TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.4, April 201
4, pp. 2697 ~ 2
7
0
6
DOI: http://dx.doi.org/10.11591/telkomni
ka.v12i4.4251
2697
Re
cei
v
ed Au
gust 11, 20
13
; Revi
sed O
c
t
ober 1
9
, 201
3; Acce
pted
No
vem
ber 8,
2013
The Optimal Design of Communication Module for
Campus Smart Card
Zhu Lin
Sichua
n Col
l
e
ge of Architect
u
ral T
e
chnol
og
y,
Net
w
ork Ma
nag
ement C
e
n
t
er, De
yang 6
1
800
0, Chi
n
a
E-mail: zhu
lin
_
z
lqq@
16
3.com
A
b
st
r
a
ct
Ca
mp
us Car
d
is an
i
m
port
a
n
t
part in th
e d
i
gital c
a
mpus
li
fe. How
e
ver, d
ue to th
e pr
obl
ems
of
current
netw
o
rk desi
gn,
in th
e co
urse
of ca
rd usi
ng,
th
e si
gna
l ch
an
nels
are se
i
z
e
d
by
a lar
ge
nu
mbe
r
of
communic
a
tio
n
tags, w
h
ich leads to the p
hen
o
m
en
a of block
ed co
mmunic
a
tion
proc
ess and u
n
s
m
oot
h
communic
a
tio
n
.
In order to
solve th
is pro
b
le
m,
the c
o
mmu
n
ic
ation
mo
dul
e of car
d
s
ystem
has b
e
e
n
optim
i
z
ed in this paper. The
module c
onsists
of
two parts, which
are writi
ng system
and reading system
.
A
mu
lti-taskin
g
mu
lti-po
int ma
ppi
ng d
e
co
mp
ositio
n techn
i
que w
a
s intr
oduc
ed to th
e desi
g
n
a
tion
of
communic
a
tio
n
mo
dul
e. Usin
g
the metho
d
ba
sed o
n
t
he co
mb
in
ation
of Map a
nd R
educ
e
function to
ma
k
e
the task
dec
o
m
p
o
siti
on w
h
ic
h is
fro
m
a
su
dde
n i
n
cre
a
se
of netw
o
rk traf
fic char
acteristi
cs an
d d
a
tab
a
s
e,
obtain a lar
g
e number
of s
ub-tasks,
and accom
p
lis
h t
he
management
of
m
u
tation network traffic.
Experi
m
ental
results sh
ow
that in th
e task s
c
hed
uli
ng
o
f
op
ti
mi
z
ed
m
o
du
le
, th
e
so
lution
tim
e
is short a
nd
respo
n
se
is f
a
st. It can so
lve th
e co
mmunic
a
tion
bl
oc
ked
prob
le
ms
in th
e
proce
ss of card
us
ing
.
F
u
rthermore, it provid
es theo
retical refer
enc
es for
the impr
ove
m
e
n
t of commu
n
icati
on s
ystem des
ig
n o
f
smart card, a
n
d
pro
m
otes th
e
construction of
digita
l ca
mpus
evolvi
ng.
Ke
y
w
ords
:
smart card, hardw
are pl
atform, s
o
ftw
are desig
n. multi-
poi
nt ma
ppi
ng
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
With the de
velopment of
compute
r
tech
nolo
g
y, the com
pute
r
-ba
s
ed n
e
twork a
nd
comm
uni
cati
on develop
ed
quickly. Fro
m
the develo
p
m
ent of the camp
us n
e
twork to a vari
e
t
y
o
f
appli
c
ation
s
spread, the digital cam
p
u
s
con
s
tr
uctio
n
is moving forwa
r
d at a
n
unprecede
nted
rate. The e
s
sence of Digit
a
l Camp
us i
s
to opt
imize the integratio
n of fragmen
ted informati
on
resou
r
ces to
achieve d
a
ta shari
ng, th
ereby i
n
crea
sing t
he leve
l of man
age
ment withi
n
t
h
e
camp
us an
d
teachi
ng. A
m
ong th
em,
the ca
mpu
s
ca
rd system con
s
tru
c
tion
i
s
the ba
sis
f
o
r
digitizatio
n
project
s
. It plays an impo
rta
n
t role
in the progress of intelligent uni
versity cam
p
us
con
s
tru
c
tion.
Camp
us
Card is a unified
set of
the e
n
tire scho
ol comm
unity finan
cial se
rvi
c
e
s
,
internal mi
cro-paym
ent transac
tio
n
s, a
u
thentication,
and bu
sine
ss man
agem
e
n
t in one of the
school
-wi
de
unified ca
mp
us card
syste
m
. It take
s a
d
vantage of the powerful feature
s
card
and
offline tran
sa
ction
s
of sma
r
t IC. And un
der the
environment of
ca
mpus
network o
r
campu
s
card
private n
e
two
r
k, the
on
e
card
ca
n b
e
a
v
ailabl
e
th
rou
ghout
th
e ca
mpus. Ca
mp
us ca
rd syst
em
has
be
com
e
an imp
o
rtant
mean
s for th
e variety of i
n
stitution
s
an
d unive
rsitie
s to enh
an
ce t
h
e
manag
eme
n
t level and imp
r
ove ca
mpu
s
servi
c
e
s
.
Ho
wever, as the
num
ber of
cam
p
u
s
card us
er a
n
d
function
increases, the
n
u
mbe
r
of
netwo
rk no
d
e
s al
so
incre
a
se
d, the n
e
t
work
stru
ctu
r
e i
s
be
co
ming mo
re
co
mplex, and
each
manag
eme
n
t system
s
run
on the
networks
platform
have differe
nt perfo
rma
n
ce
req
u
ire
m
e
n
ts.
Therefore,
th
e contra
dictio
n of
si
gnifi
ca
nt increa
se
in
network
si
ze
and
existin
g
netwo
rk servi
c
e
cap
abilities
a
nd environm
ent ha
s be
come in
cre
a
si
ngly promi
n
e
n
t. This is m
a
inly due to
the
disa
dvantag
e
desi
gn of
communi
catio
n
mod
u
le in
cam
p
u
s
card. Since
no
des
with limi
t
ed
resou
r
ces, th
ere a
r
e a la
rg
e numbe
r tag
s
to sei
z
e the
commu
nication ch
ann
el, whi
c
h lea
d
ing
to
dynamic sch
edulin
g man
a
gement ta
sks are
affected,
cau
s
in
g blo
c
ked
co
mmuni
cation
process
and un
smo
o
th comm
uni
ca
tion, limiting t
he scop
e of application of the ca
rd.
Curre
n
tly, researche
r
s hav
e u
s
ed va
riou
s meth
od
s to
deal
with the
resou
r
ce
sch
edulin
g
of virtual
net
work sy
stem. Gen
e
ral
spe
a
kin
g
, it
can
be divid
ed i
n
to two
ki
nd
s a
l
gorithm
s
whi
c
h
are stati
c
alg
o
rithm an
d dynamic al
go
rithm. In
static algorith
m
, the virt
ual netwo
rk ma
p doe
s
not
cha
nge
with
time, resulting in lo
w ut
ilization
of
p
h
ysical re
so
u
r
ce
s.
While i
n
the dynam
ic
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 4, April 2014: 2697 – 2
706
2698
algorith
m
, it can
re
allocate the ma
ppe
d re
so
urce
s
i
n
the runni
ng
pro
c
e
s
s of v
i
rtual n
e
two
r
k to
optimize th
e use of the u
nderlyin
g phy
sical netwo
rk resource
s to
ensu
r
e n
e
twork
co
nne
ctivity
[1]. Literature
prop
osed to
re-co
n
figure the virt
ual n
e
twork m
appi
n
g
peri
odi
cally
and
sele
ctively
to optimize the phy
sical
netwo
rk
overload p
a
rt
s. Ho
wever, th
e num
ber
of
virtual net
work
resource real
location will
affect the stability
of the
network and
comp
utational overhead [2];
literature also
desig
ned
a redistri
bution of
resource
s mappin
g
algo
rithm. The al
gorithm
rema
ps
the lowe
r pri
o
rity requ
est
of virtual network
o
n
bottlene
ck n
ode
s
and bottlen
eck link to imp
r
ove
the acce
ptan
ce
ratio of v
i
rtual net
wo
rk an
d
re
du
ce the p
r
essu
re on t
he u
s
e of re
sou
r
ces
Ho
wever, in this algo
rithm,
only in the e
v
ent of
a virtual netwo
rk reque
sts
cann
ot be mappe
d, it
make
s
rem
a
p
p
ing a
d
ju
stment of virtual
netwo
rk. In
th
e migration of
a virtual n
o
d
e
, it is al
so n
e
ed
to migrate virtual link con
necte
d to the node.
Thu
s
, the overal
l pays for ca
lculatio
n of the
algorith
m
is
still large [3]. Literatu
re
prese
n
ted a
d
y
namic
adap
tive virtual n
e
twork m
app
ing
algorith
m
for
a customi
z
e
d
Internet. T
h
e
algo
rithm
i
s
execute
d
pe
ri
odical reallo
cation in
multiple
virtual gate
w
ays to
sha
r
e
the ba
nd
wid
t
h between
t
he virtual
lin
ks of the
sa
me phy
sical
link.
Ho
wever, the
algorithm do
es not co
nsi
der the a
llo
cation of virtual netwo
rk n
ode
s, and in
this
algorith
m
, ph
ysical n
e
two
r
k link n
eed t
o
kno
w
t
he p
e
rform
a
n
c
e o
b
jective fun
c
tion of all virtual
netwo
rks in
a
d
vance [4, 5
]. Literature
p
r
es
ents
a di
stributed
auton
omy re
sou
r
ce mana
gem
e
n
t
mech
ani
sm
s based on self-organi
zation
of techni
ca
l, in the mech
a
n
ism, the phy
sical node
ca
n
identify of the overloa
d
traf
fic a phy
sical
link, an
d mig
r
ate throug
h the virtual n
o
d
e
s to mini
mize
su
ch flo
w
e
x
perien
c
e
d
b
y
the num
b
e
r of
physi
cal links, to
maintain l
o
a
d
bala
n
ci
ng
and
band
width
sa
vings
of re
so
urces u
s
ed.
But this
m
e
chani
sm d
o
e
s
not expl
ain
the virtual
ho
st
node
sele
cti
on method
o
f
objectives t
o
be migrate
d
[6]. VMCTune is a dy
namic
re
sou
r
ce
allocation ba
sed o
n
the vi
rtual cl
uste
r l
oad bal
an
cin
g
mechani
sm
, which
ca
n
be monito
red
in
real
-time virt
ual ma
chin
es and p
h
ysi
c
al
machi
n
e
s
re
sou
r
ce utiliza
t
ion, and the
n
by calli
ng t
h
e
resou
r
ce re-a
llocatio
n
alg
o
r
ithm, on
the
same
phy
sical ma
chin
e, to obtai
n pa
rti
a
l load
bal
an
cing
of virtual m
a
chi
n
e
s
; sim
u
ltaneo
usly,
on multip
le
physi
cal ma
chin
es, virtu
a
l machine
live
migratio
n to
obtain the
gl
obal lo
ad-bal
anci
ng virt
ual
clu
s
ter. T
h
is mechani
sm
can
effectivel
y
allocate resources to virtual machines,
but also
improve the utilization of
physi
cal resources.
But
this alg
o
rithm
to sel
e
ct th
e
target
ho
st, did not
co
nsi
der th
e imp
a
c
t of mig
r
atio
n on li
nk load
[7,
8].
To solve the
s
e p
r
obl
em
s, we p
r
o
p
o
s
e
a de
com
p
o
s
ition method
based o
n
th
e multi-
taski
ng a
nd
multi-poi
nt mappin
g
[9]. In this metho
d
, the introd
uct
i
on of Map
R
e
duce fun
c
tion
is
applie
d in the
task d
e
comp
osition of mut
a
tions traffic in cam
p
u
s
ca
rd net
work to
compl
e
te the
resou
r
ce sch
edulin
g of virtual net
work n
ode
s an
d the
simulatio
n
e
x
perime
n
ts
so as to
optimi
z
e
the com
m
uni
cation m
odul
e desi
gn of
curre
n
t cam
p
us net
wo
rk
card an
d the
r
eby solve the
blocke
d pro
b
l
e
ms in the ne
twork co
mmu
nicatio
n
s.
A new an
d co
nvenient mod
e
rn life is brou
g
h
t
to student
s and teache
rs, so as to enhan
ce
working efficien
cy, and improv
e the modern
manag
eme
n
t level of sch
oo
l [10, 13].
2.
Ov
erall Designs of Camp
us Smart Ca
rd
2.1. Design
Philosoph
y
of the Sy
stem
Digital camp
us card syst
em nee
ds to
reflect the d
o
minant id
ea
of integrate
d
desi
g
n
sufficie
n
tly, all the data so
urces of
informationized
campus a
r
e fro
m
the sha
r
ed
databa
se, when
the variou
s d
epartm
ents a
r
e processin
g
data ma
int
enan
ce in future, only the
data of shared
databa
se n
e
e
d
to be maint
a
ined, all the
informat
io
n source
s of ea
ch appli
c
ation
sub
s
ystem
are
from the
sha
r
ed data
b
a
s
e,
whe
n
const
r
ucting th
e
su
bsyste
ms, th
e integration
and u
n
ificatio
n o
f
sha
r
ed d
a
ta
cente
r
, sma
r
t
card system
s and p
o
rt
al i
n
formatio
n sy
stem
s sh
ould
be con
s
id
ered,
in order t
o
achi
eve
co
nsi
s
ten
c
y of informatio
n
from top t
o
bottom a
nd unify ide
n
tity
authenti
c
atio
n. System a
r
chite
c
tu
re
should
co
nsid
er the
prin
ci
ple of ind
e
p
ende
nce, wh
ich
mean
s it is
not re
stri
cted
by
sp
ecifi
c
types a
dop
ted, like
dat
aba
se
s, ope
rating sy
stem
s,
prog
ram
m
ing
langua
ge
s,
card type, co
mmuni
ca
tion
netwo
rks, an
d ca
n be a
d
apted to vari
ou
s
operational e
n
vironm
ents
and ma
ke full use of existi
ng re
sou
r
ce
s.
2.2. Require
ments o
f
Sy
stem Design
Smart card
tech
nolo
g
y for universities i
s
a
clo
s
e i
n
te
gration
of co
mputer
and
h
i
gh-te
ch
means based on intelligent recogniti
on principle. T
he inherent i
n
formation characteristi
c
s and
attributive ch
ara
c
ter
of the ca
rd a
r
e
use
d
to
process rel
a
ted
operati
on. A
c
cordi
ng to t
he
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
The Optim
a
l De
sign of Co
mm
unication
Module fo
r Cam
pus Sm
art Card
(Zhu Li
n)
2699
cha
r
a
c
teri
stics of embed
d
ed tech
nolo
g
y
and two-
di
mensi
onal
co
de rea
d
e
r
techn
o
logy and
its
appli
c
ation i
n
the field
of testing, th
e
study det
e
r
min
ed the
overal
l stru
ctu
r
e
of the
sma
r
t card
read
er
syste
m
based on
embed
ded
system. The e
n
t
ir
e system i
s
divided into
three p
a
rts: t
h
e
read
er, e
m
be
dded
processing
platform
and L
C
D di
sp
lay.
The entire proces
s of t
he unive
rsitie
s
card rea
d
e
r
systems is a
c
compli
sh
ed b
y
embedde
d system. The
main wo
rk p
r
oce
s
s inclu
d
e
s
informatio
n re
ad & written, informat
io
n co
mpared, located and o
pera
t
ed.
By c
o
ns
truc
ting
s
m
art c
a
rd projec
ts
to form
n
e
two
r
k o
perating
environ
ment
whi
c
h i
s
suitabl
e to scho
ol cam
p
u
s
ca
rd
syste
m
, build
a b
a
si
c ca
rd
system software and ha
rd
ware
architectu
re. Card
d
a
ta ce
nter
a
nd ca
rd
man
age
me
nt
ce
nter ba
sis
platform are co
nst
r
u
c
ted,
camp
us
ca
rd
serv
i
c
e
cent
e
r
,
ban
k t
r
an
sf
er sy
st
em,
se
t
t
l
ement
sy
st
ems a
nd s
e
lf
-
s
erv
i
ce
sy
st
e
m
s
are
built; ca
mpus card p
e
rsonal
ide
n
tification
and
e
l
ectro
n
ic pu
rse payme
nt fu
nction
s
are
u
s
ed
to achieve
campu
s
di
ning
ch
arg
e
s,
su
perm
a
rket o
r
sm
all bu
sin
e
s
s di
stri
ct sh
oppin
g
cha
r
g
e
s,
Libra
r
y fee
s
,
clini
c
fee
s
a
n
d
othe
r unifie
d
co
ns
umptio
n and
settlement; identity a
u
thentication
is
reali
z
ed,
like
sig
n
-i
n fo
r meetin
gs,
stude
nt atte
ndan
ce,
phy
sical exe
r
cise; integ
r
atio
n of
comp
uter
ro
om man
age
ment, libra
ry manag
eme
n
t, and oth
e
r third-p
a
rt
y systems
are
compl
e
ted, in
orde
r to achi
eve a comp
re
hen
sive unifie
d
camp
us
ca
rd.
3.
Hard
w
a
re
De
sign of Sy
st
em
Backgroun
d system
con
s
tructi
on of the sma
r
t ca
rd is divide
d
into hard
w
are an
d
sof
t
w
a
r
e c
o
n
s
t
r
u
c
t
i
on.
The
sy
st
em sele
c
t
s a S
a
msun
g ARM9 pro
c
essor a
nd fra
m
ewo
r
k of Linux
system
s, an
d
the outp
u
t m
odule
and
co
mmuni
cati
on
module
are a
ttached, th
e card i
n
form
ation
is collecte
d
, pro
c
e
s
sed a
n
d
run i
dentifying inform
atio
n, informatio
n extractio
n
, and
sub
s
eq
u
ent
operation
s
in
depe
ndently. The pr
opo
se
d
system sel
e
cts syst
em
s
based on A
R
M9 & LINUX
and
dual
USB interface read
er whi
c
h
contai
ns the fu
ncti
o
n
of tag ident
ification a
nd i
s
sta
r
ted at t
he
same
time to
ensure
the
synchrono
us
o
peratio
n
of th
e syste
m
. A SDRAM m
e
mory a
r
chitecture
is ad
opted,
software
de
sig
n
incl
ude
s th
ree p
a
rt
s:
informatio
n re
a
d
ing, informa
t
ion com
p
a
r
ison,
and i
n
form
ation o
p
e
r
ation.
Wh
ere the
sampl
e
i
s
se
lected
by the
CUP a
u
tom
a
tically, relat
ed
operation
s
are pro
c
e
s
sed
according to
the different
types of information, part
of the card tag
s
with same
si
ze a
r
e
sel
e
ct
ed a
s
the t
r
a
i
ning
sam
p
le
s an
d
sampl
e
s to
be id
e
n
tified, the two-
dimen
s
ion
a
l i
n
formatio
n in
the ca
rd is
read
by the card reade
r,
and
recogni
zed by que
ryi
n
g
comp
ari
s
o
n
i
n
formatio
n in
the data
b
a
s
e, then the fi
nal ide
n
tificati
on result is o
u
tput. The d
e
v
ice
is po
rtable, l
o
w p
o
wer
co
nsum
ption a
n
d
ca
n be
ap
plied to oth
e
r are
a
s
by mean
s of software
desi
gn, su
ch
as card lo
ss reportin
g
, ca
rd
dynamic mo
dification.
Hardware ci
rcuit is the fo
u
ndation of the
embed
ded
card read
syst
em, and the
carri
er to
achi
eve its variou
s fun
c
ti
ons. S3
C24
4
0
is a hi
gh-q
uality pro
c
e
s
sor
belo
n
g
s
to ARM9 series
manufa
c
tured
by Samsun
g, whi
c
h
use
s
a five
-st
a
g
e
pipeli
ne
structure, ha
s h
i
gh efficie
n
cy
of
instru
ction
ru
nning. It sup
ports m
u
ltiple
operatin
g sy
stem
s inclu
d
i
ng LINUX. The inform
atio
n is
store
d
u
s
ing
SDRAM
and
FLASH me
m
o
ry. The F
D
S netwo
rk
co
ntrolle
rs
emb
ed the T
C
P /
IP
proto
c
ol i
n
to t
he
cont
rolle
r
for the
first
time, an
d reali
z
e Inte
rn
et surfing t
h
ro
ug
h the
TCP /
IP. It
is co
nvenient
to acce
ss th
e Internet ne
twork and e
a
s
y to achieve
network ma
nagem
ent. The
controlle
r u
s
es
a hi
gh
-p
erform
an
ce
real-time
op
e
r
ating
sy
ste
m
, and
sup
ports multi
-
task
operation
whi
l
e en
su
re th
e
real
-time fe
a
t
ure of th
e ta
sk.
Di
splay
p
a
rt u
s
e
s
Sh
arp's
3.5-i
n
ch T
F
T
LCD, co
mm
unication int
e
rface u
s
e
s
the RS
-23
2
and
USB
to reali
z
e
human
-comp
u
ter
interac
t
ion.
3.1. S3C244
0
Chip
S3C24
40
pro
c
e
s
sor i
s
develope
d by th
e Sam
s
ung
with ARM
9
a
s
its
co
re. In
orde
r to i
m
prove
the comp
atib
ility with traditional von Neuman
n
arch
itecture, ARM920T u
s
e
s
Harvard Ca
che
stru
cture. Wit
h
great ex
p
a
n
sibility, the
pro
c
e
s
sor ca
n meet th
e requireme
nt o
f
future
syste
m
upgradin
g
. ARM9 mi
cro
p
roce
ssor h
a
s t
he followi
ng chara
c
te
risti
c
s:
(1) Five
pip
e
line: fetch,
decode, exe
c
ute,
sto
r
ag
e
and write
operation. In
stru
ction
executio
n efficien
cy is hig
h
(2) Provide
s
a Harva
r
d a
r
chitecture
(3) Sup
p
o
r
ts
dual in
stru
ction set
(4) Sup
p
o
r
ts
32-bit hig
h
-sp
eed AMBA bus interfa
c
e
(5) Sup
p
o
r
t Linux and othe
r embedd
ed systems
(6) MP
U su
p
ports
real
-tim
e operating system;
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 4, April 2014: 2697 – 2
706
2700
3.2. The Car
d
Read
er Selection
RFID (radio
freque
ncy
i
d
entification
)
t
e
ch
ni
qu
e is
employed
in
rea
d
ing
an
d
writing
system of ca
mpus
card. The techn
o
log
y
is an
automatic identification techni
q
ue, wirel
e
ss non-
conta
c
t meth
od is u
s
e
d
to com
m
uni
cate in two-way, so as to
enable
rea
d
i
ng in bat
ch
and
remote, an
d i
dentify fast moving obje
c
ts, each o
b
je
ct has a
uniqu
e
identifier to
meet the nee
ds
of incre
a
si
ng
flow of information and the incr
ea
sing
speed of informatio
n pro
c
e
ssi
ng, thereby
increa
sing eff
i
cien
cy and redu
cing
co
sts. RFID techn
o
logy adopt
s
a
large scale integrate
d
ci
rcuit
cal
c
ulatio
n, electroni
c
id
entific
ation,
comp
uter co
mmuni
cation
tech
nolo
g
y, to a
c
hieve
non
-
c
o
n
t
ac
t
c
a
r
r
i
er
id
en
tific
a
tion
a
n
d
da
ta
exc
h
a
n
ge
by the read
er an
d the
RFI
D
t
ag mo
unted
on a
carrie
r. RFI
D
technol
ogy
has the
adva
n
tage
s wh
i
c
h
bar
code
do
es not h
a
ve, like waterpro
of,
antimagn
etic,
remote read
distan
ce, larg
e st
ora
ge cap
a
city of labeli
ng data, etc.
Typical
RFID rea
der sy
ste
m
co
nsi
s
ts
o
f
th
ree p
a
rt
s: the ele
c
tro
n
i
c
tag
s
, reade
rs
and
comp
uter
co
mmuni
cation
s ce
nter. The
system
blo
c
k
diagram is
sh
own in Fig
u
re
1.
Figure 1. RFI
D
System Blo
ck
Diag
ram
(1) Ele
c
tro
n
ic
tag
Electro
n
ic
ta
g
is kno
w
n as radio
-
fre
q
uen
cy
tag o
r
tran
spon
de
r, and it is t
he data
informatio
n carri
er of the
RFID
system,
whi
c
h is
usu
a
lly installed
on the ide
n
tified obje
c
t u
s
e
d
to
store
the
obj
ect info
rmatio
n. According
to differ
ent
a
pplication
en
vironme
n
ts, t
he el
ectroni
c
tag
has diffe
rent
shap
e, si
ze
and op
erati
ng freq
uen
cy
, and sp
ace
coupli
ng wit
h
RF si
gnal
is
achi
eved to
complete
data
co
mmuni
cati
on a
nd
ene
rg
y tran
sfer th
rough
coupli
n
g elem
ents a
nd
read
er.
Du
e t
o
the
metho
d
of p
o
we
r
su
pply, ele
c
tron
ic tag
can
be
divided
into
active tag
an
d
passive tag.
(2) Rea
der
The read
er,
also
kn
own a
s
a
rea
d
he
a
d
, is respon
si
ble to comm
unicate with
electroni
c
tag, re
ad
or
write
ele
c
tro
n
i
c tag
info
rma
t
ion,
and
sen
d
the
data
to
the comp
uter co
mmuni
cati
on
network. In t
he process
of wo
rk, the
establi
s
hm
ent of the com
m
unication,
anti-collisi
on
and
authenti
c
atio
n are compl
e
ted by the rea
der.
(3)
Comp
uter commu
nication ce
nter
The comp
ute
r
co
mmuni
ca
tion ce
nter i
s
mainly re
sp
onsi
b
le to se
nd comman
d
s
to the read
er,
c
o
llec
t
da
ta r
e
c
e
ive
d
by r
e
a
d
e
r
,
an
d
th
en
pr
oc
es
s d
a
t
a in
ac
co
r
d
a
n
ce
w
i
th d
i
ffe
re
n
t
requi
rem
ents.
Usually it a
c
hieves
data t
r
an
sfer
with t
he read
er th
r
ough RS-232
,
RJ-45,
USB
,
Ethernet an
d Wi-Fi.
3.3. LCD
Dis
p
lay
S3C24
40 co
ntains
i
n
tern
al
built-in L
CD co
ntrolle
r
whi
c
h su
p
ports
4,
8
a
n
d
16-bit
colo
r
LCD,
su
ch
as STN an
d TFT. T
a
ki
n
g
the
e
ffectivene
ss an
d
co
st into
acco
u
n
t, the p
r
opo
sed
system
sele
ct
s Sharp
3.5-i
n
ch 3
20X24
0
TFTLCD
. Th
e expan
sion i
s
co
nvenie
n
t becau
se of the
LCD controll
e
r
and
drive
r
o
f
S3C244
0. (1) VD0-1
5
: 1
6
-bit data li
ne
; (2) VCLK: LCD
clo
c
k sig
nal,
whi
c
h i
s
used
to send
the
data of
ea
ch
point into
the
shift re
giste
r
;
(3) VLine: li
n
e
si
gnal
u
s
ed
to
indicate com
p
letion of the
data-in
-line
transmissio
n
from the
shif
t bit registe
r
s to the displ
a
y
driver chip, and ma
ke
s the line point
er. In t
he mode of TFT
16, that is, the horizo
n
t
a
l
synchro
n
ization sig
nal; (4
) VFram
e
: frame sig
nal
i
s
used to in
dicate the
start of person
nel
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
The Optim
a
l De
sign of Co
mm
unication
Module fo
r Cam
pus Sm
art Card
(Zhu Li
n)
2701
informatio
n, while the
ro
w pointer i
s
put
in the firs
t lin
e of the display. In 16-bit mode of TFT,
it is
a verti
c
al
syn
c
hroni
zing
si
gnal;
(5) L
C
D_PWRE
N:
In 16
-bit m
o
d
e
of T
FT, the
ena
ble
sig
n
a
l is
output. A data sign
al for in
dicatin
g
the synchroni
za
tio
n
of the clock signal, the la
tch to the pin.
3.4. Communication Inte
rfac
e and Storage Mod
u
le
The com
m
uni
cation
interfa
c
e
s
of ha
rdware sy
st
em in
clud
e
RS-2
32
se
rial i
n
terfa
c
e
an
d
USB interfa
c
e. The
CLIENT TO
HOST se
rial d
a
ta
transmissio
n
can
be a
c
hi
e
v
ed ea
sily through
the RS-2
32 seri
al port which i
s
use
d
by mo
st embedd
ed in
strument
s and
compute
r
d
a
ta
transmissio
n. In o
r
de
r to
conne
ct the
RS-232
se
rial
port to
TlL l
e
vel device, M
AX232 inte
grated
circuit
chip i
s
em
ployed
to swit
ch S3
C24
10
se
rial
level. The
cha
r
a
c
ter
m
edian
of UA
RT
transceive
r
i
s
co
nfigurable
and
ca
n be
5
to 8. In
tran
sition, 32-byte
data
written f
r
om data
bu
s i
s
transmitted t
o
FIFO,
and
then i
s
fe
d t
o
the
shi
ft
re
gister
a
nd co
nverted
to se
rial data outp
u
t
throug
h TXD pin. Data is received fro
m
RXD pin
serial. It first enter the re
cei
v
e shift regist
er,
and the
n
fed
into the FIF
O
with h
a
lf-word d
ept
h. T
he re
ceivin
g
and tra
n
sitin
g
of FIFO h
a
ve
maskabl
e int
e
rruption
whi
c
h
coul
d be i
s
sued
wh
en
data of the FI
FO re
ache
s
a pre-set am
ount.
At the same
time, the receiving and transitin
g of
FIFO ca
n re
qu
est the DMA.
UART tra
n
smit
baud
rate d
e
pend
s o
n
the
input divide
r
whi
c
h h
a
ve
configurable
cl
ock an
d
software. The
UART
also
have
uni
t circuit fo
r b
aud
rate
self
-detectio
n
whi
c
h co
uld be use
d
to co
nfigure
the
sto
p
bit
and p
a
rity check. Lin
u
x operat
ing
system
sup
port
s
mo
st com
m
on ha
rd
wa
re devi
c
e
s
a
n
d
provide
s
th
e approp
riate d
r
ivers,
so all
of OHCI
com
p
liant
USB d
e
vice i
n
terfa
c
e can
be
driv
en.
The sy
stem
sup
port
s
the
USB interfa
c
e. St
ora
ge
equipm
ent select
s SDRA
M (syn
ch
ron
ous
DRAM
)
whi
c
h
ca
n b
e
synchroni
ze
d
with
the exte
r
nal
clo
c
k of the
CPU. The
SDRAM and
sy
ste
m
CPU u
s
e the same
clo
ck, so if
the CPU external cl
ock is 100MHZ
the frequ
en
cy sent to memo
ry
is al
so 10
0M
HZ. Throug
h
this way me
mory efficie
n
c
y ca
n be im
proved. T
he
system h
a
rd
ware
platform use
s
K4S5616
3
2
-T
C75 chip
manufactu
re
d by SAMSUNG, with monolithi
c cap
a
city
32M *
16
bit and
SDRA
M mod
u
le
s. CUP
come
s
with F
L
A
S
H el
ectri
c
al
ly era
s
a
b
le, an
d
informatio
n is not lost when po
we
r-d
own. The
h
a
r
dware desi
g
n of the sy
stem is sh
own
in
Figure 2.
Figure 2. Hardwa
re Desi
gn
of the Comm
unication System
4.
Soft
w
a
r
e
De
sign of th
e Sy
stem
4.1. Initializ
ation Design
The m
a
in
p
r
og
ram
a
c
co
mplish
e
s initi
a
lizatio
n
work a
c
co
rding
to the
ba
ckgroun
d
comm
and.
If the RF module re
ceive
s
sendin
g
co
mmand
s,
it
will call the initialization
subro
u
tine.
Whe
n
CO
NF
IG re
giste
r
P
R
IM_RX
bit i
s
0,
the t
r
an
smit mo
de i
s
begi
n, u
s
e
r
s sh
ould
set t
he
address of the sen
d
ing & receiving an
d
the minimu
m address wi
dth of the lo
we
st two bits of
SETUP_AW
regi
ster
("01"
means 3 by
t
e
s,
"10" re
prese
n
ts 4
byt
e
s,
"11" is 5
bytes,
"00" is
an
invalid state).
If the reader commu
nication ha
s t
ag of
nRF2
4L01, t
he last bit of EE_AA regist
er
will be set 1
to open automatic ACK
function.
If
reader
comm
u
n
icate with t
he other mo
dels’
comp
atible ta
gs
(n
RF24
01,
nRF
240,
nRF24E1, n
R
F2
4E2), thi
s
bit
must b
e
set t
o
"0"
to turn
off
the automatic ACK. Part of
t
he cod
e
is shown as follo
ws:
void TX_Mod
e()
{
IO_Clear_CE();
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 4, April 2014: 2697 – 2
706
2702
SPI_Write_B
uf(WRITE_REG +
RX_A
DDR_P0,
TX_ADDRESS, TX_ADR_WI
D
TH); // Set
the sen
d
ing a
nd re
ceiving
addresse
s
S
P
I_RW_Reg(WRITE_REG +
EN_A
A,
0x01); // Start the autom
atic ACK function
SPI_RW_
R
eg
(WRITE_
R
EG + SETUP_AW,0x02)
;
// Set the width of the address
SPI_RW_
R
eg
(WRITE_
R
EG +
RF_S
ETUP, 0x0F
);
// Set the data rate and o
u
t
put powe
r
SPI_RW_
R
eg
(WRITE_
R
EG +
CO
NF
IG, 0x0A); //
Set the sendi
ng mode
IO_Set_CE(); // Set CE at high level, enable
sen
d
i
ng device
}
If the RF m
odule
re
ceiv
es
re
ceiving
comm
and
s,
i
t
will call the
re
ceiver initializatio
n
routine. Th
e CO
NFIG re
gi
ster PRIM_
R
X bit is se
t to
0, and enter receive mo
de.
The initializat
ion
of the other registe
r
s a
nd
sen
d
pro
g
ra
m are
the same. Part of the c
o
de is
as
follows
:
vo
id
R
X
_
M
ode
(
)
{
IO_Clear_CE();
S
P
I_Write_Buf(WRITE
_REG
+
RX_ADD
R_P0, TX_ADDRESS, TX_ADR_WIDTH); //Set
the sen
d
ing a
nd re
ceiving
addresse
s
SPI_RW_Reg(WRITE_RE
G +
EN_AA, 0x01); // the automatic ACK
function
SPI_RW_
R
eg
(WRITE_
R
EG + SETUP_A
W,0x02);
// Set the width of the address
SPI_RW_
R
eg
(WRITE_
R
EG +
RF_S
ETUP,
0x0F);
// Set the data rate and o
u
t
put powe
r
S
P
I_RW_Reg(WRITE_REG +
CONFIG
, 0x0B); //
Set the receiv
ing mode
IO_Set_CE(); // Set CE at high level, enable recei
v
ing device
}
4.2. Reading
Program De
sign
The coll
ecte
d tag inform
ation is se
nt fr
om nRF2
4
L01 to micro
c
ontrolle
r through SPI
port, and the
n
to PC, finally proce
s
sed
by the background p
r
og
ra
ms.
MSP430F1
4
9
write
s
the
d
a
ta to be
se
nt and th
e re
ceiving
add
re
ss i
n
to the
T
X
_FIFO
regi
ster
of n
R
F24
L01, the
n
se
nt CE to
high,
at this ti
me, nRF
24L
0
1
begin to
se
nd data.
Whe
n
compl
e
ted, the tran
smitting device waits for AC
K
from the re
ceptio
n sid
e
. If the ACK is not
received
with
in a
ce
rtain
time, the dat
a will
be aut
omatically ret
r
an
smitted. Retra
n
smissi
on
interval an
d the num
ber
of retra
n
smi
s
si
on are det
e
r
mined by the
high fou
r
ADR an
d the lo
w four
ARC of
SET
UP_RET
R re
gister. The
n
u
mbe
r
of re
transmi
ssion
is up to
15.
If
AC
K succ
essfully
received
or t
he retran
smi
ssi
on i
s
time
out, the dev
i
c
e ente
r
s sta
n
dby mod
e
to
wait for the n
e
xt
data tran
smi
s
sion.
In the tran
sm
issi
on, 7 to 4
bits of OBS
E
R
VE_TXP registe
r
a
r
e d
e
fined a
s
pa
cket loss
cou
n
ter PLO
S
_CNT used
to reco
rd th
e numbe
r of
data pa
ckets lost in the transmitted. 3
to 0
bits a
r
e d
e
fined a
s
data
retra
n
smissi
on counte
r
A
R
C_CNT
re
set whe
n
n
e
w data p
a
cket
is
available. Th
e su
ccess rate
of data re
ceived ca
n be
obtaine
d thro
ugh PLOS_
C
NT reco
rd val
u
e
whi
c
h can be
use
d
to asse
ss the
reliabili
ty of
the equi
pment.
4.3. Multi-ta
s
k
Multi-poin
t
Mapping De
composition
Algorithm
Virtual net
wo
rk m
appi
ng i
s
to requ
est
for
di
stributio
n of the corresp
ondi
ng u
nderlyin
g
physi
cal
network resource
s to th
e virtu
a
l net
work
which
ha
s
different virtu
a
l n
ode
s a
nd virt
ua
l
link resource
con
s
trai
nts
and topol
ogy
requi
rem
ent
s. From th
e perspe
c
tive
of infrast
r
u
c
ture
provide
r
s, ma
pping
sho
u
ld
as mu
ch a
s
p
o
ssible
to me
et the need
s
of the virtual netwo
rk i
n
the
ca
se of
spe
n
d
ing minim
u
m re
sou
r
c
e
co
st
.
V
i
rt
ual
Net
w
or
k Ma
p
p
ing i
s
just
a
sub
-
f
u
n
c
t
i
on
of
Network virtu
a
lizatio
n, how to us
e the
shorte
st po
ssi
ble time to co
mplete the re
sou
r
ce map
p
i
n
g
is a ne
ce
ssa
r
y conditio
n
to ensure
sm
ooth depl
oy
ment of servi
c
e
s
and a
ppl
ication
s
. Virtual
Network Map
p
ing Pro
b
lem
is also
NP-h
ard even in
t
he co
ndition
of static dem
and In the early
studie
s
, rese
arche
r
s u
s
e
d
vario
u
s of
method i
n
va
rying d
e
g
r
ee
s, to limit the
probl
em,
su
ch a
s
ignori
ng the
virtual nod
e
or virtual lin
k resou
r
ce
re
quire
ment
s, or supp
ose
all of the virtual
netwo
rk
req
u
e
sts a
r
e
kno
w
n, or o
n
ly con
c
e
r
ne
d with a particul
a
r virtual net
work top
o
log
y
requ
est, or t
he assu
mptions u
nde
rlying physi
ca
l netwo
rk ha
s
sufficient
re
sou
r
ces, without
rega
rd to th
e
requ
este
d virtual net
wo
rk re
sou
r
ces li
mit, or to ign
o
re the
po
sition of the virt
ual
node req
u
ire
m
ents
The
r
e
f
ore,
re
se
archers h
a
ve propo
sed
a nu
mber
of heu
ri
stics to
study
su
ch
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
The Optim
a
l De
sign of Co
mm
unication
Module fo
r Cam
pus Sm
art Card
(Zhu Li
n)
2703
probl
em
s. Th
ese
he
uri
s
tic
strategi
es ca
n be
divided
i
n
to two
categ
o
rie
s
: the
de
mand
ba
sed
on
static ma
ppin
g
probl
em an
d need
s ba
se
d on dynami
c
mapping p
r
o
b
lem.
The sm
art
card
s' lab
e
l m
anag
ement
method for i
n
formatio
n m
u
tation de
co
mpositio
n
based o
n
mu
lti-point multi-task
de
com
p
osition
te
ch
ni
que, complet
e
de
comp
osit
ion on
Ha
do
op
platform
whi
c
h is
co
mpo
s
e
d
mainly by t
he di
stri
b
u
ted
file syste
m
a
nd Ma
pRedu
ce ta
sk pa
rall
el
comp
uting m
odel. The pla
tform parallel
calculate
s
card ta
sk b
e
h
a
vior data, wi
thout con
s
id
e
r
ing
singl
e node t
a
sk sche
dulin
g and data
storag
e proble
m
s.
(1)
Card label inf
o
rmatio
n extraction
In the ta
sk d
e
com
p
o
s
ition
process, first, t
he label
is coll
ecte
d a
n
d
tra
n
smitted
to the
netwo
rk,
to be rest
ore
d
t
o
the co
nne
ction re
co
rd
s of T
C
P / I
P
layer, a
n
d
its featu
r
e
s
are
extracted. Th
e details of which a
r
e de
scribed b
e
lo
w
1)
The network
comm
uni
cati
on modul
e is
set to
open, so that it can condu
ct su
rvei
llance on
tag read a
nd
con
n
e
c
t card monitori
ng sy
stem and
co
mmuni
cation
netwo
rk.
2)
Pre-p
r
o
c
e
s
s of the operati
ng stru
ctu
r
e o
f
the tag data, and ca
pture
data acco
rdin
g to the
data pa
ckets
conve
r
ted fr
o
m
tag operation data.
3)
Extract data packet of net
work op
eratio
ns
from the m
odule a
nd co
nvert its form
at, then
store in the d
a
taba
se.
4)
Acco
rdi
ng to the above p
r
o
c
ed
ure to iterative
process the card labe
l until it reach
e
s the sto
p
con
d
ition to e
nd the ope
rati
on.
5)
The ca
rd
com
m
unication m
odule i
s
co
nfigure
d
to normal mode, re
ceiving d
a
ta in buffer to
obtain net
work ope
rating
chara
c
te
risti
c
s
and provid
e accurate data
foundation fo
r
manag
eme
n
t of decom
po
si
tion of network traffic mutat
i
on.
(2)
Mut
a
tions assignm
ents decompo
sitio
n
Manag
eme
n
t
Card ta
g d
a
ta colle
cted
a
r
e i
n
itialize
d
t
o
con
s
titute
a data
colle
ction that
cont
ains the
central nod
e
and se
rial
numbe
r. Among them,
the central no
de re
co
rd
s the name
sto
r
ed
locatio
n
of t
he file
syste
m
and
clie
nt acce
ss
tra
c
es, an
d
seri
al num
ber is use
d
to
sto
r
e
manipul
ating
data in the
n
e
twork
and t
he u
s
er's
r
e
q
uest fo
r processing. A
cco
rding to the
serial
numbe
r of th
e netwo
rk op
eration
a
l dat
a, we c
an create,
delete, and
copy
the
data. The d
a
ta
pro
c
e
ssi
ng fo
r netwo
rk feature set above
has a st
rong
fault toleran
c
e and thro
ug
hput.
The
combi
nat
ion of Ma
p fu
nction
and
th
e Re
du
ce fu
n
c
tion tran
sfer data
set co
mposed
by label
feat
ure
to the
di
stribute
d
net
work file
syst
em,
thereby redu
ci
n
g
n
e
twork file
storage
time. The main step
s of data pro
c
e
ssi
ng method are
that: first divide the large
r
tasks into a la
rge
numbe
r of su
b-tasks; then
carry out in
depe
ndent
d
a
ta ope
ration
to sub-ta
sks and coll
ect t
he
results; finally get the
initial process results.
Duri
ng the proce
s
s of tag data with Ma
p f
unction, the tasks sh
oul
d be decomp
o
se
d. The
sizes of obtai
ned sub-ta
sks are
simila
r whi
c
h can be
describ
ed a
s
s
p
lit
. Each indep
ende
nt sub
-
task ca
n
be d
i
vided
into co
rre
sp
ondi
ng key
value
pai
r
(1
,
1
)
key
va
l
u
e
whi
c
h
will be transmitted to
Map fun
c
tion.
After the pro
c
e
ss, we can
get the new
key value pai
r
(2
,
2
)
ke
y
v
al
ue
w
h
ic
h
is
a
l
s
o
the input of Redu
ce fun
c
tio
n
. After proce
ssi
ng t
he inp
u
t data, the correspon
ding
key-value p
a
i
rs
(3
,
3
)
ke
y
v
al
ue
ca
n be
obt
ained, a
nd t
hen
acco
rdin
g to the
act
ual ne
ed
s o
f
use
r
s du
ri
ng
operation, the
pairs a
r
e o
u
tputted to the target lo
cation
in HDFS dat
aba
se.
The ste
p
s
of method
comb
ining Ma
p fun
c
tion
with the
Redu
ce fu
nction to pass d
a
ta set
c
o
ns
tituted by network
c
h
arac
teris
t
ic
to the dis
t
ribute
d
netwo
rk file
syst
em are shown as follo
ws:
Figure 3. Initializatio
n of Data Processin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 4, April 2014: 2697 – 2
706
2704
Acco
rdi
ng to the method describ
ed above,
the multi-taskin
g
multi-point mappin
g
decompo
sitio
n
techniqu
e
ca
n be
u
s
ed in
ma
ss label
ca
rd
matchin
g
ta
sks an
d pe
rf
orm
cal
c
ulatio
ns.
Data
pa
ckets co
rrespon
din
g
to the
net
work op
erating
ch
ara
c
te
risti
c
s are extract
e
d
from HDFS cach
e, and the
similarity ma
tching p
r
o
c
e
s
s is carried o
u
t on the rem
a
ining ele
m
e
n
ts
in the data
b
a
se. T
he results obtai
ned
are
reo
r
d
e
red an
d re-di
v
ided to obt
ain the
simil
a
rity
sortin
g of th
e
keys. T
he first
P
key valu
e p
a
irs are extra
c
ted from the
above d
a
ta th
en written to
the HDFS.
5. Test
Analy
s
es
The purpose of testing is to verify the f
easibility of the system
.
This paper
takes a
vocation
al
co
llege fo
r
exa
m
ple to
verif
y
the p
r
op
osed
strate
gy o
f
allocation
o
f
acq
u
ired ta
sks
and no
de co
mmuni
cation
resou
r
ce sch
edulin
g.
Manag
eme
n
t modul
e
sele
cts ta
sk with
the hig
h
e
s
t
prio
rity from
the sch
edulin
g task
table Task_li
s
t and publi
s
hes ta
sk to a
b
le CA1,
CA
2, and CA3 a
c
cordi
ng to the inform
atio
n in
the kno
w
le
dg
e base by multicastin
g
way
.
Once the
CA1, CA2, an
d CA3
re
cei
v
ing ta
sk inf
o
rmatio
n of
a card
issue
d
, the
information will be stored in Task_li
s
t as a reco
rd. Reference database
tasks related data,
pre-
sched
uling re
sults of p
r
o
c
e
ssi
ng unit CA
1 is sh
own in Figure 6
Thro
ugh a
scheduli
ng, three unit Agent
return
biddin
g
informatio
n
to the Mana
gement
Agent (i
ncl
udi
ng bi
d valu
e,
the loa
d
rate
and
com
p
leti
on time
), the
pro
c
e
s
s is sh
own
in Fi
gu
re
1.
Finally, Man
ageme
n
t Ag
ent sele
cts a
nd n
o
tifies u
n
it CA1to
co
mplete
Ord
e
r1. Wh
en
get
the
reply from
CA1, CA1
will
be fo
rmally
a
ssi
gne
d to
co
mplete th
e ta
sk.
The
simul
a
tion
re
sults
are
s
h
ow
n
in
F
i
gu
r
e
7
.
Manag
eme
n
t Agent sel
e
ct
one of two l
o
sin
g
Agent
units a
s
a b
a
c
kup p
r
o
c
e
s
s, and the
relevant i
n
formation
will b
e
save
d, in o
r
de
r to
redu
ce the p
r
o
c
e
s
sing tim
e
of t
he a
cci
dent
and
redu
ce th
e a
cci
dent da
ma
ge. Mean
whil
e, the mana
gement Ag
en
t also noti
c
e
s
unit CA
2 a
nd
CA3 to end th
e missi
on ne
gotiation a
ssi
gnment.
Figure 4. Unit
Cell Agent1
Sched
uling P
r
ocess
Figure 5. Card Task Sch
e
d
u
ling Simulati
on
Re
sults
Experimental
re
sults sho
w
that the
sma
r
t card
sy
ste
m
is
able
to
complete th
e
expecte
d
sched
uled tasks. Althou
gh the
re
su
lts are not
the best sche
duling, the solutio
n
and
comm
uni
cati
on time of M
ASS scheduli
ng problem ar
e
contai
ned i
n
the
whole processing
course
of the workpi
ece
s
, the sol
u
tion time is short
and h
a
s f
a
st re
spo
n
se, so it has cert
ain advanta
g
e
s
whe
n
applie
d
to dynamic sche
duling.
6. Conclu
sion
Camp
us
ca
rd syste
m
is a larg
e-scal
e info
rm
ation
integratio
n
system
whi
c
h is the
system of
software
-
b
a
se
d and h
a
rd
ware an
d terminal e
quip
m
ent su
pple
m
ented. It is an
importa
nt pa
rt of digital
ca
mpus. At thi
s
stag
e,
si
ne t
he n
u
mbe
r
of
cam
p
u
s
ca
rd u
s
er in
cre
a
s
e,
tasks i
s
more
and
mo
re
diversity, a
nd i
n
the sp
ec
ific
pe
r
i
o
d
th
e us
ag
e
pe
ak
eme
r
g
e
s
.
T
h
er
e
f
ore
,
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
The Optim
a
l De
sign of Co
mm
unication
Module fo
r Cam
pus Sm
art Card
(Zhu Li
n)
2705
it can e
a
sily l
ead to la
rge
-
scale
sign
al tags
to
sei
z
e t
he commu
nication chan
nel
and imp
a
ct t
he
camp
us life a
nd the t
r
an
sa
ction m
ana
ge
ment. It i
ndicates that t
r
ad
itional net
wo
rk a
r
chitectu
re
s
can
not fully a
dapt to
the
n
e
w
era of
the
develo
p
ment
nee
ds of th
e
ca
mpu
s
ca
rd
. It is a
n
urg
e
n
t
need to esta
b
lish a flexible, agile netwo
rk se
rvice
e
n
vironm
ent. So,
it is very necessary to use
a
rea
s
on
able
calcul
ation me
thod to optimize the re
so
urce sch
eduli
n
g
of virtual network card.
This pa
pe
r introdu
ced
a multi-tasking a
nd m
u
lti-point ma
pping de
co
mpositio
n
techni
que. It mainly u
s
e
s
the M
a
p
R
edu
ce
co
mp
uting mod
e
l
to quickly
decompo
se
the
informatio
n o
f
mutations t
ags i
n
the
card n
e
two
r
k. MapRedu
ce
is a p
r
o
g
ra
mming mo
de
l
prop
osed by
Googl
e for
distribute
d
pa
rallel comput
i
ng of la
rge
-
scale data. M
a
p
R
ed
uce mod
e
l is
inspi
r
ed
by
map
and
redu
ce
fun
c
tions commo
nly used in
functio
nal
prog
ram
m
ing
．
A
Map
／
Redu
ce job usually splits the in
p
u
t data set
int
o
indep
end
en
t chun
ks
whi
c
h are p
r
o
c
e
s
sed
by the map tasks in a
co
mpletely parallel mann
er
．
The re
du
ce
tasks me
rge
all intermedi
ate
values
gen
erated by the
map ta
sks
．
Users
only dev
ote them
selv
es to
how to
spe
c
ify the m
a
p
function
s an
d red
u
ce fu
nction
s
．，
Th
e details
of partitionin
g
the input d
a
ta
sche
duling
the
prog
ram’
s
executio
n a
c
ross a
set of
m
a
chi
n
e
s
，，
han
dling m
a
chin
e failu
re
s
a
n
d
ma
nagin
g
t
h
e
requi
re
d int
e
r-m
achine
comm
uni
cati
on a
r
e ta
k
en
care of
by the
ru
ntime sy
ste
m
o
f
MapRedu
ce
．
It reduces t
he overall diff
iculty of prog
ramming in a l
a
rge extent.
MapRedu
ce f
unctio
n
ha
s three di
stin
ct advant
ag
es: first, com
p
a
r
e
d
to the other parallel
prog
ram
m
ing
model
s, Ma
pRe
d
u
c
e ha
s highe
r prog
ramming
effici
ency, which has tran
spa
r
ent
details
whe
n
cal
c
ulatin
g the parallel, loa
d
balan
ci
n
g
a
nd fault tolerance of tasks; seco
nd, it can
fully and efficiently utilize
the resou
r
ces of ea
ch
machi
ne in t
he clu
s
te
r, and is
suitabl
e for
pro
c
e
ssi
ng
computing ta
sks
of larg
e-scale data i
n
th
e clu
s
ter, it m
a
ke
s the
giga
bit-level data
can
be ope
rated
in an ordin
a
ry PC, which coul
d only be operated in large com
m
ercial ha
rd
ware
formerly; third, good fault
handlin
g me
cha
n
ism
allo
ws M
a
p
R
ed
u
ee with hi
gh
reliability. Th
ese
advantag
es
make it b
e
co
me the main
strea
m
programming m
o
d
e
l of a variet
y of computi
ng
platform
s in
creasi
ngly, whe
n
faci
ng l
a
rg
e
-
scale
data
computing
task, Ma
pRedu
ce p
r
og
rammi
ng
model will pl
a
y
a more imp
o
rtant rol
e
.
Simulation re
sults
sho
w
th
at this de
com
positio
n tech
nique u
s
e
d
in
this mappi
ng
can be
an effective solution to ana
lyze large dat
a sets
and
ot
her p
r
obl
em
s. And the sp
e
ed of sol
u
tion
is
quick a
nd the
re
spon
se tim
e
is
sho
r
t. It is capa
ble
of rational man
a
gement
a
nd config
uratio
n of
netwo
rk
re
so
urces. It can
also re
du
ce
network
traf
fic disruption,
provide a re
feren
c
e for the
resou
r
ce
sch
edulin
g of vi
rtual
network and
offer n
e
w i
dea
s to
optimizatio
n
desi
gn
of
ca
rd
comm
uni
cati
on mo
dule. T
hus, th
ere
is
certai
n p
r
a
c
ti
cal
appli
c
atio
n si
gnificance
.
Mean
while,
the
system can effectively
promote
the proce
s
s of sch
ool edu
catio
n
informati
z
a
t
ion and digi
tal
camp
us con
s
truction, i
m
prove man
age
ment leve
l
of scho
ol, en
h
ance
servi
c
e
quality of
b
a
ck
office, achi
eve downsi
z
ing
to improve effici
en
cy, and become a
n
integral p
a
r
t of staff and
stude
nts’ working, stu
d
ying
and living.
Ho
wever, du
e to the research of re
source ma
na
gement an
d
mapping m
e
thod of
camp
us
card
network is still in early stage,
and the
con
s
traint
s of rese
arch l
e
vel and ene
rgy
limited, there must be al
so
many issue
s
that
need furt
her stu
d
y an
d improvem
e
n
t in this pap
er.
Future
research will be available
from the following two aspect
s:
(1) Thi
s
p
a
p
e
r p
r
e
s
ent
s
a multi-ta
skin
g and
multi
-
point ma
ppin
g
of virtual
reso
urce
sched
uling m
e
thod, but the details
of the desi
gn is
not sufficient,
such a
s
for
trouble
s
h
ooti
n
g
,
mobility man
ageme
n
t, dynamic
re
sou
r
ce
sched
ulin
g algo
rithms.
In many usa
ge sce
nari
o
s,
the
basi
c
p
r
in
cipl
e of the reso
urce
scheduli
ng is
th
e
sam
e
, but the
sp
eed a
nd
stabi
lity perform
an
ce
are very different, So for trouble
s
ho
oting,
mobility manage
me
nt applicatio
n scena
rio
s
, the
developm
ent of appro
p
riat
e resou
r
ce sche
duling al
g
o
rithm ha
s great resea
r
ch value, but also
the one of ne
xt step in this resea
r
ch wo
rk.
(2)
The
effective distin
gu
ish b
e
twe
en
physi
cal
re
so
urces is th
e
importa
nt factor for
improvem
ent
of virtual n
e
twork
mapp
ing alg
o
ri
thm
.
This
pap
er studie
s
th
e
link n
ode
-b
a
s
ed
pre
s
sure and
the pre
s
sure
of the virtual netwo
rk o
p
timization m
a
p
p
ing alg
o
rith
m, and ma
ki
ng
the virtual ne
twork map
p
in
g pro
c
e
s
s ca
n be ba
sed
o
n
t
he cha
r
a
c
t
e
rist
i
cs
of the virtual netwo
rk
to allo
cate m
o
re t
han t
en
pairs of
physi
cal
re
sou
r
ce
s. Therefore, to imp
r
ove th
e pe
rform
a
n
c
e of
virtual netwo
rk mappi
ng al
gorithm i
s
the
next step of this work.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 4, April 2014: 2697 – 2
706
2706
Referen
ces
[1]
K
w
a
n
g
C
h
e
u
l
Shin, K
y
u
Jo
n
g
Oh. Smartca
r
d-Base
d R
e
m
o
te Auth
entic
ation
Schem
e Pr
eservi
ng
Use
r
Anon
ym
it
y
.
IJIPM
. 2013; 4(2)
: 10-18.
[2]
MA Jia
n
-ho
ng,
JI Li-
x
ia. Stu
d
y
o
n
A
gent
Immune
Net
w
o
r
k Mon
i
torin
g
S
y
stem
Mod
e
l
.
Co
mp
ute
r
Simulation
. 20
13; 30(5): 2
13-
216.
[3]
A Bel
y
aev, AA
Leksikov, AM Serzhantov, V
V
T
y
ur
nev, Ya
F
Bal’va, An
A Leksikov. Band
pass filte
r
w
i
t
h
a
n
ultra-
w
i
de stop
ba
nd
d
e
sig
ned
on m
i
niatur
ized c
o
a
x
ial res
o
n
a
tors.
Journ
a
l of C
o
mmu
n
icati
ons
T
e
chno
logy an
d
Electron
ics.
201
3; 58(2): 11
0-11
7.
[4]
W
U
Chun-q
i
o
ng. Net
w
ork
Intrusion D
e
te
ction Mod
e
l
Based o
n
F
eature Sel
e
ctio
n.
Compute
r
Simulation
. 20
12; 29(6): 1
36-
139.
[5]
Shen
g-Ch
in Yu
, F
ong-Lin
g
F
u
, Chia-
j
en T
i
ng,
Hs
ing-
Chu
an
Lu. Older-Us
e
r
s
Acceptanc
e of Smartcar
d
Pay
m
ent S
y
stems: An Investigat
io
n of an Ol
d-street Vend
e
r
s.
JCIS
. 2013; 3(3): 110-1
17.
[6]
EV Sukhon
in. On the issue
of combin
ed a
c
tive-
pass
i
ve radi
o soun
di
ng
of
precipitati
o
n on rad
i
o-
occultati
on p
a
ths.
Journa
l of Co
mmun
icati
o
ns
T
e
chno
lo
gy and El
ectron
ics
. 2013; 58(
2): 124-
127.
[7]
Jian-mi
ng C
u
i,
Z
ong-she
ng
L
a
i, Xiao-
ju
n Z
hang.
Cr
ypt
ana
l
y
sis
and Impr
o
v
ement of A R
e
mote User
Authentic
atio
n Scheme Bas
e
d
on D
y
nam
ic ID Using Sm
art Card.
JDCTA
. 201
3; 7(1): 828
-837.
[8]
SHAN Don
g
-h
ong, Z
H
AO Wei-tin
g
. Resear
ch on
Intrusio
n Detectio
n System Ne
ural
Net
w
orks a
n
d
Princip
a
l C
o
mp
one
nt Anal
ys
is.
Computer Si
mulati
on
. 20
11; 28(6): 15
3-1
5
6
.
[9]
Jian
g Z
hen
g-T
ao, Z
hou M
e
i-
Ron
g
, W
en Li.
An Effi
cient Au
thenticati
on Sc
heme
w
i
th F
o
r
w
a
r
d S
e
curi
t
y
for ST
B-based
DT
V.
AISS
. 2013; 5(7): 17
2-1
77.
[10]
HUANG Ka
i- f
eng, Z
H
AO T
ao.
Ap
plic
atio
n
of Particl
e
S
w
a
rm Optimiz
a
tion
Cl
usterin
g
in
Net
w
o
r
k
Securit
y
.
C
o
mputer Si
mu
latio
n
. 2012; 2
9
(12)
: 144-14
7.
[11]
VI Zubkov, VI
Shche
g
lov. T
he
magn
etic sus
c
eptib
ilit
y of a compos
it
e med
i
um cons
isting
of anisotro
p
i
c
ferrite partic
l
es
w
i
t
h
differ
entl
y
or
der
ed or
ie
ntations
of an
i
s
otrop
y
a
x
es.
Journ
a
l of C
o
mmu
n
icati
o
n
s
T
e
chno
logy an
d
Electron
ics
. 201
3; 58(2): 14
3-14
8.
[12]
AS Zav’y
a
lov,
VI Koshelev, VM
T
a
rnovskii.
Radi
atio
n po
wer loss i
n
the
output u
n
it
of the multi
w
a
v
e
Cher
enkov g
e
n
e
rator.
Journ
a
l
of Commun
i
cat
i
ons T
e
ch
nol
og
y and Electro
n
i
c
s
. 2013; 58(
2)
: 149-15
7.
[13]
Ren-J
unn
H
w
a
ng, F
e
n
g
-F
u S
u
, Yan
g
-Yi
Ch
en.
A
Ne
w
T
w
o-factor D
y
n
a
m
ic ID-bas
ed
Remote
User
Authentic
atio
n Scheme.
JCIT
. 2013; 8(3): 8
3
7
-84
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.