Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
8
,
N
o.
6
,
D
ece
m
ber
201
8
, pp.
4412
~
44
22
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
8
i
6
.
pp
4412
-
44
22
4412
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Ellip
tic Curve
Crypt
ogra
ph
y Ba
s
ed Dat
a Transmi
ssion again
st
Blackhol
e Attack
in MAN
ET
Jeen
at Sult
ana
1
, T
as
nu
va
A
hmed
2
1
Depa
rtment of
Com
pute
r
Scie
n
ce
and Engi
ne
ering,
Chi
tt
agong
Univer
sit
y
of
En
gine
er
ing
and
T
e
chnol
og
y
,
Banglade
sh
2
Depa
rtment of
Com
pute
r
Scie
n
ce
and Engi
ne
ering,
Southern
Un
ive
rsit
y
Banglad
esh
,
Bang
la
d
esh
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Oct
12
, 201
7
Re
vised
Ju
l
4
,
201
8
Accepte
d
J
ul
18
, 2
01
8
Mobile
nod
es
r
oaming
aro
und
in
the
hosti
le
en
vironment
of
m
obil
e
adhoc
net
work
(MA
N
ET
)
play
the
rol
e
of
route
r
as
well
as
term
ina
l.
W
hil
e
ac
t
in
g
as
a
rou
te
r
,
a
no
de
ne
eds
to
cho
ose
a
r
el
i
abl
e
ro
uti
ng
proto
col.
Beside
s,
an
enc
r
y
pt
ion
al
gor
it
hm
is
nee
ded
to
sec
ure
d
ata
t
o
be
convey
e
d
t
hrough
th
e
unfrie
ndl
y
at
m
o
sphere
whil
e
acting
as
a
te
rm
i
nal
.
W
e
hav
e
i
m
ple
m
ent
ed
El
li
p
ti
c
Curve
C
r
y
ptogr
aph
y
(
EC
C)
al
ong
with
A
dhoc
On
Dem
an
d
Multi
pa
th
Distanc
e
Ve
ct
or
(AO
MD
V)
rou
ti
ng
proto
col
to
sec
ure
d
at
a
tr
ansm
ission
aga
inst
bl
ac
khol
e
at
t
ac
k
in
a
M
AN
ET
.
ECC,
a
publi
c
ke
y
cr
y
p
t
ogra
ph
y
tha
t
works
on
discre
te
loga
ri
thm
proble
m
with
a
m
uch
sm
al
le
r
ke
y
si
ze,
has
bee
n
used
to
enc
r
y
pt
dat
a
p
ac
ke
ts
at
source
node
bef
ore
tra
nsm
ission.
W
e
hav
e
used
AO
M
DV
,
a
rel
i
abl
e
rou
ti
n
g
protoc
ol
co
m
par
ed
to
it
s
par
en
t
protoc
ol
,
Adhoc
On
Dem
and
Distance
V
ec
tor
(AO
DV
),
with
a
m
ult
ip
ath
ext
ension
,
for
routi
ng.
Th
e
enc
r
y
pt
ed
pac
k
e
ts
tra
nsferri
ng
b
et
wee
n
nod
es
via
AO
MD
V,
has
bee
n
prove
d
sec
ure
d
against
bla
ckhole
at
t
ack.
The
per
form
a
nce
of
the
sec
ure
d
pro
to
col
has
bee
n
ana
l
y
z
ed
in
t
erms
of
diffe
ren
t
p
erf
orm
a
nce
m
et
ri
cs
and
in
t
erms
of
v
ar
y
ing
num
ber
o
f
blackhole attac
ker
nodes
.
Ke
yw
or
d:
AO
M
DV
Crypto
gr
a
phy
NS
-
2.5
Secu
rity
Copyright
©
201
8
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights
reserv
ed
.
Corres
pond
in
g
Aut
h
or
:
Jeenat S
ultana
,
Dep
a
rtm
ent o
f C
om
pu
te
r
Scie
nce a
nd E
ng
i
ne
erin
g,
Southe
rn Univ
ersit
y B
ang
la
de
sh
,
22 S
hah
ee
d
Mi
rza La
ne (E)
, M
ehed
i
bag Ro
ad,
C
hitt
agon
g 400
0,
Ba
ngla
de
sh
.
Em
a
i
l:
swar
na
040405
6@
ya
hoo.com
1.
INTROD
U
CTION
A
Mo
bile
A
d
-
hoc
netw
ork
(MA
NET
)
c
on
sist
s
of
a
nu
m
ber
of
no
des
c
omm
un
ic
at
ing
in
an
infr
a
struct
ur
el
e
ss
m
ann
er
us
i
ng
t
he
ra
dio
frequ
e
ncy
c
hann
el
.
Be
sides
pla
yi
ng
the
ro
le
of
a
te
rm
inal,
every
node
nee
ds
to
act
as
ro
uter
.
The
prot
ocol
s
us
ed
f
or
da
ta
transm
iss
i
on
in
MA
NE
T
assum
e
a
t
ru
st
e
d
env
i
ronm
ent
a
nd
t
hu
s
sec
ur
it
y
issues
are
no
t
in
bu
il
t.
The
env
i
ronm
ent
c
an
be
c
onsider
ed
ad
versary
a
s
data
transm
itted
by
a
source
node
can
be
colle
ct
ed
by
a
c
om
prom
ise
d
node
within
it
s
di
re
ct
transm
issi
on
range.
Secu
rity
m
igh
t
be
ac
hieve
d
by
avo
i
ding
the
at
ta
cker
nodes
thr
ough
intr
us
i
on
detect
io
n
te
chn
i
qu
e
s.
T
he
oth
e
r
way
is
to
sec
ure
the
data
to
be
transm
it
te
d
over
a
ho
sti
le
e
nv
i
ronm
ent
so
that
the
data
is
ke
pt
protect
ed
eve
n
after
at
ta
ck.
F
ollow
i
ng
t
he
s
econd
te
c
hn
i
que,
we
ha
ve
use
d
Ell
ipti
c
Cu
rv
e
C
rypto
gr
a
ph
y
(ECC)
al
ong
with
Adh
oc
on
De
m
and
M
ulti
path
Dista
nce
Ve
ct
or
(AOMD
V
)
pr
oto
c
ol
an
d
hav
e
s
how
n
th
e
i
m
pact
on
bl
ackhol
e
at
ta
ck.
Mob
il
e
node
s
travell
in
g
in
m
ob
il
e
ad
hoc
ne
tworks
a
re
vul
ner
a
ble
to
at
ta
cks.
T
hus,
sec
ur
it
y
bec
om
es
a
ver
y
sensiti
ve
issue
in
MANET.
T
o
ens
ure
secur
e
tra
nsm
issi
on
,
va
rio
us
enc
ryptio
n
al
gorithm
s
hav
e
al
read
y
been
us
e
d
with
va
rio
us
prot
oco
ls
.
Sti
ll
th
e
fr
agile
nat
ure
of
the
pr
otoc
ols
does
not
le
t
the
m
keep
dat
a
transm
issi
on
s
ecur
e
f
or
a
lo
ng
pe
rio
d
of
t
i
m
e.
In
t
his
w
ork,
we
ha
ve
sh
ow
n
th
e
im
pact
of
ECC
bas
e
d
AO
M
DV
on
bl
ackhole
at
ta
ck
an
d
a
naly
zed
it
s
perform
ance
us
in
g
va
rio
us
pe
rfor
m
ance
m
et
rics
in
res
pe
ct
of
tim
e
and
a
gain
st
var
yi
ng
nu
m
ber
of
blac
khole
at
ta
c
ker
node
s.
In
the
pr
evio
us
w
ork
[
1],
we
on
ly
d
e
al
t
with
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ell
ipti
c Curve
Crypto
graphy
Base
d Da
t
a
T
r
an
s
missio
n a
gainst
…
(
Jee
nat Su
lt
an
a
)
4413
the
perf
or
m
ance
against
ti
m
e.
Sonal
S
hr
iv
ast
ava
et
al
[2
]
propose
d
a
sc
hem
e
of
In
tr
usi
on
Detect
ion
Syst
e
m
(IDS)
t
hat
us
es
a
hop
co
unt
m
echan
ism
to
detect
the
at
ta
cker
.
T
he
pro
pose
d
I
DS
sc
hem
e
keeps
rec
ord
of
the
routin
g
in
form
at
ion
,
t
he
inter
m
edi
at
e
node
that
receive
s
th
e
data,
the
ne
xt
hop
of
t
hat
node
.
It
gathe
rs
the
at
ta
cker
inf
orm
at
ion
,
for
wards
to
the
ne
tw
ork
an
d
de
fe
nds
f
ur
t
her
par
t
ic
ipati
on
.
Vandan
a
Aror
a
et
al
[3
]
pr
ese
nted
a
tr
ust
ed
a
nd
aut
he
ntica
te
en
viron
m
ent
that
is
cr
eat
ed
by
RSA
base
d
sig
natu
r
e
an
d
im
ple
m
e
nted
with
AOD
V
protoc
ol.
T
hey
ha
ve
us
e
d
RS
A
encr
y
ption
al
gorithm
that
us
e
s
a
la
r
ger
siz
ed
key
a
nd
the
r
outi
ng
protoc
ol,
A
ODV
p
r
ovides
no
m
ul
ti
path
facil
it
y.
Pr
iy
ank
a
Ba
ns
al
et
al
[4
]
stud
y
the
ef
fec
t
of
black
hole
at
ta
ck
and
nei
ghbor
a
tt
ack
on
AO
M
DV
prot
oc
ol.
An
y
s
olu
ti
on
to
the
at
ta
cks
ha
s
not
been
pr
opos
e
d.
M
Ja
na
rdha
na
Ra
j
u
et
al
[5
]
pro
po
se
d
a
sy
stem
based
on
Ell
ipti
c
cur
ve
ElGam
al
Encr
ypti
on
te
ch
niq
ue
an
d
us
ed
it
with
AOD
V.
In
our
pro
po
se
d
syst
e
m
,
we
em
ph
asi
zed
on
t
he
e
ncr
y
ption
a
l
gorithm
that
is
m
or
e
secur
e
a
nd
t
he
protoc
ol, A
O
MDV wit
h
le
ss
over
hea
d.
The
re
st
of
t
he
pap
e
r
is
or
ga
nized
as
fo
ll
ow
s;
Sect
i
on
2
disc
us
ses
t
he
pro
per
ti
es
of
A
OMD
V
protoc
ol,
blac
khole
at
ta
ck
,
t
he
detai
l
of
El
li
ptic
Curve
C
rypto
gr
a
phy
(
ECC
),
the
rea
so
ns
beh
i
nd
c
hoos
i
ng
AO
M
DV
a
s
r
ou
ti
ng
protoc
ol
and
ECC
as
encr
y
ption
te
chn
i
qu
e
,
the
l
at
er
of
t
his
s
ect
ion
is
ab
ou
t
the
m
et
ho
dolo
gy
and
c
on
tri
buti
on,
sect
io
n
3
shows
the
sim
ulati
on
scena
rio
a
nd
re
su
lt
s
an
d
perform
ance
analy
sis
of AOMD
V be
fore a
nd after
s
ecur
it
y i
m
p
lemen
te
d. I
n
sect
i
on
4,
we
c
oncl
ude
our w
ork wit
h
f
uture sc
ope
.
2.
RESEA
R
CH MET
HO
D
2.1.
Pr
op
er
ties of
AOMD
V
The
key
of
the
AO
M
DV
prot
oco
l
li
es
in
en
su
ri
ng
t
hat
m
u
lt
iple
paths
dis
cov
e
re
d
are
lo
op
-
f
ree
a
nd
disjoint.
The
r
ou
te
disco
ve
ry
is
m
ai
ntained
by
a
fl
ood
-
ba
sed
m
echan
is
m
.
AO
MDV
r
ou
te
update
rul
es
are
i
m
ple
m
ented
t
o
eve
ry
node
i
n
orde
r
to
m
ain
ta
in
lo
op
-
f
ree
do
m
and
dis
jo
intness.
AO
M
DV
us
es
the
r
ou
ti
ng
inf
or
m
at
ion
al
read
y
avail
able
in
the
basic
A
ODV
prot
oco
l
[6
]
,
there
by
r
edu
ci
ng
the
overh
ea
d
cause
d
wh
il
e
disco
ver
i
ng m
ulti
ple p
at
hs
[7]
. Th
e
pro
per
ti
es that are
m
ain
ta
ined
b
y
A
O
MDV p
ro
t
oco
l
are
as
foll
ows:
a.
Lo
op Fr
ee
dom
Lo
op
fr
ee
dom
in
A
OMD
V
is
m
ai
ntained
by
the
pa
ram
et
er
‘Ad
ver
ti
sed
_h
op
c
ount
’.
T
he
adv
e
rtise
d
hopc
ount
repre
sents
th
e
m
axim
u
m
ho
pc
ount
of
the
m
ul
ti
pl
e
pat
hs
for
t
he
destinat
io
n
av
ai
la
ble
at
the
s
ource
node.
T
he
a
dverti
sed
hopc
ount
rem
ai
ns
co
nst
ant
f
or
the
s
a
m
e
sequ
e
nce
nu
m
ber
.
The
prot
oco
l
only
acce
pt
s
al
te
rn
at
e r
oute
s w
it
h
l
ow
e
r h
opco
unts, t
hu
s
guara
ntees lo
op
fr
ee
do
m
.
b.
Disjoint
Path
On
e
of t
he
f
ollow
i
ng tw
o
ty
pe
s of
disjoint
pa
ths is m
ai
ntain
ed
b
y
A
OMD
V protoc
ol:
1.
Node
-
dis
j
oi
nt
path: T
he path
that d
oes n
ot have a
ny no
de
i
n
c
omm
on
.
2.
Link
-
dis
joint
pa
th:
Lin
k
-
disjo
int
path
s
do
not
ha
ve
a
ny
c
om
m
on
li
nk
s
.
T
hough,
li
nk
-
di
sjoint
paths
m
ay
hav
e
node
s
in
c
omm
on
.
Link
disjoint
ness
is
m
ai
ntained
by
AO
M
DV
as
node
-
disjointn
es
s
is
m
uch
stric
t
er
than
li
nk
-
dis
jointness
and
pres
ents lesser
nu
m
ber
o
f dis
j
oi
nt
r
oute
s. Th
us
,
is l
ess ef
fecti
ve
.
c.
Mult
iple Pat
h M
ai
ntenan
ce
AO
M
DV
proc
esses
eve
ry
RR
EQ
pac
ket
to
m
ai
ntain
m
ul
t
iple
paths
.
Whi
le
it
s
par
ent
node
,
A
O
D
V
processes
the
f
irst
RR
EQ
se
nt
from
a
so
ur
c
e
node
t
o
c
rea
te
a
com
m
un
ic
at
ing
path
bet
ween
tw
o
no
de
s
an
d
discar
d
the
o
t
he
r on
es
.
d.
Fr
es
h
r
oute
m
a
intenanc
e
AO
M
DV
c
hoose
s
routin
g
pat
h
by
co
ns
ide
ri
ng
the
m
axi
m
um
sequ
ence
nu
m
ber
and
th
us
m
ai
ntains
fr
es
h
route.
I
n
case
of
sam
e
seq
ue
nce
nu
m
ber
,
it
co
ns
iders
m
ini
m
u
m
hop
c
ount.
T
he
ty
pes
of
se
qu
e
nce
nu
m
ber
m
ai
ntained by
AO
M
DV are:
1.
Sour
ce
Se
qu
e
nc
e Num
ber
: Ev
ery n
ode m
ai
nt
ai
ns
a m
onoton
ic
al
ly
incr
easi
ng se
quence
nu
m
ber
.
2.
Desti
natio
n
Se
qu
e
nce
Nu
m
be
r:
Eve
ry
node
m
ai
ntains
th
e
highest
kn
own
seq
ue
nce
nu
m
ber
f
or
ea
ch
destinat
io
n
in
the
routin
g
ta
bl
e.
2.2.
Bl
ackhole
A
t
t
ack
Bl
ackhole
Atta
ck
is
a
kind
of
at
ta
ck
w
her
e
t
he
at
ta
cke
r
no
de
se
nds
a
fals
e
RR
EP
as
a
n
answer
t
o
t
he
RR
EQ
a
nd
sho
ws
it
sel
f
as
a
r
ou
ti
ng
node
wi
th
the
fr
e
sh
est
route
by
ge
nerat
ing
a
la
rg
e
r
s
equ
e
nce
num
ber
.
In
this
at
ta
ck
[8
]
[9
]
,
the
a
dvers
ary
node
at
tra
ct
s
pack
et
s
bu
t
do
es
not
tran
sm
it
any
pack
et
to
the
destinat
io
n,
rather
dr
op
s
t
hem
al
l.
Du
e
to
this
at
ta
ck
the
pac
kets
se
nd
s
by
the
no
des
do
no
t
re
ach
th
ei
r
pro
pose
d
destinat
io
n.
A
OMD
V
ge
neral
ly
cho
os
es
an
al
te
rn
at
e
path
when
a
pat
h
is
blo
cke
d
by
an
at
ta
cker
node
.
The
pro
blem
m
ai
nly
occu
rs
wh
e
n
eve
ry
oth
er
pat
h
is
bloc
ked
by
a
nu
m
ber
of
at
ta
cker
nodes
an
d
the
at
ta
cker
node
act
ing
as
a
ro
uter
cam
ou
fla
ges
it
sel
f
as
a
valid
on
e
.
The
Fig
ur
e
1
sh
ows
a
MA
N
ET
wh
e
re
the
pack
et
transm
itted
by
the
sou
rce,
S
de
sti
ned
for
D
i
s
achie
ved
by
M,
a
blac
khole
at
ta
cker
node
play
ing
t
he
r
ol
e
of
a
router
with
gre
at
er s
eq
ue
nce
nu
m
ber
.
A, B a
nd C
are
the i
nt
erm
ediat
e n
odes.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
87
08
In
t J
Elec
&
C
om
p
En
g,
V
ol.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4412
-
4422
4414
Figure
1. Bl
ackhole at
ta
ck
R
REQ: R
oute
r
e
qu
e
st, RR
EP:
Rou
te
rep
ly
, se
q_no: se
qu
e
nc
e num
ber
2.3.
El
li
pt
ic
C
urve Cr
yptog
raphy
Ell
ipti
c
Curv
e
Crypto
gr
a
phy
(ECC)
is
us
ed
to
descr
i
be
a
gr
ou
p
of
cry
ptogra
phic
too
ls
and
protoc
ols
wh
e
re
the
sec
ur
it
y
is
base
d
on
t
he
discre
te
log
arit
hm
pro
blem
.
ECC
is
based
on
s
et
s
of
num
ber
s
an
d
equ
at
io
ns t
hat
are ass
ociat
ed wit
h
el
li
ptic cu
rv
es
[1
0].
E
CC
works i
n
the
foll
ow
i
ng phase
s:
a.
Key Ge
ne
rati
on
1.
Global
P
ub
li
c
Elem
ents
a)
E
p
(a,
b)
ell
ipti
c
curve
with
para
m
et
ers
a, b &
p
in
the e
quat
ion
:
Y
2
m
od
p=(X
3
+aX+
b) m
od
p
b)
G
is t
he base
point
on ell
ipti
c curve
2.
User A
K
ey
G
ener
at
io
n
a)
Sele
ct
p
ri
vate
key
n
A
; n
A
<
n
b)
Ca
lc
ulate
p
ubli
c k
ey
, P
=
n
A
×
G
3.
User B Key
G
e
ner
at
io
n
a)
Sele
ct
p
ri
vate
key,
n
B
; n
B
<
n
b)
Ca
lc
ulate
p
ubli
c k
ey
, M
=
n
B
×
G
4.
Gen
e
rati
on
of
Secret
Key b
y
us
er
A,
P
1
=
k
=
n
A
×
M
5.
Gen
e
rati
on
of
Secret
Key
by
us
er
B,
P
2
=
k
=
n
B
×
P
The
t
wo cal
cul
at
ion
s
pro
du
ce
the sam
e resu
lt
b
eca
us
e
n
A
×
M
=
n
A
×
(n
B
×
G)
=
n
B
×
(n
A
×
G)
=
n
B
×
P.
b.
ECC
En
c
ryptio
n
1.
Con
si
der a m
e
ssage ‘Pm
’
sen
t from
A
to
B.
2.
User ‘
A’ ch
oos
es a r
a
ndom
p
osi
ti
ve
intege
r
‘
k’, a
pr
i
vate ke
y ‘n
A
’
and
ge
ne
rates the
publi
c k
ey
,
3.
P
A
=
n
A
×
G
.
4.
Chooses
G, the
b
ase
point sele
ct
ed
on t
he
Ell
ipti
c Curve
E
p
(
a, b).
5.
Pr
od
uces t
he
ci
ph
e
rtext,
consi
sti
ng
of p
ai
r of p
oin
ts,
Cm
={
kG, P
m
+
kP
B
}
=
(C
1
, C
2
) w
here,
6.
P
B
=
n
B
×
G.
7.
The p
ub
li
c
key of B
with
pr
i
va
te
k
ey
‘n
B
’.
c.
ECC
D
ecry
ption
1.
To decry
pt the
ci
ph
e
rtext,
Cm
, B m
ulti
plies th
e
first
po
i
nt in
the
pair, C
1
by
B’s
sec
ret
key.
2.
Subtract
s the
re
su
lt
fro
m
the
seco
nd point,
C
2,
P
m
+ k
P
B
-
n
B
(
kG)
=
Pm
+ k
(n
B
G)
–
n
B
(
kG)
=
Pm
.
2.4. Wh
y AO
MDV
?
AO
M
DV,
A
dh
oc
on
Dem
and
Mult
ipath
Dis
ta
nce
Vecto
r
P
ro
t
oco
l
[
7],
is
the
m
ulti
path
extensi
on
of
the
re
now
ne
d
AOD
V
protoc
ol.
Be
in
g
t
he
m
ul
ti
path
exte
nd
e
d
ve
rsion,
it
pro
vid
es
m
or
e
al
te
rn
at
e
pat
hs
a
nd
thu
s
is
m
or
e
re
li
able.
It
reduc
es
the
over
hea
d
cause
d
by
li
nk
fail
ur
e
a
nd
r
oute
disc
ov
e
r
y.
In
AO
M
DV,
route
disco
ver
y i
s
on
ly
n
eede
d wh
e
n
e
ver
y al
te
r
na
te
li
nk
is
bro
ke
n.
2.5. Wh
y
EC
C?
Ell
ipti
c
Curv
e
Crypto
gr
a
ph
y
is
a
reli
able
and
e
ff
ic
ie
nt
public
key
cr
yptogra
ph
ic
te
chn
i
qu
e
t
hat
pro
vid
es
e
quivale
nt
sec
ur
it
y
as
oth
e
r
public
key
c
r
ypto
gr
a
phic
te
chn
iq
ues
bu
t
with
sm
al
l
er
key
siz
e
[10]
,
[11].
More
over,
It
i
s
base
d
on
the
discrete
lo
ga
rithm
pr
oble
m
[1
2]
w
hich
sta
te
s
that
if
K
=
nx
G
,
the
n
it
is easy
to d
er
ive K
w
it
h giv
en
n
a
nd
G,
but t
ou
gh to
re
ve
al
n
w
it
h
gi
ven K
and G
. T
hu
s w
it
h
ECC
,
it
w
il
l be
tough f
or the
bl
ackhole at
ta
ck
er
node
t
o
ret
rieve the
private
key f
ro
m
g
ive
n sec
ret
key an
d p
ub
li
c
key.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ell
ipti
c Curve
Crypto
graphy
Base
d Da
t
a
T
r
an
s
missio
n a
gainst
…
(
Jee
nat Su
lt
an
a
)
4415
2.6.
Me
thod
olo
gy
The
AO
M
DV
protoc
ol
fir
st
di
scov
e
rs
the
m
ulti
path
r
oute
to
the
destinat
i
on
t
hro
ugh
rou
te
disco
ver
y
process
an
d
s
ends
the
pa
ck
et
s
to
it
.
The
flo
wch
a
rt
of
the
disco
ve
ry
is
sho
wn
in
Fi
gure
2.
At
the
ve
ry
beg
i
nn
i
ng
of
da
ta
transm
issi
on
,
t
he
s
ource
node
broa
dcas
ts
RR
EQ
a
nd
the
ne
xt
hop
t
hat
ha
s
r
oute
to
th
e
destinat
io
n
se
nds
bac
k
RR
EP
.
O
n
the
c
ontr
ary
to
A
ODV,
A
OMD
V
pro
cesses
e
ver
y
R
REQ.
Whi
le
A
O
D
V
processes
on
ly
the
first
RR
E
Q
an
d
discar
ds
the
rest.
By
pr
ocessin
g
eve
ry
RR
EQ,
AO
M
DV
e
ns
ures
m
ulti
path
connecti
vity
.
The
destinat
io
n
node
c
hoose
s
on
e
of
the
m
ulti
paths
by
co
m
par
in
g
the
seq
uence
num
ber
.
The
fr
e
sh
est
r
ou
te
is
gu
a
ra
nt
eed
by
c
on
si
de
rin
g
the
la
rg
e
st
sequ
e
nce
num
ber
.
In
cas
e
of
eq
ual
seq
uen
ce
nu
m
ber
,
pat
h
with
le
ss
hop
count
is
c
onsidere
d
t
o
en
sur
e
the
s
hortest
path.
T
he
de
sti
nation
f
orwards
th
e
RR
EP to
t
he
s
ource
no
de
a
nd the
path
is
d
e
fin
ed fo
r data
tra
nsm
issi
on
.
Af
te
r
tra
ns
m
it
t
ing
nu
m
ber
of
pack
et
s
,
A
OM
DV
a
vo
i
ds
the
m
al
iciou
s
no
de
as
it
is
design
ed
s
o
a
nd
restrains
the
s
ource
node
f
r
om
transm
it
ti
n
g
pack
et
s
to
t
he
at
ta
cke
r
node.
I
n
the
m
e
anwhil
e,
s
om
e
of
the
pack
et
s
ha
ve
a
lready
bee
n
ret
rieve
d
by
the
a
tt
acker
no
de.
Our
goal
is
t
o
secur
e
t
hose
re
trie
ved
pac
kets
from
rev
eal
in
g.
T
hus
to
sec
ure
the
data
pac
kets,
the
pac
kets
a
r
e
enc
rypted
by
ECC
.
This
is
do
ne
by
cre
at
ing
a
secur
e
age
nt
t
hat
ge
ner
at
es
the
e
ncr
ypte
d
pa
cket.
T
he
pac
ket
is
the
n
rea
ched
the
destinat
ion
thr
ough
one
of
the s
el
ect
ed
m
ulti
paths.
Figure
2.
Flo
w
ch
a
rt of
route
disco
ver
y
with
AOMD
V prot
oco
l.
RR
EQ: R
ou
te
re
qu
est
,
R
REP:
Ro
ute r
e
ply,
seq
_no: se
qu
e
nc
e num
ber
Ever
y
ti
m
e
a
node
w
orks
as
a
sou
rce,
it
ge
ner
at
es
a
pri
va
te
/pu
blic
key
pair
.
Firstl
y,
i
t
chooses
a
rand
om
pr
ivat
e
key,
ge
ne
rates
a
sec
ret
ke
y
from
it
s
own
pr
i
vate
key
and
recei
ver’s
public
key.
T
hen
it
encr
y
pts
the
pa
cket
with
the
ne
wly
gen
e
rated
secret
ke
y
an
d
anno
un
ces
t
he
public
key.
T
he
encr
ypte
d
pa
cket
is
sent
via
AOM
DV
pr
oto
c
ol.
Af
te
r
receivi
ng
the
e
ncr
y
pted
pac
ket,
the
destinat
io
n
no
de
ge
ner
at
es
th
e
sa
m
e
secret
key
wit
h
the
help
of
it
s
own
pr
i
vate
ke
y
and
the
sen
der’s
new
publ
ic
key.
The
n
usi
ng
it
s
s
har
e
d
secret
key
an
d
pr
ivat
e/
public
key
pa
ir,
the
receive
r
decr
y
pts
the
pa
cket
to
get
the
or
i
gin
al
data
as
show
n
in
Fig
ure
3.
It
will
be
toug
h
f
or
the
m
al
iciou
s
node
to
re
trie
ve
the
pri
va
te
key
fr
om
giv
en
sec
ret
key
and
the
public
key.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
87
08
In
t J
Elec
&
C
om
p
En
g,
V
ol.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4412
-
4422
4416
Th
us
,
if
an
at
ta
cker
m
anag
es
to
rev
eal
the
secret
key,
it
would
be
a
ch
al
le
ng
e
to
retri
eve
the
pr
i
vat
e
key
without
wh
ic
h decry
pting t
he packet is
not
possible.
Figure
3.
Pac
ke
t encry
ption b
y EC
C
3.
RESU
LT
S
A
ND AN
ALYSIS
We
ha
ve
us
e
d
NS
-
2.3
5
f
or
the
sim
ulatio
n
of
t
he
pro
po
sed
syst
em
,
a
discrete
-
e
ve
nt
si
m
ulati
on
instru
m
ent
that
has
bee
n
pro
ve
d
f
un
ct
io
nal
in
stu
dying
the
vib
ra
nt
natu
re
of
MA
NET.
The
pe
rfo
rm
ance
is
m
easur
ed
i
n
three
case
s:
w
it
ho
ut
m
al
ic
io
us
no
de,
with
three
b
la
ck
hole
at
ta
cker
node
s
an
d
wit
h
ECC
i
m
ple
m
entat
io
n.
The
m
et
rics that are
us
ed
for
analy
zi
ng
t
he p
erfor
m
ance [5] o
f
the
pro
t
oco
l
are desc
ribe
d belo
w:
a.
Av
e
ra
ge
T
hro
ughp
ut
Av
e
ra
ge
nu
m
ber
of
pac
kets
pa
ssing t
hro
ugh
the n
et
wor
k pe
r un
it
of
ti
m
e. I
t i
s m
easur
ed
i
n kbps.
Av
e
ra
ge
T
hro
ughp
ut
=
s
e
co
n
d
s
)
T
i
m
e
(
i
n
ly
s
u
c
c
e
s
s
f
u
l
s
e
n
t
p
ac
k
e
t
s
of
N
u
m
b
er
b.
Packet
Deli
very
Rat
io
It
is
the
rati
o
of
total
nu
m
ber
of
pack
et
s
s
ucc
essfu
ll
y
receiv
ed
at
the
destin
at
ion
node
s
to
the
nu
m
be
r
of p
ac
kets
for
warde
d by the
so
urce
node
s t
hro
ughout t
he sim
ulati
on
.
Packet
Deli
very
Rat
io
=
P
a
c
k
e
t
s
S
e
n
t
of
N
u
m
b
e
r
P
a
c
k
e
t
s
R
e
c
e
i
v
e
d
of
N
u
m
b
e
r
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ell
ipti
c Curve
Crypto
graphy
Base
d Da
t
a
T
r
an
s
missio
n a
gainst
…
(
Jee
nat Su
lt
an
a
)
4417
c.
Av
e
ra
ge
E
nd
-
to
-
E
nd
Delay
This
is
the
a
ve
rag
e
delay
bet
ween
t
he
se
nd
i
ng
of
pac
kets
by
the
s
ource
and
t
hose
rece
ived
by
the
receiver
.
Av
e
ra
ge
E
nd
-
to
-
E
nd
Delay
=
n
i
0
e
i
v
e
d
P
a
c
k
e
t
s
R
e
c
of
N
u
m
b
e
r
T
o
t
a
l
S
e
n
t
)
P
a
c
k
e
t
of
T
i
m
e
-
R
e
c
i
v
e
d
p
a
c
k
e
t
of
(
T
i
m
e
d.
Norm
al
iz
ed
Rou
ti
ng L
oad
The n
um
ber
of
r
ou
ti
ng
pac
kets transm
i
tt
ed
pe
r data
p
ac
ket
delivere
d
at
t
he
d
est
inati
on.
3.1. Si
mula
tio
n S
cen
ario
In
N
S
-
2.3
5,
w
e
hav
e
co
ns
i
de
red
a
n
area
of
1186
x
58
4
m
et
ers
and
the
wireless
to
po
l
og
y
we
hav
e
us
e
d
c
on
ta
in
s
25
no
des
i
nclu
ding
3
black
hole
at
ta
cke
r
nodes.
T
he
m
axim
u
m
sp
eed
of
the
m
ob
il
e
nodes
i
s
0.1Mb
ps
.
T
he
t
otal t
i
m
e u
sed for sim
ulati
on
is 1
00 seco
nds.
Th
e
sim
ulati
o
n
sce
nar
i
o
is s
how
n
i
n
Ta
ble
1.
Table
1.
Sim
ul
at
ion
E
nvir
onm
ent
Si
m
u
latio
n
Para
m
e
ter
Valu
e
Si
m
u
lato
r
NS
-
2
.35
Area
1
1
8
6
x
5
8
4
Ro
u
tin
g
pro
t
o
co
l
AOMD
V
Attack
Black
h
o
le attack
Pack
et size
1
0
0
0
,1
5
0
0
Nu
m
b
e
r
o
f
no
d
es
25
Nu
m
b
e
r
o
f
attacke
r
n
o
d
es
3
Tr
af
f
ic ty
p
e
CB
R,UDP
No
d
e Place
m
en
t
Ran
d
o
m
Si
m
u
latio
n
ti
m
e
1
0
0
sec
The
out
pu
t
N
A
M
file
is
sh
own
in
Fi
gure
4
wh
ic
h
is
ge
nerat
ed
as
a
res
ult
of
sim
ulati
on
.
NA
M
sta
nds
for
Net
work
An
im
at
or
that
sh
ows
t
he
gr
aph
ic
al
re
pres
entat
ion
of
pa
cket
trans
fer
a
m
on
g
the
no
de
s
in
a
MANET
.
Figure
4. O
utput N
AM
file
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
87
08
In
t J
Elec
&
C
om
p
En
g,
V
ol.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4412
-
4422
4418
In
Fig
ur
e
4,
th
e
blu
e
no
de
re
pr
ese
nts
t
he
de
sti
nation,
the
s
ources
a
re
de
note
d
by
gr
ee
n
and
the
re
d
on
e
s
are
the
bl
ackhole
at
ta
cker
node
s.
F
our
source
no
de
s
are
tra
ns
m
itti
ng
pack
et
s
i
n
tim
e
intervals
of
20
seco
nd
s
.
F
r
om
the
generate
d
trace
file
s,
diff
e
re
nt
pe
rfo
rm
ance
m
et
ric
s
are
m
easur
e
d
s
uc
h
as
a
ve
rage
thr
oughput, e
nd
-
to
-
en
d delay
,
p
ac
ket d
el
i
very
r
at
io a
nd norm
al
iz
ed
routin
g
loa
d
a
s s
how
n
in
Ta
ble 2.
Table
2.
Per
for
m
ance Analy
sis o
f AOM
DV,
with
Atta
ck
a
nd
with Sec
uri
ty
Para
m
eters
AOMD
V witho
u
t
Maliciou
s n
o
d
es
AOMD
V with
Black
h
o
le Attack
AOMD
V with
EC
C
(again
st Black
h
o
le A
ttack
)
Total CB
R p
ackets
sen
t
1241
1241
1492
CB
R p
ackets
r
ecei
v
ed
1240
48
238
Pack
et deliv
er
y
r
a
t
io
0
.99
9
2
0
.03
8
7
0
.15
9
5
End
-
to
-
en
d
delay
(
m
s
)
8
5
6
9
0
.5
8
5
6
5
6
.3
8
5
6
6
1
No
r
m
aliz
ed
Ro
u
tin
g
L
o
ad
0
.11
0
2
.79
2
0
.54
2
Av
erage T
h
rou
g
h
p
u
t (
k
b
p
s)
3
5
4
.969
2
2
0
.408
2
9
3
.089
The
data
f
r
om
Table
2
has
be
en
us
ed
to
plo
t
the
perform
ance
m
e
tric
s.
Th
e
pe
rfor
m
ance
m
et
rics
are
m
easur
ed
in
e
ve
ry 10 sec
onds
and is a
naly
zed a
gainst ti
m
e
and num
ber
of
m
al
ic
iou
s nod
e
s consec
utively
.
3.2. Per
f
orm
ance
Anal
ys
is
agains
t
Ti
me
The
c
urve
i
n
F
igure
5
s
hows
that
the
th
r
oughput
without
a
ny
m
alici
ou
s
a
ct
ivit
y
rises
to
ab
ou
t
35
5
kbps
,
the
t
hro
ughput
dro
pp
e
d
to
a
bout
22
0
kbps
on
th
e
pr
e
sence
of
m
a
li
ci
ou
s
no
des.
B
ut
with
ECC
i
m
ple
m
entat
io
n
in
the elem
entary A
OMD
V protoc
ol,
t
he v
al
ues
are
in
c
re
ased t
o 290 k
bps.
Figure
5. Tim
e
v
s.
th
rou
ghput cu
rv
e
. nom
al
_th
: Th
rou
ghput wit
h n
o
m
al
ic
i
ou
s
no
de, m
al
_
th: Th
rou
ghput
with m
al
ic
iou
s nod
e
, s
ec
_th
:
Thro
ughput
wi
th sec
ur
it
y
As
sho
wn
i
n
F
igure
6,
t
he
pa
cket
delive
ry
r
at
io
kep
t
re
duc
ing
on
t
he
pres
ence
of
blac
khole
at
ta
cker
nodes
in
the
MANET
a
nd
dro
ps
to
ab
out
0.
04
with
th
ree
at
ta
cker
s.
But
wh
e
n
the
pr
op
os
e
d
m
e
t
hod
is
i
m
ple
m
ented
t
hen
e
ve
n
with
the
at
ta
cker
nodes
bein
g
present
in
the
netw
ork,
the
pack
et
deliver
y
rati
o
i
m
pr
oved
cons
idera
bly t
o
a
bove 0.
15.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ell
ipti
c Curve
Crypto
graphy
Base
d Da
t
a
T
r
an
s
missio
n a
gainst
…
(
Jee
nat Su
lt
an
a
)
4419
Figure
6. Tim
e
v
s.
pac
ket
delivery
rati
o
c
urv
e. nom
al
_p
kt:
PD
R
with
no
m
al
ic
iou
s nod
e
, m
al
_p
kt: PD
R wit
h
m
al
ic
iou
s nod
e
, s
ec
_pkt: P
DR w
it
h sec
uri
ty
The
e
nd
-
to
-
en
d
delay
is
alm
os
t
sam
e
in
al
l
the
th
ree
cases
as
seen
i
n
Fi
gure
7.
I
niti
al
ly
t
h
e
delay
is
a
bit
highe
r
with
ECC
im
plem
entat
ion
.
O
n
an
a
ve
rag
e
,
t
he
pro
po
se
d
m
et
ho
d
does
not
af
fect
the
e
nd
-
to
-
en
d
delay
.
Figure
7. Tim
e
v
s.
del
ay
cu
r
ve
.
nom
al
_d
el
ay
: En
d
-
to
-
e
nd del
ay
w
it
h
no m
al
ic
iou
s
node,
m
al
_d
el
ay
: En
d
-
to
-
end d
el
ay
with
m
al
ic
iou
s
node
, s
ec
_d
el
ay
: E
nd
-
to
-
e
nd d
el
ay
with sec
ur
it
y
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
87
08
In
t J
Elec
&
C
om
p
En
g,
V
ol.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4412
-
4422
4420
The
gra
ph
i
n
Figure
8
il
lust
rated
the
norm
al
iz
ed
routin
g
load
f
or
the
th
ree
m
entioned
cases.
The
Norm
al
iz
ed
Rou
ti
ng
Loa
d
is
ab
out
0.1
wit
hout
at
ta
ck;
it
raises
high
a
bo
ve
2.0
i
n
t
he
presence
of
at
ta
ck
a
nd
r
ed
uces t
o 0.5
with ECC i
m
ple
m
entat
ion
.
Figure
8. Tim
e
v
s.
no
rm
alized rou
ti
ng l
oad c
urve.
nom
al
_n
or_r
t
_ld
:
Norm
al
iz
ed
r
ou
ti
ng l
oad w
it
h n
o
m
al
ic
iou
s nod
e
, m
al
_n
or_r
t
_l
d: No
rm
alized r
outi
ng loa
d wi
th m
alici
ou
s
node
, s
ec
_nor_
r
t_ld: N
orm
aliz
ed
routin
g
loa
d wi
th sec
ur
it
y
3.3.
Perf
orm
ance
Anal
ys
is
agains
t Num
be
r of
Ma
li
ci
ou
s
Nodes
As
in
Fi
gure
9,
we
ob
se
r
ve
that
with
var
yi
ng
num
ber
of
m
al
ic
iou
s
node
s,
the
a
ver
a
ge
throu
ghput
do
e
sn’t dec
rea
se r
a
pid
ly
, inst
ead
with the
ne
wly de
velo
ped
protoc
ol,
t
he
th
rou
gh
put cha
nges
ver
y
sli
gh
tl
y.
Figure
9. N
umber
of m
al
ic
iou
s nodes
vs
. A
ve
rag
e
Thr
ough
pu
t
with
secu
re
d AO
M
DV
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ell
ipti
c Curve
Crypto
graphy
Base
d Da
t
a
T
r
an
s
missio
n a
gainst
…
(
Jee
nat Su
lt
an
a
)
4421
The
Pac
ket
D
el
ivery
rati
o
i
s
def
i
nitel
y
hig
he
st
with
no
m
a
li
ci
ou
s
no
de
in
t
he
e
nv
i
ronm
ent.
It
decr
ease
s
slowl
y
with
increasing
num
ber
of
m
al
ic
iou
s
node
s
as
in
Figu
re
10
.
T
he
aver
a
ge
en
d
-
to
-
e
nd
delay
sh
ow
n
in
Fi
gur
e
11
i
ncr
eases
gr
a
dual
ly
with
increm
ented
m
al
ic
iou
s
no
des
as
tim
e
is
ta
ken
by
the
e
ncr
y
ption
process
w
it
h E
CC
.
Figure
10. N
um
ber
o
f
m
al
ic
i
ou
s
no
des vs.
Packet
Deli
very
Rat
io w
it
h
se
cur
e
d A
OMD
V
Figure
11. N
um
ber
o
f
m
al
ic
i
ou
s
no
des vs.
Av
e
ra
ge Del
ay
w
it
h sec
ured
AO
M
DV
As
s
how
n
in
F
igure
12,
t
he
norm
al
iz
ed
rout
ing
loa
d
gro
w
s
with
the
nu
m
ber
of
blac
khole
at
ta
cker
nodes
pr
e
sent
in
the
sit
uatio
n,
th
ough
the
norm
al
iz
ed
routing
l
oad
m
ay
var
y
de
pe
nding
on
t
he
nu
m
ber
of
pack
et
t
ran
sm
issi
on.
Figure
12. N
um
ber
o
f
m
al
ic
i
ou
s
no
des vs.
Norm
al
iz
ed
Rou
ti
ng L
oad
wit
h
sec
ur
e
d A
O
MDV
4.
CONCL
US
I
O
N
A
MA
NET
is
su
bject
to
a
va
riet
y
of
at
ta
cks
.
This
pa
per
proposes
a
way
t
o
sec
ur
e
data
transm
issi
on
in
MA
NET
s
o
that
data
m
ay
rem
ai
n
confide
nt
w
he
n
a
ne
goti
at
ed
node
ge
ts
it
s
pro
pr
ie
t
ary.
We
ha
ve
chosen
black
ho
le
at
ta
c
k,
a
ve
ry
com
m
on
on
e
,
to
an
al
yz
e
the
i
m
p
act
of
secu
red
AO
M
DV,
th
ou
gh
secu
rity
is
need
e
d
Evaluation Warning : The document was created with Spire.PDF for Python.