Co
m
pu
ter Sci
ence a
nd Inf
or
mat
i
on
Tec
h
no
lo
gies
Vo
l.
2
, No
.
1
,
Ma
rch
2021
,
pp.
33
~
42
IS
S
N:
27
22
-
3221
,
DOI: 10
.11
591
/
csi
t.v
2i1
.p
33
-
42
33
Journ
al h
om
e
page
: htt
p:
//
ia
esprime
.com/i
ndex.
php/csit
Su
rv
ey o
f
wormh
ole atta
ck
in wire
less sens
or netwo
rks
Umash
an
k
ar Ghu
ga
r
, Jay
ar
am
Pr
adha
n
Depa
rtment
o
f
C
om
pute
r
Scie
n
ce,
Berh
ampur Uni
ver
sit
y
,
Brahma
pur,
Odisha
760
007,
Indi
a
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Ma
y
12, 20
2
0
Re
vised
Ju
l
2
,
20
2
0
Accepte
d
J
ul
2
9
, 2
0
2
0
From
the
la
st
de
ca
de
,
a
wir
el
ess
sensor
net
work
(
W
SN
)
has
a
ver
y
importan
t
role
ov
er
the
n
et
works
.
The
p
rimar
y
fe
at
ure
s
of
W
SN
inclu
de
sat
el
l
ite
comm
unic
at
ion,
broa
dca
st
ch
ann
el
,
hosti
le
envi
ro
nm
ent
,
m
edica
l
s
y
stem
an
d
dat
a
g
at
her
ing.
T
her
e
ar
e
a
lo
t
of
a
tt
a
cks
ava
i
la
bl
e
i
n
W
SN
.
In
worm
hole
atta
ck
sce
nar
io
is
brutal
from
othe
r
a
tta
cks,
which
is
smoothl
y
resolv
ed
i
n
net
works
but
tough
to
observe
.
Th
is
surve
y
pape
r
is
an
ex
per
iment
to
obse
rving
thr
eats
and
a
lso f
ocuse
s
on
som
e
diffe
r
en
t
m
et
hod
to ide
nt
if
y
th
e
worm
hol
e
a
tt
a
cks.
Ke
yw
or
d
s
:
IDS
MANET
Sensor
no
de
Worm
ho
le
att
acks
WSN
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
B
Y
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
Um
ashan
ka
r G
huga
r
Dep
a
rtm
ent o
f C
om
pu
te
r
Scie
nce
Be
rh
am
pu
r
Un
iversity
Brahm
apu
r
, O
dish
a
76
0007, In
dia
Em
a
il
:
ug
huga
r@gm
ail.co
m
1.
INTROD
U
CTION
W
i
reless
se
nso
r
net
work
(
WSN)
buil
t
a
netw
ork,
w
hich
is
a
sp
rea
d,
a
ut
om
a
ti
c
governin
g
ne
twork
a
nd
it
correspo
nd
i
ng
with
se
veral
sens
or
node
s
in
sp
eci
fic
env
i
ronm
ent.
Nodes
are
obs
erv
e
d
by
t
he
natu
ral
conditi
ons,
su
c
h
as
hu
m
idit
y,
com
pr
essio
n,
he
at
,
wa
ve
an
d
direct
io
n
at
dif
fer
e
nt
ar
eas
[1]
.
It
is
a
ti
ny
de
vic
e
wh
ic
h
has
a
li
m
it
ed
m
easur
em
ent
resou
rce.
They
a
re g
ra
dual
ly
arr
a
ng
e
d
in
a w
irel
ess
se
ns
or
e
nviro
nm
ent
[
2].
WSN
a
re
broa
dly
util
iz
ed
on
dif
fere
nt
a
ppli
cat
ion
s
su
c
h
as
,
area
obser
ving,
de
fense
s
urv
ei
ll
ance,
he
al
th
care
syst
e
m
,
ho
m
e
aff
irm
at
ion
a
nd
sat
el
li
te
com
m
un
ic
at
ion
.
WSN
s
uffer
s
from
var
ious
secur
it
y
issue
because
us
ua
ll
y
it
is
de
plo
ye
d
i
n
haza
r
dous
e
nv
ir
onm
ent.
Se
ns
or
no
de
has
s
om
e
lim
it
a
ti
on
s
uc
h
a
s
li
m
it
ed
li
feti
m
e,
le
ss
com
pu
ti
ng
ca
pa
bili
ty
and
lo
w
m
e
m
o
ry
sp
ace
[3
,
4].
Ba
se
d
on
t
hese
li
m
it
at
ion
s,
they
a
re
ar
range
d
in
no
isy
env
i
ronm
ent,
it
is
hig
hly
affe
ct
ed
an
d
sen
s
it
ive
to
sever
al
ty
pes
of
at
ta
cks
[
5].
Ba
sic
al
ly
,
sensor
no
de
s
are
cat
egory
by
f
our
s
ub
-
syst
em
s
[
6
-
13,
14]
.
P
r
oc
esso
r
a
nd
m
e
m
or
y,
transcei
ver
,
se
nsor
an
d
batte
ry.
He
re
we
ha
ve
discusse
d t
he
s
ever
al
ty
pes of
at
ta
cks.
Ma
inly
att
acks a
re cla
ssifie
d by t
wo
par
ts.
Fir
st pa
rt
is the
att
ack
a
gainst
secur
it
y
m
echan
ism
and
a
no
t
her
is
r
outi
ng
m
echan
ism
.
Nu
m
ber
s
of
at
ta
cks
a
re
li
ste
d
as
bel
ow
but
we
a
re
fo
c
us
es
on w
orm
ho
le
att
ack.
Worm
ho
le
A
tt
ack
Sybil
A
tt
ack
Bl
ackhole Att
a
ck
Hell
o flo
od A
tt
ack
Sinkh
ole A
tt
ac
k
Den
ia
l
of Servi
ce
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol. 2, N
o.
1, M
arch 2
021
:
33
–
42
34
Th
us
,
t
hese
s
urvey
papers
ba
sic
al
ly
fo
cus
on
va
rio
us
a
ppr
oac
hes
to
de
te
ct
worm
ho
le
at
ta
cks.
I
n
Sect
ion
2
disc
us
se
d
t
he
i
ntr
usi
on
detect
ion
syst
e
m
in
WSN;
In
Sect
io
n
3
disc
us
se
d
t
he
w
orm
ho
le
at
ta
ck
in
wireless
se
nso
r
net
wor
ks
;
I
n
Sect
ion
4
dis
cusse
d
va
rio
us
detect
ion
a
pp
ro
ac
hes
of
t
he
worm
ho
le
at
ta
ck
in
wireless
sens
or
net
works
with
su
m
m
ary.
Finall
y
in
Sect
io
n
5,
w
e
hav
e
discu
ssed
t
h
e
f
utu
re
researc
h
c
halle
ng
e
s
and co
nclu
sio
n.
2.
INTRUSI
ON
DETE
CTIO
N
SYSTE
MS
An
i
ntrusio
n
detect
ion
syst
e
m
(I
DS)
is
use
d
f
or
obser
vi
ng
t
he
net
wor
k
dia
gnos
is
a
gainst
na
sty
m
ov
e
m
ents
and
inf
orm
ed
to
the
bas
e
sta
ti
ons.
Ma
inly
it
is
div
ided
by
tw
o
ty
pes:
m
isuse
I
DS
a
nd
a
nom
aly
ID
S.
Mi
su
se
IDS
In
this sy
ste
m
, th
e ab
norm
al
p
at
te
rn
is
calc
ulat
ed
a
nd c
ontrast
w
it
h t
he
pre
vio
us
data
[5
]
.
Si
gn
al
’s
energy
is
us
e
d
to
detect
t
he
m
al
ic
iou
s
node,
wh
e
re
if
the
en
erg
y
is
c
olli
sio
n
with
th
e
act
ua
l
posit
ion
s
the
n
th
e
m
essage
tr
a
nsm
issi
on
is
c
ons
idere
d
as
do
ub
t
fu
l
[
6].
Anom
a
ly
ID
S
-
It
is
det
ect
ed
by
pr
oto
c
ol,
w
her
e
prev
e
ntion
m
et
ho
d i
s
us
ed
befor
e
t
he det
ect
ion
sta
ge
. H
ere
prot
oco
ls
a
re
act
ivate
d
on
the
data
wit
h r
espect
to
the
ne
tw
or
k
perform
ance.
Wh
e
n
t
he
data
is
sat
isfyi
ng
th
e
ru
le
th
en
it
is
cal
le
d
as
nor
m
al
no
de
el
se
m
al
ic
iou
s
node
.
Whe
n
the
int
rude
r
is
detect
ed,
the
n
inf
or
m
s
to
t
he
syst
e
m
[7,
8].
In
r
outi
ng
,
va
r
iou
s
m
ulti
path
routin
g
te
c
hn
i
qu
e
is
us
e
d
f
or
best
r
edun
dan
cy
pat
h
wit
h
hi
gh
en
erg
y
e
ff
ic
ie
ncy
eff
ic
ie
ncy
[9
]
.
W
at
ch
dog
te
c
hn
i
qu
e
is
a
det
ect
ion
te
chn
iq
ue, w
he
re eac
h node c
an obse
r
ver
s
th
ei
r
nei
ghbori
ng nod
e
s w
it
hin
t
he radi
o
ra
nge
[10].
3.
WORMH
OL
E ATTA
CK
The
worm
ho
le
at
ta
cks
are
m
os
t
brutal
in
na
ture.
Ge
ner
al
ly
,
m
or
e
than
tw
o
m
alici
ou
s
no
des
create
a
secrete
r
out
e
is
cal
le
d
tu
nn
e
l.
Her
e
the
at
t
ackers
are
buil
t
a
connecti
on
to
e
ach
node
,
so
that
the
y
ca
n
com
m
un
ic
at
e
a
t
a
high
s
peed
ov
e
r
the
net
work
s
with
oth
e
r
nodes
.
A
w
or
m
ho
le
at
ta
cks
ca
n
be
f
reely
carr
ie
d
out
acro
s
s rou
ti
ng
in the s
e
ns
or
ne
tworks
. Routi
ng
prot
oco
ls
ha
s no
m
echan
ism
to
pr
e
ven
t a
gainst it
[11
]
. I
n
ot
her
words,
w
hen
t
he
worm
ho
le
at
ta
cks
occ
urs,
it
droppin
g
a
ll
the
pac
kets
an
d
ca
us
e
ne
twork
i
nter
rup
ti
on
.
Worm
ho
le
at
ta
ck
is
al
so
us
e
d
in
the
f
or
m
of
m
erg
ing
of
se
le
ct
ive
f
orward
an
d
Sybil
at
ta
ck
[
12
]
.
In
Fig
ur
e
1,
the d
at
a
pac
ket
accepte
d
N
od
e D
from
N
ode
A
a
nd
vice v
e
r
sa.
Fig
ure
1.
Wo
r
m
ho
le
at
ta
ck
in
WSN
3.1.
Types
of wor
mho
le
att
ack
Her
e
,
we
ha
ve
cat
eg
or
ie
s
the
w
orm
ho
le
at
ta
ck
est
ablishe
d
on
the
se
veral
te
ch
niques.
Nu
m
ber
s
of
nodes
are pa
rtic
ipati
ng
f
or
est
ablishin
g
t
he m
et
ho
d f
or
w
orm
ho
le
into f
ollow
in
g
ty
pe
s [15
]
.
Using
pac
ket
encapsulat
io
n:
T
he
num
ber
of
data
packet
an
d
node
are
encapsulat
ed
be
tween
tw
o
na
sty
nodes
.
Using
out
-
of
-
ba
nd
c
ha
nn
el
:
On
ly
sin
gle
na
sty
node
is
oc
currin
g
with
t
he
hi
gh
s
pee
d
of
c
omm
un
ic
ation
sco
pe.
Using
pac
ket
r
el
ay
:
The
nasty
no
de
gi
ves
r
e
pl
ay
s
to
al
l
data
pack
et
s
betwee
n
t
he
tw
o
c
omm
un
ic
at
ed
no
de
s.
Finall
y,
the
du
plica
te
nod
e
is
create
d by
nast
y node.
Using
prot
oc
ol
dist
or
ti
on:
Sin
gle
nasty
node
is
trie
s
f
or
cra
c
king
t
he
at
ta
ck
,
wh
ic
h
is
at
ta
c
k
by
the
r
outi
ng
protoc
ol.
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Su
r
vey
of wor
mhole
attack i
n
wi
rel
ess se
nsor
networks
…
(
Um
ash
ankar
Ghu
ga
r
)
35
3.2.
Routin
g
pr
oto
cols for w
orm
ho
le
atta
c
k
Most
of
the
r
outi
ng
prot
oc
ols
are
us
ed
in
W
SN
.T
he
r
ou
ti
ng
protoc
ols
are
cat
egories
int
o:
Proact
iv
e
and
Re
act
ive
[
16
]
.
A
ODV,
S
ecur
e
-
A
O
DV
a
nd
D
SR
are
pr
oacti
ve
routin
g
prot
oco
ls
w
he
re
as
DSD
V,
OLS
R,
OS
P
F ar
e
Reac
ti
ve
r
ou
ti
ng
pr
o
toc
ols.
Fig
ure
2. Ro
uting Pr
ot
oco
ls
3.2.1
Aod
v
r
outi
ng
p
ro
t
oco
l
The
A
O
DV
st
and
for
ad
-
ho
c
on
-
dem
and
distance
vect
or
.
Its
is
a
m
os
tl
y
us
e
d
prot
oco
l
and
cal
le
d
a
s
dynam
ic
reactiv
e
r
outi
ng
protoco
l
[
17
-
18]
,
t
hat
create
a
sel
f
r
oute
on
cal
l
su
pp
or
t.
Wh
e
n
a
sen
der
no
de
s
ends
a
data
pac
ket
to
destinat
io
n
no
de,
it
m
us
t
ta
ke
the
hel
p
of
r
ou
ti
ng
ta
ble.
I
f
the
no
de
gai
n
recent
paths
t
he
n
the
data
pac
ket
a
re
forw
a
r
ded
to
de
sti
nation
el
se
i
t
us
es
t
he
r
ou
te
disco
v
ery
proc
ess.
I
n
A
ODV
protoc
ol,
tw
o
c
on
t
ro
l
m
essage
is
us
e
d
by
r
oute
disc
ov
e
ry
proce
ss
i.e
route
re
ques
t
(RREQ)
a
nd
Rou
te
Re
ply
(
RR
EP).
T
o
obt
ai
n
the
recent
pat
h,
R
REQ
an
d
RR
E
P
c
on
t
ro
l
m
essages
are
us
e
d.
Wh
e
n
the
rout
e
disco
ver
y
pr
ocess
is
over
,
t
han
t
he
da
ta
p
ac
ket
of
so
urce a
nd
des
ti
nation node
c
an be c
onnecte
d.
3.2.2
Saodv r
outi
ng
protoc
ol
AOD
V
protoc
ol
on
e
xtensi
on
le
ads
to
S
AODV
protoc
ol
[19].
T
his
has
a
great
er
util
it
y
in
the
sec
ur
it
y
to
protect
t
he
r
ou
te
disc
ov
e
ry
m
echan
is
m
.
Fr
om
desira
ble
asym
m
e
tric
cryp
tosyst
em
,
ea
ch
node
has
a
coupl
e
of
sig
natu
re
ke
y
and
it
is
abili
ty
to
ver
ify
t
he
assum
ption
be
tween
giv
e
n
a
ddress
a
nd
publ
ic
key
of
the
s
a
m
e
node. S
o
S
A
O
DV h
as
the
task
of k
ey
m
anag
em
ent sch
em
e [
20
]
.
3.2.3
Dynam
ic
so
ur
c
e r
ou
ti
ng
proto
col (DSR
)
DS
R
prot
oco
l
i
s
us
e
t
o
up
date
the
cac
he
m
em
or
y
of
r
oute
by
route
disc
overy
proce
ss.
I
t
updates
the
inf
or
m
at
ion
ab
ou
t
al
l
li
nk
s
between
t
he
sen
de
r
an
d
recei
ver
node.
In
orde
r
to
tran
sm
i
t
data,
a
well
def
ine
d
route
is
ta
ken
i
nto
a
ccount
by
the
route
disc
over
y
process
for
n
ode.
This
m
ot
ive
is
achieve
d
by
r
ou
te
dis
cov
e
ry
process
and
route m
ai
ntenan
c
e proces
s.
Rou
te
D
isc
ov
e
ry
process:
Wh
en
a
se
nder
no
de f
orward
a
da
ta
to a
nothe
r node
ove
r net
work, it
h
as
t
o
go
th
r
ough
it
s
route
cac
he.
I
n
case
of
unava
il
abili
ty
of
r
ou
te
s
bet
we
en
th
e
recei
ver
an
d
sen
der
tha
n
r
oute
is
discar
ded
a
nd
i
t
broa
dcast
RR
EP
(Route
Re
pl
y).
Wh
en
the
r
ecei
ver
no
de
or
any
i
nterm
ediat
e
node
has
re
cei
ved
the fres
h path
f
ro
m
the sende
r
nod
e
, th
e
n
RR
EP (
R
oute
Rep
ly
)
is ge
ner
at
e
d[2
1].
Rou
te
Ma
inten
ance
P
ro
ces
s:
W
it
h
the
i
niti
ation
of
data
tra
ns
m
issi
on
proc
ess,
it
is
the
ta
sk
of
se
nde
r
node
to
co
nf
i
r
m
that
ver
y
ne
xt
hop
receive
d
both
t
he
data
an
d
t
ran
sm
it
t
he
route
t
o
rec
ei
ver
.
I
n
case
sen
der
didn’t
get
a
c
onfirm
ation
m
essage
tha
n
it
ge
n
erates
r
ou
te
e
rror
m
essage.
Aft
er
that
the
hop
again
sta
rts
the
rout
e
disco
ver
y
proc
ess.
3.2.4
Desti
natio
n
se
qu
e
nce
d dist
an
ce vecto
r routi
ng pr
oto
c
ol (D
SDV)
As
pe
r
the
the
ory
of
Be
ll
m
an
al
gorithm
,
it
is
a
ta
ble
dr
ive
n
r
ou
ti
ng
progra
m
.
Her
e
the
authors
descr
i
be
the
co
ncep
t
of
routin
g
loop
prob
le
m
us
ing
th
ei
r
al
gorithm
.
In
this
al
gorith
m
ro
utin
g
ta
bl
e
store
the
se
quenc
e
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol. 2, N
o.
1, M
arch 2
021
:
33
–
42
36
nu
m
ber
.
Ba
sic
al
ly
the
sequ
e
nce
nu
m
ber
is
us
e
d
e
ven
nu
m
ber
fo
r
the
a
ct
ive
netw
ork
and
od
d
nu
m
ber
f
or
inact
ivate
net
work.
Wh
e
n
th
e
routin
g
inf
or
m
at
ion
ci
rcu
la
te
d
am
on
g
inac
ti
ve
node,
at
t
hat
tim
e
occu
r
s
m
or
e
sen
ding tr
oubl
es [22
]
.
3.2.5
Op
ti
m
iz
ed
li
nk stat
e routin
g p
ro
t
oco
l
(OLSR
)
It
is
a
proacti
ve
r
ou
ti
ng
prot
oc
ol
an
d
us
e
d
f
or
IP
r
ou
ti
ng
[
23
]
.Basi
cal
ly
it
is
com
patibil
ity
with
m
ob
il
e
ad
hoc
netw
orks
an
d
a
d
hoc
netw
orks.
T
o
i
den
ti
fy
a
nd
set
up
t
he
tran
sm
i
ssion
li
nk
ov
e
r
netw
ork
the
n
it
m
us
t
us
e
d
hello
a
nd co
nt
ro
l
m
essag
e
of
t
opology. In
OL
SR,
e
ach
node
is
cal
c
ulati
ng
t
he
nex
t
h
op d
est
inati
on u
si
ng
sh
ort
est
fo
rw
a
r
ding
path.
3.2.6
Op
e
n
s
hortest
path first
It
is
us
e
d
t
o
fin
d
t
he
le
ast
–
c
ost
path
from
a
so
urce
no
de
t
o
a
desti
nation
node
within
a
gro
up
of
no
de
s
.
As
s
how
n
i
n
Fi
g
ure
3,
a
group
of
r
oute
rs
us
i
ng
the
sam
e
rou
ti
ng
protoc
ol
f
or
al
l
intr
oduce
d
to
an
a
utonom
ou
s
syst
e
m
(A
S)
.
U
pon
j
oi
ning
the
AS
,
a
node
use
s
t
he
hello
pro
tocol
to
disco
ve
r
nei
ghbori
ng
nodes
.
The
n
it
form
s
adj
ace
ncies
wit
h i
ts
ne
w
neig
hbors
to
exc
ha
nge
routin
g
inf
orm
ation
[24]
.
Abo
ve a
ll
,
it
is
fau
lt
y
f
or
e
very
node
on
a
netw
ork
c
onnect
t
o
al
l
oth
er
no
de
of
th
e
net
work.
T
o
pr
e
ve
nt
this
sit
uation,
a
no
de
is
co
ns
ide
red
a
s
the
destinat
io
n
node.
It
is
c
onsid
ered
to
be
nei
ghbori
ng
no
de
of
eac
h
node
ov
e
r
netw
ork
and
com
m
un
ic
at
e
the
inf
or
m
at
ion
b
e
tween t
hem
.
Fig
ure
3
.
Co
nnect
ed
aut
onom
ou
s
syst
em
(
AS
S)
A
ba
ck
up
no
de
,
w
hich
is
al
ways
m
ai
ntain
update
record
s
for
s
uccess
f
ul
tran
sact
ion
so
that
if
th
e
pr
im
ary
desig
nated
no
de
c
r
ash
ca
n
be
re
placed
im
m
ediat
el
y.
At
the
t
i
m
e
of
re
gu
la
r
proce
ss,
each
no
de
rep
eat
e
dly flo
ods
update
d
m
e
ssages
t
o neig
hbori
ng
nodes
of
eve
ry
node. T
his m
essage in
dicat
es it
s stat
us an
d
pro
vid
es
t
he
c
os
t
f
or
t
opolog
ic
al
database.
Wh
e
n
flo
od
i
ng
m
essage
are
pro
ve
d
ack
nowl
edg
em
ent
that
m
eans
syst
e
m
is
reli
able.
A
no
de
ca
n
c
hec
k
w
heth
er
t
he
i
nco
m
ing
li
nk
is
old
e
r
or
ne
wer
us
in
g
se
que
nce
nu
m
be
r.
Wh
e
n
the
cost
is
cha
ng
e
the
n
it
sends
al
l
the
s
e
m
essages.
Da
ta
base
Me
ssa
ge
s
pr
ov
i
de
the
seq
uen
ce
num
ber
f
or
the
entire
c
ha
nnel
,
wh
ic
h
is
he
ld
by
se
nder
.
Wh
e
n
t
he
valu
e
is
com
par
in
g
with
t
he
se
nder,
the
n
receiv
er
can
reso
l
ve
the
m
os
t
cu
rr
e
nt
val
ue
s.
When
a
li
ne
is
delive
rin
g
then
this
m
essage
is
fu
ll
y
us
e
d
i
n
the
syst
em
as
the
resu
lt
of
this
al
gorithm
is
that
each
pai
r
of
nei
ghbori
ng
node
s
detect
s
the
m
os
t
rece
nt
data
and
ne
w
inf
orm
at
ion
is t
ran
sm
it
te
d
on this
way [
25, 2
6].
4.
DETE
CTIO
N
A
PP
ROA
CHE
S OF
T
HE
WORMH
OL
E ATTA
CK
We
ha
ve
discu
ssed
t
he
dif
fere
nt
te
ch
nique
of
intr
us
io
n
de
te
ct
ion
syst
em
f
or
w
or
m
ho
le
s
at
ta
ck
a
nd
cat
egorized
t
he
dif
fer
e
nt
te
c
hn
i
qu
e
in
asce
nd
i
ng
or
der
from
ye
ar
2013
to
20
16.
In
[
27]
,
a
wise
s
olut
ion
is
pr
esc
ribe
d
to
e
rad
ic
at
ed
w
orm
ho
le
a
tt
acks
for
ad
-
hoc
network
by
pr
ov
i
ding
dir
ect
ion
a
l
antenn
a
t
o
th
e
nodes
.
Node
us
es
t
he
def
i
nite
reg
i
ons
of
t
heir
a
nten
na
in
est
a
blishi
ng
c
onne
ct
ion
a
m
on
g
t
hem
.
Each
pa
ir
of
node
has
evaluate
d
the
di
recti
on
of
rece
iving
the
in
for
m
at
ion
f
ro
m
ei
ther.
He
nce
re
l
at
ion
betwee
n
consecuti
ve
ne
ighbors
is
est
ablished
on
ly
if
the
dire
ct
ion
of
in
form
at
ion
flo
w
of
both
the
nodes
i
s
in
arr
a
ngem
e
nt
with
one
a
nothe
r.
This
ad
diti
on
al
inform
at
ion
enab
le
worm
ho
le
disco
ver
y
a
nd
intr
oduces
t
he
netw
ork
fl
uctuati
on.
S
o
that
it
can
be
sm
oo
thly
s
po
t.
I
n
[
28
]
,
th
e
aut
hors’
pro
po
s
ed
a
m
or
e
si
m
ple
too
l
kn
own
as
“
Packe
t
le
ashes“
acco
rd
a
nc
e
with
t
he
recog
niti
on
of
ge
ographical
an
d
tem
po
ral
le
as
he
s.
T
he
i
nform
at
ion
pro
vide
d
t
o
the
packet
s
tha
t
con
t
ro
ls
the
t
r
ansm
issi
on
dis
ta
nce
cal
le
d
L
each.
The
distance
of
se
nd
e
r
an
d
t
he
receiv
er
is
s
pecifie
d
by
the
geog
raphical
le
ash.
Whe
n
the
receivin
g
node
s
acce
pt
t
he
data
pac
kets,
it
cal
culat
es
the
distance
a
nd
tim
e
of
t
he
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Su
r
vey
of wor
mhole
attack i
n
wi
rel
ess se
nsor
networks
…
(
Um
ash
ankar
Ghu
ga
r
)
37
transm
issi
on
.
The
recei
ver
a
naly
ze
now
on
com
par
in
g
thi
s
in
form
at
ion
can
det
ect
w
he
ther
the
pa
ck
et
has
forw
a
r
ded
th
r
ough
w
orm
ho
le
at
ta
cks
or
no
t.
Her
e
the
pac
ke
t
lim
it
a
ti
on
is
known
by
te
m
poral
le
as
hes,
wh
ic
h
is
determ
ining
the
distance
it
can
c
over
the
m
os
t.
In
t
his
te
chn
i
qu
e
the
po
sit
ion
of
no
de
is
not
that
im
po
rta
nt
rather
tha
n
ti
m
e
facto
r
play
s
an
im
po
rtant
r
ole.
It
can
acce
ss
the
ti
m
e
calcu
la
ti
on
an
d
it
s
com
par
iso
n
up
to
an
order
of
na
nos
econd.
O
n
eac
h
packet
,
the
s
end
e
r
m
ention
an
a
uthorize
d
t
i
m
e
bar
,
w
hich
is
com
par
e
d
by
the
receiver
a
nd
th
e
pac
ket
f
orwa
r
ding
distance
is
sim
ply
giv
en
by
the
pr
oduct
of
velocit
y
of
li
ghts
a
nd
tran
sm
i
ssion
tim
e.
In
case
of
a
la
r
ge
tim
e
di
ff
ere
nce
it
in
di
cat
es
the
prese
nce
of
w
or
m
hole
.
I
n
[
29]
,
t
he
auth
or
s
pu
t
f
or
ward
a
“gr
a
ph
the
or
e
ti
cal
”
app
r
oach
to
preve
nt
w
orm
ho
le
at
ta
cks.
This
co
nce
pt
is
pur
el
y
est
ablished
on
the
“
lo
c
at
io
n
awar
e
guar
d
no
de
”
(LAG
Ns).
Wh
e
n
t
he
key
est
ablishm
ent
process
is
us
e
d
for
detect
ing
worm
ho
le
at
ta
ck
a
nd
it
al
so
us
ed
the
decode
d
m
essage.
If
sam
e
m
essage
is
he
ard
f
ro
m
on
e
gu
a
r
d
or
tw
o
LAGNs
are
he
ard
f
ro
m
diff
e
re
nt f
a
r
a
way
L
AGNs
t
he
n worm
ho
le
is d
et
ect
ed
.
In
[
30
]
,
the
a
ut
hors
pro
posed
that
w
orm
ho
le
at
ta
cks
in
sta
ti
on
ary
se
nsor
net
work
are
in
vest
igate
d
us
i
ng
netw
ork visual
iz
at
ion
. In
this
m
et
ho
d, the si
gnal
stre
ng
t
h de
te
rm
ines the
distance. Eac
h se
ns
or c
onveys a
ll
the
gathe
red
inf
or
m
at
ion
to
the
m
ai
n
sta
ti
on
.
T
he
c
on
t
ro
ll
er
c
om
pu
te
s
the
ne
tworks
physi
cal
topolo
gy
us
i
ng
se
nsor
pr
e
dicte
d dist
a
nce.
If a
wor
m
ho
le
att
ack is
present
the
n
it
is
seen
that
a stri
ng
pu
ll
in
g
th
e
netw
ork
te
rm
in
al
s,
if
no
t
the
n
t
he
t
opol
og
y
is
flat
.
In
[
31]
,
t
he
a
ut
hors
a
dopte
d
li
gh
twei
ght
co
un
te
rm
easur
e
f
or
w
orm
ho
le
a
tt
ack
cal
le
d
LITE
W
ORP
an
d
this r
esult
has
a
dv
a
nt
ages
of
ve
ry
quic
k
detect
ion
o
f
worm
ho
le
at
ta
cks
an
d
the
l
os
s o
f
fr
act
io
n
of
pac
kets
is
ver
y
le
ss.
I
n
[
32
]
,
here
the
a
uthor
’s
e
m
ph
asi
s
on
the
“r
ound
–
tri
p
t
rav
el
ti
m
e”
(RTT
)
m
essage,
wh
ic
h
pro
vid
es
the
m
axi
m
u
m
tim
e
s
re
qu
i
re
for
t
he
tran
sm
issi
on
.
Wh
e
n
this
tim
e
is
m
ulti
plied
wit
h
sp
ee
d
of
li
gh
ts
it
giv
es
t
he
distanced
tra
velle
d.
N
ow
this
dist
ance
is
t
o
be
c
om
par
ed
with
th
e
predict
ed
distance
.
If
t
her
e
is
a
la
r
ge
diff
e
rence
then
it
threat
worm
ho
le
at
ta
cks.
In
[
33
]
,
t
he
a
uth
ors
desc
ribe
t
hat,
worm
ho
le
at
ta
cks
in
f
ound
in
m
ulti
path
r
outi
ng.
I
n
cas
e
of
ne
w
r
oo
t
re
quirem
ents
s
ource
ex
cess
by
usi
ng
r
ou
te
re
quest
(
RR
EQ)
in
the
netw
ork
an
d
the
n
t
he
respo
n
se
is
wa
it
ed.
T
he
inte
r
m
ediat
e
node
on
ly
pass
a
wa
y
this
r
ou
te
re
qu
e
st
(RREQ
).
O
n
th
e
sa
m
e
tim
e
th
e
receiver
will
wait
to
get
r
oute
after
getti
ng
r
ou
te
request
(RREQ)
.Stat
ist
ic
al
An
al
ysi
s
of
M
ulti
-
path
(S
A
M)
is
i
ntr
oduc
ed,
that
us
e
P
m
ax
an
d
.
wh
i
ch
a
re
hi
gh
e
r
if
worm
ho
le
a
tt
ack
is
pr
ese
nt.Pm
ax
giv
es
the
pro
ba
bili
ty
of
t
he
r
ou
te
s
out
of
al
l
possi
ble
r
oute
and
(t
heta)
is
t
he
dif
fer
e
nce
be
tween
top
tw
o
fr
e
qu
e
ntly
pa
per
e
d
li
nk
s
.
If
a
w
or
m
ho
le
s
at
ta
ck
is
m
or
e
than
PM
F
(
pro
bab
il
it
y
m
ass
functi
on)
the
n
it
giv
es
h
i
gh
f
requen
cy
.
H
e
re a
ut
hors
al
s
o
a
naly
sis t
he
m
ulti
path
r
outi
ng and
D
SR
with
f
ine
co
m
par
iso
ns
.
In
[
34
]
a
”
hel
lo
co
ntr
ol
m
es
sage”
is
us
e
d
to
detect
wor
m
ho
le
at
ta
cks
as
co
ns
e
nt
wit
h
OLS
R
in
par
ti
cula
r.
He
us
e
d
the
a
ggre
gate
of
hello
m
essage
ti
m
e
int
erv
al
(H
MT
I)
that
li
e
within
a
j
it
te
r.
A
range
r=
[T
-
α
,
T+∞]
is
coat
ed.
In
ra
nge
H
MTI
are
co
ns
id
ered
valid
or
el
se
it
is
out
of
s
et
of
r
ules.
In
c
ase
of
unusual
HMTI
seco
nd
a
ry
che
cks
a
re
do
ne.
I
n
a
dd
it
io
n
to
t
his
an
untr
ue
posit
ive
al
arm
i
n
neg
at
e
d
in
c
ase
of
wea
k
w
ork
ing
node
wh
ic
h
ha
s
m
any
pack
et
s
but
this
is
not
the
case
of
a
nd
at
ta
ckin
g
node.
In
[
35
]
,
t
he
auth
or
s
im
ple
m
ented
delay
pe
r
ho
p
ind
ic
at
io
n
(
D
el
PH
I
)
to
dete
ct
worm
ho
le
at
ta
cks.
It
is
al
so
wor
k
on
t
he
sam
e
pr
inciple
of
com
par
ison
of
path
ti
m
e
distance
an
d
pre
dicte
d
distance.
T
his
process
w
orks
in
t
wo
ph
a
ses,
first
is
c
ollec
ti
on
of
route
pat
h
by
the
receive
rs
and
sen
ders
in
cl
ud
e
a
DRE
Q
pack
et
s
sim
il
ar
to
the
c
on
ce
pt
of
SA
M
a
nd
s
ign
it
befor
e
se
ndin
g.
On
t
he
getti
ng
the
pac
ket
th
e
receiver
has
to
ad
d
it
s
ID
a
nd
1
ho
p
co
unt
is
increm
ente
d.
T
he
m
ini
m
u
m
delay
an
d
hope
c
ount
in
f
or
m
at
ion
are
util
iz
ed
for
the
m
ini
m
u
m
detect
ion
.
I
n
t
he
seco
nd
ph
ase
,
“
round
–
trip
t
rav
el
ti
m
e
”
(RTT
)
is
us
ed
to
cal
culat
e
the
tim
e
diff
e
r
ence
betwee
n
t
he
total
nu
m
ber
of
se
nt
in
for
m
at
ion
and
ack
nowle
dgem
ent
receive
d.
I
n t
his
proce
ss
the
delay
pe
r
hop
value
(
D
PH
)
is
cal
culat
ed a
s
RTT/
2h,
wh
e
re
h
is
the
hop
c
ount
to
t
he
defi
nite
con
se
cuti
ve.
In
norm
al
case
ti
ny
hops
ha
ve
ti
ny
RT
T
w
her
e
a
s
in
case
of
worm
ho
le
at
ta
ck
the
ti
ny
ho
ps
a
re
gia
nt
R
TT.
I
f
one
del
ay
per
hop
val
ue
(
DPH
)
cr
osse
s
the
t
hr
es
ho
ld
val
ue
then
al
l
path
s
nex
t
to
t
his
tre
at
ed
as
unde
r
worm
ho
le
s
at
ta
cks
I
n
[
36]
,
the
aut
hors
us
e
d
a
uniq
ue
te
c
hn
i
qu
e
of
rad
i
o
fin
ge
r
pri
nting.
It
init
ia
te
s
with
the
rad
i
o
sig
nal
receivi
ng
by
the
fin
ge
rprin
ti
ng
de
vic
e
and
t
hen
t
he
sign
al
is
conve
rted
t
o
the
di
gital
for
m
.
The
sign
al
passing
is
posit
ion
e
d
a
nd
it
s
char
act
erist
ic
s
a
re
de
scrib
ed
.
A
set
of
char
act
e
rs
f
rom
fing
er
pri
nts
is
la
te
r
us
ed
for
ap
par
at
us
ide
ntific
at
ion
.
In
[
37
]
,
the
a
utho
r
s
pro
po
se
d
a
m
et
h
od,
wh
e
n
a
se
nd
e
r
sen
d
a
RR
EQ
m
essage
to
r
e
cei
ver
,
the
n
it
wait
s
f
or
the
RR
EP.
Ou
t
of
the
num
ber
of
RR
EP
receive
d
by
th
e
source,
the
R
REP
with
high
est
fr
e
qu
e
ncy
is
com
par
e
wit
h
the
pr
e
de
fine
d
val
ue.
If
t
he
pack
et
dro
p
rati
o
is
la
r
ger
tha
n
pack
et
sent
rati
o
n
the
n
it
im
plies
that
w
or
m
ho
le
is
present.
I
n
[38],
the
aut
hors
pro
po
s
e
d
that, tw
o plot
node
s ar
e
con
ne
ct
ed
by t
unnel
su
c
h
as t
hey ar
e n
ei
gh
bors.
The
r
oute
requ
est
(
RR
E
Q
)
a
nd
t
opol
og
y
co
nt
ro
l
m
essages
(
TCM
)
a
re
c
on
ve
y
am
on
g
thes
e
pl
ot
node
s
thr
ough
tu
nne
ls.
By
usi
ng
t
he
e
xtra
t
unne
l
nodes
,
these
node
s
hav
e
the
s
hortest
pa
th.
Af
te
r
the
li
nk
i
s
est
ablishin
g,
the
at
ta
cke
r
sel
ect
one
a
no
t
he
r
as
m
ulti
po
int
relay
s
(MRP
s
).
As
res
ult
fe
w
to
polo
gies
c
on
t
ro
l
m
essages
a
nd
data
pac
kets
a
r
e
le
aked
th
rou
gh
the
tu
nn
el
.
A
s
co
ns
e
qu
e
nce
false
to
po
l
og
y
inf
or
m
at
ion
is
sp
rea
d
thr
ough
the
ne
tworks
.
I
n
[
39
]
,
the
auth
or’s
propose
d
a
tr
us
t
base
d
m
odel
fo
r
detect
io
n
in
wireless
s
ens
or
netw
orks.
In
t
r
us
t
base
d
syst
e
m
,
each
no
de
has
s
om
e
values,
w
hich
is
ca
ll
ed
trust
val
ue
.
By
us
in
g
this
trust
values
the
s
ou
rce
node
is
cal
culat
ed
t
he
act
ual
r
oute
to
t
he
destinat
io
n.
Wh
e
n
t
he
tra
nsm
issi
on
occur
s
over
netw
ork,
in
w
hi
ch
nu
m
ber
of
pack
et
s
dr
op
ra
ti
o
is
high
m
eans
t
ru
st
val
ue
is
le
ss
a
nd
worm
ho
le
at
ta
cks
is
pr
ese
nt
in
the
netw
ork
.
If
the
t
r
us
t
val
ue
is
high
m
eans,
al
l
th
e
pa
ck
et
s
wh
ic
h
is
rec
ei
ved
by
the
de
sti
nation,
it
indi
cat
es
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol. 2, N
o.
1, M
arch 2
021
:
33
–
42
38
that
the
tr
us
t
va
lue
is
hi
gh
on
the
nei
ghbori
ng
no
de
of
s
our
ce
node
betwe
en
the
s
ource
t
o
de
sti
nation.
I
n
[
40]
,
the
a
uthor
’s
pr
opos
e
d
a d
ist
ri
bu
te
d
i
ntell
igent
a
ge
nt
-
base
d sy
stem
.
Her
e
t
he
am
biti
on
is
the u
se of g
e
ne
rali
zed
intru
si
on
detec
ti
on
syst
em
(
IDS
)
f
ram
ewo
r
k
wh
ic
h
is
s
o
li
gh
t
weig
ht
that
it
can
r
un
on
t
he
sens
ors
no
de
and
it
identifie
s
the
worm
ho
le
at
ta
cks
al
on
g
with
it
s
at
ta
cker
s.
Wh
e
n
t
hat
at
ta
cker’s
node
i
s
fou
nd
i
n
the
ne
twork
,
then
it
is
in
for
m
ed
with
a
n
i
nd
ic
at
io
n
m
essage.
Af
te
r
that
each
node
m
akes
thei
r
co
ncl
us
io
n
on
t
he
ba
se
o
f
consecuti
ve
no
de
re
peat.
I
n
[
41]
,
it
is
assum
ed
that
be
hav
i
ors
of
a
node
a
re
con
t
ro
l
by
it
s
con
s
ecuti
ve
no
de
s.
A
node
us
es
it
s
ne
ighbor
node
t
o
se
nd
r
ou
te
re
qu
e
st
(
RR
EQ
)
m
essage
to
the
destinat
io
n
no
de
.
I
f
the
se
nder
didn’t
get
route
re
ply
(
RR
EP
)
m
essage
withi
n
predi
ct
ed
tim
e,
then
sen
der
c
on
cl
ude
the
pr
e
sence
of
worm
ho
le
s
at
ta
c
k
and
e
nclo
se
thi
s
route
in
the
li
st
of
w
orm
ho
le
at
ta
cks
li
st.
A
conj
ug
at
iv
e
no
de
that
is
m
ana
ged
by
eve
ry
node
that
co
ns
ist
s
of
RR
EQ
seq
ue
nce
nu
m
ber
,
Neig
hbors
node
I
D,
se
nd
i
ng
&
recei
ving
tim
e
of
RR
E
Q.
T
he
m
axi
m
u
m
tim
e
li
m
it
equ
al
t
o
WPT/
2
is
wait
ed
by
the
se
nder
if
RR
E
Q
is
delay
ed
m
or
e
t
han
th
us
it
i
nd
i
c
at
es
the wo
rm
ho
le
att
acks and
e
nt
irel
y i
t do
es
n’
t
su
pp
or
t
DS
R R
ou
ti
ng
prot
oco
l
.
In
[
42
]
,
Al
t
he
sen
der’s
node
s
wait
f
or
a
c
knowle
dgm
ent
(
ACK
)
m
essage.
I
f
ACK
m
essage
is
not
receive
d
t
hen
t
he
ne
xt
node
is
at
ta
ck,
wh
ic
h
is
worm
ho
le
a
tt
acks.
ACK
m
essage
sho
ul
d
no
t
retrace
the
pat
h
and
se
nt
betwe
en
t
he
se
pa
rati
on
by
tw
o
ho
ps.
Now
Tim
e
t
o
Li
ve
(TT
L)
play
s
a
great
r
ole
si
nce
t
he
pa
th
is
diff
e
re
nt.
I
f
t
he
ACK
m
essage
is
no
t
receive
d
within
TTL
t
he
n
worm
ho
le
at
ta
cks
is
detect
ed.
In
[
43]
,
the
a
uthor
s
us
e
d
two
ste
p
m
echan
ism
fo
r
the
detect
ing
t
he
w
or
m
ho
le
a
tt
acks.
The
firs
t
ste
ps
co
ns
ist
of
tw
o
m
et
ho
ds.
I
n
the
first
m
et
ho
d,
th
e
no
de
an
d
his
nex
t
node
are
i
den
ti
fie
d
by
us
i
ng
rou
nd
-
trip
-
ti
m
e
(RTT)
a
nd
i
n
t
he
sec
ond
m
et
hod
their
li
st
is
m
a
de
a
nd
if
the
de
sti
nation
node
i
s
not
i
n
that
li
st
then
it
is
do
ub
t
f
ull
in
natu
re.
In
the
sec
ond
ste
p
m
echan
ism
,
after
detect
ion
of
doubt
f
ull
li
nk
t
he
at
ta
ck
is
con
cl
uded
usi
ng
RTS/
CTS
m
et
ho
d.
I
n
[
44]
,
the
auth
or
s
us
e
d
A
ODV
an
d
D
S
R
r
ou
ti
ng
pr
oto
col.
Her
e
al
so
a
Tr
us
t
base
d
secur
it
y
m
od
el
is
us
e
d
f
or
det
ect
ing
intru
si
on.
T
his
m
od
el
has
be
en
intr
oduce
d
to
identify
the
at
ta
cks,
w
hic
h
is
cal
le
d
sta
ti
cal
ly
m
et
ho
d.
If
a
ny
connecti
on
get
s
doubtful,
t
he
n
t
he
tru
st
value
is
cal
culat
e
d
t
o
determ
ine
the
worm
ho
le
at
ta
cks.
In
th
e
tr
us
t
m
od
el
,
nodes
m
on
it
or
ed
thei
r
neig
hbou
rin
g
on
t
he
basis
of
pack
et
drop
pat
te
rn
.
If
a
ny
node
is
fo
un
d
to
be
doubt
then
st
oc
k
tr
ust
is
identifie
d
by
the
no
de,
w
he
ther
t
he
node
is
aff
ect
e
d
by
worm
ho
le
at
ta
ck
or
not.
In
[
45]
,
the
auth
or
s
pro
po
s
ed
Digital
i
nvest
igati
on
to
de
te
ct
w
or
m
ho
le
at
ta
cks
in
W
SN
s.
WSN
ar
e
e
xp
la
ine
d
th
at
a
dd
gen
e
rati
on
a
nd
protect
s
flo
w
of
e
vid
e
nces
about
se
nsors
node
cha
racter
ist
ic
s
in
the
ne
twork
.
A
gro
up
of
detect
ive
node
s
are
sprea
d
ov
er
the
net
wor
ks
to
co
ntr
ols
th
e
topolo
gy
an
d
datag
ram
passing
by
sens
or
nodes
.
Ob
se
r
vation
node
an
d
base
s
ta
ti
on
no
de
j
oi
ntly
form
s
different
WSN
net
works
cal
le
d
obser
vatio
n
network.
Fr
e
qu
e
ncy b
an
ds
a
re u
sed
to establi
sh
li
nk bet
ween
observ
ers
a
nd
the b
as
e
sta
ti
on
b
ut
t
hi
s
is
not
sup
por
te
d
by
sens
or
s
node
.
The
detect
ion
s
ensiti
vity
of
se
ns
or
node
is
le
ss
than
the
ob
s
erv
e
r.
In
[
46
]
,
t
he
aut
hors
pro
po
s
ed
a
'
con
flic
ti
ng
-
set
'
for
eac
h
node
is
m
ade
to
filt
erin
g
the
false
m
easur
em
ent
of
dista
nce
bu
t
i
ts
biggest
li
m
i
t
at
ion
was
that,
it
w
or
ks
only
wh
e
re
there
is
no
pac
ke
t
loss
b
ut
w
he
n
at
ta
c
ker
s
at
ta
cks
the
n
t
he
Pa
cket
dro
ps
is
ce
rtai
n
to h
a
ppen
. So t
he
syst
em
is u
nder
a
worm
ho
le
att
acks.
In
[
47]
,
t
he
a
ut
hors
pr
opos
e
d
a
m
od
el
,
w
hich
c
reate
a
cl
ust
er
us
in
g
no
of
nodes
in
MA
N
ET.
I
n
t
his
pap
e
r
va
rio
us
data
str
uctu
re
are
e
xp
la
ine
d
a
nd
al
gorithm
is
al
so
pro
posed
.
He
re
tw
o
la
ye
rs
a
re
m
ention
in
the
cl
us
te
r,
wh
e
re
on
e
node
is
tre
at
ed
as
cl
us
te
r
head
am
ong
se
ver
al
no
des.
W
hen
a
node
is
a
ff
ect
ed
by
a
w
orm
ho
l
e
at
ta
ck
in
the
la
ye
r1
,
t
he
n
w
hi
ch
in
form
s
to
the
cl
us
te
r
he
ad
of
la
ye
r
1.A
fter
tha
t
cl
us
te
r
hea
d
of
la
ye
r
1
will
ind
ic
at
es
the
cl
us
te
r
hea
d
of
la
ye
r2
a
bout
the
abno
rm
al
no
de
.
S
o
that
cl
ust
er
hea
d
of
la
ye
r
2
ind
ic
at
e
the
m
essage
to
al
l
the
cl
us
t
er
head
of
la
ye
r1,
the
n
t
he
cl
ust
er
he
ad
of
la
ye
r1
i
nfor
m
th
e
m
essages
to
their
resp
ect
iv
e
node
within
their
cl
ust
er.
In
[
48]
,
th
e
aut
hors
pro
po
sed
l
ocali
zat
ion
-
base
d
syst
em
s,
wh
ic
h
a
re
vu
lnera
ble
to
w
orm
ho
le
at
ta
cks
as
t
hey
m
anipu
la
te
th
e
local
iz
at
ion
m
et
ho
d
To
pre
ven
t
t
he
w
or
m
ho
le
at
ta
c
k,
a
'
distance
-
c
onsis
te
ncy
-
base
d
sec
ur
e
locat
ion
'
sc
hem
e
was
im
ple
m
e
nted,
T
his
w
orks
on
the
detec
ti
on
,
e
xact
loc
at
ion
a
nd
tra
pping
of
worm
ho
le
at
ta
cks
In
[
49
]
,
t
he
auth
ors
use
d
te
chn
i
qu
es
th
at
identify
the
worm
ho
le
at
ta
cks.
I
n
the
fir
st
way
al
gorithm
us
es
hop
c
ountin
g
m
et
ho
d,
re
bu
il
t
local
m
aps
at
ever
y
nodes
a
nd
the
n
a
diam
eter
featu
res
to
i
den
ti
fy
by
the
prob
le
m
s
due
to
worm
ho
le
at
ta
cks.
T
he
evaluate
d
r
ound
t
rip
ti
m
es
(
RTT)
bet
wee
n
the
co
ns
ec
utiv
e
node
s
are
us
e
d t
o co
m
par
e in
the
se
cond
w
ay
.
Its
m
ajo
r
a
dv
a
nta
ges
is
not
requi
red a
dd
it
io
nal
har
dware
a
nd c
on
s
um
e
le
ss
en
e
r
gy.
In
[50],
t
he
a
utho
rs
pro
posed
th
at
at
ta
cker
s
m
a
y
rec
ord
t
he
l
oc
at
ion
of
pac
ke
ts
in
WSN
an
d
se
nd
them
to
on
e
m
or
e
l
ocati
on
an
d
a
gain
tra
ns
m
it
the
m
in
to
th
e
netw
ork.
When
it
f
ound
the
roots,
t
he
w
or
m
ho
le
detect
ion
proc
ess
is
goin
g
on,
w
hich
c
ounts
dif
f
ere
nce
between
t
he
neig
hbour
nodes
to
ano
t
her
node
?
If
t
he
diff
e
re
nce is m
or
e t
ha
n
the
de
sti
nation n
od
e
detect
the
wor
m
ho
le
s.
In [51
]
, th
e a
uthor
s
propose
d
the
stat
ist
ic
al
analy
sis t
o
id
e
ntify t
he w
or
m
ho
le
att
ac
ks
in
WSN.T
he pr
op
os
e
d
al
gorithm
is cat
eg
or
i
es
by
thr
ee
par
ts
.i.
e.
Stat
ist
ic
al
an
alysis
m
et
ho
d, w
hich
is
u
se
d f
or rou
ti
ng i
nfor
m
at
ion
for det
ect
ing
t
he worm
ho
le
att
acks.
Determ
inati
on
of the
vulnera
bl
e w
orm
ho
le
s.
Ti
m
e con
strai
nt
s is u
se
d for
va
li
dation
i
n worm
ho
le
att
acks.
It
us
es
m
ulti
-
path
routin
g,
ti
m
e
co
ns
tr
ai
nts
a
nd
sta
ti
sti
cal
analy
sis
to
ver
if
y
the
vulne
ra
ble
co
nnect
io
n.
It
do
es
n’t
need
tim
e
synchron
i
sat
ion
,
directi
onal
a
nten
na
a
nd
GPS.
I
n
t
his
m
et
ho
d
it
ca
n
worm
ho
le
at
ta
cks
wit
h
high
qual
it
y
of
acc
ur
acy
.
I
n
[
52
]
,
the
a
uthor
s
propose
th
e
secur
it
y
em
erges
as
a
ce
ntrall
y
in
M
ANET.
T
he
app
li
cat
io
ns
of
MANE
T
we
re
dep
l
oyed
i
n
va
rio
us
fiel
ds.
Worm
ho
le
at
ta
ck
is
a
se
ver
e
destr
uctive
in
natu
re,
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Su
r
vey
of wor
mhole
attack i
n
wi
rel
ess se
nsor
networks
…
(
Um
ash
ankar
Ghu
ga
r
)
39
wh
ic
h
is
sm
oo
thly
res
olv
e
d
in
netw
orks
bu
t
t
ough
t
o
obse
rv
e
.
It
is
visible
e
ve
n
if
t
he
intr
ud
er
has
no
t
ne
go
ti
at
ed
at
any
sit
uatio
n
an
d
re
st
of
al
l
com
m
un
ic
at
ion
giv
e
s
s
ecu
rity
,
no
velty
,
a
uth
entic
it
y
an
d
c
onfide
ntly
.
In
[
53
]
,
the
auth
or’s
prese
nts
diff
e
re
nt
ty
pes
of
se
nsor
node
s
a
nd
m
any
la
ye
r
wise
at
ta
cks
m
us
t
be
present
in
the
networ
k.
Worm
ho
le
att
acks a
re
us
e
d i
n t
his
pa
per in
a
tt
ack m
od
el
, w
hich
is sm
oo
thl
y res
olv
e
d
i
n n
et
work
s
bu
t t
ough
to
ob
s
er
ve.
Her
e
t
he
aut
hors
pro
pose
d
a
m
et
ho
d,
wh
ic
h
is
us
e
d
t
he
Mi
nt
r
oute
prot
oco
l.
In
[
54
]
,
the
a
uthors
a
ddress
the
m
ult
iple
–
hop
Mo
bile
ad
ho
c
netw
orks
,
wh
e
re
each
node
act
s
as
a
ho
st
a
nd
r
ou
te
r
in
the
route.
Au
t
hor
pro
po
se
d
a
te
c
hn
i
qu
e
,
w
hich
i
s
ide
ntify
the
at
ta
cks
without
usi
ng
sync
hro
nizat
ion
re
qu
ire
m
ents.
T
he
bas
ic
thi
ng
is
to
fin
d a
noth
er
way
from
so
ur
ce
t
o
nex
t
ho
p a
nd
finall
y
it
cal
culat
es
the
no
of
hops
for
detect
ing
w
orm
ho
le
s
at
ta
cks.
I
n
[
55]
,
it
us
es
pac
ket
encapsulat
io
n
te
chn
iq
ue.
He
r
e
pack
et
s
a
re
e
ncapsulat
ed
i
n
AOD
V
prot
oc
ol.
I
n
this
te
ch
nique,
le
ss
hop
c
ount
is
cr
eat
ed
an
d
it
is
c
om
par
ed
t
o
ot
her
no
rm
al
li
nk
s.
M
LD
W
m
ai
ntain
a
bi
g
structu
re,
whic
h
is
di
vi
ded b
y
04 parts
,
i.e
:
Exam
inati
on
lay
er.
Discl
osure
lay
er.
Re
organ
iz
at
io
n l
ay
er.
Segregati
on la
ye
r.
Her
e
the
First
03
la
ye
rs
w
ork
as
a
Detect
or
and
la
st
la
ye
r
works
as
a
Pr
e
ven
te
r
for
w
orm
ho
le
at
ta
cks
in
MA
NET
usi
ng
AOD
V
protoc
ol.
I
n
[56]
,the
a
utho
r
’s
pro
po
se
d
a
te
c
hn
i
qu
e
,
wh
ic
h
is
giv
es
secu
r
e
data
transm
issi
on
usi
ng
neig
hbou
r
node
analy
sis
concept
to
i
den
ti
fy
the
w
or
m
ho
le
at
ta
cks
in
M
AN
E
T.
Thi
s
te
chn
iq
ue
a
naly
ze
the
neig
hbourin
g
no
des
.
so
that
it
chec
ks
the
reli
a
bili
t
y
of
the
nodes
for
data
tra
nsm
issi
on
on
t
he
netw
ork
,
Acc
ordin
g
t
o
this
te
chn
i
que,
a
node
s
en
d
a
r
equ
e
st
to
it
s
ne
ighbou
r
no
des
and
it
m
ai
ntain
the
request
a
nd
res
pons
e
syst
em
.
Her
e
node
m
ain
ta
ins
a
ta
ble
f
or
t
racin
g
the
tim
e
ou
t.
If
a
node
does
n’
t
ge
t
th
e
rep
ly
tim
e
that
m
eans
at
ta
cks
occurs
in
t
he
ne
twork
.
The
e
nt
ire
node
f
r
om
so
urce
to d
est
i
nation
is
a
naly
zed
t
o
detect
the
w
orm
ho
le
s
at
ta
ck
us
in
g
AOD
V
prot
oco
l
i
n
MA
NET.
I
n
[57],
t
he
a
uthors
pr
opose
a
te
c
hn
i
que,
w
hich
is
li
able
to
dete
ct
worm
ho
le
s
at
ta
cks
in
MA
N
ET
us
i
ng
a
naly
sis
of
t
he
m
isbehav
i
ng
nodes con
ce
pt.
Acc
ordin
g
to the au
t
hors,
it
co
ncen
t
rates
on
t
he
detect
io
n
of the a
bnor
m
al
n
od
es a
nd
pr
e
ve
ntion
of
t
he
w
or
m
ho
le
a
tt
acks.
The
r
ou
te
discov
e
ry
process
i
s
us
ed
,
w
hich
i
s
a
sen
de
r
node
wa
nt
to
data
s
end
i
ng
process
with
a
nother
node
i
n
the
netw
ork,
it
ha
s
t
o
go
t
hro
ugh
it
s
r
oute
c
ache.
I
n
case
of
unavail
abili
ty
of
r
ou
te
s
betw
een
t
he
receive
r
a
nd
sen
der
tha
n
r
oute
is
disca
rd
e
d
a
nd
it
broa
dc
ast
RR
EP.
T
he
RR
EP
is
gen
e
rated,
wh
e
n
th
e
recei
ver
node
or
any
interm
ediat
e
no
de
has
got
the
rece
n
t
route
t
o
the
recei
ver
no
de.
A
nothe
r
im
portant
is
that
DS
R
prot
oco
l
is
use
d
to d
et
ect
the
node
s where
t
he
m
isbehav
i
ng
node
s ar
e sim
ple d
isc
ar
ded
a
nd not inclu
ding
into
the
routin
g
ta
ble
of D
SR.
I
n
D
S
R, p
aram
et
er
is
use
d
f
or
e
valu
at
ing
t
he
net
w
ork
pe
rfor
m
a
nce
i.e
j
it
te
r,
th
r
oughput
a
nd
de
la
y
.
In
[58],
her
e
the
a
uthors
use
d
a
g
ener
al
m
echan
i
sm
, which i
s
use
d
with
ou
t
ha
rdwar
e
.
It
ex
pl
ai
ns
the
detai
ls
ab
ou
t
pack
et
detect
io
n
te
c
hn
i
qu
e
.
T
hat
pack
et
ho
l
ds
t
he
inf
orm
a
ti
on
of
l
ocali
zat
ion
a
nd
cl
oc
k
sy
nc
hron
iz
at
ion
f
or
detect
ing
a
ff
ec
te
d
node
in
M
AN
E
T.
Detect
i
on
Pa
cket
ha
s
f
our
fiel
ds
:
total
ho
p
co
unt,
pro
cessi
ng
bit,
c
ount
to
reach
ne
xt
hop
and
ti
m
est
a
m
p
.T
his
fiel
ds
are
ad
de
d
to
t
he
hea
de
r
of
d
et
ect
ion
pac
ket
.
I
n
[
59
]
,
t
he
a
uthors
pro
po
se
d
a
nor
m
al
iz
ed
wo
rm
ho
le
local
i
ntr
us
io
n
detect
ion
al
gorithm
,
w
hich
is
up
gr
a
da
ti
on
ve
rsion
of
local
intru
si
on
detec
ti
on
r
ou
ti
ng
se
cur
it
y
in
M
A
NET.
In
this
t
echn
i
qu
e
a
n
in
te
rm
ediat
e
neighb
or
nodes
a
re
us
es
disco
ver
y
m
ec
han
ism
proces
s
an
d
pac
ket
dro
p
cal
c
ulator
.
Ba
sed
on
the
isolat
io
n
te
c
hniq
ue,
at
the
ti
m
e
of
transm
issi
on
over
the
netw
ork,
w
her
e
each
nod
e
receive
d packet
f
or
t
he c
onfirm
ed Worm
ho
le
n
odes.
In
[
60]
,
the
aut
hors
pro
pose
d
te
chn
iq
ue,
wh
i
ch
is
base
d
on
Hash
base
d
c
om
pr
ession
f
unct
ion
(H
C
F).
It
is
ba
sic
al
ly
us
ed
for
sec
ure
has
h
f
unct
ion
t
o
cal
culat
e
t
he
value
of
hash
fi
el
d
f
or
r
oute
re
qu
e
st
(RRE
Q)
passe
s
ov
e
r
the
net
w
orks.
He
re
A
O
DV
r
ou
ti
ng
protoc
ol
is
us
e
d
.A
s
per
the
a
ut
hors.
S
ource
node
sta
rts
the
route
disco
ver
y
proc
ess
f
or
sea
rch
i
ng
t
he
destinat
io
n
node
.
T
he
n
t
he
s
ource
no
de
com
pu
te
the
H
CF
an
d
al
so
co
m
pu
te
the
val
ue
of
ha
sh
fiel
d
with
R
REQ
an
d
it
pa
sses
to
his
nei
ghbo
rin
g
no
de.
If
the
val
ue
of
neig
hbori
ng
no
de
is
sam
e
to
the
value
of
destinat
i
on
no
de
.
At
th
at
sit
uation
t
he
destinat
io
n
node
rec
ei
ves
t
he
no
of
RR
E
Q.
Finall
y
the
destinat
io
n
no
de
im
ple
m
e
nt
the
HCF
c
oncept.
Othe
rw
i
se
the
oth
er
s
i
nterm
ediat
e
node
betwee
n
s
ource
to
destinat
io
n,
th
ey
will
i
m
ple
m
ent
HCF
has
h
fiel
ds
an
d
pa
sses
to
it
s
ne
xt
node.
If
t
he
c
al
culat
ed
has
h
value
is
com
par
ed
to
a
pp
e
nd
ha
sh
val
ue
an
d
gets
the
sam
e
resu
lt
then
the
destinat
ion
node
sen
d
ba
ck
RR
EP
m
es
sage
to
the
s
ource.
Othe
rw
ise
if
ca
lc
ulate
the
has
h
value
is
not
s
a
m
e
with
t
he
a
pp
e
nd
has
h
val
ue
t
hen
the
des
ti
nation
node dete
ct
s th
e RR
EQ
a
nd it
treat
ed
as
af
fec
te
d
no
de by w
orm
ho
le
att
ackers.
In
[
61
]
,
the
a
ut
hors
us
e
d
a
hy
br
id
te
ch
niqu
e
“
w
or
m
ho
le
r
esi
sta
nt
hybri
d
te
chn
i
qu
e
(
W
RHT)”
.
It
bas
ed
on
watch
dog
an
d
Delhi
C
on
c
ept.
It
giv
es
in
f
orm
at
ion
ab
out
t
he
pac
ket
drop
and
the
delay
pe
r e
ac
h
hops
a
nd
us
ed
for
the
f
ull
pha
se
r
ou
te
proces
s
in
wireless
se
ns
or
netw
ork.
Her
e
t
he
a
utho
rs
bu
il
d
up
m
eth
od
w
hich
is
use
d
for
worm
h
ole
dete
ct
ion
in
e
ve
ry
sens
or
de
vices
with
lo
w
c
os
t
s.
WH
RT
is
a
n
exte
ns
io
n
ve
rsion
A
O
DV
r
ou
ti
ng
protoc
ol.
The
pro
posed
m
et
ho
d
is
t
o
al
lo
w
f
or
cal
culat
in
g
t
he
w
or
m
ho
le
presence
pro
ba
bili
ty
(W
PP
)
f
or
a
pat
h
in
a
dd
it
io
n
to
hop
co
unt
in
form
ation
in
the
so
urce
no
de
over
the
se
nsor
ne
tworks
.
Durin
g
t
he
route
dis
cov
e
ry
process
,
pe
r
ho
p
tim
e
delay
prob
a
bili
ty
(TD
P
H)
a
nd
tim
e
d
el
ay
pr
oba
bili
ty
(TDPP)
is
cal
c
ulate
d
f
or
d
et
e
ct
ing
worm
ho
le
at
ta
cks.
I
n
the
ne
xt
par
t
of
the
WH
RT
,
an
othe
r
pa
ram
et
er
is
cal
culat
e
d,
w
hich
is
cal
le
d
per
ho
p
pack
et
los
s
pro
bab
il
it
y
(P
LPP
).
T
he
val
ues
of
PLPP
a
nd
TD
PP
are
us
ed
for
decisi
on
m
aking
,whethe
r
a
pa
th
P
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol. 2, N
o.
1, M
arch 2
021
:
33
–
42
40
is
aff
ect
e
d
by
worm
ho
le
at
ta
cks
or
not.
S
o
that
the
r
ou
ti
ng
pr
oto
c
ol
A
ODV
is
ta
king
co
rr
ect
way
f
or
the
transm
issi
on
over
the
se
nsor
s
netw
orks.
We
prese
nted
se
ver
al
w
or
m
ho
le
s
at
ta
ck
s
in
WSN.Fi
na
ll
y,
by
evaluati
ng
the
po
sit
ive
a
nd
ne
gative
asp
ect
s
of
al
l
existi
ng
te
chn
i
qu
e
s,
ti
ll
date
open
resea
rch
c
halle
nges
stud
ie
d
are
require
d
f
or
de
te
ct
ion
worm
ho
le
at
ta
cks.
I
n
Ta
bles
1,
t
he
m
os
t
im
po
rtant
de
te
ct
ion
m
et
hods
a
nd
requirem
ents are
el
ab
or
at
e
d
i
n
se
quentia
ll
y wit
h resp
ect
t
o y
ear.
Tables
1.
The
m
os
t im
po
rtant d
et
ect
io
n
m
eth
ods
an
d req
ui
rem
ents are
el
aborated
in
se
qu
entia
ll
y
with
resp
ect
t
o y
ear
Res
earc
h
er
Year
Metho
d
Too
ls
Proto
co
l
Req
u
ire
m
en
ts/C
o
m
m
en
ta
r
y
H.
Lu
,
D.
Evan
s
[
2
7
]
2003
Direction
al
An
ten
n
a
-
Direction
al
n
eig
h
b
o
r
d
iscovery
p
roto
co
l
Direction
al anten
n
as o
n
eac
h
node
with
GPS
Y.
C. Hu
an
d
D.
B.
Jh
an
so
n
[
2
8
]
2003
Pack
et leash
es
an
d
end
-
to
-
en
d
NS2
TI
K p
roto
co
l
GPS Co
o
rdin
ato
r
a
n
d
L
o
o
sely
Sy
n
ch
ron
ized
cloc
k
.
L.
lazo
s,
R. Poo
v
en
d
ra
m
[
2
9
]
2004
Localizatio
n
-
-
Bas
ed
on
locatio
n
awar
e
‘
g
u
ard
n
o
d
es’(L
AGNs
)
,
n
o
t
ap
p
licab
le to M
ANET
W
.
Wan
g
an
d
B.
Bh
argav
a
[
3
0
]
2004
Netwo
rk
v
isu
alizatio
n
-
-
Cen
tralized c
o
n
trol, see
m
s p
ro
m
isin
g
,
wo
rks
bas
ed
on
den
se n
etwo
rks
,
m
o
b
ility
is
n
o
t stu
d
ied
Issa Kh
alil,
Saurab
h
Bag
ch
i,
Ness
B.
S
h
rof
f
[
3
1
]
2005
LI
T
E
W
ORP
NS2
Key
m
an
ag
e
m
en
t
p
roto
co
l
Ap
p
licab
le on
ly
in static n
etwo
rks
,
A.
Baru
ch
,
R.
Cu
r
m
o
l
a,
C. Nita
-
Ro
taru,
D
.
Ho
l
m
er
,
H
.
Ru
b
en
s
[
3
2
]
2005
Ti
m
e
of
f
lig
h
t
NS2
ODSBR
Hardwar
e enab
lin
g
one
-
b
it
m
ess
ag
es
an
d
i
m
m
ed
i
ate rep
ly
w
ith
o
u
t CPU inv
o
lv
e
m
en
t
N.
So
n
g
,
L
.
Qian
,
X.
Li
.
[
3
3
]
2005
Statistical
Ap
p
roach
es
NS2
MR and
DSR
W
o
rks
on
ly
wi
th
m
u
ltip
ath
on
de
m
a
n
d
p
roto
co
l
H.S.
Ch
iu
and
K
.
Lui
[
3
5
]
2006
Delp
h
i
NS2
AODV
No
t con
sid
ered
K.B.
Ras
m
u
ss
en
a
n
d
S.
Cap
k
u
n
,
[36
]
2007
Rad
io
Fin
g
erprint
in
g
-
-
Fin
g
erprint
in
g
Devices is n
eeded
.
Kh
in
Sand
ar
W
in
.
[
3
7
]
2008
DAW
NS2
DSR
,
L
F
an
al
y
sis
Delay Par
a
m
e
ter
S.
Ch
o
i,
D.
Ki
m
,
D.
L
ee,
J. Ju
n
g
[
4
1
]
2008
W
AP
CBR
DSR
Maxi
m
u
m
tr
an
s
m
i
ss
io
n
dis
tan
ce is
calculated
H.
Vu
,
A.
Kulk
arni, K.
Sarac, N.
M
itt
al
[
4
3
]
2008
W
OR
ME
ROS
-
-
Ti
m
e
sy
n
ch
ron
izat
io
n
is r
eq
u
ired.
Top
o
lo
g
ical chan
g
e is no
t con
sid
ered
M.
S.
Sank
aran,
S.
Po
d
d
ar,
P.
Das
,
[
4
4
]
2009
SAW
-
AODV
No
t con
sid
ered
H.
Ch
en
,
W
.
L
o
u
,
X.
Su
n
,
an
d
Z
.
W
an
g
[
4
8
]
2010
Secu
re
lo
calization
NS2
Co
n
f
lictin
g
the
set
-
b
ased
resistan
ce
lo
calization
,
Distrib
u
ted
detecti
o
n
sy
ste
m
Gu
p
ta S,
Ka
r
S
,
Dh
ar
m
a
raja
[
5
0
]
2011
W
HO
P
NS2
W
HO
P,
AODV
No
t r
eq
u
ired any
hard
su
p
p
o
rt
an
d
clock
synch
ron
izatio
n
C.
P.
v
an
d
an
a
,
A
.F.
S.
Dev
raj
[
5
5
]
2013
ML
D
W
NS2
AODV
No
t r
eq
u
ired any
s
p
ecialized h
ard su
p
p
o
rt
an
d
clock
sy
n
ch
ron
izatio
n
R. sin
g
h
,
J, sin
g
h
,
Rav
in
d
ar
sin
g
h
[61]
2016
W
RHT
NS2
AODV
It
b
ased
on
the co
m
b
in
atio
n
of
two
tech
n
iq
u
es, i.
e.
W
a
tch
d
o
g
and
Delph
i.
5.
CONCL
US
I
O
N
Worm
ho
le
at
ta
cks
i
n
WSNs
a
re
one
of
the
brutal
at
ta
cks
t
ha
t
can
be
im
pl
e
m
ented
easi
ly
in
se
nsor
s
netw
orks.
I
n
t
hi
s
pa
per
num
ber
s
of
m
et
ho
dolog
ie
s
is
disc
usse
d
for
detect
ing
w
or
m
ho
le
a
tt
ack.
H
oweve
r
,
it
is
no
t l
ess in
f
or
m
at
ion
. T
he
refore
w
e b
el
ie
ve
th
at
the an
al
ysi
s o
n
t
his p
a
per
is
h
el
pin
g us
f
or
dev
el
op
i
ng
t
he
n
e
w
m
et
ho
d
t
o dete
ct
w
orm
ho
le
.
REFERE
NCE
S
[
1
]
M.
Tiwa
ri,
K.
V
eer
Ar
y
a,
R.
Ch
o
u
d
h
a
ri,
K.
Sid
h
arth
Ch
o
u
d
h
ary
.
“
Desig
n
in
g
Intru
sio
n
Detecti
o
n
to
Detect
Black
h
o
le
an
d
Selectiv
e
Fo
rwar
d
in
g
Attack
in
W
SN
b
ased
o
n
l
o
cal
Inf
o
r
m
atio
n
.”
Fou
rth
Inter
n
a
ti
o
n
a
l
Co
n
feren
ce
o
n
Co
mp
u
ter
S
cien
ces
a
n
d
C
o
n
verg
en
ce
Info
rma
tio
n
Technolo
g
y
” 20
0
9
.
[
2
]
E.
N
a
m
Hu
h
and
T
.
Ho
n
g
Hai
.
“
Ligh
tweig
h
t I
n
trus
io
n
Detec
tio
n
f
o
r
W
ir
ele
ss
Sens
o
r
Netwo
rk
s” iTech
.20
1
1
.
[
3
]
J.
Du
,
J.
Li,
“
A
St
u
d
y
o
f
Secu
rity
R
o
u
tin
g
Proto
co
l
f
o
r
Wi
reless
Sen
so
r
Netwo
rk
.”
Inter
n
a
tio
n
a
l
Co
n
fere
n
ce
o
n
Ins
tru
men
ta
tio
n
,
Mea
su
remen
t,
Co
mp
u
ter, Co
mmu
n
ica
tio
n
and Co
n
tro
l
,
2011
.
[
4
]
F.
Bao
,
I.
Ra
y
Ch
en
,
M.
Jeo
n
g
Ch
an
g
,
an
d
J.
-
Hee
Ch
o
.
“
Hierarch
ical
T
rus
t
M
an
ag
e
m
en
t
f
o
r
W
i
reless
Sen
so
r
Net
wo
rks
an
d
i
ts
Ap
p
licati
o
n
s
to
Tr
u
st
-
Bas
ed
Ro
u
tin
g
an
d
Intru
sio
n
Det
ectio
n
.”
I
EE
E
Tra
n
sa
ctio
n
s
o
n
Netw
o
rk
a
n
d
S
ervice
Ma
n
a
g
ement
,
v
o
l.
9
,
n
o
.
2
,
p
p
.
1
6
9
-
1
8
3
,
Ju
n
e 20
1
2
.
[
5
]
M.
A.
Ras
sa
m
,
M.
A.
Maa
rof
an
d
A.
Zainal
.
“
A
Su
rvey
o
f
Intru
sion
Detection
Sch
e
m
es
in
W
ireless
Sen
so
r
Net
wo
rks
.
”
A
m
e
rican
Jo
u
rnal o
f
App
lied
Sciences,
v
o
l.
9
,
no
.
1
0
,
p
p
.
1
6
3
6
-
1
6
5
2
,
Jan
u
ary
20
1
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Com
pu
t. Sci.
I
nf. Tec
hnol.
Su
r
vey
of wor
mhole
attack i
n
wi
rel
ess se
nsor
networks
…
(
Um
ash
ankar
Ghu
ga
r
)
41
[
6
]
W
.
Rib
eiro
Pir
es
J´
u
n
io
r,
T.
H
.
d
e
P
au
la
Fig
u
eiredo
H.
C
h
i
W
o
n
g
,
A.
A.F
.
Lou
reir
o
.
“
Malicio
u
s
No
d
e
Detection
in
W
ir
eless
Sen
so
r
Netwo
rks
.”
P
rocee
d
in
g
s o
f
the
1
8
th
Inter
n
a
tio
n
a
l P
a
ra
l
lel an
d
Distrib
u
ted
P
ro
cess
in
g
Symp
o
siu
m
(
IP
DPS’0
4
),
IE
E
E
,
2004
.
[
7
]
V.
K.
Jatav
,
M.
Tr
i
p
ath
i,
M
S
G
au
r
an
d
V.
L
ax
m
i
.
“
W
i
re
less
Sen
so
r
Netwo
rks
:
Atta
ck
Mod
els
an
d
D
etectio
n
.”
I
ACS
IT
Ho
n
g
Ko
n
g
Co
n
feren
ces IP
CS
IT,
IA
CS
IT
P
res
s
,
Sin
g
ap
o
re
,
v
o
l
. 3
,
2
0
1
2
.
A.
Pau
la
R.
d
a
Silv
a,
M.
H.
T.
Mar
tin
s
Br
u
n
o
,
P.S.
Ro
ch
a,
A.
A.F.
Lou
reir
o
,
L
.
B.
Ru
iz,
H.
Ch
i
W
o
n
g
.
“
Decentralize
d
in
trus
io
n
d
etectio
n
in
wireless
sen
so
r
n
etwo
rks
.”
Procee
d
in
g
s
o
f
th
e
1
st
ACM
Inter
n
a
tio
n
a
l
W
o
rks
h
o
p
o
n
Qu
a
l
ity
o
f
S
ervice
a
n
d
S
ecur
it
y
in
W
ireles
s
a
n
d
Mo
b
ile Networks
,
2
0
0
5
.
[
8
]
G.
Sa
ravan
an
,
P
.
R
.
Patil
,
M.
R.
Ku
m
a
r
.
“
Su
rvey
o
n
Intru
sio
n
Detection
Sy
s
te
m
in
Hete
rog
en
eo
u
s
W
SN
Usin
g
M
u
ltip
ath
Ro
u
tin
g
.”
IOSR Jou
rn
a
l of Co
mp
u
ter Eng
in
eeri
n
g
(
I
OS
R
-
JC
E)
,
v
o
l.
1
6
,
no.
2
,
p
p
.
2
6
-
3
1
,
2
0
1
4
.
[
9
]
U
Gh
u
g
ar,
J
Prad
h
an
,
M
Bis
wal
.
“
A
No
v
el
Intru
sio
n
Detection
Sy
ste
m
f
o
r
Detecting
Black
Ho
le
Attack
s
in
W
ireless
Sen
so
r
Netwo
rk u
sin
g
AO
DV Pr
o
to
co
l
.
”
IJC
S
N Inter
n
a
tio
n
a
l J
o
u
rn
a
l of Co
m
p
u
te
r S
cien
ce an
d
Netwo
rk
,
v
o
l.
5
,
no.
4
,
Aug
u
st 2
0
1
6
.
[
1
0
]
Y.
Sab
ri
,
N.
El
Ka
m
oun
.
“
GRPW
-
M
u
S
-
s: A Secure
En
h
an
ced Tr
u
st Awar
e Ro
u
tin
g
again
st
W
o
r
m
h
o
le
Attack
s
in W
irel
ess
Sens
o
r
Netwo
rks
.
”
Co
mm
u
n
ica
tio
n
s o
n
A
p
p
l
ied
E
lectro
n
ics
(
C
AE
)
,
pp.
2394
–
4
7
1
4
,
2
0
1
6
.
[
1
1
]
Sin
g
h
,
M.
,
D
as,
R.
Sah
o
o
.
“
A
Su
rvey
o
f
Dif
f
erent
Tec
h
n
iq
u
es
f
o
r
Detection
o
f
W
o
r
m
h
o
le
Attack
in
W
i
reless
Sen
so
r
Netwo
rk
.”
Inter
n
a
tio
n
a
l Jo
u
r
n
a
l of S
cien
tific and Eng
in
eerin
g
R
esea
rch
.
v
o
l.
3
,
n
o
.
1
0
,
2
0
1
2
.
A.
Bh
arathid
asan
and
V.
A
.
S.
Pon
d
u
ru
. “
Sen
so
r
Netwo
rks
: An Ov
erview
.”
Te
ch
n
ical Rep
o
rt,
De
p
t.
o
f
Co
m
p
u
ter
S
c
ien
ce,
Un
iv
ersit
y
of
Calif
o
rnia at Da
v
is
,
2
0
0
2
.
[
1
2
]
M.
Tub
aish
at and
S.
M
ad
ria
.
“
Sen
so
r
netwo
rks
: an o
v
er
v
iew
.”
I
EE
E
Poten
tia
ls
,
v
o
l.
2
2
,
p
p
.
2
0
-
23,
2
0
0
3
.
[
1
3
]
Priy
a
Maida
m
w
ar
an
d
Nek
ita
Ch
av
h
an
.
“
A
Su
rvey
o
n
Secu
rity
Issu
es
to
d
etect
wo
r
m
h
o
le
Attack
in
W
ireless
Sen
so
r
n
etwo
rk
.
”
Inter
n
a
tio
n
a
l Jo
u
r
n
a
l on
A
d
Hoc Netw
o
rkin
g
Sys
tems (IJANS
)
,
v
o
l.
2
,
no
.
4
,
Octo
b
er
2
0
1
2
.
[
1
4
]
R.H.
Kh
o
k
h
ar,
M
d
.
A.
Ng
ad
i
,
S.
M
an
d
a.
“
A
Rev
iew
o
f
Cu
rr
en
t
Ro
u
tin
g
Attack
s
in
Mob
ile
Ad
Ho
c
Net
wo
rks
.”
Inter
n
a
tio
n
a
l
Jo
u
rn
a
l of Co
mp
u
t
er S
cien
ce an
d
Secu
rity
,
v
o
l.
2
,
n
o
.
3
,
p
p
.
1
8
-
2
9
,
2
0
0
8
.
[
1
5
]
Jali,
Ka
m
arularif
in
Ab
d
,
Ja
m
alu
l
-
Lai
l
Ab
Manan
Zaid
Ah
m
ad
.
“
Mitigatio
n
o
f
Black
Ho
le
Attack
s
f
o
r
Ao
d
v
Ro
u
tin
g
Prot
o
co
l.”
Internatio
n
al Jo
u
rnal o
f
New
Co
m
p
u
t
er
A
rchitectu
res
an
d
T
h
eir
Ap
p
licatio
n
s, vo
l.
1
,
p
p
.
3
3
6
-
3
4
3
,
2
0
1
1
.
[
1
6
]
Ku
m
ar
,
V.
“
Si
m
u
l
atio
n
and
Co
m
p
ari
so
n
of
AODV
an
d
DSR Ro
u
tin
g
Pr
o
to
co
ls in
M
ANE
Ts
.
”
Ma
ster
Thes
is
,
2
0
0
9
.
[
1
7
]
S.
Lu,
L.
Li,
K.
La
m
an
d
L.
Jia.
“
SAODV:
A
MAN
ET
Ro
u
tin
g
Proto
co
l
th
at
can
W
ith
stan
d
Black
Ho
le
Attack
.
”
2
0
0
9
Inter
n
a
tio
n
a
l
Co
n
feren
ce on
Co
mp
u
ta
tio
n
a
l
Intellig
en
ce an
d
Secu
rity
,
Beijin
g
,
p
p
.
4
2
1
-
4
2
5
,
2
0
0
9
.
[
1
8
]
Manel
Gu
er
rer
o
Z
ap
ata,”
Secu
r
e
Ad
h
o
c
On
-
De
m
an
d
Distan
ce
Vector
R
o
u
tin
g
.
”
in
ACM
S
IGMOB
ILE
Mo
b
ile
Co
mp
u
tin
g
a
n
d
Co
mmu
n
ica
tio
n
s R
eview
,
v
o
l.
6
,
no
.
3
,
p
p
.
1
0
6
-
1
0
7
,
Ju
n
e
20
0
2
.
[
1
9
]
Ch
u
n
h
u
i
Zhu
,
M
y
u
n
g
J.
Lee,
Ta
rek
Saad
aw
.
“
RTT
-
Ba
sed
Op
ti
m
al
W
aiti
n
g
Ti
m
e
f
o
r
Bes
t
Ro
u
te
Selectio
n
In
Ad
Ho
c
Ro
u
tin
g
Proto
co
ls
.”
IE
EE
Milita
ry Co
mmu
n
ica
tio
n
Co
n
feren
ce
,
vol
.
2,
pp1054
-
1
0
5
9
,
Oct
2
0
0
3
.
[
2
0
]
K.U.
R
Kh
an
,
A.
V.
Red
d
y
,
R.U
.
Z
a
m
an
,
K.A
Red
d
y
,
T.
S
Harsha
.
“
An
Ef
ficient
DSDV
Ro
u
t
in
g
Proto
co
l
for
W
irelessMo
b
ile
Ad
Ho
c
Net
wo
rks
an
d
its
Perf
o
r
m
an
c
e
Co
m
p
a
riso
n
.”
S
ec
o
n
d
UKS
IM
Eur
o
p
ea
n
S
ymp
o
siu
m
o
n
Co
mp
u
ter
Mo
d
eli
n
g
a
n
d
S
imu
la
tio
n
,
Ind
ia,
p
p
.
5
0
6
-
5
1
1
,
20
0
8
.
[
2
1
]
B
Kan
n
h
av
o
n
g
,
H
Nak
ay
a
m
a,
Y
Ne
m
o
to
,
N
Kato
.
“
A
Su
rvey
of
Ro
u
tin
g
Attack
s
In
Mob
ile
Ad
Ho
c
Netwo
r
k
s
.
”
IE
E
E
W
ireles
s
Co
mmu
n
ica
tio
n
,
v
o
l.
14,
n
o
.
5
,
Octo
b
er
2
0
0
7
.
[
2
2
]
Ab
h
ish
ek
Ver
m
a
a
n
d
Neh
a
Bh
ardwaj
.
“
A
Rev
iew
o
n
Ro
u
tin
g
In
f
o
r
m
atio
n
Proto
co
l
(RIP)
an
d
Op
en
Sh
o
rtest
Path
First
(O
SPF)
Ro
u
tin
g
Pr
o
to
co
l
.
”
Inter
n
a
tio
n
a
l
Jo
u
r
n
a
l of Futu
re Gener
a
tio
n
Co
mmu
n
ica
tio
n
an
d
Netwo
rki
n
g
.
vol
.
9
,
p
p
.
1
6
1
-
170,
2
0
1
6
.
[
2
3
]
h
ttp
s://en
.wikip
ed
i
a.org
/wik
i/Op
en
_
Sh
o
rtest_
Path
_
First.
[
2
4
]
E
K
af
f
ash
i,
Ai
Mou
sav
i,
H
Rah
v
ard
.
“
A
n
ew
attack
o
n
l
in
k
-
state
d
atab
ase
in
o
p
en
sh
o
rtest
p
ath
f
irst
rou
tin
g
p
roto
co
l
.
”
Jo
u
rn
a
l
o
f
Electrica
l an
d
E
le
ctro
n
ic E
n
g
in
eering
,
v
o
l.
3
,
n
o
.
2
-
1
,
p
p
.
39
-
45
,
2
0
1
5
.
[
2
5
]
L.
Hu
,
D.
Evan
s
.
“
Usin
g
Direction
al
An
ten
n
as
to
Preve
n
t
W
o
r
m
h
o
le
At
ta
ck
s
.
”
1
4
Proce
ed
in
g
s
o
f
th
e
1
1
th
Net
wo
rk
an
d
Distrib
u
t
e
d
Sy
ste
m
Secu
rity S
y
m
p
o
siu
m
,
pp
.
2
0
0
3
.
[
2
6
]
Y.
C.
Hu
,
A.
Per
ri
g
,
an
d
D.
B.
Jo
h
n
so
n
.
“
Pack
et
Leashes
:
A
Def
en
se
ag
ai
n
st
W
o
r
m
h
o
le
At
t
acks
in
W
ire
less
N
etwo
rks
.
”
I
n
P
roc
.,
o
f I
EE
E
-
INF
OCO
M
,
v
o
l.3, pp
.
1
9
7
6
-
1
9
8
6
,
2
0
0
3
.
[
2
7
]
Lazo
s,
Lou
k
as
&
Po
o
v
en
d
ran,
Rad
h
a.
“
SeRLo
c:
Secu
r
e
Ran
g
e
-
Ind
ep
en
d
en
t
Localizatio
n
f
o
r
W
ir
eless
Sen
so
r
Netwo
rks
.
”
AC
M
Tra
n
sa
ctio
n
s o
n
Se
n
so
r Netw
o
rks
,
v
o
l
.
1
,
p
p
.
21
-
30
,
2
0
0
4
.
[
2
8
]
[
3
0
]
W
.
W
an
g
,
B.
Bh
argav
a.
“
Visu
ali
zatio
n
o
f
wo
r
m
h
o
les
in
sen
so
r
n
etwo
r
k
s
.”
Proceed
in
g
s
o
f
th
e
2
0
0
4
ACM
worksh
o
p
o
n
W
ireles
s
S
ecur
ity
,
p
p
.
5
1
-
6
0
,
2
0
0
4
.
[
2
9
]
Kh
alil,
S
au
r
ab
h
Ba
g
ch
i
an
d
N.
B.
Sh
r
o
ff
.
“
LI
TE
W
ORP:
a
lig
h
tweig
h
t
co
u
n
ter
m
e
asu
re
f
o
r
th
e
wo
r
m
h
o
le
attack
in
m
u
ltih
o
p
wi
reles
s
n
etwo
rks
.”
2
0
0
5
I
n
tern
a
tio
n
a
l Co
n
fe
ren
ce on
Depen
d
a
b
le Sys
tems a
n
d
N
etwo
rks
(
DS
N'
0
5
)
,
Yo
k
o
h
a
m
a,
Jap
an
,
p
p
.
6
1
2
-
6
2
1
,
2
0
0
5
.
[
3
0
]
B.
Awe
rbu
ch
,
R.
Cu
r
t
m
o
l
a,
D.
Ho
l
m
e
r
,
H
.
Ru
b
en
s
an
d
C.
Nita
-
Ro
taru.
“
On
th
e
Su
rviv
ab
ility
o
f
Ro
u
tin
g
Proto
co
ls
in
Ad
Ho
c
W
ire
less
N
etwo
rk
s.”
F
irs
t
Inter
n
a
ti
o
n
a
l
Co
n
feren
ce
o
n
S
ecur
ity
a
n
d
Pri
va
cy
fo
r
Emerg
in
g
Area
s
in
Co
mmu
n
ica
tio
n
s
Netw
o
r
k
s
(
S
ECUR
ECOMM
'
0
5
)
,
Ath
en
s, pp
.
3
2
7
-
3
3
8
,
2
0
0
5
.
[
3
1
]
N.
So
n
g
,
L.
Qian
,
X.
Li.
“
W
o
r
m
h
o
le
Attack
s
Det
ecti
o
n
in
W
ireless
A
d
Ho
c
Netwo
rks
:
A
Statistical
An
al
y
sis
Ap
p
roach
”.
I
n
Proceed
in
g
s o
f
the
19
th
I
EE
E
Inter
n
a
tio
n
a
l P
a
ra
llel an
d
Distrib
u
ted
P
ro
cess
in
g
Symp
o
siu
m
,
p
p
.
8
-
1
5
,
2
0
0
5
.
[
3
2
]
M.
A.
Go
rlatov
a,
P
.
C.
Maso
n
,
M
.
W
an
g
,
L.
L
a
m
o
n
t,
R.
Li
scan
o
.
“
Detecting
W
o
r
m
h
o
le
Attack
s
in
Mob
ile
Ad
Ho
c
Netwo
rks
th
rou
g
h
Proto
co
l Breakin
g
and
Packet Ti
m
in
g
Analysis
.”
In
IE
E
E
Milita
ry Co
mmu
n
ica
tio
n
s Confer
en
ce
,
p
p
.
1
-
7
,
2
0
0
6
.
[
3
3
]
H.S.
Ch
iu
an
d
K.
L
u
i.
“
DelPHI:
W
o
r
m
h
o
le
Det
ectio
n
Mechan
is
m
f
o
r
Ad
Ho
c
W
ireless
Net
wo
rks
”.
In
Pro
cee
d
in
g
s
o
f
Inter
n
a
ti
o
n
a
l
S
ymp
o
siu
m on
W
ireles
s Perva
sive Co
mp
u
tin
g
,
p
p
.
6
-
1
1
,
2
0
0
6
.
[
3
4
]
K.B.
Ras
m
u
ss
en
a
n
d
S.
Cap
k
u
n
.
“
I
m
p
li
catio
n
s
o
f
radio
f
in
g
erprint
in
g
o
n
th
e
secu
rity
o
f
s
en
so
r
n
etwo
rks
.
”
Third
Inter
n
atio
n
a
l
Co
n
f
erence on
Security
and
P
rivacy
i
n
Co
m
m
u
n
icatio
n
Networks
and
the
W
o
rks
h
o
p
s, pp
.
3
3
1
-
3
4
0
,
Sep
.
2
0
0
7
.
[
3
5
]
Kh
in
San
d
ar
W
in
.
“
An
aly
sis
o
f
Detec
tin
g
W
o
r
m
h
o
le
Att
ack
in
W
ireless
Ne
two
rks
”,
W
o
rld
A
c
ad
e
m
y
o
f
Scien
ce,
Eng
in
eering
an
d
Techn
o
lo
g
y
,
v
o
l.
4
8
,
p
p
.
4
2
2
-
4
2
8
,
2
0
0
8
.
[
3
6
]
F.
N
ait
-
Ab
d
ess
ela
m
,
B.
Ben
sao
u
an
d
T.
Tal
eb
.
“
Detec
tin
g
an
d
av
o
id
in
g
wo
r
m
h
o
le
at
tack
s
in
wirel
ess
ad
h
o
c
n
etwo
rks
.”
in
IE
E
E
Co
m
m
u
n
icatio
n
s Magazin
e,
v
o
l.
4
6
,
no
.
4
,
p
p
.
1
2
7
-
1
3
3
,
Ap
ril
2
0
0
8
.
[
3
7
]
S.
Özd
e
m
i
r,
M.
M
eg
h
d
ad
i,
an
d
Ý.
G
ü
ler.
“
A
ti
m
e
an
d
trus
t
-
b
ased
wo
r
m
h
o
le
d
etectio
n
alg
o
r
ith
m
f
o
r
wir
eless
sen
so
r
n
et
-
wo
rks
.”
(
m
an
u
scrip
t in
Tur
k
ish
),
in 3
rd Inf
o
rm
a
tio
n
Securit
y
an
d
Cry
p
to
lo
g
y
.
[
3
8
]
Kron
tiris,
T.
Giann
etso
s, and
T
.
Di
m
it
riou
.
“
Lidea: A
dis
tribu
ted
ligh
tweig
h
t intru
sio
n
detectio
n
ar
ch
itectu
re
f
o
r
s
en
so
r
n
etwo
rks
.
”
in
S
ECURE
C
OM
M
’08
:
Fou
rth
Inte
rn
a
tio
n
a
l
Co
n
fere
n
ce
o
n
S
ecur
ity
a
n
d
Priva
cy
fo
r
Co
m
mu
n
ica
tio
n
Netw
o
rks
,
Istan
b
u
l,
Turk
ey
,
Sep
te
m
b
er
2
2
-
2
5
20
0
8
.
[
3
9
]
S.
Ch
o
i,
D.
Ki
m
,
D.
Lee
,
J.
Ju
n
g
.
“W
AP:
W
o
r
m
h
o
le
Attack
Preven
tio
n
Alg
o
rith
m
in
Mob
ile
Ad
Ho
c
Netwo
r
k
s
.
”
In
Inter
n
a
tio
n
a
l
Co
n
feren
ce on
Senso
r Netw
o
rks
,
Ub
iq
u
ito
u
s a
n
d
Tru
st
wo
rth
y Co
mp
u
tin
g
,
p
p
.
3
4
3
-
3
4
8
,
2
0
0
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2722
-
3221
Com
pu
t. Sci.
I
nf. Tec
hnol.
,
V
ol. 2, N
o.
1, M
arch 2
021
:
33
–
42
42
[
4
0
]
S
Özd
e
m
i
r,
M
M
e
g
h
d
ad
i,
Ý
Gü
ler.
“
Detectio
n
alg
o
rithm
f
o
r
wirel
ess
sen
so
r
n
etwo
rks
”
in
3
rd
Inf
o
r
m
atio
n
Sec
u
rity
an
d
Cr
y
p
to
lo
g
y
Co
n
f
erence (
ISC’0
8
).
[
4
1
]
H.
Vu
,
A.
Ku
lk
arni
,
K.
Sar
ac,
N.
Mitt
al.
“
W
ORM
EROS
:
A
New
Fra
m
ewo
rk
f
o
r
Def
en
d
i
n
g
a
g
ain
st
W
o
r
m
h
o
le
Attack
s
o
n
W
ireles
s
Ad
Ho
c
Netwo
rks
.
”
In
Pro
ceedin
g
s
o
f
Inter
n
a
tio
n
a
l
Conf
ern
ce
o
n
W
ireles
s
Algo
rith
ms
S
yste
ms
a
n
d
App
lica
ti
o
n
s
,
LNCS
5
2
5
8
,
p
p
.
491
-
5
0
2
,
2
0
0
8
.
[
4
2
]
M.
S.
San
k
aran,
S.
Po
d
d
ar,
P.S.
D
as, S.
Selv
ak
u
m
ar
.
“
A
No
v
el
Secu
rity
M
o
d
el
SaW
:
Secu
ri
t
y
ag
ain
st
W
o
r
m
h
o
le
attack
in
W
i
reles
s
Sen
so
r
Netwo
rks
.
”
I
n
Pr
o
ceedin
g
s o
f
I
n
ternati
o
n
al Co
n
fer
en
ce on
PDCN
,
2
0
0
9
.
[
4
3
]
B.
T
riki
,
S.
Rek
h
i
s
,
an
d
N.
Bo
u
d
riga.
“
Dig
ital
Inv
estig
atio
n
o
f
W
o
r
m
h
o
le
Att
acks
in
W
i
re
less
Sen
so
rNet
wo
r
k
s
.
”
Eigh
th
IE
E
E
Internatio
n
al Sy
m
p
o
siu
m
on
Networ
k
Co
m
p
u
tin
g
and
App
licatio
n
s, 20
0
9
.
[
4
4
]
H.
Ch
en
,
W
.
Lou
,
an
d
Z.
W
an
g
.
“
Co
n
f
lictin
g
-
set
-
b
ased
wo
r
m
-
h
o
le
at
tack
resistan
t
lo
caliz
ati
o
n
in
wireless
sen
s
o
r
n
etwo
rks
.
"
Bo
o
k
Ch
ap
ter
Lecture
N
o
tes in
Co
mp
u
ter S
cien
ce−Ub
iq
u
ito
u
s
I
n
tellig
en
ce an
d
Co
mp
u
tin
g
,
v
o
l.
5
5
8
5
/2
0
0
9
,
p
p
.
2
9
6
−3
0
9
,
2
0
0
9
.
[
4
5
]
D.Bar
m
an
Ro
y
,
R.
Ch
ak
i
,
N.
Ch
ak
i
,”
A
New
Clu
ster
-
b
as
ed
W
o
r
m
h
o
le
Intru
sio
n
Detection
alg
o
rith
m
f
o
r
Mob
ile
Ad
h
o
c
Netwo
rks
”,
Inter
n
a
tio
n
a
l Jo
u
r
n
a
l of Netw
o
rk S
ecur
ity
&
I
ts App
lica
tio
n
s (
I
JNS
A)
,
v
o
l
1
,
no
1
,
Ap
ril
2
0
0
9
.
[
4
6
]
H.
Ch
en
,
W
.
Lou
,
X.
Su
n
,
an
d
Z.
W
an
g
.
“
A
secu
re
lo
c
alizatio
n
ap
p
roach
ag
ain
st
w
o
r
m
h
o
le
attacks
u
sin
g
d
istan
ce
co
n
sis
ten
cy
.”
EUR
ASIP
Jo
u
rn
a
l
o
n
W
ireles
s
Co
m
mu
n
ica
tio
n
a
n
d
N
et
-
wo
rkin
g
-
S
p
ecia
l
Issu
e
o
n
W
irele
ss
Netw
o
rk
Algo
ri
th
ms,
S
ystems
a
n
d
App
lica
tio
n
s
,
v
o
l.
1,
p
p
.22
−3
2
,
2
0
1
0
.
[
4
7
]
Prasan
n
ajit
B,
Ven
k
atesh
,
An
u
p
a
m
a
S
,
Vin
d
h
y
k
u
m
a
ri
K,
S
u
b
h
ash
i
n
i
S
R,
Vin
ith
a
G
.
“
An
ap
p
roach
to
wards
Detection
o
f
W
o
r
m
h
o
le
Attack
in Sen
so
r
N
etwo
rks
.
”
First
I
n
t
ern
a
tio
n
a
l Co
n
fere
n
ce on
I
n
teg
ra
ted
I
n
tellig
en
t Co
mp
u
ti
n
g
(
IC
II
C
)
,
p
p
.28
3
-
2
8
9
,
2
0
1
0
.
[
4
8
]
S.
Gu
p
ta,
S.
Ka
r
an
d
S.
Dh
a
r
m
a
raj
a.
“
W
H
OP:
W
o
r
m
h
o
le
attack
d
ete
ct
io
n
p
roto
co
l
u
sin
g
h
o
u
n
d
p
acket.”
2
0
1
1
Inter
n
a
tio
n
a
l
Co
n
feren
ce on
I
n
n
o
va
tio
n
s in
I
n
fo
rm
a
tio
n
Techn
o
lo
g
y
,
Ab
u
Dhab
i,
p
p
.
2
2
6
-
2
3
1
,
2
0
1
1
.
[
4
9
]
Z.
Zhao
,
B
.
W
ei,
X.
Do
n
g
,
L.
Yao
an
d
F.
Gao
.
“
Det
ectin
g
W
o
r
m
h
o
le
Attack
s
in
W
i
rele
ss
Sen
so
r
Netwo
r
k
s
with
Statistical
An
aly
sis
.”
2
0
1
0
WAS
E
Inter
n
a
tio
n
a
l Co
n
feren
ce on
I
n
fo
rma
tio
n
E
n
g
in
eer
in
g
,
Beid
aih
e,
Heb
ei,
p
p
.
2
5
1
-
2
5
4
,
2
0
1
0
.
[
5
0
]
Bin
tu
Kad
h
iwala
a
n
d
Harsh
Sh
ah
.
“
Exp
lo
ration
o
f
W
o
r
m
h
o
le
Attack
w
ith
i
ts
Detection
an
d
P
reven
tio
n
Techn
iq
u
es
in
W
irel
ess
Ad
-
h
o
c
Netwo
rks
.”
IJ
CA
Pro
ceedin
g
s
o
n
Inter
n
a
tio
n
a
l
Confer
en
ce
o
n
Rece
n
t
Tren
d
s
in
Informa
t
io
n
Tech
n
o
l
o
g
y
a
n
d
Co
mp
u
ter
S
cien
ce
2
0
1
2
I
CRTITCS
,
n
o
.
9
,
p
p
.
1
-
6
,
Feb
ru
ary 2
0
1
3
.
[
5
1
]
K.
P
atel
,
T
.
Mano
ranjith
a
m
.
“
De
tect
io
n
o
f
wo
r
m
h
o
le
attack
in
w
ireless
sen
so
r
n
etwo
rk
.”
I
n
tern
a
tio
n
a
l
Jo
u
rnal
o
f
Eng
in
eerin
g
Resear
ch
&
T
echn
o
lo
g
y (
I
JER
T)
,
2
0
1
3
.
[
5
2
]
D.S
Ku
sh
wah
a,
A.
Kh
are,
J.
L.
Ran
a
.
“
I
m
p
rov
ed
T
rus
tf
u
l
Ro
u
tin
g
Proto
co
l
to
Detect
W
o
r
m
h
o
le
Attack
in
MAN
ET
.
”
Inter
n
a
tio
n
a
l
Jo
u
rn
a
l of Co
mp
u
t
er App
lica
tio
n
s
,
v
o
l.
6
2
,
n
o
.
7,
p
p
.
2
1
-
25,
Jan
u
ary
20
1
3
.
[
5
3
]
C.P.
V
an
d
an
a,
A.
F.S.
D
ev
raj.
“
M
L
DW
-
a
Multila
y
e
re
d
Detection
m
ech
an
is
m
f
o
r
W
o
r
m
h
o
le
attack
in
AOD
V
b
ased
MAN
ET
.
”
Inter
n
a
tio
n
a
l Jo
u
r
n
a
l of S
ecur
ity,
Pri
va
cy an
d
Tru
st Ma
n
a
g
ement
,
vol
.
2
,
n
o
.
3,
p
p
.
2
9
-
41,
Ju
n
e
2
0
1
3
.
[
5
4
]
S.
Go
y
al
,
H.
Ro
h
il,
“
Secu
ring
MANE
T
ag
ain
st
W
o
r
m
h
o
le
A
ttack
u
sin
g
Neig
h
b
o
r
No
d
e
An
aly
sis
.
”
Inter
n
a
tio
n
a
l
Jo
u
r
n
a
l
o
f
Co
mp
u
ter App
lica
t
io
n
s
,
v
o
l.
8
1
,
no.
1
8
,
p
p
.
4
4
-
4
8
,
No
v
em
b
er
2
0
1
3
.
[
5
5
]
Y.
Sin
g
h
,
A
.
Kh
a
tk
ar,
P
.
Ran
i,
“
W
o
r
m
h
o
le
Attack
A
v
o
id
an
ce
Techn
iq
u
e
in
Mob
ile
Ad
h
o
c
Netwo
rks
”,
T
h
ird
Internatio
n
al
Co
n
f
erence on
Ad
v
an
ced C
o
m
p
u
tin
g
&
Co
m
m
u
n
icatio
n
T
echn
o
lo
g
ies, Ro
h
tak
,
6
-
7
April 20
1
3
.
[
5
6
]
P.
N
ay
ak
,
A
.
S
a
h
ay
,
Y.
Pan
d
ey
.
“
Detection
an
d
Preven
tio
n
o
f
W
o
r
m
h
o
le
Atta
ck
s
i
n
MA
NE
Ts
u
sin
g
Dete
ctio
n
Pack
et
.
”
Inter
n
a
tio
n
a
l Jo
u
r
n
a
l of S
cien
tific
&
Eng
in
eerin
g
R
esea
rch
,
v
o
l.
4
,
no.
6
,
J
u
n
e
-
2
0
1
3
.
[
5
7
]
N.
Ch
o
u
d
h
ary
,
S.
Ag
rawa
l
.
“
An
aly
si
s
o
f
W
o
r
m
-
Ho
le
Attack
in
MAN
E
T
u
sin
g
AODV
R
o
u
tin
g
Proto
co
l
.”
S
S
RG
Inter
n
a
tio
n
a
l
Jo
u
rn
a
l of Electro
n
ic
s a
n
d
Co
mmu
n
i
ca
tio
n
E
n
g
in
eering (
S
S
RG
-
IJ
ECE
)
,
v
o
l.
1
,
n
o
.
1
0
,
Dec
2
0
1
4
.
A.
Patel,
N.
Pat
el
a
n
d
R.
Pa
tel.
“
Def
en
d
in
g
ag
ain
st
W
o
r
m
h
o
le
Att
ack
i
n
MA
NET
.”
2
0
1
5
Fifth
Inter
n
a
tio
n
a
l
Co
n
feren
ce
o
n
Co
mmu
n
ica
tio
n
Sy
stems
an
d
Netwo
rk Tech
n
o
lo
g
ies
,
Gw
alio
r,
pp
.
6
7
4
-
678
,
2
0
1
5
.
[
5
8
]
R.
S
in
g
h
,
J
,
sin
g
h
,
Rav
in
d
ar
sin
g
h
.
“
W
HRT
:
A
H
y
b
rid
Techn
iq
u
e
f
o
r
Detectin
g
of
W
o
r
m
h
o
l
e
Attack
in
W
ir
eles
s
Sen
so
r
Netwo
rks
.
”
Mo
b
ile I
n
fo
rma
tio
n
Sys
tems
,
Hin
d
a
wi
p
u
b
lish
in
g
Co
rp
o
a
rtio
n
,
v
o
l.
2
0
1
6
,
n
o
.
1
3
,
p
p
.
1
-
1
3
,
2
0
1
3
.
Evaluation Warning : The document was created with Spire.PDF for Python.