I
ndo
ne
s
i
a
n J
o
ur
na
l
o
f
E
l
e
c
t
r
i
c
a
l
E
ng
i
ne
e
r
i
ng
a
nd C
o
m
p
u
t
e
r
Sc
i
e
nc
e
V
o
l.
11
,
N
o.
2
,
A
ug
us
t
201
8
,
pp.
50
4
~
51
4
I
S
S
N
:
2502
-
4752,
D
O
I
:
10.
11
591/
i
j
e
e
c
s
.
v
1
1
.i
2
.p
p
504
-
5
14
504
Jou
r
n
al
h
om
e
p
age
:
h
ttp
:
//ia
e
s
c
or
e
.
c
om
/
j
our
nal
s
/
i
nde
x
.
php/
i
j
e
e
c
s
M
ult
i
-
Ag
en
t
ba
s
e
d M
a
pReduce
M
o
del f
o
r E
f
f
ici
ent
Ut
ili
z
a
t
io
n
o
f S
y
s
te
m
R
es
o
ur
ces
H
een
a
k
o
u
s
a
r,
B
.
R
.
P
r
as
ad
B
ab
u
D
e
pa
r
t
m
e
nt
of
C
S
E
,
E
a
s
t
P
oi
n
t
C
ol
l
e
g
e
of
E
ng
i
ne
e
r
i
ng
&
T
e
c
hnol
o
gy
B
a
ng
a
l
or
e
,
I
ndi
a
A
rt
i
cl
e I
n
f
o
AB
S
T
RAC
T
A
r
tic
le
h
is
to
r
y
:
R
ecei
v
ed
No
v
21
,
201
7
Re
v
i
se
d
J
an
8
,
201
8
A
ccep
t
ed
J
an
2
7,
2
01
8
R
e
c
e
nt
l
y
w
i
t
h i
nc
r
e
a
s
e
d a
dopt
i
o
n
of
bi
g da
t
a
,
I
nt
e
r
ne
t
of
T
hi
ng
s
a
nd s
e
ns
or
t
e
c
hnol
og
y
b
y
va
r
i
ous
or
g
a
ni
z
a
t
i
on f
or
pr
ov
i
s
i
o
ni
ng
s
m
a
r
t
i
nt
e
l
l
i
g
e
nt
s
e
r
v
i
c
e
s
f
or
v
a
r
i
ous
a
ppl
i
c
a
t
i
on
us
e
s
.
D
a
t
a
pr
oc
e
s
s
i
ng on r
e
a
l
-
tim
e
s
o
c
ia
l
m
ed
i
a an
d
s
en
s
o
r
d
at
a i
s
b
een
a k
e
y
ar
e
a o
f
r
es
e
ar
ch
i
n
r
ecen
t
t
i
m
e
s
an
d
t
h
es
e
da
t
a
a
r
e
m
a
s
s
i
ve
a
nd c
ont
i
n
uo
us
.
S
m
a
r
t
a
ppl
i
c
a
t
i
on
us
i
ng
s
e
ns
or
a
nd s
oc
i
a
l
m
ed
i
a d
at
a c
an
b
e cl
as
s
i
f
i
ed
i
n
t
o
t
h
r
ee cl
as
s
:
1
)
o
n
l
i
n
e p
r
o
c
es
s
i
n
g
o
f
s
tr
e
a
m
in
g
d
a
ta
; 2
)
onl
i
ne
pr
oc
e
s
s
i
ng
of
hi
s
t
or
i
c
a
l
da
t
a
;
a
n
d 3)
hy
br
i
d
pr
oc
e
s
s
i
ng
of
bot
h.
T
he
e
x
i
s
t
i
ng
m
ode
l
a
r
e
de
s
i
g
ne
d c
ons
i
de
r
i
ng
s
t
r
e
a
m
or
ba
t
c
h pr
oc
e
s
s
i
ng
.
F
or
pr
ov
i
s
i
oni
ng
r
e
a
l
-
t
i
m
e
pr
oc
e
s
s
i
ng
M
a
pR
e
duc
e
f
r
a
m
e
w
or
k
us
i
ng
H
a
doop f
r
a
m
e
w
or
k i
s
c
ons
i
de
r
e
d by
s
t
a
t
e
-
of
-
ar
t
t
e
c
hni
que
f
or
da
t
a
i
nf
l
ow
f
or
e
c
a
s
t
i
ng
.
H
ow
e
v
e
r
,
t
he
H
a
doo
p
ba
s
e
d f
or
e
c
a
s
t
i
ng
m
ode
l
ar
e n
o
t
ef
f
i
ci
en
t
i
n
f
u
l
l
y
u
t
i
l
i
zi
n
g
s
y
s
t
e
m
r
es
o
u
r
ce.
A
g
en
t
b
as
ed
M
ap
R
ed
u
ce
f
o
r
eca
s
t
i
n
g
m
o
d
el
i
s
ad
o
p
t
ed
b
y
s
t
at
e
-
of
-
ar
t
t
ech
n
i
q
u
e t
o
u
t
i
l
i
ze s
y
s
t
e
m
ef
f
i
ci
en
t
l
y
.
H
o
w
e
v
er
,
t
he
y
i
nc
ur
s
hi
g
h c
om
put
a
t
i
on ov
e
r
he
a
d,
t
hu
s
i
nc
r
e
a
s
e
c
os
t
of
c
om
put
i
ng
c
os
t
.
T
o
ov
e
r
c
om
e
t
hi
s
w
or
k
pr
e
s
e
nt
a
n a
g
e
nt
b
a
s
e
d D
a
t
a
I
nf
l
ow
F
or
e
c
a
s
t
i
ng
(
D
I
F
)
m
ode
l
f
or
bot
h s
t
r
e
a
m
a
nd non
-
s
tr
e
a
m
(
h
is
to
r
ic
a
l)
d
a
ta
b
y
u
s
in
g
M
u
ltiv
a
r
ia
te
G
a
u
s
s
ia
n
M
ix
tu
r
e
(
M
G
M
)
m
o
d
e
l
.
T
hi
s
w
or
k
pr
e
s
e
nt
a
n A
ge
nt
ba
s
e
d M
a
pR
e
duc
e
(
A
M
R
)
f
r
a
m
e
w
or
k
t
o pr
oc
e
s
s
da
t
a
i
n
r
eal
-
t
i
m
e an
d
u
t
i
l
i
ze s
y
s
t
e
m
r
es
o
u
r
ce ef
f
i
ci
en
t
l
y
.
T
o
p
r
o
v
i
d
e s
cal
ab
i
l
i
t
y
f
o
r
p
r
o
ces
s
i
n
g
s
o
ci
al
m
ed
i
a an
d
s
en
s
o
r
d
at
a D
I
F
-
A
M
R
m
ode
l
a
dopt
s
c
l
o
ud
co
m
p
u
t
i
n
g
ar
ch
i
t
ect
u
r
e.
E
x
p
er
i
m
en
t
ar
e co
n
d
u
ct
ed
t
o
ev
al
u
at
e p
e
r
f
o
r
m
an
c
e
o
f D
I
F
-
A
M
R
of
ov
e
r
e
x
i
s
t
i
ng
m
ode
l
s
how
s
s
i
g
ni
f
i
c
a
nt
pe
r
f
or
m
a
nc
e
im
p
r
o
v
e
m
e
n
t in
te
r
m
s
o
f
c
o
m
p
u
ta
tio
n
tim
e
.
Ke
y
wo
rd
s
:
A
ge
nt
C
l
ou
d c
o
m
put
i
ng
F
o
r
c
a
s
tin
g
H
a
doop
M
ap
R
ed
u
ce
P
a
r
a
lle
l c
o
m
p
u
t
in
g
S
t
r
e
a
m c
o
mp
u
t
i
n
g
C
opy
r
i
g
ht
©
201
8
I
ns
t
i
t
ut
e
o
f
A
d
v
anc
e
d E
ngi
ne
e
r
i
ng
an
d
Sc
i
e
nc
e
.
A
l
l
ri
g
h
t
s re
se
rv
e
d
.
Co
rre
sp
o
n
d
i
n
g
Au
t
h
o
r
:
H
een
a k
o
u
s
ar
,
D
ep
ar
t
m
en
t
o
f
C
S
E
,
E
as
t
P
o
i
n
t
C
o
l
l
eg
e o
f
E
n
g
i
n
eer
i
n
g
&
T
ech
n
o
l
o
g
y
,
B
a
nga
l
o
r
e
,
I
nd
i
a
.
E
m
a
il:
hk
he
e
na
ko
us
a
r
7
3
@
g
m
a
i
l
.
c
o
m
1.
I
NT
RO
D
UCT
I
O
N
W
ith
i
nc
r
e
a
s
i
ng a
d
o
p
t
i
o
n o
f
I
nt
e
r
ne
t
o
f
T
hi
n
gs
(
I
o
T
)
t
o
pr
o
vi
s
i
o
n r
e
a
l
-
t
i
m
e
s
er
v
i
ce
s
f
o
r
v
ar
i
o
u
s
s
m
a
r
t a
p
p
lic
a
tio
n
u
s
e
s
.
S
u
c
h
a
s
s
m
a
r
t tr
a
n
s
p
o
r
ta
tio
n
s
y
s
te
m
,
d
is
a
s
te
r
m
a
n
a
g
e
m
e
n
t s
y
s
te
m
e
tc
.
w
h
e
r
e
s
e
n
s
o
r
a
r
e
de
pl
oy
e
d a
c
r
os
s
t
h
e
gl
obe
t
o pr
ov
i
s
i
on
r
e
a
l
-
t
i
m
e s
er
v
i
c
es
t
o
u
s
er
s
.
D
at
a p
r
o
ces
s
i
n
g
b
as
ed
o
n
r
eal
-
t
i
me
s
en
s
o
r
d
at
a i
s
b
een
a k
e
y
r
es
e
ar
ch
ar
ea o
f
v
ar
i
o
u
s
acad
e
m
i
a
,
i
n
d
u
s
t
r
y
an
d
g
o
v
er
n
m
e
n
t
o
r
g
an
i
zat
i
o
n
i
n
r
ecen
t
ti
m
e
s
.
A
d
a
ta
d
r
iv
e
n
p
a
r
a
d
ig
m
o
f
s
m
a
r
t
i
n
f
o
ta
in
m
e
n
t
a
p
p
lic
a
tio
n
s
,
a
i
m
i
n
g
a
t
m
i
n
i
n
g
th
e
e
s
s
e
n
tia
l
v
a
l
ue
o
f
s
e
ns
or
da
t
a
a
n
d
pr
om
ot
i
ng
s
m
a
r
t
s
e
r
v
i
c
e
s
,
i
s
a
r
i
s
e
n i
n s
m
a
r
t
i
n
t
e
l
l
i
g
e
n
t
t
r
a
n
s
por
t
do
m
a
i
n [
1]
.
H
o
w
e
v
e
r
,
s
e
n
s
or
d
at
a an
d
i
t
s
r
eal
-
t
i
m
e ap
p
l
i
cat
i
o
n
s
er
v
i
ce
s
h
a
v
e s
o
m
e
n
e
w
c
h
a
r
a
c
te
r
is
tic
s
c
o
m
p
a
r
i
n
g
to
tr
a
d
itio
n
a
l s
e
n
s
o
r
d
a
ta
a
n
d
a
p
p
lic
a
tio
n
s
:
1)
M
as
s
i
v
en
e
s
s
: W
it
h
th
e
g
r
o
w
th
o
f
s
m
a
r
t in
f
o
ta
in
m
e
n
t a
p
p
lic
a
tio
n
s
e
r
v
ic
e
s
,
la
r
g
e
a
m
o
u
n
t o
f
d
a
ta
is
g
en
er
at
ed
b
y
d
i
f
f
er
en
t
s
e
n
s
o
r
p
er
i
o
d
i
cal
l
y
.
F
o
r
e
x
a
m
p
l
e,
l
o
cat
i
o
n
an
d
t
r
af
f
i
c
m
a
n
ag
e
m
en
t
s
er
v
i
ces
g
en
er
at
e
s
s
e
n
s
o
r
d
at
a at
v
er
y
h
i
g
h
r
at
e,
w
h
i
c
h
l
ead
t
o
s
i
ze o
f
p
et
ab
y
te
o
f
d
a
ta
in
a
s
h
o
r
t te
r
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
I
SSN
:
2502
-
4752
M
u
lti
-
A
ge
nt
bas
e
d M
ap
R
e
duc
e
M
ode
l
f
or
E
f
f
i
c
i
e
nt
U
t
i
l
i
z
at
i
on of
S
y
s
t
e
m
…
(
H
e
e
na k
ous
ar
)
505
2)
S
t
r
e
a
mi
n
g
:
S
e
n
s
o
r
d
at
a
f
r
o
m
d
i
f
f
er
en
t
s
o
u
r
ces
s
u
ch
as
u
s
er
w
ear
ab
l
e a
n
d
i
t
s
l
o
cat
i
o
n
an
d
m
o
b
i
l
i
t
y
i
n
f
o
r
m
at
i
o
n
ar
r
i
v
es
i
n
t
o
t
h
e
s
er
v
er
co
n
t
i
n
u
o
u
s
l
y
f
r
o
m
v
ar
i
o
u
s
s
o
u
r
ces
as
a
s
t
r
ea
m
s
at
r
ap
i
d
r
at
e.
E
f
f
i
ci
en
t
m
ec
h
a
n
i
s
m
i
s
n
eed
ed
t
o
p
r
o
ces
s
/
p
er
f
o
r
m
a
n
al
y
s
i
s
o
n
s
u
ch
s
t
r
ea
m
d
at
a i
n
r
eal
-
t
i
m
e
w
i
t
h
out
a
ny
in
te
r
r
u
p
tio
n
.
3)
R
eal
-
T
i
me
:
I
n
m
a
n
y
a
p
p
lic
a
tio
n
s
e
r
v
ic
e
s
s
u
c
h
a
s
p
o
ll
u
tio
n
m
o
n
ito
r
i
n
g
,
li
v
e
v
e
h
ic
le
tr
a
f
f
ic
m
o
n
ito
r
i
n
g
co
m
p
u
t
i
n
g
,
w
e n
eed
r
eact
q
u
i
ck
l
y
t
o
s
u
c
h
l
ar
g
e s
e
n
s
o
r
d
at
a o
n
ce t
h
e
y
ar
e g
e
n
er
at
e
d
.
S
i
n
ce t
h
es
e
i
n
f
o
r
m
at
i
o
n
b
eco
m
e
s
o
b
s
o
l
et
e/
o
u
t
d
at
ed
q
u
i
ck
l
y
.
P
er
f
o
r
m
i
n
g
an
al
y
s
i
s
o
n
s
u
c
h
l
ar
g
e
d
at
a
i
n
n
ear
r
eal
-
t
i
m
e r
eq
u
i
r
es
an
e
f
f
i
ci
en
t
m
ec
h
an
i
s
m
t
o
co
l
l
ect
i
n
p
u
t
s
t
r
ea
m
,
o
r
g
an
i
ze t
h
e d
at
a an
d
p
er
f
o
r
m
co
m
p
u
t
i
n
g
an
d
d
at
a an
al
y
s
i
s
.
4)
E
ve
r
-
E
x
pa
n
di
ng
:
A
s
y
ear
s
p
as
s
es
n
e
w
t
y
p
e
a
n
d
n
u
m
b
er
o
f
s
en
s
o
r
s
a
v
ai
l
ab
l
e i
n
m
ar
k
et
i
n
cr
eas
e
s
.
A
s
cal
ab
l
e r
eal
-
t
i
m
e p
r
o
ces
s
i
n
g
d
es
i
g
n
i
s
r
eq
u
i
r
ed
t
o
h
an
d
l
e s
u
ch
e
v
er
i
n
cr
eas
i
n
g
s
e
n
s
o
r
d
at
a v
o
l
u
m
e.
T
ha
t
i
s
,
t
he
f
ut
ur
e
r
e
a
l
-
t
i
m
e
pr
e
di
c
t
i
on
m
ode
l
s
h
o
u
l
d
n
ot
onl
y
pr
oc
es
s
l
ar
g
e
v
o
l
u
m
e o
f
s
e
n
s
o
r
d
at
a a
n
d
g
u
ar
a
n
t
ee r
eal
-
t
i
m
e
q
u
a
l
i
t
y
b
y
a
ddi
n
g
c
o
m
put
i
ng
n
ode
s
,
bu
t
a
l
s
o n
e
w
t
y
pe
s
of
s
e
n
s
or
da
t
a
e
a
s
i
l
y
.
A
d
d
i
t
i
o
n
al
l
y
,
t
h
er
e
ar
e
m
a
n
y
ap
p
l
i
cat
i
o
n
b
as
ed
o
n
s
en
s
o
r
d
at
a
p
r
o
ces
s
i
n
g
w
h
i
c
h
ca
n
b
e
cl
as
s
i
f
i
ed
i
n
t
o
t
h
r
ee cl
as
s
e
s
b
as
ed
o
n
h
o
w
t
h
e s
e
n
s
o
r
d
at
a
ar
e
p
r
o
ces
s
ed
as
s
h
o
w
n
i
n
T
ab
l
e
1
.
O
nl
i
ne
p
r
o
c
e
s
s
i
n
g o
f
s
tr
e
a
m
i
n
g
d
a
ta
,
o
f
f
lin
e
p
r
o
c
e
s
s
in
g
o
f
h
i
s
to
r
ic
a
l/s
ta
t
is
tic
a
l
d
a
ta
a
n
d
h
y
b
r
id
p
r
o
c
e
s
s
in
g
o
f
b
o
th
h
i
s
to
r
ic
a
l a
n
d
s
tr
e
a
m
i
n
g
d
a
ta
.
T
o
c
o
p
e
w
it
h
d
if
f
e
r
e
n
t a
p
p
lic
a
tio
n
s
in
s
m
a
r
t
in
f
o
t
a
i
n
m
e
nt
d
o
m
a
i
n,
t
he
p
r
o
c
e
s
s
i
ng
s
ys
t
e
m
ne
e
d
t
o
d
eal
w
i
t
h
b
o
t
h
m
as
s
i
v
el
y
l
a
r
g
e h
i
s
t
o
r
i
cal
d
at
a an
d
s
t
r
ea
m
i
n
g
d
at
a.
T
h
e h
i
s
t
o
r
i
cal
s
e
n
s
o
r
d
at
a i
s
co
n
s
i
d
er
ed
t
o
b
e m
o
r
e v
al
u
ab
l
e an
d
h
e
n
c
e n
eed
t
o
b
e m
i
x
ed
w
i
t
h
s
t
r
e
a
m
d
at
a i
n
m
an
y
cas
e
s
f
o
r
p
er
f
o
r
m
i
n
g
s
t
at
i
s
t
i
cal
c
o
m
p
u
ta
tio
n
,
p
a
tte
r
n
d
is
c
o
v
e
r
y
a
n
d
d
a
ta
in
f
lo
w
p
r
e
d
ic
tio
n
.
T
ab
l
e
1
.
S
en
s
o
r
D
at
a A
p
p
l
i
cat
i
o
n
T
y
p
es
T
y
pe
A
ppl
i
c
a
t
i
o
n
C
h
a
r
a
ct
e
ri
s
t
i
c
O
n
l
i
n
e
p
r
o
c
e
s
s
i
n
g
o
f
r
e
a
l
-
tim
e
s
tr
e
a
m
in
g
s
en
s
o
r
d
at
a
F
a
k
e l
i
c
en
s
e p
l
a
t
e d
et
ect
i
o
n
,
b
l
a
ck
l
i
s
t
e
d
v
eh
i
c
l
e/
u
s
er
s
S
i
n
g
l
e
r
o
w
q
u
e
r
i
e
s
f
o
r
m
s
e
n
s
o
r
d
a
t
a
s
t
r
ea
m
R
ea
l
-
tim
e
m
o
n
i
t
o
r
in
g
o
f
tr
a
f
f
ic
a
n
d
t
r
a
v
e
l
tim
e
p
r
e
d
ic
tio
n
M
u
l
t
i
p
l
e
r
o
w
s
a
g
g
r
e
g
a
t
i
o
n
f
r
om
w
i
n
d
o
w
e
d
d
a
t
a
o
n
s
t
r
e
a
m
O
f
f
l
in
e
p
r
o
c
e
s
s
in
g
o
f
h
is
to
r
i
c
a
l
/s
ta
ti
s
t
ic
a
l
s
en
s
o
r
d
at
a
M
i
n
i
n
g
v
e
h
i
c
l
e
a
n
d
u
s
e
r
d
at
a
A
g
g
r
e
ga
t
i
on
on
b
a
t
c
h
d
a
t
a
H
y
b
r
i
d
p
r
o
c
e
s
s
i
n
g
o
f
b
o
t
h
h
i
s
t
o
r
i
c
a
l
a
n
d
st
r
e
a
m
se
n
so
r
d
a
t
a
O
v
e
r
s
p
e
e
d
i
n
g
v
e
h
i
c
l
e
d
i
s
c
o
v
e
r
y
,
C
l
o
n
e
d
l
i
c
en
s
e
p
l
a
t
e d
et
ec
t
i
o
n
S
i
n
g
l
e
s
t
r
e
a
m
i
n
g
r
o
w
s
c
o
m
p
a
r
i
s
o
n
f
r
o
m
st
a
t
i
st
i
c
a
l
se
n
so
r
d
a
t
a
I
n
o
r
d
er
cat
er
p
er
f
o
r
m
a
n
ce r
eq
u
i
r
e
m
e
n
t o
f
s
m
a
r
t i
n
f
o
ta
i
n
m
e
n
t s
y
s
te
m
a
p
p
lic
a
tio
n
s
,
h
o
w
to
p
r
o
v
id
e
s
cal
ab
l
e an
d
r
eal
-
t
i
m
e
pr
oc
e
s
s
i
ng
s
u
ppor
t
f
or
m
a
s
s
i
v
e
l
y
l
a
r
g
e
a
n
d
c
on
t
i
nu
ous
s
e
n
s
or
da
t
a
i
s
be
c
o
m
i
ng
on
e
s
s
e
n
tia
l
is
s
u
e
in
s
m
a
r
t i
n
f
o
ta
in
m
e
n
t
s
y
s
te
m
.
N
o
v
e
l
m
e
th
o
d
o
lo
g
ie
s
a
r
e
r
e
q
u
ir
e
d
to
s
u
p
p
o
r
t
a
bov
e
m
e
n
t
i
on
e
d
a
p
p
l
i
c
a
t
i
o
ns
a
nd
t
o
o
ve
r
c
o
m
e
und
e
r
l
yi
n
g c
ha
l
l
e
nge
s
.
C
h
al
l
en
g
es
i
n
d
es
i
g
n
i
n
g
s
en
s
o
r
d
at
a p
r
o
ces
s
i
n
g
s
y
s
t
e
m
:
T
o
co
p
e w
i
t
h
ab
o
v
e
m
e
n
t
i
o
n
ed
f
e
at
u
r
es
a
n
d
a
ppl
i
c
a
t
i
on
s
of
s
e
n
s
or
d
a
ta
,
th
e
r
e
e
x
is
t c
h
a
lle
n
g
e
s
in
d
e
s
i
g
n
i
n
g
a s
en
s
o
r
d
at
a p
r
o
ces
s
i
n
g
s
ys
t
e
m
t
ha
t
c
a
n
p
r
o
ces
s
m
as
s
i
v
e
l
y
l
ar
g
e s
e
n
s
o
r
d
at
a i
n
n
ear
r
eal
-
t
i
m
e an
d
h
a
n
d
l
e d
i
f
f
er
en
t
t
y
p
e o
f
s
e
n
s
o
r
d
at
a ap
p
l
i
cat
i
o
n
.
T
h
e
ch
al
l
e
n
g
e
s
ar
e s
u
m
m
ar
i
zed
:
.
1.
F
i
r
s
t
l
y
,
m
a
n
ag
e
m
e
n
t
o
f
b
o
t
h
h
i
s
t
o
r
i
cal
an
d
s
t
r
ea
m
s
e
n
s
o
r
d
at
a,
es
p
eci
al
l
y
co
n
s
i
d
er
i
n
g
t
h
e m
as
s
i
v
e v
o
l
u
me
o
f
h
i
s
t
o
r
i
cal
d
at
a an
d
co
n
t
i
n
u
o
u
s
l
y
ar
r
i
v
i
n
g
s
t
r
ea
m
d
at
a.
T
h
e s
en
s
o
r
d
at
a i
s
co
m
p
o
s
ed
o
f
d
i
f
f
er
en
t
di
m
e
ns
i
ons
a
n
d
n
e
e
ds
t
o be
t
r
a
n
s
f
e
r
r
e
d t
o di
f
f
e
r
e
nt
pr
oc
e
s
s
i
n
g t
a
s
k of
s
m
a
r
t
i
nf
ot
a
i
nm
e
n
t
s
y
s
t
e
m
.
S
o
m
e
d
at
a m
a
y
b
e s
h
ar
ed
acr
o
s
s
d
i
f
f
er
en
t
ap
p
l
i
cat
i
o
n
s
.
A
s
a r
es
u
l
t
l
ead
t
o
co
m
p
l
ex
i
t
y
i
n
m
an
a
g
i
n
g
s
e
n
s
o
r
d
at
a.
2.
S
e
c
on
dl
y
,
bui
l
di
ng
of
a
n
i
n
t
e
g
r
a
t
e
d a
r
c
h
i
t
e
c
t
u
r
e
t
o s
u
ppor
t
on
l
i
n
e
pr
oc
e
s
s
i
ng
of
s
t
r
e
a
m
da
t
a
,
of
f
l
i
ne
pr
oc
e
s
s
i
n
g of
h
i
s
t
or
i
c
a
l
da
t
a
,
a
n
d hy
br
i
d pr
oc
e
s
s
i
ng
of
bot
h s
t
r
e
a
m
a
n
d
h
i
s
t
or
i
c
a
l
da
t
a
.
S
i
n
c
e
s
t
r
ea
m
d
at
a
i
s
co
n
t
i
n
u
o
u
s
a
n
d
h
i
s
t
o
r
i
cal
d
at
a i
s
m
as
s
i
v
e,
t
h
e
y
n
eed
t
o
b
e s
eg
m
en
t
ed
an
d
f
u
s
ed
f
o
r
p
r
o
ces
s
i
n
g
i
n
d
is
tr
ib
u
te
d
a
r
c
h
i
te
c
tu
r
e
in
a
c
o
o
r
d
in
a
te
d
m
a
n
n
e
r
.
F
o
r
e
x
a
m
p
le
,
to
p
e
r
f
o
r
m
c
o
m
p
u
ta
tio
n
o
f
c
lo
n
e
p
la
te
v
e
h
ic
le
id
e
n
ti
f
ic
a
tio
n
,
m
illio
n
s
o
f
h
i
s
to
r
ic
a
l
r
e
c
or
ds
a
n
d
s
t
r
e
a
m
da
t
a
i
n
s
pe
e
d of
10,
000
r
e
c
or
ds
pe
r
s
eco
n
d
s
n
eed
s
t
o
b
e ex
ec
u
t
ed
ev
er
y
s
eco
n
d
.
3.
T
h
i
r
dl
y
,
t
h
e
r
e
i
s
i
n
c
r
e
a
s
i
n
g
i
nn
ov
a
t
i
v
e
s
m
a
r
t
a
ppl
i
c
a
t
i
on
t
h
a
t
de
pe
n
d on
pr
oc
e
s
s
i
ng
of
s
e
ns
or
da
t
a
.
S
om
e
o
f
t
he
s
e
m
a
y b
e
s
ho
r
t
-
te
r
m
a
p
p
lic
a
tio
n
s
.
A
s
a
r
e
s
u
lt
,
ef
f
ect
i
v
e an
d
s
i
m
p
l
i
f
i
ed
d
at
a i
n
f
l
o
w
p
at
t
er
n
o
f
ap
p
l
i
cat
i
o
n
an
d
s
en
s
o
r
d
at
a p
r
o
ces
s
i
n
g
s
y
s
t
e
m
ar
e n
eed
ed
.
T
he
s
i
gni
f
i
c
a
nt
gr
o
w
t
h o
f
B
i
gD
a
t
a
a
nd
c
l
o
ud
t
e
c
hno
l
o
gi
e
s
ha
s
i
nc
ur
r
e
d
s
i
g
ni
f
i
c
a
nt
c
h
a
l
l
e
n
ge
s
i
n
n
et
w
o
r
k
ar
c
h
i
t
ect
u
r
e i
n
d
at
ac
en
t
er
t
o
p
r
o
ces
s
s
en
s
o
r
d
at
a
i
n
r
eal
-
t
i
m
e
.
I
n [
2]
a
dopt
e
d t
r
a
f
f
i
c
e
ng
i
n
e
e
r
i
ng
t
ech
n
i
q
u
e t
o
ad
d
r
es
s
t
h
e b
a
n
d
w
i
d
t
h
r
eq
u
i
r
e
m
en
t
o
f
d
at
a
cen
t
er
n
et
w
o
r
k
.
I
n
[
3
]
p
r
es
e
n
t
ed
a d
at
a i
n
f
l
o
w
pr
e
di
c
t
i
on
m
ode
l
by
a
dopt
i
ng
r
a
t
e
c
on
t
r
ol
m
e
c
h
a
n
i
s
m
b
a
s
e
d on
pr
e
di
c
t
e
d t
r
a
f
f
i
c
i
n n
e
t
w
or
k
.
R
e
c
e
n
t
l
y
r
es
ear
ch
er
h
av
e co
n
s
i
d
er
ed
t
i
g
h
t
i
n
t
e
g
r
at
i
o
n
o
f
ap
p
l
i
cat
i
o
n
an
d
n
et
w
o
r
k
l
a
y
er
f
o
r
o
p
t
i
m
i
zi
n
g
n
et
w
o
r
k
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
25
02
-
4752
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
,
V
o
l.
11
, N
o
.
2
,
A
ug
us
t
2018
:
5
04
–
514
506
r
o
u
tin
g
la
y
e
r
f
o
r
p
r
e
d
ic
tin
g
a
p
p
lic
a
tio
n
tr
a
f
f
ic
[
4
]
.
A
ll t
h
e
s
e
m
o
d
e
l
r
e
q
u
ir
e
s
p
r
io
r
u
n
d
e
r
s
ta
n
d
in
g
o
f
a
p
p
lic
a
tio
n
t
r
af
f
i
c i
n
d
at
a cen
t
er
n
et
w
o
r
k
s
(
i
.
e.
t
h
e ab
i
l
i
t
y
t
o
f
o
r
ecas
t
d
at
a i
n
f
l
o
w
b
ef
o
r
e p
ack
et
ar
r
i
v
es
i
n
n
et
w
o
r
k
)
.
H
o
w
e
v
e
r
,
it is
d
if
f
ic
u
lt to
p
r
e
d
ic
t d
a
ta
in
f
lo
w
o
f
a
p
p
lic
a
tio
n
a
c
c
u
r
a
te
.
A
l
l
e
x
i
s
t
i
n
g
m
e
t
h
odol
ogy
[
5]
f
oc
us
e
d on
pr
e
di
c
t
i
n
g
da
t
a
i
nf
l
o
w
ba
s
e
d on
n
e
t
w
or
k
l
e
v
e
l
p
a
r
a
m
e
t
e
r
s
us
i
n
g he
ur
i
s
t
i
c al
g
o
r
i
t
h
m
.
F
o
r
ex
a
m
p
l
e,
[
5
]
co
m
p
u
t
es
d
at
a i
n
f
l
o
w
u
s
i
n
g
f
l
o
w
co
u
n
t
er
m
eas
u
r
e
m
e
n
t
o
n
s
w
i
t
c
h
es
an
d
[
6
]
u
s
ed
s
o
ck
e
t
b
u
f
f
er
o
ccu
p
an
c
y
at
eac
h
n
o
d
es
t
o
co
m
p
u
t
e d
at
a i
n
f
l
o
w
f
o
r
each
n
o
d
es
.
H
o
w
e
v
e
r
,
t
h
e
s
e
m
e
t
h
odol
og
i
e
s
h
a
v
e
n
um
be
r
of
dr
a
w
ba
c
ks
.
F
i
r
s
t
l
y
,
m
o
s
t
of
t
h
e
m
ca
n
n
o
t
p
r
ed
i
ct
t
h
e d
at
a
i
n
f
l
o
w
d
e
m
an
d
b
ef
o
r
e d
at
a en
t
er
t
h
e n
et
w
o
r
k
.
S
eco
n
d
l
y
,
p
er
f
o
r
m
a
n
ce r
eq
u
i
r
e
m
e
n
t
i
d
en
t
i
f
i
ed
o
n
n
et
w
o
r
k
p
at
h
can
n
o
t
acc
u
r
at
el
y
r
ef
l
ect
t
h
e
act
u
al
d
e
m
an
d
o
f
ap
p
l
i
cat
i
o
n
d
u
e
t
o
co
n
g
e
s
t
i
o
n
co
n
t
r
o
l
at
en
d
n
o
d
es
an
d
p
r
es
en
ce o
f
n
o
i
s
e
of
ba
c
kg
r
oun
d f
l
o
w
.
T
h
i
r
dl
y
,
t
h
e
y
f
a
i
l
t
o i
de
n
t
i
f
y
pr
i
or
i
t
y
i
nf
or
m
a
t
i
on
a
n
d f
i
n
e
-
g
r
ai
n
ed
d
at
a
i
nf
l
o
w
de
pe
n
de
n
c
i
e
s
i
m
pos
e
d
by
a
ppl
i
c
a
t
i
ons
.
A
s
a
r
e
s
u
l
t
,
n
e
t
w
or
k
-
l
a
y
er
b
as
ed
ap
p
r
o
ach
es
ar
e s
h
o
w
n
t
o
b
e
pe
r
f
or
m
i
ng
poor
l
y
i
n
pr
e
di
c
t
i
ng
t
h
e
r
e
a
l
-
ti
m
e
a
p
p
lic
a
tio
n
d
e
m
an
d
[
7
]
,
t
h
u
s
af
f
ect
i
n
g
s
y
s
t
e
m
p
er
f
o
r
m
a
n
ce.
A
g
e
n
t
b
as
ed
f
o
r
ecas
t
i
n
g
m
o
d
el
ai
d
i
n
ach
i
ev
i
n
g
f
i
n
e
-
g
r
ai
n
ed
p
er
f
o
r
m
an
ce
f
o
r
p
er
f
o
r
m
i
n
g
r
eal
-
ti
m
e
an
al
y
s
i
s
o
n
s
o
ci
al
m
ed
i
a an
d
s
en
s
o
r
d
at
a ad
o
p
t
i
n
g
Map
R
ed
u
ce f
r
a
m
e
w
o
r
k
u
s
i
n
g
cl
o
u
d
a
r
ch
i
t
ect
u
r
e [
8
]
,
[
9
]
a
nd
[
1
0]
.
H
a
doop f
r
a
m
e
w
or
k
[
11]
i
s
a
w
i
de
l
y
u
s
e
d M
a
pR
e
du
c
e
f
r
a
m
e
w
or
k
t
h
a
t
a
do
pt
s
c
l
ou
d pl
a
t
f
or
m
.
M
ap
R
ed
u
ce f
r
a
m
e
w
o
r
k
i
s
co
m
p
o
s
ed
o
f
t
w
o
s
t
ag
e.
I
n
t
h
e i
n
i
t
i
al
s
t
a
g
e,
i
n
p
u
t
d
at
a t
o
b
e p
r
o
ces
s
ed
i
s
f
r
ag
m
e
n
t
ed
i
n
t
o
s
e
g
m
e
n
t
s
.
E
ach
s
eg
m
e
n
t
i
s
a
s
s
o
ci
at
ed
w
i
t
h
a
m
ap
co
m
pu
t
i
ng
w
or
k
e
r
t
h
a
t
pr
ov
i
de
s
⟨
K
e
y |
V
a
l
ue
⟩
p
ai
r
s
as
o
u
t
p
u
t
s
.
T
h
e o
u
t
p
u
t
s
o
b
t
ai
n
ed
ar
e s
o
r
t
ed
o
n
t
h
e b
as
i
s
o
f
t
h
e K
e
y
v
al
u
es
as
s
o
ci
at
ed
.
T
h
e s
o
r
t
ed
va
l
ue
s
a
r
e
t
he
i
np
ut
t
o
r
e
d
uc
e
c
o
m
p
ut
i
ng
w
o
r
ke
r
s
,
t
ha
t
i
s
,
⟨
K
e
y
|
S
or
t
e
d
L
i
s
t
(
V
al
u
e)
⟩
.
R
ed
u
ce
co
m
p
u
t
i
n
g
w
o
r
k
er
s
k
eep
s
t
h
e o
u
t
co
m
e i
n
H
ad
o
op
di
s
t
r
i
bu
t
e
d f
i
l
e
s
y
s
t
e
m
(
H
D
F
S
)
.
T
h
e
m
a
p a
n
d r
e
du
c
e
c
om
pu
t
i
n
g
w
o
r
k
er
s
ar
e
g
en
er
al
l
y
v
i
r
t
u
al
m
ach
i
n
e
s
(
V
Ms
)
i
n
p
u
b
l
i
c
cl
o
u
d
en
v
i
r
o
n
m
e
n
t
s
.
A
s
i
m
p
l
e
M
ap
R
ed
u
ce
m
o
d
el
d
ep
l
o
y
ed
o
n
t
h
e V
M
b
as
ed
co
m
p
u
t
i
n
g e
n
vi
r
o
n
m
e
nt
i
s
s
ho
w
n i
n
F
i
g
ur
e
1.
Fi
g
ur
e
1
.
A
r
c
h
ite
c
t
u
r
e
of
M
a
p
R
e
du
c
e
f
r
a
m
e
w
or
k
de
pl
o
y
e
d on
pu
bl
i
c
c
l
ou
d pl
a
t
f
or
m
I
n
[
8
]
p
r
es
en
t
ed
ag
en
t
b
as
ed
f
o
r
ecas
t
i
n
g
m
o
d
el
t
o
p
r
ed
i
ct
a
i
r
p
o
l
l
u
t
i
o
n
l
e
v
el
acr
o
s
s
d
i
f
f
er
en
t
r
eg
i
o
n
.
T
o
im
p
r
o
v
e
a
c
c
u
r
a
c
y
o
f
p
r
e
d
ic
tio
n
t
h
e
ir
m
o
d
e
l
a
d
o
p
te
d
a
r
tif
ic
ia
l
i
n
te
lli
g
e
n
c
e
a
n
d
to
f
o
r
ecas
t
s
h
o
r
t
es
t
p
at
h
D
i
j
k
s
t
r
a al
g
o
r
i
t
h
m
.
T
h
e p
er
f
o
r
m
an
a
l
y
s
i
s
o
n
r
eal
-
t
i
m
e H
ad
o
o
p
M
ap
R
ed
u
ce f
r
a
m
e
w
o
r
k
i
s
ad
o
p
t
ed
.
H
o
w
ev
er
,
t
h
e
y
c
o
n
s
i
de
r
e
d pe
r
f
or
m
a
n
c
e
e
v
a
l
u
a
t
i
on
onl
y
f
or
of
f
l
i
n
e
da
t
a
.
I
n
[
9]
pr
e
s
e
n
t
e
d a
hy
br
i
d
m
ode
l
t
o pr
oc
e
s
s
bot
h
s
t
r
e
a
m
a
nd
no
n
-
st
e
a
m
s
e
n
s
o
r
d
at
a b
y
ad
o
p
t
i
n
g
M
ap
R
ed
u
ce f
r
a
m
e
w
o
r
k
.
T
h
ei
r
m
o
d
el
ach
i
ev
e
s
s
cal
ab
l
e
pe
r
f
or
m
a
n
c
e
ov
e
r
t
r
a
di
t
i
on
ope
n
s
ou
r
c
e
s
t
r
e
a
m
i
n
g
m
ode
l
S
t
or
m
[
12]
.
H
o
w
e
v
e
r
,
t
h
e
i
r
m
ode
l
i
n
c
u
r
s
c
o
m
pu
t
i
ng
o
v
er
h
ead
.
A
s
a r
es
u
l
t
i
n
cu
r
h
i
g
h
co
s
t
o
f
p
r
o
ces
s
i
n
g
.
T
o
m
i
n
i
m
i
ze co
s
t
[
1
0
]
p
r
es
en
t
ed
an
ag
e
n
t
b
as
ed
f
or
e
c
a
s
t
i
n
g
m
ode
l
on
t
op of
H
a
doop.
T
h
e
m
ode
l
i
n
c
u
r
s
s
l
i
g
ht
ov
e
r
h
e
a
d f
or
a
ppl
i
c
a
t
i
on
bu
t
a
t
t
h
e
r
e
du
c
t
i
on
o
f
c
o
m
p
ut
i
ng t
i
m
e
.
T
hus
,
r
e
d
uc
i
ng c
o
s
t
o
f
c
o
m
p
ut
i
n
g.
H
o
w
e
ve
r
,
r
e
s
ul
t
p
r
e
s
e
nt
e
d
i
n [
1
3]
s
ho
w
s
a
d
o
p
t
i
n
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
I
SSN
:
2502
-
4752
M
u
lti
-
A
ge
nt
bas
e
d M
ap
R
e
duc
e
M
ode
l
f
or
E
f
f
i
c
i
e
nt
U
t
i
l
i
z
at
i
on of
S
y
s
t
e
m
…
(
H
e
e
na k
ous
ar
)
507
H
a
doop M
a
pR
e
du
c
e
f
or
pr
oc
e
s
s
i
ng
i
nc
ur
s
c
o
m
p
ut
i
n
g o
ve
r
he
a
d
.
S
i
nc
e
,
i
t
d
o
e
s
no
t
f
ul
l
y
ut
i
l
i
z
e
t
he
r
e
s
o
ur
c
e
av
ai
l
ab
l
e.
T
o
co
p
e w
i
t
h
r
es
ear
ch
c
h
al
l
en
g
e
s
i
n
d
es
i
g
n
i
n
g
ef
f
i
ci
e
n
t
r
eal
-
t
i
m
e s
o
ci
al
m
ed
i
a an
d
s
en
s
o
r
d
at
a
pr
oc
e
s
s
i
n
g a
n
d D
a
t
a
I
nf
l
o
w
F
or
e
c
a
s
t
i
ng
(
D
I
F
)
m
ode
l
,
t
h
i
s
w
or
k
pr
e
s
e
nt
an
ag
e
n
t
b
as
ed
d
at
a i
n
f
l
o
w
f
or
e
c
a
s
t
i
n
g
m
ode
l
f
or
bot
h
s
t
r
e
a
m
a
n
d n
o
n
-
s
t
r
e
a
m
(
hi
s
t
o
r
i
c
a
l
)
d
a
t
a
b
y u
s
i
ng M
ul
t
i
va
r
i
a
t
e
G
a
us
s
i
a
n M
i
xt
ur
e
(
M
G
M
)
m
o
d
el
.
T
o
p
er
f
o
r
m
an
al
y
s
i
s
o
n
m
a
s
s
i
v
el
y
l
ar
g
e s
o
ci
al
m
ed
i
a a
n
d
s
e
n
s
o
r
d
at
a,
t
h
i
s
w
o
r
k
ad
o
p
t
p
ar
al
l
el
c
o
mp
u
t
i
n
g
f
r
a
me
w
or
k
.
T
h
i
s
w
or
k pr
e
s
e
n
t
a
n
A
g
e
n
t
ba
s
e
d
M
a
pR
e
du
c
e
(
A
M
R
)
f
r
a
m
e
w
or
k
t
o pr
oc
e
s
s
da
t
a
i
n
r
eal
-
t
i
m
e an
d
u
t
i
l
i
ze
s
y
s
t
e
m
r
es
o
u
r
ce ef
f
i
ci
e
n
t
l
y
.
T
o
p
r
o
v
i
d
e s
cal
ab
i
l
i
t
y
f
o
r
p
r
o
ces
s
i
n
g
s
o
ci
al
m
ed
i
a a
n
d
s
en
s
o
r
d
at
a A
M
R
m
o
d
el
ad
o
p
t
s
cl
o
u
d
co
m
p
u
t
i
n
g
ar
ch
i
t
ect
u
r
e.
T
he
C
o
n
tr
ib
u
tio
n
o
f
r
e
s
e
a
r
c
h
w
o
r
k
i
s
a
s
f
o
llo
w
s
:
a)
A
n
o
v
el
A
g
e
n
t
b
as
ed
m
o
d
el
M
ap
R
ed
u
ce f
r
a
m
e
w
o
r
k
i
s
p
r
es
en
t
ed
f
o
r
d
at
a i
n
f
l
o
w
f
o
r
ecas
t
i
n
g
.
b)
T
h
e
pr
o
pos
e
d D
I
F
-
A
M
R
s
u
ppor
t
pr
oc
e
s
s
i
n
g
of
s
t
r
e
a
m
,
n
on
-
s
t
r
ea
m
d
at
a an
d
co
m
b
i
n
at
i
o
n
o
f
b
o
t
h
i
n
r
eal
-
t
i
me
.
c)
T
he
D
I
F
-
A
M
R
pr
ov
i
de
s
s
ca
l
a
b
l
e p
r
o
ces
s
i
n
g
b
y
ad
o
p
t
i
n
g
cl
o
u
d
co
m
p
u
t
i
n
g
ar
ch
i
t
ect
u
r
e.
d)
DI
F
-
A
M
R
u
ti
liz
e
s
y
s
te
m
r
e
s
o
u
r
c
e
e
f
f
ic
ie
n
tl
y
a
n
d
e
x
p
e
r
i
m
e
n
t
o
u
tc
o
m
e
s
h
o
w
s
s
i
g
n
if
ic
a
n
t
r
e
d
u
c
tio
n
i
n
c
o
m
p
ut
i
ng t
i
m
e
.
T
h
e r
es
t
o
f
t
h
e p
ap
er
i
s
o
r
g
a
n
i
zed
as
f
o
l
l
o
w
s
.
I
n
s
ec
t
i
o
n
I
I
t
h
e p
r
o
p
o
s
ed
ag
en
t
b
as
ed
f
o
r
ecas
t
i
n
g
m
ode
l
i
s
pr
e
s
e
n
t
e
d.
I
n
pe
n
u
l
t
i
m
a
t
e
s
e
c
t
i
on
e
x
pe
r
i
m
e
n
t
a
l
s
t
udy
i
s
c
a
r
r
i
e
d ou
t
.
T
h
e
c
on
c
l
us
i
on
a
n
d
f
ut
u
r
e
w
or
k
is
d
e
s
c
r
ib
e
d
in
la
s
t s
e
c
t
io
n
.
2.
P
R
O
P
O
S
E
D
A
G
EN
T B
A
S
E
D
F
O
R
EC
A
S
TI
N
G
M
O
D
EL
T
h
i
s
s
ect
i
o
n
p
r
es
en
t
an
ef
f
i
c
i
en
t
s
t
r
ea
m
a
n
d
n
o
n
-
st
r
e
a
m
da
t
a
i
n
f
l
o
w
f
or
e
c
a
s
t
i
ng
m
ode
l
by
a
dopt
i
ng
a
ge
nt
b
as
ed
M
ap
R
ed
u
ce
f
r
a
m
e
w
o
r
k
.
T
h
e p
r
o
ces
s
o
f
p
er
f
o
r
m
i
n
g
s
t
r
ea
m
a
n
d
n
o
n
-
s
t
r
ea
m
d
at
a i
n
f
l
o
w
f
o
r
ecas
t
i
n
g
m
o
d
el
o
n
ag
e
n
t
b
as
ed
M
ap
R
ed
u
ce f
r
a
m
e
w
o
r
k
co
n
s
i
s
t
o
f
f
o
l
l
o
w
i
n
g
s
t
ep
s
:
a)
D
a
ta
in
f
lo
w
p
r
e
d
ic
to
r
s
e
le
c
tio
n
D
a
ta
in
f
lo
w
pr
e
di
c
t
or
can
b
e d
ef
i
n
ed
b
y
Mu
l
t
i
v
ar
i
at
e G
a
u
s
s
i
a
n
M
i
x
t
u
r
e (
M
G
M
)
m
o
d
el
,
w
h
i
ch
i
s
c
om
pos
e
d
of
num
e
r
ou
s
w
e
i
ght
e
d
G
a
u
s
s
i
a
n
l
i
k
e
l
i
h
ood
de
n
s
i
t
y
m
ode
l
.
T
h
e
G
M
m
ode
l
w
i
t
h
v
ar
i
ab
l
es
ca
n
b
e
s
ta
te
d
a
s
f
o
llo
w
s
:
(
|
)
=
ℓ
.
=
1
(1
)
w
h
er
e
i
s
t
h
e co
l
l
ect
i
o
n
o
f
d
at
a s
t
r
eam
s
,
ℓ
is
th
e
lik
e
li
h
o
o
d
d
e
n
s
it
y
f
o
r
th
e
ℎ
v
ar
i
ab
l
e,
is
th
e
q
u
a
lif
ie
r
s
p
eci
f
i
er
f
o
r
t
h
e
ℎ
v
a
r
ia
b
le
,
s
a
tis
f
y
i
n
g
>
0
,
∑
=
1
=
1
,
=
(
1
,
…
,
,
1
,
…
,
)
i
s
th
e
q
u
a
l
if
ie
r
s
p
eci
f
i
er
s
et
f
o
r
G
M
m
o
d
el
.
b)
Qu
al
i
f
i
er
s
p
eci
f
i
er
ap
p
r
o
x
i
m
at
i
o
n
E
x
pe
c
t
a
t
i
on
m
a
x
i
m
i
z
a
t
i
o
n
a
l
g
or
i
t
hm
i
s
on
e
of
t
h
e
e
x
t
e
ns
i
v
e
l
y
u
s
e
d
m
e
t
h
odol
ogy
f
or
qu
a
l
i
f
i
e
r
s
p
e
c
if
ie
r
a
p
p
r
o
x
i
m
a
t
io
n
.
E
x
p
e
c
ta
tio
n
m
a
x
i
m
iz
a
tio
n
a
l
g
o
r
ith
m
a
id
in
m
i
n
i
m
iz
i
n
g
th
e
c
o
m
p
l
e
xi
t
y o
f
m
a
xi
m
u
m
lik
e
li
h
o
o
d
e
s
ti
m
a
t
io
n
.
E
x
p
ect
at
i
o
n
m
a
x
i
m
i
zat
i
o
n
al
g
o
r
i
t
h
m
i
s
co
m
p
o
s
ed
o
f
t
w
o
p
h
a
s
es
s
u
ch
a
s
E
x
p
ect
at
i
o
n
p
h
as
e an
d
Max
i
m
i
za
t
i
o
n
p
h
a
s
e.
I
n
E
x
p
ect
at
i
o
n
p
h
a
s
e t
h
e
r
es
p
o
n
d
i
n
g
d
eg
r
ee i
s
co
m
p
u
t
ed
u
s
i
n
g
p
r
es
en
t
q
u
a
lif
ie
r
s
p
e
c
if
ie
r
a
n
d
in
M
a
x
i
m
iz
at
i
o
n
p
h
as
e,
q
u
al
i
f
i
er
s
p
eci
f
i
er
ar
e u
p
d
at
ed
b
as
ed
o
n
r
es
p
o
n
d
i
n
g
d
eg
r
ee.
F
o
r
o
ne
K
-
v
a
r
ia
b
le
G
M
m
o
d
e
l,
in
w
h
ic
h
c
o
lle
c
tio
n
o
f
d
a
ta
s
tr
e
a
m
is
=
1
,
…
,
an
d
t
h
e q
u
al
i
f
i
er
s
p
eci
f
i
er
ar
e
=
,
,
=
1
,
…
,
,
on
e
i
t
e
r
a
t
i
on
of
t
h
e
e
xpe
c
ta
tio
n
m
a
x
i
m
iz
a
tio
n
a
l
g
o
r
ith
m
is
e
x
p
r
e
s
s
e
d
a
s
f
o
llo
w
s
:
E
x
p
e
c
ta
tio
n
p
h
a
s
e
:
,
=
∗
∑
∗
(2
)
M
ax
i
m
i
zat
i
o
n
p
h
as
e:
=
1
,
=
1
(3
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
25
02
-
4752
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
,
V
o
l.
11
, N
o
.
2
,
A
ug
us
t
2018
:
5
04
–
514
508
=
∑
∗
,
=
1
∑
,
=
1
(4
)
=
∑
,
=
1
∗
−
∗
−
ℬ
∑
,
=
1
.
(5
)
T
h
e ex
p
ect
at
i
o
n
m
a
x
i
m
i
zat
i
o
n
w
i
l
l
s
t
o
p
i
f
co
n
v
er
g
en
ce
co
n
d
itio
n
is
s
a
ti
s
f
ie
d
o
r
p
r
e
d
e
f
in
e
d
ite
r
a
tio
n
is
c
o
m
p
le
te
d
.
T
h
e
lo
c
a
l d
a
ta
in
f
lo
w
f
o
r
a
ll n
o
d
e
is
o
b
ta
i
n
e
d
p
o
s
t c
o
m
p
le
tio
n
/te
r
m
i
n
a
t
io
n
o
f
e
x
p
e
c
ta
tio
n
m
a
x
i
m
iz
a
tio
n
a
l
g
o
r
ith
m
.
c)
G
lo
b
a
l in
f
lo
w
p
r
e
d
ic
tio
n
m
o
d
e
l
I
n
or
de
r
t
o pr
e
di
c
t
t
h
e
g
l
oba
l
i
nf
l
o
w
of
da
t
a
s
t
r
e
a
m
,
t
h
e l
o
cal
i
n
f
l
o
w
o
f
d
at
a
s
t
r
ea
m
o
b
t
ai
n
ed
b
y
e
x
p
e
c
ta
tio
n
m
a
x
i
m
iz
a
tio
n
a
l
g
o
r
ith
m
s
h
o
u
ld
b
e
p
o
o
le
d
.
F
o
r
th
a
t
f
ir
s
tl
y
,
t
h
e
s
i
m
ila
r
it
y
a
m
o
n
g
d
i
f
f
e
r
e
n
t lo
c
a
l
m
o
d
el
s
i
s
o
b
t
ai
n
ed
.
S
eco
n
d
l
y
,
t
h
e l
o
cal
m
o
d
el
s
ar
e cl
as
s
i
f
i
e
d
.
A
n
d
l
a
s
t
l
y
,
t
h
e
l
o
cal
m
o
d
el
s
ar
e p
o
o
l
e
d
b
as
ed
on
w
e
i
gh
t
of
e
a
c
h
c
l
a
s
s
t
o f
or
m
g
l
oba
l
m
ode
l
s
.
L
e
t
ℋ
1
,
ℋ
2
r
ep
r
es
en
t
t
w
o
l
o
cal
m
o
d
el
s
,
an
d
t
h
e d
i
s
t
an
c
e
a
m
ong
t
h
e
m
i
s
c
o
m
put
e
d a
s
f
o
l
l
o
w
s
:
(
ℋ
1
,
ℋ
2
)
=
1
(
ℋ
1
,
ℋ
2
)
.
(6
)
W
h
er
e,
ℬ
(
ℋ
1
,
ℋ
2
)
=
1
∑
1
⃗
+
∞
−
∞
∗
1
∑
2
⃗
⃗
(7
)
T
h
e s
i
m
i
l
ar
i
t
y
a
m
o
n
g
t
w
o
l
o
ca
l
m
o
d
el
s
i
s
co
n
s
i
d
er
ed
t
o
b
e h
i
g
h
i
f
t
w
o
l
o
cal
m
o
d
el
s
ar
e cl
o
s
e t
o
each
o
t
h
er
.
T
h
en
cl
as
s
i
f
i
cat
i
o
n
p
r
o
b
l
e
m
o
f
l
o
cal
m
o
d
el
s
can
b
e s
t
a
t
ed
as
a
m
i
n
i
m
u
m
s
p
an
n
i
n
g
t
r
ee p
r
o
b
l
em
,
an
d
c
an
b
e ad
d
r
es
s
ed
u
s
i
n
g
[
8
]
.
L
e
t
t
h
e
w
ei
g
h
t
o
f
l
o
cal
m
o
d
el
s
b
e ex
p
r
es
s
ed
as
:
,
=
1
,
…
,
,
(8
)
C
o
v
ar
i
an
ce m
at
r
i
x
o
f
l
o
cal
m
o
d
el
s
i
s
ex
p
r
es
s
ed
as
:
,
=
1
,
…
,
,
(9
)
A
n
d
av
er
ag
e
v
al
u
e o
f
l
o
cal
m
o
d
el
s
i
s
ex
p
r
es
s
ed
as
:
,
=
1
,
…
,
.
(
10)
T
h
e
n
to
id
e
n
tif
y
w
h
ic
h
s
e
t t
h
e
m
o
d
e
l
ℋ
b
e
lo
n
g
s
to
is
e
x
p
r
e
s
s
e
d
b
y
f
o
llo
w
in
g
f
u
n
c
tio
n
=
ℋ
,
=
1
,
ℋ
∈
ℋ
ℛ
0
,
ℋ
∉
ℋ
ℛ
.
(
11)
L
et
co
n
s
i
d
er
a s
et
o
f
d
at
a s
t
r
ea
m
f
o
r
l
o
cal
m
o
d
el
ℋ
be
ℋ
.
I
f
d
at
a s
t
r
ea
m
s
i
ze o
f
l
o
cal
m
o
d
el
s
ar
e n
o
t
eq
u
al
,
t
h
e
w
ei
g
h
t
o
f
g
l
o
b
al
m
o
d
el
s
can
b
e co
m
p
u
t
ed
as
:
=
∑
∙
ℋ
∙
ℋ
,
=
1
∑
∙
ℋ
=
1
,
(
12)
T
h
e co
v
ar
i
an
ce
m
at
r
i
x
o
f
g
l
o
b
al
m
o
d
el
s
i
s
co
m
p
u
t
ed
as
:
=
∑
∙
ℋ
∙
∙
ℋ
,
=
1
∑
∙
ℋ
=
1
∙
ℋ
,
,
(
13)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
I
SSN
:
2502
-
4752
M
u
lti
-
A
ge
nt
bas
e
d M
ap
R
e
duc
e
M
ode
l
f
or
E
f
f
i
c
i
e
nt
U
t
i
l
i
z
at
i
on of
S
y
s
t
e
m
…
(
H
e
e
na k
ous
ar
)
509
T
h
e av
er
ag
e v
al
u
e o
f
g
l
o
b
al
m
o
d
el
s
i
s
co
m
p
u
t
ed
as
:
=
∑
∙
ℋ
∙
∙
ℋ
,
=
1
∑
∙
ℋ
=
1
∙
ℋ
,
(1
4
)
d)
A
g
e
n
t
b
as
ed
Map
R
ed
u
ce
m
o
d
el
T
h
e
pr
opos
e
d a
g
e
n
t
ba
s
e
d M
a
pR
e
du
c
e
m
ode
l
i
s
c
o
m
pos
e
d of
f
ol
l
o
w
i
ng
a
g
e
n
t
a
s
s
h
o
w
n i
n
F
i
g
ur
e
2.
Fi
g
ur
e
2
.
A
r
ch
i
t
ect
u
r
e o
f
p
r
o
p
o
s
ed
A
g
e
n
t
b
as
ed
M
ap
R
ed
u
ce
(
A
MR
)
M
o
d
el
F
i
l
e/
S
t
rea
m
R
ea
d
A
g
en
t
:
h
e
r
e
th
e
c
l
ie
n
t
s
y
s
te
m
(
F
ile
/S
t
r
ea
m
R
ead
A
g
en
t
)
r
ead
a
t
ex
t
f
i
l
e/
s
t
r
ea
m
d
at
a an
d
s
en
d
s
each
l
i
n
e o
f
t
ex
t
as
m
es
s
a
g
e t
o
t
h
e C
l
i
en
t
A
g
e
n
t
.
C
l
i
e
nt
A
g
e
nt
:
t
h
e
C
l
i
en
t
A
g
e
n
t
h
as
t
h
e r
e
f
er
en
ce t
o
t
h
e
R
e
m
o
t
e
A
g
en
t
(
M
ap
R
ed
u
ce
A
g
e
n
t
)
an
d
t
h
e
m
es
s
ag
e i
s
p
as
s
ed
o
n
t
h
e R
e
m
o
t
e A
g
en
t
.
S
erv
er/
M
a
p
R
ed
u
c
e A
g
en
t
:
t
h
e
S
er
v
er
/
Map
R
ed
u
ce A
g
e
n
t
g
et
s
t
h
e m
e
s
s
a
g
e f
r
o
m
t
h
e C
l
i
en
t
A
g
e
n
t
.
T
h
e
A
g
e
n
t
u
s
e
s
P
r
io
r
it
y
S
p
e
c
i
f
ie
r
to
s
e
le
c
t t
h
e
p
r
io
r
ity
o
f
th
e
m
e
s
s
a
g
e
a
n
d
f
ilt
e
r
s
t
he
q
ue
ue
s
a
c
c
o
r
d
i
ngl
y
.
T
he
P
ri
o
ri
t
y
S
p
eci
f
i
er
i
s
u
s
ed
t
o
s
eg
r
e
g
at
e t
h
e
m
es
s
ag
e a
m
o
n
g
t
h
e M
ap
R
ed
u
ce r
eq
u
es
t
s
a
n
d
r
et
r
i
ev
i
n
g
t
h
e l
i
s
t
o
f
o
u
t
co
m
es
(
O
U
T
_
L
I
S
T
)
m
es
s
ag
e
f
r
o
m
t
h
e
A
g
gr
e
g
a
t
o
r
A
ge
nt
.
M
a
pR
e
duc
e
A
g
e
nt
:
s
e
n
ds
a
c
r
os
s
t
h
e
m
e
s
s
a
g
e
t
o t
h
e
M
a
p
A
g
e
n
t
f
or
m
a
ppi
ng
t
h
e
w
or
ds
.
T
h
i
s
M
a
p
A
g
e
n
t
u
s
e
s
R
o
u
n
d R
obi
n
R
ou
t
e
r
di
s
pa
t
c
h
e
r
.
A
f
t
e
r
m
a
ppi
ng
t
h
e
w
or
ds
,
t
h
e
m
e
ss
a
g
e
s i
s
s
e
n
t
a
c
r
o
ss t
o
t
h
e
R
e
du
c
e
A
g
e
n
t
f
or
r
e
du
c
i
ng
t
h
e
w
or
ds
.
S
i
m
i
l
a
r
t
o M
a
p
A
g
e
n
t
t
h
e
R
e
du
c
e
A
g
e
n
t
a
l
s
o
u
s
e
s
R
oun
d R
obi
n
R
o
u
t
er
d
i
s
p
at
ch
er
.
T
he
r
ed
u
ced
r
es
u
l
t
s
ar
e s
en
t
t
o
t
h
e
A
ggr
e
ga
t
o
r
A
ge
n
t
t
h
at
d
o
es
an
i
n
-
m
e
m
or
y
a
gg
r
e
g
a
t
i
on
o
f
th
e
r
e
s
u
lt.
e)
A
ge
n
t
b
as
ed
M
ap
R
ed
u
ce i
m
p
l
e
m
en
t
at
i
o
n
o
f
d
at
a i
n
f
l
o
w
f
or
e
c
a
s
t
i
ng
m
ode
l
T
h
is
s
e
c
tio
n
p
r
e
s
e
n
t
th
e
d
e
ta
il
i
m
p
le
m
e
n
ta
tio
n
o
f
d
a
ta
i
n
f
lo
w
f
o
r
e
c
a
s
ti
n
g
o
n
a
g
e
n
t b
a
s
e
d
M
a
p
R
e
d
u
c
e
f
r
a
m
e
w
or
k
.
F
i
r
s
t
l
y
,
t
h
e
da
t
a
s
t
r
e
a
m
s
h
ou
l
d be
r
ou
t
e
d i
nt
o one
di
r
e
c
t
e
d g
r
a
ph
,
i
n
w
hi
c
h
a
r
c
s
r
ep
r
es
en
t
n
u
m
b
er
o
f
li
n
e
s
a
n
d
th
e
v
e
r
tic
e
s
r
e
p
r
e
s
e
n
t
n
u
m
b
e
r
o
f
f
ile
s
.
E
a
c
h
a
r
c
is
r
e
p
r
e
s
e
n
te
d
u
s
i
n
g
t
w
o
th
i
n
g
s
,
th
e
to
ta
l
v
o
lu
m
e
a
n
d
p
r
e
s
e
n
t lo
a
d
o
f
th
e
s
tr
e
a
m
s
/
f
ile
s
,
w
h
ic
h
a
r
e
d
e
r
iv
e
d
f
r
o
m
d
i
f
f
e
r
e
n
t c
o
m
p
u
ta
tio
n
a
l
a
p
p
lic
a
tio
n
s
tu
d
ie
s
.
F
ir
s
tl
y
,
w
e
i
n
p
u
t t
h
e d
at
a t
h
a
t
h
o
l
d
s
an
ar
c o
f
t
h
e
g
r
ap
h
s
a
n
d
i
t
s
co
n
t
e
n
t
.
T
h
en
t
h
e M
ap
o
p
er
at
i
o
n
p
r
o
d
u
ces
t
w
o
co
n
t
e
n
t
s
f
o
r
eac
h
ar
c
d
at
a,
o
n
e
k
e
y
ed
u
n
d
er
each
o
f
t
h
e
v
er
t
i
ces
t
h
a
t
f
o
r
m
t
h
e
ar
c.
T
h
e v
al
u
e eac
h
r
eco
r
d
pos
s
e
s
s
a
ugm
e
n
t
i
ng
f
l
o
w
an
d
p
at
h
.
E
ach
is
e
q
u
a
l to
to
ta
l v
o
lu
m
e
m
in
u
s
p
r
e
s
e
n
t lo
a
d
.
P
o
s
t
c
o
m
p
le
t
io
n
o
f
M
ap
o
p
er
at
i
o
n
,
a s
eq
u
e
n
ce o
f
co
n
t
ai
n
er
s
ar
e cr
eat
ed
,
each
c
o
n
t
ai
n
er
p
o
s
s
es
s
r
eco
r
d
s
f
o
r
e
ach
ar
c ad
j
acen
t
t
o
i
t
s
c
o
r
r
e
s
p
o
nd
i
n
g/
ne
i
g
hb
o
ur
i
n
g
ve
r
t
e
x.
I
n R
e
d
uc
e
s
s
t
a
ge
,
r
e
c
o
r
d
s
w
i
t
h
s
a
m
e
k
e
y
g
en
er
at
e
d
i
n
M
ap
s
t
a
g
e ar
e
cu
m
u
l
at
ed
b
y
p
er
f
o
r
m
i
n
g
s
h
u
f
f
l
e an
d
s
o
r
t
o
p
er
at
i
o
n
t
o
s
ea
r
ch
t
h
e p
at
h
an
d
co
m
p
u
t
e
.
P
o
s
t c
o
m
p
le
tio
n
o
f
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
25
02
-
4752
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
,
V
o
l.
11
, N
o
.
2
,
A
ug
us
t
2018
:
5
04
–
514
510
r
ed
u
ce o
p
er
at
i
o
n
,
al
l
t
h
e 2
-
d
i
s
t
an
ce p
at
h
s
a
m
o
n
g
p
at
h
s
/
n
o
d
e p
ai
r
ar
e g
en
er
at
ed
.
A
g
a
i
n
t
h
e
o
u
t
p
u
t
o
f
p
r
ev
i
o
u
s
r
ed
u
ce o
p
er
at
i
o
n
an
d
t
h
e r
aw
ar
c i
s
co
m
b
i
n
ed
t
o
f
o
r
m
i
n
p
u
t
an
d
p
er
f
o
r
m
i
t
er
at
i
v
e co
m
p
u
t
a
t
i
o
n
.
S
i
m
i
l
ar
l
y
,
t
h
e
M
ap
o
p
er
at
i
o
n
p
r
o
d
u
ces
s
eq
u
en
ce
s
o
f
r
eco
r
d
s
an
d
r
eco
r
d
s
w
i
t
h
s
a
m
e
k
e
y
ar
e
s
e
n
d
t
o
a s
p
eci
f
i
c r
ed
u
ce
w
o
r
k
er
.
P
o
s
t
co
m
p
l
et
i
o
n
o
f
M
ap
p
h
as
e,
al
l
3
-
d
i
s
t
a
nc
e
p
a
t
hs
a
nd
ar
e co
m
p
u
t
ed
.
S
i
m
i
l
a
r
l
y
,
t
h
e
p
r
o
ces
s
i
s
r
ep
eat
ed
.
T
h
e i
t
er
at
i
o
n
p
r
o
ces
s
i
s
car
r
i
ed
o
u
t
u
n
t
i
l
o
u
t
p
u
t
o
f
t
w
o
r
ed
u
ce
f
u
n
ct
i
o
n
s
ar
e s
a
m
e
.
T
h
i
s
m
ea
n
s
a
nd
p
at
h
s
a
m
o
n
g
each
n
o
d
e p
ai
r
a
r
e en
cl
o
s
ed
i
n
al
l
t
h
e o
u
t
p
u
t
p
r
o
d
u
ced
b
y
each
r
o
u
n
d
d
ecr
ea
s
e.
A
l
l
w
e n
ee
d t
o
do i
s
t
o s
or
t
t
h
e
s
e
ou
t
c
o
m
e
s
by
a
n
d n
ode
pa
i
r
.
I
n
n
ex
t
s
ec
t
i
o
n
ex
p
er
i
m
en
t
ar
e co
n
d
u
ct
ed
t
o
ev
al
u
at
e t
h
e p
er
f
o
r
m
a
n
ce D
I
F
-
A
M
R
o
ve
r
e
xi
t
i
ng
m
o
d
e
l
co
n
s
i
d
er
i
n
g
d
i
f
f
er
en
t
ex
p
er
i
m
en
t
s
.
3.
RE
S
U
L
T
AND ANA
L
YS
I
S
T
h
i
s
s
ect
i
o
n
d
es
cr
i
b
es
t
h
e ex
p
er
i
m
e
n
t
al
a
n
al
y
s
i
s
of
pr
opos
e
d D
a
t
a
I
n
f
l
o
w
F
or
e
c
a
s
t
i
ng
A
M
R
(
D
I
F
-
A
M
R
)
p
er
f
o
r
m
a
n
ce ac
h
i
ev
ed
o
v
er
ex
i
t
i
n
g
m
o
d
el
[
1
0
]
.
T
h
e D
I
F
-
A
M
R i
s i
m
p
l
e
m
e
n
t
e
d
u
s
i
n
g
C#
,
D
o
t
N
e
t
f
r
a
m
e
w
or
k
4.
5 a
n
d de
pl
oy
e
d on
M
i
c
r
os
of
t
A
z
u
r
e
C
l
o
u
d c
o
m
pu
t
i
ng
pl
a
t
f
or
m
.
a)
W
or
d f
r
eq
u
en
c
y
d
et
ect
i
o
n
co
m
p
u
t
at
i
o
n
T
he
DI
F
-
A
M
R
f
r
a
m
e
w
or
k
i
s
de
pl
oy
e
d on
D
3 V
M
i
n
s
t
a
n
c
e
c
om
pos
e
d of
4 v
i
r
t
u
a
l
c
o
m
pu
t
i
n
g
c
or
e
s
,
14 G
B
R
A
M
a
n
d 200 G
B
l
oc
a
l
S
S
D
h
a
r
d dr
i
v
e
s
pa
c
e
.
T
h
e
D
I
F
-
A
M
R
f
r
a
m
e
w
or
k
i
s
de
pl
oy
e
d on
M
i
c
r
os
of
t
azu
r
e cl
o
u
d
co
m
p
u
t
i
n
g
p
l
at
f
o
r
m
co
m
p
o
s
ed
o
f
o
n
e
m
a
s
t
er
n
o
d
es
an
d
f
o
u
r
v
i
r
t
u
a
l
c
o
m
put
i
ng
n
ode
s
t
o pe
r
f
or
m
m
a
p a
n
d r
e
du
c
e
ope
r
a
t
i
on
w
hi
c
h
r
u
ns
on
W
i
n
do
w
s
S
e
r
v
e
r
2012
R
2 op
e
r
a
t
i
n
g
s
y
s
t
e
m
.
H
D
I
ns
i
g
ht
c
l
u
s
t
e
r
[
14]
,
[
15]
i
s
c
on
s
i
de
r
e
d f
or
e
xi
t
i
n
g
m
ode
l
o
n
M
i
c
r
os
of
t
a
z
u
r
e
c
l
ou
d
c
o
m
put
i
ng
pl
a
t
f
or
m
.
I
de
n
t
i
c
a
l
c
om
p
u
t
i
n
g
pl
a
t
f
or
m
a
n
d c
onf
i
g
u
r
a
t
i
o
n
i
s
c
on
s
i
de
r
e
d f
or
bot
h
pr
op
os
e
d a
n
d e
x
i
s
t
i
ng
m
ode
l
.
T
h
e
w
or
d
f
r
e
qu
e
n
c
y
s
t
a
t
i
s
t
i
c
a
ppl
i
c
a
t
i
on i
s
de
v
e
l
ope
d u
s
i
n
g
C
# pr
og
r
a
m
i
n
g
l
a
ng
u
a
g
e
a
n
d D
ot
.
n
e
t
f
r
a
m
e
w
or
k
4.
5.
A
n
d
f
or
e
x
i
s
t
i
ng
H
a
doop ba
s
e
d
m
o
de
l
i
t
i
s
de
s
i
g
n
e
d us
i
ng
J
a
v
a
pr
og
r
a
m
i
ng
l
a
n
gu
a
g
e
.
T
h
e W
i
k
i
p
ed
i
a d
at
as
et
[
1
6
]
i
s
co
n
s
i
d
er
ed
f
o
r
ex
p
er
i
m
e
n
t
an
al
y
s
i
s
.
T
h
e W
i
k
i
p
ed
i
a d
at
as
et
i
s
h
u
g
e i
n
s
i
ze (
i
.
e.
>1
0
0
G
B
)
an
d
i
s
s
p
l
i
t
i
n
t
o
5
1
2
M
B
each
an
d
s
t
o
r
ed
i
n
A
z
u
r
e cl
o
u
d
co
n
t
ai
n
er
.
F
o
r
ex
p
er
i
m
e
n
t
al
an
al
y
s
i
s
t
h
i
s
w
o
r
k
co
n
s
i
d
er
8
G
B
o
f
d
at
a.
T
h
e
w
or
d f
r
e
qu
en
c
y
s
t
at
i
s
t
i
cs
ap
p
l
i
cat
i
o
n
s
w
er
e e
x
ec
u
t
ed
o
n
t
h
e
D
I
F
-
A
M
R
a
nd e
x
i
s
t
i
n
g
H
a
doop
d
ep
l
o
y
m
en
t
s
an
d
t
h
e r
es
u
l
t
s
o
b
t
ai
n
ed
ar
e n
o
t
ed
.
T
h
e M
ap
o
p
er
at
i
o
n
co
m
p
u
t
at
i
o
n
t
i
m
e at
each
co
m
p
u
t
i
n
g
n
o
d
e i
s
n
o
t
ed
an
d
r
es
u
l
t
o
b
t
ai
n
ed
ar
e s
h
o
w
n
in
F
ig
ur
e
3
.
F
r
o
m
r
e
s
u
l
t o
b
ta
in
e
d
i
t
ca
n
b
e s
ee
n
t
h
at
D
I
F
-
A
M
R
ac
h
i
e
v
es
f
as
t
er
ex
ec
u
t
i
o
n
t
i
m
e t
h
an
ex
i
t
i
n
g
m
o
d
el
.
R
ed
u
ce o
p
er
at
i
o
n
co
m
p
u
t
at
i
o
n
t
i
m
e
at
eac
h
co
m
p
u
t
i
n
g
n
o
d
e i
s
n
o
t
ed
an
d
r
es
u
l
t
o
b
t
ai
n
ed
ar
e s
h
o
w
n
i
n
Fi
g
ur
e
4
.
F
r
o
m
r
e
s
u
lt o
b
ta
in
e
d
it c
a
n
b
e
s
e
e
n
th
a
t D
I
F
-
A
M
R
ach
i
ev
e
s
f
a
s
t
er
ex
e
c
ut
i
o
n t
i
m
e
t
ha
n e
xi
t
i
n
g
m
o
d
el
.
F
r
o
m
r
e
s
u
l
t
o
b
t
ai
n
ed
i
t
can
b
e s
een
t
h
at
F
I
F
-
A
M
R
e
x
h
ib
it
s
f
a
s
te
r
c
o
m
p
u
ta
tio
n
ti
m
e
t
h
a
n
e
x
is
tin
g
H
ad
o
o
p
b
as
ed
m
o
d
el
[
1
0
]
.
T
h
e av
er
ag
e co
m
p
u
t
at
i
o
n
t
i
m
e o
f
Map
an
d
R
ed
u
ce o
p
er
at
i
o
n
co
m
p
u
t
at
i
o
n
t
i
m
e i
s
n
ot
e
d a
n
d i
s
s
h
o
w
n i
n F
i
g
ur
e
5.
T
h
e
c
om
pu
t
a
t
i
on
t
i
m
e
of
M
a
p ope
r
a
t
i
on
i
s
r
e
du
c
e
d b
y
8.
16%
by
D
I
F
-
AM
R
ov
e
r
e
x
i
s
t
i
ng
m
ode
l
.
T
h
e
c
om
p
u
t
a
t
i
on
t
i
m
e
of
R
e
du
c
e
op
e
r
a
t
i
on
i
s
r
e
du
c
e
d b
y
85.
23%
by
D
I
F
-
A
M
R
o
v
er
e
x
is
t
in
g
m
o
d
e
l.
T
h
e
to
ta
l c
o
m
p
u
ta
t
io
n
ti
m
e
f
o
r
p
e
r
f
o
r
m
in
g
w
o
r
k
f
r
e
q
u
e
n
c
y
s
ta
tis
ti
c
s
c
o
m
p
ut
a
t
i
o
n o
n
A
z
ur
e
c
lu
s
te
r
is
s
h
o
w
n
i
n
F
ig
ur
e
6.
T
h
e
c
om
p
u
t
a
t
i
on t
i
m
e
of
m
a
p
ph
a
s
e
c
ons
i
de
r
i
ng
D
I
F
-
A
M
R
i
s
100.
29 s
e
c
on
ds
a
n
d f
or
e
x
i
s
t
i
n
g
m
ode
l
i
s
198.
0 s
e
c
on
ds
.
T
h
e
c
om
p
u
t
i
n
g
t
i
m
e
r
e
du
c
t
i
on
of
49.
34%
a
n
d 43.
36%
i
s
a
c
h
i
e
v
e
d b
y
pr
op
os
e
d D
I
F
-
A
M
R
o
v
er
ex
i
s
t
i
n
g
m
o
d
el
f
o
r
Map
an
d
R
ed
u
ce o
p
er
at
i
o
n
r
es
p
ect
i
v
el
y
.
T
h
e t
o
t
al
co
m
p
u
t
at
i
o
n
t
i
m
e o
f
m
ap
p
h
as
e co
n
s
i
d
er
i
n
g
D
I
F
-
A
M
R
i
s
107.
0
5 s
e
c
on
ds
a
n
d f
or
e
x
i
s
t
i
ng
m
ode
l
i
s
210.
0
s
e
c
on
ds
.
T
h
e
t
ot
a
l
c
om
p
u
t
i
n
g
t
i
m
e
r
e
du
c
t
i
on
of
49.
023%
i
s
a
c
h
i
e
v
e
d b
y
pr
opos
e
d D
I
F
-
AM
R
o
v
e
r
e
xi
s
t
i
ng
m
o
d
e
l
.
Fi
g
ur
e
3.
M
a
p ope
r
a
t
i
on
c
om
p
u
t
a
t
i
on
t
i
m
e
o
n
w
or
k
e
r
n
ode
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
I
SSN
:
2502
-
4752
M
u
lti
-
A
ge
nt
bas
e
d M
ap
R
e
duc
e
M
ode
l
f
or
E
f
f
i
c
i
e
nt
U
t
i
l
i
z
at
i
on of
S
y
s
t
e
m
…
(
H
e
e
na k
ous
ar
)
511
Fi
g
ur
e
4
.
R
ed
u
ce o
p
er
at
i
o
n
co
m
p
u
t
at
i
o
n
t
i
m
e o
n
w
o
r
k
er
n
o
d
es
F
i
g
ur
e
5
.
A
v
er
ag
e co
m
p
u
t
at
i
o
n
t
i
m
e o
n
w
o
r
k
er
n
o
d
es
o
n
M
a
p
an
d
R
ed
u
ce p
h
as
e
Fi
g
ur
e
6
.
C
o
m
p
u
ta
t
io
n
ti
m
e p
er
f
o
r
m
an
ce e
v
al
u
at
i
o
n
co
n
s
i
d
e
r
i
n
g
D
I
F
-
A
M
R
a
n
d e
x
i
s
t
i
ng
m
ode
l
b)
Ho
t
-
W
or
d de
t
e
c
t
i
on
c
om
put
a
t
i
on
T
he
D
I
F
-
A
M
R
f
r
a
m
e
w
or
k
i
s
de
pl
oy
e
d on
D
3 V
M
i
n
s
t
a
n
c
e
c
om
pos
e
d of
1 v
i
r
t
u
a
l
c
o
m
pu
t
i
n
g
c
or
e
s
,
14 G
B
R
A
M
a
n
d 200 G
B
l
oc
a
l
S
S
D
h
a
r
d dr
i
v
e
s
pa
c
e
.
T
h
e
D
I
F
-
A
M
R
f
r
a
m
e
wo
r
k
i
s
de
pl
oy
e
d on
M
i
c
r
os
of
t
a
z
u
r
e
c
l
ou
d c
o
m
p
u
t
i
n
g pl
a
t
f
or
m
c
o
m
pos
e
d o
f
on
e
m
a
s
t
e
r
n
o
de
s
a
n
d
f
ou
r
v
i
r
t
u
a
l
c
o
m
put
i
ng
n
ode
s
t
o pe
r
f
or
m
m
a
p a
n
d r
e
du
c
e
ope
r
a
t
i
on
w
hi
c
h
r
u
ns
on
W
i
n
do
w
s
S
e
r
v
e
r
2012
R
2 op
e
r
a
t
i
n
g
s
y
s
t
e
m
.
H
D
I
ns
i
g
ht
c
l
u
s
t
e
r
[
14]
,
[
15]
i
s
c
on
s
i
de
r
e
d
f
or
e
xi
t
i
n
g
m
ode
l
o
n
M
i
c
r
os
of
t
a
z
u
r
e
c
l
ou
d
c
o
m
put
i
ng
pl
a
t
f
or
m
.
I
de
n
t
i
c
a
l
c
om
p
u
t
i
n
g pl
a
t
f
or
m
a
n
d c
onf
i
g
u
r
a
t
i
o
n
i
s
c
o
n
s
i
de
r
e
d f
or
bot
h
pr
opos
e
d a
n
d e
x
i
s
t
i
ng
m
ode
l
[
17]
.
T
h
e
h
ot
-
w
or
d
de
t
e
c
t
i
on
a
l
g
or
i
t
hm
[
18]
i
s
de
v
e
l
ope
d u
s
i
ng
C
# pr
og
r
a
m
i
ng
l
a
n
gu
a
g
e
a
n
d D
ot
.
n
e
t
f
r
a
m
e
w
or
k
4.
5.
A
n
d f
or
e
x
i
s
t
i
ng
H
a
doop ba
s
e
d m
ode
l
i
t
i
s
de
s
i
g
n
e
d us
i
ng
J
a
v
a
pr
og
r
a
m
i
ng
l
a
n
gu
a
g
e
.
T
h
e
“
M
ov
i
e
t
w
e
e
t
i
n
gs
”
d
at
as
et
[
1
9
]
i
s
co
n
s
i
d
er
ed
f
o
r
ex
p
er
i
m
e
n
t
a
n
al
y
s
i
s
an
d
s
t
o
r
ed
i
n
A
z
u
r
e cl
o
u
d
co
n
t
ai
n
er
.
T
w
eet
s
co
n
s
i
s
t
i
n
g
o
f
2500
0,
50
000
,
750
00 a
n
d
100
000 m
ov
i
e
s
i
s
c
on
s
i
de
r
e
d a
n
d i
s
r
e
pr
e
s
e
n
t
e
d a
s
25K
,
50K
,
75K
a
n
d 100K
.
T
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN
:
25
02
-
4752
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
,
V
o
l.
11
, N
o
.
2
,
A
ug
us
t
2018
:
5
04
–
514
512
ho
t
-
w
o
r
d
d
et
ect
i
o
n
al
g
o
r
i
t
h
m
w
er
e ex
ecu
t
ed
o
n
t
h
e D
I
F
-
A
M
R
a
n
d e
x
i
s
t
i
ng
H
a
doop d
e
pl
oy
m
e
n
t
s
a
n
d t
h
e
r
es
u
l
t
s
o
b
t
ai
n
ed
ar
e n
o
t
ed
.
T
h
e t
o
t
al
co
m
p
u
t
at
i
o
n
t
i
m
e o
f
D
I
F
-
A
M
R
a
nd
e
xi
s
t
i
ng
m
od
e
l
i
s
n
ot
e
d a
n
d i
s
s
h
o
w
n
i
n F
i
g
ur
e
7
.
E
x
p
er
i
m
en
t
an
al
y
s
e
s
s
h
o
w
s
as
n
u
m
b
er
o
f
t
w
e
et
s
i
n
cr
ea
s
es
t
h
e co
m
p
u
t
at
i
o
n
t
i
m
e o
f
b
o
t
h
D
I
F
-
A
M
R
an
d
ex
i
s
t
i
n
g
m
o
d
el
i
n
cr
eas
es
.
T
h
e co
m
p
u
t
at
i
o
n
p
er
f
o
r
m
an
ce i
m
p
r
o
v
e
m
e
n
t
o
f
D
I
F
-
A
M
R
o
ve
r
e
xi
s
t
i
ng
m
ode
l
f
or
25K
i
s
a
bou
t
10.
6
%
,
f
or
50K
i
s
a
bou
t
16.
69%
,
f
or
75K
i
s
a
bou
t
58
.
41%
a
n
d f
or
10
0K
i
s
a
bou
t
5
7
.
9
8
%
i
s
ach
i
ev
ed
.
A
n
a
v
er
ag
e co
m
p
u
t
at
i
o
n
p
er
f
o
r
m
an
ce
i
m
p
r
o
v
e
m
en
t
o
f
4
6
.
9
4
%
i
s
ach
i
ev
ed
D
I
F
-
AM
R
o
v
e
r
e
x
is
ti
n
g
m
o
d
e
l.
B
a
s
e
d
o
n
th
e
r
e
s
u
lts
o
b
ta
in
e
d
i
n
th
e
h
o
t
w
o
r
d
d
e
te
c
tio
n
a
lg
o
r
ith
m
i
t
can
b
e co
n
cl
u
d
ed
t
ha
t
t
he
D
I
F
-
A
M
R
m
o
d
el
ex
h
i
b
i
t
s
l
o
w
er
co
m
p
u
t
at
i
o
n
t
i
m
e
w
h
en
co
m
p
ar
ed
t
o
t
h
e e
x
i
s
t
i
n
g
m
o
d
el
.
Fi
g
ur
e
7
.
C
o
m
p
u
t
at
i
o
n
t
i
m
e p
er
f
o
r
m
an
ce e
v
al
u
at
i
o
n
co
n
s
i
d
e
r
i
n
g
D
I
F
-
A
M
R
a
n
d e
x
i
s
t
i
ng
m
ode
l
H
D
I
n
s
i
gh
t
c
l
u
s
t
e
r
[
14]
,
[
15
]
i
s
c
on
s
i
de
r
e
d
f
or
e
x
i
t
i
ng
m
ode
l
on
M
i
c
r
os
o
f
t
a
z
u
r
e
c
l
ou
d
c
om
p
u
t
i
ng
pl
a
t
f
or
m
.
I
de
n
t
i
c
a
l
c
o
m
pu
t
i
ng
pl
a
t
f
or
m
a
n
d c
onf
i
gu
r
a
t
i
on i
s
c
on
s
i
de
r
e
d f
or
bot
h
pr
o
p
os
e
d a
n
d e
x
i
s
t
i
n
g
m
ode
l
.
T
h
e
h
ot
-
w
or
d de
t
e
c
t
i
on
a
l
g
or
i
t
hm
[
18]
i
s
de
v
e
l
ope
d u
s
i
n
g C
# pr
og
r
a
m
i
ng
l
a
ng
u
a
g
e
a
n
d D
ot
.
n
e
t
f
r
a
me
w
or
k
4.
5.
A
n
d f
or
e
x
i
s
t
i
n
g H
a
doop ba
s
e
d
m
ode
l
i
t
i
s
de
s
i
gn
e
d us
i
n
g
J
a
v
a
pr
og
r
a
m
i
n
g l
a
ngu
a
g
e
.
T
he
“
M
o
v
i
et
w
eet
i
n
g
s
” d
at
as
et
[
1
7
]
i
s
co
n
s
i
d
er
ed
f
o
r
ex
p
er
i
m
e
n
t
an
al
y
s
i
s
an
d
s
t
o
r
ed
i
n
A
z
u
r
e
cl
o
u
d
co
n
t
ai
n
er
.
T
w
e
e
t
s
c
on
s
i
s
t
i
n
g
of
25000,
5
0000,
7
500
0 a
n
d 1
000
00 m
o
v
i
es
i
s
co
n
s
i
d
er
ed
an
d
i
s
r
ep
r
es
en
t
ed
as
2
5
K
,
5
0
K
,
75K
a
n
d 1
00K
.
T
h
e
h
ot
-
w
o
r
d
d
et
ect
i
o
n
al
g
o
r
i
t
h
m
w
er
e ex
ecu
t
ed
o
n
t
h
e D
I
F
-
A
M
R
a
nd e
x
i
s
t
i
ng
H
a
doop
d
e
p
lo
y
m
e
n
t
s
a
n
d
th
e
r
e
s
u
lts
o
b
ta
in
e
d
a
r
e
n
o
te
d
.
T
h
e
to
ta
l c
o
m
p
u
ta
tio
n
c
o
s
t o
f
D
I
F
-
A
M
R
a
n
d e
x
i
s
t
i
ng
m
ode
l
i
s n
o
t
e
d
a
nd
i
s
s
ho
w
n i
n
T
ab
l
e 2
.
E
x
p
er
i
m
en
t
a
n
al
y
s
es
s
h
o
w
s
a
s
n
u
m
b
er
o
f
t
w
eet
s
i
n
cr
eas
es
t
h
e co
m
p
u
t
at
i
o
n
c
os
t
of
bot
h
D
I
F
-
A
M
R
an
d
ex
i
s
t
i
n
g
m
o
d
el
i
n
cr
ea
s
es
.
T
h
e co
m
p
u
t
at
i
o
n
co
s
t
i
m
p
r
o
v
e
m
en
t
o
f
D
I
F
-
AM
R
o
v
e
r
e
x
i
s
t
i
ng
m
ode
l
f
or
25K
i
s
a
bo
u
t
32.
67%
,
f
or
5
0K
i
s
a
bo
u
t
3
7.
34%
,
f
or
75K
i
s
a
bou
t
63
.
0
5%
a
n
d f
or
100K
i
s
a
bou
t
62.
23%
i
s
a
c
hi
e
v
e
d.
A
n a
v
e
r
a
g
e
c
o
m
p
u
t
a
t
i
on c
os
t
i
m
pr
ov
e
m
e
n
t
of
48.
82%
i
s
a
c
hi
e
v
e
d D
I
F
-
AM
R
o
v
e
r
e
x
is
t
in
g
m
o
d
e
l.
B
a
s
e
d
o
n
th
e
r
e
s
u
lt
s
o
b
ta
in
e
d
in
t
h
e
h
o
t
w
o
r
d
d
e
te
c
tio
n
a
lg
o
r
ith
m
it c
a
n
b
e
c
o
n
c
lu
d
e
d
th
a
t
t
he
DI
F
-
A
M
R
m
ode
l
e
x
h
i
bi
t
s
l
o
w
e
r
c
om
put
a
t
i
on
c
os
t
w
he
n
c
o
m
p
a
r
e
d
to
th
e
e
x
is
tin
g
m
o
d
e
l.
T
ab
l
e
2
.
C
os
t
c
om
put
a
t
i
on
u
s
i
n
g
D
I
F
-
A
M
R
t
e
c
hn
i
q
ue
N
u
m
b
er
o
f
T
w
i
t
t
er
F
e
ed
s
C
o
s
t c
o
m
p
u
ta
tio
n
(
$
)
E
xi
s
t
i
n
g
D
I
F
-
AM
R
I
m
p
r
o
v
e
m
e
n
t
(
%
)
2
5
K
0
.
4
2
9
1
0
.
2
8
8
9
3
2
.
67
5
0
K
0
.
8
4
6
0
0
.
5
3
0
1
3
7
.
34
7
5
K
1
.
5
2
7
5
0
.
5
6
4
4
6
3
.
05
1
0
0
K
1
.
6
7
1
1
0
.
6
3
1
1
6
2
.
23
I
n
th
i
s
s
e
c
tio
n
th
e
e
x
e
c
u
tio
n
o
f
th
e
i
m
p
r
e
c
is
e
a
p
p
lic
a
tio
n
s
n
a
m
e
l
y
w
o
r
d
f
r
e
q
u
e
n
c
y
s
ta
tis
ti
c
s
a
n
d
h
o
t
w
or
d de
t
e
c
t
i
on
i
s
p
r
es
en
t
ed
.
T
h
e r
es
u
l
t
s
p
r
es
en
t
ed
h
er
e
p
r
o
v
e t
h
at
t
h
e D
I
F
-
A
M
R
m
o
d
el
r
ed
u
ces
t
h
e
co
m
p
u
t
at
i
o
n
t
i
m
e o
b
s
er
v
ed
d
u
e t
o
t
h
e n
o
v
el
ag
e
n
t
b
as
ed
p
ar
al
l
el
c
o
m
p
u
t
at
i
o
n
m
et
h
o
d
i
n
co
r
p
o
r
at
ed
.
A
n
a
v
e
r
a
g
e
r
e
du
c
t
i
on
of
49.
02%
f
or
w
or
d f
r
e
qu
e
n
c
y
s
t
a
t
i
s
t
i
c
s
a
n
d 46.
94%
f
or
t
h
e
h
o
t
w
o
r
d
d
e
te
c
tio
n
is
r
e
por
t
e
d c
on
s
i
de
r
i
n
g
t
h
e
D
I
F
-
A
M
R
m
ode
l
w
he
n
c
o
m
pa
r
e
d t
o t
h
e
e
x
i
s
t
i
n
g
m
ode
l
[
10]
.
T
h
e
c
um
u
l
a
t
i
v
e
an
al
y
s
i
s
o
v
er
s
t
at
e
-
of
-
a
r
t te
c
h
n
iq
u
e
i
n
T
a
b
le
2
s
ho
w
s
t
he
e
f
f
i
c
i
e
nc
y
o
f
D
I
F
-
A
M
R o
v
e
r
s
t
a
t
e
-
of
-
a
r
t te
c
h
n
iq
u
e
i
n t
e
r
m
s
o
f
r
o
b
us
t
ne
s
s
a
nd
s
c
a
la
b
ilit
y
.
S
i
n
c
e
,
D
I
F
-
A
M
R
s
u
ppor
t
pa
r
a
l
l
e
l
pr
oc
e
s
s
i
n
g
o
f
s
t
r
e
a
m
a
n
d
n
on
-
st
r
e
a
m
da
t
a
by
a
dopt
i
n
g
M
a
pR
e
du
c
e
f
r
a
m
e
w
or
k
.
A
dopt
i
on
of
a
g
e
n
t
ba
s
e
d c
om
p
u
t
i
n
g
a
i
d i
n
f
u
l
l
y
u
t
i
l
i
z
i
n
g
s
y
s
t
e
m
r
es
o
u
r
ces
.
S
u
p
p
o
r
t
accu
r
at
e p
r
o
ces
s
i
n
g
o
f
a
l
l
t
y
p
e o
f
d
at
a
s
h
o
w
s
r
o
b
u
s
t
n
es
s
o
f D
I
F
-
A
M
R
.
A
dopt
i
on
c
l
ou
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nd
o
ne
s
i
a
n J
E
l
e
c
E
ng
&
C
o
m
p
S
c
i
I
SSN
:
2502
-
4752
M
u
lti
-
A
ge
nt
bas
e
d M
ap
R
e
duc
e
M
ode
l
f
or
E
f
f
i
c
i
e
nt
U
t
i
l
i
z
at
i
on of
S
y
s
t
e
m
…
(
H
e
e
na k
ous
ar
)
513
pl
a
t
f
or
m
a
i
d i
n
pr
ov
i
ng
s
c
a
l
a
b
i
l
i
t
y
o
f
pr
oc
e
s
s
i
ng
of
l
a
r
g
e
a
m
ou
n
t
of
da
t
a
o
f
v
a
r
i
ous
t
y
pe
s
on
l
a
r
g
e
c
o
m
p
u
t
i
n
g
cl
u
s
t
er
s
.
A
l
l
t
h
es
e
f
eat
u
r
e at
t
r
i
b
u
t
ed
t
o
t
h
e p
er
f
o
r
m
a
n
ce i
m
p
r
o
v
e
m
e
n
t
o
f D
I
F
-
A
M
R
ov
e
r
e
x
i
s
t
i
ng
m
ode
l
,
s
ho
w
n
i
n T
a
b
l
e
3
.
T
ab
l
e 3
.
C
o
m
p
a
r
is
o
n
w
i
th
s
ta
t
e
o
f
a
r
t te
c
h
n
iq
u
e
[8
]
[9
]
[
1
0
]
DI
F
-
AM
R
M
ap
R
ed
u
c
e p
l
a
t
f
o
r
m
c
o
n
s
i
d
er
ed
H
a
d
o
o
p
S
t
or
m
H
a
d
o
o
p
C
u
st
o
m
A
g
en
t
b
as
ed
Y
es
No
Y
es
Y
es
C
l
o
u
d
a
d
o
p
t
e
d
Y
es
Y
es
Y
es
Y
es
N
on
-
s
t
r
e
a
m
d
a
t
a
p
r
oc
e
s
s
i
n
g
s
u
p
p
o
r
t
Y
es
Y
es
Y
es
Y
es
S
t
r
ea
m
d
a
t
a
p
r
o
c
e
s
s
i
n
g
s
u
pp
or
t
No
Y
es
No
Y
es
H
y
b
r
i
d
p
r
o
c
e
s
s
i
n
g
s
up
p
o
r
t
No
Y
es
No
Y
es
F
o
r
e
cas
t
i
n
g
acc
u
r
a
cy
Y
es
No
Y
es
Y
es
4.
CO
NCL
U
S
I
O
N
T
h
e m
ai
n
co
n
t
r
i
b
u
t
i
o
n
o
f
t
h
i
s
w
o
r
k
i
s
p
r
es
en
t
i
n
g
an
a
g
e
n
t
b
as
ed
s
cal
ab
l
e s
t
r
ea
m
an
d
n
o
n
-
st
r
e
a
m
p
r
o
ces
s
i
n
g
o
f
s
o
ci
al
m
ed
i
a a
n
d
s
en
s
o
r
d
at
a u
s
i
n
g
Map
R
ed
u
ce f
r
a
m
e
w
o
r
k
.
E
f
f
i
ci
e
n
t
d
at
a
i
n
f
l
o
w
f
o
r
ecas
t
i
n
g
m
o
d
el
ad
o
p
t
i
n
g
Map
R
ed
u
ce a
n
d
cl
o
u
d
p
l
at
f
o
r
m
i
s
p
r
es
e
n
t
e
d
.
M
u
l
t
i
v
ar
i
a
t
e G
au
s
s
i
a
n
Mi
x
t
u
r
e (
M
G
M
)
m
o
d
el
i
s
de
s
i
gn
e
d t
o i
m
pr
ov
e
t
h
e
a
c
c
u
r
a
c
y
of
f
or
e
c
a
s
t
i
n
g
f
or
bot
h
s
t
r
e
a
m
a
n
d
n
on
-
st
r
e
a
m
d
at
a.
D
I
F
-
AM
R
o
v
e
r
c
o
m
e
s
t
h
e
li
m
ita
tio
n
o
f
s
t
a
te
-
of
-
ar
t
t
ech
n
i
q
u
e i
n
t
er
m
s
o
f
co
m
p
u
t
at
i
o
n
o
v
er
h
ead
w
h
i
ch
ar
e ex
p
er
i
m
e
n
t
al
p
r
o
v
en
.
E
x
p
er
i
m
e
n
t
ar
e co
n
d
u
ct
ed
t
o
ev
al
u
at
e t
h
e p
er
f
o
r
m
an
ce o
f
p
r
o
p
o
s
ed
D
I
F
-
A
M
R
o
v
er
s
t
at
e o
f
ar
t
-
te
c
h
n
iq
u
e
in
te
r
m
s
o
f
c
o
m
p
u
t
a
tio
n
t
i
m
e
c
o
ns
i
d
e
r
i
n
g d
i
f
f
e
r
e
nt
e
xp
e
r
i
m
e
nt
s
uc
h
a
s
w
o
r
d
f
r
e
q
ue
nc
y
s
t
a
t
i
s
t
i
c
pr
e
di
c
t
i
on
a
n
d
h
ot
w
or
d
de
t
e
c
t
i
on
.
E
x
pe
r
i
m
e
n
t
o
u
t
c
o
m
e
s
s
h
o
w
s
D
I
F
-
A
M
R
i
m
p
r
o
ve
s
c
o
m
p
ut
a
t
i
o
n
t
i
m
e
b
y
49.
02%
ov
e
r
e
x
i
s
t
i
ng
m
ode
l
f
or
w
or
d f
r
e
qu
e
n
c
y
s
t
a
t
i
s
t
i
c
s
a
nd 46.
94%
f
or
h
ot
w
or
d de
t
e
c
t
i
on
.
T
h
e
e
x
pe
r
i
m
e
nt
o
u
t
co
m
es
s
h
o
w
s
s
i
g
n
i
f
i
can
t
p
er
f
o
r
m
a
n
ce i
m
p
r
o
v
e
m
en
t
o
f
D
I
F
-
A
M
R
o
v
er
ex
i
t
i
n
g
m
o
d
el
i
n
t
er
m
s
o
f
c
o
m
p
u
ta
tio
n
ti
m
e
c
o
n
s
id
e
r
in
g
e
x
e
c
u
tio
n
o
n
c
lo
u
d
p
la
tf
o
r
m
f
o
r
r
e
la
tiv
e
l
y
la
r
g
e
s
tr
e
a
m
a
n
d
n
o
n
-
s
t
r
ea
m
d
at
a.
T
hus
D
I
F
-
A
M
R
i
s
s
cal
ab
l
e i
r
r
es
p
ect
i
v
e o
f
d
at
a s
i
ze a
n
d
co
m
p
u
t
i
n
g
cl
u
s
t
er
s
i
ze.
T
h
e f
u
t
u
r
e
w
o
r
k
w
o
u
l
d
co
n
s
i
d
er
p
er
f
o
r
m
an
ce e
v
al
u
at
i
o
n
co
n
s
i
d
er
i
n
g
l
ar
g
e d
at
as
et
a
n
d
al
s
o
co
n
s
i
d
er
s
e
c
u
r
i
t
y
pr
ov
i
s
i
on
i
ng
f
or
c
o
m
pu
t
i
ng
da
t
a
on
c
l
ou
d pl
a
t
f
or
m
s
.
T
h
i
s
w
or
k
w
o
u
l
d
f
u
r
t
h
e
r
c
o
n
s
i
de
r
opt
i
m
i
z
a
t
i
on
o
f
M
ap
R
ed
u
ce s
c
h
ed
u
l
er
f
o
r
f
u
r
t
h
er
r
ed
u
ct
i
o
n
o
f
co
m
p
u
t
at
i
o
n
t
i
m
e.
R
EF
ER
EN
C
ES
[
1
]
L
.
T
a
ng
e
t
a
l
.
,
`
`
A
f
r
a
m
e
w
or
k of
t
r
a
v
e
l
i
ng
c
o
m
pa
ni
on di
s
c
ov
e
r
y
on t
r
aj
ect
o
r
y
d
at
a s
t
r
e
a
m
s
,
'
'
A
C
M
T
r
ans
.
I
nt
e
l
l
.
Sy
s
t
.
T
e
c
hnol
.
,
v
ol
.
5,
n
o.
1,
A
r
t
.
I
D
3
,
D
e
c
.
201
3.
[
2
]
T
.
B
e
ns
on,
A
.
A
na
nd,
A
.
A
ke
l
l
a
,
a
nd M
.
Z
ha
ng
,
“
M
i
c
r
o T
E
:
F
i
ne
gr
ai
ne
d t
r
af
f
i
c
e
n
gi
ne
e
r
i
n
g f
or
d
at
a c
e
nt
e
r
s
,
”
i
n
P
r
oc
.
C
oN
E
X
T
,
A
r
t
.
n
o.
8,
20
11.
[
3
]
K
.
C
h
en
et
al
.
,
“
O
S
A:
An
o
p
tic
a
l
s
w
itc
h
in
g
a
r
c
h
ite
c
t
u
r
e
fo
r
d
a
t
a
c
e
n
te
r
n
e
tw
o
r
k
s
w
ith
u
n
p
r
e
c
e
d
e
n
te
d
fle
x
ib
il
ity
,
”
in
P
r
oc
.
N
S
D
I
,
p.
18
,
2
01
2.
[
4
]
A
.
D
.
F
e
r
g
us
on,
A
.
G
uha
,
J
.
P
l
a
c
e
,
R
.
F
ons
e
c
a
,
a
nd
S
.
K
r
i
s
h
na
m
ur
t
hi
,
“
P
a
r
tic
ip
a
to
r
y
n
e
tw
o
r
k
in
g
,
”
i
n P
r
oc
.
H
ot
-
I
C
E
,
p.
2,
2
01
2.
[
5
]
M
.
Al
-
F
ar
es
,
S
.
R
ad
h
ak
r
i
s
h
n
a
n
,
B
.
R
ag
h
av
an
,
N
.
H
u
an
g
,
an
d
A
.
V
ah
d
at
,
“
H
e
de
r
a:
D
y
nam
i
c
f
l
o
w
s
c
he
dul
i
ng
f
or
dat
a c
e
nt
e
r
ne
t
w
or
k
s
,
”
i
n
P
r
oc
.
N
S
D
I
,
p.
1
9,
20
10.
[
6
]
A
. C
u
r
t
i
s
, W
. K
i
m
, a
n
d
P
. Y
a
l
a
g
a
n
d
u
l
a
,
“
M
a
ho
ut
:
L
ow
-
ov
e
r
he
a
d d
at
ac
e
nt
e
r
t
r
a
f
f
i
c
m
a
na
ge
m
e
nt
us
i
n
g e
nd
-
hos
t
-
bas
e
d
e
l
e
p
ha
nt
de
t
e
c
t
i
on
,
”
i
n
P
r
o
c
. I
E
E
E
I
N
F
O
C
O
M
,
p
p
.
1
6
2
9
–
163
7
,
2
01
1.
[
7
]
H
.
H
.
B
azzaz et
al
.
,
“
Sw
i
t
c
hi
n
g t
he
opt
i
c
al
di
v
i
de
:
F
un
dam
e
nt
al
c
hal
l
e
nge
s
f
or
hy
br
i
d e
l
e
c
t
r
i
c
al
/
op
t
i
c
al
d
at
ac
e
nt
e
r
n
et
w
o
r
ks
,
”
i
n
P
r
oc
.
S
O
C
C
,
A
r
t
.
n
o.
3
0,
20
11
.
[
8
]
A
. E
l
F
a
z
z
i
k
i
, D
.
B
en
s
l
i
m
an
e,
A
.
S
ad
i
q
,
J
.
O
u
ar
zazi
an
d
M
.
S
ad
g
al
,
"
A
n
A
ge
nt
B
a
s
e
d T
r
a
f
f
i
c
R
e
g
ul
a
t
i
on
S
y
s
t
e
m
f
or
t
he
R
oa
ds
i
de
A
i
r
Q
ua
l
i
t
y
C
ont
r
ol
,
" i
n
I
E
E
E
A
c
c
e
s
s
,
vo
l
.
5
,
p
p.
13
192
-
1
32
01,
20
17
.
[
9
]
Z
.
Z
ha
o,
W
.
D
i
ng
,
J
.
W
a
ng a
nd Y
.
H
a
n,
"
A
H
y
br
i
d P
r
oc
e
s
s
i
ng
S
y
s
t
e
m
f
or
L
a
r
g
e
-
S
ca
l
e
T
r
a
f
f
i
c
S
en
s
o
r
D
at
a,
"
i
n
I
E
E
E
A
c
c
e
s
s
,
vo
l
.
3,
pp
.
2
34
1
-
23
51,
20
15
.
[
1
0]
W
ei
D
ai
,
P
en
g
H
u
,
“
R
e
s
e
a
r
c
h on
P
e
r
s
o
na
l
i
z
e
d B
e
ha
v
i
or
s
R
e
c
om
m
e
nda
t
i
on
S
y
s
t
e
m
B
a
s
e
d on C
l
o
ud C
om
put
i
ng
”
,
T
E
L
K
O
M
N
I
K
A
I
nd
one
s
i
an J
our
n
al
o
f
E
l
e
c
t
r
i
c
al
E
n
gi
ne
e
r
i
n
g
V
ol
.
12,
N
o.
2,
F
e
br
ua
r
y
201
4
,
pp.
1
48
0
~
14
86
.
[
1
1]
H
a
doo
p,
ht
t
p
:
/
/
ha
do
op.
a
pa
c
he
.
or
g
,
a
c
c
e
s
s
e
d O
c
t
.
22,
20
17.
[
1
2]
H
a
doo
p.
[
O
n
l
i
ne
]
.
A
v
a
i
l
a
bl
e
:
ht
t
p
:
/
/
ha
d
oo
p.
a
pa
c
he
.
or
g
/
,
a
c
c
e
s
s
e
d O
c
t
.
21
,
2
01
7.
[
1
3]
H
ai
w
en
H
an
,
W
ei
p
i
n
g
Z
h
en
g
,
“
A
P
r
i
v
ac
y
D
at
a
-
o
r
i
en
t
ed
H
i
er
ar
ch
i
cal
M
ap
R
ed
u
ce P
r
o
g
r
am
m
i
n
g
M
o
d
el
”,
T
E
L
KO
M
NIKA
I
nd
one
s
i
a
n J
our
n
al
of
E
l
e
c
t
r
i
c
al
E
ng
i
ne
e
r
i
ng
,
V
ol
.
11,
N
o
.
8,
A
ug
us
t
20
13,
pp
.
45
8
7~
4
59
3 e
-
I
SSN
:
208
7
-
27
8X
.
[
1
4]
H
di
ns
i
g
ht
(
ha
d
oo
p o
n a
z
ur
e
)
,
"
ht
t
ps
:
/
/
w
w
w
.
ha
doo
po
na
z
ur
e
.
c
om
/
,
a
c
c
e
s
s
e
d O
c
t
.
20
,
2
01
7.
Evaluation Warning : The document was created with Spire.PDF for Python.