Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 5
,
O
c
tob
e
r
201
6, p
p
. 2
396
~240
2
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
5.9
678
2
396
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Trust-B
as
ed Pri
v
acy f
o
r Intern
et of Things
Vera Sur
yani,
Selo
Sulistyo,
Widyawan
Department o
f
Electrical Engin
e
eri
ng & Information Technolog
y
,
Universita
s Gad
j
ahmada, Yog
y
ak
arta, Indon
esia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Dec 9, 2015
Rev
i
sed
Jun
27,
201
6
Accepte
d
J
u
l 15, 2016
Internet of Things or
widely
kn
ow
n as
IOT
m
a
kes
s
m
art objects
becom
e
act
ive part
ic
ipan
ts
in the com
m
u
nica
tion proces
s
between obje
c
t
s
and their
environm
ent. Io
T servic
es that u
tili
ze Int
e
rne
t
co
nnect
ion requir
e
solutions to
a new probl
em
:
s
ecurit
y
and priv
ac
y.
S
m
art obje
c
t
s
and m
achin
e-t
o
-m
achine
communications
in IOT now b
ecome in
ter
e
sting research
, in
cluding th
at
related
to secu
rity
. Privacy
, which is a
safe
condition in
which ob
ject is fr
ee
from interfer
e
nce from other
objects, is one
of th
e im
portan
t
aspe
cts in IO
T
.
Privacy
can be
implemente
d using various way
s
for examples by
app
l
y
i
ng
encr
y
p
tion algor
ithms, restrictio
ns on acce
ss to
data or
users,
as well as
im
plem
enting r
u
les or spe
c
if
ic
poli
c
y.
Trustable ob
ject selection is
on
e
techn
i
que to improve privacy
.
Th
e pro
ces
s
of s
e
lecting a trus
tab
l
e
objec
t can
be done based
on past activ
ities or tr
ust histor
y
of the object, also b
y
apply
i
ng a thr
e
shold value to determine
whether
an object is "trusted" or not.
S
o
m
e
res
earch
e
r
s
have s
t
ud
ied
this
app
r
oach
.
In this
s
t
ud
y,
t
h
e s
e
l
ect
ion
processes of tru
s
table ob
jects are ca
lcu
l
ated us
ing Modified
Ant Colo
n
y
algorithm. The
simulation was
performed
and r
e
sulted
in declin
ing graphic
trend bu
t st
abil
iz
ed in
c
e
rta
i
n
trust va
lue
.
Keyword:
An
t co
lon
y
algo
rith
m
IoT
Objects
Privacy
Trust
Copyright ©
201
6 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Vera
S
u
ry
ani,
Depa
rt
m
e
nt
of
El
ect
ri
cal
Engi
neeri
n
g
&
In
f
o
rm
ati
on Tec
h
n
o
l
o
gy
,
Uni
v
ersitas Ga
dja
h
m
a
da,
Jl
Gra
f
i
k
a
N
o
2,
Sl
em
an, Yo
gy
aka
r
t
a
, I
n
do
nesi
a.
Em
a
il: v
e
ra.s3
t
e1
4@m
a
il.u
g
m
.ac.id
1.
INTRODUCTION
The c
o
ncept
of
t
h
e
Int
e
r
n
et
of
Thi
ngs
(
I
O
T
)
l
a
b wa
s fi
rst
d
e
vel
o
ped
by
A
u
t
o
-I
D C
e
nt
er
at
M
I
T
i
n
1
998
. Th
e
o
b
jects, p
e
rson
s, o
r
t
h
ing
s
are id
en
tified
wi
th
a v
i
su
al rep
r
esen
tatio
n on
th
e
In
tern
et
[1
]
.
Inform
atio
n
Tech
no
log
y
(IT) in
frastru
c
ture will b
e
n
e
ed
ed
(su
c
h
as
co
m
p
u
t
ers and teleco
mm
u
n
i
catio
n
net
w
or
ks
) f
o
r
dat
a
exc
h
a
n
gi
n
g
am
ong
o
b
j
ec
t
s
or
t
h
i
n
gs
i
n
t
h
e
IOT
.
In a
g
reem
ent with the
de
fini
tion
of "t
hings
"
,
the
IOT
communication c
a
n take
place
betwee
n the
objects and the object to humans. Th
e objects that
may
consist of sens
ors ca
n be numerous
, can be either
st
at
i
c
or m
obi
l
e
, an
d ev
ery
w
here
di
st
ri
b
u
t
e
d. M
a
ny
ap
pl
i
cat
i
ons ca
n be
appl
i
e
d
i
n
I
o
T, suc
h
as
real
-t
im
e
med
i
cal
m
o
n
ito
ri
n
g
, in
tellig
en
t transpo
r
tat
i
o
n
system
, i
n
tellig
en
t app
l
ian
ce and
in
t
e
llig
en
t ag
ricultu
re.
Ano
t
h
e
r app
licatio
n
th
at is n
o
t less i
m
p
o
r
tant to
d
e
v
e
lop
in
th
e IOT is a smart g
r
id
,
wh
ere th
e estu
ary en
d
of
the existing res
earch i
n
this a
r
ea is hum
an energy sa
vi
ng for sm
art appliances. Se
ns
ors tech
no
log
i
es take an
i
m
p
o
r
tan
t
ro
le in
so
m
e
Io
T
ap
p
lication
s
.
Fo
r ex
am
p
l
e, for m
o
n
ito
ring
en
v
i
ron
m
en
tal co
nd
itio
ns purpo
ses
req
u
i
r
e
d
m
a
ny
sens
or
s;
suc
h
as h
u
m
i
di
t
y
senso
r
s,
t
e
m
p
erat
ure se
ns
ors
,
a
n
d
ot
her s
e
ns
o
r
s.
Sens
o
r
s are
al
so
neede
d
i
n
e
-
he
al
t
h
;
for e
x
am
pl
e, i
s
a bl
o
od
press
u
re sens
o
r
(s
phy
gm
o
m
anom
et
er), sen
s
or
heart
rat
e
(
E
C
G
)
,
sens
or m
u
scle
(EM
G
), and
accelerom
eter [2].
These
se
nsors
or
obje
cts should
be c
o
nnecte
d
t
o
a
public
n
e
two
r
k
or commu
n
i
cate wirelessly with
o
t
h
e
r obj
ects [3
]. Th
is co
nd
itio
n
m
a
k
e
s o
b
j
ects b
eco
m
e
v
u
l
n
e
rab
l
e
to attack [4]. Access c
ont
rol play
s an i
m
port
a
nt
rol
e
here
, t
o
cl
as
sify anyone
who is allowed to
comm
unicate or e
x
cha
n
ge
data.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Trust
B
a
se
d
Pr
i
v
acy f
o
r
I
n
t
e
r
n
et
of
Thi
n
g
s
(
V
era
S
u
ry
ani
)
2
397
Dat
a
nee
d
s a s
a
fe an
d rel
i
a
bl
e
m
eans t
o
be
sent
bet
w
een objects beca
use ob
jects are
vulnera
b
le t
o
b
e
ing
attack
ed [5
]. Asp
ects o
f
priv
acy, con
f
i
d
en
tia
lity, a
u
th
en
ticity an
d
in
tegrity in
th
e IOT are am
o
n
g
param
e
t
e
rs t
h
a
t
can
be
used
as secu
ri
t
y
per
f
o
r
m
a
nce i
n
e
x
cha
n
ge
of
su
ch
dat
a
. M
a
ny
way
s
ca
n
be
use
d
t
o
im
ple
m
ent privacy [6], eithe
r
centra
l
i
zed o
r
di
st
ri
b
u
t
e
d
.
The use o
f
key
di
st
ri
b
u
t
i
on i
s
one ce
nt
ral
i
zed wa
y
t
o
im
prov
e pri
v
acy
;
bot
h sy
m
m
e
t
r
i
c
and asym
m
e
t
r
i
c
key di
st
ri
but
i
o
ns.
Whi
l
e
gi
vi
ng t
h
e val
u
e
of t
r
u
s
t
or
scori
ng
has a d
i
ffere
nt
way
t
o
im
prove
pri
v
a
c
y
,
and t
h
i
s
wa
y
use di
st
ri
b
u
t
e
d way
i
n
st
ea
d
of ce
nt
ral
i
zed
one
.
Obj
ects t
h
at are po
ten
tial to
act so
m
e
attack
s or
h
a
v
e
carried
o
u
t
t
h
e attack
will b
e
g
i
v
e
n a v
e
ry low
score
of
t
r
ust
val
u
e, ca
usi
n
g ot
her
ob
ject
s t
o
rec
ons
i
d
er t
h
ei
r
deci
si
on
on m
a
ki
ng t
h
e co
n
n
ect
i
on t
o
t
h
ese o
b
j
ect
s
.
Lik
e
wise, an
ob
j
ect th
at
was
n
e
v
e
r carried
ou
t an
attack
and
h
a
s a
go
od
track
reco
rd
will b
e
g
i
v
e
n
t
h
e valu
e
of hi
gh trust.
Trust m
e
tric c
a
lculations
bec
o
m
e
an
im
portant issue
to
be
res
o
lv
e
d
in the IOT
because
the
num
bers o
f
co
nnect
e
d
o
b
j
ect
s are dy
nam
i
cal
l
y
change
d o
r
t
h
ey
can m
ove t
o
di
ffe
rent
net
w
o
r
ks f
r
ee
l
y
. To
m
a
ke dat
a
e
x
c
h
an
ge
bet
w
ee
n
o
b
ject
s
can
be
d
one
safely it
requires
appropriate re
commendations
.
Ob
ject
s
wi
t
h
m
a
ny
expe
ri
en
ces o
f
vi
si
t
i
ng
som
e
net
w
o
r
k
s
causi
ng m
a
ny
rec
o
m
m
e
ndat
i
ons
can
b
e
give
n to the object.
In
othe
r words,
hie
r
arc
h
ical trust scoring is nee
d
ed
t
o
m
a
ke an accurate trust scoring.
Fo
r ex
am
p
l
e, tr
u
s
t sco
r
ing
can
b
e
fo
und o
n
t
h
e fr
iend’
s r
e
co
mm
en
d
a
tio
n
in
so
cial
n
e
two
r
k
s
su
ch
as
Facebook a
nd
Linke
d In. In IOT, this calcul
a
tion is
useful for
provisioning
rec
o
m
m
e
nda
tions am
ong
objects
to
ex
ch
an
g
e
data secu
rely.
M
e
t
a
heuri
s
t
i
c
m
e
t
hods
suc
h
as Ge
net
i
c
Al
g
o
ri
t
h
m
,
Ant
C
o
l
o
ny
,
Harm
on
y
Search
, Pa
rt
i
c
l
e
Swarm
Op
tim
izat
io
n
,
can
b
e
u
s
ed
to assess tru
s
t
metric [7
]-[9
] objectively, instead of
s
u
bjecti
v
ely [10],[11]. But
not
m
a
ny
of t
h
ese st
udi
es ha
v
e
sol
v
e
d
t
h
e p
r
obl
em
s of t
r
ust
m
e
t
r
i
c
s for dy
nam
i
cal and hi
erarc
h
i
cal
ob
je
ct
s.
Metaheuristic m
e
thod will be
used in this study for sc
or
i
n
g or assessing trust
m
e
tric for dynam
i
c and m
obile
ob
ject
s i
n
I
o
T.
The usa
g
e of t
h
i
s
m
e
t
hod f
o
r
t
r
ust
m
e
t
r
i
c
s cal
c
ul
at
i
on i
s
ex
pect
ed t
o
be a cont
ri
b
u
t
i
on t
o
t
h
e
pri
v
acy aspects
of IoT
.
The
pape
r i
s
o
r
ga
ni
zed a
s
f
o
l
l
ows:
Sect
i
o
n
II
descri
bes t
h
e defi
ni
t
i
on
of
t
r
ust
a
nd
p
r
i
v
acy
i
n
Io
T
and al
so
Ant
C
o
l
o
ny
al
gori
t
hm
, Sect
i
on II
I descri
bes t
h
e
sim
u
l
a
t
i
on envi
r
onm
ent
,
Sect
i
on IV e
xpl
ai
ns t
h
e
resul
t
of
si
m
u
lat
i
on a
n
d
i
t
s
an
al
y
s
i
s
, and
fi
na
l
l
y
Sect
i
on V
c
oncl
ude
s t
h
e
c
oncl
u
si
o
n
s.
2.
PRIV
ACY
AND T
RUS
T IN T
H
E I
N
TER
NET
OF THIN
GS
An object ca
n be
re
pres
ente
d
digitally, and
whe
n
it is co
nn
ected to the In
tern
et, th
en
it can
b
e
accessed a
nd c
ont
rolled
from an
ywhere
. Some applications take be
nefit
of these ki
nds
of
obj
ects: Inte
lligent
Tran
sp
ort
a
t
i
o
n
Sy
st
em
(ITS
)
whe
r
e
c
o
m
m
u
n
i
cat
i
on occ
u
r
s
bet
w
ee
n ve
hi
cl
es
(V
2
V
) or
bet
w
ee
n Vehi
c
l
e
and
Infrast
ru
ct
u
r
e (V2I),
h
ealth
m
o
n
itoring
i
n
tele
m
e
d
i
cin
e
where attache
d
se
nsors
can be
re
m
o
tely
m
onitore
d,
and gene
ral
machine
-
to-m
ach
i
n
e
com
m
uni
cat
i
ons.
P
r
i
o
r
t
o
t
h
e c
o
m
m
uni
cation esta
blishm
ent, an
obje
ct can
choose
ot
her
objects
that
a
r
e t
r
ust
a
bl
e o
r
n
o
t
.
2.
1.
Pri
vac
y
IETF
In
tern
et
Security Glo
s
sary d
e
fi
n
e
s
p
r
i
v
acy as "th
e
rig
h
t
o
f
an
en
tity (u
su
ally a p
e
rson
), acting
o
n
t
h
eir
o
w
n
b
e
h
a
l
f
, to
d
e
term
in
e th
e ex
ten
t
to
wh
ich
it
will in
teract with
its env
i
ron
m
en
t, in
clu
d
i
n
g
t
h
e
ex
ten
t
t
o
wh
ich
th
e en
tity is willin
g
t
o
sh
are in
fo
rm
ati
o
n
ab
ou
t t
h
em
sel
v
es
with
o
t
h
e
rs" [1
2
]
. Based
o
n
th
is
defi
ni
t
i
on,
p
r
i
v
acy
can
be a
p
p
l
i
e
d i
n
[6]
:
a.
Devices
Devices
in IOT are any type of de
vices that is
connecte
d
to The
In
tern
et, in
clud
ing
sens
ors
,
act
uat
o
rs
,
cam
e
ras
an
d
a
n
y
ot
hers
net
w
or
ke
d devi
ces
. If
t
h
e de
vi
ces are
m
a
ni
pul
at
e
d
fo
r
s
p
eci
fi
c pu
r
poses
,
t
h
en
t
h
e
data tra
n
sm
ission m
a
y becom
e
unsafe
.
b.
C
o
m
m
uni
cat
i
o
n
Data e
n
cryption ca
n
be
done
to e
n
sure
confide
n
tia
lity
during
t
h
e
t
r
ansm
ission proce
ss takes place. For
th
is v
a
riou
s encryp
tion
algorith
m
s
can
b
e
app
lied
.
c.
S
t
o
r
ag
e
Access con
t
ro
l
is o
n
e
o
f
th
e
mech
an
ism
s
to
m
o
n
ito
r who
is allo
wed
to
access th
e d
a
tab
a
se. En
cryp
tion
of
d
a
ta is
u
s
efu
l
en
oug
h to
cop
e
with
th
e in
tru
s
i
o
n of t
h
e
d
a
tabase.
d.
Pr
ocessing
The
dat
a
m
u
st
be e
n
s
u
re
d a
safe
pr
ocess s
o
ot
her
can
n
o
t
m
i
suse t
h
e
d
a
t
a
. Di
gi
t
a
l
R
i
ght
M
a
nag
e
m
e
nt
(DRM) is a m
e
th
od
th
at can be u
s
ed
h
e
re.
In
gene
ral
,
t
h
e
pri
v
acy
as
pec
t
i
s
cl
osel
y
li
nked t
o
t
h
e l
e
v
e
l
of co
nfi
d
en
ce or t
r
ust
.
T
h
e hi
g
h
er t
h
e
l
e
vel
s
of t
r
ust
t
o
an o
b
ject
, t
h
e hi
g
h
e
r
t
h
e l
e
vel
of
pri
v
ac
y
of t
h
e ob
ject
. A hi
g
h
de
gre
e
of p
r
i
v
acy
can b
e
d
e
scri
b
e
d
as the g
u
i
d
a
n
ce
fo
r
an
obj
ect to
cho
o
s
e wh
ich
ob
jects it will co
mmu
n
i
cate with
,
b
y
sup
porting
d
a
ta
in
th
e
fo
rm
o
f
rep
u
t
ation
and
t
r
u
s
t m
e
trics o
f
th
e cand
i
date ob
j
ects.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
239
6
–
24
02
2
398
2.
2.
Trust Calc
ulation
Trust is a con
c
ep
t related
t
o
th
e belief and
ex
p
ecta
tion
s
o
f
reliab
ility, in
teg
r
ity, secu
rity, as well as
th
e ab
ility o
f
an
en
tity [13
]
. Tru
s
t
b
e
co
m
e
s an
im
p
o
r
tan
t
th
i
n
g th
at sho
u
l
d
b
e
im
p
l
e
m
en
te
d
in IOT b
ecause its
usef
ul
ne
ss t
o
secure t
h
e c
o
m
m
uni
cat
i
on bet
w
een
ob
ject
s;
for exam
ple helping the
objec
t
s to choose anothe
r
t
r
ust
a
bl
e
ob
jec
t
duri
ng t
h
e co
m
m
uni
cat
i
on [
14]
. M
ean
w
h
i
l
e
, rep
u
t
a
t
i
on i
s
used t
o
det
e
r
m
i
n
e t
h
e l
e
vel
of t
r
ust
,
and i
t
can be
m
easured
base
d o
n
pri
o
r
kn
o
w
l
e
d
g
e o
f
t
h
e i
n
t
e
ract
i
on
wi
t
h
ot
he
rs o
b
j
ec
t
s
and base
d o
n
t
h
e
expe
ri
ences
o
f
ot
he
r o
b
j
ect
s.
R
e
put
at
i
o
n can
al
so be
used a
s
a param
e
t
e
r for asse
ssi
n
g
t
h
e l
e
vel
of t
r
ust
of a
n
object.
Dynamic
trust m
echanism
is
usef
ul
for the object as a control for
se
lectin
g
th
e ap
p
lication
services in
t
h
e I
O
T.
C
h
a
r
a
c
t
e
ri
st
i
c
of t
r
us
t
as desc
ri
be
d i
n
[8]
,
[1
5]
:
a.
Asymmetric an
d
su
bj
ectiv
e
b.
Depe
n
d
ent
o
r
has a
pa
rt
i
c
ul
ar
co
nt
ext
c.
Dynam
i
c; which m
eans that its val
u
e m
i
ght
increase
,
decrea
se, or rem
a
in
stable
Trust is intransitive, for e
x
a
m
ple if
A trusts B and B trusts C, then
A is not
necessa
rily to trust C
.
B
u
t
B
can
pr
o
v
i
d
e a
rec
o
m
m
e
ndat
i
o
n
o
f
C
rep
u
t
a
t
i
on t
h
at
m
i
ght
be
bene
fi
ci
al
fo
r
A i
n
t
a
ki
n
g
a
deci
si
o
n
wh
et
h
e
r to
tru
s
t C o
r
no
t.
A s
o
ci
ol
ogi
st
Di
eg
o
Gam
b
et
ta de
fi
nes t
r
ust
as f
o
l
l
o
w
s
:
"
.
.. tru
s
t is
a
certa
i
n
d
e
g
r
ee
o
f
p
r
ob
ab
ility th
a
t
is
o
w
n
e
d
by someon
e
o
r
so
met
h
ing
(sub
jectively
)
to
perf
or
m
a
n
a
c
t
i
on,
t
o
m
oni
t
o
r
t
h
e
act
i
ons
(
a
l
t
ho
ug
h
n
o
t
di
r
ect
l
y
)
and c
a
n
i
n
f
l
u
ence t
h
e
a
c
t
i
ons
of
a
pers
on
or
t
h
i
n
g
"
[
1
6
]
.
Th
is
d
e
fin
ition, wh
en
to
b
e
ap
p
lied to
t
h
e co
m
p
u
t
er science, will lead
t
o
th
e t
r
u
s
t m
o
d
e
lin
g, tru
s
t
man
a
g
e
m
e
n
t
, an
d qu
an
titati
v
e
d
ecision
-m
ak
ing
.
Tru
s
t
m
o
d
e
lin
g
relat
e
d
to th
e aspects o
f
co
m
p
u
t
atio
n
a
l
represen
tatio
n
o
f
th
e tru
s
t v
a
l
u
e, wh
ile tru
s
t
man
a
g
e
m
e
nt is
a collection of evide
n
ce and risk assessm
e
n
t for
d
ecision
-m
ak
in
g [15
]
. Th
e tru
s
t
v
a
lu
e is u
s
u
a
lly written
i
n
th
e n
u
m
b
e
r or
lab
e
l,
and
can
b
e
i
n
b
i
n
a
ry, d
i
screte
or co
nt
i
n
uo
us
fo
rm
s. For exa
m
pl
e i
n
t
h
e for
m
of bi
nary
re
prese
n
t
a
t
i
on
of
t
h
e num
ber 1 i
ndi
cat
es "t
rust
abl
e
"
and 0 indicates "un-trusta
ble".
To
calcu
late th
e tru
s
t v
a
lu
e we ap
p
l
y Mo
d
i
fi
ed
An
t Co
lon
y
Alg
o
rith
m
.
Th
e alg
o
r
ith
m
is
su
itab
l
e
for
tru
s
t calcu
latio
n
s
b
ecau
s
e
of
its
m
e
th
o
d
which
invo
lv
ing
p
r
i
o
r kno
wledg
e
, th
e t
r
ail of o
t
h
e
r an
ts left on
a
track
th
at h
a
s
b
een
p
a
ssed
.
Th
is p
r
i
o
r
k
nowled
g
e
invo
lv
es th
e calcu
latio
n
of th
e tru
s
t valu
es th
at h
a
v
e
b
e
en
save
d f
r
o
m
previ
o
us i
n
t
e
ract
i
ons
, ei
t
h
er
di
r
ect
l
y
or i
n
di
rectly. Si
m
ilari
ti
es way
o
f
work
ing
is t
h
e
reaso
n
for
th
e d
e
v
e
lop
m
e
n
t o
f
An
t Co
l
o
n
y
alg
o
rith
m
f
o
r tru
s
t m
e
trics calcu
latio
n
.
So
m
e
m
o
d
i
ficatio
n
s
of An
t Co
lon
y
alg
o
rith
m
are
mad
e
in
o
u
r
t
r
u
s
t m
e
trics c
a
lcu
l
atio
n
.
T
h
i
s
m
odi
fi
cat
i
o
n
bri
e
fl
y
m
a
de for t
h
e p
h
e
r
o
m
one
dep
o
si
t
or t
r
us
t
val
u
e dep
o
si
t
as seen i
n
Eq.
(4)
,
(5
) an
d (
6
)
.
The f
o
l
l
o
w
i
ng su
bsect
i
o
n
s
expl
ai
n t
h
e o
v
eral
l
pr
ocess
o
f
t
h
e
pr
o
pose
d
t
r
ust
assessm
ent
m
e
t
h
o
d
.
2.
2.
1.
Pre-pr
ocessing
Ob
ject
s a
r
e
de
fi
ne
d i
n
t
h
e
f
o
r
m
of m
a
t
r
i
ces: o
n
e c
o
nt
ai
ni
n
g
t
h
e t
r
ust
val
u
e an
d
ot
he
r c
o
nt
ai
ni
n
g
t
h
e
con
n
ect
i
o
n am
on
g o
b
j
ect
s. D
e
faul
t
t
r
ust
val
u
e f
o
r al
l
ob
je
ct
s i
s
0.5 by
consi
d
eri
ng t
h
at
al
l
new ob
ject
s hav
e
th
e sam
e
p
r
ob
ab
ility o
f
tru
s
t
valu
e.
2.
2.
2.
Trust
Va
lue Ca
lculatio
n Process
a.
An
ob
ject
can
m
ovi
ng fr
om
one net
w
o
r
k t
o
ot
he
r net
w
or
ks
and can
be co
nnect
e
d
t
o
a n
u
m
b
er of o
b
j
ec
t
s
.
B
e
fo
re t
h
e co
n
n
ect
i
on t
o
a t
a
rget
o
b
j
ect
, a t
r
ust
assessm
ent
i
s
done. I
f
t
h
e t
a
rget
ob
ject
i
s
l
o
cat
ed i
n
the
sam
e
net
w
or
k
or
i
n
t
r
a
-
net
w
o
r
k, t
h
en
t
h
e t
r
us
t
val
u
e c
a
l
c
ul
at
i
on i
s
pe
rf
orm
e
d as
f
o
l
l
o
w
s
:
∑
,
(1
)
whe
r
e
∀
,
,
∈
1,
b.
If t
h
e target
object is locate
d
and connected
to
d
i
f
f
e
r
e
n
t
n
e
t
w
or
k or
i
n
ter
-
netw
or
k.
Wh
en
o
t
h
e
r
n
e
twork
s
ex
ist
mo
re th
an
on
e ob
j
ect th
at
eve
r
comm
unicate
with the
target object
or
objec
t
j
,
th
e m
o
st tru
s
tfu
l
obj
ect will b
e
ch
o
s
en
u
s
ing
m
a
x
i
m
a
l v
a
l
u
e of
.
Tru
s
t v
a
lu
e calcu
latio
n in
d
i
fferent
n
e
two
r
k
is
written
as
fo
llows:
∑
,
(2)
whe
r
e
∀
,
,
∈
1,
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Trust
B
a
se
d
Pr
i
v
acy f
o
r
I
n
t
e
r
n
et
of
Thi
n
g
s
(
V
era
S
u
ry
ani
)
2
399
c.
Th
e equ
a
tio
n fo
r all In
t
r
a an
d in
tern
etwo
rk
s:
∑
(3
)
whe
r
e
∀
,
,
,
,
,
,
∈
1,
d.
Ph
ero
m
o
n
e
d
e
p
o
s
it
o
r
tru
s
t valu
e d
e
po
sit:
1
Δ
(4)
∑
(5
)
e.
Phe
r
om
one
e
v
apo
r
at
i
o
n or
t
r
ust
val
u
e red
u
c
t
i
on:
1
1
(6
)
3.
SIMULATION AND RESULTS
To dem
onst
r
at
e t
h
e i
d
ea we
have i
m
pl
em
ent
e
d t
h
e
pr
o
p
o
s
ed m
odi
fi
ed
al
go
ri
t
h
m
on M
A
TLA
B
versi
on
7.
13 a
nd
usi
n
g a si
m
p
l
e
t
opol
o
g
y
as show
n i
n
Fi
gu
re 1. T
h
e
t
opol
ogy
use
d
i
n
t
h
e sim
u
l
a
t
i
on
consists of thre
e networks
where eac
h
n
e
twor
k h
a
s thr
ee m
e
m
b
er
s o
f
obj
ects.
Fi
gu
re
1.
Net
w
or
k T
o
pol
ogy
B
a
sed o
n
t
o
pol
ogy
s
h
o
w
e
d
i
n
Fi
gu
re 1
,
a si
m
u
l
a
t
i
on was
per
f
o
r
m
e
d usi
n
g a scena
r
i
o
as
fol
l
o
ws.
A
n
object
A1 wa
nt to connect t
o
obj
ect B
1
. So
A1
calculates the trust va
l
u
e t
o
wa
rds object
B1.
T
h
e trust
value
calculations underta
ken
from
object
A1 are a
s
follow:
a.
Pre-proces
sing
Th
e co
nn
ection
m
a
trix
an
d in
itial tru
s
t v
a
l
u
e m
a
trix
w
e
re in
itialized
in
th
is step. Connectio
n
m
a
trix
was filled
up
with
a
v
a
lu
e
of ‘1’
if t
h
e object was co
nn
ected
,
an
d
v
a
lue o
f
‘0’
fo
r
v
i
ce v
e
rsa. R
o
ws o
f
th
e
matrix desc
ribe the a
v
ailable
net
w
orks
, a
nd col
u
m
n
s of
the m
a
trix show objects th
at a
r
e connected t
o
each
n
e
two
r
k
.
Thu
s
th
e conn
ection
matrix
fro
m
A1
p
o
i
n
t
of
v
i
ew will b
e
:
1
2
3
1
2
3
1
2
3
11
1
00
0
10
0
Defau
lt tru
s
t
valu
e
for all
obj
ects i
n
t
h
e
b
e
g
i
nn
ing
is
0
.
5
.
So
t
h
e con
t
ent of tru
s
t m
a
trix
fro
m
A1
p
o
i
n
t
of v
i
ew will
b
e
:
1
2
3
1
2
3
1
2
3
0.5
0
.5
0
.
5
0.5
0
.5
0
.
5
0.5
0
.5
0
.
5
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
239
6
–
24
02
2
400
b.
T
r
ust
V
a
l
u
e C
a
l
c
ul
at
i
on P
r
oce
ss
B
a
sed
on
eq
ua
t
i
on
(1
),
o
b
ject
A
1
t
r
i
e
d t
o
fi
n
d
out
w
h
et
he
r
ob
ject
B
4
i
s
l
o
cat
ed o
n
t
h
e sa
m
e
net
w
o
r
k
w
ith
obj
ect
A1
or
n
o
t
. I
f
ob
j
ect
B
1
is situ
ated
in
t
h
e sa
m
e
n
e
twork
with
obj
ect
A1
, th
en
o
b
j
ect
A1
will
calcu
late th
e tru
s
t v
a
l
u
e of
o
b
j
ect B1
u
s
ing
eq
u
a
tion
(1).
Ho
wev
e
r
,
if it turn
s t
h
at obj
ect B1
is no
t lo
cat
ed
in
th
e sam
e
n
e
two
r
k
with
th
e obj
ect
A1, an
d
t
h
en
o
b
j
ect
A1
will ask
o
t
h
e
rs
ob
j
ects in
t
h
e
n
e
twork
t
h
at co
nn
ected
with
o
b
j
ect B1. Fro
m
th
e to
po
log
y
in
Fi
g
u
re1
,
ob
j
ect
A3 is connecte
d
to object
B
1
. Furtherm
ore,
object
A1
will g
i
v
e
weigh
t
to
ob
j
ect
A3 b
a
sed
o
n
prev
i
o
u
s
h
i
st
o
r
ical
tru
s
ts
v
a
lu
e of
o
b
j
ect
A3
.
If
ob
j
ect
A3
is a t
r
u
s
tab
l
e
o
b
j
ect, th
en
o
b
j
ect
A1
will g
i
v
e
a wei
g
h
t
o
f
0
.
8
,
o
r
g
i
v
e
a
weigh
t
of 0.3
o
t
h
e
rwise.
Ob
ject
A3
i
n
v
e
stigates all
con
n
ect
ed
ob
je
ct
s t
h
at
ha
ve a
hi
st
ory
o
f
t
r
us
t
t
o
wa
r
d
s
B
1
.
T
r
ust
val
u
e
of
objects
which
located i
n
t
h
e
sam
e
n
e
two
r
k
with
ob
j
ect B
1
will be calcu
lated
usin
g
eq
u
a
ti
o
n
(2), an
d
add
e
d
al
to
g
e
th
er
with
oth
e
r
o
b
j
ects located
i
n
di
f
f
e
r
ent
net
w
o
r
k
s
usi
n
g e
q
uat
i
o
n
(
3
)
.
c.
T
r
ust
V
a
l
u
e
Sc
ori
n
g
Pr
ocess
Th
e scoring pro
cess of tru
s
t
v
a
lu
e will lead to
2
resu
lts;
wh
et
h
e
r an
o
t
her
ob
j
ect
is t
r
ustab
l
e en
oug
h
o
r
no
t. Repu
tatio
n
is ano
t
h
e
r p
a
ram
e
ter th
at can b
e
u
s
ed to d
e
term
in
e
a trustab
l
e
ob
ject.
W
e
igh
ting th
e
rep
u
t
a
t
i
on ca
n
be d
o
n
e usi
ng
a ran
g
e o
f
val
u
e 0 t
o
1
,
an
d s
e
t
a t
h
resh
ol
d
val
u
e f
o
r det
e
r
m
i
n
i
ng t
h
e l
i
m
i
t
of a
t
r
ust
a
bl
e
o
b
j
ec
t
.
A
pa
rt
i
c
ul
ar
con
s
t
a
nt
val
u
e
was
use
d
i
n
t
h
i
s
si
m
u
l
a
t
i
on;
a t
r
ust
a
bl
e
o
b
je
ct
was
gi
ve
n a
val
u
e
of
0.
8,
a
n
d
a
u
n
-t
r
u
st
a
b
l
e
ob
j
ect
was
gi
ve
n
a
val
u
e
o
f
0
.
3
.
The
p
r
ocess
o
f
t
r
ust
val
u
e
sc
o
r
i
n
g i
n
t
h
i
s
si
m
u
l
a
t
i
on
was per
f
o
rm
ed usi
n
g p
h
er
o
m
one
dep
o
si
t
eq
uat
i
o
n
as
descri
bed
i
n
e
quat
i
o
n (
4
).
M
o
re
ove
r
,
gra
n
t
i
n
g
a
r
e
pu
tatio
n
w
a
s d
o
n
e
u
s
i
n
g
equatio
n
(
5
)
.
The t
r
ust
val
u
e
can
be
red
u
ce
d
or i
n
crease
d
ove
r t
i
m
e, dep
e
ndi
ng
o
n
t
h
e r
e
put
at
i
o
n
of t
h
e o
b
ject
.
Th
e
pr
ocess
o
f
t
r
ust
val
u
e
re
d
u
ct
i
o
n
was
do
ne
usi
n
g
eq
uat
i
o
n
(6
)
.
Th
e sim
u
latio
n was aim
e
d
fo
r calcu
latin
g
t
h
e tru
s
t
v
a
lu
e, an
d it was
do
n
e
serially an
d ran
d
o
m
ly in
100 tim
es. Initial trust val
u
e
for e
v
ery
obj
e
ct was set
of
0.5, and t
h
is value ca
n
be c
h
anged according
to t
h
e
num
ber
o
f
o
b
j
ect
s t
h
at
p
r
o
v
i
d
e a
n
assessm
ent
of
t
h
e
am
ount
o
f
t
r
ust
t
o
war
d
s
o
b
j
ect
B
1
.
Fi
g
u
r
e
2
s
h
ows
t
h
e
resu
lts
o
f
th
e t
r
u
s
t
v
a
lu
e
o
f
obj
ect B1.
Fi
gu
re
2.
Tr
ust
Val
u
e
o
f
O
b
je
ct
B
1
T
r
end ch
art
from
Fig
u
r
e 3
d
i
sp
lays th
e
redu
cin
g
tr
en
d
o
f
p
h
ero
m
o
n
e
v
a
l
u
e, in
wh
ich so
me po
i
n
t
will
st
abl
e
i
n
a ce
rt
ai
n
val
u
e
.
T
h
i
s
red
u
ci
n
g
t
r
ust
val
u
e i
s
cause
d
by
e
v
a
p
o
r
at
i
o
n
pr
oce
ss w
h
e
n
t
h
ere
i
s
n
o
con
n
ect
i
o
ns
ha
ppe
ne
d
d
u
ri
ng
si
m
u
l
a
t
i
on. R
e
put
at
i
o
n
val
u
e al
so
c
ont
ri
b
u
t
i
n
g
t
o
t
h
e
d
ecreasi
n
g
o
f
t
h
e t
r
us
t
value,
where t
h
e val
u
es
of
reputation wa
ne whe
n
th
e
r
e
are no trust a
ssessm
ents done for
som
e
objects
.
W
e
igh
ting
process in th
e sco
r
i
n
g repu
tatio
n
g
i
v
e
s a
sig
n
i
fican
t i
n
fl
uen
ce i
n
d
e
termin
in
g
of t
h
is v
a
l
u
e.
Sel
ect
i
ng t
h
e a
p
p
r
op
ri
at
e m
e
tho
d
f
o
r s
c
o
r
i
n
g
re
put
at
i
o
n ca
n
m
a
ke t
h
e i
m
provem
e
nt
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Trust
B
a
se
d
Pr
i
v
acy f
o
r
I
n
t
e
r
n
et
of
Thi
n
g
s
(
V
era
S
u
ry
ani
)
2
401
Fi
gu
re
3.
Tre
n
d C
h
art
of
P
h
er
om
one Val
u
e
Ano
t
h
e
r tru
s
t
assessm
en
t research, wh
ich
was in
sp
ir
ed
by ant colony,
was al
so
don
e b
y
[
8
]. U
s
i
n
g
di
ffe
re
nt
rep
u
t
a
t
i
on an
d t
r
ust
fo
rm
ul
a as wel
l
as sel
ect
ed perform
a
nce param
e
ters,
research in t
h
is area are
very
p
o
t
e
nt
i
a
l
t
o
be devel
ope
d
.
4.
CO
NCL
USI
O
N
Secu
ri
t
y
aspec
t
s pl
ay
an i
m
port
a
nt
r
o
l
e
i
n
Int
e
r
n
et
of T
h
i
ngs
. S
o
m
e
appl
i
cat
i
ons s
u
c
h
as IT
S, e
-
heal
t
h
,
fl
eet
m
a
nagem
e
nt
, sm
art
m
e
t
e
ri
ng,
h
o
m
e
aut
o
m
a
ti
on m
i
ght
re
qui
r
e
secu
ri
t
y
im
pl
em
ent
a
t
i
on t
o
secure
t
h
e dat
a
t
r
a
n
s
f
erre
d a
nd t
o
i
m
prove
pri
v
ac
y
.
In
t
h
i
s
pap
e
r, a m
odi
fi
ca
t
i
on
of
A
n
t
C
o
l
o
ny
al
g
o
ri
t
h
m
was
pr
o
pose
d
fo
r
scori
ng t
r
ust
val
u
e
of
o
b
je
ct
s i
n
I
n
t
e
rnet
of
Thi
ngs
. T
h
e si
m
u
l
a
t
i
on sh
ows t
h
at
a
ddi
ng
a
p
a
ram
e
ter n
a
mely rep
u
t
ation
n
eeds an
app
r
o
p
riate weigh
t
i
n
g
m
e
th
od
so
th
at th
e tru
s
t valu
es will b
e
sco
r
ed
fairly for all ob
j
ects, reg
a
rd
i
n
g th
eir
p
r
ev
i
o
u
s
activ
ities during
th
e co
mmu
n
i
cation
p
r
o
c
ess. Im
p
r
o
v
e
m
e
n
t
s on
secu
rity m
o
d
e
l fo
r t
r
u
s
t
v
a
l
u
e scoring
will be a goo
d alte
rnativ
e so
l
u
tio
n in
fu
t
u
re
wo
rk
s, esp
ecially th
e on
e,
whic
h
has
better re
sistance a
g
ainst attacks.
AC
KN
OWLE
DG
MENT
Th
is
r
e
sear
ch
w
a
s
p
a
r
tially su
ppo
r
t
ed
b
y
M
i
n
i
str
y
of
Resear
ch, Techn
o
l
og
y, an
d
H
i
gh
er Edu
cation
,
In
d
onesi
a
u
nde
r t
h
e
P
U
PT
g
r
a
n
t
(
7
51/
UN
1-
P
.
II
I/
LT/
D
I
T-
LI
T/
20
1
6
).
REFERE
NC
ES
[1]
R. H. Web
e
r, “I
nternet of
Thing
s
–
New s
ecuri
t
y
and priv
ac
y
c
h
all
e
nges
,
”
Computer Law and
Security Review
,
vol/issue: 26(1), pp.
23–30
,
2010
.
[2]
H.
S.
Kim a
nd J.
S.
Se
o,
“A Da
ily
Activity
Monitoring S
y
s
t
em for Internet
o
f
Things-Assisted Living in Ho
me
Area Networks,”
International Jo
urnal
of Electr
ical and Computer Engin
eering
, vo
l/issue: 6
(
1), pp.399–405, 2016
.
[3]
S. Tozlu,
et al.
, “
W
i
-Fi enable
d sensors for i
n
terne
t
of thing
s
: A practica
l
a
pproach,
”
IEEE Communication
Magazine
, vo
l/is
sue: 50(6), pp
. 1
34–143, 2012
.
[4]
J. Pescator
e, “Securing
the ‘ In
ter
n
et of
Th
ings ’ S
u
rvey
”, Repor
t,
SANS
Institute, 2014.
[5]
J. M. S. Koppula, “Secure
Digital Signature Sch
e
me Ba
sed on Elliptic Curv
es for
Internet of Thin
gs,”
Internat
ion
a
l
Journal of Electrical and
Computer
Eng
i
neer
ing
,
vol/issue: 6(3),
2
016.
[6]
Y. Cheng,
et
al
.
,
“
P
rivac
y
in
m
achine-to-m
a
c
h
ine
com
m
unications
A s
t
ate-o
f-the-ar
t
s
u
rve
y
,
”
in
2012 IEEE
International Co
nference on
Co
mmunication S
y
stems, ICCS
2012
, pp
. 75–79
, 201
2.
[7]
D.
Chen,
et al.
,
“TRM-IoT: A tr
ust management
model based on
fuzzy
r
e
putatio
n
for internet of th
ings,”
Computer
Scien
c
e and Info
rmation System
,
vol/issue:
8(4), p
p
. 1207–1228
, 2
011.
[8]
P. Bedi and R
.
Sharma, “Trust
b
a
sed recommender s
y
stem using
ant
co
lon
y
for tr
ust com
putation
,
”
Ex
pe
rt Sy
ste
m
with App
l
i
c
ation
, vol/issue: 39(1)
, pp
. 1183–1190
, 2012.
[9]
S. Zafar
and M.
K. Soni, “T
rust based QOS protocol(TBQP)
using meta-heuristic gene
tic
algor
it
hm
for optim
izi
ng
and secur
i
ng MANET,”
ICR
O
IT 2014 - P
r
oc
.
2014 Int
e
rnat
i
o
nal
Conf
e
r
e
n
c
e
on
R
e
li
abi
li
t
y
,
Opt
i
m
i
z
at
i
on,
and
Information Technology
, pp. 173
–177, 2014
.
[10]
M. Nitti,
et al.
,
“
A
subjective
m
odel for trustworthiness eval
uation in the social Internet of Things,”
IEEE
International Symposium on Per
s
onal, Indoor,
a
nd Mobile Radio Communicatio
n PIMRC
, pp
. 18
–23, 2012
.
[11]
Z. Chen
,
et al.
,
“Fuz
zy
T
h
e
o
ry
for the
P2P S
ubject Trust Ev
aluatio
n Model,”
In
tern
ational Journal
of Advan
cemen
t
in Computing
Technology.
, vol/issue: 4(8), pp
. 67
–74, 2012
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 5
,
O
c
tob
e
r
20
16
:
239
6
–
24
02
2
402
[12]
RFC 2828, “Internet Security
Glosar
y
.
”
[13]
Z. Yan
and
P. Zhang, “A Survey on Trust Ma
nag
e
ment for
Intern
et of
Th
ings,” 20
14.
[14]
R. Roman,
et al.
, “Securing
the In
ternet of
things,”
IEEE Computer Society.
, vol/issue: 44(9)
, pp
. 51
–58, 2011
.
[15]
S.
Rie
s
,
et
al
.
,
“A Cla
ssifica
tion
of Trust S
y
ste
m
s,
”
Mo
ve to
Mean
ingful
Internet S
y
stem 2006
, pp
.
894–903, 2006
.
[16]
D. Gambetta, “Trust: Making
an
d Break
ing Coo
p
erative R
e
latio
ns,” 2000
.
BIOGRAP
HI
ES OF
AUTH
ORS
Vera S
u
r
y
ani r
ece
ived M
a
s
t
e
r
degree
ini In
f
o
rmation Techo
l
og
y
from Institut Tekno
logi
Bandung, Indon
esia in 2009.
She joined as
a
Lecturer in
the School of Computing and
Informatics, Telkom University
, in 2003. She is currently
member of Sistem Elektron
is
laborator
y
at Un
iversitas Gadjah
Ma
da
.
He
r re
sea
r
c
h
inte
re
sts in
clude wireless sensor network,
distributed s
y
stem, and Intern
et of Things. Cu
rrently
she
is working for Ph.D program at
Departem
ent of
Ele
c
tri
cal Eng
i
n
eering & Infor
m
ation Techno
l
o
g
y
, Univers
itas
Gadjah M
a
da,
Indonesia.
Selo Sulisty
o is
an associate pr
ofessor in
Infor
m
ation and Co
mmunication Technolog
y
at th
e
Department
of Electrical Engin
eering
and Infor
m
ation Technolog
y
.
He is also
Head of Sistem
Electronis labor
ator
y
at Univer
s
ita
s Ga
dja
h
Ma
da
. His re
se
a
r
ch
inter
e
sts inc
l
u
d
ing Software
Modeling, Mob
ile app
lication
development
an
d
S
ecurit
y
for
the Intern
et o
f
Things
and
connected
objects.
Wid
y
awan is an assistant prof
essor in In
formation and Com
m
unication Technolog
y
at th
e
Department of Electrical Eng
i
neering and Info
rmation Technolog
y Univer
sitas Gad
j
ah Mada. He
is also Director
of Center of S
y
s
t
em and Info
rmation Resource at Universitas Gadjah Mada. His
res
earch
int
e
res
t
s
including pe
rv
as
ive com
puting
,
computer secur
i
ty
, ub
iquitous computing, and
w
i
r
e
l
e
s
s
sy
st
e
m
.
Evaluation Warning : The document was created with Spire.PDF for Python.