I
nte
rna
t
io
na
l J
o
urna
l o
f
E
lect
rica
l a
nd
Co
m
p
ute
r
E
ng
in
ee
ring
(
I
J
E
CE
)
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t
201
8
,
p
p
.
2
5
2
1
~2
5
3
0
I
SS
N:
2
0
8
8
-
8708
,
DOI
: 1
0
.
1
1
5
9
1
/
i
j
ec
e
.
v8
i
4
.
p
p
2
5
2
1
-
2530
2521
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ia
e
s
co
r
e
.
co
m/
jo
u
r
n
a
ls
/in
d
ex
.
p
h
p
/
I
JE
C
E
I
m
pa
ct
of
Pac
k
et
Inter
-
a
rriv
a
l Ti
me Fe
a
tu
res for
O
n
line
P
eer
-
to
-
P
e
e
r
(P
2
P)
C
la
ss
ificatio
n
B
us
hra
M
o
ha
m
m
e
d Ali
Ab
d
a
lla
1
,
M
o
s
a
b H
a
m
da
n
2
,
M
o
ha
mm
ed
Su
lt
a
n M
o
ha
m
m
e
d
3
,
J
o
s
eph St
ephen B
a
s
s
i
4
,
I
s
m
a
ha
ni I
s
m
a
i
l
5
,
M
uh
a
m
m
a
d N
a
dzir
M
a
rso
no
6
1,
2,
3,
5
,
6
De
p
a
rtm
e
n
t
o
f
El
e
c
tro
n
ic a
n
d
C
o
m
p
u
ter E
n
g
in
e
e
rin
g
,
F
a
c
u
lt
y
o
f
El
e
c
tro
n
ic E
n
g
in
e
e
ri
n
g
,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
sia
,
8
1
3
1
0
,
J
o
h
o
r
Ba
h
r
u
,
M
a
lay
sia
4
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter E
n
g
in
e
e
rin
g
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsit
y
o
f
M
a
id
u
g
u
ri,
Bo
r
n
o
sta
te,
Nig
e
ria
Art
icle
I
nfo
AB
ST
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
A
p
r
12
,
2
0
1
8
R
ev
i
s
ed
J
u
l
20
,
2
0
1
8
A
cc
ep
ted
J
u
l
2
6
,
2
0
1
8
Id
e
n
ti
f
ica
ti
o
n
o
f
b
a
n
d
w
id
t
h
-
h
e
a
v
y
In
tern
e
t
traff
ic
is
i
m
p
o
rtan
t
f
o
r
n
e
tw
o
rk
a
d
m
in
istrato
rs
to
th
ro
tt
le
h
ig
h
-
b
a
n
d
w
id
th
a
p
p
l
ica
ti
o
n
traf
f
ic.
F
lo
w
fe
a
tu
re
s
b
a
se
d
c
las
sif
ic
a
ti
o
n
h
a
v
e
b
e
e
n
p
r
e
v
io
u
sly
p
ro
p
o
se
d
a
s
p
ro
m
isin
g
m
e
th
o
d
t
o
id
e
n
ti
f
y
In
tern
e
t
traff
ic
b
a
se
d
o
n
p
a
c
k
e
t
sta
ti
stica
l
f
e
a
tu
re
s.
T
h
e
se
lec
ti
o
n
o
f
sta
ti
stica
l
f
e
a
tu
re
s
p
la
y
s
a
n
im
p
o
rtan
t
ro
le
f
o
r
a
c
c
u
ra
te
a
n
d
ti
m
e
l
y
c
las
si
f
ica
ti
o
n
.
In
t
h
is
w
o
rk
,
w
e
i
n
v
e
stig
a
te
th
e
i
m
p
a
c
t
o
f
p
a
c
k
e
t
in
ter
-
a
rriv
a
l
ti
m
e
fe
a
tu
re
f
o
r
o
n
li
n
e
P
2
P
c
l
a
ss
if
i
c
a
ti
o
n
in
term
s
o
f
a
c
c
u
ra
c
y
,
Ka
p
p
a
sta
ti
stic
a
n
d
t
im
e
.
S
im
u
latio
n
s
w
e
r
e
c
o
n
d
u
c
ted
u
si
n
g
a
v
a
il
a
b
le
trac
e
s
f
ro
m
Un
iv
e
rsit
y
o
f
Bre
sc
ia,
Un
iv
e
rsit
y
o
f
A
a
lb
o
rg
a
n
d
Un
iv
e
rsit
y
o
f
C
a
m
b
rid
g
e
.
Ex
p
e
rime
n
tal
re
su
lt
s
sh
o
w
th
a
t
th
e
in
c
lu
sio
n
o
f
in
ter
-
a
rriv
a
l
ti
m
e
(I
A
T
)
a
s
a
n
o
n
li
n
e
f
e
a
tu
re
in
c
re
a
s
e
s si
m
u
latio
n
ti
m
e
a
n
d
d
e
c
re
a
se
s cl
a
ss
i
f
ica
ti
o
n
a
c
c
u
ra
c
y
a
n
d
Ka
p
p
a
sta
ti
st
ic.
K
ey
w
o
r
d
:
Featu
r
es
s
elec
tio
n
Ma
ch
i
n
e
lear
n
i
n
g
On
li
n
e
f
ea
tu
r
es
P
2
P
Co
p
y
rig
h
t
©
2
0
1
8
In
stit
u
te o
f
A
d
v
a
n
c
e
d
E
n
g
i
n
e
e
rin
g
a
n
d
S
c
ien
c
e
.
Al
l
rig
h
ts
re
se
rv
e
d
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Mu
h
a
m
m
ad
Nad
zir
Ma
r
s
o
n
o
,
Dep
ar
t
m
en
t o
f
E
lectr
o
n
ic
an
d
C
o
m
p
u
ter
E
n
g
in
ee
r
i
n
g
,
Facu
lt
y
o
f
E
lectr
o
n
ic
E
n
g
in
ee
r
in
g
,
Un
i
v
er
s
iti T
ek
n
o
lo
g
i M
ala
y
s
ia
,
8
1
3
1
0
,
J
o
h
o
r
B
ah
r
u
,
Ma
lay
s
ia
.
E
m
ail:
n
ad
zir
@
f
k
e.
u
t
m
.
m
y
1.
I
NT
RO
D
UCT
I
O
N
T
o
d
ay
,
p
ee
r
-
to
-
p
ee
r
(
P
2
P)
is
as
an
ar
ch
itec
tu
r
e
f
o
r
s
h
ar
i
n
g
a
w
id
e
r
an
g
e
o
f
m
ed
ia
o
n
t
h
e
I
n
ter
n
et.
P
2
P
tr
af
f
ic
r
ep
r
esen
ts
ab
o
u
t
2
7
%
to
6
0
%
o
f
th
e
to
tal
I
n
ter
n
et
tr
af
f
ic,
d
ep
en
d
in
g
o
n
g
eo
g
r
ap
h
ic
lo
ca
tio
n
[1
]
,
[
2
]
.
T
h
e
h
ig
h
v
o
l
u
m
e
o
f
P
2
P
tr
af
f
ic
is
d
u
e
to
f
ile
s
h
ar
i
n
g
,
v
i
d
eo
s
tr
ea
m
in
g
,
o
n
-
li
n
e
g
a
m
i
n
g
an
d
o
th
er
ac
tiv
ities
th
at
clie
n
t
-
s
er
v
er
ar
ch
itect
u
r
e
ca
n
n
o
t
ac
co
m
p
li
s
h
a
s
f
a
s
t
o
r
as
ef
f
icie
n
t
a
s
th
e
P
2
P
a
r
ch
itect
u
r
e.
R
ap
id
p
r
o
g
r
ess
io
n
o
f
P
2
P
tr
af
f
ic
v
o
lu
m
e
t
h
r
o
u
g
h
o
u
t
t
h
e
y
ea
r
s
h
a
v
e
r
es
u
lted
i
n
d
eter
io
r
ated
n
e
t
w
o
r
k
p
er
f
o
r
m
a
n
ce
an
d
co
n
g
es
tio
n
d
u
e
to
th
e
h
ig
h
b
an
d
w
id
th
co
n
s
u
m
p
ti
o
n
o
f
P
2
P
ap
p
licatio
n
s
[
3
]
.
T
h
er
ef
o
r
e,
tr
af
f
ic
id
en
ti
f
icatio
n
i
s
r
eq
u
ir
ed
to
i
m
p
r
o
v
e
tr
af
f
ic
m
an
a
g
e
m
e
n
t.
First
g
e
n
er
atio
n
P
2
P
ap
p
licatio
n
tr
af
f
ic
w
er
e
r
elati
v
el
y
ea
s
y
to
b
e
id
en
t
if
ied
d
u
e
to
th
e
u
s
e
o
f
f
i
x
ed
p
o
r
ts
n
u
m
b
er
s
.
Ho
w
e
v
er
,
cu
r
r
en
t
P
2
P
a
p
p
licatio
n
s
ar
e
ab
le
to
cir
cu
m
v
en
t
p
o
r
t
-
b
ased
id
en
tif
icat
io
n
b
y
u
s
i
n
g
an
o
n
y
m
o
u
s
p
o
r
t
n
u
m
b
er
s
o
r
p
o
r
t
d
is
g
u
is
e
[
4
]
,
[
2
]
.
B
esid
es,
m
et
h
o
d
s
th
at
r
el
y
o
n
i
n
s
p
ec
tin
g
ap
p
licatio
n
p
ay
lo
ad
s
i
g
n
at
u
r
es
h
a
v
e
al
s
o
b
ee
n
p
r
o
p
o
s
ed
[
5
]
.
Fo
r
p
r
iv
ac
y
a
n
d
i
m
p
r
ac
tic
al
r
ea
s
o
n
s
,
t
h
is
m
et
h
o
d
is
in
e
f
f
ec
t
iv
e.
T
h
e
ef
f
ec
ti
v
e
n
es
s
o
f
t
h
e
p
o
r
t
-
b
ased
an
d
p
a
y
lo
ad
-
b
ased
m
et
h
o
d
s
p
r
o
m
p
te
d
th
e
u
s
e
o
f
f
lo
w
s
tatis
t
ics
a
s
f
ea
t
u
r
es
f
o
r
tr
a
f
f
ic
id
en
ti
f
icatio
n
.
T
h
ese
s
tr
ateg
ie
s
o
f
f
er
f
le
x
ib
ilit
y
to
d
etec
t
P
2
P
tr
af
f
ic
co
m
p
ar
ed
to
u
s
i
n
g
s
ig
n
at
u
r
e
-
b
ased
a
n
d
p
o
r
t
-
b
ased
m
et
h
o
d
s
.
Sev
er
al
tec
h
n
iq
u
es
h
av
e
b
ee
n
p
r
o
p
o
s
ed
o
v
er
th
e
last
t
w
o
d
ec
ad
es
th
at
f
o
cu
s
ed
o
n
th
e
attain
ab
le
id
en
ti
f
icatio
n
ac
c
u
r
ac
y
u
s
in
g
s
ev
er
al
m
ac
h
in
e
lear
n
i
n
g
(
M
L
)
alg
o
r
it
h
m
s
.
Ho
w
e
v
er
,
th
e
i
m
p
ac
t
o
f
e
x
p
lo
r
in
g
th
e
ef
f
ec
t
o
f
d
is
ti
n
ct
s
ets
o
f
s
ta
tis
tical
f
ea
tu
r
es
h
as
n
o
t
b
ee
n
r
esear
ch
ed
in
-
d
ep
th
.
W
o
r
k
in
[
6
]
h
as
r
ep
o
r
ted
th
at
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
5
2
1
–
2
5
3
0
2522
f
ea
t
u
r
e
s
elec
t
io
n
is
a
v
ital
ta
s
k
to
i
m
p
r
o
v
e
t
h
e
cla
s
s
i
f
icat
i
o
n
an
d
id
en
ti
f
icat
io
n
p
er
f
o
r
m
an
ce
co
m
p
ar
ed
to
s
elec
tio
n
o
f
t
h
e
cla
s
s
i
f
icat
io
n
alg
o
r
ith
m
.
P
r
esen
t
l
y
,
s
ev
er
al
f
ea
tu
r
e
s
elec
tio
n
al
g
o
r
ith
m
s
h
a
v
e
b
ee
n
in
tr
o
d
u
ce
d
,
e.
g
.
,
[
7
]
-
[
1
1
]
.
Ho
w
e
v
er
,
m
o
s
t
o
f
t
h
e
i
n
tr
o
d
u
ce
d
m
et
h
o
d
s
do
n
o
t
co
n
si
d
er
th
e
i
m
p
ac
t
o
f
in
te
g
r
ati
n
g
o
n
li
n
e
f
ea
t
u
r
es
w
it
h
in
ter
-
ar
r
iv
a
l ti
m
e
(
I
A
T
)
f
o
r
o
n
lin
e
P
2
P c
lass
if
ic
atio
n
.
T
h
is
p
ap
er
p
r
o
p
o
s
es
an
ap
p
r
o
ac
h
b
ased
o
n
an
al
y
tic
m
e
th
o
d
s
o
n
e
-
w
a
y
a
n
al
y
s
is
o
f
v
ar
ian
ce
an
d
in
cr
e
m
e
n
tal
tr
af
f
ic
clas
s
i
f
icati
o
n
al
g
o
r
ith
m
.
O
n
e
-
w
a
y
an
a
l
y
s
is
o
f
v
ar
ian
ce
is
i
m
p
le
m
e
n
t
ed
u
s
i
n
g
KN
AM
E
to
o
l
an
d
Ho
ef
f
d
i
n
g
T
r
ee
in
cr
em
en
tal
m
ac
h
in
e
lear
n
in
g
al
g
o
r
ith
m
i
s
i
m
p
le
m
en
ted
u
s
in
g
MO
A
(
Ma
s
s
iv
e
O
n
-
lin
e
An
al
y
s
is
)
to
o
l in
o
r
d
er
to
in
v
e
s
ti
g
ate
t
h
e
i
m
p
ac
t o
f
p
ac
k
et
I
A
T
f
ea
tu
r
e
f
o
r
o
n
lin
e
P
2
P
class
i
f
icatio
n
.
T
h
e
r
e
m
ain
d
er
o
f
t
h
is
p
ap
er
is
o
r
g
a
n
ized
as
f
o
llo
w
s
.
Secti
o
n
2
i
n
tr
o
d
u
ce
s
r
elate
d
w
o
r
k
s
in
cl
u
d
i
n
g
ML
co
n
ce
p
ts
,
tr
a
f
f
ic
cla
s
s
i
f
ic
ati
o
n
a
n
d
f
ea
tu
r
e
s
elec
tio
n
.
S
ec
tio
n
3
d
is
c
u
s
s
es
t
h
e
m
et
h
o
d
o
lo
g
y
to
in
v
es
tig
a
te
th
e
i
m
p
ac
t
o
f
p
ac
k
et
i
n
ter
-
ar
r
iv
al
ti
m
e
f
ea
t
u
r
e
f
o
r
o
n
li
n
e
P
2
P
class
if
icat
io
n
.
T
h
e
ex
p
er
i
m
en
tal
s
e
tu
p
,
r
es
u
lt
an
d
d
is
cu
s
s
io
n
ar
e
d
is
c
u
s
s
ed
in
Sectio
n
4
.
Sectio
n
5
p
r
esen
ts
th
e
co
n
cl
u
s
io
n
.
2.
RE
L
AT
E
D
WO
RK
Ma
ch
i
n
e
lear
n
i
n
g
(
M
L
)
is
ap
r
o
m
is
i
n
g
tech
n
iq
u
e
t
h
at
h
as
b
e
en
u
s
ed
f
o
r
d
ata
m
in
i
n
g
an
d
k
n
o
w
led
g
e
d
is
co
v
er
y
[
1
2
]
.
Un
s
u
p
er
v
is
ed
lear
n
in
g
s
tr
ateg
ie
s
b
asicll
y
clu
s
ter
s
f
lo
w
s
w
it
h
s
i
m
ilar
p
ar
t
ter
n
b
eh
av
io
u
r
.
Su
p
er
v
i
s
ed
lear
n
in
g
n
ee
d
s
a
s
et
o
f
lab
eled
d
ata
t
o
tr
ai
n
its
m
o
d
el
in
ad
v
a
n
ce
f
o
r
id
en
tif
icat
io
n
an
d
class
i
f
icatio
n
o
f
d
ata
[
1
2
]
.
C
las
s
i
f
icatio
n
u
s
i
n
g
f
lo
w
f
e
atu
r
es
m
ain
l
y
d
ep
lo
y
s
m
ac
h
in
e
lear
n
i
n
g
to
p
er
f
o
r
m
tr
ai
n
in
g
a
n
d
class
i
f
icatio
n
.
Fro
m
th
e
e
x
tr
ac
ted
f
lo
w
f
ea
tu
r
e
s
,
th
e
clas
s
if
ie
r
p
r
e
d
ict
s
th
e
class
o
f
n
e
w
f
lo
w
.
T
h
is
p
r
o
ce
s
s
is
ca
lled
a
d
ata
m
in
i
n
g
p
r
o
b
lem
.
T
h
e
f
ir
s
t
w
o
r
k
u
s
i
n
g
th
i
s
tec
h
n
iq
u
e
w
as
b
y
[
1
3
]
.
Gen
er
ally
,
class
if
icatio
n
ca
n
b
e
p
er
f
o
r
m
ed
in
t
h
r
ee
s
tep
s
,
e
x
tr
ac
ti
n
g
t
h
e
f
ea
tu
r
es,
s
elec
tio
n
o
f
f
ea
tu
r
e
a
n
d
g
en
er
ati
n
g
cla
s
s
i
f
ier
[
1
4
]
.
Mo
o
r
e
et
a
l
.
[
1
5
]
h
as
s
u
g
g
e
s
te
d
2
4
9
f
ea
tu
r
es t
h
at
ca
n
b
e
p
o
te
n
tiall
y
u
s
ed
i
n
M
L
tr
af
f
ic
id
e
n
tif
icat
io
n
.
Ho
w
e
v
er
m
o
s
t
o
f
t
h
ese
f
ea
t
u
r
es
ca
n
o
n
l
y
b
e
o
b
tain
ed
in
a
n
o
f
f
-
li
n
e
m
o
d
e.
Of
f
-
li
n
e
f
ea
t
u
r
e
s
s
u
c
h
as
m
ax
i
m
u
m
an
d
m
i
n
i
m
u
m
b
y
te
s
in
p
ac
k
e
t
o
n
l
y
ca
n
b
e
o
b
tain
ed
w
it
h
co
m
p
lete
f
lo
w
s
.
W
o
r
k
in
[
1
6
]
em
p
lo
y
ed
all
2
4
9
f
ea
t
u
r
es
s
u
g
g
e
s
ted
in
[
1
5
]
d
er
iv
ed
f
r
o
m
p
ac
k
et
s
tr
ea
m
s
co
n
s
is
ti
n
g
o
f
o
n
e
o
r
m
o
r
e
p
ac
k
e
t
h
ea
d
er
s
.
Mo
s
t
o
f
th
ese
f
ea
t
u
r
es c
an
n
o
t b
e
ex
tr
ac
ted
o
n
lin
e
f
r
o
m
li
v
e
tr
af
f
ic
f
o
r
o
n
lin
e
tr
a
f
f
ic
id
en
ti
f
ic
at
io
n
.
Featu
r
e
s
elec
tio
n
(
FS
)
is
u
s
ed
to
s
elec
t
o
p
ti
m
al
s
u
b
s
et
f
ea
t
u
r
es
f
r
o
m
t
h
e
i
n
p
u
t
w
h
ic
h
ca
n
ef
f
icien
tl
y
d
escr
ib
e
th
e
i
n
p
u
t
d
ata
w
h
il
e
r
ed
u
cin
g
e
f
f
ec
ts
f
r
o
m
ir
r
el
ev
an
t
o
r
n
o
i
s
e
f
ea
tu
r
e
s
y
et
s
till
p
r
o
v
id
e
g
o
o
d
p
r
ed
ictio
n
o
f
its
cla
s
s
[
7
]
,
[
1
7
]
.
T
r
af
f
ic
id
en
ti
f
icatio
n
ca
n
b
e
i
m
p
r
o
v
ed
w
it
h
r
e
f
er
en
c
e
to
co
m
p
u
tatio
n
al
p
er
f
o
r
m
a
n
ce
a
n
d
ac
cu
r
ac
y
b
y
u
s
i
n
g
th
e
m
o
s
t r
ele
v
an
t
f
ea
t
u
r
es [
1
8
]
.
L
o
o
et
a
l
.
[
8
]
p
r
o
p
o
s
ed
1
2
o
n
li
n
e
f
ea
tu
r
e
s
w
it
h
o
u
t
f
ea
tu
r
e
s
r
elate
d
ti
m
e.
Mo
n
e
m
i
e
t
a
l
.
[
1
9
]
h
as
p
r
o
p
o
s
ed
3
5
r
ea
l
-
ti
m
e
f
lo
w
f
ea
tu
r
es
t
h
at
ca
n
b
e
ea
s
il
y
e
x
tr
ac
ted
f
r
o
m
f
lo
w
r
ec
o
r
d
s
.
T
h
ese
f
lo
w
s
in
cl
u
d
e
n
u
m
b
er
o
f
p
ac
k
et
s
,
p
o
r
t
ad
d
r
ess
,
p
r
o
to
co
l,
o
v
er
all
T
r
an
s
m
is
s
io
n
C
o
n
tr
o
l
P
r
o
to
co
l
(
T
C
P
)
f
la
g
s
,
a
v
er
ag
e
v
o
lu
m
e
i
n
b
y
te,
v
o
lu
m
e
i
n
b
y
te
p
er
p
ac
k
et,
f
lo
w
d
u
r
atio
n
,
p
ay
lo
ad
v
o
lu
m
e
i
n
b
y
te,
f
lo
w
d
u
r
atio
n
,
av
er
ag
e
n
u
m
b
er
o
f
p
ac
k
et
p
er
s
ec
o
n
d
,
av
er
ag
e
v
o
l
u
m
e
i
n
b
y
te
p
er
s
ec
o
n
d
,
av
er
a
g
e
p
a
y
lo
ad
v
o
lu
m
e
in
b
y
te
p
e
r
s
ec
o
n
d
,
av
er
a
g
e
p
a
y
lo
ad
v
o
l
u
m
e
i
n
b
y
te
p
er
p
ac
k
et,
an
d
av
er
ag
e
ti
m
e
i
n
ter
v
al.
E
r
m
an
et
a
l
.
[
1
6
]
h
as
p
er
f
o
r
m
ed
b
ac
k
w
ar
d
g
r
ee
d
y
s
ea
r
ch
o
n
v
ar
io
u
s
d
atasets
a
n
d
f
o
u
n
d
th
at
t
h
e
u
s
e
o
f
ti
m
e
-
r
ela
ted
f
ea
tu
r
es
s
u
c
h
a
s
d
u
r
atio
n
,
I
A
T
an
d
f
lo
w
t
h
r
o
u
g
h
p
u
t a
r
e
n
o
t
u
s
e
f
u
l in
tr
a
f
f
ic
cl
ass
i
f
icatio
n
.
On
li
n
e
f
ea
tu
r
es
tech
n
iq
u
e
s
h
a
v
e
b
ee
n
p
r
o
p
o
s
ed
in
[
7
]
,
[
2
0
]
.
T
h
ese
w
o
r
k
s
u
s
ed
C
a
m
b
r
id
g
e
d
atasets
an
d
N
aiv
e
B
a
y
e
s
to
ev
alu
a
te
t
w
o
f
ea
t
u
r
e
s
elec
tio
n
alg
o
r
it
h
m
s
n
a
m
ed
B
ias
C
o
e
f
f
ic
ien
t
R
es
u
lts
(
B
FS
)
an
d
S
elec
ted
O
n
li
n
e
F
ea
tu
r
e.
T
h
e
s
e
w
o
r
k
s
ac
h
ie
v
ed
ac
cu
r
ac
y
o
f
9
0
.
9
2
%
an
d
9
3
.
2
0
%,
r
esp
ec
tiv
el
y
.
B
esid
es,
t
h
e
w
o
r
k
i
n
[
7
]
h
as
co
n
s
id
er
ed
I
A
T
as o
n
e
o
f
th
e
p
r
o
p
o
s
ed
o
n
-
li
n
e
f
ea
t
u
r
es.
Mo
s
t
r
esear
ch
es
h
a
v
e
f
o
c
u
s
e
d
o
n
o
n
lin
e
f
ea
t
u
r
es
w
it
h
I
AT
as
s
u
g
g
ested
i
n
[
7
]
,
[
11
]
,
[
19
]
,
[
2
1
]
.
Ho
w
e
v
er
,
th
e
i
m
p
ac
t
o
f
p
ac
k
e
t
in
ter
-
ar
r
iv
al
ti
m
e
f
ea
t
u
r
e
f
o
r
o
n
lin
e
P
2
P
class
i
f
icatio
n
s
till
p
lay
s
a
n
i
m
p
o
r
tan
t
r
o
le
f
o
r
ac
cu
r
ate
an
d
ti
m
el
y
cl
ass
i
f
icatio
n
.
3.
O
VE
RVI
E
W
O
F
T
H
E
M
E
T
H
O
DS
Ou
r
p
r
o
p
o
s
ed
m
et
h
o
d
to
in
v
i
s
ticate
t
h
e
i
m
p
ac
t
o
f
p
ac
k
et
I
A
T
f
ea
t
u
r
e
f
o
r
o
n
li
n
e
P
2
P
cl
ass
i
f
icatio
n
co
n
s
is
t
o
f
t
w
o
m
ai
n
s
ta
g
es
,
te
s
t
th
e
s
i
g
n
f
ica
n
ce
o
f
p
ac
k
et
I
A
T
f
ea
t
u
r
e
an
d
i
n
v
e
s
ti
g
ate
t
h
e
i
m
p
ac
t
o
f
p
ac
k
et
I
A
T
f
ea
t
u
r
e
f
o
r
o
n
li
n
e
P
2
P
id
en
ti
f
icatio
n
w
it
h
r
e
f
er
en
ce
to
ac
cu
r
ac
y
,
k
ap
p
a
s
tati
s
tic
an
d
t
i
m
e.
T
h
e
f
ir
s
t
s
ta
g
e
o
n
e
-
w
a
y
a
n
al
y
s
is
o
f
v
ar
ian
ce
an
al
y
t
i
cs
u
s
i
n
g
KN
A
ME
to
o
l to
test
th
e
s
i
g
n
f
ica
n
ce
o
f
I
A
T
.
I
n
t
h
e
s
ec
o
n
d
s
tag
e,
Ho
ef
f
d
i
n
g
T
r
ee
in
cr
em
e
n
tal
m
ac
h
in
e
lear
n
i
n
g
al
g
o
r
ith
m
i
s
i
m
p
le
m
e
n
ted
in
MO
A
to
o
l.
A
ll
s
ta
g
es
w
il
l
b
e
d
is
cu
s
s
ed
in
d
etails
in
Sec
tio
n
3
.
1
.
Fig
u
r
e
1
s
h
o
w
s
th
e
o
v
er
v
ie
w
o
f
th
e
p
r
o
p
o
s
ed
m
et
h
o
d
to
in
v
est
ig
ate
t
h
e
i
m
p
ac
t o
f
p
ac
k
et
I
A
T
f
e
atu
r
e
f
o
r
o
n
lin
e
P
2
P
class
if
icatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
I
mp
a
ct
o
f P
a
ck
et
I
n
ter
-
a
r
r
iva
l Time
F
ea
tu
r
es fo
r
...
(
B
u
s
h
r
a
Mo
h
a
mme
d
A
li A
b
d
a
lla
)
2523
Fig
u
r
e
1
.
P
r
o
p
o
s
ed
m
eth
o
d
3
.
1
.
T
ec
hn
i
qu
e
s
f
o
r
a
na
ly
zing
f
e
a
t
ures
Ko
n
s
ta
n
z
I
n
f
o
r
m
atio
n
Mi
n
er
(
Kn
i
m
e)
is
a
r
ec
en
t
o
p
en
-
s
o
u
r
c
e
d
ata
an
al
y
tic
s
p
latf
o
r
m
th
a
t
allo
w
s
f
o
r
u
n
d
er
ta
k
in
g
co
m
p
lete
s
ta
tis
ti
cs
an
d
d
ata
m
i
n
i
n
g
an
al
y
s
i
s
.
On
e
-
w
a
y
A
NOV
A
i
s
i
m
p
le
m
en
ted
in
KNI
ME
b
en
ch
m
ar
k
[
2
2
]
.
W
E
KA
w
o
r
k
s
p
ac
e
to
o
ls
also
is
u
s
ed
f
o
r
class
i
f
icatio
n
[
2
3
]
.
On
e
-
w
a
y
A
N
OV
A
is
th
e
m
o
s
t
ef
f
ec
tiv
e
m
eth
o
d
a
v
ailab
le
f
o
r
an
al
y
zi
n
g
t
h
e
m
o
r
e
co
m
p
lex
d
ata
s
ets
[
2
4
]
.
I
n
th
i
s
w
o
r
k
,
we
co
m
p
u
ted
th
e
F
-
s
tatis
t
ic
u
s
i
n
g
A
NOV
A
.
E
q
u
a
tio
n
s
(
5
)
an
d
(
1
)
r
ep
r
esen
ts
s
u
m
o
f
s
q
u
ar
e
(
S
S)
in
A
NOV
A
.
W
h
ile
t
h
e
s
u
m
o
f
s
q
u
ar
es
f
o
r
T
r
ea
tm
e
n
t
(
SS
T
)
is
g
i
v
en
b
y
E
q
u
at
io
n
(
2
)
.
Su
m
o
f
s
q
u
ar
es
f
o
r
E
r
r
o
r
(
SS
E
)
is
co
m
p
u
ted
u
s
i
n
g
E
q
u
atio
n
(
3
)
.
T
h
e
Var
ian
ce
b
et
w
ee
n
T
r
ea
tm
e
n
ts
(
MST
)
is
co
m
p
u
ted
b
y
E
q
u
atio
n
(
4
)
.
T
h
e
Var
ian
ce
W
it
h
i
n
T
r
ea
tm
en
t
s
(
MS
E
)
is
co
m
p
u
ted
u
s
in
g
E
q
u
atio
n
(
5
)
.
F
-
s
tati
s
t
ic
is
o
b
tain
ed
b
y
d
iv
id
in
g
MS
T
t
o
MSE
is
g
iv
e
n
b
y
E
q
u
a
tio
n
(
6
)
.
Usi
n
g
9
5
% c
o
n
f
id
e
n
ce
in
ter
v
al
f
o
r
m
ea
n
d
i
f
f
er
en
ce
,
A
NOV
A
i
s
ca
lc
u
late
d
as:
∑
∑
(
)
(
1
)
∑
∑
(
)
∑
∑
(
)
∑
∑
(
)
∑
∑
(
)
(
2
)
∑
∑
(
)
(
3
)
T
h
en
(
4
)
(
5
)
(
6
)
T
h
u
s
,
w
it
h
A
NOV
A
te
s
t
n
u
ll
h
y
p
o
t
h
esi
s
,
w
h
ic
h
m
ea
n
s
th
a
t
th
er
e
ar
e
n
o
tr
ea
tm
e
n
t
ef
f
ec
ts
.
W
h
er
e
b
ar
is
th
e
s
a
m
p
les
m
ea
n
,
is
th
e
s
a
m
p
le
s
ize,
is
th
e
s
p
ec
if
ied
p
o
p
u
latio
n
m
e
an
.
Ma
s
s
i
v
e
O
n
li
n
e
An
al
y
s
i
s
(
M
OA
)
[
2
5
]
:
MO
A
i
s
a
d
ata
s
tr
e
a
m
m
i
n
i
n
g
s
u
ite
t
h
at
w
a
s
w
r
it
ten
i
n
J
av
a.
User
s
ca
n
u
s
e
MO
A
u
s
i
n
g
Gr
ap
h
ic
User
I
n
ter
f
ac
e
(
GUI
)
o
r
th
r
o
u
g
h
co
m
m
an
d
li
n
es.
Di
f
f
er
en
t
f
r
o
m
W
E
KA
[
2
3
]
w
h
ic
h
is
f
o
r
b
atch
d
ata
m
i
n
in
g
,
MO
A
s
p
ec
ializes
o
n
p
r
o
ce
s
s
in
g
a
n
d
an
a
l
y
zi
n
g
d
ata
s
tr
ea
m
s
.
T
h
e
s
u
it
e
in
cl
u
d
es
e
v
al
u
atio
n
to
o
ls
s
u
c
h
as
co
n
ce
p
t
d
r
if
t
e
v
al
u
atio
n
,
a
n
d
i
n
ter
leav
e
-
te
s
t
-
th
e
n
-
tr
ain
e
v
alu
a
tio
n
.
I
t
is
a
ls
o
b
u
ilt
w
i
th
a
co
llectio
n
o
f
d
ata
s
tr
ea
m
id
en
ti
f
icatio
n
tec
h
n
iq
u
es s
u
c
h
as
Nai
v
e
B
a
y
e
s
,
Ho
ef
f
d
in
g
T
r
ee
,
B
ag
g
in
g
an
d
B
o
o
s
tin
g
tech
n
iq
u
e
s
.
I
n
th
is
p
ap
er
,
MO
A
is
u
s
ed
to
an
aly
ze
th
e
i
m
p
ac
t
o
f
i
n
teg
r
at
in
g
o
n
lin
e
f
ea
tu
r
e
s
w
i
t
h
I
A
T
f
o
r
o
n
lin
e
P
2
P
class
if
icati
o
n
.
4.
E
XP
E
R
I
M
E
NT
A
L
SE
T
UP
,
RE
SU
L
T
S AN
D
D
I
SCU
SS
I
O
N
T
h
is
s
ec
tio
n
,
p
r
esen
t
s
an
d
d
icu
s
s
es
t
h
e
n
et
w
o
r
k
tr
a
f
f
ic
d
atas
ets
u
s
ed
a
n
d
th
e
ev
al
u
atio
n
m
eth
o
d
u
s
ed
to
ev
alu
ate
t
h
e
i
m
p
ac
t o
f
in
teg
r
atin
g
o
n
li
n
e
f
ea
t
u
r
es
w
it
h
i
n
te
r
-
ar
r
iv
al
ti
m
e
f
o
r
o
n
lin
e
P
2
P
c
lass
i
f
icatio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
5
2
1
–
2
5
3
0
2524
4.
1.
Da
t
a
s
et
N
et
w
o
r
k
tr
ac
es
ar
e
u
s
ed
to
v
a
lid
ate
th
e
p
er
f
o
r
m
a
n
ce
o
f
t
h
e
p
r
o
p
o
s
ed
tech
n
iq
u
e.
T
h
ese
d
atasets
ar
e
P
A
M
[
2
6
]
,
UNI
B
S
[
2
7
]
an
d
C
a
m
b
r
id
g
e
[
1
5
]
.
T
ab
le
1
s
u
m
m
ar
izes
th
e
u
s
ed
d
ataset
s
,
w
h
ic
h
t
h
e
d
escr
ip
tio
n
o
f
ea
ch
d
ataset
as
f
o
llo
w
s
:
a.
P
A
M
tr
ac
es
w
as
ca
p
tu
r
ed
i
n
A
alb
o
r
g
Un
i
v
er
s
i
t
y
f
r
o
m
2
5
th
Feb
r
u
ar
y
2
0
1
3
to
1
s
t
Ma
y
2
0
1
3
an
d
r
ep
o
r
ted
in
[
2
6
]
.
T
h
e
lab
el
o
f
th
e
d
atas
et
w
a
s
co
llected
u
s
i
n
g
Vo
lu
n
t
ee
r
-
B
ased
S
y
s
te
m
(
VB
S).
A
t
o
tal
o
f
1
,
2
6
2
,
0
2
2
f
lo
w
s
w
er
e
ca
p
tu
r
ed
,
w
h
er
e
5
3
5
,
4
3
8
f
lo
w
s
w
er
e
lab
eled
as
r
ep
o
r
ted
i
n
[
2
6
]
.
Ho
w
e
v
er
,
o
n
l
y
3
3
9
,
0
6
1
f
lo
w
s
co
u
ld
b
e
u
s
ed
as
m
o
s
t
f
lo
w
s
h
av
e
les
s
t
h
an
f
iv
e
p
ac
k
ets
a
n
d
th
e
n
e
tf
lo
w
an
d
f
ea
t
u
r
e
e
x
tr
ac
to
r
m
o
d
u
les
o
n
l
y
e
x
tr
ac
t
f
lo
w
s
t
h
at
co
n
ta
i
n
f
iv
e
p
ac
k
et
s
o
r
m
o
r
e.
B
y
u
s
in
g
t
h
e
p
r
o
v
id
ed
in
f
o
r
m
atio
n
f
iles
,
th
e
f
lo
w
s
ar
e
lab
eled
i
n
to
f
o
u
r
class
e
s
:
W
E
B
,
F
T
P
,
P
2
P
,
an
d
Oth
er
s
.
b.
T
h
e
UNI
B
S
d
atasets
[
2
7
]
w
er
e
o
b
tain
ed
f
r
o
m
a
s
er
ies
o
f
wo
r
k
s
tatio
n
s
at
th
e
Un
i
v
er
s
i
t
y
o
f
B
r
esciaf
r
o
m
30
th
Sep
tem
b
er
2
0
1
6
to
2
n
d
O
cto
b
er
2
0
1
6
.
T
h
e
tr
ac
es
ar
e
c
o
llected
o
n
ed
g
e
r
o
u
ter
,
w
h
er
e
th
e
tr
af
f
ic
w
a
s
g
en
er
a
ted
b
y
2
0
w
o
r
k
s
ta
tio
n
s
r
u
n
n
i
n
g
GT
to
o
ls
et.
I
n
th
i
s
w
o
r
k
,
th
e
tr
ac
es
w
er
e
p
r
o
ce
s
s
ed
u
s
i
n
g
n
etf
l
o
w
an
d
f
ea
t
u
r
e
ex
tr
ac
to
r
b
ased
o
n
1
m
i
n
u
tes
ti
m
eo
u
t
an
d
f
lo
w
s
w
it
h
a
m
in
i
m
u
m
o
f
f
i
v
e
p
ac
k
e
ts
ar
e
ex
tr
ac
ted
.
A
to
tal
o
f
7
7
,
3
0
3
f
lo
w
s
ar
e
ex
tr
ac
ted
an
d
all
f
lo
w
s
f
e
at
u
r
es
ar
e
ex
tr
ac
ted
b
ased
o
n
o
n
l
y
t
h
e
f
ir
s
t
f
i
v
e
p
ac
k
ets
o
f
ea
ch
f
lo
w
.
T
h
e
ac
co
m
p
a
n
ied
g
r
o
u
n
d
tr
u
t
h
lab
els
,
w
er
e
u
s
e
to
clas
s
i
f
y
f
lo
w
s
i
n
to
f
i
v
e
clas
s
es,
P
2
P
,
Sk
y
p
e,
W
eb
,
Oth
er
s
,
an
d
Ma
il.
c.
T
h
e
C
a
m
b
r
id
g
e
d
atasets
w
er
e
o
b
tain
ed
f
r
o
m
tr
ac
es
ca
p
tu
r
e
d
o
n
th
e
Ge
n
o
m
e
C
a
m
p
u
s
n
et
w
o
r
k
in
Au
g
u
s
t
2
0
0
3
in
th
e
Un
i
v
er
s
it
y
o
f
C
a
m
b
r
id
g
e
[
1
5
]
.
T
h
er
e
ar
e
ten
d
if
f
er
en
t
d
ataset
s
ea
ch
f
r
o
m
a
d
i
f
f
er
en
t
p
er
io
d
o
f
th
e
2
4
-
h
o
u
r
d
ay
.
T
h
ese
d
atase
ts
co
n
s
i
s
t
o
f
T
C
P
f
lo
w
.
Fu
r
t
h
e
r
m
o
r
e,
ev
er
y
f
lo
w
s
a
m
p
le
is
h
i
g
h
d
i
m
en
s
io
n
al
s
in
ce
it
co
n
tai
n
s
2
4
8
f
ea
tu
r
es.
T
h
e
d
ataset
ap
p
licatio
n
s
w
it
h
n
eg
li
g
ib
le
c
lass
e
s
s
u
c
h
as
g
a
m
e
s
a
n
d
in
ter
ac
ti
v
e
w
er
e
e
x
cl
u
d
ed
as
i
t
is
in
s
u
f
f
icie
n
t
f
o
r
tr
ai
n
in
g
a
n
d
test
i
n
g
.
T
h
ese
i
n
clu
d
e
cla
s
s
es
s
u
c
h
as
FT
P
-
P
asv
,
A
t
tack
,
P
2
P
,
Data
b
ase,
Mu
lti
m
ed
ia,
W
eb
,
Ma
il,
FTP
-
C
o
n
tr
o
l,
an
d
Ser
v
ice
s
.
T
ab
el
1
.
Data
s
ets Statis
t
ics
U
N
I
B
S
P
A
M
C
a
m
b
r
i
d
g
e
#
F
l
o
w
i
n
st
a
n
c
e
s
7
7
,
3
0
3
3
3
9
,
0
6
1
3
9
7
,
0
3
0
#
C
l
a
sse
s
5
4
10
#
F
l
o
w
f
e
a
t
u
r
e
s e
x
t
r
a
c
t
e
d
9
9
9
4
.
1
.
1
.
Da
t
a
s
et
prepro
ce
s
s
ing
On
li
n
e
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
an
d
o
n
li
n
e
f
ea
t
u
r
es
w
it
h
I
AT
an
d
w
it
h
o
u
t
I
A
T
as
s
u
g
g
e
s
ted
in
o
u
r
p
r
ev
io
u
s
w
o
r
k
[
2
8
]
ar
e
s
elec
ted
.
Fo
r
th
e
UNI
B
S
an
d
P
A
M
d
atasets
,
th
e
f
ea
t
u
r
es
ar
e
ex
tr
ac
ted
b
ased
o
n
th
e
f
ir
s
t
f
i
v
e
p
ac
k
e
ts
s
tatis
tic
o
f
ea
ch
f
lo
w
.
Ho
w
ev
er
,
f
o
r
th
e
C
a
m
b
r
id
g
e
d
ataset,
t
h
e
s
ta
ti
s
tics
o
f
t
h
e
f
ir
s
t
5
p
ac
k
ets
ar
e
n
o
t
av
ailab
le
w
ith
o
u
t
ac
ce
s
s
to
th
e
r
a
w
p
ac
k
et
s
.
T
h
u
s
,
f
o
r
th
is
d
ataset,
t
h
e
co
m
p
lete
f
lo
w
s
tatis
t
ic
is
u
s
ed
(
n
o
t
o
n
l
y
f
ir
s
t
5
p
ac
k
ets).
I
n
o
r
d
er
to
h
av
e
a
f
air
c
o
m
p
ar
is
o
n
o
f
al
l
d
atasets
,
t
h
e
m
ea
n
f
ea
t
u
r
es
i
n
C
a
m
b
r
id
g
e
d
atase
t a
r
e
m
o
d
if
i
ed
to
to
tal
f
ea
tu
r
es.
T
ab
le
2
s
h
o
w
s
t
h
e
li
s
t o
f
f
ea
tu
r
e
t
h
at
h
ad
b
ee
n
ex
tr
ac
ted
.
T
ab
le
2
.
On
lin
e
f
ea
t
u
r
es
w
i
th
I
A
T
#
N
a
me
D
e
scri
p
t
i
o
n
1
P
o
r
t
_
a
S
o
u
r
c
e
p
o
r
t
n
u
mb
e
r
2
P
o
r
t
_
b
P
o
r
t
b
D
e
st
i
n
a
t
i
o
n
p
o
r
t
n
u
mb
e
r
3
P
l
y
_
si
z
e
_
b
a
T
o
t
a
l
b
y
t
e
i
n
I
P
p
a
c
k
e
t
(
D
o
w
n
l
i
n
k
)
4
P
l
y
_
si
z
e
T
o
t
a
l
b
y
t
e
i
n
I
P
p
a
c
k
e
t
5
P
c
k
_
si
z
e
_
b
a
T
o
t
a
l
b
y
t
e
i
n
Et
h
e
r
n
e
t
p
a
c
k
e
t
(
D
o
w
n
l
i
n
k
)
6
P
c
k
_
si
z
e
T
o
t
a
l
b
y
t
e
i
n
Et
h
e
r
n
e
t
p
a
c
k
e
t
7
i
a
t
_
b
a
T
o
t
a
l
p
a
c
k
e
t
i
n
t
e
r
-
a
r
r
i
v
a
l
t
i
me
(
d
o
w
n
l
i
n
k
)
8
I
a
t
T
o
t
a
l
p
a
c
k
e
t
i
n
t
e
r
-
a
r
r
i
v
a
l
t
i
me
9
C
l
a
ss
P
r
o
t
o
c
o
l
4
.
1
.
2
.
E
v
a
lutio
n
P
r
eq
u
en
tial
ev
al
u
atio
n
u
s
i
n
g
f
ad
in
g
f
ac
to
r
s
f
o
r
g
etti
n
g
m
ec
h
an
is
m
p
r
o
p
o
s
ed
b
y
Ga
m
a
e
t
a
l
.
[
2
9
]
is
ad
o
p
ted
as
th
e
ev
al
u
atio
n
m
et
h
o
d
.
T
h
is
m
et
h
o
d
is
s
u
itab
le
f
o
r
ev
alu
ati
n
g
i
n
cr
e
m
e
n
tal
lear
n
in
g
a
lg
o
r
it
h
m
.
T
h
e
p
r
eq
u
en
tial p
ar
a
m
eter
s
u
s
ed
i
n
o
u
r
ex
p
er
i
m
en
t a
r
e
as st
ated
b
elo
w
,
u
n
le
s
s
s
p
ec
i
f
ied
o
th
er
w
i
s
e:
a.
C
las
s
i
f
ier
to
tr
ain
: H
o
ef
f
d
in
g
T
r
ee
b.
Stre
a
m
to
lear
n
f
r
o
m
: P
AM
,
C
a
m
b
r
id
g
e
an
d
UNI
B
S d
ataset
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
I
mp
a
ct
o
f P
a
ck
et
I
n
ter
-
a
r
r
iva
l Time
F
ea
tu
r
es fo
r
...
(
B
u
s
h
r
a
Mo
h
a
mme
d
A
li A
b
d
a
lla
)
2525
c.
T
r
ain
in
g
a
n
d
tes
tin
g
o
n
a
to
tal
o
f
2
5
0
,
0
0
0
s
a
m
p
les
f
o
r
P
A
M
a
n
d
C
a
m
b
r
id
g
e
,
w
h
ile
UNI
B
S
5
0
0
0
0
s
a
m
p
le
s
.
d.
T
esti
n
g
ev
er
y
t
w
o
h
u
n
d
r
ed
s
am
p
les
e.
I
n
s
ta
n
ce
s
b
et
w
ee
n
m
e
m
o
r
y
b
o
u
n
d
ch
ec
k
s
:
1
9
3
,
0
0
0
s
a
m
p
l
es
f
o
r
P
A
M
a
n
d
C
a
m
b
r
id
g
e
,
w
h
ile
U
NI
B
S
4
0
0
0
0
s
am
p
le
s
f.
E
v
alu
a
te
P
r
eq
u
en
tial P
ar
a
m
ete
r
s
: W
in
d
o
w
C
las
s
i
f
icatio
n
P
er
f
o
r
m
a
n
ce
E
v
al
u
ato
r
g.
Size
o
f
s
lid
i
n
g
i
s
1
,
0
0
0
T
h
e
p
er
f
o
r
m
an
c
e
i
n
d
icato
r
s
u
s
ed
in
th
is
r
esear
ch
ar
e
clas
s
i
f
icatio
n
ti
m
e
,
Kap
aa
s
tatis
tic
K
=
1
an
d
av
er
ag
e
ac
c
u
r
ac
y
(
A
c
c
)
.
Av
er
ag
e
ac
c
u
r
ac
y
i
s
t
h
e
o
v
er
all
ac
cu
r
ac
y
f
o
r
a
d
ataset.
L
e
t
th
e
to
tal
co
r
r
ec
t
id
en
ti
f
icatio
n
i
n
a
d
ataset
w
it
h
(
N
)
f
lo
w
in
s
ta
n
ce
s
is
η
.
T
h
e
p
er
f
o
r
m
an
ce
i
n
d
icato
r
s
u
s
ed
in
th
is
p
ap
er
ar
e:
(
7
)
(
8
)
w
h
ile
:
class
i
f
ier
’
s
p
r
eq
u
en
ti
al
ac
cu
r
ac
y
is
:
p
r
o
b
ab
ilit
y
o
f
co
r
r
ec
t
p
r
ed
ictio
n
.
Kap
p
a
h
as
p
r
ef
er
ab
l
e
p
r
o
p
er
ties
s
u
ch
th
a
t
v
al
u
e
o
f
1
w
it
h
p
er
f
ec
t
ag
r
ee
m
en
t
(
)
is
u
s
ed
.
T
h
e
v
alu
e
ap
p
r
o
x
i
m
ate
l
y
ze
r
o
w
h
en
th
e
o
b
s
er
v
ed
ag
r
ee
m
e
n
t
is
al
m
o
s
t
t
h
e
s
a
m
e
a
s
w
o
u
ld
b
e
e
x
p
ec
ted
b
y
c
h
an
ce
(
)
.
Fu
r
th
er
m
o
r
e,
Kap
p
a
s
tatis
t
ic
d
o
es n
o
t a
s
s
u
m
e
m
ar
g
in
al
p
r
o
b
ab
ilit
ies to
b
e
th
e
s
am
e
f
o
r
d
if
f
er
e
n
t o
b
s
er
v
er
s
.
4
.
2
.
O
ne
-
w
a
y
ANO
VA
t
est
re
s
ults
T
h
is
s
u
b
s
ec
tio
n
e
x
p
lai
n
s
t
h
e
s
ig
n
i
f
ica
n
t
o
f
s
elec
ted
f
ea
t
u
r
es
b
y
u
s
in
g
ANOV
A
te
s
t
w
i
th
9
5
%
co
n
f
id
e
n
ce
in
ter
v
al
f
o
r
th
e
m
ea
n
d
if
f
er
en
ce
.
T
h
e
r
es
u
lt
e
x
p
lain
s
a
ll
s
elec
ted
f
ea
t
u
r
es
ar
e
s
ig
n
i
f
ica
n
t
b
ec
au
s
e
af
ter
te
s
ted
w
it
h
A
N
OV
A
t
h
e
P
-
v
al
u
e
le
s
s
th
a
n
0
.
0
5
.
Al
s
o
,
th
i
s
te
s
t
e
x
p
lain
s
t
h
e
I
AT
f
ea
tu
r
es
ar
e
le
s
s
s
ig
n
i
f
ica
n
t th
a
n
o
th
er
f
ea
t
u
r
es
as sh
o
w
n
i
n
Fi
g
u
r
e
2
.
4
.
3
.
O
nli
ne
cla
s
s
if
ica
t
io
n r
es
ults
T
h
e
ex
p
er
im
e
n
tal
r
es
u
lts
p
r
es
en
ted
in
Fi
g
u
r
e
3
to
Fig
u
r
e
8
,
illu
s
tr
a
te
t
h
e
ef
f
ec
t
o
f
I
A
T
in
clu
s
io
n
as
an
o
n
li
n
e
f
ea
t
u
r
e
f
o
r
P
2
P
id
en
tif
icat
io
n
.
T
h
e
r
esu
lt
as
p
r
ese
n
ted
in
T
ab
le
3
in
d
icate
s
th
at
p
ac
k
et
I
A
T
f
ea
tu
r
e
as
o
n
li
n
e
f
ea
t
u
r
e
d
ec
r
ea
s
es
i
d
en
tific
atio
n
ac
c
u
r
ac
y
an
d
K
ap
aa
s
tatis
t
ic.
F
u
r
th
er
m
o
r
e,
p
ac
k
et
I
A
T
f
ea
t
u
r
e
in
cr
ea
s
es
t
h
e
e
x
p
er
i
m
e
n
tal
e
v
a
lu
atio
n
ti
m
e.
T
h
is
is
a
s
a
r
esu
l
t o
f
p
ac
k
et
I
A
T
f
ea
t
u
r
e
m
o
r
p
h
in
g
w
h
ich
i
n
v
o
l
v
es
alter
n
atio
n
o
n
d
ir
ec
tio
n
p
atter
n
w
h
ich
i
s
d
ep
en
d
en
t
o
n
n
et
w
o
r
k
lo
ca
tio
n
s
.
A
l
s
o
th
e
s
e
r
esu
lt
s
p
r
o
v
e
p
r
ev
io
u
s
o
f
f
li
n
e
s
tu
d
ies t
h
at:
a.
T
im
e
-
r
elate
d
f
ea
t
u
r
es d
o
n
o
t h
elp
to
d
is
tin
g
u
is
h
a
m
o
n
g
ap
p
licatio
n
s
[
2
0
]
,
[
3
0
]
.
b
.
T
h
e
u
s
e
an
d
s
tatis
tica
l
f
ea
t
u
r
es
o
f
ap
p
licatio
n
d
ep
en
d
en
t
o
n
l
y
o
n
in
ter
-
p
ac
k
et
ti
m
e
is
a
ch
alle
n
g
i
n
g
tas
k
d
u
e
to
th
e
ti
m
e
r
eq
u
ir
ed
b
y
an
ap
p
licatio
n
to
g
en
er
ate
an
d
tr
an
s
f
er
p
ac
k
ets
to
th
e
tr
an
s
p
o
r
t
lay
er
is
m
a
s
k
ed
b
y
t
h
e
f
ac
t th
a
t a
d
d
itio
n
al
ti
m
e
is
ad
d
ed
d
u
e
to
th
e
n
et
w
o
r
k
co
n
d
itio
n
s
a
n
d
th
e
T
C
P
lay
er
[
3
1
]
.
T
ab
le
3
.
C
lass
if
icatio
n
R
es
u
lts
C
a
m
b
r
i
d
g
e
O
n
l
i
n
e
f
e
a
t
u
r
e
s w
i
t
h
o
u
t
I
A
T
O
n
l
i
n
e
f
e
a
t
u
r
e
s w
i
t
h
I
A
T
A
c
c
u
r
a
c
y
me
a
n
9
8
.
8
6
9
8
.
8
0
K
a
p
p
a
s
t
a
t
i
s
t
i
c
me
a
n
5
9
.
4
6
5
9
.
4
6
C
P
U
t
o
t
a
l
t
i
me
p
e
r
se
c
o
n
d
me
a
n
1
.
4
1
2
.
2
8
U
N
I
B
S
d
a
t
a
se
t
O
n
l
i
n
e
f
e
a
t
u
r
e
s w
i
t
h
o
u
t
I
A
T
O
n
l
i
n
e
f
e
a
t
u
r
e
s w
i
t
h
I
A
T
A
c
c
u
r
a
c
y
me
a
n
9
4
.
1
3
9
3
.
1
5
K
a
p
p
a
s
t
a
t
i
s
t
i
c
me
a
n
8
7
.
8
7
8
6
.
1
6
C
P
U
t
o
t
a
l
t
i
me
p
e
r
se
c
o
n
d
me
a
n
0
.
5
1
0
.
7
4
P
A
M
d
a
t
a
se
t
O
n
l
i
n
e
f
e
a
t
u
r
e
s w
i
t
h
o
u
t
I
A
T
O
n
l
i
n
e
f
e
a
t
u
r
e
s w
i
t
h
I
A
T
A
c
c
u
r
a
c
y
me
a
n
9
2
.
4
2
9
2
.
2
2
K
a
p
p
a
s
t
a
t
i
s
t
i
c
me
a
n
1
7
.
7
1
1
5
.
4
2
C
P
U
t
o
t
a
l
t
i
me
p
e
r
se
c
o
n
d
me
a
n
1
.
5
6
1
.
8
2
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
5
2
1
–
2
5
3
0
2526
Fig
u
r
e
2
.
Scr
ee
n
s
h
o
t o
f
test
s
t
atis
tic
A
NO
V
A
Fig
u
r
e
3
.
UNI
B
S d
ataset
m
ea
n
clas
s
if
icatio
n
ac
c
u
r
ca
y
Fig
u
r
e
4
.
UNI
B
S d
ataset
m
ea
n
k
ap
aa
s
tat
is
tic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
I
mp
a
ct
o
f P
a
ck
et
I
n
ter
-
a
r
r
iva
l Time
F
ea
tu
r
es fo
r
...
(
B
u
s
h
r
a
Mo
h
a
mme
d
A
li A
b
d
a
lla
)
2527
Fig
u
r
e
5
.
UNI
B
S d
ataset
ev
al
u
tio
n
ti
m
e
Fig
u
r
e
6
.
P
A
M
d
ataset
m
ea
n
c
lass
i
f
icatio
n
ac
cu
r
ca
y
Fig
u
r
e
7
.
P
A
M
d
ataset
m
ea
n
k
ap
aa
s
tatis
tic
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
5
2
1
–
2
5
3
0
2528
Fig
u
r
e
8
.
P
A
M
d
ataset
ev
al
u
ti
o
n
ti
m
e
5.
CO
NCLU
SI
O
N
I
n
th
is
p
ap
er
,
w
e
i
n
v
e
s
ti
g
ated
th
e
i
m
p
ac
t
o
f
p
ac
k
et
I
A
T
f
e
atu
r
e
f
o
r
o
n
lin
e
P
2
P
class
if
ic
atio
n
w
it
h
r
ef
er
en
ce
to
ac
c
u
r
ac
y
,
k
ap
p
a
s
tatis
tic
a
n
d
e
v
al
u
atio
n
ti
m
e.
T
h
e
s
i
m
u
latio
n
r
es
u
lt
s
i
n
d
icate
th
at
t
h
e
p
ac
k
et
I
A
T
f
ea
t
u
r
e
s
f
o
r
o
n
li
n
e
P
2
P
class
if
icatio
n
d
ec
r
ea
s
e
ac
c
u
r
ac
y
a
n
d
Kap
p
a
s
tat
is
tic,
an
d
a
ls
o
in
cr
ea
s
e
e
v
al
u
atio
n
ti
m
e.
T
h
ese
r
esu
lt
s
b
ec
au
s
e
I
A
T
m
o
r
p
h
in
g
u
s
u
al
l
y
i
n
v
o
l
v
e
s
alter
n
atio
n
o
n
d
ir
ec
tio
n
p
att
er
n
an
d
d
ep
en
d
o
n
d
if
f
er
e
n
t
n
et
w
o
r
k
lo
ca
tio
n
s
.
T
h
e
ac
k
n
o
w
led
g
m
e
n
t
s
ec
t
io
n
is
o
p
tio
n
al.
T
h
e
f
u
n
d
i
n
g
s
o
u
r
ce
o
f
th
e
r
esear
ch
ca
n
b
e
p
u
t h
er
e.
RE
F
E
R
E
NC
E
S
[1
]
D.
L
.
Jo
h
n
so
n
,
E.
M
.
Be
ld
i
n
g
,
a
n
d
G
.
V
a
n
S
tam
,
“
Ne
t
w
o
rk
tra
ff
i
c
lo
c
a
li
ty
in
a
ru
ra
l
a
f
rica
n
v
il
la
g
e
,
”
in
Pro
c
e
e
d
in
g
s
o
f
t
h
e
fi
ft
h
in
ter
n
a
ti
o
n
a
l
c
o
n
fer
e
n
c
e
o
n
in
f
o
rm
a
ti
o
n
a
n
d
c
o
mm
u
n
ica
ti
o
n
tec
h
n
o
lo
g
ies
a
n
d
d
e
v
e
lo
p
m
e
n
t.
ACM
,
2
0
1
2
,
p
p
.
2
6
8
-
2
7
7
.
[2
]
J.
S
.
Ba
ss
i,
L
.
H.
Ru
,
I.
Ism
a
il
,
B.
M
.
Kh
a
m
m
a
s,
a
n
d
M
.
N.
M
a
rso
n
o
,
“
On
li
n
e
p
e
e
r
-
to
-
p
e
e
r
traf
f
i
c
id
e
n
ti
f
ica
ti
o
n
.
b
a
se
d
o
n
c
o
m
p
lex
e
v
e
n
ts
p
ro
c
e
ss
in
g
o
f
traff
ic
e
v
e
n
t
sig
n
a
tu
re
s,”
J
o
u
rn
a
l
T
EKNO
L
OG
I
,
v
o
l.
7
8
,
n
o
.
7
,
p
p
.
9
-
1
6
,
2
0
1
6
.
[3
]
R.
D.
T
o
rre
s,
M
.
Y.
Ha
jj
a
t,
S
.
G
.
Ra
o
,
M
.
M
e
ll
ia,
a
n
d
M
.
M
.
M
u
n
a
f
o
,
“
In
f
e
rrin
g
u
n
d
e
sira
b
le
b
e
h
a
v
io
r
f
ro
m
p
2
p
traff
ic
a
n
a
l
y
si
s,”
b
o
o
k
ti
tl
e
is
A
CM
S
IG
M
ET
RICS
P
e
rf
o
rm
a
n
c
e
Ev
a
lu
a
ti
o
n
Re
v
iew
,
v
o
l.
3
7
,
n
o
.
1
.
A
CM
,
2
0
0
9
,
p
p
.
2
5
-
3
6
.
[4
]
C.
W
a
n
g
,
Z.
W
a
n
g
,
Z.
Ye
,
a
n
d
H
.
Ch
e
n
,
“
A
p
2
p
traf
f
ic
id
e
n
ti
f
ica
ti
o
n
a
p
p
ro
a
c
h
b
a
se
d
o
n
sv
m
a
n
d
b
fa
,
”
In
d
o
n
e
si
a
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
, v
o
l.
1
2
,
n
o
.
4
,
p
p
.
2
8
3
3
-
2
8
4
2
,
2
0
1
4
.
[5
]
A
.
W
.
M
o
o
re
a
n
d
K.
P
a
p
a
g
ian
n
a
k
i,
“
T
o
w
a
rd
th
e
a
c
c
u
ra
te
id
e
n
ti
f
ica
ti
o
n
o
f
n
e
tw
o
rk
a
p
p
li
c
a
ti
o
n
s,”
i
n
In
ter
n
a
ti
o
n
a
l
W
o
rk
sh
o
p
o
n
Pa
ss
ive
a
n
d
Active
Ne
two
rk
M
e
a
su
re
me
n
t.
S
p
ri
n
g
e
r,
2
0
0
5
,
p
p
.
4
1
-
5
4
.
[6
]
P
.
V
a
n
De
r
P
u
tt
e
n
a
n
d
M
.
V
a
n
S
o
m
e
re
n
,
“
A
b
ias
-
v
a
rian
c
e
a
n
a
ly
sis
o
f
a
re
a
l
w
o
rld
lea
rn
i
n
g
p
r
o
b
lem
:
T
h
e
c
o
i
l
c
h
a
ll
e
n
g
e
2
0
0
0
,
”
J
o
u
rn
a
l
o
f
M
a
c
h
in
e
L
e
a
rn
i
n
g
,
v
o
l.
5
7
,
n
o
.
1
-
2
,
p
p
.
1
7
7
-
1
9
5
,
2
0
0
4
.
[7
]
H.
A
.
Ja
m
il
,
A
.
M
o
h
a
m
m
e
d
,
A
.
Ha
m
z
a
,
S
.
M
.
No
r,
a
n
d
M
.
N.
M
a
rso
n
o
,
“
S
e
lec
ti
o
n
o
f
o
n
-
li
n
e
f
e
a
t
u
re
s
f
o
r
p
e
e
r
-
to
-
p
e
e
r
n
e
tw
o
rk
tra
ff
ic
c
la
ss
i
f
ic
a
ti
o
n
,
”
i
n
Rec
e
n
t
Ad
v
a
n
c
e
s
in
In
tell
ig
e
n
t
In
f
o
rm
a
ti
c
s.
S
p
rin
g
e
r,
2
0
1
4
,
pp.
3
7
9
-
3
9
0
.
[8
]
H.
R.
L
o
o
a
n
d
M
.
N.
M
a
rso
n
o
,
“
On
li
n
e
n
e
tw
o
rk
tra
ff
ic
c
las
sifica
ti
o
n
w
it
h
in
c
re
m
e
n
tal
lea
rn
in
g
,
”
J
o
u
rn
a
l
o
f
Evo
lvin
g
S
y
ste
ms
,
v
o
l.
7
,
n
o
.
2
,
p
p
.
1
2
9
-
1
4
3
,
2
0
1
6
.
[9
]
H.
Zh
a
n
g
,
e
t
a
l
.
,
“F
e
a
tu
re
se
lec
ti
o
n
f
o
r
o
p
ti
m
izin
g
tra
ff
ic
c
las
sif
ic
a
ti
o
n
,
”
J
o
u
rn
a
l
o
f
C
o
mp
u
ter
Co
mm
u
n
ica
ti
o
n
s
,
v
o
l.
3
5
,
n
o
.
1
2
,
p
p
.
1
4
5
7
-
1
4
7
1
,
2
0
1
2
.
[1
0
]
S
.
Jo
se
p
h
,
e
t
a
l
.,
“
Co
o
p
e
ra
ti
v
e
lea
r
n
in
g
f
o
r
d
istri
b
u
ted
i
n
-
n
e
tw
o
rk
traff
ic
c
las
si
f
ic
a
ti
o
n
,
”
in
IOP
Co
n
fer
e
n
c
e
S
e
rie
s:
M
a
ter
ia
ls
S
c
ien
c
e
a
n
d
En
g
in
e
e
rin
g
,
v
o
l
.
1
9
0
,
n
o
.
1
.
IOP
Pu
b
li
sh
in
g
,
2
0
1
7
,
p
.
0
1
2
0
1
0
.
[1
1
]
J.
-
j.
Zh
a
o
,
e
t
a
l
.
,
“
Re
a
l
-
ti
m
e
f
e
a
t
u
re
se
lec
ti
o
n
in
traf
f
ic
c
l
a
ss
i
f
ica
ti
o
n
,
”
J
o
u
rn
a
l
o
f
C
h
in
a
Un
ive
rs
it
i
e
s
o
f
Po
sts
a
n
d
T
e
lec
o
mm
u
n
ica
ti
o
n
s
,
v
o
l.
1
5
,
p
p
.
6
8
-
7
2
,
2
0
0
8
.
[1
2
]
T
.
T
.
Ng
u
y
e
n
a
n
d
G
.
A
r
m
it
a
g
e
,
“
A
su
rv
e
y
o
f
tec
h
n
iq
u
e
s
f
o
r
in
tern
e
t
traff
ic
c
las
si
f
ic
a
ti
o
n
u
si
n
g
m
a
c
h
in
e
lea
rn
in
g
,
”
J
o
u
rn
a
l
IEE
E
Co
mm
u
n
ica
t
io
n
s S
u
rv
e
y
s
&
T
u
to
ria
ls
,
v
o
l.
1
0
,
n
o
.
4
,
p
p
.
5
6
-
7
6
,
2
0
0
8
.
[1
3
]
M
.
G
.
S
c
h
u
lt
z
,
e
t
a
l
.
,
“
Da
ta
m
in
in
g
m
e
th
o
d
s
f
o
r
d
e
tec
ti
o
n
o
f
n
e
w
m
a
li
c
io
u
s
e
x
e
c
u
tab
les
,
”
in
S
e
c
u
rity
a
n
d
Priva
c
y
,
2
0
0
1
.
S
&
P
2
0
0
1
.
Pro
c
e
e
d
in
g
s.
2
0
0
1
IE
EE
S
y
mp
o
si
u
m o
n
.
IE
EE
,
2
0
0
1
,
p
p
.
3
8
-
4
9
.
[1
4
]
G.
T
a
h
a
n
,
e
t
a
l
.,
“
M
a
l
-
id
:
A
u
to
m
a
ti
c
m
a
l
wa
re
d
e
tec
ti
o
n
u
sin
g
c
o
m
m
o
n
se
g
m
e
n
t
a
n
a
l
y
sis
a
n
d
m
e
ta
-
f
e
a
tu
re
s,”
J
o
u
rn
a
l
o
f
M
a
c
h
i
n
e
L
e
a
rn
in
g
Res
e
a
rc
h
,
v
o
l.
1
3
,
n
o
.
A
p
r,
p
p
.
9
4
9
-
9
7
9
,
2
0
1
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J
E
lec
&
C
o
m
p
E
n
g
I
SS
N:
2
0
8
8
-
8708
I
mp
a
ct
o
f P
a
ck
et
I
n
ter
-
a
r
r
iva
l Time
F
ea
tu
r
es fo
r
...
(
B
u
s
h
r
a
Mo
h
a
mme
d
A
li A
b
d
a
lla
)
2529
[1
5
]
A
.
M
o
o
re
,
e
t
a
l
.,
“
Cro
g
a
n
,
Disc
ri
m
in
a
to
rs
f
o
r
u
se
in
f
lo
w
-
b
a
s
e
d
c
l
a
ss
if
i
c
a
ti
o
n
,
”
Qu
e
e
n
M
a
ry
a
n
d
W
e
stfi
e
ld
Co
ll
e
g
e
,
De
p
a
rtme
n
t
o
f
Co
m
p
u
ter
S
c
ien
c
e
,
T
e
c
h
n
ica
l
Rep
o
rt ,
2
0
0
5
.
[1
6
]
T
.
A
u
ld
,
e
t
a
l
.
,
“
Ba
y
e
sia
n
n
e
u
ra
l
n
e
tw
o
rk
s
f
o
r
in
tern
e
t
traff
i
c
c
las
si
f
ica
ti
o
n
,
”
J
o
u
rn
a
l
IEE
E
T
ra
n
s
a
c
t
io
n
s
o
n
n
e
u
ra
l
n
e
two
rk
s
,
v
o
l.
1
8
,
n
o
.
1
,
p
p
.
2
2
3
-
2
3
9
,
2
0
0
7
.
[1
7
]
A
.
T
jah
y
a
n
to
,
e
t
a
l
.
,
“
S
p
e
c
tral
-
b
a
se
d
f
e
a
tu
re
s
ra
n
k
in
g
f
o
r
g
a
m
e
lan
in
stru
m
e
n
ts
id
e
n
ti
f
ica
ti
o
n
u
sin
g
f
il
ter
tec
h
n
iq
u
e
s,”
J
o
u
rn
a
l
T
EL
KOM
NIKA
(
T
e
lec
o
mm
u
n
ica
ti
o
n
Co
m
p
u
ti
n
g
El
e
c
tro
n
ics
a
n
d
Co
n
tro
l)
,
v
o
l.
1
1
,
n
o
.
1
,
p
p
.
9
5
-
1
0
6
,
2
0
1
3
.
[1
8
]
O.
He
n
c
h
iri
a
n
d
N.
Ja
p
k
o
w
icz
,
“
A
f
e
a
tu
re
se
lec
ti
o
n
a
n
d
e
v
a
lu
a
ti
o
n
sc
h
e
m
e
f
o
r
c
o
m
p
u
ter
v
iru
s
d
e
tec
ti
o
n
,
”
i
n
D
a
t
a
M
in
in
g
,
2
0
0
6
.
ICDM
’0
6
.
S
ixth
I
n
t
e
rn
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
.
IEE
E
,
2
0
0
6
,
p
p
.
8
9
1
-
8
9
5
.
[1
9
]
A
.
M
o
n
e
m
i
,
e
t
a
l
.
,
“
On
li
n
e
n
e
tf
p
g
a
d
e
c
isio
n
tree
sta
ti
stica
l
traff
i
c
c
las
si
f
ier,”
J
o
u
rn
a
l
o
f
Co
m
p
u
ter
Co
mm
u
n
ica
ti
o
n
s
,
v
o
l
.
3
6
,
n
o
.
1
2
,
p
p
.
1
3
2
9
-
1
3
4
0
,
2
0
1
3
.
[2
0
]
L
.
Zh
e
n
a
n
d
L
.
Q
io
n
g
,
“
A
n
e
w
f
e
a
tu
re
se
lec
ti
o
n
m
e
th
o
d
f
o
r
in
tern
e
t
traff
ic
c
las
sif
i
c
a
ti
o
n
u
sin
g
m
l,
”
J
o
u
rn
a
l
o
f
Ph
y
sic
s
Pro
c
e
d
ia
,
v
o
l.
3
3
,
p
p
.
1
3
3
8
-
1
3
4
5
,
2
0
1
2
.
[2
1
]
Y.
Zh
a
o
,
e
t
a
l
.
,
“
Hie
ra
rc
h
ica
l
re
a
l
-
ti
m
e
n
e
tw
o
rk
tra
ff
ic
c
las
si
f
ica
ti
o
n
b
a
se
d
o
n
e
c
o
c
,
”
In
d
o
n
e
sia
n
J
o
u
rn
a
l
o
f
El
e
c
trica
l
En
g
in
e
e
rin
g
a
n
d
Co
m
p
u
ter
S
c
ien
c
e
,
v
o
l.
1
2
,
n
o
.
2
,
p
p
.
1
5
5
1
-
1
5
6
0
,
2
0
1
4
.
[2
2
]
h
.
KN
IM
E,
“
,
”
h
tt
p
s://
tec
h
.
k
n
im
e
.
o
rg
/f
o
ru
m
/b
io
in
f
o
rm
a
ti
c
s/,
”
A
c
c
e
c
e
d
2
2
De
c
2
0
1
6
”
.
[2
3
]
h
.
W
EKA
,
“
,
”
h
tt
p
:/
/w
ww
.
c
s.
wa
i
k
a
to
.
a
c
.
n
z
/m
l/
w
e
k
a
/,
”
A
c
c
e
c
e
d
0
1
De
c
2
0
1
6
.
”
.
[2
4
]
L
.
L
.
Ku
p
p
e
r,
“
A
p
p
li
e
d
re
g
re
ss
io
n
a
n
a
ly
sis a
n
d
o
th
e
r
m
u
lt
iv
a
riate
m
e
th
o
d
s
,”
Du
x
b
u
ry
Pre
ss
,
B
o
sto
n
,
1
9
7
8
[2
5
]
A
.
Bifet
a
n
d
R.
Kirk
b
y
,
“
Da
ta
st
re
a
m
m
in
in
g
a
p
ra
c
ti
c
a
l
a
p
p
r
o
a
c
h
,
”
Un
ive
rs
it
y
o
f
W
AIKA
T
O,
T
e
c
h
n
ica
l
Rep
o
rt
,
2
0
0
9
.
[2
6
]
V
.
Ca
re
la
-
Esp
a
n
o
l,
e
t
a
l
.
,
“
Is
o
u
r
g
ro
u
n
d
-
tru
t
h
f
o
r
traf
f
ic
c
las
si
f
ica
ti
o
n
re
li
a
b
le
?
”
in
˜
I
n
ter
n
a
ti
o
n
a
l
Co
n
fer
e
n
c
e
o
n
Pa
ss
ive
a
n
d
Active
Ne
two
rk
M
e
a
su
re
me
n
t.
S
p
rin
g
e
r,
2
0
1
4
,
p
p
.
9
8
-
1
0
8
.
[2
7
]
F
.
G
rin
g
o
li
,
e
t
a
l
.,
“
G
t:
p
ick
in
g
u
p
t
h
e
tru
t
h
f
ro
m
th
e
g
ro
u
n
d
f
o
r
in
tern
e
t
traf
f
ic,”
J
o
u
rn
a
l
o
f
A
CM
S
IGCO
M
M
Co
mp
u
ter
C
o
m
mu
n
ica
ti
o
n
Rev
ie
w
,
v
o
l.
3
9
,
n
o
.
5
,
p
p
.
1
2
-
1
8
,
2
0
0
9
.
[2
8
]
B.
M
.
A
.
A
b
d
a
ll
a
,
e
t
a
l
.
,
“
M
u
lt
i
-
s
tag
e
fe
a
tu
re
se
lec
ti
o
n
f
o
r
o
n
-
li
n
e
f
lo
w
p
e
e
r
-
to
-
p
e
e
r
traff
ic
id
e
n
ti
f
ic
a
ti
o
n
,
”
in
Asia
n
S
imu
l
a
t
io
n
C
o
n
fer
e
n
c
e
.
S
p
rin
g
e
r,
2
0
1
7
,
p
p
.
5
0
9
-
5
2
3
.
[2
9
]
J.
G
a
m
a
,
e
t
a
l
.
,
“
On
e
v
a
lu
a
ti
n
g
stre
a
m
lea
rn
in
g
a
lg
o
rit
h
m
s,”
J
o
u
rn
a
l
o
f
M
a
c
h
i
n
e
lea
rn
i
n
g
,
v
o
l.
9
0
,
n
o
.
3
,
p
p
.
3
1
7
-
3
4
6
,
2
0
1
3
.
[3
0
]
J.
Kö
g
e
l,
”
On
e
-
w
a
y
d
e
la
y
m
e
a
s
u
re
m
e
n
t
b
a
se
d
o
n
f
lo
w
d
a
ta
in
larg
e
e
n
terp
rise
n
e
t
w
o
rk
s
,”
Ph
D
th
e
sis
.
Un
iv
.
S
tu
tt
g
a
rt,
I
n
st.
F
ü
r
K
o
m
m
u
n
ik
a
ti
o
n
sn
e
tze
u
n
d
Re
c
h
n
e
rsy
ste
m
e
,
2
0
1
3
.
[3
1
]
B.
Qu
,
e
t
a
l
.
,
“
A
n
e
m
p
iri
c
a
l
st
u
d
y
o
f
m
o
rp
h
in
g
o
n
b
e
h
a
v
io
r
-
b
a
se
d
n
e
tw
o
rk
tra
ff
ic
c
la
ss
i
f
ica
ti
o
n
,
”
J
o
u
r
n
a
l
o
f
S
e
c
u
rity a
n
d
C
o
mm
u
n
ica
ti
o
n
Ne
t
wo
rk
s
,
v
o
l.
8
,
n
o
.
1
,
p
p
.
6
8
-
7
9
,
2
0
1
5
.
B
I
O
G
RAP
H
I
E
S
O
F
AUTH
O
RS
B
u
sh
r
a
M
o
h
a
m
m
e
d
Ali
,
is
a
P
h
D
c
a
n
d
id
a
te
a
t
th
e
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
Un
iv
e
r
siti
T
e
k
n
o
lo
g
i
M
a
lay
sia
.
H
e
o
b
tain
e
d
B.
S
c
.
a
n
d
M
.
S
c
.
in
C
o
m
p
u
ter
En
g
in
e
e
rin
g
a
n
d
Ne
tw
o
rk
s
-
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
a
n
d
T
e
c
h
n
o
lo
g
y
,
Un
iv
e
rsit
y
o
f
Ge
z
ira,
S
u
d
a
n
.
He
is
a
lec
t
u
re
r
a
t
th
e
f
a
c
u
lt
y
o
f
Co
m
p
u
ter
a
n
d
S
tatisti
c
s
S
tu
d
ies
,
U
n
iv
e
rsity
o
f
Ko
rd
o
f
a
n
.
His
re
se
a
rc
h
in
ter
e
sts
in
c
lu
d
e
c
o
m
p
u
ter
a
rc
h
it
e
c
tu
r
e
,
Ne
tw
o
rk
T
ra
ff
ic cl
a
ss
i
f
ica
ti
o
n
a
n
d
c
o
n
tro
l
,
A
rti
f
icia
l
In
telli
g
e
n
c
e
a
n
d
o
p
t
im
iza
ti
o
n
tec
h
n
i
q
u
e
s.
M
o
sa
b
H
a
m
d
a
n
is
a
P
h
D
st
u
d
e
n
t
a
t
V
e
c
a
d
re
se
a
rc
h
g
ro
u
p
i
n
Un
iv
e
rsity
T
e
c
h
n
o
lo
g
y
M
a
la
y
sia
(UT
M
).
He
o
b
tain
e
d
Ba
c
h
e
lo
r
d
e
g
re
e
in
El
e
c
tro
n
ic
a
n
d
El
e
c
tri
c
a
l
e
n
g
in
e
e
rin
g
f
ro
m
Un
iv
e
rsit
y
o
f
S
c
ien
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
(S
u
d
a
n
)
in
2
0
1
0
,
a
n
d
M
sc
in
c
o
m
p
u
ter
a
rc
h
it
e
c
tu
re
a
n
d
n
e
tw
o
rk
i
n
g
a
t
Un
iv
e
rsit
y
o
f
Kh
a
rto
u
m
(S
u
d
a
n
)
in
2
0
1
4
.
His
c
u
rre
n
t
re
se
a
r
c
h
in
tere
sts
a
re
S
o
f
t
wa
re
D
e
f
i
n
e
d
Ne
tw
o
rk
in
g
,
L
o
a
d
Ba
lan
c
in
g
,
T
ra
ff
ic Cl
a
ss
i
f
ic
a
ti
o
n
,
a
n
d
F
u
t
u
re
Ne
tw
o
rk
.
M
o
h
a
m
m
e
d
S
u
lta
n
M
o
h
a
m
m
e
d
re
c
e
iv
e
d
h
is
BS
c
in
C
o
m
p
u
ter
E
n
g
in
e
e
rin
g
f
ro
m
Ho
d
e
id
a
h
Un
iv
e
rsit
y
(Ye
m
e
n
)
in
2
0
0
5
.
He
re
c
e
iv
e
d
th
e
M
S
c
in
Co
m
p
u
ter
En
g
in
e
e
rin
g
a
n
d
Ne
tw
o
rk
s
f
ro
m
th
e
Un
iv
e
rsit
y
o
f
Jo
rd
a
n
(Jo
rd
a
n
)
i
n
2
0
1
5
.
He
is
c
u
rre
n
t
ly
p
u
rsu
in
g
h
i
s
P
h
.
D.
stu
d
y
a
t
Un
iv
e
siti
Tek
n
o
lo
g
i
M
a
la
y
sia
.
His
re
se
a
rc
h
in
tere
sts
a
re
p
a
ra
ll
e
l
p
ro
c
e
ss
in
g
,
m
u
lt
i
-
c
o
re
e
m
b
e
d
d
e
d
sy
ste
m
s,
S
y
ste
m
-
on
-
Ch
ip
(
S
o
C).
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
0
8
8
-
8708
I
n
t J
E
lec
&
C
o
m
p
E
n
g
,
Vo
l.
8
,
No
.
4
,
A
u
g
u
s
t 2
0
1
8
:
2
5
2
1
–
2
5
3
0
2530
J
o
se
p
h
S
te
p
h
e
n
B
a
ss
i
re
c
e
iv
e
d
h
is
P
h
.
D.
d
e
g
re
e
in
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
f
ro
m
th
e
Un
iv
e
rs
it
i
T
e
k
n
o
lo
g
i
M
a
lay
sia
,
in
2
0
1
7
,
M
.
En
g
.
d
e
g
re
e
in
El
e
c
tri
c
a
l
&
El
e
c
tro
n
ics
En
g
i
n
e
e
rin
g
(El
e
c
tro
n
ic
s)
f
ro
m
Un
iv
e
rsit
y
o
f
M
a
id
u
g
u
ri,
Nig
e
ria
in
2
0
1
2
a
n
d
B.
T
e
c
h
d
e
g
re
e
in
Co
m
p
u
ter
S
c
ien
c
e
&
M
a
th
e
m
a
ti
c
s
f
ro
m
F
e
d
e
ra
l
Un
iv
e
rsit
y
o
f
Tec
h
n
o
lo
g
y
M
in
n
a
,
Nig
e
ria
in
2
0
0
0
.
He
is
c
u
rre
n
tl
y
a
L
e
c
tu
re
r
w
it
h
th
e
De
p
a
rt
m
e
n
t
o
f
Co
m
p
u
te
r
En
g
in
e
e
rin
g
,
F
a
c
u
lt
y
o
f
En
g
in
e
e
rin
g
,
Un
iv
e
rsit
y
o
f
M
a
id
u
g
u
ri
,
Nig
e
ria.
His
re
se
a
rc
h
in
tere
sts
a
re
in
Ne
tw
o
rk
a
lg
o
rit
h
m
ic,
A
rti
f
icia
l
In
telli
g
e
n
c
e
&
o
p
ti
m
iza
ti
o
n
tec
h
n
iq
u
e
s an
d
c
o
m
p
u
ter co
m
m
u
n
ica
ti
o
n
n
e
tw
o
rk
s.
Is
m
a
h
a
n
i
Is
m
a
il
re
c
e
i
v
e
d
h
e
r
P
h
D
d
e
g
re
e
in
El
e
c
tri
c
a
l
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
f
ro
m
Un
iv
e
r
siti
T
e
k
n
o
lo
g
i
M
a
lay
sia
in
2
0
1
3
.
S
h
e
is
a
S
e
n
io
r
L
e
c
tu
re
r
w
it
h
th
e
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Un
iv
e
rsiti
T
e
k
n
o
lo
g
i
M
a
lay
sia
.
He
r
f
ield
s are
in
d
ig
it
a
l
sy
ste
m
s an
d
n
e
tw
o
rk
a
lg
o
rit
h
m
ic.
M
u
h
a
m
m
a
d
N
a
d
z
ir
B
in
M
a
r
so
n
o
re
c
e
iv
e
d
th
e
P
h
D
i
n
El
e
c
tri
c
a
l
a
n
d
Co
m
p
u
ter
En
g
in
e
e
rin
g
.
Un
iv
e
rsit
y
o
f
V
icto
ria,
Ca
n
a
d
a
in
2
0
0
7
.
He
is
n
o
w
a
n
A
ss
o
c
iate
P
ro
f
e
ss
o
r
a
t
th
e
F
a
c
u
lt
y
o
f
El
e
c
tri
c
a
l
En
g
in
e
e
rin
g
,
Un
iv
e
risti
T
e
k
n
o
lo
g
i
M
a
la
y
sia
.
His
r
e
se
a
r
c
h
in
tere
sts
in
s
y
st
e
m
le
v
e
l
in
teg
ra
ti
o
n
,
w
o
rk
in
g
in
m
u
lt
ip
le
a
re
a
s
o
f
e
m
b
e
d
d
e
d
sy
ste
m
s,
sp
e
c
ializ
e
d
c
o
m
p
u
ter
a
rc
h
it
e
c
tu
re
s,
V
L
S
I
d
e
sig
n
,
n
e
tw
o
rk
a
lg
o
rit
h
m
ic,
n
e
t
w
o
rk
-
on
-
c
h
ip
,
a
n
d
n
e
tw
o
rk
p
ro
c
e
ss
o
r
a
rc
h
i
tec
tu
re
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.