TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol.12, No.6, Jun
e
201
4, pp. 4557 ~ 4
5
6
2
DOI: 10.115
9
1
/telkomni
ka.
v
12i6.540
1
4557
Re
cei
v
ed
De
cem
ber 2
8
, 2013; Re
vi
sed
F
ebruary 25,
2014; Accept
ed March 1
1
, 2014
Security Evalu
a
tion for RFID System: Security
Evaluation Index Architecture and Evaluation Model
Heng
fen
g
Lu
o*
1
, Ruiqi Liu
1
, Yingkai
Wang
1
，
Jin
g
hao Ch
en
2
1
T
he 5th Electronics R
e
searc
h
Instit
ute of the Ministr
y
of Informati
o
n
Indus
tr
y
of Chi
na, Guan
gd
ong,
Guangz
ho
u 51
061
0
2
T
he net
w
o
rk s
e
curit
y
an
d def
end
ing te
am of Chon
gqi
ng Pu
blic Sec
u
rit
y
B
u
rea
u
,
Cho
ngq
in
g, Ch
ong
qin
g
,40
1
1
2
1
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: lhf@cepr
ei.bi
z
A
b
st
r
a
ct
With the rapid
development and a
pplication
of RFID syst
em
, people would pay
m
o
re
attention
t
o
the secur
i
ty of RF
ID, w
h
ich is the ess
ent
ial iss
ue
and
key techn
o
lo
g
y
in RF
ID sys
tem. T
h
is p
a
p
e
r
establ
ishes th
e eval
uati
on in
dex arch
itectur
e
of RF
ID security, then acc
o
rdi
ng to thes
e ind
e
xes for th
e
evaluation of t
he RFID system
sec
u
rity which bas
ed on fu
z
z
y
sy
nthetic
evaluation model, this paper
prop
oses a A
H
P (T
he Anal
ytic Hierarchy
Process) for weig
hs, and the
n
app
lies fu
zzy compr
e
h
ensi
v
e
eval
uatio
n met
hod
for mor
e
accurate and effective
ev
a
l
u
a
tion res
u
lts, w
h
ich is use
d
to evalu
a
te th
e
security of RF
ID system. Our secu
rity ev
alu
a
tion
in
dex
architectur
e
and
eval
uati
o
n
mo
del w
i
th R
F
ID
system is very
practica
l and
effective in the re
al life.
Ke
y
w
ords
: index system
, evaluation
m
o
del,
fu
z
z
y
c
o
mprehensive ev
aluation
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
RFID
(Radio
frequ
ency i
d
entification
)
t
e
ch
nol
o
g
y, which
uses
sp
atial co
uplin
g
of RF
sign
al o
r
tra
n
s
missio
n p
r
o
pertie
s
of
ra
d
a
r, en
ablin
g t
he a
u
tomatic
obje
c
t re
co
gn
ition, re
cog
n
ition
without hum
a
n
intervention
,
works in a v
a
riet
y of harsh environ
men
t
s, and othe
r advantag
es.
RFID inc
l
udes
three
part
s
: RF
ID tag
s
, RFID rea
der a
nd RFI
D
data
processin
g
system.
1) RFID
ta
gs,
by
co
upling
comp
one
nts and ch
ip
s, which co
ntain
s
built-in ante
nna,
an
d
use
s
for
com
m
unication wi
th RF antenn
a of reade
r.
2) RFI
D
re
ad
er, whi
c
h rea
d
s or
write
s
tag inform
atio
n of device
s
.
3) Data pro
c
essing
syste
m
for RFID,
whic
h is
RFID devices
for data tra
n
s
missio
n,
pro
c
e
ssi
ng a
nd appli
c
atio
ns.
RFID se
cu
rity
issu
es
in
cl
uding som
e
basi
c
se
curit
y
features, such as confi
dentiality,
integrity, avai
lability, authentic
ation, authori
z
ation and anonymity
. RFID sy
stem of the
security
probl
em ha
s
cau
s
e
d
parti
cular con
c
e
r
n
of all aspe
cts, comprehe
nsive eval
uatio
n of the secu
rity
of RFID syst
ems is al
so the hot issue
s
in
this are
a
,
and this paper would a
methodol
ogy
to
evaluate RFI
D
syste
m
se
curity as follows.
1.1.
Ev
aluation Index Archite
c
ture M
odel
This pa
pe
r wi
ll adopt a fuzzy comp
re
he
nsive
evaluati
on method fo
r se
cu
rity evaluation
of R
F
ID
s
y
s
t
em. The pr
oc
es
s
of fuzzy c
o
mpr
e
he
ns
ive evaluation method is
to s
t
ar
t w
i
th
qualitative fuzzy selectio
n
and then ob
tain resu
lts b
y
operation throu
gh fuzzy
transfo
rmati
o
n
prin
ciple.
Whe
n
eval
ua
ting security
of RFI
D
sy
st
ems, a
s
th
ey
involve a va
ri
ety of differe
nt facto
r
s
and prope
rtie
s of system, so the ev
alua
tion must take account of
all its fields a
nd decre
ase the
impact of
sub
j
ective facto
r
s. So this pa
per divi
d
e
levels of the
RF
ID system
se
curity, and
e
a
ch
level would
o
w
n
so
me fa
ctors with
corresp
ondi
ng
weight, which
evaluate th
e
se
curity
of th
e
entire RFID system com
p
rehen
sively. It is difficult
to comp
are the pro
s
and
con
s
of the orde
r of
impact fa
ctors by usi
ng single-l
e
vel factors evalu
a
tion, whi
c
h i
s
difficult to identify uniform
weig
hts, so y
ou ca
n use the multi-
level
evaluation m
e
thodol
ogy in
stead.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4557 – 4
562
4558
This
se
cu
rity factors
can
b
e
divided
into
two
level
s
: th
e first level
consi
ders fo
ur
indexe
s
of RFID Security: phys
i
cal s
e
curity, c
o
mmunic
a
tion
s
e
c
u
rity, data s
e
curity, performanc
e
, and
indexe
s
su
bdi
vided into se
cond level are indicated in T
able 1.
Table 1. RFI
D
System Evaluation Ind
e
x
Architectu
re
First Level
（）
R
Weight
Second
Level
（）
C
Weight
1. Ph
y
s
ical
Security
U
1
1.
Tags
U
11
2.
Securit
y
of re
ade
r devices
U
12
3. Fai
l
u
re
rec
o
v
e
r
y
U
13
2. Communication
Securit
y
U
2
1.
Interfer
ence bet
w
e
en rea
der a
n
d
writer device
U
21
2. Access
contr
o
l
U
22
3.
Tags’ encr
y
ption
and decr
y
ption
U
23
4. Protocol
securit
y
U
24
3. Data
Securit
y
U
3
1.
Data
encr
y
ption
U
31
2. Data
integrit
y
U
32
4. Performance
U
4
1.
Tag
capacit
y
U
41
2. Access
t
i
me
U
42
3.
Reader la
rgest a
ccess speed
U
42
4.
Reader la
rgest a
ccess capacity
U
44
2. RFID Sy
stem Securit
y
Ev
aluation Mode
l Ba
se
d on Fuzz
y
Compre
hens
iv
e Ev
aluation
2.1. Ev
aluation Flo
w
The flow
cha
r
t of this evaluation model i
s
sh
ow in Fig
u
re 1.
Figure 1. Evaluation Flo
w
Chart
2.2. AHP for
Weigh
t
s
Determinatio
n of weight coefficient
s pl
ay an
import
ant role in
RFID system
s security
evaluation m
odel, o
c
cupie
s
an
impo
rtan
t position,
We
ight co
efficien
ts of chan
ge
s in the valu
e to
be evaluate
d
will lead directly to cha
nge
s in ord
e
r
of evaluation, whi
c
h di
rectly affect
the
evaluation re
sults.
This p
ape
r use
s
hie
r
a
r
chy analysi
s
pro
c
e
ss to d
e
termin
e wei
ghts. The A
H
P (Th
e
Analytic Hi
erarchy Pro
c
e
s
s) is
a
Unite
d
States University of Pittsbu
rgh
Profe
s
sor T. L. Sa
aty
prop
osed a
combi
nation
of qualitative and qua
ntit
ative analysi
s
of multi-ob
jective de
cisi
on
method. It h
a
s
cha
nge
d the p
r
eviou
s
optimizatio
n tech
niqu
es
ca
n only ad
dre
ss th
e issu
e
of
quantitative a
nalysi
s
of th
e
tradition
al
co
nce
p
ts, a
nd
h
eade
d into
th
e lon
g
stay in
many
scienti
f
ic
studie
s
on th
e qu
alitative
analysi
s
of le
vel of terr
itory, provide
s
a
simple
meth
o
d
for qu
antita
t
ive
analysi
s
of
n
on-q
uantitative event. In th
e analytic
hie
r
archy p
r
o
c
e
s
s ap
plied
to the pe
rforman
c
e
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Secu
rity Eval
uation for RFID System
: Se
curit
y
Eval
uat
ion Inde
x Architecture… (Hen
gfeng L
u
o
)
4559
evaluation, its bigg
est ad
vantage is
th
e accu
rate d
e
termin
ation
of
weight
s of indexes
so t
hat
indicators co
uld be rea
s
o
nably refle
c
t the relative importan
c
e
and laid the
foundation
s
of
sci
entific pe
rforma
nce evaluation
syste
m
. The flow chat of AHP
method for weighting proced
ure
is show in Figure 2:
Figure 2. The
Proce
s
s of AHP Method to
Determi
ne th
e Weig
hts
2.2.1. Cons
tr
uct Jud
g
em
ent Ma
trices
Acco
rdi
ng to
Table 1, it sh
ows that in t
he
hie
r
a
r
chi
c
al relatio
n
shi
p
of Up
per
a
nd lo
wer
levels have been
ide
n
tified.
It’s
a
s
su
med
th
at
R i
s
o
ne
eleme
n
t of up
pe
r l
e
vel, wh
ose
the
effec
t
ive value of lower level is
C1,
C2... Cm.
So f
o
llow the
R
weights
c
o
rres
ponding to t
he
relative weight assigned
to the C1,
C
2,...,C
m. Index system,
C1,C2,...,Cm don’t have
fixe
d
quantitative relation
ship,
which
ne
ed
expert ju
dgm
ent
s
fo
r imp
o
rta
n
ce
of
Ci
an
d Cj
a
c
cordi
n
g to
the 1-9
-
scal
e
assi
gnme
n
t of importan
c
e level
s
are sho
w
n in
Table 2. First level inde
x R
judgem
ent m
a
trix in [1]:
mm
m
m
m
m
r
r
r
r
r
r
r
r
r
R
...
...
...
...
...
...
...
2
1
2
22
21
1
12
11
(1)
Ri wei
ghts
co
uld be obtai
n
ed:
m
i
i
i
i
W
W
W
1
/
(2)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4557 – 4
562
4560
The app
roxim
a
tion of weig
ht vector is:
m
im
i
i
r
ri
r
W
*
...
*
*
2
1
(3)
Table 2. The
Scale Ta
ble
Scale
Definition
Introduction
1
Equally
impo
rtan
t
T
w
o elements a
r
e compared to
e
qually
impo
rtant.
3
Slightly
important
T
w
o elements co
mpare to each
ot
her. One elem
en
t is slightly
mo
re important th
an th
e
other
5
Obviously
impo
rt
ant
T
w
o elements co
mpare to each
ot
her. One elem
en
t is obviously
mo
re important t
han
the other
7
Much more
important
T
w
o elements co
mpare to each
ot
her. One elemen
t is much more important th
an th
e
other
9
Extremel
y import
ant
T
w
o elements co
mpare to each
ot
her. One elem
en
t is extremel
y
m
o
re important t
han
the other
2,4,6,8
Intermediate valu
es
The effect of i-th
factor relative to t
he i-
th factors is bet
w
een ab
ove tw
o
adjacent levels.
The main p
r
o
b
lem of Com
puting is to solve the maximum eige
nvalue of judgme
n
t matrix and
its
corre
s
po
n
d
ing eige
nve
c
tor. Wh
en relative
weig
h
t
vector is o
b
tained, i
n
o
r
de
r to
en
su
re
rea
s
on
able
concl
u
si
on of
AHP, it
req
u
ires
ch
ecking
the
con
s
iste
ncy of judgm
ent
matri
c
e
s
. So
by
usin
g con
s
ist
ency in
dicat
o
rs
1
-
m
m
-
CI
max
＝
for con
s
i
s
ten
c
y ch
eck, the bigge
r
CI is, the m
o
re
seri
ou
s i
n
con
s
iste
ncy j
udg
ment mat
r
ix i
s
. If th
e
che
c
k pa
ssed,
the ei
genve
c
t
o
r i
s
th
e
wei
ght
vector; Othe
rwises, it need
s to re
co
n
s
tru
c
t judgme
n
t matrices [2, 3
]
.
2.2.2. The Weights o
f
the
Second Lev
e
l Index (Ci)
The wei
ghts
are corre
s
po
nding to the p
r
odu
ct of its level:
Ci=Wi*Wij
(4)
2.2.3. The Weights o
f
the
Second Lev
e
l Index Corr
espondin
g
to the First L
e
v
e
l
j indi
cators i
s
esta
blished i
n
level
R, we
ights fo
r va
ri
ous ind
e
xes
coul
d b
e
obt
ained
by
usin
g the abo
ve AHP method re
spe
c
tive
ly:
9
1
i
il
1
W
＝
＝
(
il
W
is
the weigt
h
s
of firs
t index I to
the first level)
9
1
9
1
i
i2
2
;......;
W
i
ij
i
W
＝
，
＝
(5)
2.3. Fuzzy
Comprehen
s
iv
e
Ev
a
l
uation Method
The fuzzy comprehe
nsiv
e evaluation
method
[4, 5] is ba
sed
on fuzzy ma
thematics
theory a
nd th
e pri
n
ci
ple of
maximum m
e
mbe
r
ship d
egre
e
[6, 7],
whi
c
h
Con
s
id
ers the va
rio
u
s
factors a
s
sociated with th
e evaluation
to make
a comprehe
nsiv
e evaluatio
n. The evalu
a
tion
methods will take into acco
unt all relevant factors.
2.3.1. Ev
alua
tion Elements Set
All the factors a
r
e divide
d into S subset
s, whi
c
h
is
s
i
Y
Y
Y
,......,
,
2
, s
a
tis
f
ying the
condition
s
i
Y
Y
Y
,......,
,
2
,
)
(
j
i
F
Y
Y
s
i
.
E
a
ch
set
s
i
Y
i
,......,
2
,
1
,
co
uld be eval
u
a
ted
by the factor
sets of its n
e
x
t level
in
X
, that is
s
i
X
X
X
Y
in
i
i
i
,...,
2
,
1
},
,...,
,
{
2
1
, whe
r
e n i
s
the numbe
r o
f
i
Y
.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Secu
rity Eval
uation for RFID System
: Se
curit
y
Eval
uat
ion Inde
x Architecture… (Hen
gfeng L
u
o
)
4561
2.3.2. Establish Commen
t
s Set
The commen
t
set of all the facto
r
s a
r
e establi
s
h
e
d
as
m
V
,...,
V
,
V
2
1
if there
are m
comm
ent
s.
2.3.3. Fuzzy
Matrice
s
Each su
bset of
evaluatio
n factor
)
,...,
2
,
1
(
s
i
Y
i
would
be eval
uated
by sin
g
le to
o
b
tain
singl
e facto
r
evaluation m
a
trice
s
nm
k
ij
i
r
R
)
(
,
, where
m
k
n
j
s
i
,...,
2
,
1
;
,...,
2
,
1
;
,...,
2
,
1
.
k
ij
r
,
is the degree
of membership of single
factor to com
m
ent
k
V
. In a c
o
mprehe
nsiv
e evaluation,
whe
n
ma
king
a comp
reh
e
n
s
ive evalu
a
tio
n
, based
on
a
c
tual singl
e
fa
ctor evaluatio
n
matrix coul
d
be obtain
ed
based on
act
ual fields, h
e
r
e is
Delp
hi
method. The
spe
c
ific p
r
o
c
edure is: on
the
basi
s
of
man
-
ma
chin
e inte
gration
data
acq
u
isitio
n,
e
a
ch
expe
rt score th
e fa
ctors of
comm
en
ts
based on
coll
ected d
a
ta, which
scorin
g range i
s
in
the interval [0,1]. For exam
ple, each exp
e
rt
sc
ore
s
ij
X
factors, whi
c
h sho
u
ld be satisf
ied
1
V
. The degree of me
mbershi
p
is
the
corre
s
p
ondin
g
score, an
d obtain
s
sin
g
le
factor evalua
tion fuzzy ma
trice
s
i
R
.
2.3.4. Establish Weight Set
}
,...,
,
{
2
1
in
i
i
i
a
a
a
A
are the wei
g
hts of each evaluation fact
or
)
,...,
2
,
1
(
s
i
Y
i
,
whe
r
e
n
j
ij
a
1
1
. Dete
rminatio
n of
weig
ht coefficient
s i
s
very
impo
rtant; it dire
ctly affe
cts th
e
final evaluati
on re
sult
s. There
are
ma
ny comm
on method
s
of determi
ning weig
hts,
su
ch
as
AHP, binary
comp
ari
s
on
function a
nd so
n. But determini
ng
weight
s is an ever-m
ore
comp
re
hen
si
ve pro
c
e
ss,
so acco
rdin
g to the sp
ecif
i
c
ity of information se
cu
rity, weig
ht settin
g
in
pra
c
tical
ap
pl
ication
s
can
be carried
ou
t simultan
eou
sly with th
e e
v
aluation divi
sion. T
h
is
pa
per
use AHP to d
e
termin
e wei
ghts.
2.3.5. Fuzzy
Compre
hens
iv
e Ev
aluation
From ab
ove, the co
mpre
hen
sive
evaluation
vector of
)
,...,
2
,
1
(
s
i
Y
i
is
}
,...,
,
{
2
1
in
i
i
j
i
i
b
b
b
R
A
B
.
Bec
a
us
e many fac
t
ors
affec
t
th
e
evalu
a
tion results,
in orde
r to
avoid losi
ng
valuable info
rmation an
d b
e
truly unbia
s
ed, it shoul
d take into
acco
unt a variety of
factors influe
nce,
so
}
,...,
2
,
1
(
m
k
b
ik
sho
u
ld be
comp
uted by the
weig
hted ave
r
age m
e
thod.
So
ik
b
of each
}
,...,
2
,
1
(
s
i
B
i
could be o
b
tai
ned a
s
follows:
n
j
k
ij
ij
ik
m
k
r
a
b
1
,
,
,...,
2
,
1
,
(6)
2.3.6. Multi-lev
e
l
Fuz
z
y
Comprehensiv
e
E
v
a
l
uation
)
,...,
2
,
1
(
s
i
Y
i
is re
ga
rde
d
as a sin
g
le factor an
d
i
B
is regarded a
s
the singl
e fact
or
evaluation m
a
trice
s
of
i
Y
, which
are con
s
tructed
the
comprehe
nsiv
e evaluatio
n
matrix betwe
en
Y and V.
(7)
11
12
1
1
22
1
2
2
2
12
...
..
..
m
m
s
ss
s
m
bb
b
B
B
bb
b
B
B
bb
b
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 6, June 20
14: 4557 – 4
562
4562
Acco
rdi
ng to
the wei
ghts o
f
Yi in Y,
}
,...,
,
(
2
1
s
a
a
a
A
, where
s
i
i
a
1
1
. Finally, the
final comm
et vector of Y is:
(8)
3. Conclusio
n
Scientific a
n
d
effective evaluation of security
of
RFI
D
sy
st
em
s is
o
ne of the imp
o
rtant
measures to
gua
rantee
system securit
y
. This
pa
pe
r
intro
d
u
c
e
s
the
ba
sic ele
m
ents of
RF
ID
system
s and
general se
curity r
equi
re
ments an
alysis of RFI
D
systems. T
o
evaluate RFID
sy
st
em
s,
it
is impo
rt
an
ce
of
the esta
blishm
ent of
the eval
uati
on ind
e
x system of syste
m
s
se
curit
y
.
Th
e ind
e
x
sy
s
t
em is
ba
si
cally
b
u
ilt
RFID system
evaluatio
n based on
f
u
zzy
comp
re
hen
si
ve evaluation
model, which is very h
e
l
p
ful and
effective in ou
r
pratical work.
In
prac
tic
e
more s
c
i
entific
,
more e
ffic
i
ent RFID s
y
s
t
em
sec
u
rity as
sess
ment is
further direc
t
ion.
Referen
ces
[1]
Schadsc
h
n
e
id
er A, Klingsc
h
W
,
Klüpfel H, et al
. Enc
y
c
l
o
ped
ia of Com
p
le
xit
y
an
d S
y
stem Scienc
e
.
Berlin H
e
i
del
be
rg, Ne
w
Y
o
rk. 200
9.
[2]
W
u
CG, W
ang
YM, T
ang YM, et al. Vari
ab
l
e
fuzz
y rec
o
g
n
i
tion m
o
d
e
l for
the flo
od
hazar
d eva
l
u
a
tio
n
base
d
on set p
a
ir an
al
ysis.
J Northw
est A&F Univ.
201
2; 40
(1): 221-2
26.
[3]
Z
hang M, Yan
g
YQ. Performance An
al
ys
is of Au
tomob
ile
F
r
iction Cl
utch
Based o
n
Ent
r
op
y F
u
zz
y
Compre
he
nsiv
e Evalu
a
tio
n
. Advanc
ed Mater
i
als Res
earc
h
. 201
2; 479: 23
0
1
-23
04.
[4]
Hua
ng S,
Xia
H, T
an L. T
he F
u
zz
y C
o
mpr
ehe
nsive
Eva
l
uatio
n for
Infor
m
ation
Secur
i
ty.
C
o
mp
uter
T
e
chno
logy an
d
Deve
lop
m
ent
.
2010; 1: 048.
[5]
Anastassi
ou GA. F
u
zzy
mat
h
ematics: Appro
x
im
atio
n theor
y.
Springer
. 2010.
[6]
Liu
X, Pedr
ycz
W
.
Axiomatic fu
zz
y
s
e
t theor
y and its ap
plic
a
t
ions.
Sprin
ger.
2009.
[7]
Liu S, Proctor
W
,
Cook D. Using
an
integ
r
ated
fuzz
y se
t and d
e
li
berat
ive multi-cr
iteri
a
eva
l
uati
o
n
appr
oach
to fa
cilitate
dec
isio
n-makin
g
in
inv
a
sive
spec
ies
mana
geme
n
t.
Eco
l
og
i
c
al
Econ
om
i
c
s
. 201
0;
69(1
2
): 237
4-2
382.
12
(,
,
,
)
m
TA
B
t
t
t
Evaluation Warning : The document was created with Spire.PDF for Python.