TELKOM
NIKA
, Vol. 11, No. 9, September 20
13, pp.
5267
~52
7
6
ISSN: 2302-4
046
5267
Re
cei
v
ed Fe
brua
ry 24, 20
13; Re
vised
June 9, 201
3; Acce
pted Jun
e
21, 2013
Mobile Computing Clouds Interactive Model and
Algorithm based on Multi-core Grids
Liu Lizhao*
1
,
Zhu Shunzh
i
1
, Sun Zhonghai
2
, Li Qi
3
1
Departme
n
t of Computer Sci
ence a
nd T
e
chnol
og
y, Xi
ame
n
Univ
ersit
y
of T
e
chnolog
y,
Xi
amen, Ch
ina
2
Departme
n
t of Information C
enter, Xiame
n
Healt
h
Bure
au,
Xiam
en 3
6
1
0
0
5
, Chin
a
3
KOLLZ
OK Intelli
ge
nce Sci
e
n
c
e and T
e
chno
log
y
C
o
., Ltd
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: 4931
07
149
@
qq.com
A
b
st
r
a
ct
Multi-core tec
hno
logy
is the key techn
o
l
ogy
of
mob
i
l
e
clou
d co
mputin
g, w
i
th the b
o
o
m
development of cloud tec
h
nology, the
authors focus on
the pr
oblem
of
how to
make the target code
computed by m
o
bile
c
l
oud
t
e
rminal multi-c
o
re
c
o
m
p
il
er t
o
us
e cloud
multi-core
system
c
onstruction, to
ensur
e sync
h
roni
z
a
t
i
on
of d
a
t
a cross-val
i
dat
ion c
o
mpi
l
ati
o
n
,
and
prop
ose t
he co
nce
p
t of
end
mobi
le c
l
o
u
d
entity ind
i
rect s
y
nchro
n
i
z
a
t
i
on
and d
i
rect syn
chron
i
z
a
ti
o
n
; u
s
e w
a
ve ormation e
ner
gy con
v
ersio
n
, give o
u
r
a
meth
od to
c
a
lcul
ate
ind
i
re
ct synchro
ni
z
a
tion v
a
lu
e a
n
d
direct sy
nchr
oni
z
a
t
i
on
val
u
e acc
o
rdi
ng to
th
e
cross exper
ien
c
e and cross
time of co
mpil
ation e
n
tity
; construct functi
on rel
a
tive lev
e
l al
gorith
m
w
i
t
h
Hell
in
ger dista
n
ce, and g
i
ve
an alg
o
rith
m meth
od of compre
hens
ive
synchron
i
z
a
ti
on val
ue. T
h
rou
g
h
exper
iment st
a
t
istics an
d
ana
l
ysis, take t
h
res
hol
d l
i
m
it
v
a
lu
e
as th
e
aver
ag
e, self-sync
h
ro
ni
z
a
ti
on v
a
l
u
e
as
devi
a
tion, th
e upd
ate functi
o
n
of ind
i
rect sy
nchro
n
i
z
a
t
i
on v
a
lu
e is co
nstru
c
ted; an i
n
ter-
do
ma
in
mu
lti-core
synchro
ni
z
a
ti
o
n
flow
chart is
give
n; then
int
e
r-do
m
ain
c
o
mpilati
on data
sy
nchro
n
i
z
a
t
i
on u
pdate exp
e
ri
ment
is carried
out w
i
th more tha
n
300
0 en
d mobil
e
clo
ud
mu
lti-core co
mpil
ation e
n
viro
n
m
ent. T
h
roug
h t
h
e
ana
lysis
of dat
a co
mp
ilati
on
oper
ation
proc
ess an
d res
u
lt
s, the synchr
o
ni
z
a
ti
on
alg
o
rit
h
m is
prove
d
t
o
b
e
reaso
nab
le a
n
d
effective.
Ke
y
w
ords
:
mobil
e
c
l
ou
d c
o
mp
utin
g,
multi-
core tec
h
n
o
l
o
g
y
, mu
lti-core
c
o
mpil
atio
n, sy
nchro
nous,
cro
ss
valid
atio
n
Copy
right
©
2013 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
The
develop
ment of
clo
u
d
co
mputin
g i
s
not limit
e
d
to
the P
C
, with
the mo
bile i
n
ternet
of
boomin
g dev
elopme
n
t, ba
sed o
n
a m
obile termi
n
a
l
of the clou
d com
puting
servi
c
e
s
ha
s
emerged. Ba
sed
on
the
d
e
finition of
cl
oud
com
put
in
g, mobile
co
mputing
clo
u
d
s i
s
th
rou
g
h
the
mobile net
wo
rk with o
n
-d
emand, ea
sy
to expand way to obtain the infrastructure, platform,
softwa
r
e
(o
r
appli
c
ation
)
a
s
well
a
s
a
I
T
reso
urce
s or (info
r
matio
n
) se
rvice
d
e
livery
an
d
u
s
e
mode. Mobil
e
cloud
comp
u
t
ing is clo
ud computing te
chnolo
g
y in mobile Intern
et appli
c
ation [1
].
Multinucl
e
a
r
comp
uting te
chn
o
logy i
s
the key tech
n
o
logy of mo
b
ile com
puting
clou
ds,
The termin
al
entity of mobile clo
ud
pool can
be
a mononu
cl
ear comp
ute
r
or multinu
c
lear
comp
uter, th
e wh
ole m
o
bile cl
oud
po
ol is
a movi
ng cl
oud
co
mputing te
rm
inal multin
ucl
ear
platform [2
-3
]. "Mobile M
e
" roll
ed
out
by Appl
e, Inc. i
s
a
clou
d ba
se
d o
n
mobile te
rmi
nal
multinucl
e
a
r
cal
c
ulatio
n
a
nd
mo
bile cl
oud storage
solutio
n
s. A
c
cording
to A
pple'
s
whol
e
idea,
the prog
ram
lets mobile
cloud term
inal sy
ne
rgi
s
tic treatment
E-mail, notepad p
r
oje
c
t,
comm
uni
cati
on bo
ok,
ph
otos a
nd
oth
e
r file
s at th
e sa
me time,
use
r
s would
do thin
gs to
be
automatic int
e
ra
ction, u
p
d
a
te to iM
ac, i
P
od, iPhon
e and all kind
s of
mobile
terminal
inte
rfa
c
e
[4]
by Apple Inc. An importa
nt link is mu
ltinucle
ar co
mpiling of th
e mobile
clo
ud multinu
c
l
ear
environ
ment
informatio
n
intera
ction.
The imp
o
rt
ant goal
of multinucl
e
a
r
compili
ng i
s
the
multinucl
e
a
r
of the synergi
stic inte
ractio
n and multinu
c
lea
r
Credi
bility compiling.
The compile
r cre
d
ible g
o
a
l
is from the
point of view of the ope
ration of the
system
guarantee
co
mpiled th
e credibility of t
he softwa
r
e,
He Ya
nXia
ng p
r
ofe
s
sors an
d the te
am
membe
r
s to
compile fram
e
w
ork an
d pa
rt cre
d
ible
the
o
r
etical
details of the syste
m
summ
ary a
n
d
define [1], the main in
clu
d
ing two me
aning
s: on
o
ne ha
nd
we
must en
su
re
that the com
p
iler
itself is credible, the compil
er itself credi
bility is
mainly refers to the co
rrectness
of the translat
i
on
pro
c
e
ss.
First of all, the
co
mpiler fu
nctio
n
mu
st
be
co
rrect, that i
s
th
e so
urce
co
d
e
and
compil
ed
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NIKA
Vol. 11, No
. 9, September 201
3: 526
7 – 5276
5268
cod
e
on
beh
avior e
quival
ence with; S
e
co
nd, en
su
re
that the
co
mpiler i
n
the
build p
r
o
c
e
s
s will
not to bring a
n
y compute
r
system
security probl
em
s, prevent mali
cious atta
ck b
y
modifying the
compil
er, d
u
ri
ng
compil
atio
n exe
c
utive
malicio
us co
d
e
. On th
e oth
e
r h
and,
we
must e
n
sure that
the
compiler compil
ed obj
ect code
i
s
credible,
that
is, through the
compil
er
provide
credibility to
the gu
ara
n
te
e sy
stem, mu
st en
su
re th
at all throug
h t
he
compil
atio
n, testing
and
verificatio
n
a
nd
the su
ccess
of the genera
t
i
on of prog
ra
m code, in th
e system plat
form is runnin
g
on a safe a
nd
reliabl
e, at least, is sche
du
led to meet the requi
rem
e
n
t
s of the faithful. To meet the requi
rem
e
n
t
s
of the above two a
s
pe
cts
compile
r calle
d cre
d
ible
co
mpiler.
Espe
cially in the cloud te
chnolo
g
y the prospe
rou
s
de
velopment of the backg
rou
nd, the
mobile
te
rmi
nal
p
a
rall
el compil
ation, moving clou
ds clo
ud co
mpile
n
e
cessary and
m
obile
terminal di
stributed clo
ud
terminal mo
re nucle
ar co
mpiler techn
o
logy is wid
e
ly fusion, more
nucl
ear compiled and
cloud computing
will
show
the vitality of the speci
a
l together.
Com
pared
with the t
r
adi
tional compil
er, mo
bile
cl
oud te
rminal
to co
nsid
er
moving mo
re
nu
clea
r p
r
og
ram
clou
d termin
a
l
of the synch
r
oni
zation, m
u
tually
exclu
s
ive, data com
petiti
on, the d
ead lo
ck, lo
ad
balan
ce,
etc,
make
the
obj
ect
cod
e
com
p
iler
out to
m
a
ke
full u
s
e
of mobile
clo
ud m
u
ch of t
he
system stru
ct
ure,
gu
ar
ante
e
nu
clea
r sy
nch
r
oni
ze
dat
a co
nsi
s
ten
cy requi
reme
n
t
s [2] and m
a
ke
more
und
er t
he envi
r
onm
ent of nu
cle
a
r
cl
oud
credi
ble compil
ed
into a ve
ry
valuable
and
the
chall
enge of t
he su
bje
c
t.
Information
compil
ed by the mobil
e
clo
ud termin
al credibl
e multi-core cro
s
s-va
lidation
pro
c
e
s
s e
s
sentially the
same
verifi
cation p
r
o
c
e
s
s a
nd truste
d computin
g
grid
inform
ation
excha
nge, credible
m
u
lticore co
m
p
iled
synchro
nou
s lin
k me
sh
mobile
clou
d
entities to
work
together,
To
the verificatio
n
of inte
ra
ctive info
rm
atio
n [3-1
0]. In this
pape
r, th
e eme
r
gin
g
cloud
techn
o
logy di
stributio
n to verify the calculatio
ns, the
trusted gri
d
cro
s
s-validat
ion cal
c
ulatio
n
method i
s
in
trodu
ced i
n
to
the mobile
cloud te
rmi
nal credi
ble
multi-core
co
mpile proble
m
,
Propo
se
d an
d solve the
probl
em of h
o
w to u
s
e hi
story to com
p
ile expe
rien
ce to calcula
t
e a
cre
d
ible
co
m
p
iler
synchro
n
izatio
n valu
e, the credi
bility of self-sy
n
ch
roni
zatio
n
value, the a
gent
indire
ctly tru
s
ted syn
c
hroni
zation val
ue f
unctio
n
sets,
Cro
s
s-vali
dati
on process to
form a m
obile
cloud terminal credible mul
t
i-core envi
r
onment, t
hus
achi
eving the key step
s of
the credibility of
multico
r
e syn
c
hroni
zation
compil
ed [11,
12].
2. Credible Compilation Interaction Model Desc
rip
t
ion base
d o
n
Multi-cor
e
Mobile Clou
d
Clou
d com
p
u
t
ing take virtu
a
lizatio
n tech
nology to define and di
stin
guish hardwa
r
e and
softwa
r
e mo
bile entities i
n
the pool, mobile
cl
oud
entities hav
e been defin
ed from diffe
rent
pers
p
ec
tives
,
s
u
c
h
as
: Grid and unit. In the
pe
rspe
ctive of
credi
ble
com
p
ilation, the
botto
m
hard
w
a
r
e
pla
tform of the
clou
d po
ol
can b
e
bu
ilt
b
a
se
d o
n
tru
s
ted technol
o
g
y, all elem
e
n
ts
colle
cted by software
comp
ilation in the mobile cl
oud
pool is multi-core mobil
e
cl
oud co
mpilati
on
mobile clo
ud entity,
if
the
mobile clo
ud colle
cti
on of entities need
to
be
fully cre
d
ible,
the upp
er
colla
boration
compil
ation
sho
u
ld b
e
cre
d
ible, wh
ile the key point of
mobile
cloud entities
should
be compile
d
synchro
nou
s,
the problem
depe
nd
s on
the formul
ate of the mo
b
ile clo
ud e
n
tity
intera
ction
co
mpile
agreem
ent an
d th
e v
e
rificatio
n
of i
n
tera
ctive inf
o
rmatio
n [13
-
16], the l
a
tter is
also
an imp
o
rtant foundatio
n of the form
er on
e.Th
i
s
p
aper
of syn
c
h
r
ono
us i
n
tera
ction validatio
n
of synch
r
oni
zation usi
ng th
e followin
g
de
finition model
:
Synchrono
us (o
r n
o
t syn
c
hron
ou
s) is e
v
aluation
sev
e
ral
mobile
e
n
tity cond
uct
ed joi
n
tly
compil
ed a
cl
oud the
beh
a
v
ior of the o
b
j
e
ctive po
ssibi
lity degree, th
e evaluatio
n
of the co
mmo
n
behavio
r in
compile m
onit
o
r
(or impo
ssible to m
onito
r the
beh
avio
r)
before a
n
d
the compil
ation
and be
havio
r in its own influen
ce be
havior. Mo
ve
the relatio
n
s
hip b
e
twe
e
n
clou
ds
ent
ity
synchro
n
ization can be d
i
vided
into di
rect and
indi
rect
syn
c
h
r
o
n
izatio
n syn
c
hron
ou
s of t
w
o
types: dire
ct synchro
n
ization is a movi
ng clo
ud ent
it
y acco
rdin
g to the past ea
ch othe
r'
s direct
intera
ction
a
nd d
e
termi
n
e
to compile
experie
nc
e o
f
other mobil
e
cl
oud
entit
y synchro
n
o
u
s
degree; Indirect syn
c
h
r
oni
zation i
s
a m
o
ving clo
ud b
a
se
d on pa
st
the compil
ation of the enti
t
y
results an
d d
e
termin
e to recom
m
en
d the other
m
o
bile clo
ud en
tity recomme
ndation
s
of the
synchro
n
ization of informat
ion co
m
p
iled
experie
nce le
vel [17, 18].
A mobile
clo
ud entity hist
ory syn
c
h
r
oni
zati
on
of the
clou
d is
movi
ng entity a b
ehavior
compil
ed ex
pectatio
n
s, b
a
se
d on the
expected in
a spe
c
ific ti
me and m
a
n
y
other nu
cl
ear
environ
ment
for the m
obil
e
entity mobi
le clo
u
d
s
clo
ud entity in t
he pa
st the
behavio
r of t
h
e
compil
er o
b
servation. If mobile cl
oud
e
n
tity a need
to make sure
to a dire
ct int
e
ra
ction
with
no
experie
nce o
r
in quite a
lo
ng time n
o
di
rect i
n
te
ra
ctio
n
expe
rien
ce mobile clou
d the
co
mpilati
on
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Mobile Com
puting Clo
u
d
s
Intera
ctive M
o
del and Algo
ri
thm
based on
Multi-co
re…
(Liu Li
zha
o
)
5269
of the
entity b syn
c
h
r
o
n
o
u
s
deg
ree,
can a
c
co
rdin
g
to indi
re
ct
synchrono
us relation
ship
m
o
re
comp
re
hen
si
ve other nu
cl
ear
mobile
cl
oud of
mobi
l
e
clo
u
d
entity b entity indi
rect
synchro
n
o
us
informatio
n that clou
d the
history of m
obile
entity b
synchro
n
ou
s value, and
according to
the
history
synch
r
ono
us valu
e to take a co
m
p
iler be
havior [19-20].
Interactive
co
mpilation exp
e
rien
ce
can
deci
de a co
mpiler mo
bile
cloud e
n
tity
or mo
re
than a nucl
e
ar co
mpiled t
he deg
ree of
region
al
syn
c
hroni
z
ation. Synchr
ono
us value accord
ing
to the intera
ctive experi
e
nce
and
ca
n
last from
th
e interval
wh
en evalu
a
tin
g
intera
ction
b
y
function
, One
x repre
s
ent
s experie
nce,
τ
repres
ent
s
the time value.
One x rep
r
e
s
ents expe
rien
ce,
τ
r
e
pr
es
en
ts
th
e
time
va
lu
e
.
W
h
en
mo
vin
g
a
c
l
ou
d
w
i
th
this mu
ch
ent
ity to nucl
ear
domain
mobil
e
cl
oud
ent
ity intera
ct b,
m
obile
entity of a
clou
d ba
se
d
on previo
us
synchro
nou
s
evaluation fu
nction first
of the storag
e of inte
ractive
experien
ce
and
finally a and
b interactio
n to the time now
cal
c
ul
ate to b dire
ctly synch
r
on
ization of val
u
e
And then mo
ve to a more
cloud
s
clou
d
entity nucle
ar re
gion
al synchrono
us
a
g
ent for
inquires the h
i
story of the e
n
tity
b mobile cloud
synchronou
s value.
Acco
rdi
ng to
a re
ce
nt syn
c
hrono
us ag
ent for
this domain S
in t
h
e other
mobile
cloud
entity
and
b
intera
ction experie
nce cal
c
ulation
i
∈
S,
Then
th
e history of
comp
uting
b synchro
nou
s value:
ri For syn
c
h
r
onou
s age
nt is moving clou
ds
of in
dire
ct sync
hronou
s value
entity I.
Synchrono
us agent will
history
synchron
ou
s val
u
e a
cco
rdi
n
g to b use
type updat
e all
recomme
nde
d the indire
ct synchro
nou
s
movement cl
oud entity value:
Among them
is the upd
ate
d
sync
hro
nou
s age
nt for m
obile cl
oud of
indire
ct syn
c
hron
ou
s
value entity I,
, function
θ
(Sub) i
s
e
x
ponential d
r
ab dimini
shin
g function
s,
function
η
(S
ub) i
s
expon
e
n
tial mon
o
ton
ous in
cre
a
si
n
g
fun
c
tion,
σ
as th
e a
c
cept
ed recomme
nd
data erro
r th
reshold. M
o
b
ile clou
d enti
t
y for a mo
bi
le clou
d the
history of th
e entity b after
synchronous
value will be
used to
cal
c
ul
ate the type to b compre
hensive synchro
nous value:
, On be
half o
f
a Tab to
b
of the comp
rehen
s
ive
syn
c
hrono
us
value, and al
pha rep
r
e
s
en
ts self
synch
r
ono
us va
l
u
e
,
the scope f
o
r [0, 1], by many nu
clea
r
mobile cl
oud
the entity initially sure,
and al
pha
in the value of a calculated de
cisi
on
comp
re
hen
si
ve synchrono
us
whe
n
di
re
ct expe
rien
ce
and f
r
om ot
her m
obile
cl
oud of i
ndire
ct
experie
nce e
n
tity weight o
f
the match.
Mobile
cl
ou
d
entity accordi
ng to the val
ue of the T
a
b a
will make the final decisi
on to sync directly affe
cts i
t
s in the translatio
n process of
interact
ive
compil
ation b
ehavior.
3. Wav
e
Ene
r
g
y
Information Theor
y
,
t
h
e Int
roducti
on and Analy
s
is
of Hellinger Distance
Literatu
re [21
,
22] of mobile cloud e
n
tity
model analy
s
is a
r
e given
a kind when
sele
cted
area
cal
c
ul
ation on
e mobil
e
clo
ud entity
is the infl
u
e
n
c
e of the
sum
function, d
e
fined a
s
volatili
ty
informatio
n e
nergy
fun
c
tion; And
at the
sa
me time
in th
e
ar
ea
ar
e th
e
c
a
lc
u
l
a
t
io
n
o
f
s
o
me
lo
c
a
l
area h
a
s the
influen
ce and
the propo
rtio
n of t
he total influence
cal
c
ulatio
n meth
od, defined a
s
fluctuation
s
e
nergy than i
n
formation; An
d explai
n the
function of
mutual co
nne
c
tion de
gre
e
b
y
function
s rel
a
ted informatio
n said.
f(x
,
)
ib
i
b
Sub
R
t
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NIKA
Vol. 11, No
. 9, September 201
3: 526
7 – 5276
5270
Theo
rem 1: for eithe
r
a given functio
n
f (s), if it can be
expresse
d a
s
:
Whe
r
e
. For
contin
uou
s di
fferentiable f
unctio
n
,
the functio
n
of th
e fluctuatio
n
of energy
transfo
rmatio
n for informati
on:
Wave en
ergy
than informat
ion for:
Theo
rem 2: set of a system for physi
ca
l transfe
r fun
c
tion
, the transfe
r functio
n
for identificati
on
, {S1, S2... Sn} for test sample
s, if:
Among them
,
as the fitting deviation ratio, Then:
Theo
rem 3
Functio
n
I ( S ) and M ( S ) of the correlat
ed deg
r
ee i
s
;
Func
tion I ( S ) in
M ( S ) under the co
nditions of the correl
ated de
gree i
s
,
If
M ( S ) is
a
multivalued functio
n
, said
.
Theo
rem 4:
Hailin gri
d
di
stance i
s
abl
e to reflect the tw
o di
stributions
between simil
a
rity m
e
trics [4,
16]. Hypothe
sis in
metri
c
spa
c
e
,
P and Q represent re
spe
c
tively the co
rre
spo
n
din
g
para
m
eters of
two co
ntinuou
s di
stri
b
u
tion, so T
h
e two
di
strib
u
tion b
e
twe
e
n
the
Helli
n
ger
distan
ce i
s
de
fined as
:
Equivalent to
For counta
b
l
e
spa
c
e, Tha
t
is when the
distri
butio
n is discrete, two distrib
u
tion
betwe
en
the Hailin g
r
id
distan
ce can
be defined a
s
:
1
()
(
)
i
n
i
f
s
f
s
()
i
f
s
1
2
1
()
(
)
()
2
ii
i
n
i
fs
s
s
Ef
s
&
g
1
[(
)
]
[
(
)
]
li
m
[(
)
]
ik
i
i
K
nK
EF
S
E
F
S
tE
EF
S
()
WS
()
GS
()
()
()
J
MP
J
M
P
J
K
SM
P
MP
J
GS
WS
E
WS
M
p
11
1
()
(
)
()
()
()
()
M
PM
P
nn
MP
i
M
P
i
MP
J
M
P
J
ii
n
MP
J
MP
i
i
GS
WS
GS
WS
WS
WS
[(
)
,
(
)
]
I
SM
S
[(
)
|
(
)
]
IS
M
S
1
[(
)
|
(
)
]
n
i
i
IS
M
S
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Mobile Com
puting Clo
u
d
s
Intera
ctive M
o
del and Algo
ri
thm
based on
Multi-co
re…
(Liu Li
zha
o
)
5271
We can obtai
n the Hailin lattice dist
ance
has the following properties:
Helling
e
r di
stance in
the
simila
rity of
data
sets tha
n
the
Eucli
d
ean
dista
n
ce
ha
s th
e
advantag
e on
the form and
meanin
g
, Ci
esla
k et al
, [4] use it to design the corre
s
pondi
ng de
si
gn
deci
s
io
n tre
e
algorith
m
. [4]
and
analy
z
ed
and
verified
the effe
ctiven
ess of th
e di
stance
m
e
a
s
u
r
e
to sele
ct the
approp
riate f
eature
s
, Esp
e
cially in th
e i
m
balan
ce
d d
a
ta; Lee et
al [5] to apply it on
t
he A
d
jace
nt
cla
ssif
i
e
r
.
4. Validation Model and Algorith
m Design
of Multi-c
o
r
e
Compiled
Sy
nchronous
Interaction w
i
thin
Domain
3.1. Indirect Upda
te Fun
c
t
ion and Self Sy
nchronous Value
Acco
rdi
ng to
the theo
ry
of fluctuation
s
info
rmation
ene
rgy we
rai
s
e the
followin
g
rea
s
oni
ng:
Rea
s
o
n
ing.1:
Suppo
se
tha
t
move the m
obile
clou
d e
n
tity A and m
obile
clou
d e
n
tity B in
the domai
n A to gain experie
nce
duri
ng the
previous di
re
ct in
teractio
n, every
intera
ction from the
cu
rre
n
t cal
c
ulatio
n
time
,
,
for the histo
r
y o
f
mobile
clou
d
entity A mobile clou
d entity B synchroni
zation value
.
Note: whe
n
you move wit
h
in the regio
n
A
cloud en
tity b
b evaluation of mobile cloud
entity a hi
story of interactiv
e expe
rien
ce
in
a, whi
c
h
is equivalent
to b
to apply exce
ssive
impa
ct
of a region th
e sum of. Every time interaction a
s
so
ci
ated with the distan
ce fro
m
the current time
interval, with the further e
x
perien
c
e of
intera
cti
on, the histo
r
ical experie
n
ce o
f
the impact will
decay, so
f d
e
rivative fun
c
tion on
be
hal
f of the atten
uation fu
nctio
n
can
be
co
mpiled
und
e
r the
different m
u
lti-co
re
enviro
n
ment
settin
g
. Re
asoni
n
g
,
are
a
A
syn
c
hroni
zation
agent
on th
e
i-th
mobile
cloud
entity b indire
ct syn
c
hroni
zation
erro
r for an acce
ptabl
e error
thre
sh
old for
σ
is self-syn
ch
ronization valu
e of alpha, t
he indire
ct syn
c
hroni
zation
update al
gorit
hm:
Description:
When
or
will have the overall evaluation of the
synchro
n
ization ag
ent with
in mobile
clo
ud entity
b, hi
gher than th
e
agent'
s
o
p
timal expe
ctati
ons
may be provi
ded by the a
gent dire
ct sy
nch
r
oni
zatio
n
value is too low, are m
o
re
likely to entities
in the mobile
cloud mi
sju
dgment
s or
update to in
t
e
ra
ct with b not timely Centre, whi
c
h
will
indire
ctly affect the indi
rect simultane
ou
s a
s
se
ssme
n
t
of a to b; thre
shol
d for
the two case
s
reflect
the
ag
ency
and
the
re
st of
the
mobile
clo
u
d
entities
on th
e mo
bile
clo
u
d
entity the
final
evaluation of
b deviates fro
m
the optimal
assess
me
nt. Deviation fro
m
the asse
ssed value of the
update
of the impa
ct of i
ndire
ct
syn
c
h
r
oni
zation val
ue
from the
mobile
clou
d
entity a self-
synchro
nou
s value, to ju
dge the imp
a
ct of t
he size of error, so alph
a ca
n be expre
s
sed
deviation
s in the normal di
stribut
ion fun
c
tion, the updat
e pro
c
e
ss.
When
there
is no
thre
sh
o
l
d value
co
n
s
traint,
we fo
und in
the e
x
perime
nt ba
sed
on
updatin
g alg
o
r
ithm for i
ndirect syn
c
h
r
o
n
i
z
ation
of t
he
t distrib
u
tion
error i
s
the l
a
rge
r
value
ha
s
12
{,
}
n
x
xx
12
{,
}
n
2
1
1
()
(
)
(,
)
2
n
ii
i
i
fx
x
hi
s
t
or
y
a
b
i
Sub
2
2
2
2
()
1
ex
p
,
2
2
,
()
1
ex
p
,
2
2
i
ii
ii
i
i
ii
Sub
r
Sub
r
r
Sub
Sub
r
Sub
i
Sub
i
Sub
i
r
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NIKA
Vol. 11, No
. 9, September 201
3: 526
7 – 5276
5272
better co
nvergen
ce effect, can reflect the moving
clo
uds of gre
a
te
r self-synchro
nizatio
n
and the
synchro
n
ization age
nt syn
c
hroni
zation
entities re
co
mmend
ed rol
e
con
s
tru
c
t are as follo
ws:
Also find up
dates
ba
sed
on
distributio
n functio
n
ca
n be in
error
whe
n
sm
alle
r rapid
res
pon
se
ef
f
e
ct
of
self
-
s
y
n
ch
ron
o
u
s
sy
nch
r
oni
zatio
n
error and
re
comm
end
ed values, stru
ct
ure
is
as
follows
:
Rea
s
o
n
ing 3.
If a and b repre
s
e
n
t two
discrete poi
nt set spa
c
e,
its relevan
c
e
,
with Hellinger (A,B) say
s
.
De
scription:
If function
and respe
c
tively
rep
r
e
s
en
tative two a
spa
c
e
discrete
poin
t
s set of ra
nge, s for d
e
fined dom
ai
n, its related
of with Hel
linger di
stan
ce
cal
c
ulatio
n Shi, distan
ce
more l
a
rg
e is related
of
m
o
re
small, p
o
i
n
ts set and it
self of di
stan
ce fo
r
zero an
d relat
ed of for i
s
m
a
ximum; co
n
d
itions
relate
d of ca
n with
two a p
o
ints
set of differen
c
e
s
and which a points
set of Helling
e
r di
st
ance t
han Shang two a p
o
ints set Zhijian of Hellin
g
e
r
distance to calcul
ation, Points
set by reduction letter to Hellinge
r
distance f
r
om
the ori
g
in and
the point set
above an
d eq
ual to the dist
anc
e of the Hellinge
r dista
n
ce b
e
twe
e
n
and
.
)]
(
),
(
[
)]
(
),
(
[
S
M
S
I
Hellinger
S
M
S
I
]
)
(
),
(
[
}
)
(
],
)
(
)
(
{[
]
)
(
|
)
(
[
1
1
1
1
n
i
i
n
i
i
n
i
i
n
i
i
S
M
S
I
Hellinger
S
M
S
M
S
I
Hellinger
S
M
S
I
]
)
(
),
(
[
}
)
(
,
{
]
)
(
|
)
(
[
1
1
1
1
n
i
i
n
i
i
n
i
i
S
M
S
I
Hellinger
S
M
O
Hellinger
S
M
S
I
Rea
s
o
n
ing 4.
If domain A mobile clou
d entit
y a direct syn
c
h
r
o
n
izatio
n withi
n
other
mobile
cloud
entity value
, other mobil
e
clou
d entiti
e
s on th
e m
obile cl
oud e
n
tity a direct
synchro
n
ization value
, the proxy dire
ctly to ot
her mobile clo
ud e
n
tity synchro
n
izatio
n value
, of a self-syn
chroni
zation
value
.
De
scription: After
simul
a
tion
exp
e
rime
n
t
,
an
d
Can
be
used
as
α
i
n
the
initial calcul
ation and
re
al-time up
dat
e algo
rithm. Here a
s
a
n
example, accordin
g to the Rea
s
o
n
ing 3.
)
,
Hellinger(
]
),
(
Hellinger[
]
|
[
1
1
1
1
1
1
1
n
i
mi
n
i
ai
n
i
mi
n
i
mi
n
i
ai
n
i
mi
n
i
ai
t
t
t
t
t
t
t
1
2
1
()
2
`(
1
)
()
2
i
ii
Su
b
rr
2
x
1
22
2
1
`
2(
)
2
i
Su
b
ii
i
rS
u
b
e
r
[,
]
A
B
()
I
S
()
M
S
()
IS
()
M
S
ai
t
i
a
t
mi
t
ai
m
i
[t
|
t
]
ai
m
i
[t
|
t
]
ai
i
a
[t
|
t
]
ai
m
i
[t
|
t
]
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Mobile Com
puting Clo
u
d
s
Intera
ctive M
o
del and Algo
ri
thm
based on
Multi-co
re…
(Liu Li
zha
o
)
5273
3.2. Inter-do
main Multi-c
o
re Compile
Sy
nc
hronou
s Cros
s-v
a
lidation Mod
e
l and Algorithm
Process
5. Model par
a
meter ev
aluation algorithm
Step1: Sets the e
r
ror thre
shold val
ue
an
d inte
rval for
update
fun
c
tions, yo
u
can
cho
o
se
to set up self
-synchronous
(In long
-term
stability can be selected).
Step2
:
Set the time decay
function, com
puting
hi
story
and experi
e
n
c
e the differe
nce, to
be evaluatin
g
mobile clo
u
d
computin
g a
nd histo
r
y sy
nch
r
oni
zatio
n
direct
synchronization val
u
e
value of the entity. Calculat
ion
re
sults if you wa
nt one
of the f
ourth step, or go to third ste
p
.
Step 3: colle
ct the indi
re
ct synch
r
oni
za
tion
syn
c
hron
ous
age
nt value an
d ea
ch
mobile
clou
d entity to move to
e
v
aluate the
synchrono
us
value di
re
ctly clou
d entity li
st, at the
sa
me
time, apply for each mobile
cloud of dire
ct synchro
n
ization of entity value list. If s
e
lf synch
r
on
o
u
s
value ha
s set
the turn step
5, or turn ste
p
4.
Step 4: a
c
cording to t
he th
ird
step
data
cal
c
ul
atio
n a
nd
set the
sel
f
synchro
nou
s valu
e,
data normali
zation. Turn th
e third step.
Step 5: set of
the values
of the indirect syn
chroni
zatio
n
update fu
nction (if the re
straint
s
on the limit attack
can
sele
ct different up
date functio
n
).
Step 6: calculation of the co
mp
reh
e
n
s
ive synch
r
on
ous valu
es (and tran
sfe
r
to other
deci
s
io
n-m
a
ki
ng functio
n
)
update all m
obile cl
oud o
f
synch
r
on
ou
s value an
d
agen
cy entity
recomme
nde
d value, turn the first ste
p
.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NIKA
Vol. 11, No
. 9, September 201
3: 526
7 – 5276
5274
Data er
ror
threshold
initial
i
zatio
n
Histor
y
upd
ate t
he data interval is set to 0.5 time units
The
first
roun
d
of
updates
The second
round of u
pdate
....
i
=
850
round of
i = 851 round of
Multi-
cor
e
label
r1
t1
R2
T2
....
R850
T850
R851
T851
1
6
....
2013
4005
0.50000.53
980.57930.
61790.655
4 0.6915
....
0.98930.99
18 0.99380
.99530.996
50.99740.9
981
0.5040
0.5438
0.5832
0.6217
0.6591
0.6950
....
0.9896
0.9920
0.9940
0.9955
0.9966
0.9975
0.9982
0.5080
0.5478
0.5871
0.6255
0.6628
0.6985
....
0.9898
0.9922
0.9941
0.9956
0.9967
0.9976
0.9982
0.5120
0.5517
0.5910
0.6293
0.6664
0.7019
....
0.9901
0.9925
0.9943
0.9957
0.9968
0.9977
0.9983
.... 0.5239
0.5636
0.6026
0.6406
0.6772
0.7123
....
0.9909
0.9931
0.9948
0.9961
0.9971
0.9979
0.9985
0.5279
0.5675
0.6064
0.6443
0.6808
0.7157
....
0.9911
0.9932
0.9949
0.9962
0.9972
0.9979
0.9985
0.5319
0.5714
0.6103
0.6480
0.6844
0.7190
....
0.9913
0.9934
0.9951
0.9963
0.9973
0.9980
0.9986
0.5359
0.5753
0.6141
0.6517
0.6879
0.7224
....
0.9916
0.9936
0.9952
0.9964
0.9974
0.9981
0.9986
Subi
0.6950
0.0595
0.0555
0.0518
0.0484
0.0452
....
0.0422
0.0210
Ri
2.8784
2.8609
2.8453
2.8313
2.8187
2.8073
....
2.7969
2.7874
6. Applicatio
ns of multi-c
o
re s
y
nchronous compil
e model
Setting simul
a
tion p
r
og
ra
m: A domai
n
nucl
ear comp
iled mo
re th
a
n
the n
u
mb
er of 300
0
mobile cl
oud
entity,
that is equivale
nt to hav
ing 30
00 com
p
iled
with nu
clea
r compil
ed; Da
ta
compil
ed
error threshold ini
t
ialization
σ
= 0.550
0, syn
c
hron
ou
s
re
co
mmend
up
dat
e ba
se
d o
n
th
e
function of no
rmal di
stributi
on co
nstructo
r,
.
Time unit i
s
three time
s
resp
ecti
vely set to
. Computing mobile
clou
d entity synchroni
zatio
n
with anoth
e
r
mo
ve in any
cloud e
n
tity
2011 hi
story
value.
Rea
s
o
n
ing o
ne, history sy
nch
r
oni
zatio
n
value re
sults in the settings of the different time
units. Thi
s
time for the
0
.
7833, Di
re
ct and
simulta
neou
s value
of 0.9896
(Space i
s
limi
t
ed
empiri
cal
dat
a, see S
c
he
d
u
le); Th
ese two valu
es
as the first rou
nd of up
date
synchro
n
ization
initial value, a
s
the a
bove t
able t1 a
s
sh
own i
n
;
4 too
,
base
d
on
re
aso
n
ing, Self-synchro
nization
value of 0.69
50
,
set the in
itial value in
orde
r to
spe
e
d
up the
up
d
a
te rate
0.7,
update
d
in
re
al
time synchro
n
izatio
n valu
es, upd
ate Agent dire
ct synchroni
zatio
n
value for
all mobile
cl
oud
entity first ro
und of in
dire
ct syn
c
h
r
oni
zation value
s
r1 and
mobile
clou
d entity
2013 t1, Into
the
se
con
d
roun
d
of
cal
c
ulatio
ns. Sp
ecifi
c
rene
wal
pro
c
e
s
s an
d the
d
a
ta sho
w
n i
n
the ab
ove ta
ble,
The unit of time is S, the current upd
ate
to the 341.
Visible
wh
en
Subi redu
ce
hou
rs of mo
bile
ag
ent
cl
oud
of indi
re
ct syn
c
h
r
on
o
u
s val
ue
entity 2013 in
cre
a
se gradu
ally, indire
ct synchr
o
nou
s
value in the
update p
r
o
c
e
ss
and influ
e
n
ce
to mobile clo
ud 6 again a
nd other mo
bile entit
y to
the valuation of 2013 cl
oud entity value,
based o
n
the
mobile
clo
u
d
of dire
ct exp
e
rien
ce
value
on e
n
tity 2013 mo
bile
clo
uds i
n
several
roun
ds of up
d
a
ting entity with large syn
c
hron
ou
s valu
e; At the same time can
See mo
bile
cloud 6
to the
valuation
of
2013
entities and
othe
r m
obile
clou
d o
f
2013
evaluation e
n
tity synchro
nou
s increa
se or de
cre
a
s
e so at this mome
nt, according to
the
rea
s
oni
ng 4 t
o
mobile
clo
u
d
6 entity self
sync
hronou
s
value
in
the update pro
c
e
ss ba
sic rem
a
in
s
the same, in this mo
re nu
cl
ear a compile
d environm
en
t can be rou
g
h
set to fixed
value. Now th
e
big adj
ustme
n
ts to ag
en
cy found the
20
13 indi
re
ct
value when th
e
synchro
n
ization, mobile
cl
oud
6 entity of s
e
lf synch
r
on
ous value i
s
fluctuatio
n
s
,
so st
ri
ct
ly
spea
king s
h
o
u
ld maint
a
in
self
synchro
nou
s
value of the real-time u
pda
tes.
1
()
,
0
.
5
(
)
f
t
1,
1
,
1
t
s
tm
th
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Mobile Com
puting Clo
u
d
s
Intera
ctive M
o
del and Algo
ri
thm
based on
Multi-co
re…
(Liu Li
zha
o
)
5275
7. Conclusio
n
So accordi
n
g to the flu
c
tuation of energy
of information th
eory set up
by the
synchro
nou
s
cal
c
ulatio
n re
sults me
et th
e mo
bile
clo
u
d
termi
nal m
u
ltinucl
eated
trusted
compi
l
er
real time
syn
c
hroni
zation i
s
ne
ede
d, for in the
traditi
onal g
r
id
co
mputing a
s
a
fixed value
data
(su
c
h a
s
eq
u
i
valent self confiden
ce val
ue) c
an al
so
realize syn
c
hron
ou
s upd
ate so that the
update p
r
o
c
e
ss m
o
re a
c
cu
rate an
d perf
e
ct; at the
sa
me
time
can the
cla
ssi
cal mathemati
cs in
mature
algo
ri
thms a
nd m
o
re
grid
com
puting inte
ra
ctive algo
rith
m into mo
bil
e
cl
oud te
rm
inal
multinucl
eate
d
trusted
co
mpiler. The
same we al
so
see some li
mitations, su
ch a
s
is currently
unabl
e to find an effective transfo
rm d
a
ta error
thre
shol
d value u
pdate, and a
c
tually thre
sh
old
sho
u
ld be
co
nstru
c
ted
by fluctuation e
n
e
rgy ratio,
bu
t in the experiment with ot
her pa
ram
e
te
rs
matchin
g
, so
did not di
scu
ss;
we
co
nstruct the
i
ndire
ct syn
c
h
r
oni
zation up
date f
unctio
n
is
ba
sed
on a la
rge
numbe
r of
simulatio
n
and expe
rim
ent
al stati
s
tical results f
o
r different grid
environ
ment, sele
ct the ap
prop
riate up
d
a
te func
tion relies on exp
e
r
t experie
nce, future need
s to
establi
s
h a
co
mplete theo
re
tical fram
e
w
o
r
k throug
h de
rivation prove
s
that.
Ackn
o
w
l
e
dg
ments
This
work
wa
s su
ppo
rted
by: The natio
nal natu
r
al scien
c
e fou
n
d
a
tion (6
1070
151), the
national n
a
tural scie
nce
foundation
(609
032
03)
,
the national
natural sci
ence found
a
t
ion
(611
032
46),
and the n
a
tional natu
r
al
sci
en
ce f
oun
dation of Fuji
an (2
010
J0
1
353), the XM
UT
resea
r
ch foun
dation (YKJ1
1024
R),
XM scientific resea
r
ch fou
ndatio
n (350
2Z2
0
1
2303
7).
Referen
ces
[1] http://
w
w
w
.
cl
ou
d
w
h
y
.com/mi
n
gci/20
11/0
307/
91.html
[2]
Peng W
a
n
g
, Cl
oud com
puti
n
g
technol
og
y a
n
d
app
licati
on e
x
am
pl
es.
Posts & telecom
press.
2010; 1.
[3]
Micha
e
l M
ill
er. Cl
oud
C
o
mp
uting: W
e
b-Ba
sed
A
p
p
licati
o
ns T
hat Ch
an
ge th
e W
a
y
Y
ou W
o
rk
a
n
d
Coll
ab
orate On
line. QUE PUB
.
2009, 4.
[4] http://
w
w
w
.
appl
e.com/mobileme/
[5]
Yan
x
i
ang
He,
W
e
i W
u
, T
ao Liu, Qi
nga
n
Li, Min
g
h
ao H
u
, Jian
bo
Li
u,Qian Sh
i. T
heor
y an
d K
e
y
Impleme
n
tatio
n
T
e
chniqu
es
of T
r
usted Compiler: A S
u
rv
e
y
.
J
ourn
a
l of
F
r
ontiers
of C
o
mp
uter Sci
enc
e
and T
e
ch
no
log
y
.
1673-9
4
1
8
/2
011/0
5
(0
1)-00
01-2
2
[6]
Yan
x
i
ang H
e
, W
e
i W
u
,
T
ao
Liu. Res
earch
on the Ke
y T
e
chno
log
y
of T
r
usted Com
p
il
e
r
.
Comp
uter
Engi
neer
in
g & Scienc
e.
201
0; 32(8).
[7]
I Foster, C Kesselman eds.
T
he Grid: Blue
print for
a N
e
w
Comp
utin
g Inf
r
astructure. Sa
n F
r
ansisc
o
,
CA: Morgan K
aufman
n
. 199
9
.
[8]
F
Azzedin, M
Mahes
w
a
ra
n.
Evolving and
m
a
naging trust
in grid comput
ing system
s
. In
: Proceed
in
gs
of the IEEE Canad
ian C
onfer
ence o
n
El
ectri
c
al Com
puter
Engi
neer
in
g. 2002.
[9]
T
he Globus
Securit
y
T
e
a
m
. Globus to
olkit vers
io
n
4 gri
d
sec
u
rit
y
infrastructur
e
: a stan
dar
d
s
persp
ective. Versio
n 4 [S/OL]. [2005-09
-12].http:
∥
globus.org/toolkit/docs/
4. 0/securit
y
/GT
4-GSI-
Overvie
w
.pdf
[10]
Novotn
y J, T
uecke S, W
e
lch
V. An onl
ine cr
ede
nt
ial r
e
p
o
sit
o
r
y
for th
e gr
id:
M
y
Pro
x
y[R/O
L
]. [2001-0
8
-
01]. http:
∥
w
w
w
.
g
l
o
bus.org/
a
l
lianc
e/pu
blic
ati
ons/m
ypro
xy.p
df
[11]
Jian
C, W
a
n
g
C, L
i
u
X,
et
al.
A s
u
rvey
o
f
job
sch
edu
2
lin
g
in
gri
d
s.
Proce
edi
ngs
of t he
9t
h
Internatio
na
l Asia2P
acific W
eb Confer
ence
(APW
eb20
07).
Berlin, Germa
n
y
: Spri
ng
er2V
erla
g, 2007
:
419
24
27.
D
a
ta
e
r
ro
r
threshold
value
initial
i
zation
The num
ber
of update interv
als Hist
orical dat
a. Set to 0.5 time
units
t=1s
α
=0.7
0.3
Multi-
cor
e
label
1
2
3
4
5
7
.
...
1997
1998
1999
2000
Ti
a
0.8303
0.8216
0.8215
0.7924 0.7583
0.7521
.
...
0.0232
0.0210
0.0185
0.0103
Tai
0.9306
0.9237
0.9179
0.8964 0.8883
0.8623
.
...
0.0356
0.0317
0.0294
0.0240
Tmi
0.7950
0.7895
0.7851
0.7818
0.7784
0.7752
.
...
0.0421
0.0410
0.0402
0.0387
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NIKA
Vol. 11, No
. 9, September 201
3: 526
7 – 5276
5276
[12] Ab
w
a
j
y
j
H.
Fault2tolerant sc
heduling polic
y
for grid com
p
uting system
s.
Proceed
in
gs of t he 18t
h
IEEE International Parall
el and Distributed
Processing S
y
mpos
ium. Los
Alamitos, Cal.
, USA: IEEE
Comp
uter Soci
et
y
Press. 20
0
4
: 2382
24
4.
[13]
Yulo
ng
Liu, Y
u
and
a C
ao, Ji
an
Li. A
ne
w
t
y
p
e
of in
dir
e
ct s
y
n
c
hron
izatio
n m
ode
l.
Co
mpute
r
Engi
ne
erin
g
and Ap
plic
atio
ns
. No. 200
4 of 2004
[14]
Shan W
a
n
g
, Ying Yu
an, T
a
o
y
u
an Ch
en
g
,
Kunlon
g Z
h
a
ng. Dou
b
le s
y
nchr
o
n
izati
on
mode
l of the
beh
avior-
bas
e
d
service gr
id e
n
viro
nment.
Jo
urna
l of Co
mp
uter Appl
icatio
ns
. 2005; 2
5
(9)
.
[15]
Don
gan W
a
n
g
,
Hao Xu, Kai Nan, Bao
p
in
g
Yan. Grid computi
ng b
a
s
ed on the r
e
comme
nde
d
s
y
nc
hro
n
izati
o
n mode
l.
Appli
c
ation R
e
se
arch of Co
mputer
s
. 2006.
[16]
Yanj
un Gen, Hon
gqi Z
h
a
ng,
Jing Z
hou. Sync
hro
n
izati
o
n
model
bas
ed
on trusted computin
g gr
i
d
beh
avior.
Jo
urnal of Co
mpute
r
Researc
h
an
d Deve
lop
m
ent
. 2008.
[17]
A Abdul-
R
ahm
an, S Hail
es.
Supp
ortin
g
T
r
ust in Virtual C
o
mmuniti
es
. In: Proceedi
ngs
of the Ha
w
a
i
i
Internatio
na
l C
onfere
n
ce o
n
System Scie
nce
s
, Maui Ha
w
a
ii,
Jan 4-7
,
20
00
.
[18]
W
enju
an
Li,
Xi
aod
on
g W
a
n
g
, Yan
g
g
en F
u
,
Z
h
ixia
ng
F
U
.
S
e
vera
l stud
ies
of the
gri
d
s
y
n
c
hron
izatio
n
mode
l.
Journa
l
of F
u
z
h
o
u
Uni
v
ersity (Natura
l
Science Ed
itio
n)
. 2006; 3
4
(2)
:
2006.
[19]
T
C
G.T
CG Mai
n
Specific
atio
n Versio
n 1.1b [
EB/OL].
[20] wwww
.
t
ruste
d
c
o
mputi
nggr
ou
p
.
org/home, 2
0
0
4
-06-
30.
[21]
Z
h
iqi
n
Gan, G
uosu
n
Z
en.
S
i
multan
eous
as
sessment
of b
ehav
ior-b
ase
d
mode
l in
gri
d
envir
onme
n
t.
Co
mp
uter Appl
icatio
ns an
d Softw
are
. 2005; 12.
[22]
Lipi
ng
W
a
n
g
,
Shou
ba
o Ya
ng
. A s
y
nchr
ono
us mo
del
in
th
e gri
d
envir
on
ment.
Co
mput
er En
gin
eeri
n
g
and Ap
plic
atio
ns.
2004: 2
3
.
[23]
Xi
aofe
ng W
a
n
g
, Jing Z
han
g,
Shang
pi
ng W
ang, Yal
i
n
g
Z
hang. Simu
ltan
e
ous assessm
e
n
t of the grid
computi
ng s
y
stem.
Computer Engi
neer
in
g
. 2006; 22.
[24]
Xi
n Li, Yan Li
u
,
Kefei Chen. S
y
nc
hro
n
izati
o
n model i
n
grid
comput
in
g. Computer En
gin
eeri
ng. 200
8
;
31(3).
[25]
Jin W
a
n
g
,
Xin
Yang,
Xiaos
o
ng L
i
u. S
y
nc
hr
ono
us mo
del
of a n
e
w
ki
nd
of grid
be
havi
o
r.
Co
mp
ute
r
Engi
neer
in
g an
d Appl
icatio
ns.
200
3; 21.
[26]
Liu L
i
zha
o
, Li
u Yingc
ha
o,
Hua
ng Z
h
il
i, Li Maoqi
ng.
T
h
e Cha
o
tic Dis
turbanc
e of UAV System'
s
Co
mmun
icati
o
n And Co
pi
ng
Strategy
. ICCC
AS08 proc
ee
di
ng. 200
8: 155-
160.
[27] Liu
L
i
zh
ao.
T
h
e Cha
o
tic C
h
a
r
acters an
d N
e
w
Control Str
a
tegy of U
n
mann
ed Air
p
la
n
e
Informatio
n
System Intern
ation
a
l Sy
mp
o
s
iu
m
. Comp
utation
a
l Intel
lig
ence a
nd D
e
s
i
gn 2
0
0
8
proc
eed
ing. 2
0
0
8
;
120-
123.
Evaluation Warning : The document was created with Spire.PDF for Python.