Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
5, N
o
. 5
,
O
c
tob
e
r
201
5, p
p
. 1
102
~111
0
I
S
SN
: 208
8-8
7
0
8
1
102
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
An Empirical Critique of
On-Demand Routing P
r
ot
ocols
against Rushing Att
a
ck i
n
MANET
S. Ash
o
k Ku
mar
*
,
E.
Sure
s
h
Babu
*, C.
Nagaraju
**, A. Ped
a
G
o
pi
*
*
Dep
a
rtment of
Computer Scien
ce
and Engineering,
K L
Univ
ersity
**Departm
ent
of
Com
puter S
c
i
e
n
ce
and
E
ngineering, Yogi Vemana University
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Apr 2, 2015
Rev
i
sed
Jun
14,
201
5
Accepted
Jun 29, 2015
Over the last decade, researcher
s ha
d proposed
numerous
mo
bile ad h
o
c
routing proto
c
ols for which
are o
p
erate
in
an on-d
e
mand way
,
as standard
on-
demand routing
protocols
such as AODV,
DSR
and TO
RA, etc., have been
shown to often
have faster reaction tim
e and lower overhead th
an proact
ive
protocols. However,
the open
n
ess of
the routing environm
ent and th
e
absence of centralized s
y
stem and infrastructu
r
e make them exposed to
s
ecurit
y
att
acks
in larg
e ex
tent
. In par
ticu
l
ar
,
one s
u
ch kind
of att
acks
is
rushing atta
ck,
which is m
o
stl
y
h
a
rd to det
ect due to th
ei
r inherit
e
d
properties,
that alters th
e n
e
twork sta
tist
i
cs
radic
a
ll
y.
In
thi
s
paper,
w
e
m
odeled a rus
h
ing att
ack whi
c
h is
a powerfu
l att
ack
that
e
xploits
the
weaknesses of th
e secur
e
rou
ting
protoc
ols. Moreover,
to know th
e weakn
e
ss
and strength
of
these proto
c
ols,
it is ne
ces
s
a
r
y
t
o
tes
t
th
eir p
e
rf
orm
a
nce i
n
hostile env
i
ron
m
ents. Subsequentl
y
,
th
e perfor
m
ance is m
easured with the
various metrics
,
some ot them ar
e av
er
age
throug
hput, packet deliver
y
ratio,
averag
e end-
to
-end del
a
y
an
d etc
.
,
to
co
m
p
are and
ev
alua
te th
eir
performance.
Keyword:
A
ODV
DSR
M
ANET
s
R
out
i
n
g Pr
ot
oc
ol
s
Ru
sh
i
n
g Attack
Copyright ©
201
5 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
s:
C. Na
gara
ju,
Depa
rt
m
e
nt
of
C
o
m
put
er Sci
e
nce a
n
d
E
ngi
n
eeri
n
g,
Yo
gi Vem
a
na
Uni
v
ersity
,
V
e
m
a
n
a
p
u
r
a
m
,
K
a
d
a
p
a
,
A
ndhr
a Pr
ad
esh 5162
16
, Ind
i
a
Em
a
il:
cnrcse@yahoo.com
A. Peda
G
o
pi
,
Depa
rt
m
e
nt
of
C
o
m
put
er Sci
e
nce a
n
d
E
ngi
n
eeri
n
g,
K L Un
iv
ersity,
V
a
dd
esw
a
r
a
m
,
Gu
n
t
u
r
, Andh
ra Pr
ad
esh
5
2200
1,
I
n
d
i
a
Em
ail: gopiare
p
alli2@gm
a
il.com
1.
INTRODUCTION
:
W
i
rel
e
ss c
o
m
m
uni
cat
i
on er
a has
de
vel
o
p
e
d
great
l
y
o
v
e
r t
h
e l
a
st
fe
w de
cades
,
w
h
i
c
h
br
o
u
g
h
t
essen
tial ch
ang
e
s to
t
h
e fiel
d
of teleco
mmu
n
i
cation
a
nd
net
w
or
ki
n
g
.
P
a
rt
i
c
ul
arl
y
, t
h
e
wi
rel
e
ss a
n
d
m
obi
le
com
m
uni
cat
i
ons
have
bec
o
m
e
an i
n
t
e
gr
al
part
o
f
ou
r dai
l
y
l
i
f
e.
Ho
we
ver
,
exc
h
an
gi
n
g
t
h
e
r
e
qui
red
in
fo
rm
atio
n
wi
th
fu
ll-fledg
e
d wireless co
nnectiv
ity o
v
e
r
wireless
n
e
twork
s
is still an
o
p
e
n
p
r
ob
lem
.
In
t
h
is
reg
a
rd, th
ere is a n
eed
to
estab
lish
th
e in
stan
t co
mm
u
n
i
catio
n
to
ex
chan
g
e
th
e informatio
n
b
e
tween
th
e
wi
rel
e
ss
de
vi
ces. M
o
bi
l
e
A
d
h
o
c
net
w
or
k
[1]
i
s
one
s
u
ch
ki
n
d
of
net
w
o
r
k
t
h
at
pr
o
v
i
d
es i
n
st
ant
com
m
uni
cat
i
on
bet
w
ee
n t
h
e
m
obi
l
e
users
whi
c
h f
o
rm
s a dy
nam
i
cal
ly chan
ge
d
net
w
o
r
k
wi
t
h
o
u
t
any
fi
xe
d
n
e
t
w
o
r
k
in
frastru
ct
u
r
e. Mo
re im
p
o
r
tantly, th
ese n
e
two
r
k
s
ch
an
ges d
y
n
am
i
cal
l
y
and u
n
p
r
e
d
i
c
t
a
bl
y
t
h
at
m
a
kes rout
i
n
g
as a chal
l
e
ngi
n
g
i
ssue. T
o
ad
dress t
h
i
s
i
ssue
,
researc
h
er
s h
a
ve p
r
o
p
o
sed
n
u
m
e
rous r
o
ut
i
n
g pr
ot
ocol
s D
S
R
[
2]
,
AO
D
V
[
3
]
,
DS
DV
,
OLSR
,
T
O
R
A
,
et
c.
H
o
weve
r,
t
h
e
o
p
e
nne
ss
of
t
h
e
r
o
ut
i
n
g
e
nvi
r
o
n
m
ent
an
d t
h
e a
b
se
nce
of
fi
xe
d i
n
frast
ru
ct
ure m
a
ke t
h
e
m
exposed
t
o
s
ecuri
t
y
at
t
acks
i
n
l
a
r
g
e e
x
t
e
nt
. I
n
part
i
c
ul
ar,
m
o
st
of t
h
e at
t
acks
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
110
2
–
11
10
1
103
suc
h
as Black
hole attack, worm
hole att
ack, rus
h
ing attacks
,
which are pa
rticul
arly h
a
rd
to
d
e
tect du
e to th
eir
in
h
e
rited
p
r
op
erties, th
at alters th
e
n
e
two
r
k
st
atistics rad
i
cally.
This pa
per com
p
ares the perform
ance analysis a
nd evaluation of three
reactive (T
ORA, DSR a
nd
AODV)
p
r
o
t
oco
l
s in
ho
stile en
v
i
ron
m
en
t. Sp
ecifically,
th
e ru
sh
ing
attack
will b
e
evalu
a
ted
ag
ai
n
s
t
th
ese
reactive prot
oc
ols to validate
the pe
rform
ance of the M
A
NET th
ro
ugh
si
m
u
la
tio
n
s
. Mor
e
ov
er, ru
sh
ing attack
doe
s
not
c
o
nsu
m
e t
h
e l
o
t
of
r
e
so
urces
o
r
c
o
st
t
o
s
u
b
v
e
r
t
t
h
e
no
rm
al
beh
a
vi
o
u
r
o
f
t
h
e
n
e
t
w
o
r
k
.
Pa
rt
i
c
ul
arl
y
,
ru
sh
ing
attack
is o
n
e
of th
e den
i
al-of-serv
i
ce attack
s
th
at
ex
p
l
o
it th
e v
u
l
n
e
rab
ility ag
ai
n
s
t stan
d
a
rd
dyn
amic
ro
ut
i
n
g pr
ot
oc
ol
s. I
n
react
i
v
e
rout
i
ng
pr
ot
oc
ol
, gl
o
b
al
r
out
e di
sco
v
ery
p
r
oced
u
r
e i
s
i
n
i
t
i
at
ed by
t
h
e ori
g
i
n
at
o
r
th
at g
e
n
e
r
a
tes
a r
o
u
t
e r
e
q
u
e
st
(
RREQ
)
and
fo
rw
ard
s
th
e R
R
EQ
p
a
ck
ets to
th
e
n
e
igh
bour
no
d
e
s. Accord
ing
to
th
e prop
erty, t
h
e in
term
ed
iate n
o
d
e
s
will tak
e
th
e
first
ru
sh
ed
RREQ pack
et in
to
th
e con
s
id
eratio
n
and d
i
scard
th
e fo
ll
o
w
i
n
g
RREQ p
a
ck
ets. Th
e attack
er
will ex
p
l
o
it th
i
s
p
r
op
erty in
ev
ery d
y
n
a
m
i
c
rou
ting
pro
t
o
c
o
l
to
commence the
rus
h
ing attac
k
.
The rel
a
t
e
d w
o
rk
i
s
he
re
i
n
S
ect
i
on 2. Sect
i
o
n
3
di
sc
uss a
b
o
u
t
react
i
v
e r
out
i
n
g p
r
ot
oco
l
s.
Sect
i
o
n 4
bri
e
fl
y
ex
pl
ai
n
s
abo
u
t
r
u
s
h
i
n
g
at
t
ack and
Sec
t
i
on 5
d
eals wi
th the pe
rform
a
nce ev
al
uat
i
o
n o
f
t
h
ree o
n
-
d
em
and
ro
ut
i
n
g
pr
ot
oc
ol
s. S
ect
i
o
n
6
pr
o
v
i
d
es t
h
e si
m
u
l
a
t
i
on w
o
rk
.
I
n
t
h
e
en
d,
we
co
ncl
u
de i
n
se
ct
i
on
7.
2.
RELATED WORK
In
m
oder
n
t
i
m
es,
num
ero
u
s
r
e
sol
u
t
i
o
n a
r
e
p
r
o
p
o
sed
f
o
r
M
ANE
T
[4]
ro
ut
i
ng
p
r
o
p
o
sal
s
t
o
m
oderat
e
th
e troub
le of ro
u
ting
in
terruptio
n
.
Ei
t
h
e
r
sy
m
m
e
t
r
i
c
key
b
a
sed
pr
ot
oc
ol
s
(e.
g
., T
E
SL
A [
5
]
)
[
6
]
o
r
p
u
b
l
i
c
key
base
d di
gi
t
a
l
si
gnat
u
res
[7]
a
r
e use
d
t
o
di
st
i
n
g
u
i
s
h t
h
e g
e
nui
ne m
e
m
b
ers fr
om
t
h
e st
range
rs.
Su
bse
q
uent
l
y
network elem
e
n
ts re
ject to a
g
ree or
fo
rw
ar
d an
y un
au
t
h
or
i
zed
d
a
ta
p
a
ck
et. Co
nv
ersely, th
o
s
e cr
yp
tograp
h
i
c
co
un
term
easu
r
es can
no
t fu
l
l
y resp
ond
to th
e rou
ting
co
mm
o
t
io
n
ch
allen
g
e
. Th
e in
tru
s
i
o
n
d
e
t
ectio
n
tech
n
i
qu
es
p
r
op
o
s
ed
b
y
Brad
l
e
y et a
l
[8
] an
d Ch
eun
g
and
Lev
itt [9
] fo
r iden
tifyin
g
an
d
detectin
g
rou
t
ers th
at
forward
fak
e
rou
t
e up
d
a
te
messag
e
p
a
ckets. Here, we
illu
strate o
n
e
in
v
a
rian
t of no
d
e
p
e
rfo
rm
an
ce an
d
prese
n
t
a
scat
t
e
red
m
e
t
hod t
o
avoi
d t
h
e
h
o
st
s
t
h
at
ha
ve
bee
n
t
r
ap
pe
d i
n
fri
n
g
e t
h
at
i
n
v
a
ri
a
n
t
.
A m
u
lti-p
a
th
ap
pro
ach
st
u
d
i
ed
b
y
Pap
a
d
i
m
i
trato
s
and
Haas [10
]
is to
allev
i
ate th
e rou
t
e
co
mm
o
t
io
n
attacks. T
h
e message
pac
k
ets are enc
o
ded into
rem
oval co
des, t
h
e recei
ver is able to
re
cupe
rate the se
nde
r’s
in
fo
rm
atio
n
o
n
ly after g
e
ttin
g a th
resho
l
d
d
i
v
i
sion
of en
cod
i
ng
sym
b
o
l
s t
h
at h
a
v
e
b
e
en
carried
al
o
n
g
with
th
e
m
u
l
tip
le p
a
th
s.
The
ro
ut
i
n
g
pr
ot
oc
ol
o
f
Pe
rl
m
a
n’s Fl
oo
di
n
g
NPB
R
[
1
1]
f
o
r
wi
re
d c
o
nne
ct
i
on
net
w
or
ks
d
o
es
not
g
o
t
h
r
o
u
g
h
t
h
i
s
at
t
ack, as t
h
e proce
d
ure d
o
es
not
de
pen
d
o
n
t
h
e ge
nui
ne
pat
h
of t
h
e
o
v
er
fl
o
w
fo
r r
o
ut
i
ng;
sl
i
ght
l
y
, i
t
hav
e
need
of t
h
at
every
pac
k
et
b
e
swam
ped t
h
r
o
u
g
h
t
h
e net
w
or
k.
Awe
rb
uc
h
et
al
. [12]
p
r
o
pos
ed
an
an
alytical sch
e
m
e
an
d
m
u
lti-p
a
th
ev
al
u
a
tio
n
to
d
e
tect th
e m
a
l
i
cio
u
s
no
d
e
s. If a m
a
l
e
v
o
l
en
t no
d
e
can
not
d
i
scr
i
m
i
n
a
te th
e in
fo
r
m
atio
n
p
ack
ets w
ithout sn
oop
ing
p
i
gg
yb
ack
s
f
r
o
m
th
o
s
e w
ith, th
en
th
e or
ig
i
n
ator
can
recognize the
failure
ra
nge
on a
rout
e. Though, none
of
the allied
work accepts
our
restricted a
p
proach t
o
pr
ot
ect
t
h
e bes
t
pat
h
ex
p
o
se
d
by
t
h
e u
n
d
erl
y
i
ng r
o
ut
i
ng
pr
ot
oc
ol
. T
h
e ab
ove
p
r
o
p
o
sal
s
are n
o
t
su
p
p
o
r
t
e
d o
n
presc
r
i
b
e
d
m
odel
s
.
3.
ON-
D
EM
AND
(R
EA
C
T
IVE)
ROU
T
I
N
G
PROTOCOL
S IN MANETs
From
the last decade
,
m
a
ny routin
g protoc
ols ha
ve
bee
n
devel
ope
d for ad-hoc
net
w
orks by
kee
p
ing
th
e b
a
sic fu
n
c
t
i
o
n
a
lities o
f
rou
tin
g, wh
ich
su
ppo
rts th
e commu
n
i
catio
n
ov
er selected
p
a
th
s and
tran
sm
it th
e
packet
s
ove
r t
h
e r
out
e. T
h
i
s
pape
r at
t
e
m
p
t
s
t
o
un
der
s
tand the perform
a
nce issues
of some ex
istin
g
reactiv
e
protoc
ols
unde
r m
obility and va
rying
traffi
c conditions
[13]. Recently a
gr
eat attention showed towards t
h
e
on
-
d
em
and pr
ot
oc
ol
s f
o
r i
t
s
pot
e
n
t
i
a
l
t
o
wa
rds l
o
w
r
out
i
n
g o
v
e
r
hea
d
.
T
h
e o
n
-
dem
a
nd
ro
ut
i
n
g p
r
ot
o
c
ol
s are
so
urce-in
itiated
th
at d
i
scov
ers th
e rou
t
e as an
“as n
e
ed
ed
” b
a
sis in
stead
o
f
conv
en
tio
n
a
l p
r
o
activ
e
p
r
oto
c
o
l
s.
As t
h
ese
ro
ut
i
ng
pr
ot
oc
ol
s
effect
i
v
el
y
f
u
l
f
i
l
s
t
h
e ge
neral characte
r
istics of ad-hoc networks s
u
ch as
con
n
ect
i
v
i
t
y
agai
nst
dy
n
a
m
i
cal
l
y
changi
n
g
t
opol
ogy
, fi
t
s
t
h
e l
o
w ba
nd
wi
dt
h
of wi
rel
e
ss l
i
nks [
14]
,
qui
ck
con
v
e
r
ge
nce, e
t
c,. A
O
D
V
,
T
O
R
A
a
n
d DS
R
are fe
w o
f
t
h
e dy
nam
i
c (react
i
v
e) r
o
ut
i
n
g p
r
ot
oc
ol
s w
h
i
c
h a
r
e
expl
ai
ne
d b
r
i
e
f
l
y
i
n
t
h
e fol
l
o
w
i
ng su
bsect
i
o
ns
.
3.
1
A
d
-ho
c
On-Dema
nd D
i
st
ance Vec
t
or
(AODV)
AO
D
V
i
s
a dy
nam
i
c rout
i
n
g pr
ot
oc
ol
t
h
a
t
creat
es
an
d
main
tain
s th
e rou
t
es on
ly wh
en
th
ey are
neede
d
.
AO
D
V
i
s
com
posed of t
w
o ess
e
nt
i
a
l
funct
i
o
n
s
nam
e
l
y
rout
e di
scove
ry
p
r
oce
d
ure an
d
ro
ut
e
maintenance procedure.
Ro
ut
e di
scove
r
y:
G
l
ob
al Route d
i
scov
er
y pr
o
c
edu
r
e is in
i
tiated
wh
en
there is no
ro
u
t
e is av
ailab
l
e to
th
e
targ
et no
d
e
from th
e o
r
ig
in
at
o
r
. Th
e
r
efore t
h
e source broa
dcast the route
request pac
k
et
s (RREQ)
[15]
to all
th
e neigh
bou
r
n
o
d
e
s.
O
n
r
eceip
t of
RREQ
,
t
h
e
n
e
ighb
ou
r node
s create a
reverse
e
n
try to the
ori
g
inat
or. The
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
An E
m
pi
ri
cal
C
r
i
t
i
que of
O
n
-De
m
a
n
d
R
o
ut
i
n
g
Pr
ot
oc
ol
s
a
gai
nst
R
u
s
h
i
n
g
At
t
a
ck i
n
MA
N
E
T
(
C
. N
a
gar
aj
u)
1
104
interm
ediate host receivi
ng R
R
EQ pac
k
et wi
ll forward to
it
s neighbour a
n
d this
procedure will go
on
until the
req
u
est
arri
ve at
t
h
e t
a
rget
node
. O
n
ce t
h
e t
a
rget
no
de re
cei
ve t
h
e R
R
E
Q t
h
en i
t
resp
ons
e wi
t
h
ro
ut
e repl
y
packet
(RREP) over the
reve
rs
e lin
k toward
s th
e
o
r
i
g
in
ator.
Ro
ut
e m
a
i
n
t
e
n
ance
:
On
ce th
e
p
a
th
is est
a
b
lish
e
d
in
between
t
h
e
o
r
ig
in
ator an
d
t
h
e targ
et, the rou
t
e
main
ten
a
n
ce is in
trodu
ced to
v
e
rify t
h
e validity o
f
th
e
rou
t
e b
ecau
s
e th
e no
d
e
s are arb
itrarily
m
o
v
e
in
an
d ou
t
of t
h
e n
e
t
w
or
k
.
Thi
s
pr
ot
o
c
ol
m
a
kes use of
HELL
O m
e
ssages t
o
ens
u
re t
h
e ro
ut
es are a
c
t
i
v
e by
peri
o
d
i
cal
ly
br
oa
dcast
t
h
e m
e
ssages bet
w
een t
h
e n
odes
[1
6]
. If t
h
e n
o
d
e in
th
e n
e
two
r
k
do
es no
t get a data packet from
t
h
e i
n
t
e
rm
edi
a
te wi
t
h
i
n
t
h
e st
i
pul
at
ed t
i
m
e t
h
en t
h
e l
i
n
k bet
w
een i
n
t
e
rm
edi
a
t
e
node a
n
d i
t
s
el
f i
s
consi
d
e
r
ed a
s
br
o
k
en
. Thi
s
pr
ot
oc
ol
pa
rt
i
c
ul
arl
y
uses t
h
e l
o
cal
re
pai
r
m
echani
s
m
t
o
re
bui
l
d
t
h
e
ro
ut
e t
o
war
d
s t
h
e
d
e
stin
ation
.
3.
2
Dynamic
S
o
ur
ce Rou
t
ing (DSR)
The
Dy
nam
i
c So
urce R
o
ut
i
n
g (
D
SR
)
pr
ot
o
c
ol
i
s
an
ef
fi
ci
ent
an
d si
m
p
l
e
ro
ut
i
n
g
pr
ot
oc
ol
desi
g
n
ed
fo
r ex
pl
oi
t
i
n
m
u
lt
i
-
ho
p M
A
NETs
. Thi
s
pr
ot
oc
ol
l
e
t
t
h
e net
w
or
k t
o
be
ut
t
e
rl
y
sel
f-c
on
fi
g
u
ri
ng a
n
d
sel
f
-
o
r
g
a
n
i
zin
g
,
with
ou
t an
y cen
t
r
alised
ad
m
i
n
i
stratio
n or an
act
iv
e n
e
t
w
ork
infrastru
c
tu
re.
The key
el
em
ent
of D
S
R
i
s
the use o
f
so
ur
ce rout
i
ng i
n
whi
c
h t
h
e ori
g
i
n
at
or
kn
ow
s t
h
e com
p
l
e
t
e
sequ
en
ce or
d
e
r o
f
th
e nod
es i
n
th
e ro
u
t
e to
targ
et
n
o
d
e
and
ev
ery
d
a
ta
pack
et carries t
h
is inform
atio
n
in its
packet
heade
r
[17]. The global rout
e discovery and route
maintenance ar
e the two im
portant m
echanism
s
of
DSR
.
I
n
R
o
ut
e
Di
sco
v
e
r
y
,
ori
g
i
n
at
o
r
br
oa
dc
ast
s
a R
R
E
Q
p
acket
t
h
at
i
s
s
w
am
ped i
n
a
c
ont
rol
l
e
d
m
a
nner
al
l
t
h
e way
t
h
r
o
ug
h t
h
e
net
w
o
r
k
and
i
s
re
pl
i
e
d
by
a R
R
E
P
(
r
o
u
t
e
re
pl
y
)
fr
om
ei
t
h
er t
h
e t
a
r
g
et
no
de
o
r
a
n
y
ot
he
r
i
n
t
e
rm
edi
a
t
e
node t
h
at
rec
o
gni
ze a
pat
h
t
o
t
h
e t
a
r
g
et
no
de.
To
decr
ease t
h
e cost
of R
out
e
Di
scove
ry
p
r
o
c
ed
ure, ev
ery no
d
e
k
e
ep
s
a set of so
urce rou
t
es it
h
a
s
ov
erh
eard
or
learn
e
d
wh
ich
it in
sisten
tly u
tilizes to
li
mit th
e p
r
op
ag
atio
n and
frequ
en
cy
of RREQs.
The anothe
r ba
sic
m
echanis
m of DSR is Route Main
ten
a
nce in
wh
ich
there is n
o
con
c
ep
t o
f
Hell
o
messag
e
s w
h
i
c
h
ar
e r
e
sp
onsib
le
fo
r p
e
r
i
od
ical
upd
atin
g
o
f
ro
ut
es as
l
i
k
e i
n
A
O
D
V
pr
ot
oc
ol
.
It
i
s
t
h
e
r
e
spon
sib
l
e
f
o
r ev
er
y m
o
b
ile n
o
d
e
of
m
a
in
ta
in
in
g th
e ro
u
t
i
n
g pro
t
o
c
o
l
b
e
tw
een
t
h
e
no
de and
n
e
x
t
hop in
the
p
a
th
fro
m
th
e
o
r
i
g
in
ator to
t
h
e targ
et no
d
e
. In
case, if an
y rou
t
e fro
m
th
e sou
r
ce is
b
r
ok
en, orig
inato
r
is
in
fo
rm
ed
with
a ROUTE ER
ROR (RERR).
Then t
h
e ori
g
i
n
at
o
r
t
r
i
e
s t
o
m
a
ke use o
f
any
ot
he
r pat
h
t
o
t
a
rge
t
n
o
d
e
alread
y in
its ro
u
t
e co
llectio
n
o
r
can
ag
ain
in
itiate
ro
u
t
e d
i
scov
ery p
r
o
c
ed
ure ag
ai
n
to
d
i
scov
er a fresh
route. Route Discovery m
e
chanism
and Route Ma
intenance m
echanis
m
both can work com
p
letely
on
dem
a
nd
nat
u
re
an
d
DSR
d
o
es
n
o
t
re
q
u
i
r
e a
n
y
peri
od
pac
k
e
t
s at
any
l
e
vel
i
n
t
h
e
net
w
o
r
k.
3.
3
Tempor
ally Ordered
Rout
ing Algorithm
(TORA)
The Tem
poral
l
y
Orde
red R
o
ut
i
ng
Al
g
o
ri
t
h
m
[18]
(TOR
A) r
o
ut
i
ng p
r
o
t
ocol
i
s
an we
l
l
-
or
gani
ze
d,
scalab
le d
i
ssemin
a
ted
and
hig
h
l
y ad
ap
tiv
e ro
u
ting algo
rith
m
p
r
o
p
o
s
ed fo
r m
u
lti-h
o
p
and
h
i
gh
ly
vib
r
an
t
m
o
b
ile n
e
two
r
k
s
. TORA is a sen
d
e
r-i
n
itiate
d
d
y
n
a
m
i
c ro
u
tin
g
pro
t
o
c
o
l
th
at d
i
scov
ers nu
m
e
ro
u
s
rou
t
es fro
m
the ori
g
inator to the target
node. T
h
e m
a
in facet of TOR
A
is that th
e localization of the c
ont
rol m
e
ssages to a
sm
all set of hosts whe
n
the topol
ogy
of the network is about to change
.
To accom
p
lish this, the inform
ation
o
f
ro
u
ting
abou
t th
eir adj
acen
t
nod
es m
u
st b
e
m
a
in
tain
ed
b
y
all th
e
no
d
e
s in
t
h
e
n
e
two
r
k
.
The T
O
R
A
r
o
ut
i
ng
pr
ot
oc
ol
has t
h
ree
pri
m
ary
funct
i
o
n
s
such as R
o
u
t
e C
r
eat
i
on, R
out
e M
a
i
n
t
e
na
nce an
d
R
out
e E
r
as
ure
[1
9]
. T
h
e
fi
rst
fu
nct
i
o
n o
f
T
O
R
A
i
s
r
o
ut
e cre
a
t
i
on
pr
oce
d
u
r
e w
h
i
c
h i
s
res
p
onsi
b
l
e
f
o
r
sel
e
ct
i
ng
app
r
op
ri
at
e hei
ght
s
f
o
r
r
out
er
s an
d f
o
rm
i
ng a seq
u
en
ce o
f
l
i
nks
w
h
i
c
h a
r
e
di
rect
l
y
l
eadi
n
g t
o
t
h
e t
a
r
g
et
no
de
.
Th
e second
fun
c
tio
n o
f
TOR
A
is ro
u
t
e
m
a
i
n
tain
ing
wh
ich r
e
sp
ond
s
wh
en
n
e
two
r
k topo
log
y
alter
s
. Sin
ce
every
m
obi
l
e
n
ode
m
u
st
have
a pa
rt
i
c
ul
ar
hei
ght
,
any
m
obi
l
e
n
ode
wi
t
h
nu
l
l
hei
g
ht
i
s
c
o
n
s
i
d
ere
d
as
an
e
r
ased
n
o
d
e
. TORA
pro
t
o
c
o
l
h
a
s a distin
ctiv
e qu
ality o
f
m
a
in
ta
in
in
g nu
m
e
ro
u
s
path
s to th
e target n
o
d
e
so th
at
th
e
to
po
log
i
cal chan
g
e
s of
th
e
netw
or
k
do
no
t
r
e
qu
ir
e an
y
respo
n
s
e at all. TORA
resp
on
d
s
on
ly wh
en all th
e
paths
to t
h
e target node a
r
e lost.
4.
MO
DELIN
G
A RU
SHI
N
G ATTA
CK
R
u
shi
n
g
i
s
a z
e
ro
del
a
y
at
t
ack s
h
o
w
n i
n
Fi
g
u
re
1.
It
i
s
m
o
r
e
da
nge
r
ous
w
h
en
at
t
acker
ne
arby
s
o
urc
e
or
dest
i
n
at
i
o
n.
R
eact
i
v
e ro
ut
i
n
g
p
r
ot
ocol
s
ar
e m
o
re susc
ep
tib
le to
th
e ru
shin
g
attack b
ecau
s
e
wh
en
ev
er so
urce
node broadcast
the RREQ pac
k
ets, the
attacker or m
a
lic
ious
node receive
s
that and forwa
r
d without any
hop-
cou
n
t
u
pdat
e
a
nd
del
a
y
i
n
t
o
t
h
e net
w
or
k [
2
0]
.
When l
e
gi
t
i
m
a
t
e
nod
es re
cei
ve ori
g
i
n
al
R
R
E
Q pac
k
et
s, t
h
ey
are droppe
d
because it alrea
d
y receive
d
pa
cket from
adversary a
nd t
r
ea
t th
is as a duplicate packets. Thus,
attacker included in active
route and
di
sr
upt
s dat
a
f
o
r
w
ar
di
ng
. T
h
e at
t
ack
er
can
gain
high s
p
eed i
n
acc
ess of
req
u
est
by
sl
o
w
i
n
g
do
w
n
t
h
e
res
p
o
n
se
t
i
m
e
of
ot
her
n
ode
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
JECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
110
2
–
11
10
1
105
Fi
gu
re 1.
R
u
s
h
i
ng At
t
ack
The attacker c
a
n increas
e the
traffic
i
n
net
w
or
k by
kee
p
i
n
g
t
h
e net
w
o
r
k t
r
ansm
i
ssi
on qu
eues [
21]
f
u
l
l
of t
h
e
n
earb
y
n
o
d
e
s. Hen
ce
n
o
d
e
s
will resp
ond
to th
e req
u
e
st la
te d
u
e
to
h
e
av
y traffic. Th
e auth
en
tic n
o
d
e
s
will b
e
b
u
s
y
v
a
lid
ating req
u
e
st con
t
ain
i
ng
false au
t
h
en
ticatio
n
s
t
h
us slowing
d
o
wn
th
ei
r respon
se ab
ility.
5.
PERFO
R
MA
NCE
EV
AL
UATI
O
N
O
F
O
N
-
D
EM
AN
D
RO
UT
ING
PR
OT
OCOL
S
UN
DER
RU
SHI
N
G A
TTACK
5.
1
R
u
shing
At
ta
ck Aga
i
nst AOD
V
In AODV routing prot
ocol the orig
inat
or
node initiate a route disc
overy for the target node in the
network.
If the
intruder
forwa
r
ds t
h
e RREQs
and a
r
e the
first to arri
ve at each nei
g
hbour of the
destination,
then any pat
h
discovere
d
by
this global route discove
ry process will e
m
brace [22]
a hop through
the intruder.
Th
at is, if a n
e
ig
hbo
ur no
de of th
e d
e
stin
ation
no
d
e
re
ceive
s the rus
h
ed R
R
EQ pac
k
et
fro
m
th
e in
trud
er, th
en
it fo
rward
s
th
at RREQ, an
d
d
i
scard
s
an
y fu
rt
h
e
r leg
itim
a
t
e REQUESTs fro
m
th
e n
o
n
-
attack
in
g
n
o
d
e
s [23
]
.
As a co
nse
que
nce, t
h
e
ori
g
i
n
at
or i
s
u
n
abl
e
t
o
fi
n
d
a
n
y
fu
n
c
t
i
onal
ro
ut
es.
In
gene
ral
w
o
r
d
s, a
n
i
n
t
r
ude
r
no
de
th
at forwards
RREQs m
o
re h
a
stily th
an
v
a
lid
n
o
d
e
s, ca
n
b
o
o
s
t th
e
p
r
obab
ility o
f
th
at ro
u
t
es th
at in
corpo
r
ate
th
e in
trud
er
wi
ll b
e
d
i
scov
ered
rath
er th
an
t
h
e o
t
h
e
r
rou
t
es.
Fi
gu
re
2.
1
0
-
N
ode
s P
D
F
Vs
P
a
use Ti
m
e
Fi
gu
re
3.
3
0
-
N
ode
s P
D
F
Vs
P
a
use Ti
m
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
An E
m
pi
ri
cal
C
r
i
t
i
que of
O
n
-De
m
a
n
d
R
o
ut
i
n
g
Pr
ot
oc
ol
s
a
gai
nst
R
u
s
h
i
n
g
At
t
a
ck i
n
MA
N
E
T
(
C
. N
a
gar
aj
u)
1
106
Fi
gu
re
4.
5
0
-
N
ode
s P
D
F
Vs
P
a
use Ti
m
e
Fig
u
r
e
5
.
10
-No
d
e
s Th
rou
ghpu
t V
s
Pau
s
e
Time
Ho
we
ver t
h
e a
b
o
v
e c
o
n
v
e
r
sat
i
on
has
used
t
h
e case o
f
m
obi
l
e
host
s
i
n
t
h
e
net
w
or
k t
h
at
f
o
r
w
ar
d
ju
st
th
e
first
R
R
E
Q
from
any
gl
ob
al
rout
e di
sc
ov
ery
pr
ocess, t
h
e po
wer
f
ul
r
u
s
h
i
n
g at
t
ack can
al
so be used a
g
ai
nst
any
r
o
ut
i
n
g
p
r
ot
oc
ol
t
h
at
i
n
e
v
i
t
a
bl
y
f
o
r
w
ar
ds
an
y
part
i
c
ul
ar R
R
E
Q
f
o
r
e
ach
gl
o
b
al
r
o
ut
e di
sc
ove
ry
p
r
ocess.
5.
2
R
u
shing
At
ta
ck Aga
i
nst DSR
In
DSR, th
e i
n
ten
tio
n
o
f
th
e
attack
er nod
e is to
rush
t
h
e RREQ to
th
e
n
e
i
g
hbo
urs
o
f
th
e targ
et no
de
so that it first affects the route
discovery
mechanism
an
d later it can
misbehave the
route
m
a
inte
nanc
e
mechanism
,
routing
etc.
W
e
can disc
uss t
h
e bot
h m
ech
anism
s
affected by the rus
h
ing attack.
The
Route
di
sco
v
ery
m
e
chani
s
m
cont
ai
ns t
w
o segm
ent
s
cal
l
e
d rout
e
request (RReq) and rout
e r
e
ply (
RRly)
. Th
er
efore
t
h
e r
u
s
h
i
n
g at
t
ack ca
n t
a
ke a
c
t
i
v
e part
i
c
i
p
at
i
on i
n
t
w
o
pha
ses. I
n
t
h
e
gl
obal
ro
ut
e
di
sc
ove
ry
m
echani
s
m
,
a
malevolent int
e
rm
ediate node without
veri
fying t
h
e rout
e request in it
s cache,
forwa
r
ds t
h
e route reque
st
who
s
e aim is
t
o
d
i
rectly ru
sh th
e ro
u
t
e req
u
est to
ward
s the n
e
ig
hbo
ur o
f
th
e targ
et n
ode. During
th
e rou
t
e
reply
of the
de
stination, the
malicious node takes a
pl
ace as a
n
i
n
term
e
d
iary
node i
n
t
h
e
route acc
um
ula
tion
of t
h
e
gl
o
b
al
r
out
e
di
sc
ove
ry
pr
ocess
.
As a
con
s
eq
ue
nce, a
pat
h
co
nt
ai
ni
n
g
r
u
s
h
i
n
g
n
ode
i
s
creat
ed
bet
w
e
e
n
so
ur
ces t
o
d
e
st
in
atio
n
i
n
th
e netw
or
k.
W
h
en
a n
e
ighb
our
no
de
of t
h
e
dest
i
n
at
i
o
n
get
s
t
h
e
ru
she
d
R
R
E
Q
fr
om
th
e in
trud
er, it fo
rward
s
th
at
RREQ p
ack
et
.
W
h
en
th
e v
a
lid
requ
ests fro
m
o
t
h
e
r n
o
d
e
s ap
p
e
ar later at th
e
n
e
igh
bou
r
ho
sts, th
ey will b
e
d
i
scard
e
d.
DSR rou
tin
g pro
t
o
c
o
l
p
r
ov
id
es
a rou
t
e m
a
in
te
n
a
n
ce m
ech
an
i
s
m
in
whi
c
h i
t
can kno
w t
h
at
a
m
obi
l
e
no
de i
s
respo
n
si
bl
e for c
o
rr
ob
o
r
at
i
ng t
h
a
t
t
h
e packet
t
h
at
i
t
for
w
ar
d has
bee
n
receive
d by t
h
e ne
xt hop along t
h
e
legitimate path.
If no ack
nowledgment is receive
d after
forwa
r
ding t
h
e
dat
a
pac
k
et
a
p
a
rt
i
c
ul
ar m
o
re
num
ber
of t
i
m
es, t
h
i
s
pa
rt
i
c
u
l
ar n
ode
sen
d
a ro
ut
e er
r
o
r m
e
ssage
(R
ER
R
)
bac
k
to
th
e
orig
in
at
o
r
. Since th
e
malicio
u
s
no
de is th
e i
n
termed
iate n
o
d
e
, so
it cau
s
es t
h
e
p
o
werfu
l
Den
i
al of
Ser
v
i
ce at
t
ack
by
t
u
m
b
l
e
d t
h
i
s
r
out
e e
r
r
o
r (R
ER
R
)
m
e
ssage pac
k
et
.
5.
3
R
u
shing
At
ta
ck Aga
i
nst TOR
A
TORA pro
t
o
c
ol is p
r
o
p
o
s
ed
fo
r m
u
lti-h
o
p
an
d ex
tr
em
ely d
y
n
a
mic
m
o
b
ile n
e
two
r
k
s
and
is a send
er-
in
itiated
d
y
n
a
mic ro
u
ting
pro
t
o
c
o
l
. TORA
p
r
o
t
o
c
o
l
d
i
scov
er
s
nu
m
e
ro
u
s
rou
t
es fro
m
an
orig
i
n
ato
r
t
o
a targ
et
no
de.
T
h
e m
a
in fa
cet
o
f
T
O
R
A
p
r
ot
oc
ol
i
s
t
h
at
t
h
e c
o
ntrol m
e
ssages are restricted to
a sm
all set of
host
s
wh
en
t
h
e topo
lo
g
y
o
f
th
e
n
e
t
w
ork is ab
ou
t
to
ch
ang
e
.
TORA is also vu
l
n
erab
le to ru
shin
g
attack.
Due to
its
ope
n
n
ess
an
d l
ack
of
ce
nt
ral
aut
h
ori
t
y
,
rus
h
i
n
g
at
t
ack i
s
l
a
unc
he
d i
n
T
O
R
A
du
ri
n
g
r
o
u
t
e creat
i
o
n
pha
se b
y
forward
i
ng
QRY
m
e
ssag
e
qu
ick
l
y to
th
e targ
et nod
e, when
th
e d
e
stin
at
io
n
nod
e g
e
ts
th
is QRY m
e
s
s
ag
e it
will create rou
t
e th
rou
g
h
t
h
e i
n
tru
d
e
r
n
o
d
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
110
2
–
11
10
1
107
Fig
u
r
e
6
.
30
-No
d
e
s Th
rou
ghpu
t V
s
Pau
s
e
Time
Fig
u
r
e
7
.
50
-No
d
e
s Th
rou
ghpu
t V
s
Pau
s
e
Time
Fi
gu
re
8.
1
0
-
N
ode
s D
e
l
a
y
Vs
Pause
Ti
m
e
Fi
gu
re 9.
3
0
-
N
ode
sDel
ay
Vs Pause
Ti
m
e
Fig
u
r
e
10
. 50
-N
od
es D
e
lay V
s
Pau
s
e
Tim
e
6.
SIM
U
LATI
O
N
RESULTS
AN
D A
NAL
Y
S
IS
Th
e Network Si
m
u
lato
r 2
(Ns-2
)
too
l
is u
s
ed
f
o
r
per
f
o
r
m
i
ng t
h
e si
m
u
l
a
t
i
ons
u
n
d
er
t
h
e
U
b
u
n
t
u
ope
rat
i
o
n sy
st
em
. Duri
n
g
t
h
e sim
u
l
a
t
i
on unde
r NS
2 t
o
ol
, every
n
ode i
n
t
h
e net
w
or
k
m
oves dy
na
m
i
cal
ly
towa
rds t
h
e target node.
Once
the target
node is reache
d
, t
h
e
m
obile host ta
kes a
particul
ar pe
riod
of ti
me (in
seco
nds
) f
o
r
r
e
st
and sel
ect
a new ra
n
dom
t
a
rget
n
ode
is after th
at Pause ti
m
e
. Du
ring
th
e sim
u
lati
o
n
, th
is
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
An E
m
pi
ri
cal
C
r
i
t
i
que of
O
n
-De
m
a
n
d
R
o
ut
i
n
g
Pr
ot
oc
ol
s
a
gai
nst
R
u
s
h
i
n
g
At
t
a
ck i
n
MA
N
E
T
(
C
. N
a
gar
aj
u)
1
108
p
r
o
c
ed
ure
rep
e
ats so
t
h
at contin
u
o
u
s
top
o
l
og
ical ch
an
g
e
s
will o
ccur i
n
t
h
e
n
e
two
r
k
.
Differen
t
scen
ari
o
s
for
di
ffe
re
nt
n
u
m
b
er
of m
obi
l
e
ho
st
s an
d
di
ffe
re
nt
Pa
use t
i
m
es
are
gene
rat
e
d
.
The
res
u
l
t
s
f
o
r
si
m
u
l
a
ti
on a
r
e
gi
ve
n
i
n
t
h
e a
b
o
v
e
su
bsect
i
o
n i
n
t
h
e
g
r
a
phs
. G
r
a
phs
dem
onst
r
at
e t
h
e
com
p
ari
s
on
of
t
h
e t
h
ree
re
act
i
v
e r
out
i
n
g
p
r
ot
ocol
s
by
va
ry
i
n
g
num
ber
o
f
h
o
st
s
on
t
h
e
ba
si
s
of
t
h
e
per
f
o
r
m
a
nce m
e
t
r
i
c
s w
h
i
c
h a
r
e m
e
nt
i
oned
be
l
o
w.
6.
1
Perfor
mance Metrics
The pe
rf
orm
a
nce m
e
t
r
i
c
s whi
c
h are gi
ven
bel
o
w
are conside
r
ed for the evaluation
of MANET
pr
ot
oc
ol
s:
1)
Packet
Del
i
v
e
r
y
Fract
i
on:
Th
e rat
i
o
of n
u
m
ber of dat
a
packet
s del
i
ve
red t
o
t
h
e t
a
rg
et
node t
o
t
h
o
s
e
pr
o
duce
d
by
t
h
e o
r
i
g
i
n
at
or.
2)
End
to
End
Delay: Th
is represen
ts th
e tim
e
tak
e
n
for a
d
a
ta p
ack
et to mo
v
e
fro
m
th
e orig
in
at
o
r
t
o
th
e
t
a
rget
no
de.
3)
Th
rou
ghp
u
t: Th
is m
e
trics represen
ts
the a
v
e
r
age
num
b
er of
bits arri
ved
per
seco
n
d
at
dest
i
n
at
i
o
n a
n
d
measured in
bps.
We
ha
ve
obse
r
ved
t
h
ree
react
i
v
e r
o
ut
i
n
g
p
r
o
t
ocol
s as
al
rea
d
y
s
u
m
m
a
ri
zed
Dy
nam
i
c So
urce
R
o
ut
i
n
g
(DSR
)
pr
ot
oc
o
l
, Tem
porary
Or
dere
d R
out
i
n
g
Al
go
ri
t
h
m
(TOR
A
)
pr
ot
oc
ol
an
d
A
d
h
o
c
On
-Dem
and
D
i
st
ance
Vect
o
r
R
o
ut
i
n
g (
A
OD
V
)
p
r
o
t
ocol
.
We
ha
v
e
m
a
ke use o
f
sim
u
l
a
t
i
ons t
h
at
are
gi
ve
n
b
e
l
o
w t
o
e
v
al
ua
t
e
t
h
e
resu
lt .NAM ed
ito
r wh
ich is
a n
e
twork an
i
m
ato
r
to
illu
st
rate th
e an
im
a
t
e
d
represen
tatio
n
o
f
th
e t
h
ree dyn
amic
ro
ut
i
n
g p
r
ot
oc
ol
s T
O
R
A
,
A
O
D
V
a
n
d
D
S
R
,
t
h
ei
r
ro
ut
i
n
g
pat
h
s a
n
d t
h
ei
r pe
rf
orm
a
nces an
d
NS
2 n
e
t
w
o
r
k
si
m
u
lato
r. Mo
reo
v
e
r,
X-g
r
aph is u
s
ed
to
represen
t th
e
p
a
cket d
e
liv
er
y r
a
tio
, thr
oug
hpu
t, an
d
av
g. end
-
t
o
-
e
nd
del
a
y
g
r
ap
hi
cal
l
y
for
t
h
e t
h
ree
dy
nam
i
c ro
ut
i
n
g
p
r
ot
oc
ol
s a
n
d
t
h
e
r
ef
ore
co
m
p
ari
ng t
h
em
.
Tabl
e 1. Si
m
u
lat
i
on
Pa
ram
e
t
e
rs used
i
n
NS
-
2
NS-2 Para
m
e
t
ers
Si
m
u
lation Ti
m
e
Nu
m
b
er
of Nodes
Si
m
u
lation Are
a
Routing Pr
otocols
T
r
a
ffic
Pause Ti
m
e
Packet Size
M
ovem
e
nt M
odel
500 (
s
)
25,
50,
75,
1
00,
12
5,
150
1000 x 1
000m
AODV,
DSR
and TORA
CBR(Constant Bit Rate)
10 (m
s)
512 by
tes
Rando
m
Wa
y
Poin
t
Fr
o
m
th
e Figur
e 2,
3
& 4 it is
m
o
n
ito
r
e
d
t
h
at in
itially th
e PDF is
v
e
r
y
h
i
gh
in th
e case o
f
AODV
whe
n
com
p
are
d
wi
t
h
t
h
e ot
he
r t
w
o p
r
ot
oc
ol
s but
i
t
reduces
si
gni
fi
ca
nt
l
y
i
f
t
h
ere i
s
i
n
crease i
n
t
h
e num
ber of
si
m
u
latin
g
nod
es.
Wh
ile i
n
th
e case
o
f
DSR sim
u
latio
n
th
e
PDF is
h
a
v
i
n
g
h
i
g
h
i
n
fi
rst situ
ation
b
u
t
it
decrease
s
at the starting of the second
scen
ario
if th
e nu
mb
er
o
f
sim
u
la
ting
hosts increa
ses. PDF in the case
of T
O
RA is also increa
sed
by the
increase in the num
b
er
of
si
m
u
latin
g
nod
es. Th
e PDF o
f
A
O
D
V
, D
S
R an
d
TOR
A
r
out
i
n
g
pr
ot
ocol
s
u
n
d
e
r r
u
s
h
i
n
g at
t
a
ck s
h
ows
t
h
at
t
h
e
packet
del
i
very
rat
i
o
i
n
t
h
ese r
out
i
n
g
p
r
o
t
ocol
s
ado
p
t
i
n
g si
m
i
lar
pat
t
e
rns
as i
n
creasi
n
g
t
h
e
n
u
m
b
er o
f
no
de
s d
u
e t
o
on
-de
m
and
nat
u
re
of
t
h
ese
pr
ot
oc
ol
s.
In
Figu
re 5, 6
& 7 th
rou
ghp
uts illu
strate th
e lo
ss
rate and
reflect th
e accuracy and
co
m
p
leten
e
ss of
t
h
e dy
nam
i
c rout
i
ng
pr
ot
o
c
ol
.
From
t
h
e abo
v
e g
r
ap
hs i
t
i
s
very clear that
the decr
ease in th
e th
ro
ugh
put with
th
e in
crease in
n
o
d
e
s m
o
b
ility
. As th
e p
a
ck
et
d
r
o
p
at su
ch
a h
eav
y lo
ad
traffic is
m
u
ch
h
i
g
h
. TOR
A
proto
c
o
l
sh
ows b
e
tter
perfo
r
m
a
n
ce at
th
e h
i
g
h
m
o
b
ility b
u
t
in
fu
rther situ
ation
s
it
sho
w
s a l
o
wer throug
hpu
t.
TORA
sh
ow
s a h
i
gh
th
ro
ugh
pu
t w
h
en
co
m
p
ared t
o
AODV and DSR. Avera
g
e
t
h
ro
ug
h
put
of t
h
e t
h
ree r
out
i
n
g
pr
ot
oc
ol
s i
s
de
crease i
n
t
h
e
p
r
esence
of
r
u
s
h
i
n
g
at
t
ack
but
D
S
R
has
t
h
e l
e
ss
t
h
r
o
ug
h
put
t
h
a
n
AO
D
V
.
Fi
gu
re
8,
9
&
10
sh
o
w
s t
h
e a
v
era
g
e e
n
d-t
o
-
e
nd
g
r
ap
hs
an
d
fr
om
t
hose
gr
aph
s
we
ha
ve s
een t
h
at
t
h
e
avera
g
e pac
k
et
delay increase
s
with
th
e raise in
th
e nu
m
b
er o
f
sim
u
latin
g
m
o
b
ile h
o
s
ts
wh
ich
are
waitin
g
in
the interface
queue
while the
dynam
i
c r
outing
protoc
ols
make an atte
mpt to
discove
r
the path to the
target
no
de.
DSR
an
d A
O
D
V
sh
o
w
s t
h
e p
o
o
r d
e
l
a
y
feat
ures as t
h
ei
r pat
h
s
i
n
t
h
e net
w
or
k are us
ual
l
y
not
t
h
e
sho
r
t
e
st
. E
v
e
n
i
f
t
h
e
fi
rst
gl
o
b
a
l
ro
ut
e
di
sco
v
e
ry
p
r
oce
ss
di
scove
rs t
h
e s
h
o
r
t
e
st
pat
h
,
t
h
at
pat
h
m
a
y
not
r
e
m
a
i
n
th
e sho
r
test over a ti
m
e
p
h
a
se d
u
e
to
m
o
b
ility o
f
th
e h
o
s
t. TORA pro
t
o
c
o
l
to
o
h
a
s th
e worst d
e
lay featu
r
es
because of the
distance i
n
formation
loss
with growth. Als
o
in t
h
e T
O
RA
protoc
ol pat
h
creation m
a
y not
h
a
pp
en
rap
i
d
l
y. Th
is will lead
s to
th
e
po
ssib
l
e ex
ten
s
i
v
e d
e
lays wh
i
l
e waitin
g
fo
r th
e n
e
w p
a
ths to
b
e
reso
l
v
ed. Howev
e
r, fo
r larg
e
n
e
two
r
k
s
TOR
A
illu
strates a
su
perior
p
e
rforman
ce with
low m
o
b
ility rat
e
. Th
e
avera
g
e e
n
d t
o
end
delay in
AODV,
DSR
a
n
d T
O
RA is in
crease i
n
the
e
x
istence
of
rus
h
ing attack but
delay
of
DSR
i
n
sl
i
g
ht
l
y
hi
g
h
er
t
h
a
n
AO
D
V
due
t
o
cac
he
ove
r
h
e
a
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
110
2
–
11
10
1
109
7.
CO
NCL
USI
O
N
Th
e
p
a
st d
ecad
e
w
itn
essed
a d
r
astic ch
ange in
th
e ub
iqu
i
to
u
s
tech
no
logy, as r
e
sear
ch
er
s pr
opo
sed
several
m
obi
l
e
ad hoc net
w
o
r
k r
o
ut
i
ng
pr
ot
ocol
s w
h
i
c
h c
a
n w
o
r
k
i
n
an
on-
dem
a
nd fa
shi
o
n. H
o
weve
r, t
h
e
ope
n nat
u
re o
f
t
h
ese rout
i
n
g
chan
nel
s
an
d t
h
e abse
nce of
fi
xe
d i
n
fra
st
ru
ct
ure m
a
ke t
h
em
expose
d
t
o
a l
a
rge
ext
e
nt
o
f
secu
ri
t
y
assaul
t
s
. Even t
h
o
u
g
h
a hu
ge am
ount
of t
i
m
e and m
a
n p
o
we
r are i
n
v
o
l
v
e
d
i
n
res
earc
h
,
so
m
e
a
ttack
s are still h
a
rd
to
d
e
fend
in
MANETs, On
e
su
ch
typ
e
o
f
attack
is ru
sh
ing
attack
s, wh
ich
are
p
a
rticu
l
arly h
a
rd
t
o
d
e
tect due to
th
eir inh
e
rited
p
r
op
er
ties, th
at alters th
e n
e
two
r
k
statistics rad
i
cally. In
th
is
pape
r,
we m
o
del
e
d a
p
o
w
e
r
f
u
l
r
u
s
h
i
n
g a
t
t
ack w
h
i
c
h i
s
use
d
a
g
ai
ns
t
t
h
ree
react
i
v
e ro
ut
i
n
g
pr
ot
ocol
s
.
Mo
reo
v
e
r,
we h
a
v
e
tested
th
eir
p
e
rform
a
n
ce in
h
o
s
tile en
v
i
ron
m
en
ts. Sub
s
equ
e
n
tly, th
e p
e
rform
a
n
ce is
measured
with the various perform
a
nce
metrics such as
th
r
o
u
g
h
p
u
t
,
en
d-t
o
-e
nd
del
a
y
,
and
packet
d
e
l
i
v
ery
ratio
etc., As o
u
r fu
t
u
re
wo
rk
, secu
rity so
lu
tion
s
fo
r t
h
is in
trusion
is n
eed
ed
fo
r
h
i
gh
pro
s
p
ecti
v
e of
so
ph
isticated
w
i
r
e
less ad
-hoc n
e
twor
k.
REFERE
NC
ES
[1]
S. Basagni, M.
Conti, S
.
Giordano a
nd I
.
Stojmenovic, “Mobile
Ad Hoc Netw
or
king”, A John
Wiley
& Sons,
I
n
c.,
Publication, 200
4, ISBN 0- 471-
37313-3.
[2]
D.B. Johnson, D.A. Maltz, Y.
Hu, and J.G. Jetchev
a
. Th
e d
y
n
a
mic sour
ce rou
ting protoco
l
fo
r mobile ad hoc
networks (DSR). Intern
et draft, F
e
br
uar
y
2002
. d
r
aft-ietf-manet-dsr-08.txt.
[3]
Eliz
abe
t
h M. Be
lding-Ro
yer
,
Ch
arles E
.
Perkins
Evolution
and fu
ture dir
ections of the ad ho
c ond
emand distance-
vector
routing
pr
otocol-
E
lsevier
2003
[4]
C.E.
Perkins.
Ad
Hoc Ne
tworkin
g
. Addison-W
e
sl
e
y
Professional
,
first ed
ition
,
200
0.
[5]
Perrig, R
.
C
a
netti, D. T
y
gar
,
and D. Song.
Th
e T
E
SLA Broadcast Auth
enti
cation
Protocol
.
RSA Cryp
toBy
te
s
,
5(2):2–13, 2002
.
[6]
Y.C. Hu, A. Perrig, and D.
B. Johnson. Ariadne: A Secure On-demand R
outing Protocol for Ad
Hoc Networks. I
n
ACM MOBICO
M
, pag
e
s 12–23
, 2002.
[7]
K. Sanzgiri, B.
Dahill, B.N. Lev
i
ne, C.
S
h
i
e
l
d
s
,
a
n
d
E
.
R
o
y
e
r
.
A
S
e
c
u
r
e
R
o
u
ting
Protocol for Ad
Hoc Networks. I
n
10th Internation
a
l Conference o
n
Network Proto
c
ols (
I
EEE ICN
P
’02)
, 2002.
[8]
Kirk A. Bradley
,
Steven Cheung
, Nick Puketza, B
i
swan
ath Mukherjee, and Ron
a
ld
A.
Olsson. Detecting Disruptiv
e
Routers: A Dist
ributed Network
Monitoring Ap
proach. In
Proc
eedings of the I
EEE Symposium on Research in
Security and
Privacy
, p
a
ges 115
–124, May
1998
.
[9]
Steven Cheung
and Karl
Levitt. Protecting Rou
ting Infra
stru
ctu
r
es from Denial of
Service Using Cooperativ
e
Intrusion Detection. In
The 1997
New Security Pa
radigms Workshop
, September 1
998.
[10]
P. Papadimitratos and Z.J. H
aas
. S
ecure Dat
a
Trans
m
is
s
i
on i
n
M
obile Ad
Hoc Networks
. In
Second ACM
Workshop on Wireless Security (
W
iSe)
, pag
e
s 41
–50, 2003
.
[11]
Radia Perlman.
I
n
terconnections:
Bridges and
Ro
uters
. Addison-
Wesley
, 1992
.
[12]
Yih-Chun Hu,
Adrian Perrig
,
and David B. Johnson. Pack
et Leashes: A Def
e
nse ag
ainst Wormhole Attacks in
Wireless Ad Hoc Networks. In
P
r
oceedings
o
f
I
E
EE IN
FOCOM 2
003
, April 2003.
[13]
T.P
.
Kum
a
r,
E.
S
u
resh, B.V. R
a
mana, and
B.S.
Shashank, “Survey
: Routing Protocols in
Cognitive Rad
i
o Mesh
Networks”, vo
l.
6, no
. 1
,
pp
. 603
–608, 2015
.
[14]
Adrian Perrig, R
obert Szewczy
k, Victor
Wen, David Culler
,
and J
.
D. Ty
gar
.
SPINS: Security
Protocols for Sensor
Networks. In
Seven
th Annual
ACM Internatio
nal Conferen
ce
on Mobile Com
puting and Net
w
orks (
M
obiCo
m
2001)
, Rome, Italy
,
July
2001.
[15]
E.
S.
Babu,
“An Implementation
and
Performance Evaluation
Study
of AODV,
MAOD
V,
RAO
DV in Mobile Ad
hoc Networks,”
vol. 4
,
no
. 9
,
pp
.
691–695, 2013
.
[16]
E.S. Babu
, C. N
a
gara
ju, a
nd M.
H.M.K. Prasad,
“
A
Co
m
p
arative
Stud
y
of Tr
ee
based Vs. Mesh based Multic
ast
Routing Proto
c
o
l
s in Mobile Ad
hoc Networks”,
vol. 2
,
no
. 6
,
pp
.
6–11, 2013
.
[17]
Yih-Chun Hu, D
a
vid B
.
Johnson, and Adrian
Perr
ig. Secure Efficien
t Distance Vector Rou
ting
in
Mobile Wireless
Ad Hoc Networks.
In
Fourth I
E
EE Workshop on
Mobile Computi
ng Systems and
Applica
tions (
WMCSA ’02)
, June
2002.
[18]
Vincent
P
a
rk
an
d
Scott
Corson. The Tem
porally
Ordered Routing
Algor
ithm for
ad-hoc networks.
[19]
Adrian Perrig
.
The BiBa One-
Time Signature
an
d Broadcast Authentication Protocol. In
Proceed
i
ngs of the
Eighth
ACM Conference on Computer and Co
mmunications Security
(
CCS-8
)
, pages 28–37, Philadelphia PA, USA,
November 2001.
[20]
E.S. Babu and
M.H.M.K. Pras
ad, “An I
m
plementation Analy
s
is
and Eval
uation
Stud
y
of DSR with Inactive DoS
Attack
in
Mobile Ad ho
c Netwo
r
ks”, vol. 2
,
no
.
6, pp
. 501–507
,
2013.
[21]
E.S. Babu, C. N
a
garaju, a
nd M
.
H.M
.
K. P
r
as
ad, “
A
n Im
plem
entation and P
e
rform
ance Evalu
a
tio
n of P
a
ss
ive DoS
Attack
on AODV Routing Proto
c
ol
in Mobile A
d
hoc
N
e
tworks PROTOCOL
OF”,
vo
l. 2, no. 4, 2
013.
[22]
J
ogendra Kum
a
r, “
Broadc
as
ting
Traff
i
c
Load P
e
rform
ance
Analy
s
is of 802
.11
MAC in Mobile Ad hoc Networ
ks
(MANET) Usin
g Random Way
point Model (R
WM)”,
International Journal of Info
rmation &
Network
Security
(I
J
I
N
S
)
ISSN: 2089-3299 Vol.1
,
No.3, August 2
012, pp
.223~227.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
An E
m
pi
ri
cal
C
r
i
t
i
que of
O
n
-De
m
a
n
d
R
o
ut
i
n
g
Pr
ot
oc
ol
s
a
gai
nst
R
u
s
h
i
n
g
At
t
a
ck i
n
MA
N
E
T
(
C
. N
a
gar
aj
u)
1
110
[23]
Yang Shengju*,
Shi Shaoting, Zh
ao Xinhui, “Research
on S
ecur
i
ty
of
Routing Pro
t
ocols Against
Wormhole Attack
in the Ad Hoc Networks”,
TELKOMNIKA Indonesian Journal
of Electrical Engineering
ISSN: 2
302-4046 Vol.12,
No.3, Mar
c
h 20
14, pp
. 2110
~ 2
117.
BIOGRAP
HI
ES OF
AUTH
ORS
M
r
.
S. Ashok
Kumar
receiv
ed his B.Tech degree in In
fo
rmation Technolog
y
from RVR & JC
College of
Engineering
,
Guntur, pursuing M.Tech
degree in C
o
mputer Networks and Security
from K.
L.
University
Guntur.
He ha
s published 3
research
papers
in
various Inter
n
ation
a
l Journal.
He has attended
10 seminars
and workshops.
His
areas of inter
e
sts are wireless networks, security
issues in MANETs and vehicu
lar
networks. He is member of v
a
rious professional societies lik
e
UACEE, IAEN
G, IACSIT and
SDIWC
Mr. E. S
u
res
h
Bab
u
received his
B.Tech degr
ee in Com
puter S
c
ience from
RGM
College of
Engine
ering, Na
nd
y
a
l, M
.
T
ech
degree in Com
puter S
c
ien
ce fro
m
V.T.Univers
i
t
y
Belg
aum
and
pursuing PhD in Computer Science & Engin
eerin
g
from
J.
N.
T
.
Uni
v
e
r
si
ty
Ka
ki
na
da
.
Curre
ntly
,
he
is
working as
an
As
s
o
ciate P
r
ofe
s
s
o
r in the Depa
rtm
e
nt of CS
E i
n
K L Univers
i
t
y
Vij
a
yawad
a
, H
e
has got 12
y
e
ars of teaching
exper
i
en
ce. H
e
has pub
lished 8 res
earch
p
a
pers
in
variou
s
International Journal and 10 r
e
search papers in v
a
r
i
ous National
an
d International C
onferences. He
has attend
ed 32
seminars and workshops. His areas
of inter
e
sts are Wireless Networks, Network
Security
, and M
ANETs. He is
member of various
professional societies
like I
A
ENG, CSTA,and
CSI .
Dr. C. Naga Raju
is
currentl
y
working as
As
socia
t
e P
r
of
essor and Head of th
e Department o
f
Computer Science and Engin
e
ering at YSR Engi
neer
ing College of Yogivemana University
,
Poddatur, Kadapa District, and
Andhra Pradesh, India. He r
eceived his B.Tech Degr
ee in
Com
puter S
c
ien
ce from
J
.
N.T
.
Univers
i
t
y
,
Anant
a
pur,
and M.Tech Degree
in C
o
mputer Science
fromJ.N.T.Univ
e
rsity
H
y
d
e
rab
a
d and PhD in di
gita
l Im
age p
r
oces
s
i
ng from
J
.
N.T.Univ
ers
i
t
y
H
y
der
a
bad
.
He
has
got 18
ye
ars
of te
aching
exp
e
rien
ce
. He re
ce
ived res
e
arch
ex
cel
lenc
e awa
r
d,
tea
c
hing exc
e
ll
e
n
ce award and Ra
yal
a
s
eem
avid
h
y
ar
atn
a
award for his
credit. He wrote text boo
k
on C & Data str
u
ctures. He h
a
s six PhD scholars.
He has published fifty
thr
ee r
e
search p
a
pers in
various Nation
a
l and Intern
ation
a
l Journals and
a
bout thir
ty
research pap
e
rs in v
a
rious Nation
a
l
and International Conferences. He has attend
ed tw
enty
seminars and workshops.
He is me
mber of
various professional societie
s
lik
e IEEE, IS
TE an
d CSI.
M
r
.
A
.
Pe
da Gopi
received h
i
s B.Tech degr
ee
in Inform
ation Technolog
y
from
KLCE College o
f
Engineering, G
untur, pursuing
M.Tech d
e
gree
in
Computer
Networks and
Security
from
K.L.University
Guntur. He h
a
s
published 5 r
e
search
p
a
pers in
v
a
rious Internatio
nal Journal. He
has attended 8
seminars
and workshops. His areas of in
terest
s
are wir
e
less networks, secur
ity
issues in MANETs and v
e
hicular
networks.
Evaluation Warning : The document was created with Spire.PDF for Python.