a
l
,
M
.
M
i
e
t
t
i
n
e
n
,
H
.
F
e
r
e
i
d
o
o
n
i
,
N
.
A
so
k
a
n
,
a
n
d
A
.
-
R
.
S
a
d
e
g
h
i
,
“
D
Ï
o
T:
a
f
e
d
e
r
a
t
e
d
se
l
f
-
l
e
a
r
n
i
n
g
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
s
y
st
e
m
f
o
r
I
o
T,
”
i
n
2
0
1
9
I
EEE
3
9
t
h
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
D
i
s
t
ri
b
u
t
e
d
C
o
m
p
u
t
i
n
g
S
y
s
t
e
m
s
(
I
C
D
C
S
)
,
2
0
1
9
,
p
p
.
7
5
6
–
767
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
D
C
S
.
2
0
1
9
.
0
0
0
8
0
.
[
3
0
]
R
.
S
a
mr
i
n
a
n
d
D
.
V
a
su
mat
h
i
,
“
R
e
v
i
e
w
o
n
a
n
o
ma
l
y
b
a
s
e
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
s
y
st
e
m,”
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
El
e
c
t
ri
c
a
l
,
E
l
e
c
t
r
o
n
i
c
s,
C
o
m
m
u
n
i
c
a
t
i
o
n
,
C
o
m
p
u
t
e
r
,
a
n
d
O
p
t
i
m
i
za
t
i
o
n
T
e
c
h
n
i
q
u
e
s
(
I
C
EE
C
C
O
T
)
,
2
0
1
7
,
p
p
.
1
4
1
–
1
4
7
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
EEC
C
O
T.
2
0
1
7
.
8
2
8
4
6
5
5
.
[
3
1
]
P
.
V
.
Th
e
e
r
t
h
a
g
i
r
i
a
n
d
M
.
Th
a
n
g
a
v
e
l
u
,
“
E
l
e
p
h
a
n
t
i
n
t
r
u
si
o
n
w
a
r
n
i
n
g
s
y
st
e
m
u
s
i
n
g
I
o
T
a
n
d
6
L
o
W
P
A
N
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
e
n
so
rs,
Wi
r
e
l
e
ss C
o
m
m
u
n
i
c
a
t
i
o
n
s
a
n
d
C
o
n
t
r
o
l
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
6
0
5
–
6
1
6
,
2
0
1
9
,
d
o
i
:
1
0
.
2
1
7
4
/
2
2
1
0
3
2
7
9
0
9
6
6
6
1
9
1
1
2
9
0
9
2
0
0
6
.
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
Hyb
r
id
in
tr
u
s
io
n
d
etec
tio
n
mo
d
el
fo
r
h
iera
r
ch
ica
l wi
r
eles
s
s
en
s
o
r
n
etw
o
r
k
…
(
S
a
th
is
h
ku
m
a
r
M
ani
)
499
B
I
O
G
RAP
H
I
E
S
OF
AUTH
O
RS
S
a
th
ish
k
u
m
a
r
Ma
n
i
h
a
s
o
b
tain
e
d
h
is
B.
E
.
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
fr
o
m
Bh
a
ra
th
iar
Un
i
v
e
rsity
,
Co
imb
a
t
o
re
,
I
n
d
ia
a
n
d
M
.
Tec
h
.
d
e
g
re
e
in
In
f
o
rm
a
ti
o
n
Tec
h
n
o
l
o
g
y
fro
m
P
u
n
jab
i
Un
iv
e
r
sity
,
P
a
ti
a
la,
In
d
ia.
He
e
a
rn
e
d
h
is
P
h
.
D
.
in
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
in
e
e
ri
n
g
fr
o
m
S
a
v
e
e
th
a
Un
iv
e
rsit
y
,
Ch
e
n
n
a
i,
In
d
ia.
He
h
a
s
o
v
e
r
25
y
e
a
rs
of
e
x
p
e
rie
n
c
e
in
m
u
lt
ip
le
d
o
m
a
in
s
l
ik
e
tea
c
h
in
g
,
re
se
a
rc
h
a
n
d
s
o
ftwa
re
d
e
v
e
lo
p
m
e
n
t.
His
re
se
a
rc
h
a
re
a
is
n
e
two
rk
se
c
u
rit
y
,
m
a
c
h
in
e
l
e
a
rn
in
g
,
a
n
d
I
o
T.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
sa
th
ish
k
u
m
a
rm
a
n
i1
7
@g
m
a
il
.
c
o
m
.
Pa
r
a
sura
m
C
h
a
n
d
r
a
se
k
a
r
a
n
K
ish
o
r
e
r
a
ja
is
a
P
r
o
fe
ss
o
r
at
t
h
e
S
c
h
o
o
l
of
In
fo
rm
a
ti
o
n
Tec
h
n
o
lo
g
y
in
Ve
ll
o
r
e
In
stit
u
te
of
Tec
h
n
o
l
o
g
y
(VIT)
,
Ve
ll
o
re
,
In
d
ia.
He
h
a
s
a
to
ta
l
e
x
p
e
rien
c
e
of
23
y
e
a
rs
in
a
c
a
d
e
m
ic
a
n
d
re
se
a
rc
h
.
He
h
a
s
p
u
b
li
sh
e
d
a
n
d
p
re
se
n
ted
v
a
rio
u
s
p
a
p
e
rs
in
th
e
in
ter
n
a
ti
o
n
a
l
jo
u
rn
a
l
a
n
d
c
o
n
fe
re
n
c
e
s.
His
re
se
a
rc
h
in
tere
st
s
in
c
lu
d
e
se
c
u
rit
y
sy
ste
m
s,
in
tern
e
t
o
f
t
h
in
g
s,
a
n
d
m
e
d
ica
l
a
rti
ficia
l
in
tell
ig
e
n
c
e
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
k
ish
o
re
ra
ja.
p
c
@v
it
.
a
c
.
i
n
Christe
e
n
a
J
o
se
p
h
is
an
A
ss
o
c
iate
P
ro
fe
ss
o
r
wo
rk
in
g
in
t
h
e
De
p
a
rtme
n
t
o
f
ECE
at
S
RM
In
stit
u
te
of
S
c
ien
c
e
a
n
d
Tec
h
n
o
l
o
g
y
,
Ra
m
a
p
u
ra
m
,
Ch
e
n
n
a
i,
In
d
ia.
S
h
e
h
a
s
tea
c
h
in
g
e
x
p
e
rien
c
e
of
17
y
e
a
rs
a
n
d
h
a
s
p
u
b
li
sh
e
d
50
re
se
a
rc
h
p
a
p
e
rs
in
n
a
ti
o
n
a
l
a
n
d
i
n
tern
a
ti
o
n
al
jo
u
r
n
a
ls.
He
r
re
se
a
rc
h
in
tere
sts
i
n
c
lu
d
e
wire
les
s
c
o
m
m
u
n
ica
ti
o
n
a
n
d
n
e
tw
o
rk
s
.
S
h
e
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
c
h
ristee
n
a
0
0
3
@g
m
a
il
.
c
o
m
.
Re
ji
Ma
n
o
h
a
r
a
n
is
a
n
As
so
c
i
a
te
P
ro
fe
ss
o
r
in
th
e
D
e
p
a
rtme
n
t
of
El
e
c
tro
n
ics
a
n
d
Co
m
m
u
n
ica
ti
o
n
En
g
in
e
e
rin
g
at
R
o
h
i
n
i
Co
ll
e
g
e
of
En
g
i
n
e
e
rin
g
a
n
d
Tec
h
n
o
l
o
g
y
Ka
n
y
a
k
u
m
a
ri
In
d
ia.
He
h
a
s
a
to
tal
e
x
p
e
rien
c
e
of
15
y
e
a
rs
in
a
c
a
d
e
m
ic
a
n
d
re
se
a
rc
h
.
He
h
a
s
p
u
b
li
s
h
e
d
a
n
d
p
re
se
n
ted
v
a
rio
u
s
p
a
p
e
rs
in
t
h
e
in
tern
a
ti
o
n
a
l
jo
u
rn
a
l
a
n
d
c
o
n
fe
re
n
c
e
s.
His
re
se
a
rc
h
in
tere
sts
in
c
lu
d
e
n
e
two
rk
se
c
u
rit
y
,
in
tern
e
t
of
th
i
n
g
s,
a
n
ten
n
a
d
e
sig
n
,
a
n
d
i
n
t
ru
sio
n
d
e
tec
ti
o
n
.
He
c
a
n
b
e
c
o
n
tac
ted
a
t
e
m
a
il
:
re
ji
e
c
e
p
e
d
@g
m
a
il
.
c
o
m
.
Pra
sa
n
n
a
v
e
n
k
a
te
sa
n
Th
e
e
r
th
a
g
iri
is
wo
rk
i
n
g
as
t
h
e
As
sista
n
t
P
ro
fe
ss
o
r
in
t
h
e
De
p
a
rtme
n
t
of
Co
m
p
u
ter
S
c
ien
c
e
a
n
d
En
g
i
n
e
e
rin
g
,
G
ITAM
De
e
m
e
d
to
be
Un
iv
e
rsity
,
Be
n
g
a
lu
r
u
,
In
d
ia.
He
wa
s
a
wa
rd
e
d
Ph
.
D
.
(F
u
ll
-
Ti
m
e
)
d
e
g
re
e
in
t
h
e
y
e
a
r
2
0
2
1
on
th
e
wo
r
k
of
wire
les
s
c
o
m
m
u
n
ica
ti
o
n
with
m
a
c
h
in
e
lea
rn
in
g
fro
m
An
n
a
Un
iv
e
r
sity
,
Ch
e
n
n
a
i,
I
n
d
ia.
He
wa
s
a
wa
rd
e
d
th
e
m
o
b
il
it
y
g
ra
n
t
a
wa
rd
by
t
h
e
Re
p
u
b
li
c
of
S
lo
v
e
n
ia
in
th
e
y
e
a
r
2
0
1
7
-
2
0
1
8
.
He
h
a
s
p
u
b
li
sh
e
d
h
is
re
se
a
rc
h
wo
rk
s
in
12
S
CI
in
d
e
x
e
d
j
o
u
r
n
a
ls,
16
S
C
OPUS
in
d
e
x
e
d
j
o
u
r
n
a
ls.
His
re
se
a
rc
h
in
tere
sts
a
r
e
d
a
ta
s
c
ien
c
e
,
AI,
Io
T,
a
n
d
M
AN
ET
.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
p
ra
sa
n
n
a
it
9
1
@g
m
a
il
.
c
o
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.