I
nd
o
ne
s
ia
n J
o
urna
l o
f
E
lect
rica
l En
g
ineering
a
nd
Co
m
pu
t
er
Science
Vo
l.
42
,
No
.
3
,
J
u
n
e
2
0
2
6
,
p
p
.
8
4
6
~8
5
5
I
SS
N:
2
5
0
2
-
4
7
5
2
,
DOI
: 1
0
.
1
1
5
9
1
/ijeecs.v
42
.i
3
.
pp
846
-
8
5
5
846
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ee
cs.ia
esco
r
e.
co
m
Deep
lea
rni
ng
in
cryp
tana
ly
sis
a
co
mprehens
iv
e
r
ev
iew
of
techniqu
es,
a
pplic
a
tions
,
cha
lleng
es
,
a
nd
f
uture
traj
e
ctories
O
us
s
a
m
a
No
ui
1
,
Am
ine
B
a
rk
a
t
2
1
F
u
n
d
a
me
n
t
a
l
Te
a
c
h
i
n
g
D
e
p
a
r
t
m
e
n
t
,
F
a
c
u
l
t
y
o
f
S
c
i
e
n
c
e
s
o
f
M
a
t
t
e
r
,
U
n
i
v
e
r
si
t
y
o
f
B
a
t
n
a
1
,
B
a
t
n
a
,
A
l
g
e
r
i
a
2
LR
E
3
S
2
L
a
b
o
r
a
t
o
r
y
,
H
i
g
h
e
r
N
a
t
i
o
n
a
l
S
c
h
o
o
l
o
f
R
e
n
e
w
a
b
l
e
E
n
e
r
g
i
e
s
,
En
v
i
r
o
n
m
e
n
t
a
nd
S
u
st
a
i
n
a
b
l
e
D
e
v
e
l
o
p
m
e
n
t
,
D
j
e
r
ma
,
A
l
g
e
r
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
2
0
,
2
0
2
5
R
ev
is
ed
Ma
r
7
,
2
0
2
6
Acc
ep
ted
Ma
y
2
6
,
2
0
2
6
Th
e
in
teg
ra
ti
o
n
of
d
e
e
p
lea
rn
i
n
g
(DL)
in
t
o
c
ry
p
tan
a
ly
sis
re
p
re
se
n
ts
a
p
a
ra
d
ig
m
sh
ift
,
c
h
a
ll
e
n
g
i
n
g
t
r
a
d
it
io
n
a
l
m
a
th
e
m
a
ti
c
a
l
a
p
p
ro
a
c
h
e
s
a
n
d
u
n
l
o
c
k
i
n
g
n
o
v
e
l
a
tt
a
c
k
v
e
c
to
r
s
a
g
a
in
st
c
r
y
p
t
o
g
ra
p
h
ic
a
l
g
o
ri
th
m
s
a
n
d
imp
lem
e
n
tatio
n
s.
T
h
is
c
o
m
p
re
h
e
n
siv
e
re
v
iew
s
y
n
t
h
e
siz
e
s
fin
d
i
n
g
s
fr
o
m
re
c
e
n
t
p
iv
o
tal
p
u
b
li
c
a
ti
o
n
s
to
p
r
o
v
i
d
e
an
in
-
d
e
p
th
a
n
a
l
y
sis
of
th
e
c
u
rre
n
t
sta
te
-
of
-
th
e
-
a
rt,
m
e
th
o
d
o
lo
g
ies
,
e
m
p
iri
c
a
l
su
c
c
e
ss
e
s,
fu
n
d
a
m
e
n
tal
li
m
it
a
ti
o
n
s,
a
n
d
fu
t
u
re
p
o
ten
ti
a
l
of
DL
in
c
ry
p
tan
a
ly
sis.
We
fo
c
u
s
e
x
ten
siv
e
ly
on
tw
o
p
rima
ry
d
o
m
a
in
s
DL
-
b
a
se
d
sid
e
-
c
h
a
n
n
e
l
a
n
a
ly
sis
(DL
-
S
CA)
a
n
d
DL
-
e
n
h
a
n
c
e
d
c
ry
p
ta
n
a
ly
sis
of
s
y
m
m
e
tri
c
p
rimiti
v
e
s
.
Th
e
re
v
iew
m
e
ti
c
u
lo
u
sly
e
x
a
m
in
e
s
a
d
v
a
n
c
e
m
e
n
ts
in
a
tt
a
c
k
e
fficie
n
c
y
(re
d
u
c
in
g
t
h
e
n
u
m
b
e
r
of
trac
e
s/q
u
e
ries
),
r
o
b
u
stn
e
ss
a
g
a
i
n
st
s
o
p
h
isti
c
a
ted
c
o
u
n
term
e
a
su
re
s,
a
u
to
m
a
ted
fe
a
t
u
re
e
x
trac
ti
o
n
,
a
n
d
t
h
e
n
a
sc
e
n
t
e
x
p
l
o
ra
ti
o
n
of
t
h
e
o
re
ti
c
a
l
f
o
u
n
d
a
ti
o
n
s.
Wh
il
e
DL
d
e
m
o
n
stra
tes
re
m
a
rk
a
b
le
c
a
p
a
b
il
it
ies
in
a
u
t
o
m
a
ti
n
g
c
o
m
p
lex
p
a
tt
e
rn
re
c
o
g
n
it
i
o
n
c
rit
ica
l
to
c
ry
p
tan
a
ly
sis,
sig
n
ifi
c
a
n
t
c
h
a
ll
e
n
g
e
s
p
e
rsist,
in
c
lu
d
i
n
g
th
e
“
b
lac
k
-
b
o
x
”
n
a
tu
re
of
m
o
d
e
ls,
d
a
ta
d
e
p
e
n
d
e
n
c
y
,
sc
a
lab
il
it
y
to
fu
ll
c
ry
p
t
o
g
ra
p
h
ic
p
rimiti
v
e
s,
a
n
d
t
h
e
c
rit
ica
l
n
e
e
d
fo
r
e
x
p
lai
n
a
b
il
i
ty
a
n
d
th
e
o
re
ti
c
a
l
g
ro
u
n
d
i
n
g
.
T
h
is
re
v
iew
se
rv
e
s
as
a
fo
u
n
d
a
ti
o
n
a
l
re
so
u
rc
e
fo
r
re
se
a
rc
h
e
rs
a
n
d
p
ra
c
ti
ti
o
n
e
rs
n
a
v
i
g
a
ti
n
g
t
h
is
ra
p
id
l
y
e
v
o
lv
i
n
g
in
ters
e
c
ti
o
n
of
a
rti
ficia
l
in
tell
ig
e
n
c
e
a
n
d
c
ry
p
to
g
ra
p
h
y
.
K
ey
w
o
r
d
s
:
Ad
v
er
s
ar
ial
m
ac
h
in
e
lear
n
in
g
C
o
n
v
o
lu
tio
n
al
n
eu
r
al
n
etwo
r
k
s
C
r
y
p
tan
aly
s
is
Deep
lear
n
in
g
E
x
p
lain
ab
le
AI
L
ig
h
tweig
h
t
cr
y
p
to
g
r
a
p
h
y
Ma
ch
in
e
lear
n
in
g
Sid
e
-
ch
an
n
el
an
al
y
s
is
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
:
Ou
s
s
am
a
No
u
i
Fu
n
d
am
en
tal
T
ea
ch
i
n
g
Dep
a
r
tm
en
t,
Facu
lty
o
f
Scien
ce
s
o
f
Ma
tter
Un
iv
er
s
ity
Of
B
atn
a
1
B
atn
a,
Alg
er
ia
E
m
ail
o
u
s
s
am
a.
n
o
u
i@
u
n
iv
-
b
a
tn
a.
d
z
.
1.
I
NT
RO
D
UCT
I
O
N
C
r
y
p
to
g
r
ap
h
y
f
o
r
m
s
th
e
f
o
u
n
d
atio
n
of
m
o
d
er
n
d
ig
ital
s
e
cu
r
ity
,
e
n
ab
lin
g
th
e
p
r
o
tectio
n
of
d
at
a
co
n
f
id
en
tiality
,
in
teg
r
ity
,
an
d
au
th
en
ticity
.
C
r
y
p
tan
aly
s
is
-
th
e
ad
v
er
s
ar
ial
co
u
n
ter
p
ar
t
—
aim
s
to
u
n
co
v
er
wea
k
n
ess
es
in
cr
y
p
t
o
g
r
ap
h
ic
alg
o
r
ith
m
s
or
th
eir
p
h
y
s
ical
i
m
p
lem
en
tatio
n
s
,
e
x
p
lo
itin
g
ei
th
er
d
esig
n
f
laws
or
v
u
ln
er
ab
ilit
ies
ar
is
in
g
f
r
o
m
s
id
e
-
ch
an
n
el
leak
ag
e
an
d
f
au
lt
in
jectio
n
[
1
]
–
[
4
]
.
T
r
ad
itio
n
al
cr
y
p
tan
aly
s
is
h
as
lo
n
g
r
elied
on
ad
v
an
ce
d
m
a
th
em
at
ical
m
eth
o
d
s
,
in
clu
d
i
n
g
d
if
f
er
e
n
tial,
lin
ea
r
,
an
d
in
teg
r
al
tech
n
iq
u
es.
Ho
wev
er
,
t
h
e
r
a
p
id
in
cr
ea
s
e
in
co
m
p
u
tatio
n
al
ca
p
ab
ilit
ies
an
d
th
e
em
er
g
en
ce
of
m
ac
h
in
e
lear
n
in
g
(
ML
)
,
p
ar
ticu
lar
ly
d
ee
p
lear
n
in
g
(
DL
)
,
h
a
v
e
i
n
tr
o
d
u
ce
d
a
p
o
wer
f
u
l
d
ata
-
d
r
iv
en
p
a
r
ad
ig
m
f
o
r
c
r
y
p
tan
aly
tic
r
esear
ch
[
5
]
–
[
7
]
.
DL
m
o
d
els,
esp
ec
ially
co
n
v
o
lu
tio
n
al
n
eu
r
al
ne
two
r
k
s
(
C
NNs),
ar
e
a
d
ep
t
at
d
etec
tin
g
co
m
p
lex
,
n
o
n
lin
ea
r
p
atter
n
s
in
h
ig
h
-
d
im
en
s
io
n
al,
n
o
is
y
d
atasets
-
a
ca
p
ab
ilit
y
d
ir
ec
tly
r
elev
an
t
to
n
u
m
er
o
u
s
cr
y
p
tan
aly
tic
ch
alle
n
g
es
[
8
]
–
[
1
0
]
.
In
s
id
e
-
ch
a
n
n
el
an
al
y
s
is
(
SC
A)
,
DL
o
f
f
er
s
s
ev
er
al
d
is
tin
ct
ad
v
a
n
ta
g
es
o
v
er
class
ical
ap
p
r
o
ac
h
es
,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
Dee
p
lea
r
n
in
g
in
cryp
t
a
n
a
lysi
s
a
co
mp
r
eh
en
s
ive
r
ev
iew
o
f te
ch
n
iq
u
es,
a
p
p
lica
tio
n
s
,
…
(
Ou
s
s
a
ma
N
o
u
i
)
847
−
Au
to
m
ated
f
ea
t
u
r
e
ex
t
r
ac
tio
n
DL
m
o
d
els
can
lear
n
d
is
cr
im
in
ativ
e
f
ea
tu
r
es
d
ir
ec
tly
f
r
o
m
r
aw
in
p
u
ts
(
e.
g
.
,
p
o
wer
tr
ac
es
,
elec
tr
o
m
a
g
n
etic
em
an
atio
n
s
,
cip
h
e
r
tex
t
p
air
s
)
,
elim
in
atin
g
th
e
n
ee
d
f
o
r
m
a
n
u
al
an
d
o
f
ten
er
r
o
r
-
p
r
o
n
e
f
ea
tu
r
e
en
g
in
ee
r
in
g
r
eq
u
ir
ed
by
tr
a
d
itio
n
al
SC
A
or
s
tati
s
tical
d
is
t
in
g
u
is
h
er
s
[
7
]
,
[
1
1
]
–
[
1
3
]
.
−
No
is
e
an
d
m
is
alig
n
m
e
n
t
to
ler
an
ce
C
NN
-
b
ased
ap
p
r
o
ac
h
es
d
em
o
n
s
tr
ate
r
o
b
u
s
tn
ess
ag
ain
s
t
n
o
is
e
an
d
tr
ac
e
d
esy
n
ch
r
o
n
izatio
n
,
ef
f
ec
tiv
ely
m
o
d
elin
g
co
m
p
lex
d
e
p
en
d
e
n
cies
th
at
ar
e
d
if
f
icu
lt
to
ca
p
tu
r
e
an
aly
tically
[
1
3
]
–
[
1
5
]
.
−
C
o
u
n
ter
m
ea
s
u
r
e
r
esil
ien
ce
t
ec
h
n
iq
u
es
s
u
ch
as
d
ata
au
g
m
en
tatio
n
en
ab
le
DL
-
b
ased
SC
A
to
m
itig
ate
th
e
ef
f
ec
ts
of
co
m
m
o
n
c
o
u
n
ter
m
e
asu
r
es,
in
clu
d
in
g
r
an
d
o
m
d
ela
y
s
,
s
h
u
f
f
lin
g
,
an
d
m
ask
in
g
[
1
5
]
–
[
1
7
]
.
−
I
m
p
r
o
v
ed
attac
k
e
f
f
icien
cy
in
m
an
y
ca
s
es,
DL
-
SC
A
ac
h
iev
es
s
u
cc
ess
f
u
l
k
ey
r
ec
o
v
er
y
wi
th
s
ig
n
if
ican
tly
f
ewe
r
tr
ac
es
co
m
p
a
r
ed
to
tem
p
late
attac
k
s
or
co
r
r
elatio
n
p
o
wer
an
aly
s
is
(
C
PA)
.
Similar
ly
,
in
alg
o
r
ith
m
ic
cr
y
p
tan
aly
s
is
,
DL
ca
n
u
n
co
v
er
m
o
r
e
e
f
f
icien
t
d
is
tin
g
u
is
h
er
s
an
d
d
if
f
er
e
n
tial
ch
ar
a
cter
is
tics
th
an
tr
ad
itio
n
al
h
eu
r
is
tic
s
ea
r
ch
m
eth
o
d
s
[
1
8
]
–
[
2
0
]
.
T
h
is
r
ev
iew
s
y
n
th
esizes
f
in
d
in
g
s
f
r
o
m
r
ec
en
t
s
tu
d
ies
to
p
r
o
v
i
d
e
an
in
teg
r
ated
p
er
s
p
ec
tiv
e
on
th
e
r
o
le
of
DL
in
m
o
d
er
n
cr
y
p
tan
aly
s
is
.
Sp
ec
if
ically
,
it
ex
am
in
es
,
−
f
u
n
d
am
e
n
tal
co
n
ce
p
ts
an
d
DL
ar
ch
itectu
r
es
em
p
lo
y
ed
in
c
r
y
p
tan
aly
s
is
(
s
ec
tio
n
s
2
–
3
)
,
−
th
e
tr
an
s
f
o
r
m
ativ
e
im
p
ac
t
of
DL
on
s
id
e
-
ch
a
n
n
el
attac
k
s
(
s
ec
tio
n
4
)
,
−
ap
p
licatio
n
s
to
th
e
cr
y
p
tan
a
ly
s
is
of
s
y
m
m
etr
ic
cr
y
p
to
g
r
ap
h
ic
p
r
im
itiv
es,
p
ar
ticu
lar
ly
b
lo
ck
cip
h
er
s
(
s
ec
tio
n
5
)
,
−
lim
itatio
n
s
,
ch
allen
g
es,
an
d
th
e
“
b
lack
-
b
o
x
”
in
ter
p
r
etab
ilit
y
p
r
o
b
lem
(
s
ec
tio
n
6
)
,
an
d
−
p
r
o
m
is
in
g
f
u
tu
r
e
r
esear
ch
d
ir
e
ctio
n
s
(
s
ec
tio
n
7
)
.
Un
lik
e
ex
is
tin
g
s
u
r
v
ey
s
th
at
p
r
im
ar
ily
f
o
c
u
s
on
eith
er
p
h
y
s
ical
s
id
e
-
ch
an
n
el
attac
k
s
or
is
o
lated
n
eu
r
al
d
is
tin
g
u
is
h
er
s
,
th
is
r
e
v
iew
p
r
o
v
id
es
a
u
n
if
ied
an
aly
s
i
s
of
DL
-
b
ased
SC
A
an
d
DL
-
a
s
s
is
ted
alg
o
r
ith
m
ic
cr
y
p
tan
aly
s
is
of
s
y
m
m
etr
ic
p
r
im
itiv
es.
In
ad
d
itio
n
to
s
y
s
tem
atizin
g
ar
ch
itectu
r
es,
a
ttack
m
o
d
els,
an
d
ev
alu
atio
n
m
etr
ics,
th
is
wo
r
k
em
p
h
asizes
ex
p
lain
ab
ilit
y
,
s
ca
lab
ilit
y
,
an
d
g
en
er
aliza
tio
n
,
wh
ich
ar
e
o
f
ten
u
n
d
er
e
x
p
lo
r
e
d
.
Fu
r
th
e
r
m
o
r
e
,
th
e
r
ev
iew
o
f
f
er
s
co
m
p
a
r
ativ
e
tax
o
n
o
m
ies
an
d
s
y
n
th
esized
t
ab
les
th
at
h
ig
h
lig
h
t
p
r
ac
tical
ca
p
ab
ilit
ies,
lim
itatio
n
s
,
an
d
o
p
en
c
h
allen
g
es.
T
h
is
p
ap
er
f
o
llo
ws
th
e
I
MRaDC
s
tr
u
ctu
r
e.
Sectio
n
1
in
tr
o
d
u
ce
s
th
e
b
ac
k
g
r
o
u
n
d
an
d
m
o
tiv
atio
n
.
Sectio
n
2
p
r
esen
ts
f
o
u
n
d
atio
n
al
co
n
ce
p
ts
an
d
th
e
r
e
v
iew
m
e
th
o
d
o
lo
g
y
.
Sectio
n
s
3
an
d
4
d
escr
ib
e
th
e
m
eth
o
d
s
an
d
a
r
ch
itectu
r
es
em
p
lo
y
e
d
in
DL
-
b
ased
cr
y
p
tan
aly
s
is
,
w
ith
a
f
o
cu
s
on
SC
A
.
Sectio
n
s
5
an
d
6
p
r
o
v
id
e
r
esu
lts
-
o
r
ien
ted
d
is
cu
s
s
io
n
a
n
d
cr
itical
an
al
y
s
is
of
DL
ap
p
licatio
n
s
to
s
y
m
m
et
r
ic
cr
y
p
t
an
aly
s
is
,
in
clu
d
in
g
lim
itatio
n
s
an
d
ch
allen
g
es.
Se
ctio
n
7
o
u
tlin
es
f
u
tu
r
e
r
esear
c
h
d
ir
ec
tio
n
s
,
a
n
d
s
ec
tio
n
8
co
n
clu
d
es
th
e
p
ap
e
r
.
2.
F
O
UNDA
T
I
O
NAL
CO
NC
E
P
T
S
2
.
1
.
Cry
pt
a
na
ly
s
is
prim
er
C
r
y
p
to
g
r
ap
h
ic
g
o
als
co
n
f
i
d
en
tiality
,
in
teg
r
ity
,
au
th
en
ticity
.
Attack
m
o
d
els
cip
h
er
tex
t
-
o
n
ly
,
k
n
o
w
n
-
p
lain
tex
t,
ch
o
s
en
-
p
lain
tex
t,
c
h
o
s
en
-
cip
h
er
tex
t
.
I
m
p
lem
e
n
tatio
n
attac
k
s
v
s
.
alg
o
r
ith
m
ic
attac
k
s
d
is
tin
g
u
is
h
in
g
attac
k
s
tar
g
etin
g
p
h
y
s
ical
leak
ag
e
(
SC
A,
f
au
lt
attac
k
s
)
f
r
o
m
th
o
s
e
tar
g
etin
g
t
h
e
m
ath
em
ati
ca
l
s
tr
u
ctu
r
e
of
th
e
cip
h
er
.
Secu
r
ity
m
etr
ics
s
u
cc
ess
r
ate
(
S
R
)
,
g
u
ess
in
g
en
tr
o
p
y
(
GE
)
,
n
u
m
b
e
r
of
tr
ac
es
to
d
is
clo
s
u
r
e
(
NT
D)
f
o
r
SC
A;
p
r
o
b
ab
ilit
y
of
d
i
f
f
er
en
ti
al/lin
ea
r
ch
ar
ac
ter
is
tics
,
d
ata/tim
e
co
m
p
lex
ity
f
o
r
alg
o
r
ith
m
ic
attac
k
s
[
21]
–
[
2
3
]
.
2
.
2
.
ML
f
un
da
m
ent
a
ls
Su
p
er
v
is
ed
lear
n
in
g
lea
r
n
in
g
a
m
ap
p
in
g
f
r
o
m
in
p
u
ts
(
tr
a
ce
s
,
cip
h
er
tex
ts
)
to
lab
els
(
k
ey
b
y
tes
,
d
is
tin
g
u
is
h
in
g
lab
els).
Do
m
i
n
ates
cr
y
p
tan
aly
s
is
ap
p
licatio
n
s
[
1
5
]
.
Pr
o
f
ilin
g
v
s
.
No
n
-
Pro
f
ilin
g
Pr
o
f
ilin
g
(
Su
p
er
v
is
ed
)
attac
k
s
r
e
q
u
ir
e
a
co
n
tr
o
lled
p
r
o
f
ilin
g
p
h
ase
u
s
i
n
g
a
clo
n
e
d
e
v
ice;
n
o
n
-
Pro
f
ilin
g
attac
k
s
attem
p
t
k
ey
r
ec
o
v
er
y
d
i
r
ec
tly
.
C
o
r
e
co
n
ce
p
ts
tr
ain
in
g
/v
alid
atio
n
/t
est
s
ets,
o
v
er
f
itti
n
g
/u
n
d
er
f
itti
n
g
,
lo
s
s
f
u
n
ctio
n
s
(
ca
teg
o
r
ical
cr
o
s
s
-
en
tr
o
p
y
co
m
m
o
n
f
o
r
class
if
icatio
n
),
o
p
ti
m
izatio
n
alg
o
r
ith
m
s
(
SGD,
Ad
am
)
[
2
4
]
.
Gen
er
a
l
DL
wo
r
k
f
lo
ws,
ar
c
h
itectu
r
es,
an
d
ap
p
licatio
n
p
ar
a
d
ig
m
s
ar
e
d
is
cu
s
s
ed
in
d
etail
in
[
2
5
]
.
2
.
3
.
DL
s
pecif
ics
Ar
tific
ial
n
eu
r
al
n
etwo
r
k
s
(
A
NNs)
co
m
p
o
s
ed
of
la
y
er
s
of
in
ter
co
n
n
ec
te
d
n
e
u
r
o
n
s
(
n
o
d
es).
T
h
e
d
ep
t
h
d
ef
in
es
“
d
ee
p
”
lear
n
in
g
.
Mu
lt
i
-
lay
er
p
er
ce
p
tr
o
n
s
(
ML
Ps
)
Fu
lly
co
n
n
ec
te
d
n
etwo
r
k
s
.
E
ar
l
y
u
s
e
in
SC
A
an
d
b
lo
ck
cip
h
e
r
d
is
tin
g
u
is
h
er
s
[
2
6
]
.
C
NNs
Sp
ec
ialized
f
o
r
g
r
id
-
lik
e
d
ata
(
im
ag
es,
tim
e
-
s
er
ies
lik
e
SC
A
tr
ac
es).
Use
co
n
v
o
l
u
tio
n
al
la
y
er
s
to
e
x
tr
ac
t
lo
ca
l
f
ea
tu
r
es,
p
o
o
lin
g
l
ay
er
s
f
o
r
d
o
wn
s
am
p
lin
g
.
Stat
e
-
of
-
th
e
-
ar
t
f
o
r
DL
-
SC
A
[
8
]
.
Activ
atio
n
f
u
n
ctio
n
s
r
ec
tifie
d
lin
ea
r
u
n
it
(
R
eL
U
)
m
o
s
t
co
m
m
o
n
,
en
a
b
lin
g
non
-
lin
ea
r
ity
.
B
ac
k
p
r
o
p
ag
atio
n
Alg
o
r
ith
m
f
o
r
ca
lcu
latin
g
g
r
a
d
ien
ts
an
d
u
p
d
ati
n
g
weig
h
ts
d
u
r
i
n
g
tr
ain
in
g
.
D
ata
Au
g
m
en
tatio
n
Ar
tific
ially
ex
p
an
d
in
g
th
e
tr
ain
in
g
d
ataset
by
ap
p
ly
i
n
g
tr
an
s
f
o
r
m
atio
n
s
(
e.
g
.
,
r
an
d
o
m
s
h
if
ts
,
jitt
er
,
n
o
is
e
ad
d
itio
n
)
to
im
p
r
o
v
e
r
o
b
u
s
tn
ess
an
d
g
en
e
r
aliza
tio
n
[
2
7
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
4
6
-
8
5
5
848
2
.
4
.
Rev
iew
m
et
ho
do
lo
g
y
w
o
rk
f
lo
w
T
h
is
r
ev
iew
f
o
llo
ws
a
s
tr
u
ctu
r
ed
s
co
p
i
n
g
r
ev
iew
m
eth
o
d
o
l
o
g
y
d
esig
n
ed
to
s
y
s
tem
atica
lly
id
en
tif
y
,
s
cr
ee
n
,
an
d
s
y
n
th
esize
r
ec
en
t
ad
v
an
ce
s
in
DL
-
b
ased
cr
y
p
ta
n
aly
s
is
.
T
h
e
s
elec
tio
n
p
r
o
ce
s
s
was
o
r
g
an
ized
in
to
f
o
u
r
s
tag
es
-
id
en
tific
ati
o
n
,
s
cr
e
en
in
g
,
elig
ib
ilit
y
,
a
n
d
s
y
n
t
h
esis
-
to
en
s
u
r
e
tr
an
s
p
ar
e
n
cy
a
n
d
r
ep
r
o
d
u
cib
ilit
y
.
−
I
d
en
tific
atio
n
:
c
a
n
d
id
ate
s
tu
d
ies
wer
e
r
etr
iev
ed
v
ia
k
ey
wo
r
d
-
b
ased
s
ea
r
ch
es
ac
r
o
s
s
m
ajo
r
d
ig
ital
lib
r
ar
ies
an
d
cr
y
p
to
g
r
ap
h
y
r
ep
o
s
ito
r
ies.
−
Scr
ee
n
in
g
:
d
u
p
licate
r
ec
o
r
d
s
wer
e
r
em
o
v
ed
an
d
wo
r
k
s
not
d
ir
ec
tly
r
elate
d
to
cr
y
p
t
an
aly
s
is
or
S
C
A
(
e.
g
.
,
g
en
er
al
ML
p
ap
er
s
with
o
u
t
cr
y
p
to
g
r
a
p
h
ic
f
o
cu
s
)
wer
e
ex
clu
d
ed
.
−
E
lig
ib
ilit
y
:
i
n
clu
d
ed
s
tu
d
ies
ap
p
lied
DL
(
o
r
a
d
v
an
ce
d
ML)
to
cr
y
p
tan
aly
s
is
or
SC
A
an
d
p
r
o
v
id
ed
em
p
ir
ical
ev
alu
atio
n
(
e.
g
.
,
GE
,
SR
,
NT
D,
or
d
is
tin
g
u
is
h
in
g
a
cc
u
r
ac
y
/ad
v
a
n
tag
e)
.
−
Sy
n
th
esis
:
s
elec
ted
wo
r
k
s
wer
e
ca
teg
o
r
ized
by
attac
k
m
o
d
el,
tar
g
et
p
r
im
itiv
e
or
d
ev
ice,
n
eu
r
al
ar
ch
itectu
r
e,
d
ataset/tra
ce
ch
a
r
ac
ter
is
tics
,
an
d
ev
alu
atio
n
m
etr
ics,
en
ab
lin
g
a
co
m
p
a
r
ativ
e
s
y
n
th
esis
of
ca
p
ab
ilit
ies,
lim
itatio
n
s
,
an
d
o
p
en
r
esear
ch
c
h
allen
g
es.
To
clar
if
y
th
e
s
co
p
e
a
n
d
o
r
g
a
n
izatio
n
of
th
is
s
u
r
v
ey
,
Fig
u
r
e
1
s
u
m
m
ar
izes
th
e
m
ain
ca
t
eg
o
r
ies
of
cr
y
p
to
g
r
ap
h
ic
v
u
ln
e
r
ab
ilit
ies
an
d
th
e
co
r
r
esp
o
n
d
in
g
a
ttack
f
am
ilies
co
n
s
id
er
ed
in
th
e
liter
atu
r
e.
Vu
ln
er
ab
ilit
ies
can
be
g
r
o
u
p
ed
in
to
,
i)
im
p
lem
en
tatio
n
-
le
v
el
wea
k
n
ess
es,
wh
er
e
p
h
y
s
ical
leak
ag
e
or
f
au
lt
s
en
s
itiv
ity
en
ab
les
s
id
e
-
ch
an
n
el
an
d
f
au
lt
-
in
jectio
n
attac
k
s
;
i
i)
alg
o
r
ith
m
ic
-
lev
el
wea
k
n
ess
es,
wh
er
e
s
tatis
tica
l
b
iases
en
ab
le
d
is
tin
g
u
is
h
er
s
an
d
k
ey
-
r
ec
o
v
er
y
m
eth
o
d
s
s
u
ch
as
d
if
f
er
e
n
tial
an
d
lin
ea
r
cr
y
p
ta
n
aly
s
is
;
an
d
iii)
p
r
o
to
c
o
l/s
y
s
tem
-
lev
el
wea
k
n
ess
es
r
elate
d
to
co
m
p
o
s
itio
n
,
m
is
u
s
e,
or
s
y
s
tem
s
id
e
ef
f
e
cts.
ML
an
d
DL
ac
t
as
cr
o
s
s
-
cu
ttin
g
e
n
ab
ler
s
ac
r
o
s
s
th
ese
ca
teg
o
r
ies:
in
DL
-
SC
A,
m
o
d
els
p
r
im
ar
ily
s
u
p
p
o
r
t
d
e
n
o
is
in
g
,
alig
n
m
en
t,
an
d
au
to
m
ate
d
f
ea
tu
r
e
ex
tr
ac
tio
n
to
en
ab
le
r
o
b
u
s
t
p
r
o
f
ilin
g
u
n
d
er
c
o
u
n
ter
m
ea
s
u
r
es;
in
alg
o
r
ith
m
ic
cr
y
p
tan
aly
s
is
,
n
eu
r
al
d
is
tin
g
u
i
s
h
er
s
an
d
lear
n
e
d
b
iases
can
be
co
m
b
in
e
d
with
class
ical
tech
n
iq
u
es
to
g
u
id
e
or
f
ilter
s
ea
r
ch
;
an
d
at
th
e
s
y
s
te
m
lev
el,
lear
n
in
g
-
b
ased
p
atte
r
n
an
aly
s
is
is
o
cc
asio
n
ally
u
s
ed
f
o
r
an
o
m
aly
or
tr
af
f
ic/tim
in
g
ch
ar
ac
ter
izatio
n
(
th
o
u
g
h
it
is
less
ce
n
tr
al
to
th
e
p
r
esen
t
s
u
r
v
ey
)
.
T
h
is
tax
o
n
o
m
y
s
u
p
p
o
r
ts
a
s
tr
u
ctu
r
ed
s
y
n
th
esis
an
d
m
o
tiv
ates
th
e
co
m
p
ar
ativ
e
an
aly
s
is
d
ev
elo
p
e
d
in
th
e
s
u
b
s
eq
u
e
n
t
s
ec
tio
n
s
.
In
p
r
ac
tice,
s
y
s
tem
-
lev
el
ex
p
o
s
u
r
e
can
al
s
o
s
tem
f
r
o
m
m
is
co
n
f
ig
u
r
atio
n
of
h
a
r
d
war
e
s
ec
u
r
ity
m
o
d
u
les
an
d
p
er
m
is
s
io
n
co
n
tr
o
ls
[
2
8
]
.
Fig
u
r
e
1
.
T
a
x
o
n
o
m
y
of
cr
y
p
to
g
r
ap
h
ic
v
u
ln
er
a
b
ilit
ies
an
d
co
r
r
esp
o
n
d
in
g
attac
k
f
am
ilies
,
h
i
g
h
lig
h
tin
g
wh
er
e
ML
an
d
DL
ar
e
i
n
teg
r
ated
ac
r
o
s
s
im
p
lem
en
tatio
n
-
,
alg
o
r
ith
m
ic
-
,
an
d
p
r
o
t
o
co
l/s
y
s
tem
-
lev
el
attac
k
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
Dee
p
lea
r
n
in
g
in
cryp
t
a
n
a
lysi
s
a
co
mp
r
eh
en
s
ive
r
ev
iew
o
f te
ch
n
iq
u
es,
a
p
p
lica
tio
n
s
,
…
(
Ou
s
s
a
ma
N
o
u
i
)
849
3.
DL
ARCH
I
T
E
CT
URE
S
F
O
R
CRYP
T
AN
AL
Y
SI
S
3
.
1
.
Arc
hite
ct
ures
f
o
r
s
ide
-
c
ha
nn
el
a
na
ly
s
is
C
NN
ar
ch
itectu
r
es
p
r
ed
o
m
in
an
t
ar
ch
itectu
r
e.
VGG
-
in
s
p
ir
ed
n
etwo
r
k
s
(
s
tack
ed
s
m
all
co
n
v
o
lu
tio
n
al
lay
er
s
)
ar
e
h
ig
h
ly
e
f
f
ec
tiv
e.
R
esid
u
al
co
n
n
ec
tio
n
s
(
R
esNet
)
h
elp
tr
ain
d
ee
p
er
m
o
d
els.
C
u
s
to
m
ar
ch
itectu
r
es
tailo
r
ed
to
tr
ac
e
ch
ar
ac
ter
is
tics
ar
e
co
m
m
o
n
[
2
9
]
.
I
n
p
u
t
r
e
p
r
esen
tatio
n
one
-
d
i
m
en
s
io
n
C
NNs
f
o
r
s
in
g
le
-
tr
ac
e
attac
k
s
(
tr
ea
tin
g
tr
ac
e
as
1D
s
ig
n
al)
.
2D
C
NNs
ar
e
s
o
m
etim
es
u
s
ed
if
tr
ac
es
ar
e
p
r
e
-
p
r
o
ce
s
s
ed
in
to
s
p
ec
tr
o
g
r
am
s
or
c
o
m
b
in
e
d
with
o
th
er
d
ata
d
im
en
s
io
n
s
.
Ou
tp
u
t
lay
er
ty
p
ically
,
a
s
o
f
tm
ax
lay
e
r
f
o
r
class
if
icatio
n
(
p
r
ed
ictin
g
a
s
p
ec
if
ic
k
ey
b
y
te
v
al
u
e)
.
R
eg
r
es
s
io
n
(
p
r
e
d
ictin
g
a
lea
k
ag
e
v
alu
e)
is
less
co
m
m
o
n
but
ex
p
l
o
r
ed
.
H
y
p
er
p
ar
am
eter
tu
n
in
g
c
r
itical
f
o
r
p
er
f
o
r
m
an
ce
.
I
n
clu
d
es
n
u
m
b
e
r
/f
ilter
s
ize
of
co
n
v
o
lu
tio
n
al
lay
er
s
,
p
o
o
lin
g
s
tr
ateg
ies,
d
e
n
s
e
lay
er
s
ize,
lear
n
in
g
r
ate,
b
atch
s
ize.
Au
to
m
ated
m
eth
o
d
s
(
g
r
id
s
ea
r
c
h
,
B
ay
esian
o
p
tim
izatio
n
)
ar
e
in
c
r
ea
s
in
g
ly
u
s
ed
[
3
0
]
.
3
.
2
.
Arc
hite
ct
ures
f
o
r
a
lg
o
ri
t
hm
ic
cr
y
pta
na
ly
s
is
Dis
tin
g
u
is
h
er
n
etwo
r
k
s
o
f
ten
ML
Ps
or
C
NN
s
.
I
n
p
u
t
is
p
air
s
of
p
lain
tex
ts
/cip
h
er
tex
ts
or
d
i
f
f
er
en
tials
.
Ou
tp
u
t
is
a
p
r
o
b
ab
ilit
y
s
co
r
e
in
d
icatin
g
wh
et
h
er
t
h
e
in
p
u
t
b
elo
n
g
s
to
th
e
cip
h
e
r
or
a
r
a
n
d
o
m
p
er
m
u
tatio
n
.
Go
h
r
’
s
Sp
ec
k
attac
k
[
1
1
]
u
s
ed
a
s
o
p
h
is
ticated
C
NN.
Key
r
ec
o
v
er
y
n
etwo
r
k
s
less
c
o
m
m
o
n
an
d
h
ar
d
e
r
.
C
an
in
v
o
l
v
e
b
u
ild
in
g
n
e
u
r
al
d
is
tin
g
u
is
h
er
s
f
o
r
k
ey
-
d
ep
en
d
e
n
t
d
if
f
er
en
tials
or
d
ir
ec
tly
p
r
ed
ictin
g
k
e
y
b
its
(
e
x
t
r
em
ely
ch
allen
g
in
g
f
o
r
f
u
ll
c
ip
h
er
s
)
[
5
]
.
Of
te
n
co
m
b
in
ed
with
tr
a
d
itio
n
al
k
e
y
s
ea
r
ch
or
r
an
k
in
g
.
Hy
b
r
i
d
ap
p
r
o
ac
h
es
co
m
b
in
in
g
n
eu
r
al
d
is
tin
g
u
is
h
er
s
with
class
ical
cr
y
p
tan
aly
tic
tech
n
iq
u
es
(
e.
g
.
,
u
s
in
g
a
n
eu
r
al
d
is
tin
g
u
is
h
er
to
f
ilter
p
r
o
m
is
in
g
d
if
f
er
en
tial
p
ath
s
f
o
r
tr
ad
itio
n
al
k
ey
r
ec
o
v
e
r
y
)
.
3
.
3
.
T
ra
ini
ng
m
et
ho
do
l
o
g
ies
a
nd
cha
lleng
es
Data
r
eq
u
ir
em
e
n
ts
m
ass
iv
e
d
atasets
n
ee
d
ed
,
esp
ec
ially
f
o
r
alg
o
r
ith
m
ic
cr
y
p
tan
aly
s
is
(
m
illi
o
n
s
of
p
lain
tex
t
-
cip
h
er
te
x
t
p
air
s
)
.
Acq
u
is
itio
n
co
s
t
f
o
r
SC
A
can
be
h
ig
h
but
m
an
ag
ea
b
le.
I
m
b
ala
n
ce
d
d
ata
in
SC
A,
class
es
(
k
ey
b
y
te
v
alu
es)
a
r
e
n
atu
r
ally
b
ala
n
ce
d
.
In
d
is
tin
g
u
is
h
er
tr
ain
in
g
,
b
alan
cin
g
“
ci
p
h
er
”
v
s
.
“
r
an
d
o
m
”
in
s
tan
ce
s
is
cr
u
cial.
Pre
v
e
n
tin
g
o
v
er
f
itti
n
g
tech
n
i
q
u
es
in
c
lu
d
e
d
r
o
p
o
u
t
lay
er
s
,
L
1
/L
2
r
eg
u
lar
izatio
n
,
ea
r
l
y
s
to
p
p
in
g
,
an
d
r
ig
o
r
o
u
s
v
alid
a
tio
n
on
h
eld
-
o
u
t
d
atasets
.
T
r
a
n
s
f
er
lear
n
i
n
g
p
o
ten
tial
to
le
v
er
ag
e
m
o
d
els
p
r
e
-
tr
ain
ed
on
s
im
ilar
task
s
or
d
ev
ices
to
r
ed
u
ce
d
ata
n
ee
d
s
,
th
o
u
g
h
ex
p
l
o
r
atio
n
is
lim
ited
.
4.
DL
F
O
R
SI
D
E
-
CH
ANN
E
L
ANALY
SI
S
(
D
L
-
SCA)
4
.
1
.
T
he
SCA
la
nd
s
ca
pe
a
nd
DL
’
s
ent
ry
po
int
T
r
ad
itio
n
al
SC
A
tem
p
late
atta
ck
s
(
o
p
tim
al
p
r
o
f
iled
)
,
C
PA
(
p
o
p
u
lar
n
o
n
-
p
r
o
f
iled
)
,
DPA.
L
im
itatio
n
s
Sen
s
itiv
ity
to
tr
ac
e
m
is
alig
n
m
en
t,
n
o
is
e,
co
u
n
ter
m
ea
s
u
r
es
;
n
ee
d
f
o
r
m
an
u
al
f
ea
tu
r
e
s
elec
tio
n
(
Po
in
ts
-
of
-
in
ter
est
(
POI
s
)
)
;
p
er
f
o
r
m
an
ce
d
eg
r
ad
atio
n
with
co
m
p
lex
le
ak
ag
e
f
u
n
ctio
n
s
.
DL
r
ev
o
lu
ti
o
n
Ma
g
h
r
e
b
i
et
al
.
[
3
1
]
a
n
d
C
ag
li
et
al
.
[
8
]
we
r
e
am
o
n
g
th
e
f
ir
s
t
to
d
em
o
n
s
tr
ate
C
NN
s
o
u
tp
er
f
o
r
m
in
g
class
ical
p
r
o
f
iled
SC
A,
h
an
d
lin
g
r
aw
tr
ac
es
d
ir
ec
tly
.
4
.
2
.
P
r
o
f
iled
DL
-
SCA
t
he
ne
w
g
o
ld
s
t
a
nd
a
rd
W
o
r
k
f
lo
w
i
)
Acq
u
ir
e
p
r
o
f
ilin
g
tr
ac
es
(
k
n
o
wn
in
p
u
ts
,
k
n
o
w
n
k
ey
s
)
.
ii
)
T
r
ain
DL
m
o
d
el
(
e.
g
.
,
C
NN)
to
p
r
ed
ict
in
ter
m
ed
iate
v
alu
es
(
e.
g
.
,
S
-
b
o
x
o
u
tp
u
t)
or
d
ir
ec
tl
y
k
ey
b
y
tes
f
r
o
m
tr
ac
es.
iii
)
Acq
u
ir
e
attac
k
tr
ac
es
(
k
n
o
wn
in
p
u
ts
,
u
n
k
n
o
wn
k
e
y
)
.
iv
)
Use
t
r
ain
ed
m
o
d
el
to
p
r
ed
ict
k
ey
ca
n
d
id
ates.
v
)
R
an
k
/
r
ec
o
v
e
r
k
ey
.
Su
p
er
io
r
p
er
f
o
r
m
an
ce
co
n
s
is
ten
tly
s
h
o
wn
to
ac
h
iev
e
lo
we
r
NT
D/GE
th
an
T
e
m
p
late
At
tack
s
,
esp
ec
ially
in
n
o
is
y
en
v
ir
o
n
m
en
ts
or
with
m
is
alig
n
m
en
t.
Kim
et
al
.
[
9
]
ex
p
licitly
d
em
o
n
s
tr
ated
C
NNs
ef
f
ec
tiv
ely
u
tili
ze
n
o
is
e.
R
aw
tr
ac
e
p
r
o
ce
s
s
in
g
a
m
ajo
r
ad
v
a
n
tag
e
is
b
y
p
ass
in
g
th
e
n
ee
d
f
o
r
e
x
p
licit
POI
s
elec
tio
n
or
tr
ac
e
alig
n
m
en
t.
C
NNs
lear
n
r
o
b
u
s
t
f
ea
tu
r
es
in
v
ar
ian
t
to
s
m
all
s
h
i
f
ts
.
Mu
lti
-
task
lear
n
in
g
tr
ain
in
g
a
s
in
g
le
m
o
d
el
to
p
r
ed
ict
m
u
ltip
le
in
ter
m
ed
iate
v
alu
es
or
k
e
y
b
y
tes
s
im
u
ltan
e
o
u
s
ly
,
im
p
r
o
v
in
g
ef
f
icien
c
y
.
4
.
3
.
Co
m
ba
t
ing
c
o
un
t
er
m
ea
s
ures
wit
h
DL
J
itter
an
d
r
an
d
o
m
d
ela
y
s
C
ag
li
et
al
.
[
8
]
p
io
n
ee
r
ed
u
s
in
g
d
ata
au
g
m
e
n
tatio
n
d
u
r
in
g
tr
ain
in
g
.
By
ar
tific
ially
ad
d
in
g
r
an
d
o
m
te
m
p
o
r
al
s
h
if
ts
(
jitt
er
)
to
p
r
o
f
ili
n
g
tr
ac
es,
th
e
C
NN
lear
n
s
to
be
in
v
a
r
ian
t
to
s
u
ch
co
u
n
ter
m
ea
s
u
r
es,
o
u
tp
e
r
f
o
r
m
i
n
g
class
ical
attac
k
s
r
eq
u
ir
in
g
ex
p
licit
r
ea
lig
n
m
e
n
t.
Desy
n
ch
r
o
n
izatio
n
Similar
to
jitt
er
,
h
an
d
led
ef
f
ec
tiv
ely
by
d
ata
au
g
m
e
n
tatio
n
(
s
h
if
tin
g
)
an
d
C
NN
’
s
in
h
er
en
t
s
h
if
t
-
in
v
ar
ian
ce
.
Ma
s
k
in
g
Mo
r
e
ch
allen
g
in
g
,
DL
-
SC
A
can
p
o
ten
tially
lear
n
h
ig
h
er
-
o
r
d
er
m
o
m
en
ts
or
co
m
p
le
x
in
ter
ac
tio
n
s
b
etwe
en
s
h
ar
es.
R
eq
u
ir
es
more
tr
ac
es
an
d
p
o
te
n
tially
more
c
o
m
p
lex
m
o
d
els
or
s
p
ec
if
ic
p
r
e
-
p
r
o
ce
s
s
in
g
.
State
-
of
-
th
e
-
ar
t
but
s
till
an
ac
tiv
e
ch
allen
g
e
co
m
p
a
r
ed
to
u
n
m
ask
e
d
s
ce
n
ar
io
s
.
Hid
in
g
(
No
is
e
a
d
d
itio
n
)
C
NNs
d
em
o
n
s
tr
ate
in
h
er
en
t
r
o
b
u
s
tn
ess
to
Gau
s
s
ian
n
o
is
e.
Per
f
o
r
m
an
ce
d
eg
r
a
d
es
g
r
ac
ef
u
lly
co
m
p
a
r
ed
to
class
ical
attac
k
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
4
6
-
8
5
5
850
4
.
4
.
No
n
-
pro
f
iled
a
nd
s
em
i
-
pro
f
iled
DL
-
SCA
No
n
-
p
r
o
f
iled
ch
allen
g
es
s
ig
n
if
ican
tly
h
ar
d
er
th
a
n
p
r
o
f
iled
.
R
eq
u
ir
es
u
n
s
u
p
er
v
is
ed
or
s
elf
-
s
u
p
er
v
is
e
d
lear
n
in
g
,
d
im
en
s
io
n
ality
r
ed
u
ctio
n
(
PC
A,
au
to
en
c
o
d
er
s
)
,
or
clu
s
ter
in
g
g
u
id
ed
by
DL
f
ea
tu
r
es.
Sem
i
-
p
r
o
f
iled
ap
p
r
o
ac
h
es
u
tili
zin
g
s
o
m
e
li
m
ited
k
n
o
wled
g
e
or
m
o
d
el
f
r
o
m
a
s
im
ilar
d
ev
ice.
An
ar
ea
of
o
n
g
o
in
g
r
esear
ch
with
p
o
ten
tial
f
o
r
r
ed
u
ci
n
g
p
r
o
f
ilin
g
co
s
ts
[
1
3
]
.
4
.
5
.
Sy
s
t
e
m
a
t
iza
t
io
n
a
nd
co
m
pa
ra
t
iv
e
a
na
ly
s
is
Su
r
v
ey
s
by
Hettwe
r
et
al
.
[
3
2
]
,
Ma
s
u
r
e
et
al
.
[
3
3
]
,
a
n
d
th
e
c
o
m
p
r
eh
e
n
s
iv
e
So
K
by
Picek
et
al
.
[
1
3
]
p
r
o
v
id
e
in
v
alu
ab
le
ta
x
o
n
o
m
i
es,
class
if
y
in
g
DL
-
SC
A
wo
r
k
s
by
attac
k
t
y
p
e
(
p
r
o
f
iled
/n
o
n
-
p
r
o
f
iled
)
,
tar
g
et
(
alg
o
r
ith
m
,
im
p
lem
e
n
tatio
n
)
,
DL
ar
ch
itectu
r
e,
co
u
n
ter
m
ea
s
u
r
es
ad
d
r
ess
ed
,
ev
alu
atio
n
m
etr
ics.
Hig
h
lig
h
tin
g
th
e
ev
o
lu
tio
n
f
r
o
m
ML
Ps
to
C
NN
s
as
th
e
d
o
m
in
an
t
ar
ch
itectu
r
e.
E
m
p
h
asizin
g
th
e
cr
i
tical
im
p
o
r
tan
ce
of
r
ig
o
r
o
u
s
ev
alu
atio
n
m
et
h
o
d
o
l
o
g
ies,
r
ep
o
r
tin
g
NT
D/GE
/SR
cu
r
v
es,
an
d
u
s
in
g
m
u
ltip
le
d
atasets
.
T
ab
le
1
p
r
esen
ts
a
s
tr
u
ctu
r
ed
tax
o
n
o
m
y
of
DL
-
b
ased
SC
A
ap
p
r
o
ac
h
es,
s
u
m
m
ar
izin
g
attac
k
m
o
d
el
s
,
tar
g
et
alg
o
r
ith
m
s
,
n
eu
r
al
ar
c
h
itectu
r
es,
co
u
n
ter
m
ea
s
u
r
es
ad
d
r
ess
ed
,
an
d
e
v
alu
at
io
n
m
etr
ics
r
ep
o
r
ted
in
t
h
e
liter
atu
r
e.
T
ab
le
1
.
Deta
iled
tax
o
n
o
m
y
of
DL
-
SC
A
ap
p
r
o
ac
h
es
(
a
d
a
p
ted
f
r
o
m
[
1
3
]
,
[
3
2
]
,
[
3
3
]
)
C
h
a
r
a
c
t
e
r
i
s
t
i
c
C
a
t
e
g
o
r
i
e
s
Ex
a
m
p
l
e
s
/
N
o
t
e
s
R
e
f
e
r
e
n
c
e
s
A
t
t
a
c
k
t
y
p
e
P
r
o
f
i
l
e
d
,
n
o
n
-
p
r
o
f
i
l
e
d
,
s
e
mi
-
p
r
o
f
i
l
e
d
P
r
o
f
i
l
e
d
d
o
mi
n
a
t
e
s
r
e
se
a
r
c
h
a
n
d
p
e
r
f
o
r
m
a
n
c
e
.
N
o
n
-
p
r
o
f
i
l
e
d
r
e
m
a
i
n
s
c
h
a
l
l
e
n
g
i
n
g
.
[
8
]
,
[
9
]
,
[
1
3
]
,
[
3
1
]
-
[
3
3
]
Ta
r
g
e
t
a
l
g
o
r
i
t
h
m
A
ES,
D
ES,
R
S
A
,
EC
C
,
Li
g
h
t
w
e
i
g
h
t
(
P
R
ESEN
T,
S
I
M
O
N
,
S
P
EC
K
)
,
A
S
C
O
N
,
e
t
c
.
A
ES
is
t
h
e
m
o
st
c
o
mm
o
n
b
e
n
c
h
mar
k
.
Li
g
h
t
w
e
i
g
h
t
c
i
p
h
e
r
s
a
r
e
f
r
e
q
u
e
n
t
t
a
r
g
e
t
s
d
u
e
to
c
o
n
s
t
r
a
i
n
e
d
i
m
p
l
e
me
n
t
a
t
i
o
n
s.
[
8
]
,
[
9
]
,
[
1
1
]
-
[
1
3
]
,
[
3
2
]
Le
a
k
a
g
e
s
o
u
r
c
e
P
o
w
e
r
c
o
n
su
mp
t
i
o
n
(DPA),
e
l
e
c
t
r
o
ma
g
n
e
t
i
c
(
EM
A
)
,
t
i
mi
n
g
,
p
h
o
t
o
n
i
c
,
a
c
o
u
st
i
c
P
o
w
e
r
a
n
d
EM
a
r
e
m
o
s
t
p
r
e
v
a
l
e
n
t
.
DL
sh
o
w
s
p
r
o
mi
s
e
f
o
r
o
t
h
e
r
s.
[
9
]
,
[
1
3
]
,
[
3
2
]
,
[
3
3
]
DL
a
r
c
h
i
t
e
c
t
u
r
e
M
LP,
C
N
N
(
1
D
,
2
D
)
,
R
e
sN
e
t
,
VGG,
A
u
t
o
e
n
c
o
d
e
r
s,
R
N
N
/
LS
TM
(
r
a
r
e
)
,
h
y
b
r
i
d
m
o
d
e
l
s
C
N
N
(
1
D
)
is
st
a
t
e
-
of
-
t
h
e
-
a
r
t
f
o
r
p
r
o
f
i
l
e
d
S
C
A
on
r
a
w
t
r
a
c
e
s
.
M
LP
u
se
d
e
a
r
l
i
e
r
/
h
y
b
r
i
d
.
A
u
t
o
e
n
c
o
d
e
r
s
e
x
p
l
o
r
e
d
f
o
r
non
-
p
r
o
f
i
l
e
d
/
f
e
a
t
u
r
e
r
e
d
u
c
t
i
o
n
.
[
8
]
,
[
9
]
,
[
1
1
]
-
[
1
3
]
,
[
3
1
]
,
[
3
3
]
C
o
u
n
t
e
r
m
e
a
s
u
r
e
s
a
d
d
r
e
sse
d
N
o
n
e
,
j
i
t
t
e
r
/
r
a
n
d
o
m
d
e
l
a
y
s
,
mas
k
i
n
g
(
1
st
o
r
d
e
r
,
h
i
g
h
e
r
-
o
r
d
e
r
)
,
sh
u
f
f
l
i
n
g
,
h
i
d
i
n
g
D
a
t
a
A
u
g
me
n
t
a
t
i
o
n
h
i
g
h
l
y
e
f
f
e
c
t
i
v
e
a
g
a
i
n
s
t
j
i
t
t
e
r
/
d
e
l
a
y
s
/
sh
u
f
f
l
i
n
g
.
M
a
s
k
i
n
g
r
e
ma
i
n
s
a
si
g
n
i
f
i
c
a
n
t
c
h
a
l
l
e
n
g
e
r
e
q
u
i
r
i
n
g
m
o
r
e
t
r
a
c
e
s/
c
o
m
p
l
e
x
m
o
d
e
l
s
.
[
8
]
,
[
9
]
,
[
1
2
]
,
[
2
4
]
,
[
2
6
]
K
e
y
c
o
n
t
r
i
b
u
t
i
o
n
f
o
c
u
s
Raw
t
r
a
c
e
p
r
o
c
e
ssi
n
g
,
r
o
b
u
s
t
n
e
ss
to
n
o
i
se
,
c
o
u
n
t
e
r
m
e
a
s
u
r
e
r
e
s
i
l
i
e
n
c
e
,
f
e
w
-
sh
o
t
l
e
a
r
n
i
n
g
,
m
o
d
e
l
i
n
t
e
r
p
r
e
t
a
b
i
l
i
t
y
[
9
]
(
N
o
i
s
e
)
,
[
8
]
(
Ji
t
t
e
r
)
,
[
1
1
]
(
A
l
g
o
r
i
t
h
mi
c
l
i
n
k
)
,
[
1
3
]
,
[
3
3
]
(
S
y
st
e
ma
t
i
z
a
t
i
o
n
)
,
n
a
sce
n
t
w
o
r
k
on
e
x
p
l
a
i
n
a
b
l
e
A
I
(
X
A
I
)
.
[
8
]
-
[
1
0
]
,
[
1
2
]
,
[
2
6
]
Ev
a
l
u
a
t
i
o
n
m
e
t
r
i
c
s
S
R
,
G
E
,
N
TD
R
e
p
o
r
t
i
n
g
f
u
l
l
S
R
/
G
E
/
N
TD
c
u
r
v
e
s
is
e
sse
n
t
i
a
l
.
N
TD
@
S
R
=
8
0
-
9
0
%
c
o
mm
o
n
.
[
8
]
,
[
9
]
,
[
1
3
]
,
[
3
2
]
,
[
3
3
]
5.
DL
F
O
R
CRYP
T
ANA
L
YS
I
S
OF
SYM
M
E
T
RI
C
P
RI
M
I
T
I
V
E
S
5
.
1
.
M
o
t
iv
a
t
io
n
a
nd
s
co
pe
W
h
ile
DL
-
SC
A
tar
g
ets
im
p
lem
en
tatio
n
s
,
DL
is
also
ap
p
lied
to
attac
k
th
e
m
ath
e
m
atica
l
s
tr
u
ctu
r
e
of
s
y
m
m
etr
ic
cip
h
er
s
(
b
l
o
ck
cip
h
er
s
,
s
tr
ea
m
cip
h
er
s
)
.
G
o
als
in
clu
d
e
d
is
tin
g
u
is
h
er
s
d
if
f
er
e
n
tiatin
g
th
e
cip
h
e
r
f
r
o
m
a
r
a
n
d
o
m
p
er
m
u
tatio
n
with
f
ewe
r
s
am
p
les
th
an
b
r
u
t
e
-
f
o
r
ce
.
I
m
p
r
o
v
ed
d
if
f
er
e
n
tial/li
n
ea
r
cr
y
p
tan
al
y
s
is
f
in
d
in
g
b
etter
d
if
f
er
e
n
tial
ch
ar
ac
ter
is
tics
or
lin
ea
r
ap
p
r
o
x
i
m
atio
n
s
[
3
4
]
.
Key
r
ec
o
v
er
y
d
ir
ec
tly
or
in
d
ir
ec
tly
r
ed
u
cin
g
th
e
co
m
p
lex
ity
of
k
e
y
s
ea
r
ch
.
5
.
2
.
Neura
l
d
is
t
ing
uis
hers
C
o
n
ce
p
t
tr
ain
a
DL
m
o
d
el
(
ML
P,
C
N
N)
to
clas
s
if
y
tu
p
les
(
e.
g
.
,
p
air
s
of
p
lain
tex
ts
an
d
co
r
r
esp
o
n
d
in
g
ci
p
h
er
tex
ts
,
or
in
p
u
t/o
u
t
p
u
t
d
if
f
er
en
ce
s
)
as
o
r
ig
in
atin
g
f
r
o
m
th
e
tar
g
et
ci
p
h
er
or
a
r
a
n
d
o
m
p
er
m
u
tatio
n
[
7
]
,
[
1
1
]
,
[
1
2
]
.
PR
E
SENT
ca
s
e
s
tu
d
y
Mish
r
a
et
al
.
[
1
2
]
d
em
o
n
s
tr
ated
th
at
ML
Ps
co
u
ld
ef
f
ec
tiv
ely
d
is
tin
g
u
is
h
r
e
d
u
ce
d
-
r
o
u
n
d
v
er
s
io
n
s
(
e.
g
.
,
5
-
7
r
o
u
n
d
s
)
of
th
e
lig
h
tweig
h
t
cip
h
er
PR
E
SENT
f
r
o
m
r
an
d
o
m
with
h
ig
h
ac
cu
r
ac
y
.
T
h
is
d
em
o
n
s
tr
ated
DL
’
s
ab
ili
ty
to
ca
p
tu
r
e
n
o
n
-
r
a
n
d
o
m
n
ess
ar
is
in
g
f
r
o
m
th
e
cip
h
er
’
s
s
tr
u
ctu
r
e,
alb
eit
f
o
r
r
o
u
n
d
s
well
b
elo
w
th
e
f
u
ll
31
r
o
u
n
d
s
.
T
h
eir
wo
r
k
h
ig
h
lig
h
te
d
th
e
p
o
ten
tial
an
d
th
e
s
ig
n
if
ican
t
ch
allen
g
e
of
s
ca
lin
g
.
Me
th
o
d
o
lo
g
y
r
eq
u
ir
es
g
en
er
atin
g
v
ast
d
atasets
of
cip
h
er
an
d
r
an
d
o
m
p
er
m
u
tatio
n
o
u
t
p
u
ts
.
C
ar
ef
u
l
b
alan
cin
g
is
cr
u
cial.
E
v
alu
ati
o
n
in
v
o
lv
es
ac
c
u
r
ac
y
on
a
s
ep
ar
ate
test
s
et
an
d
o
f
ten
ca
lcu
latin
g
th
e
ad
v
an
tag
e
o
v
er
r
an
d
o
m
g
u
es
s
in
g
.
5
.
3
.
DL
-
enha
nced
diff
er
ent
ia
l
cr
y
pta
na
ly
s
is
G
o
hr
’
s
la
nd
m
a
r
k
Attack
Sp
ec
k
3
2
/6
4
Go
h
r
[
1
1
]
p
r
esen
ted
a
g
r
o
u
n
d
b
r
ea
k
i
n
g
ap
p
licatio
n
of
C
NNs
to
d
if
f
er
en
tial
cr
y
p
tan
aly
s
is
of
th
e
lig
h
tweig
h
t
AR
X
cip
h
er
Sp
ec
k
3
2
/6
4
.
Me
th
o
d
o
lo
g
y
i
)
T
r
ain
a
C
NN
to
p
r
e
d
ict
th
e
o
u
tp
u
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
Dee
p
lea
r
n
in
g
in
cryp
t
a
n
a
lysi
s
a
co
mp
r
eh
en
s
ive
r
ev
iew
o
f te
ch
n
iq
u
es,
a
p
p
lica
tio
n
s
,
…
(
Ou
s
s
a
ma
N
o
u
i
)
851
d
if
f
er
en
ce
of
t
h
e
last
r
o
u
n
d
g
i
v
en
th
e
i
n
p
u
t
d
if
f
er
e
n
ce
to
th
e
f
ir
s
t
r
o
u
n
d
(
o
r
a
d
if
f
er
e
n
tial
ch
ar
ac
ter
is
tic
o
v
e
r
m
u
ltip
le
r
o
u
n
d
s
)
.
ii
)
Use
th
is
n
eu
r
al
d
is
tin
g
u
is
h
er
to
f
ilter
cip
h
er
tex
t
p
air
s
more
lik
ely
to
f
o
llo
w
a
h
ig
h
-
p
r
o
b
a
b
ilit
y
d
if
f
er
en
tial
ch
ar
ac
ter
is
tic.
iii
)
U
s
e
th
e
f
ilter
ed
p
air
s
in
a
clas
s
ical
k
ey
r
ec
o
v
er
y
attac
k
on
th
e
last
r
o
u
n
d
(
s
)
.
R
esu
lts
Ach
iev
ed
s
ig
n
if
ican
t
im
p
r
o
v
e
m
en
ts
o
v
er
class
ical
d
if
f
er
en
tial
attac
k
s
on
r
o
u
n
d
-
r
ed
u
ce
d
Sp
ec
k
(
e.
g
.
,
up
to
11
r
o
u
n
d
s
)
.
T
h
e
C
NN
lear
n
ed
co
m
p
lex
d
if
f
er
en
tial
p
r
o
p
er
ties
d
ir
ec
tly
f
r
o
m
d
ata,
ef
f
ec
tiv
ely
au
to
m
atin
g
a
cr
u
cial
p
ar
t
of
th
e
cr
y
p
tan
aly
tic
p
r
o
ce
s
s
.
T
h
is
d
em
o
n
s
tr
ated
DL
’
s
p
o
ten
tial
to
im
p
r
o
v
e
tr
a
d
itio
n
al
cr
y
p
ta
n
aly
tic
tech
n
iq
u
es.
Sig
n
if
ican
ce
b
r
id
g
ed
th
e
g
a
p
b
etwe
en
“
ac
ad
em
ic
”
n
eu
r
al
d
is
tin
g
u
is
h
er
s
an
d
p
r
ac
tical
k
e
y
r
ec
o
v
e
r
y
,
al
b
eit
on
r
e
d
u
ce
d
r
o
u
n
d
s
of
a
lig
h
tweig
h
t
cip
h
er
.
5
.
4
.
Crit
ica
l
e
x
a
m
ina
t
io
n
of
neura
l
cr
y
pta
na
ly
s
is
B
en
am
ir
a
et
al
.
[
7
]
p
r
o
v
i
d
e
a
cr
u
cial
cr
itical
p
er
s
p
ec
tiv
e
d
i
s
tin
g
u
is
h
er
v
s
.
k
ey
r
ec
o
v
er
y
em
p
h
asize
th
at
a
g
o
o
d
n
eu
r
al
d
is
tin
g
u
is
h
er
d
o
es
not
au
to
m
atica
lly
tr
an
s
late
in
to
an
ef
f
icien
t
k
e
y
r
ec
o
v
er
y
attac
k
.
G
o
h
r
’
s
wo
r
k
is
n
o
tab
le
f
o
r
m
ak
in
g
th
i
s
lin
k
ex
p
licit.
T
he
“
wh
y
”
q
u
e
s
tio
n
cr
itically
q
u
esti
o
n
w
h
a
t
th
e
n
eu
r
al
n
etwo
r
k
is
ac
tu
ally
lear
n
in
g
.
Is
it
ca
p
t
u
r
in
g
f
u
n
d
am
en
tal
cr
y
p
to
g
r
a
p
h
ic
p
r
o
p
er
ties
(
lik
e
h
i
g
h
-
p
r
o
b
ab
ilit
y
d
if
f
e
r
en
tials
)
or
s
u
p
er
f
icial
p
atter
n
s
s
p
ec
if
i
c
to
th
e
tr
ain
in
g
d
ata
d
is
tr
ib
u
tio
n
?
L
ac
k
of
ex
p
lain
ab
ilit
y
is
a
m
ajo
r
co
n
ce
r
n
.
D
ata
ef
f
icien
cy
an
d
g
en
er
al
izatio
n
n
eu
r
al
d
is
tin
g
u
is
h
er
s
o
f
ten
r
e
q
u
ir
e
en
o
r
m
o
u
s
d
a
tasets
(
m
illi
o
n
s
of
s
am
p
les)
an
d
th
eir
g
en
er
aliza
tio
n
to
d
if
f
er
en
t
k
e
y
class
es
or
s
lig
h
tly
m
o
d
if
ied
cip
h
e
r
s
is
not
g
u
ar
an
teed
.
Scalab
ilit
y
th
e
ce
n
tr
al
ch
allen
g
e
r
em
ain
s
s
ca
lin
g
th
ese
ap
p
r
o
ac
h
es
to
f
u
ll
-
r
o
u
n
d
v
e
r
s
io
n
s
of
s
tan
d
ar
d
ized
,
s
ec
u
r
e
cip
h
er
s
lik
e
AE
S
-
1
2
8
.
C
u
r
r
en
t
s
u
cc
ess
es
ar
e
lar
g
ely
co
n
f
in
ed
to
lig
h
tweig
h
t
cip
h
er
s
or
s
ev
er
ely
r
ed
u
ce
d
r
o
u
n
d
s
.
Sin
g
h
et
al
.
[
6
]
also
h
ig
h
lig
h
t
th
is
lim
itatio
n
.
T
ab
le
2
h
ig
h
lig
h
ts
r
ep
r
esen
tativ
e
DL
ap
p
licatio
n
s
in
th
e
cr
y
p
tan
aly
s
is
of
s
y
m
m
etr
ic
p
r
im
itiv
es
an
d
s
u
m
m
ar
izes
k
ey
f
i
n
d
in
g
s
an
d
lim
itatio
n
s
.
T
ab
le
2
.
DL
a
p
p
licatio
n
s
in
s
y
m
m
etr
ic
p
r
im
itiv
e
cr
y
p
tan
aly
s
is
-
ex
ten
d
ed
a
n
aly
s
is
Ta
r
g
e
t
Te
c
h
n
i
q
u
e
K
e
y
f
i
n
d
i
n
g
Li
mi
t
a
t
i
o
n
s/
C
h
a
l
l
e
n
g
e
s
R
e
f
e
r
e
n
c
e
s
P
R
ESEN
T
(
r
e
d
u
c
e
d
)
M
LP
D
i
st
i
n
g
u
i
s
h
e
r
D
e
mo
n
st
r
a
t
e
d
c
a
p
a
b
i
l
i
t
y
of
M
LPs
to
d
i
s
t
i
n
g
u
i
sh
5
-
7
r
o
u
n
d
s
of
P
R
ES
EN
T
f
r
o
m
r
a
n
d
o
m
w
i
t
h
h
i
g
h
a
c
c
u
r
a
c
y
.
V
a
l
i
d
a
t
e
d
DL
’
s
a
b
i
l
i
t
y
to
d
e
t
e
c
t
n
o
n
-
r
a
n
d
o
m
n
e
s
s.
A
c
c
u
r
a
c
y
d
r
o
p
s
s
i
g
n
i
f
i
c
a
n
t
l
y
t
o
w
a
r
d
s
f
u
l
l
(
3
1
)
r
o
u
n
d
s.
Li
mi
t
e
d
to
d
i
st
i
n
g
u
i
s
h
e
r
,
n
o
t
k
e
y
r
e
c
o
v
e
r
y
.
La
c
k
of
i
n
si
g
h
t
i
n
t
o
w
h
y
.
[
1
2
]
S
p
e
c
k
3
2
/
6
4
(
r
e
d
u
c
e
d
)
C
N
N
D
i
f
f
e
r
e
n
t
i
a
l
A
i
d
C
N
N
l
e
a
r
n
e
d
to
p
r
e
d
i
c
t
d
i
f
f
e
r
e
n
t
i
a
l
c
h
a
r
a
c
t
e
r
i
s
t
i
c
s,
s
i
g
n
i
f
i
c
a
n
t
l
y
i
m
p
r
o
v
i
n
g
k
e
y
r
e
c
o
v
e
r
y
on
up
to
11
r
o
u
n
d
s
v
s
.
c
l
a
ss
i
c
a
l
d
i
f
f
e
r
e
n
t
i
a
l
c
r
y
p
t
a
n
a
l
y
s
i
s.
A
t
t
a
c
k
c
o
mp
l
e
x
i
t
y
s
t
i
l
l
h
i
g
h
.
F
u
l
l
c
i
p
h
e
r
(
2
2
/
2
3
r
o
u
n
d
s)
r
e
ma
i
n
s
sec
u
r
e
.
R
e
q
u
i
r
e
s
mass
i
v
e
t
r
a
i
n
i
n
g
d
a
t
a
(
1
0
0
M
+
sam
p
l
e
s)
.
[
1
1
]
G
e
n
e
r
a
l
d
i
s
t
i
n
g
u
i
sh
e
r
s
M
LP/
C
N
N
M
u
l
t
i
p
l
e
st
u
d
i
e
s
c
o
n
f
i
r
m
DL
’
s
a
b
i
l
i
t
y
to
f
i
n
d
st
a
t
i
st
i
c
a
l
d
i
st
i
n
g
u
i
s
h
e
r
s
f
o
r
v
a
r
i
o
u
s
r
e
d
u
c
e
d
-
r
o
u
n
d
c
i
p
h
e
r
s.
P
r
a
c
t
i
c
a
l
i
mp
a
c
t
on
f
u
l
l
c
i
p
h
e
r
s
mi
n
i
ma
l
so
f
a
r
.
H
i
g
h
d
a
t
a
r
e
q
u
i
r
e
m
e
n
t
s.
E
x
p
l
a
i
n
a
b
i
l
i
t
y
g
a
p
(
“
b
l
a
c
k
b
o
x
”
).
[
6
]
,
[
7
]
,
[
1
2
]
K
e
y
r
e
c
o
v
e
r
y
H
y
b
r
i
d
(DL
+
c
l
a
ss
i
c
a
l
)
M
o
s
t
p
r
o
mi
s
i
n
g
a
p
p
r
o
a
c
h
U
s
e
DL
d
i
s
t
i
n
g
u
i
sh
e
r
to
f
i
l
t
e
r
/
r
a
n
k
c
a
n
d
i
d
a
t
e
s
f
o
r
c
l
a
ss
i
c
a
l
k
e
y
r
e
c
o
v
e
r
y
st
e
p
s
(
l
i
k
e
G
o
h
r
[
1
1
]
)
.
D
i
r
e
c
t
DL
k
e
y
p
r
e
d
i
c
t
i
o
n
is
i
n
f
e
a
si
b
l
e
f
o
r
se
c
u
r
e
c
i
p
h
e
r
s
.
I
n
t
e
g
r
a
t
i
o
n
c
o
m
p
l
e
x
i
t
y
.
R
e
q
u
i
r
e
s
c
a
r
e
f
u
l
d
e
s
i
g
n
.
S
t
i
l
l
l
i
m
i
t
e
d
to
r
e
d
u
c
e
d
r
o
u
n
d
s/
l
i
g
h
t
w
e
i
g
h
t
t
a
r
g
e
t
s
.
S
c
a
l
i
n
g
is
t
h
e
c
o
r
e
i
ss
u
e
.
[
7
]
,
[
1
1
]
6.
CRIT
I
CA
L
ANA
L
YS
I
S
AN
D
L
I
M
I
T
AT
I
O
NS
6
.
1
.
T
he
ex
pla
ina
bil
it
y
(
“
B
la
ck
B
o
x
”
)
p
ro
blem
T
h
is
is
ar
g
u
ab
ly
th
e
m
o
s
t
s
ig
n
if
ican
t
cr
iticis
m
an
d
lim
itatio
n
[
3
5
]
,
[
3
6
]
.
L
ac
k
of
I
n
s
ig
h
t
DL
m
o
d
els,
esp
ec
ially
co
m
p
lex
C
NNs,
ar
e
o
p
aq
u
e.
It
’
s
ex
tr
em
el
y
d
if
f
i
cu
lt
to
u
n
d
er
s
tan
d
wh
y
a
m
o
d
el
m
ak
es
a
s
p
ec
if
ic
p
r
ed
ictio
n
or
wh
at
s
p
ec
if
ic
cr
y
p
to
g
r
ap
h
ic
p
r
o
p
e
r
ty
it
h
as
lear
n
ed
[
5
]
,
[
7
]
,
[
3
3
]
.
T
h
i
s
h
in
d
er
s
g
ain
in
g
n
ew
f
u
n
d
am
e
n
tal
cr
y
p
t
o
g
r
ap
h
ic
k
n
o
wled
g
e
f
r
o
m
s
u
cc
ess
f
u
l
attac
k
s
[
3
7
]
.
T
r
u
s
tin
g
th
e
m
o
d
el
’
s
o
u
tp
u
t,
esp
ec
ially
in
s
ec
u
r
ity
-
c
r
itical
co
n
tex
ts
.
Deb
u
g
g
in
g
or
im
p
r
o
v
in
g
m
o
d
els
s
y
s
tem
atica
lly
.
Ad
d
r
ess
in
g
XAI
in
cr
y
p
tan
aly
s
is
ap
p
l
y
in
g
XAI
te
ch
n
iq
u
es
(
s
alien
cy
m
a
p
s
,
L
I
M
E
,
SHAP)
to
cr
y
p
tan
aly
s
is
m
o
d
els
is
n
ascen
t
an
d
ch
allen
g
in
g
due
to
th
e
co
m
p
l
ex
ity
an
d
s
en
s
itiv
ity
of
cr
y
p
t
o
g
r
ap
h
ic
f
u
n
ctio
n
s
.
B
en
am
ir
a
et
al
.
[
7
]
s
tr
o
n
g
ly
ad
v
o
ca
te
f
o
r
r
esear
ch
in
to
in
t
er
p
r
etab
le
m
o
d
els
or
m
et
h
o
d
s
to
ex
tr
ac
t
h
u
m
an
-
u
n
d
e
r
s
tan
d
ab
le
r
u
les
f
r
o
m
DL
d
is
tin
g
u
is
h
er
s
.
DL
m
o
d
els
u
s
ed
in
cr
y
p
tan
aly
s
is
-
p
ar
ticu
lar
l
y
C
NNs
-
o
f
ten
b
eh
av
e
as
b
lac
k
b
o
x
es,
m
ak
in
g
it
d
if
f
icu
lt
to
e
x
p
lain
wh
y
a
p
r
ed
ictio
n
is
m
ad
e
or
wh
ich
cr
y
p
to
g
r
a
p
h
ic/leak
ag
e
f
ea
tu
r
es
ar
e
ex
p
lo
ite
d
.
T
h
i
s
lim
its
th
e
ex
tr
ac
tio
n
of
h
u
m
a
n
-
in
ter
p
r
etab
le
r
u
les,
s
y
s
tem
ati
c
d
eb
u
g
g
in
g
,
a
n
d
tr
u
s
t
in
s
ec
u
r
ity
-
cr
itical
s
ettin
g
s
.
Acc
o
r
d
in
g
ly
,
r
ec
en
t
wo
r
k
a
d
v
o
ca
tes
in
teg
r
atin
g
XAI
to
o
ls
(
e.
g
.
,
s
alien
cy
an
al
y
s
is
an
d
attr
ib
u
tio
n
m
eth
o
d
s
)
to
id
en
tify
in
f
lu
en
tial
s
am
p
les/
o
p
er
atio
n
s
an
d
to
co
n
n
ec
t
em
p
ir
ical
s
u
cc
ess
with
cr
y
p
t
o
g
r
ap
h
ic
in
s
ig
h
t
[
7
]
,
[
3
3
]
,
[
3
5
]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
4
6
-
8
5
5
852
6
.
2
.
Sca
la
bil
it
y
to
s
ec
ure
prim
it
iv
es
Scalab
ilit
y
r
em
ain
s
th
e
p
r
im
ar
y
b
ar
r
ier
f
o
r
DL
-
ass
is
ted
alg
o
r
ith
m
ic
cr
y
p
tan
al
y
s
is
.
C
u
r
r
en
t
s
u
cc
ess
es
m
ain
ly
tar
g
et
r
ed
u
ce
d
-
r
o
u
n
d
co
n
f
ig
u
r
atio
n
s
an
d
lig
h
tweig
h
t
d
esig
n
s
;
ex
ten
d
i
n
g
n
eu
r
al
d
i
s
tin
g
u
is
h
er
s
or
DL
-
aid
ed
d
i
f
f
er
en
tials
to
f
u
ll
-
r
o
u
n
d
s
tan
d
ar
d
ized
p
r
im
itiv
es
f
ac
e
s
an
ex
p
o
n
e
n
tial
g
r
o
wth
in
co
m
p
lex
ity
a
n
d
d
ata
r
eq
u
ir
em
e
n
ts
an
d
r
em
ain
s
an
o
p
en
ch
allen
g
e
[
6
]
,
[
7
]
,
[
3
8
]
.
In
DL
-
SC
A,
th
e
an
alo
g
o
u
s
f
r
o
n
tier
is
ef
f
icien
t
k
ey
r
ec
o
v
er
y
u
n
d
er
h
ig
h
-
o
r
d
e
r
m
ask
in
g
,
v
e
r
y
lo
w
SNR
,
an
d
h
ar
d
en
e
d
s
ec
u
r
e
h
ar
d
war
e.
F
r
o
m
a
d
ep
lo
y
m
en
t
p
er
s
p
ec
tiv
e
,
p
r
ac
tical
SC
A
p
ip
elin
es
m
ay
also
r
ely
on
e
d
g
e/clo
u
d
o
f
f
lo
a
d
in
g
f
o
r
tr
a
ce
p
r
o
ce
s
s
in
g
an
d
in
f
er
en
ce
,
w
h
er
e
laten
cy
an
d
s
er
v
ice
p
lace
m
en
t
c
o
n
s
tr
ain
ts
b
ec
o
m
e
r
elev
an
t
[
3
9
]
.
7.
F
UT
UR
E
RE
SE
A
RCH
DIR
E
CT
I
O
NS
7
.
1
.
XAI
f
o
r
cr
y
pta
na
ly
s
is
Dev
elo
p
in
g
s
p
ec
ialized
XAI
tech
n
iq
u
es
to
in
ter
p
r
et
wh
at
DL
m
o
d
els
lear
n
a
b
o
u
t
cr
y
p
to
g
r
ap
h
ic
alg
o
r
ith
m
s
or
leak
ag
e
f
u
n
ctio
n
s
.
Desig
n
in
g
in
h
e
r
en
tly
m
o
r
e
in
ter
p
r
etab
le
DL
ar
ch
itectu
r
es
f
o
r
cr
y
p
tan
aly
s
is
.
Usi
n
g
in
s
ig
h
ts
f
r
o
m
XAI
to
g
u
id
e
t
r
ad
itio
n
al
cr
y
p
ta
n
aly
s
is
or
im
p
r
o
v
e
cip
h
er
d
esig
n
s
[
4
0
]
,
[
4
1
]
.
Prio
r
ity
s
h
o
u
ld
be
g
iv
en
to
XAI
m
et
h
o
d
s
tailo
r
ed
to
cr
y
p
to
g
r
a
p
h
ic
an
d
leak
ag
e
d
o
m
ain
s
to
r
e
v
ea
l
wh
ich
o
p
e
r
atio
n
s
,
tim
e
s
am
p
les,
or
b
its
d
r
iv
e
m
o
d
el
d
ec
is
io
n
s
an
d
to
ex
tr
ac
t
h
u
m
an
-
u
n
d
er
s
tan
d
ab
le
r
u
les
f
r
o
m
tr
ain
ed
m
o
d
els.
7
.
2
.
B
ridg
ing
t
he
t
heo
ry
-
pra
ct
ice
gap
A
m
ajo
r
o
p
e
n
c
h
allen
g
e
is
to
t
r
an
s
late
em
p
ir
ical
n
eu
r
al
d
is
tin
g
u
is
h
er
s
i
n
to
th
e
o
r
y
-
b
ac
k
ed
cr
y
p
tan
aly
tic
g
u
ar
a
n
tees.
B
r
id
g
in
g
th
is
th
eo
r
y
–
p
r
ac
tice
g
a
p
r
eq
u
ir
es
clea
r
er
c
o
n
n
ec
ti
o
n
s
b
etwe
en
wh
at
a
m
o
d
el
lear
n
s
,
th
e
s
tatis
tical
ad
v
an
tag
e
it
ac
h
iev
es,
an
d
h
o
w
th
is
can
be
co
n
v
er
ted
in
t
o
p
r
ac
tical
k
ey
-
r
ec
o
v
e
r
y
or
p
r
o
o
f
of
wea
k
n
ess
[
4
2
]
.
7
.
3
.
I
m
pro
v
ing
da
t
a
ef
f
iciency
a
nd
g
ener
a
liza
t
io
n
T
r
an
s
f
er
lear
n
in
g
ad
ap
tin
g
m
o
d
els
p
r
e
-
tr
ain
e
d
on
one
d
ev
ice
/cip
h
er
/tas
k
to
p
er
f
o
r
m
well
on
a
r
elate
d
tar
g
et
with
m
in
im
al
n
ew
d
ata
[
4
3
]
.
Few
-
s
h
o
t/m
eta
-
lear
n
in
g
en
a
b
lin
g
m
o
d
els
to
lear
n
ef
f
ec
tiv
ely
f
r
o
m
v
er
y
f
ew
ex
am
p
les,
c
r
u
cial
f
o
r
attac
k
in
g
r
ar
e
or
h
ig
h
ly
s
ec
u
r
ed
d
ev
ices
[
4
4
]
.
Sy
n
th
etic
d
ata
g
en
er
atio
n
e
x
p
lo
r
in
g
h
ig
h
-
f
i
d
elity
s
im
u
latio
n
or
g
en
er
ativ
e
m
o
d
els
(
GANs,
d
if
f
u
s
io
n
m
o
d
els
)
to
cr
ea
te
r
ea
l
is
tic
tr
ain
in
g
d
ata,
r
ed
u
cin
g
r
elian
ce
on
p
h
y
s
ical
ac
q
u
is
itio
n
,
esp
ec
ially
f
o
r
al
g
o
r
ith
m
ic
attac
k
s
[
4
5
]
.
Self
-
s
u
p
er
v
is
ed
lear
n
in
g
L
ev
er
ag
in
g
u
n
lab
ele
d
d
ata
f
o
r
p
r
e
-
tr
ain
in
g,
r
ed
u
cin
g
t
h
e
b
u
r
d
en
of
lab
eled
d
ata
ac
q
u
is
itio
n
.
T
r
an
s
f
er
lear
n
in
g
,
f
ew
-
s
h
o
t/m
eta
-
lear
n
in
g
,
an
d
h
ig
h
-
f
i
d
elity
s
y
n
th
etic
d
ata
g
en
er
atio
n
(
e.
g
.
,
GANs/d
if
f
u
s
io
n
)
can
r
ed
u
ce
p
r
o
f
ilin
g
co
s
t
an
d
im
p
r
o
v
e
r
o
b
u
s
tn
ess
ac
r
o
s
s
d
ev
ices
an
d
n
o
i
s
e
r
eg
im
es.
7
.
4
.
Adv
a
nced
a
rc
hite
ct
ures
a
nd
t
ec
hn
iqu
e
s
Gr
ap
h
n
eu
r
al
n
etwo
r
k
s
,
tr
an
s
f
o
r
m
er
s
,
r
ein
f
o
r
ce
m
en
t
lear
n
i
n
g
,
an
d
en
s
em
b
le/f
u
s
io
n
ap
p
r
o
ac
h
es
ar
e
p
r
o
m
is
in
g
f
o
r
m
o
d
elin
g
s
tr
u
ct
u
r
ed
d
e
p
en
d
e
n
cies
in
cip
h
er
s
t
ates
or
lo
n
g
tr
ac
es
an
d
f
o
r
au
t
o
m
atin
g
p
a
r
ts
of
th
e
cr
y
p
tan
aly
tic
s
ea
r
c
h
p
r
o
ce
s
s
[
4
6
]
–
[
4
8
]
.
E
x
te
n
d
in
g
n
eu
r
al
cr
y
p
tan
aly
s
is
b
ey
o
n
d
r
ed
u
ce
d
-
r
o
u
n
d
b
lo
c
k
cip
h
e
r
s
to
ad
d
itio
n
al
p
r
im
itiv
es
an
d
le
ak
ag
e
m
o
d
alities
is
a
p
r
o
m
is
in
g
d
ir
ec
tio
n
b
u
t
r
eq
u
ir
es
ca
r
e
f
u
l
th
r
ea
t
mod
elin
g
an
d
s
ca
lab
le
tr
ain
in
g
p
ip
elin
es
[
4
9
]
.
7
.
5
.
St
a
nd
a
rdiza
t
io
n
a
nd
re
pro
du
cibi
lity
C
o
m
m
u
n
ity
b
en
c
h
m
ar
k
s
,
s
tan
d
ar
d
ized
ev
al
u
atio
n
p
r
o
to
co
ls
,
an
d
r
ep
r
o
d
u
cib
le
o
p
e
n
-
s
o
u
r
ce
to
o
lin
g
ar
e
ess
en
tial
to
en
ab
le
f
air
co
m
p
ar
is
o
n
an
d
ac
ce
ler
ate
p
r
o
g
r
ess
.
C
o
m
p
lem
en
tar
y
to
em
p
ir
i
ca
l
b
en
ch
m
ar
k
in
g
,
f
o
r
m
al
m
et
h
o
d
s
f
o
r
v
er
if
icati
o
n
/cer
tific
atio
n
of
im
p
lem
en
t
atio
n
s
an
d
d
ef
en
s
es
can
s
tr
en
g
th
en
c
o
n
f
id
e
n
ce
in
s
ec
u
r
ity
claim
s
,
esp
ec
ially
wh
en
ML
co
m
p
o
n
e
n
ts
ar
e
u
s
ed
in
th
e
attac
k
or
d
ef
e
n
s
e
lo
o
p
[
5
0
]
,
[
5
1
]
.
8.
CO
NC
L
U
SI
O
N
DL
h
as
s
ig
n
if
ican
tly
r
esh
ap
ed
th
e
lan
d
s
ca
p
e
of
m
o
d
er
n
cr
y
p
tan
aly
s
is
.
In
t
h
e
d
o
m
ain
of
SC
A
,
DL
-
p
ar
ticu
lar
ly
CCN
s
-
h
as
em
er
g
ed
as
th
e
s
tate
of
th
e
ar
t
f
o
r
p
r
o
f
iled
attac
k
s
,
en
ab
lin
g
d
ir
e
c
t
p
r
o
ce
s
s
in
g
of
r
aw
tr
ac
es,
im
p
r
o
v
ed
r
o
b
u
s
tn
ess
to
n
o
is
e
an
d
co
m
m
o
n
co
u
n
ter
m
ea
s
u
r
es
s
u
ch
as
jitt
er
,
an
d
ef
f
ec
tiv
e
k
ey
r
ec
o
v
e
r
y
with
f
ewe
r
tr
ac
es.
T
h
ese
ad
v
a
n
ce
s
,
d
em
o
n
s
tr
ated
in
wo
r
k
s
s
u
ch
as
[
8
]
,
[
9
]
,
an
d
s
y
s
tem
atiz
ed
in
[
6
]
,
[
1
3
]
,
[
3
2
]
,
r
ep
r
esen
t
a
s
u
b
s
tan
tial
p
r
ac
tic
al
im
p
r
o
v
em
e
n
t
o
v
e
r
class
ical
tech
n
iq
u
es
.
Fo
r
alg
o
r
ith
m
ic
cr
y
p
ta
n
aly
s
is
of
s
y
m
m
etr
ic
p
r
im
itiv
es,
DL
h
as
s
h
o
wn
p
r
o
m
is
e
in
au
to
m
a
tin
g
task
s
s
u
ch
as
s
tati
s
tical
d
is
t
in
g
u
is
h
er
s
an
d
th
e
d
is
co
v
er
y
of
d
i
f
f
er
en
tial
ch
ar
ac
ter
is
tics
.
No
tab
ly
,
Go
h
r
’
s
wo
r
k
on
Sp
ec
k
[
1
1
]
illu
s
tr
ates
how
n
e
u
r
al
m
o
d
els
can
en
h
a
n
ce
clas
s
ical
cr
y
p
tan
aly
tic
wo
r
k
f
lo
ws
an
d
im
p
r
o
v
e
k
ey
r
ec
o
v
er
y
on
r
ed
u
ce
d
-
r
o
u
n
d
c
ip
h
er
s
.
Ho
wev
er
,
cu
r
r
en
t
s
u
c
ce
s
s
es
r
em
ain
lar
g
ely
co
n
f
in
ed
to
lig
h
tweig
h
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
Dee
p
lea
r
n
in
g
in
cryp
t
a
n
a
lysi
s
a
co
mp
r
eh
en
s
ive
r
ev
iew
o
f te
ch
n
iq
u
es,
a
p
p
lica
tio
n
s
,
…
(
Ou
s
s
a
ma
N
o
u
i
)
853
d
esig
n
s
or
r
ed
u
ce
d
-
r
o
u
n
d
v
ar
i
an
ts
,
an
d
s
ca
lin
g
s
u
ch
a
p
p
r
o
a
ch
es
to
f
u
ll
-
r
o
u
n
d
s
tan
d
a
r
d
i
ze
d
cip
h
er
s
co
n
tin
u
es
to
be
an
o
p
en
an
d
c
h
allen
g
in
g
p
r
o
b
lem
.
T
h
is
r
ev
iew
h
ig
h
lig
h
ts
lim
itatio
n
s
th
at
co
n
s
tr
ain
b
r
o
a
d
er
im
p
ac
t:
th
e
b
lack
-
b
o
x
n
atu
r
e
of
d
ee
p
m
o
d
els
lim
its
ex
p
lain
ab
ilit
y
an
d
t
h
e
ex
tr
ac
tio
n
of
cr
y
p
to
g
r
ap
h
ic
in
s
ig
h
t,
wh
ile
d
at
a
d
ep
en
d
en
cy
an
d
g
en
er
aliza
tio
n
is
s
u
es
h
in
d
er
r
o
b
u
s
tn
ess
ac
r
o
s
s
d
ev
ices,
im
p
lem
en
tatio
n
s
,
an
d
attac
k
s
ce
n
ar
io
s
.
Ad
d
itio
n
ally
,
co
m
p
u
tatio
n
al
co
s
ts
an
d
th
e
n
ee
d
f
o
r
r
ig
o
r
o
u
s
,
r
e
p
r
o
d
u
cib
le
ev
alu
atio
n
f
r
am
ewo
r
k
s
r
em
ain
im
p
o
r
tan
t
co
n
ce
r
n
s
.
Fu
tu
r
e
r
esear
c
h
s
h
o
u
ld
th
er
ef
o
r
e
f
o
cu
s
on
im
p
r
o
v
in
g
ex
p
lain
ab
ilit
y
th
r
o
u
g
h
XAI
,
b
r
id
g
in
g
th
e
th
eo
r
y
-
p
r
ac
tice
g
ap
,
e
n
h
an
cin
g
d
ata
ef
f
icien
cy
a
n
d
g
e
n
er
a
lizatio
n
,
an
d
ex
p
lo
r
in
g
n
o
v
el
ar
ch
itectu
r
es
an
d
h
y
b
r
id
s
tr
ateg
ies
ca
p
ab
le
of
ad
d
r
ess
in
g
more
co
m
p
lex
cr
y
p
to
g
r
ap
h
ic
tar
g
ets
.
W
h
ile
DL
is
not
a
u
n
iv
er
s
al
s
o
lu
tio
n
to
c
r
y
p
ta
n
aly
s
is
,
its
ab
ilit
y
to
au
t
o
m
ate
co
m
p
lex
p
atter
n
r
ec
o
g
n
itio
n
en
s
u
r
es
t
h
at
it
will
r
em
ain
a
p
o
wer
f
u
l
an
d
ev
o
lv
i
n
g
to
o
l
with
in
th
e
o
n
g
o
in
g
ar
m
s
r
ac
e
b
etwe
en
cr
y
p
to
g
r
ap
h
ic
d
esig
n
an
d
cr
y
p
tan
aly
tic
attac
k
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
T
h
e
au
th
o
r
s
s
tate
no
f
u
n
d
in
g
is
in
v
o
lv
ed
.
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
.
Na
m
e
of
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Ou
s
s
am
a
N
oui
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Am
in
e
B
ar
k
at
✓
✓
✓
✓
✓
✓
✓
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
OF
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
no
co
n
f
lict
of
in
ter
est
.
RE
F
E
R
E
NC
E
S
[
1
]
F
.
E
.
P
o
t
e
st
a
d
-
O
r
d
o
n
e
z
,
E.
T
e
n
a
-
S
a
n
c
h
e
z
,
A
.
J.
A
c
o
st
a
-
J
i
me
n
e
z
,
C
.
J.
Ji
m
e
n
e
z
-
F
e
r
n
a
n
d
e
z
,
a
n
d
R
.
C
h
a
v
e
s,
“
D
e
s
i
g
n
a
n
d
e
v
a
l
u
a
t
i
o
n
o
f
c
o
u
n
t
e
r
m
e
a
s
u
r
e
s
a
g
a
i
n
s
t
f
a
u
l
t
i
n
j
e
c
t
i
o
n
a
t
t
a
c
k
s
a
n
d
p
o
w
e
r
si
d
e
-
c
h
a
n
n
e
l
l
e
a
k
a
g
e
e
x
p
l
o
r
a
t
i
o
n
f
o
r
A
ES
b
l
o
c
k
c
i
p
h
e
r
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
6
5
5
4
8
–
6
5
5
6
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
1
8
3
7
6
4
.
[
2
]
T.
S
u
g
a
w
a
r
a
,
N
.
S
h
o
j
i
,
K
.
S
a
k
i
y
a
ma,
K
.
M
a
t
su
d
a
,
N
.
M
i
u
r
a
,
a
n
d
M
.
N
a
g
a
t
a
,
“
S
i
d
e
-
c
h
a
n
n
e
l
l
e
a
k
a
g
e
f
r
o
m
sen
so
r
-
b
a
se
d
c
o
u
n
t
e
r
m
e
a
s
u
r
e
s
a
g
a
i
n
st
f
a
u
l
t
i
n
j
e
c
t
i
o
n
a
t
t
a
c
k
,
”
Mi
c
r
o
e
l
e
c
t
r
o
n
i
c
s
J
o
u
rn
a
l
,
v
o
l
.
9
0
,
p
p
.
6
3
–
7
1
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
me
j
o
.
2
0
1
9
.
0
5
.
0
1
7
.
[
3
]
N
.
B
e
n
h
a
d
j
y
o
u
ssef,
M
.
K
a
r
ma
n
i
,
a
n
d
M
.
M
a
c
h
h
o
u
t
,
“
P
o
w
e
r
-
b
a
se
d
s
i
d
e
c
h
a
n
n
e
l
a
n
a
l
y
si
s
a
n
d
f
a
u
l
t
i
n
j
e
c
t
i
o
n
:
h
a
c
k
i
n
g
t
e
c
h
n
i
q
u
e
s
a
n
d
c
o
m
b
i
n
e
d
c
o
u
n
t
e
r
me
a
su
r
e
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
A
d
v
a
n
c
e
d
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
a
n
d
Ap
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
2
,
n
o
.
5
,
p
p
.
7
0
9
–
7
1
9
,
2
0
2
1
,
d
o
i
:
1
0
.
1
4
5
6
9
/
I
JA
C
S
A
.
2
0
2
1
.
0
1
2
0
5
8
3
.
[
4
]
S
.
N
a
sere
l
d
e
n
,
N
.
A
l
i
a
s,
A
.
A
l
t
i
g
a
n
i
,
A
.
M
o
h
a
me
d
,
a
n
d
S
.
B
a
d
r
e
d
d
i
n
e
,
“
A
d
v
a
n
c
e
a
t
t
a
c
k
s
o
n
A
ES
:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
o
f
si
d
e
c
h
a
n
n
e
l
,
f
a
u
l
t
i
n
j
e
c
t
i
o
n
,
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
q
u
a
n
t
u
m
t
e
c
h
n
i
q
u
e
s,
”
Ed
e
l
w
e
i
s
s
A
p
p
l
i
e
d
S
c
i
e
n
c
e
a
n
d
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
9
,
n
o
.
4
,
p
p
.
2
4
7
1
–
2
4
8
6
,
2
0
2
5
,
d
o
i
:
1
0
.
5
5
2
1
4
/
2
5
7
6
8
4
8
4
.
v
9
i
4
.
6
5
8
6
.
[5
]
N
.
M
o
h
a
me
d
,
“
A
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
a
n
d
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
i
n
c
y
b
e
r
se
c
u
r
i
t
y
:
a
d
e
e
p
d
i
v
e
i
n
t
o
s
t
a
t
e
-
of
-
t
h
e
-
a
r
t
t
e
c
h
n
i
q
u
e
s
a
n
d
f
u
t
u
r
e
p
a
r
a
d
i
g
ms
,
”
K
n
o
w
l
e
d
g
e
a
n
d
I
n
f
o
rm
a
t
i
o
n
S
y
s
t
e
m
s
,
v
o
l
.
6
7
,
n
o
.
8
,
p
p
.
6
9
6
9
–
7
0
5
5
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
0
1
1
5
-
0
2
5
-
0
2
4
2
9
-
y.
[
6
]
A
.
S
i
n
g
h
,
K
.
B
.
S
i
v
a
n
g
i
,
a
n
d
A
.
N
.
T
e
n
t
u
,
“
M
a
c
h
i
n
e
l
e
a
r
n
i
n
g
a
n
d
c
r
y
p
t
a
n
a
l
y
s
i
s:
a
n
i
n
-
d
e
p
t
h
e
x
p
l
o
r
a
t
i
o
n
o
f
c
u
r
r
e
n
t
p
r
a
c
t
i
c
e
s
a
n
d
f
u
t
u
r
e
p
o
t
e
n
t
i
a
l
,
”
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
i
n
g
T
h
e
o
r
i
e
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
1
,
n
o
.
3
,
p
p
.
2
5
7
–
2
7
2
,
2
0
2
4
,
d
o
i
:
1
0
.
6
2
4
1
1
/
j
c
t
a
.
9
8
5
1
.
[
7
]
A
.
B
e
n
a
m
i
r
a
,
D
.
G
e
r
a
u
l
t
,
T
.
P
e
y
r
i
n
,
a
n
d
Q
.
Q
.
Ta
n
,
“
A
d
e
e
p
e
r
l
o
o
k
a
t
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
-
b
a
s
e
d
c
r
y
p
t
a
n
a
l
y
si
s
,
”
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
seri
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
2
6
9
6
LN
C
S
,
p
p
.
8
0
5
–
8
3
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
7
7
8
7
0
-
5
_
2
8
.
[
8
]
E.
C
a
g
l
i
,
C
.
D
u
m
a
s,
a
n
d
E.
P
r
o
u
f
f
,
“
C
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
w
i
t
h
d
a
t
a
a
u
g
me
n
t
a
t
i
o
n
a
g
a
i
n
st
j
i
t
t
e
r
-
b
a
se
d
c
o
u
n
t
e
r
m
e
a
s
u
r
e
s:
P
r
o
f
i
l
i
n
g
a
t
t
a
c
k
s
w
i
t
h
o
u
t
p
r
e
-
p
r
o
c
e
s
si
n
g
,
”
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
1
0
5
2
9
LN
C
S
,
p
p
.
4
5
–
6
8
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
6
6
7
8
7
-
4
_
3
.
[
9
]
J.
K
i
m,
S
.
P
i
c
e
k
,
A
.
H
e
u
s
e
r
,
S
.
B
h
a
si
n
,
a
n
d
A
.
H
a
n
j
a
l
i
c
,
“
M
a
k
e
s
o
m
e
n
o
i
se.
u
n
l
e
a
s
h
i
n
g
t
h
e
p
o
w
e
r
o
f
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
u
r
a
l
n
e
t
w
o
r
k
s
f
o
r
p
r
o
f
i
l
e
d
s
i
d
e
-
c
h
a
n
n
e
l
a
n
a
l
y
si
s
,
”
I
A
C
R
T
r
a
n
sa
c
t
i
o
n
s
o
n
C
ry
p
t
o
g
ra
p
h
i
c
H
a
r
d
w
a
r
e
a
n
d
E
m
b
e
d
d
e
d
S
y
st
e
m
s
,
p
p
.
1
4
8
–
1
7
9
,
2
0
1
9
,
d
o
i
:
1
0
.
4
6
5
8
6
/
t
c
h
e
s.v
2
0
1
9
.
i
3
.
1
4
8
-
1
7
9
.
[
1
0
]
P
u
r
w
o
n
o
,
A
.
M
a
’
a
r
i
f
,
W
.
R
a
h
ma
n
i
a
r
,
H
.
I
.
K
.
F
a
t
h
u
r
r
a
h
ma
n
,
A
.
Z.
K
.
F
r
i
s
k
y
,
a
n
d
Q
.
M
.
U
.
H
a
q
,
“
U
n
d
e
r
st
a
n
d
i
n
g
o
f
C
o
n
v
o
l
u
t
i
o
n
a
l
N
e
u
r
a
l
N
e
t
w
o
r
k
(
C
N
N
)
:
a
r
e
v
i
e
w
,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
R
o
b
o
t
i
c
s
a
n
d
C
o
n
t
ro
l
S
y
st
e
m
s
,
v
o
l
.
2
,
n
o
.
4
,
p
p
.
7
3
9
–
7
4
8
,
2
0
2
2
,
d
o
i
:
1
0
.
3
1
7
6
3
/
i
j
r
c
s.v
2
i
4
.
8
8
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
5
0
2
-
4
7
5
2
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
,
Vo
l.
42
,
No
.
3
,
J
u
n
e
20
2
6
:
8
4
6
-
8
5
5
854
[
1
1
]
A
.
G
o
h
r
,
“
I
mp
r
o
v
i
n
g
a
t
t
a
c
k
s
o
n
r
o
u
n
d
-
r
e
d
u
c
e
d
s
p
e
c
k
3
2
/
6
4
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
,
”
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
su
b
s
e
ri
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
1
6
9
3
LN
C
S
,
p
p
.
1
5
0
–
1
7
9
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
030
-
2
6
9
5
1
-
7
_
6
.
[
1
2
]
G
.
M
i
s
h
r
a
,
S
.
V
.
S
.
S
.
N
.
V
.
G
.
K
r
i
sh
n
a
M
u
r
t
h
y
,
a
n
d
S
.
K
.
P
a
l
,
“
N
e
u
r
a
l
n
e
t
w
o
r
k
b
a
s
e
d
a
n
a
l
y
si
s
o
f
l
i
g
h
t
w
e
i
g
h
t
b
l
o
c
k
c
i
p
h
e
r
P
R
ESEN
T,
”
A
d
v
a
n
c
e
s i
n
I
n
t
e
l
l
i
g
e
n
t
S
y
st
e
m
s
a
n
d
C
o
m
p
u
t
i
n
g
,
v
o
l
.
7
4
1
,
p
p
.
9
6
9
–
9
7
8
,
2
0
1
9
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
981
-
13
-
0
7
6
1
-
4
_
9
1
.
[
1
3
]
S
.
P
i
c
e
k
,
G
.
P
e
r
i
n
,
L.
M
a
r
i
o
t
,
L.
W
u
,
a
n
d
L.
B
a
t
i
n
a
,
“
S
o
K
:
d
e
e
p
l
e
a
r
n
i
n
g
-
b
a
se
d
p
h
y
s
i
c
a
l
si
d
e
-
c
h
a
n
n
e
l
a
n
a
l
y
si
s
,
”
AC
M
C
o
m
p
u
t
i
n
g
S
u
r
v
e
y
s
,
v
o
l
.
5
5
,
n
o
.
1
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
4
5
/
3
5
6
9
5
7
7
.
[
1
4
]
R
.
A
.
d
e
O
l
i
v
e
i
r
a
a
n
d
M
.
H
.
J.
B
o
l
l
e
n
,
“
D
e
e
p
l
e
a
r
n
i
n
g
f
o
r
p
o
w
e
r
q
u
a
l
i
t
y
,
”
E
l
e
c
t
r
i
c
Po
w
e
r
S
y
s
t
e
m
s
Re
s
e
a
r
c
h
,
v
o
l
.
2
1
4
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
p
sr
.
2
0
2
2
.
1
0
8
8
8
7
.
[
1
5
]
K
.
S
h
a
r
i
f
a
n
i
a
n
d
M
.
A
mi
n
i
,
“
M
a
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
:
a
r
e
v
i
e
w
o
f
m
e
t
h
o
d
s
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s,
”
Wo
r
l
d
I
n
f
o
rm
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
a
n
d
En
g
i
n
e
e
r
i
n
g
J
o
u
rn
a
l
,
v
o
l
.
1
0
,
n
o
.
7
,
p
p
.
3
8
9
8
–
3
9
0
4
,
2
0
2
3
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
ssr
n
.
c
o
m/
a
b
s
t
r
a
c
t
=
4
4
5
8
7
2
3
.
[
1
6
]
A
.
M
e
h
r
i
s
h
,
N
.
M
a
j
u
md
e
r
,
R
.
B
h
a
r
a
d
w
a
j
,
R
.
M
i
h
a
l
c
e
a
,
a
n
d
S
.
P
o
r
i
a
,
“
A
r
e
v
i
e
w
o
f
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
f
o
r
s
p
e
e
c
h
p
r
o
c
e
ss
i
n
g
,
”
In
f
o
rm
a
t
i
o
n
F
u
si
o
n
,
v
o
l
.
9
9
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
i
n
f
f
u
s.
2
0
2
3
.
1
0
1
8
6
9
.
[
1
7
]
S
.
J.
D
.
P
r
i
n
c
e
,
U
n
d
e
rst
a
n
d
i
n
g
d
e
e
p
l
e
a
rn
i
n
g
,
n
o
.
M
a
y
2
9
.
C
a
m
b
r
i
d
g
e
:
M
I
T
P
r
e
ss,
2
0
2
5
.
[
1
8
]
D
.
Z
.
C
h
e
n
,
F
.
Tr
e
v
i
z
a
n
,
a
n
d
S
.
Th
i
é
b
a
u
x
,
“
R
e
t
u
r
n
t
o
t
r
a
d
i
t
i
o
n
:
l
e
a
r
n
i
n
g
r
e
l
i
a
b
l
e
h
e
u
r
i
st
i
c
s
w
i
t
h
c
l
a
s
si
c
a
l
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
,
”
Pro
c
e
e
d
i
n
g
s
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
A
u
t
o
m
a
t
e
d
P
l
a
n
n
i
n
g
a
n
d
S
c
h
e
d
u
l
i
n
g
,
I
C
AP
S
,
v
o
l
.
3
4
,
p
p
.
6
8
–
7
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
6
0
9
/
i
c
a
p
s.
v
3
4
i
1
.
3
1
4
6
2
.
[
1
9
]
E.
B
r
i
e
r
,
C
.
C
l
a
v
i
e
r
,
a
n
d
F
.
O
l
i
v
i
e
r
,
“
C
o
r
r
e
l
a
t
i
o
n
p
o
w
e
r
a
n
a
l
y
si
s
w
i
t
h
a
l
e
a
k
a
g
e
m
o
d
e
l
,
”
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
su
b
s
e
ri
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
3
1
5
6
,
p
p
.
1
6
–
2
9
,
2
0
0
4
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
540
-
2
8
6
3
2
-
5
_
2
.
[
2
0
]
P
.
B
o
t
t
i
n
e
l
l
i
a
n
d
J
.
W
.
B
o
s,
“
C
o
m
p
u
t
a
t
i
o
n
a
l
a
s
p
e
c
t
s
o
f
c
o
r
r
e
l
a
t
i
o
n
p
o
w
e
r
a
n
a
l
y
s
i
s,
”
J
o
u
rn
a
l
o
f
C
ry
p
t
o
g
r
a
p
h
i
c
E
n
g
i
n
e
e
r
i
n
g
,
v
o
l
.
7
,
n
o
.
3
,
p
p
.
1
6
7
–
1
8
1
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
3
8
9
-
0
1
6
-
0
1
2
2
-
9.
[
2
1
]
N
.
O
u
ssam
a
,
B
.
A
ss
i
a
,
a
n
d
N
.
Le
m
n
o
u
a
r
,
“
S
e
c
u
r
e
i
ma
g
e
e
n
c
r
y
p
t
i
o
n
s
c
h
e
me
b
a
se
d
o
n
p
o
l
a
r
d
e
c
o
mp
o
si
t
i
o
n
a
n
d
c
h
a
o
t
i
c
m
a
p
,
”
I
n
t
e
r
n
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
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
T
e
c
h
n
o
l
o
g
y
,
v
o
l
.
1
0
,
n
o
.
4
,
p
p
.
4
3
7
–
4
5
3
,
2
0
1
7
,
d
o
i
:
1
0
.
1
5
0
4
/
I
J
I
C
T.
2
0
1
7
.
0
8
4
3
3
9
.
[
2
2
]
N
.
O
u
ssam
a
a
n
d
N
.
L
e
m
n
o
u
a
r
,
“
A
r
o
b
u
st
w
a
t
e
r
m
a
r
k
i
n
g
s
c
h
e
me
f
o
r
o
w
n
e
r
s
h
i
p
p
r
o
t
e
c
t
i
o
n
a
n
d
d
e
a
d
l
o
c
k
p
r
e
v
e
n
t
i
o
n
,
”
AC
M
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
Pr
o
c
e
e
d
i
n
g
S
e
r
i
e
s
,
v
o
l
.
2
3
-
25
-
N
o
v
e
,
2
0
1
5
,
d
o
i
:
1
0
.
1
1
4
5
/
2
8
1
6
8
3
9
.
2
8
1
6
9
1
5
.
[
2
3
]
O
.
H
a
r
k
a
t
i
,
L.
N
o
u
i
,
a
n
d
A
.
B
e
l
o
u
c
i
f
,
“
A
l
o
ssl
e
ss
p
r
o
b
a
b
i
l
i
st
i
c
i
ma
g
e
e
n
c
r
y
p
t
i
o
n
a
l
g
o
r
i
t
h
m
b
a
se
d
o
n
Q
S
D
d
e
c
o
m
p
o
si
t
i
o
n
a
n
d
mat
r
i
x
t
r
a
n
sf
o
r
m
a
t
i
o
n
s,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
m
p
u
t
e
rs
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
4
6
,
n
o
.
1
0
,
p
p
.
8
6
8
–
8
7
9
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
8
0
/
1
2
0
6
2
1
2
X
.
2
0
2
4
.
2
3
9
2
1
6
8
.
[
2
4
]
T.
D
.
B
a
r
v
e
,
A
.
Y
.
S
a
ma
n
t
,
a
n
d
M
.
S
.
K
u
l
a
y
e
,
“
A
c
o
m
p
a
r
a
t
i
v
e
st
u
d
y
o
f
o
p
t
i
m
i
z
a
t
i
o
n
a
l
g
o
r
i
t
h
ms
i
n
d
e
e
p
l
e
a
r
n
i
n
g
:
S
G
D
,
A
d
a
m
,
A
n
d
B
e
y
o
n
d
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
rn
a
l
o
f
C
o
m
p
u
t
e
r
T
e
c
h
n
o
l
o
g
y
a
n
d
E
l
e
c
t
ro
n
i
c
s
C
o
m
m
u
n
i
c
a
t
i
o
n
(
I
J
C
T
E
C
)
,
v
o
l
.
9
0
0
1
,
n
o
.
4
,
p
.
1
1
0
1
8
,
2
0
2
5
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
w
w
w
.
i
j
c
t
e
c
e
.
c
o
m
.
[
2
5
]
M
.
M
.
Ta
y
e
,
“
U
n
d
e
r
s
t
a
n
d
i
n
g
o
f
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
w
i
t
h
d
e
e
p
l
e
a
r
n
i
n
g
:
a
r
c
h
i
t
e
c
t
u
r
e
s
,
w
o
r
k
f
l
o
w
,
a
p
p
l
i
c
a
t
i
o
n
s
a
n
d
f
u
t
u
r
e
d
i
r
e
c
t
i
o
n
s,
”
C
o
m
p
u
t
e
rs
,
v
o
l
.
1
2
,
n
o
.
5
,
2
0
2
3
,
d
o
i
:
1
0
.
3
3
9
0
/
c
o
m
p
u
t
e
r
s
1
2
0
5
0
0
9
1
.
[
2
6
]
R
.
K
r
u
se,
S
.
M
o
st
a
g
h
i
m
,
C
.
B
o
r
g
e
l
t
,
C
.
B
r
a
u
n
e
,
a
n
d
M
.
S
t
e
i
n
b
r
e
c
h
e
r
,
“
M
u
l
t
i
-
l
a
y
e
r
p
e
r
c
e
p
t
r
o
n
s
,
”
p
p
.
5
3
–
1
2
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
0
-
4
2
2
2
7
-
1
_
5
.
[
2
7
]
I
.
El
Ja
a
f
a
r
i
,
A
.
El
l
a
h
y
a
n
i
,
a
n
d
S
.
C
h
a
r
f
i
,
“
R
e
c
t
i
f
i
e
d
n
o
n
-
l
i
n
e
a
r
u
n
i
t
f
o
r
c
o
n
v
o
l
u
t
i
o
n
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
J
o
u
rn
a
l
o
f
Ph
y
si
c
s:
C
o
n
f
e
re
n
c
e
S
e
r
i
e
s
,
v
o
l
.
1
7
4
3
,
n
o
.
1
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
8
8
/
1
7
4
2
-
6
5
9
6
/
1
7
4
3
/
1
/
0
1
2
0
1
4
.
[
2
8
]
A
.
J.
C
a
b
r
e
r
a
-
G
u
t
i
e
r
r
e
z
,
E.
C
a
st
i
l
l
o
,
A
.
Es
c
o
b
a
r
-
M
o
l
e
r
o
,
J.
A
.
A
l
v
a
r
e
z
-
B
e
r
mej
o
,
D
.
P
.
M
o
r
a
l
e
s,
a
n
d
L.
P
a
r
r
i
l
l
a
,
“
I
n
t
e
g
r
a
t
i
o
n
o
f
h
a
r
d
w
a
r
e
se
c
u
r
i
t
y
mo
d
u
l
e
s
a
n
d
p
e
r
mi
ssi
o
n
e
d
b
l
o
c
k
c
h
a
i
n
i
n
i
n
d
u
s
t
r
i
a
l
I
o
T
n
e
t
w
o
r
k
s,
”
I
EE
E
Ac
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
1
1
4
3
3
1
–
1
1
4
3
4
5
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
2
1
7
8
1
5
.
[
2
9
]
S
.
H
.
W
a
n
g
,
M
.
A
.
K
h
a
n
,
a
n
d
Y
.
D
.
Zh
a
n
g
,
“
V
I
S
P
N
N
:
V
G
G
-
i
n
sp
i
r
e
d
st
o
c
h
a
st
i
c
p
o
o
l
i
n
g
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
C
o
m
p
u
t
e
rs
,
M
a
t
e
ri
a
l
s
a
n
d
C
o
n
t
i
n
u
a
,
v
o
l
.
7
0
,
n
o
.
2
,
p
p
.
3
0
8
1
–
3
0
9
7
,
2
0
2
2
,
d
o
i
:
1
0
.
3
2
6
0
4
/
c
mc
.
2
0
2
2
.
0
1
9
4
4
7
.
[
3
0
]
J.
A
I
l
e
m
o
b
a
y
o
e
t
a
l
.
,
“
H
y
p
e
r
p
a
r
a
me
t
e
r
t
u
n
i
n
g
i
n
m
a
c
h
i
n
e
l
e
a
r
n
i
n
g
:
a
c
o
m
p
r
e
h
e
n
si
v
e
r
e
v
i
e
w
,
”
J
o
u
r
n
a
l
o
f
E
n
g
i
n
e
e
ri
n
g
R
e
se
a
rc
h
a
n
d
Re
p
o
r
t
s
,
v
o
l
.
2
6
,
n
o
.
6
,
p
p
.
3
8
8
–
3
9
5
,
2
0
2
4
,
d
o
i
:
1
0
.
9
7
3
4
/
j
e
r
r
/
2
0
2
4
/
v
2
6
i
6
1
1
8
8
.
[
3
1
]
H
.
M
a
g
h
r
e
b
i
,
T
.
P
o
r
t
i
g
l
i
a
t
t
i
,
a
n
d
E.
P
r
o
u
f
f
,
“
B
r
e
a
k
i
n
g
c
r
y
p
t
o
g
r
a
p
h
i
c
i
m
p
l
e
me
n
t
a
t
i
o
n
s
u
s
i
n
g
d
e
e
p
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
ser
i
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
0
0
7
6
LN
C
S
,
p
p
.
3
–
2
6
,
2
0
1
6
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
3
1
9
-
4
9
4
4
5
-
6
_
1
.
[
3
2
]
B
.
H
e
t
t
w
e
r
,
S
.
G
e
h
r
e
r
,
a
n
d
T.
G
ü
n
e
y
s
u
,
“
A
p
p
l
i
c
a
t
i
o
n
s
o
f
ma
c
h
i
n
e
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s
i
n
s
i
d
e
-
c
h
a
n
n
e
l
a
t
t
a
c
k
s
:
a
su
r
v
e
y
,
”
J
o
u
r
n
a
l
o
f
C
ry
p
t
o
g
r
a
p
h
i
c
E
n
g
i
n
e
e
ri
n
g
,
v
o
l
.
1
0
,
n
o
.
2
,
p
p
.
1
3
5
–
1
6
2
,
2
0
2
0
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
3
3
8
9
-
019
-
0
0
2
1
2
-
8.
[
3
3
]
L.
M
a
su
r
e
,
C
.
D
u
m
a
s,
a
n
d
E.
P
r
o
u
f
f
,
“
A
c
o
m
p
r
e
h
e
n
si
v
e
st
u
d
y
o
f
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
si
d
e
-
c
h
a
n
n
e
l
a
n
a
l
y
si
s
,
”
I
AC
R
T
ra
n
s
a
c
t
i
o
n
s
o
n
C
r
y
p
t
o
g
r
a
p
h
i
c
H
a
rd
w
a
r
e
a
n
d
Em
b
e
d
d
e
d
S
y
st
e
m
s
,
v
o
l
.
2
0
2
0
,
n
o
.
1
,
p
p
.
3
4
8
–
3
7
5
,
2
0
2
0
,
d
o
i
:
1
0
.
1
3
1
5
4
/
t
c
h
e
s
.
v
2
0
2
0
.
i
1
.
3
4
8
-
3
7
5
.
[
3
4
]
R
.
W
a
t
a
n
a
b
e
,
N
.
G
h
a
f
o
o
r
i
,
a
n
d
A
.
M
i
y
a
j
i
,
“
I
mp
r
o
v
e
d
d
i
f
f
e
r
e
n
t
i
a
l
-
l
i
n
e
a
r
c
r
y
p
t
a
n
a
l
y
s
i
s
o
f
r
e
d
u
c
e
d
r
o
u
n
d
s
o
f
C
h
a
C
h
a
,
”
L
e
c
t
u
r
e
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
ser
i
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
4
4
0
2
LN
C
S
,
p
p
.
2
6
9
–
2
8
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
981
-
99
-
8
0
2
4
-
6
_
2
1
.
[
3
5
]
Y
.
T.
G
o
i
,
S
.
M
.
Le
o
n
g
,
R
.
C
.
W
.
P
h
a
n
,
S
.
La
i
,
a
n
d
A
.
S
a
l
a
g
e
a
n
,
“
U
n
v
e
i
l
i
n
g
t
h
e
b
l
a
c
k
b
o
x
:
n
e
u
r
a
l
c
r
y
p
t
a
n
a
l
y
si
s
w
i
t
h
X
A
I
,
”
C
o
n
f
e
re
n
c
e
Pro
c
e
e
d
i
n
g
s
-
I
EE
E
I
n
t
e
rn
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
S
y
s
t
e
m
s,
M
a
n
a
n
d
C
y
b
e
r
n
e
t
i
c
s
,
p
p
.
1
9
5
1
–
1
9
5
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
S
M
C
5
4
0
9
2
.
2
0
2
4
.
1
0
8
3
1
5
6
4
.
[
3
6
]
Z.
To
l
b
a
,
M
.
D
e
r
d
o
u
r
,
M
.
A
.
F
e
r
r
a
g
,
S
.
M
.
M
u
y
e
e
n
,
a
n
d
M
.
B
e
n
b
o
u
z
i
d
,
“
A
u
t
o
ma
t
e
d
d
e
e
p
l
e
a
r
n
i
n
g
B
LA
C
K
-
B
O
X
a
t
t
a
c
k
f
o
r
m
u
l
t
i
m
e
d
i
a
P
-
B
O
X
S
EC
U
R
I
TY
A
S
S
ESS
M
EN
T,
”
I
EEE
Ac
c
e
ss
,
v
o
l
.
1
0
,
p
p
.
9
4
0
1
9
–
9
4
0
3
9
,
2
0
2
2
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
2
.
3
2
0
4
1
7
5
.
[
3
7
]
N
.
Le
m
n
o
u
a
r
,
“
S
e
c
u
r
i
t
y
l
i
mi
t
a
t
i
o
n
s
o
f
S
h
a
mi
r
’
s
s
e
c
r
e
t
sh
a
r
i
n
g
,
”
J
o
u
r
n
a
l
o
f
D
i
scr
e
t
e
M
a
t
h
e
m
a
t
i
c
a
l
S
c
i
e
n
c
e
s
a
n
d
C
r
y
p
t
o
g
r
a
p
h
y
,
v
o
l
.
2
6
,
n
o
.
4
,
p
p
.
9
7
7
–
9
8
9
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
8
0
/
0
9
7
2
0
5
2
9
.
2
0
2
1
.
1
9
6
1
9
0
2
.
[
3
8
]
K
.
J
a
n
g
e
t
a
l
.
,
“
Q
u
a
n
t
u
m
i
m
p
l
e
me
n
t
a
t
i
o
n
a
n
d
a
n
a
l
y
s
i
s
o
f
S
H
A
-
2
a
n
d
S
H
A
-
3,
”
I
EEE
T
r
a
n
sa
c
t
i
o
n
s
o
n
Em
e
rg
i
n
g
T
o
p
i
c
s
i
n
C
o
m
p
u
t
i
n
g
,
v
o
l
.
1
3
,
n
o
.
3
,
p
p
.
9
1
9
–
9
3
4
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
T
ETC
.
2
0
2
5
.
3
5
4
6
6
4
8
.
[
3
9
]
K
.
G
a
smi
,
S
.
D
i
l
e
k
,
S
.
T
o
s
u
n
,
a
n
d
S
.
O
z
d
e
mi
r
,
“
A
s
u
r
v
e
y
o
n
c
o
m
p
u
t
a
t
i
o
n
o
f
f
l
o
a
d
i
n
g
a
n
d
ser
v
i
c
e
p
l
a
c
e
me
n
t
i
n
f
o
g
c
o
mp
u
t
i
n
g
-
b
a
s
e
d
I
o
T,
”
J
o
u
r
n
a
l
o
f
S
u
p
e
r
c
o
m
p
u
t
i
n
g
,
v
o
l
.
7
8
,
n
o
.
2
,
p
p
.
1
9
8
3
–
2
0
1
4
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
s1
1
2
2
7
-
0
2
1
-
0
3
9
4
1
-
y.
[
4
0
]
T.
S
e
n
e
v
i
r
a
t
h
n
a
,
V
.
H
.
L
a
,
S
.
M
a
r
c
h
a
,
B
.
S
i
n
i
a
r
sk
i
,
M
.
Li
y
a
n
a
g
e
,
a
n
d
S
.
W
a
n
g
,
“
A
s
u
r
v
e
y
o
n
X
A
I
f
o
r
5
G
a
n
d
B
e
y
o
n
d
se
c
u
r
i
t
y
:
t
e
c
h
n
i
c
a
l
a
sp
e
c
t
s,
c
h
a
l
l
e
n
g
e
s
a
n
d
r
e
se
a
r
c
h
d
i
r
e
c
t
i
o
n
s,
”
I
EEE
C
o
m
m
u
n
i
c
a
t
i
o
n
s
S
u
rv
e
y
s
a
n
d
T
u
t
o
r
i
a
l
s
,
v
o
l
.
2
7
,
n
o
.
2
,
p
p
.
9
4
1
–
9
7
3
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
S
T.
2
0
2
4
.
3
4
3
7
2
4
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
d
o
n
esian
J
E
lec
E
n
g
&
C
o
m
p
Sci
I
SS
N:
2502
-
4
7
5
2
Dee
p
lea
r
n
in
g
in
cryp
t
a
n
a
lysi
s
a
co
mp
r
eh
en
s
ive
r
ev
iew
o
f te
ch
n
iq
u
es,
a
p
p
lica
tio
n
s
,
…
(
Ou
s
s
a
ma
N
o
u
i
)
855
[
4
1
]
H
.
M
a
n
t
h
e
n
a
,
S
.
S
h
a
j
a
r
i
a
n
,
J.
C
.
K
i
mm
e
l
l
,
M
.
A
b
d
e
l
s
a
l
a
m,
S
.
K
h
o
r
sa
n
d
r
o
o
,
a
n
d
M
.
G
u
p
t
a
,
“
Ex
p
l
a
i
n
a
b
l
e
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
(
X
A
I
)
f
o
r
ma
l
w
a
r
e
a
n
a
l
y
s
i
s:
a
s
u
r
v
e
y
o
f
t
e
c
h
n
i
q
u
e
s,
a
p
p
l
i
c
a
t
i
o
n
s
,
a
n
d
o
p
e
n
c
h
a
l
l
e
n
g
e
s
,
”
I
EEE
A
c
c
e
ss
,
v
o
l
.
1
3
,
p
p
.
6
1
6
1
1
–
6
1
6
4
0
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
5
.
3
5
5
5
9
2
6
.
[
4
2
]
R
.
La
w
,
S
.
S
.
I
.
L
e
i
,
K
.
Z
h
a
n
g
,
a
n
d
A
.
La
u
,
“
B
r
i
d
g
i
n
g
t
h
e
t
h
e
o
r
y
-
p
r
a
c
t
i
c
e
g
a
p
:
a
c
r
i
t
i
c
a
l
r
e
f
l
e
c
t
i
o
n
o
n
i
n
f
o
r
mat
i
o
n
a
n
d
c
o
mm
u
n
i
c
a
t
i
o
n
t
e
c
h
n
o
l
o
g
y
r
e
sea
r
c
h
,
”
I
n
t
e
r
n
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
C
o
n
t
e
m
p
o
r
a
ry
H
o
s
p
i
t
a
l
i
t
y
M
a
n
a
g
e
m
e
n
t
,
v
o
l
.
3
6
,
n
o
.
6
,
p
p
.
1
9
8
0
–
1
9
9
0
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
8
/
I
JC
H
M
-
02
-
2
0
2
3
-
0
1
3
1
.
[
4
3
]
J.
H
a
o
,
“
D
e
e
p
l
e
a
r
n
i
n
g
-
b
a
s
e
d
me
d
i
c
a
l
i
ma
g
e
a
n
a
l
y
si
s
w
i
t
h
e
x
p
l
a
i
n
a
b
l
e
t
r
a
n
sf
e
r
l
e
a
r
n
i
n
g
,
”
Pr
o
c
e
e
d
i
n
g
s
-
2
0
2
3
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
E
n
g
i
n
e
e
ri
n
g
a
n
d
D
i
st
a
n
c
e
L
e
a
r
n
i
n
g
,
C
ED
L
2
0
2
3
,
p
p
.
1
0
6
–
1
0
9
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
C
ED
L6
0
5
6
0
.
2
0
2
3
.
0
0
0
2
9
.
[
4
4
]
L.
Q
i
a
o
,
Y
.
Z
h
a
n
g
,
Q
.
W
a
n
g
,
D
.
Li
,
a
n
d
S
.
P
e
n
g
,
“
F
a
u
l
t
d
i
a
g
n
o
s
i
s
f
o
r
w
i
n
d
t
u
r
b
i
n
e
g
e
n
e
r
a
t
o
r
s
b
a
se
d
o
n
m
o
d
e
l
-
a
g
n
o
s
t
i
c
m
e
t
a
-
l
e
a
r
n
i
n
g
:
a
f
e
w
-
s
h
o
t
l
e
a
r
n
i
n
g
m
e
t
h
o
d
,
”
Ex
p
e
rt
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
6
7
,
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
sw
a
.
2
0
2
4
.
1
2
6
1
7
1
.
[
4
5
]
S
.
B
e
n
g
e
s
i
,
H
.
E
l
-
S
a
y
e
d
,
M
.
K
.
S
a
r
k
e
r
,
Y
.
H
o
u
k
p
a
t
i
,
J.
I
r
u
n
g
u
,
a
n
d
T.
O
l
a
d
u
n
n
i
,
“
A
d
v
a
n
c
e
me
n
t
s
i
n
g
e
n
e
r
a
t
i
v
e
A
I
:
a
c
o
mp
r
e
h
e
n
s
i
v
e
r
e
v
i
e
w
o
f
G
A
N
s,
G
P
T,
a
u
t
o
e
n
c
o
d
e
r
s
,
d
i
f
f
u
si
o
n
mo
d
e
l
,
a
n
d
t
r
a
n
sf
o
r
m
e
r
s,
”
I
E
EE
Ac
c
e
ss
,
v
o
l
.
1
2
,
p
p
.
6
9
8
1
2
–
6
9
8
3
7
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
A
C
C
ESS
.
2
0
2
4
.
3
3
9
7
7
7
5
.
[
4
6
]
P
.
K
i
s
a
n
g
a
,
I
.
W
o
u
n
g
a
n
g
,
I
.
Tr
a
o
r
e
,
a
n
d
G
.
H
.
S
.
C
a
r
v
a
l
h
o
,
“
N
e
t
w
o
r
k
a
n
o
ma
l
y
d
e
t
e
c
t
i
o
n
u
si
n
g
a
g
r
a
p
h
n
e
u
r
a
l
n
e
t
w
o
r
k
,
”
2
0
2
3
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
i
n
g
,
N
e
t
w
o
rk
i
n
g
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s
,
I
C
N
C
2
0
2
3
,
p
p
.
6
1
–
6
5
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
N
C
5
7
2
2
3
.
2
0
2
3
.
1
0
0
7
4
1
1
1
.
[
4
7
]
D
.
P
e
c
i
o
s
k
i
,
V
.
G
a
v
r
i
l
o
sk
i
,
S
.
D
o
ma
z
e
t
o
v
sk
a
,
a
n
d
A
.
I
g
n
j
a
t
o
v
sk
a
,
“
A
n
o
v
e
r
v
i
e
w
o
f
r
e
i
n
f
o
r
c
e
me
n
t
l
e
a
r
n
i
n
g
t
e
c
h
n
i
q
u
e
s,
”
1
2
t
h
Me
d
i
t
e
rr
a
n
e
a
n
C
o
n
f
e
r
e
n
c
e
o
n
Em
b
e
d
d
e
d
C
o
m
p
u
t
i
n
g
,
ME
C
O
2
0
2
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
0
9
/
M
EC
O
5
8
5
8
4
.
2
0
2
3
.
1
0
1
5
5
0
6
6
.
[
4
8
]
W
.
L
i
,
Y
.
P
e
n
g
,
M
.
Z
h
a
n
g
,
L.
D
i
n
g
,
H
.
H
u
,
a
n
d
L.
S
h
e
n
,
“
D
e
e
p
mo
d
e
l
f
u
s
i
o
n
:
a
s
u
r
v
e
y
,
”
I
EEE
T
ra
n
s
a
c
t
i
o
n
s
o
n
N
e
u
r
a
l
N
e
t
w
o
rk
s
a
n
d
L
e
a
r
n
i
n
g
S
y
s
t
e
ms
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
TN
N
LS
.
2
0
2
5
.
3
6
2
8
6
6
6
.
[
4
9
]
A
.
V
.
S
h
o
s
h
i
n
a
,
G
.
I
.
B
o
r
z
u
n
o
v
,
a
n
d
E.
Y
.
I
v
a
n
o
v
a
,
“
A
p
p
l
i
c
a
t
i
o
n
o
f
b
i
o
-
i
n
sp
i
r
e
d
a
l
g
o
r
i
t
h
ms
t
o
t
h
e
c
r
y
p
t
a
n
a
l
y
s
i
s
o
f
a
s
y
mm
e
t
r
i
c
c
i
p
h
e
r
s o
n
t
h
e
b
a
s
i
s
o
f
c
o
m
p
o
s
i
t
e
n
u
m
b
e
r
,
”
Pr
o
c
e
e
d
i
n
g
s
o
f
t
h
e
2
0
2
1
I
EE
E
C
o
n
f
e
r
e
n
c
e
o
f
R
u
ssi
a
n
Y
o
u
n
g
Re
s
e
a
rc
h
e
rs i
n
E
l
e
c
t
r
i
c
a
l
a
n
d
El
e
c
t
r
o
n
i
c
E
n
g
i
n
e
e
ri
n
g
,
E
l
C
o
n
R
u
s 2
0
2
1
,
p
p
.
2
3
9
9
–
2
4
0
3
,
2
0
2
1
,
d
o
i
:
1
0
.
1
1
0
9
/
E
l
C
o
n
R
u
s
5
1
9
3
8
.
2
0
2
1
.
9
3
9
6
2
4
2
.
[
5
0
]
A
.
S
w
a
r
o
o
p
e
t
a
l
.
,
“
A
c
o
m
p
r
e
h
e
n
si
v
e
o
v
e
r
v
i
e
w
o
f
f
o
r
ma
l
me
t
h
o
d
s
a
n
d
d
e
e
p
l
e
a
r
n
i
n
g
f
o
r
v
e
r
i
f
i
c
a
t
i
o
n
a
n
d
o
p
t
i
mi
z
a
t
i
o
n
,
”
2
0
2
4
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
D
e
c
i
si
o
n
Ai
d
S
c
i
e
n
c
e
s
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s,
D
A
S
A
2
0
2
4
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
0
9
/
D
A
S
A
6
3
6
5
2
.
2
0
2
4
.
1
0
8
3
6
6
5
4
.
[
5
1
]
C
.
U
r
b
a
n
a
n
d
A
.
M
i
n
é
,
“
A
r
e
v
i
e
w
o
f
f
o
r
m
a
l
met
h
o
d
s
a
p
p
l
i
e
d
t
o
mac
h
i
n
e
l
e
a
r
n
i
n
g
,
”
2
0
2
1
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s/
2
1
0
4
.
0
2
4
6
6
.
B
I
O
G
RAP
H
I
E
S
OF
AUTH
O
RS
O
u
ss
a
m
a
No
u
i
is
an
a
ss
o
c
iate
p
r
o
fe
ss
o
r
at
th
e
Un
i
v
e
rsity
of
Ba
tn
a
1,
Al
g
e
ria.
He
h
o
l
d
s
a
Ph
.
D
.
d
e
g
re
e
in
C
o
m
p
u
ter
S
c
ien
c
e
wit
h
s
p
e
c
ializa
ti
o
n
in
Cry
p
t
o
g
ra
p
h
y
a
n
d
S
e
c
u
rit
y
.
His
re
se
a
rc
h
a
re
a
s
a
re
a
u
th
e
n
ti
c
a
ti
o
n
,
ima
g
e
e
n
c
ry
p
ti
o
n
,
a
n
d
wa
term
a
rk
in
g
.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
o
u
ss
a
m
a
.
n
o
u
i@
u
n
i
v
-
b
a
t
n
a
.
d
z
.
Am
in
e
B
a
r
k
a
t
h
a
s
a
P
h
.
D
.
fro
m
P
o
li
tec
n
ico
Di
M
il
a
n
o
with
a
m
a
in
fo
c
u
s
on
Clo
u
d
Co
m
p
u
ti
n
g
sy
ste
m
s,
v
ir
tu
a
li
z
a
ti
o
n
,
a
n
d
m
o
b
i
le
n
e
two
r
k
s
(
4
G
a
n
d
5
G
)
.
Cu
rre
n
tl
y
,
he
is
an
a
ss
o
c
iate
p
ro
fe
ss
o
r
at
th
e
Hig
h
e
r
Na
ti
o
n
a
l
S
c
h
o
o
l
of
Re
n
e
wa
b
le
En
e
rg
ies
,
En
v
ir
o
n
m
e
n
t
a
n
d
S
u
sta
in
a
b
le
De
v
e
l
o
p
m
e
n
t
,
Al
g
e
r
ia
.
With
a
l
o
n
g
b
a
c
k
g
ro
u
n
d
in
v
a
rio
u
s
tec
h
n
o
lo
g
ies
a
n
d
e
x
p
e
rien
c
e
with
u
n
iv
e
rsiti
e
s
a
n
d
c
o
m
p
a
n
ies
in
m
o
re
t
h
a
n
15
Eu
ro
p
e
a
n
c
o
u
n
tr
ies
,
Am
in
e
d
e
li
v
e
rs
train
i
n
g
a
n
d
p
ro
fe
ss
io
n
a
l
se
rv
ice
s
fo
r
c
lo
u
d
-
re
late
d
to
p
ics
,
in
c
l
u
d
i
n
g
d
a
ta
c
e
n
ter
v
irt
u
a
li
z
a
ti
o
n
,
NFV
,
a
n
d
sto
ra
g
e
n
e
two
rk
s
.
Th
e
d
e
li
v
e
re
d
se
rv
ice
s
in
c
lu
d
e
d
iffere
n
t
b
u
si
n
e
ss
lev
e
ls,
m
a
in
ly
p
re
-
sa
les
a
n
d
p
o
st
-
sa
les
.
He
can
be
c
o
n
tac
ted
at
e
m
a
il
:
a
m
in
e
.
b
a
rk
a
t@
p
o
l
imi.i
t.
Evaluation Warning : The document was created with Spire.PDF for Python.