Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 4
,
A
ugu
st
2016
, pp
. 18
57
~
1
865
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
4.9
941
1
857
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
Secure Multicast Routing Protoco
l in MANETs Using Efficient
ECGDH Algorithm
Gopi Arepalli
1
, Sures
h Bab
u
Erukul
a
2
, Arepa
lli
Peda
Go
pi
3
, C.
Na
ga
Ra
ju
4
1
Research
Scholar, K
L
University
, Guntur
, Andh
ra Pradesh
2
Department of
CSE, KL Univ
er
sity
, Guntur
, An
dhra Pradesh
3
Department of
CSE Vijan
a
n Un
iversity
, Guntur
,Andhrapradesh
4
Department of
CSE, YSR College of
YV Univ
er
sity
, Proddutur
,
Andhra Pradesh
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
Ja
n 17, 2016
Rev
i
sed
May 16
, 20
16
Accepted
May 27, 2016
An Ad-hoc Network covers a set
of autono
mous
mobile nodes that
communicates through wireless communica
tio
n in an infrastructure-
less
environment. Mostly
MANETs
are used
in group communication
m
echanism
s
like
m
ilitar
y
appli
c
a
tions, em
ergen
c
y se
arch,
rescue
operat
i
ons,
vehicu
lar ad-ho
c
communication
s
and mini
ng op
erations etc. In such ty
p
e
of
networks, group
com
m
unication
is t
a
kes p
l
ac
e
b
y
m
u
ltic
asting
te
chniqu
e
.
Communication and collabor
ation is
necessary
among the n
odes in th
e
groups in m
u
lticast protocols
.
PUMA has the best m
u
lticast rout
i
ng protocol
compared to tr
ee and mesh based mu
lticast pro
t
ocols although it suffers from
se
c
u
rity
issue
s
. PUMA
ma
inly
suffe
rs from Man In The middle
attack
.
M
I
TM
atta
ck ge
nerat
e
s
traff
i
c f
l
ow, drop the
pa
ckets
and
m
i
s
c
om
m
unicate
the neighbor nod
es with false hop count. So defen
d
ing from MIT
M
attack we
designed a n
e
w mechanism called El
liptic Cur
v
e Group Diffie-Hellman
(ECGDH).
This paper compares results
of PUM
A
[1] routing protocol with
legitimate, under attack and
after provi
ding security
ag
ainst attack. Fin
a
lly
we observed ECGDH [2] give
s efficient results
even
attack has happened.
Keyword:
Diffie
-
hellm
an
Ellip
tic
cu
rv
e g
r
ou
p
Man in the m
i
ddle attack
M
ANET
s
PUMA
Copyright ©
201
6 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
1.
INTRODUCTION
A M
obi
l
e
ad h
o
c net
w
o
r
k i
s
an aut
o
n
o
m
ous col
l
ect
i
on of
no
des d
o
n
o
t
r
e
l
y
on any
pre
-
est
a
bl
i
s
h
e
d
in
frastru
ct
u
r
e th
at fo
rm
s d
y
n
a
mic co
mm
u
n
i
cativ
e n
e
twork.
Nod
e
s i
n
th
ese n
e
two
r
k
m
a
k
e
u
s
e
of m
o
b
ilit
y an
d
wireless co
mmu
n
i
cation
to
main
tain
con
n
e
ctiv
ity. Ho
wev
e
r, th
e li
m
ited
p
r
op
ag
atio
n
rang
e of th
ese wireless
en
v
i
ron
m
en
t
m
ak
e a ch
alleng
ing
issu
e t
o
estab
lis
h th
e ro
u
t
es. Sub
s
eq
u
e
n
tly, MANETS are mu
lti-ho
p
in
frastru
ct
u
r
es
less n
e
twork th
at estab
lish
e
s th
e
rou
t
es
the
m
se
lves “on the fly”. T
h
ese
networks
are
suitable
for ap
p
licatio
ns lik
e b
a
ttlefield
,
em
erg
e
ncy search, re
scu
e
o
p
e
ration
s
, v
e
h
i
cu
lar ad-ho
c
co
mm
u
n
i
catio
n
s
and
m
i
ni
ng o
p
erat
i
ons et
c.
In s
u
c
h
ap
pl
i
cat
i
ons,
com
m
uni
cat
i
o
n an
d col
l
a
b
o
r
a
t
i
on o
f
n
ode
s
am
ong t
h
e
gr
ou
p i
s
n
ecessary. Th
erefo
r
e, m
u
ltica
s
t co
mm
u
n
i
cat
io
n
is v
e
ry mu
ch
i
n
tend
ed
t
o
th
e gro
u
p
co
mm
u
n
i
catio
n
wh
ich
sav
e
s n
e
t
w
or
k r
e
sour
ces
and
b
a
ndw
id
th.
M
o
r
e
ov
er,
M
u
lti
casting is
a se
rvice for
di
ssem
i
n
a
tin
g inform
a
tio
n
t
o
a g
r
ou
p
of hosts th
at send
s
th
e d
a
ta
fro
m
a sou
r
ce to
m
u
l
tip
le d
e
stin
atio
ns in
t
h
e n
e
two
r
k
.
Th
e u
n
i
qu
e
p
r
op
erties o
f
m
u
l
ticast
co
mm
u
n
i
catio
n
is first,
t
h
e
node
can
join anyt
im
e and ca
n l
eave a
n
ytim
e from
m
u
lt
i
cast
grou
p dy
nam
i
cal
l
y
. Seco
nd
, t
h
e n
ode
s ha
ve n
o
c
onst
r
ai
nt
s o
n
t
h
e g
r
o
u
p
re
gar
d
i
n
g i
t
s
l
o
cat
i
on a
n
d
m
e
m
b
ers i
n
t
h
e gr
ou
p. T
h
i
r
d,
a node m
a
y
b
e
a
m
e
m
b
er of several
g
r
ou
ps.
Ho
weve
r, t
h
e
no
des
have se
n
d
t
h
e
packets
to t
h
e
me
m
b
ers in
t
h
e gr
oup
, ev
en
i
t
is no
t a m
e
mb
er of
a
gr
oup
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
18
57
–
1
865
1
858
Ove
r
t
h
e last
decade
,
researchers propose
d
se
veral m
u
lticast routing
pr
ot
ocols for
M
ANETs for
effect
i
v
e m
u
l
t
i
m
e
di
a co
m
m
u
n
i
cat
i
on. M
o
re
im
port
a
nt
l
y
t
h
ese ro
ut
i
ng
pr
ot
oc
ol
s can be
cat
egori
z
e
d
i
n
t
o
t
r
ee-
base
d an
d m
e
sh-
b
ase
d
r
o
ut
i
n
g
pr
ot
oc
ol
s.
H
o
we
ve
r,
ot
he
r
m
u
lt
i
cast
rout
i
n
g
p
r
ot
ocol
s
ar
e al
so a
v
ai
l
a
bl
e, w
h
i
c
h
is o
u
t
o
f
scope o
f
th
is
p
a
p
e
r. First, t
h
e tree b
a
se
d
m
u
ltic
ast rou
ting
proto
c
o
l
s m
a
in
tai
n
s a si
n
g
l
e
p
a
t
h
an
d
establishes a
s
h
are
d
m
u
lticast routing
tree
to tra
n
sm
its the pac
k
ets
from
source t
o
recei
vers
in a m
u
lticast
g
r
ou
p. Th
e m
a
in
id
ea
b
e
h
i
nd
th
ese pro
t
o
c
o
l
s is to
m
a
in
tain
me
m
o
ry for their ch
ildren in
stead
of all th
e
n
o
d
e
s.
Add
itio
n
a
lly, th
ese pro
t
o
c
o
l
s d
o
no
t prov
ide su
fficien
t
rob
u
s
t
n
ess
d
u
e
t
o
th
e li
m
ited
b
a
ndwid
th
effi
cien
cy.
One
of t
h
e t
r
ee based m
u
l
t
i
cast
rout
i
n
g
pr
ot
oc
ol
i
s
M
A
O
D
V
[
3
]
.
Whi
l
e
, M
e
s
h
base
d m
u
l
t
i
cast
rout
i
n
g
protoc
ols esta
blishes a m
e
sh
networ
k a
n
d maintains m
u
ltiple paths bet
w
e
e
n sources to
receivers. Due
to the
m
u
lt
i
p
l
e
pat
h
s,
m
e
sh based
m
u
lt
i
cast
i
ng i
s
m
o
re sui
t
a
bl
e
for
fre
que
nt
l
y
chan
gi
n
g
t
o
p
o
l
o
gi
cal
envi
r
o
nm
ents
and
pr
o
v
i
d
e m
o
re r
o
bu
st
ness
.
PUM
A
a
nd
O
D
M
R
P [
4
]
are t
h
e ro
ut
i
ng
pr
ot
oc
ol
s t
h
at
fal
l
s
un
der m
e
sh base
d
ro
ut
i
n
g pr
ot
oc
ol
s.M
o
re
ove
r,
i
n
spi
t
e
of t
h
e ro
ut
i
n
g i
ssue m
a
ny
m
obi
l
e
adh
o
c net
w
o
r
k
appl
i
cat
i
ons r
e
qui
re
s
vari
ous m
u
lticast routing
protocols that
nee
d
to
ope
rate
c
o
rrectly even i
n
hostile
environm
ent. Because the
M
ANET
S
a
r
e
m
o
re vul
nera
bl
e t
o
di
ff
ere
n
t
r
out
i
n
g at
t
a
c
k
s
wo
rm
hol
e,
bl
ack
hol
e,
r
u
s
h
i
n
g at
t
ack
, m
a
n i
n
t
h
e
m
i
ddl
e at
t
ack,
et
c., d
u
e t
o
i
t
s
i
nhe
ri
t
e
d cha
r
act
eri
s
t
i
c
sof M
ANE
Ts. T
h
i
s
pape
r
pr
op
oses
a no
vel
m
e
t
hod t
o
secu
re th
e m
u
lticast ro
u
ting
pro
t
o
c
o
l
ag
ain
s
t
man
in
th
e
m
i
ddle attack in MANE
Ts. Furt
her, we also a
n
a
l
yze
d
with
v
a
ri
o
u
s
perfo
r
m
a
n
ce m
e
trics su
ch
as throug
hpu
t,
PDF, con
t
ro
l
o
v
e
rhead
an
d to
tal ov
erh
e
ad
.
Th
is
p
a
per
is categ
o
r
ized
in
to
several sectio
n
s
. Sectio
n II
d
e
scri
b
e
s t
h
e rel
a
t
e
d
w
o
r
k
re
gar
d
i
n
g t
h
i
s
pape
r. Sect
i
on
II
I
expl
ai
n
s
t
h
e
m
u
l
t
i
cast
ro
ut
i
n
g
p
r
ot
ocol
s
,
sec
t
i
on
IV
ex
pl
ai
ns t
h
e m
a
n i
n
t
h
e m
i
ddl
e at
t
ack,
sect
i
o
n
V
m
a
i
n
l
y
foc
u
se
d o
n
securi
t
y
t
h
ro
u
g
h
EC
GD
H [5]
and fi
nal
l
y
sect
i
on VI e
xpl
ai
ns t
h
e sim
u
l
a
t
i
on re
sul
t
s
of
PUM
A
rou
ting
p
r
o
t
o
c
o
l
with leg
itimate, un
d
e
r attack
an
d after EC
GDH security.
2.
RELATED WORK
The e
x
i
s
t
i
ng
wo
rk
s m
a
i
n
l
y
foc
u
si
ng
o
n
no
rm
al
rout
i
n
g
pr
oced
u
r
e.
B
u
t
t
h
e
nat
u
r
e
s o
f
g
r
ou
p
com
m
uni
cat
i
on
pose
m
a
ny
chal
l
e
ng
es t
o
t
h
e
real
w
o
rl
d.
In
t
h
i
s
s
ect
i
o
n
we
descri
be
di
f
f
ere
n
t
m
e
tho
d
s
o
f
m
u
lt
i
cast
rout
i
n
g
p
r
ot
oc
ol
s.
R
a
vi
n
d
ra
Vai
s
h
a
m
p
ay
an et
al
pr
o
pose
d
a
P
U
M
A
di
rec
tin
g co
nv
en
tio
n it increases
hi
g
h
i
n
f
o
rm
ation c
o
nvey
a
nce
pr
op
o
r
t
i
on
wi
t
h
rest
ri
ct
e
d
co
nt
r
o
l
ove
r
h
ead
fu
rt
he
rm
ore i
n
crease hi
ghe
r b
u
n
d
l
e
co
nv
eyan
ce propo
rtion
contrasted
with
o
t
h
e
r
m
u
ltica
s
t steering
co
nv
en
tio
ns.
Men
a
k
a
Pu
shp
a
and
K.
Kat
h
i
r
a
v
a
n
pr
o
pose
d
a
n
s
w
ers f
o
r t
w
o i
n
t
e
ri
or
assa
u
lts in
p
a
r
ticu
l
ar guar
d
dog
based in
fo
r
m
atio
n
bu
nd
l
e
dr
o
p
assa
ul
t
r
ecog
n
i
zabl
e
pr
oo
f a
nd M
A
p
a
rcel
m
a
nuf
acture as
sault. El
izabeth M. R
oyer et al., propos
es
Multicast Ad hoc On-i
nterest Distan
ce Vect
or steeri
ng c
o
nvention (M
AODV
) [6
] is an
o
n
-in
t
erest
m
u
lticast
di
rect
i
n
g c
o
n
v
e
nt
i
o
n
t
h
at
dev
e
l
ops
a
c
o
mmon conveyance
tree
to
bo
lster nu
m
e
ro
u
s
send
ers and
co
llecto
r
s
i
n
a
m
u
lticast ses
s
ion. To
give i
d
eal corres
p
ondence ca
p
aci
t
y
, a di
rect
i
n
g c
o
n
v
e
n
t
i
on
f
o
r
suc
h
a dy
nam
i
c sel
f
-
begi
nni
ng syste
m
m
u
st be e
q
uipped fo
r unic
ast, telecast, and m
u
lticast.
B
r
oa
dene
d A
d
-
h
oc O
n
-
D
e
m
and Di
st
anc
e
Vect
or R
o
u
t
i
ng (
A
O
D
V
)
[7]
,
a cal
cul
a
t
i
on f
o
r t
h
e
ope
ration of s
u
ch specially appoi
nted
system
s, to offer
novel
m
u
lticas
t ca
pacities which
take after norm
ally
fr
om
t
h
e way
AO
D
V
b
u
i
l
d
s
up
uni
cast
c
o
u
r
ses.
AO
D
V
m
a
nufact
ures
m
u
lticast trees
as require
d
(i.e., on-
interest) to ass
o
ciate
m
u
lticas
t bunc
h indi
viduals. Control of the
m
u
lticast
tree
is circulated so that there
is no
si
ngl
e pu
rp
ose
of di
sap
p
o
i
n
t
m
ent
.
AO
D
V
gi
ves
ci
rcl
e
fre
e courses to
both unicast an
d m
u
lticast, even while
repairi
ng
broken connections
.
W
e
inc
o
rporate an a
ssessment philosophy and
repr
oduct
ion re
sults to
accept
th
e r
i
gh
t and
pr
odu
ctiv
e op
eratio
n
of
the
AODV calculati
on. Yunjung
Yi
et al., propose
d ODMRP. It is a
cro
s
s section
based
,
as opp
osed
to
a trad
itio
n
a
l tree-b
a
sed
,
m
u
lt
icast p
l
o
t
and
u
tilizes a
send
ing
bun
ch id
ea
(ju
s
t a su
b
s
et
o
f
hu
b
s
adv
a
nces th
e m
u
ltic
ast p
a
rcels
b
y
mean
s o
f
ch
eck
e
d
flood
ing
)
.
It app
lies on
-i
n
t
erest
m
e
t
hod
ol
o
g
y
t
o
po
we
rf
ul
l
y
assem
b
l
e
courses a
n
d ke
ep u
p
m
u
l
t
i
c
ast
bu
nch e
n
r
o
l
l
m
e
nt
. OD
M
R
P i
s
ap
pro
p
riate for sp
ecially app
o
i
n
t
ed
rem
o
t
e
syste
m
s wit
h
v
e
rsatile h
o
sts wh
ere d
a
t
a
tran
sfer capacity is
rest
ri
ct
ed,
t
o
po
l
ogy
c
h
an
ges
m
u
ch of
t
h
e t
i
m
e
and
q
u
i
c
kl
y
,
an
d
fo
rce i
s
obl
i
g
e
d
.
Felip
e Tellez e
t
al g
a
v
e
an
swers for an
elliptic b
e
n
d
cryp
t
o
syste
m
(ECC)
[8
] is su
itab
l
e to
Ad-Ho
c
sy
st
em
s and i
t
adeq
uat
e
t
o
g
u
a
rd
, di
st
i
n
gui
s
h
,
keep a
w
ay
f
r
om
t
h
e
W
o
rm
hol
e assa
ul
t
s
.
Vadha
d
iyaJanki et al
g
i
v
e
security t
o
jo
i
n
th
e i
n
corpo
r
ating
m
o
du
le, in
t
h
at to
u
tilize d
i
stin
ctiv
e so
rts
o
f
calcu
latio
n
s
su
ch
as RSA,
MD 5
,
SHA-1, and
o
t
h
e
r en
cryp
tion d
ecod
i
ng
ca
lcu
l
atio
n
an
d ad
d
ition
a
lly steerin
g
calcu
latio
n.
Vai
d
e
h
i
P
an
wal
a
et
al
pr
op
os
ed
o
n
i
n
t
e
rest
and
be
ne
ficiary started approach
m
u
lticast steering convention
called
Ad
ap
tive Mu
lticast
to
in
crease b
e
tter Qu
ality o
f
Serv
ice in
W
i
reless Networks. Yog
e
sh
Jo
sh
i et al
p
r
op
o
s
ed
and
ex
ecu
ted
a
no
vel way to
d
eal
with
settle
m
a
n
in
t
h
e cen
t
er
assau
lt ov
er SSL wh
ich u
tilizes th
e
au
th
en
tic site URL. To
h
a
nd
le su
ch
assaults we propo
se h
a
sh
ing
th
e clien
t
secret wo
rd
with
th
e gen
e
ral
p
opu
latio
n
k
e
y
o
f
t
h
e serv
er's ad
v
a
n
c
ed
au
then
ticatio
n
.
Zh
en
Ch
eng
et al p
r
op
o
s
ed
th
e calcu
l
atio
n
for ellip
ti
c
b
e
nd
D
i
ff
ie-
H
ell
m
an
k
e
y tr
ade tak
i
ng
in
t
o
acco
un
t
D
N
A
tile self
-
g
et t
o
g
e
th
er
.
To
star
t with
w
e
g
i
v
e
t
h
e DNA
[9
] tile self-g
et to
g
e
t
h
er m
o
d
e
l to
figu
re th
e scalar d
u
p
licatio
n, an
d
t
h
en we can
effectiv
el
y ex
ecu
te t
h
e
Diffie-
Hell
m
a
n
k
e
y ex
ch
ang
e
o
v
e
r ellip
tic cu
rv
e [9] b
y
d
i
g
o
u
t
t
h
e resu
lt con
s
tituen
t of t
h
e scalar ex
pon
en
tiation
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
S
ecure Mu
lticast Rou
tin
g Pro
t
o
c
o
l
i
n
MAN
E
Ts Us
ing
Efficien
t
ECGDH Al
g
o
r
it
h
m
(Gop
i
Arepa
lli
)
1
859
3.
MULTI
CA
ST
RO
UTI
N
G
P
R
OTO
C
OLS
In
gene
ral
t
h
e m
u
lt
i
cast
rout
i
ng
pr
ot
oc
ol
s u
s
ed i
n
m
obi
l
e
ad h
o
c net
w
o
r
ks are
br
oadl
y
cl
assi
fi
ed i
n
t
o
t
w
o
br
oa
d cat
e
g
o
r
i
e
s
one i
s
T
r
ee ba
sed m
u
l
t
i
cast
ro
ut
i
ng
p
r
ot
ocol
a
n
d
ot
her i
s
M
e
s
h
ba
sed m
u
l
t
i
cast
rout
i
n
g
pr
ot
oc
ol
.
3.
1.
Tree base
d m
u
lticas
t r
o
uting pr
otoc
ol
Tree b
a
sed
m
u
l
ticast p
r
o
t
o
c
o
l
m
a
in
tain
s sh
ared
m
e
d
i
u
m
with
a sin
g
l
e link to
estab
lish
com
m
uni
cat
i
on bet
w
ee
n s
o
u
r
ce t
o
dest
i
n
at
i
o
n
.
The e
x
am
pl
es of Tree
bas
e
d m
u
l
t
i
cast
prot
oc
ol
s are A
M
R
I
S
[1
0]
a
n
d
M
A
O
D
V
.
Here
, i
n
t
h
i
s
pape
r
we
c
hos
en
M
A
OD
V t
o
c
o
m
p
are
wi
t
h
m
e
sh
bas
e
d
pr
ot
oc
ol
s.
So
we
will discuss the MAODV.
MAODV is a
receive
r i
n
itiate
d tree
base
d prot
ocol a
n
d
it is the e
x
tension of
AO
D
V
ro
ut
i
n
g p
r
ot
ocol
. M
A
O
D
V
[1
0]
i
n
heri
t
e
d t
h
e c
o
nt
r
o
l
m
e
ssages l
i
k
e R
out
e R
E
Qest
(R
R
E
Q
)
, R
o
ut
e
R
E
Pl
y
(R
R
E
P
)
, M
u
l
t
i
cast
A
c
Ti
vat
i
ons
(M
AC
T)
an
d
G
r
ou
P
Hel
l
o
(G
R
P
H)
f
r
om
A
O
D
V
[
11]
pr
ot
oc
ol
.
Connection process: MAODV creates
ro
ut
es [1
2]
o
n
dem
a
nd
. S
o
u
r
ce
no
de i
n
ject
ed t
o
br
oa
dcast
t
h
e
R
R
E
Q
packet into
network to es
ta
blishing the connection w
ith rec
e
iver. Receive
r node is unicas
ted the RREP packet
to the sender by sa
m
e
forwarded
path
. Se
nder sends MAC
T
packet to the
receiver to int
i
m
a
te
m
u
lticas
t path
is established
betwee
n them
. Initial node in the group act
s as a controlle
r of the
gr
oup
and als
o
res
p
onsible
for m
a
in
tain
in
g
an
d broad
castin
g
th
e gro
u
p
sequ
en
ce
nu
mb
ers to
m
u
ltica
s
t g
r
ou
p.
N
o
d
e
s id
en
tifies th
e group
l
eader
by
usi
n
g GR
PH
. T
h
e
m
a
i
n
goal
of
M
A
O
D
V
i
s
t
o
b
u
ild th
e tree after co
m
p
letio
n
o
f
m
u
lticast
n
e
two
r
k
.
In
M
A
ODV, cont
roller node
m
a
intain
s up to
date inform
ation
of m
u
lticast tree because if any
link is
d
e
stro
yed
i
n
a g
r
ou
p
t
h
en the p
a
th
w
ill b
e
lo
st. MAODV
is vu
ln
er
ab
le f
r
o
m
m
a
n
in
th
e m
i
d
d
l
e attack
. I
t
br
oa
dcast
ro
ut
e requ
est
pack
et
i
n
t
o
t
h
e net
w
o
r
k
,
n
oxi
ou
s
no
des are p
r
e
s
ent
i
n
t
h
e co
m
m
uni
cat
i
on p
a
t
h
and
obs
erves t
h
e
data flow.
Due t
o
this
attack, t
h
e pe
rf
orm
a
nce
of
M
A
OD
V i
s
deg
r
a
d
ed
sl
i
ght
l
y
.
3.
2.
Mesh
based m
u
lticas
t r
o
utin
g pr
otoc
ol
M
e
sh
based
m
u
lt
i
cast
rout
i
n
g
p
r
ot
oc
ol
m
a
i
n
t
a
i
n
s m
u
lt
i
p
l
e
pat
h
s
an
d f
o
rm
s a
m
e
sh
net
w
or
k.
Exam
pl
es of m
e
sh based m
u
l
t
i
cast
rout
i
n
g pr
ot
ocol
s ar
e ODM
R
P
an
d PUM
A
. T
h
e
On-
D
em
and
M
u
l
t
i
cast
R
out
i
n
g Pr
ot
oc
ol
(O
DM
R
P
) i
s
a sou
r
ce st
art
e
d o
n
-i
nt
erest
l
a
t
t
i
ce based st
eeri
n
g co
nve
nt
i
o
n
.
O
D
M
R
P [
13]
i
s
works for both unicast and multicast exchanges. T
h
e as
s
o
c
i
at
i
on st
rat
e
gy
of
ODM
R
P
c
o
m
p
ri
ses t
w
o st
ages
like piggybac
king those a
r
e s
o
licitation stage and a
n
s
w
er
stag
e. In
th
e
so
licitatio
n
st
age sender s
u
rge
s
joi
n
que
st
i
o
n
pa
rce
l
i
n
t
o
t
h
e sy
st
em
. The
b
u
n
d
l
e
achi
e
ves t
h
e
nei
g
h
b
o
r
i
n
g
h
ubs
f
u
rt
herm
or
e su
rge
s
fr
om
t
hose
h
u
b
s
lastly ach
iev
e
s th
e recipien
t h
u
b
.
After d
e
stin
ati
o
n
hub
g
e
ts th
e jo
in
in
qu
iry p
a
rcel
th
en
it p
r
odu
ces th
e
joi
n
ta
ble. J
o
in table c
o
m
p
rises m
u
lticast bunc
h a
d
dr
es
s, successi
on
of source
a
d
dres
s
and
neighboring hub
address
,
jum
p
check. On the off cha
n
ce th
at an
y h
u
b
g
e
ts jo
in
tab
l
e th
en
i
t
ch
eck
s th
e follo
wing
hu
b
location
o
f
o
n
e
o
f
t
h
e passag
e
is its lo
catio
n
th
en
it d
i
stin
gu
ish
e
s it is in
a sen
d
i
ng
way to
a so
urce hu
b. After
th
at it
adva
nces
join table to
ne
xt jum
p
hubs
. Eac
h
tim
e
networ
k hubs kee
p
up
the brea
kthrough data.
Be
that
as
it
may, this directing convention is defe
nseles
s against
m
a
n
in
th
e cen
ter assau
lt. As a resu
lt o
f
th
e
o
p
e
n
n
e
ss
and
ab
sence
o
f
t
r
ust
e
d
po
we
r
M
I
TM
assa
ul
t
i
s
pr
o
p
elled a
n
d a
d
just t
h
e sys
t
e
m
execution
measurem
ents.
The ot
her m
e
sh base
d m
u
l
t
i
cast
rout
i
n
g
pr
ot
oc
ol
i
s
Pr
ot
oc
ol
fo
r U
n
i
f
i
e
d M
u
l
t
i
cast
i
ng t
h
ro
u
g
h
An
n
o
u
n
cem
ent
s
(P
UM
A
)
[1
4]
i
n
whi
c
h Pa
nt
her i
s
cr
oss sec
t
i
on
base
d el
e
m
ent
col
l
ect
or
st
art
e
d a
p
p
r
oac
h
a
n
d
b
ack
i
n
gs to
sen
d
m
u
lticast
in
form
at
io
n
allo
ted
to
a g
i
v
e
n m
u
lt
i
cast
gat
h
er and
do
n'
t
requi
re sepa
rat
e
uni
cast
d
i
recting
system
s
i
n
ce it g
o
e
s abo
u
t
as
bo
th
m
u
l
ticast an
d
un
icast. Jag
u
a
r
u
tilizes a co
n
t
ro
l m
e
ssag
e
to
co
n
t
ro
l
for e
v
ery
one
of its
operations, i.e. m
u
lticast declar
atio
n b
u
ndle (M
A
P
).
E
v
ery
M
A
P det
e
rm
ines
arran
g
e
m
e
nt
num
ber, b
unc
h
ID (l
ocat
i
o
n o
f
t
h
e gat
h
e
r
i
n
g
)
,
ce
nt
er
I
D
(lo
catio
n of th
e
cen
ter), sep
a
ratio
n
to
t
h
e center to
h
u
b
s
(
bou
n
c
e
tally)
,
w
o
r
k
par
t
h
a
il (
e
ith
er Tr
ue or
False)
, an
d
a gu
ardian
hub
th
at ex
pr
esses t
h
e so
ugh
t
nei
g
hb
o
r
t
o
achi
e
ve t
h
e cent
e
r. Fres
he
r M
A
parcel
ha
ve a hi
g
h
er s
u
ccessi
on
num
ber t
h
a
n
g
o
i
n
g bef
o
re
M
A
P
sent
by
t
h
e sa
m
e
cent
e
r. Tak
i
ng i
n
t
o
acco
u
n
t
t
h
e dat
a
co
n
t
ai
ned i
n
s
u
ch
cont
rol
b
u
ndl
e
s
, h
ubs
pr
o
g
re
ssi
vel
y
cho
o
se t
h
e ce
n
t
ers, deci
de t
h
e cou
r
ses f
o
r n
o
n
-
part
ag
gre
g
at
e hu
bs t
o
m
u
l
t
i
cast
bunc
h, t
e
l
l
about
t
h
e
j
o
i
n
i
n
g
or l
eavi
n
g i
n
t
h
e cr
oss sect
i
o
n am
ass and k
eep u
p
t
h
e l
a
t
t
i
ce sy
st
em
of t
h
e gat
h
e
r
i
n
g.P
U
M
A
ha
ve 5
fu
nct
i
o
n
s
to
m
a
in
tain
m
e
sh
and
conn
ectiv
ity p
r
o
c
ed
ure th
o
s
e areC
o
nnectiv
ity List
reco
rd
and
tran
smissio
n
o
f
M
u
lticast
Anno
un
cem
en
ts, Mesh
Estab
l
ish
m
en
t an
d
Main
ten
a
n
c
e,
C
o
re El
ect
i
o
n
pr
ocess
,
F
o
r
w
ardi
ng M
u
l
t
i
cast
Dat
a
Packets a
n
d Re
cycling Se
quence Num
b
ers.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
18
57
–
1
865
1
860
Figure
1. Man
in the Mi
ddle a
ttack Scenario
4.
MODELING
A
MAN IN T
H
E MIDDLE
ATTACK
In t
h
e m
a
n in t
h
e m
i
ddle attack
,
t
h
e attacke
r
can
put him
s
e
l
f in
t
h
e m
i
ddl
e of
t
h
e c
o
m
m
uni
cat
i
o
n
by
i
m
p
e
rson
ating
b
o
t
h
th
e sou
r
ce n
o
d
e
and
d
e
stin
atio
n
n
o
d
e
.
Let u
s
illu
strat
e
with
an
ex
am
p
l
e h
o
w
MITM [15
]
can be l
a
unc
he
d i
n
P
U
M
A
r
o
ut
i
ng
p
r
ot
ocol
.
Here
, an M
I
T
M
no
de ca
n se
nd M
A
wi
t
h
i
t
s add
r
ess t
o
an
ot
he
r
node in the group to im
personate
the receiving node, Attack
er node can
m
odify the hop count data as it has
th
e shortest p
a
th
to
th
e d
e
stinatio
n
b
y
send
i
n
g a MA
t
o
the source
node
, The att
acke
r
repeatedly
se
nds
MA
packet t
o
t
h
e s
o
urce
node
wit
h
its ra
dio ra
nge m
o
reove
r, t
h
is m
a
licio
u
s
nod
e
d
o
no
t
forward th
e M
A
p
a
ck
ets
to the s
o
urce
node
whic
h
was receive
d
from
its inte
r
m
ediate node
.
Whic
h m
a
y le
ads to t
h
e c
o
nnection
failu
re.
Hen
ce, th
e fun
c
tion
a
lities o
f
PUM
A
m
a
k
e
s
m
o
re v
u
l
n
e
rab
l
e to
laun
ch
MITM attack
, wh
i
c
h
m
a
y
deg
r
a
d
e t
h
e
pe
rf
orm
a
nce o
f
t
h
e P
U
M
A
r
o
ut
i
ng
p
r
ot
ocol
.
5.
SECU
RI
NG
MULTI
CA
ST
GR
OU
P
CO
MM
U
N
IC
AT
ION T
H
R
O
U
GH E
C
G
D
H
5.
1.
Revi
ew
of
E
l
l
i
pti
c
Cur
v
e Cr
ypt
o
gr
aph
y
Ellip
tic cu
rv
e is a two
d
i
m
e
n
s
io
n
a
l curv
e. Th
e stand
a
rd
curv
e eq
u
a
tion
is y
2
=x
3
+ax+b with
special
constraint 4a
3
+2
7b
3
≠
0
.
On
e of
th
e pu
b
lic k
e
y
cryp
tog
r
aphy
m
ech
an
ism
i
s
Ellip
tic cu
rve cryp
t
o
graph
y
[15
]
and its use
has
been increa
se
d trem
endousl
y
in recent
years beca
use the
use of la
rge
r
key size in re
maining
p
u
b
lic k
e
y m
e
ch
an
ism
s
lik
e
RSA,
d
i
g
ital sig
n
a
tures,
Diffie-h
e
llm
an
, etc
.
Ellip
tic cu
rve cryp
to
system
s
[8
]
pr
o
v
i
d
e m
o
re
effi
ci
ency
i
n
c
o
m
put
at
i
ons a
nd
o
ffe
r st
ro
ng
equ
i
valen
t
secu
rity with
smaller k
e
y sizes. The
resources lik
e
b
a
ndwid
th, st
orag
e cap
acity,
p
r
o
cessi
n
g
sp
eed
are
u
s
ed m
o
re in Ellip
tic cu
rv
e cryp
tograp
h
y
.
EC
C
have t
w
o
fi
el
ds, t
h
ose a
r
e pri
m
e Gal
o
i
s
Fi
el
d (p
) an
d
bi
nary
e
x
t
e
nsi
on
Gal
o
i
s
Fi
el
d GF
(2
m
).
Pri
m
e fi
el
d us
es al
l
real
num
bers
,
rat
i
onal
n
u
m
b
ers an
d
co
m
p
l
e
x num
ber
s
. B
i
na
ry
fi
el
d
i
s
use
d
t
o
cal
cul
a
t
e
key
s
i
n
bi
nary
fo
rm
at
. EC
C have
so
m
e
efficient alg
o
rit
h
m
s
for fi
nite
field operations s
u
ch a
s
addition,
m
u
lt
i
p
l
i
cat
i
on and i
nve
rsi
o
n.
These
s
p
ecia
lized algorithm
s
are evalua
te
d
with
th
e
h
e
lp
o
f
th
e
d
i
screte
lo
g
a
rith
m
s
(b
y
m
o
d
op
eration
s
). Ellip
tic C
u
rv
e do
m
a
in
p
a
ram
e
ters are
u
s
ed
to
rep
r
esen
t th
e ellip
tic
cu
rve
cryptogra
p
hy. The param
e
ters
are
T= { E
(
F
p
),
n,
a,
b,
G,
p, m
i
,
,
}
E(F
p
)
:
Ellip
tic cu
rv
e eq
u
a
tion
N
: O
r
d
e
r
of
g
r
ou
p
a, b
: Curve
c
o
efficients
G
:
Gr
ou
p ge
nerat
o
r
p
o
i
n
t
(G
x,
G
y
)
P
:
Prim
e b
a
se po
in
t p
ϵ
E(
F
p
)
m
i
:
i
-
t
h
gr
ou
p m
e
m
b
er i
ϵ
[1,n]
:
Private Secret key
of
m
i
( a ra
nd
om
i
n
t
e
ger
)
:
Pub
lic
k
e
y o
f
m
i,
calcu
lated
th
rou
g
h
scalar
m
u
l
tip
licatio
n
o
p
e
ration
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
S
ecure Mu
lticast Rou
tin
g Pro
t
o
c
o
l
i
n
MAN
E
Ts Us
ing
Efficien
t
ECGDH Al
g
o
r
it
h
m
(Gop
i
Arepa
lli
)
1
861
5.
2.
Diffie Hellman
Al
g
o
r
i
t
hm Using
Elliptic Curv
e Cryptog
r
aphy
M
u
l
t
i
cast
i
ng i
s
a gr
ou
p c
o
m
m
uni
cat
i
on m
echani
s
m
.
T
o
secure these
m
u
lticast communications
,
gr
o
up
key
m
a
nagem
e
nt
key
excha
nge m
e
t
h
o
d
s are u
s
e
d
. O
n
e o
f
t
h
e
wel
l
-
k
n
o
w
n m
u
lt
i
cast
i
ng rout
i
n
g
protoc
ols is PUMA, it is a receiver initiated a
p
proac
h
li
kewise elliptic
curve
grou
p
diffie
-
hellm
an is also
receiver initiated a
p
proach it means ne
w
node acts as a
group c
o
ntroller. So we
im
ple
m
ented Elliptic curve
g
r
ou
p d
i
ffie-h
ell
m
an
m
ech
an
ism
to
p
r
ov
id
e secu
rity fo
r PUMA rou
tin
g pro
t
o
c
o
l
.
Algorithm
:
El
liptic Cur
v
e
Diffie-Hellm
an
Step 1
:
N
←P
S
and
N
←P
S
Step 2
:
Calculate
N
→P
S
N
→P
S
P
S
←
P
S
∗G
P
S
←P
S
∗G
Step 3
:
N
N
Step 4
:
N
→S
N
S
N
←P
S
∗P
S
Step 5
:
N
N
Step 6
:
N
→S
N
S
N
←P
S
∗P
S
Step 7
:
S
N
,
→
S
N
S
N
el
se
got
pr
o
b
l
e
m
i
n
com
put
a
t
i
o
n
The ab
o
v
e al
g
o
ri
t
h
m
generat
e
s secret
key
f
o
r t
w
o
part
y
com
m
uni
cat
i
ons. Let
us
Assu
m
e
Node
‘
a
’
N
a
an
d n
ode ‘
b
’
N
b
wa
nts to c
o
mm
unicate securely through a secret
key.
Firstly, N
a
an
d N
b
ra
n
dom
l
y
sel
ect
s
pri
v
ate keys
P
S
,
P
S
res
p
ect
i
v
el
y
.
Ne
xt
,
bot
h no
des
gene
rat
e
p
u
b
l
i
c
key
s
P
S
,P
S
by u
s
i
ng
gr
o
up
ge
nerat
o
r p
o
i
n
t
a
n
d
exc
h
an
gi
n
g
t
h
ei
r
pu
bl
i
c
key
s
i
n
t
o
eac
h ot
her a
f
t
e
r t
h
at
b
o
t
h
n
ode
s cal
cul
a
t
e
secret
key
s
S
N
,S
N
i
ndi
vi
dual
l
y
and
fi
nal
l
y
,
bot
h m
u
st
ge
ne
rat
e
eq
ual
resul
t
s.
Algorithm
:
Joining of
new
n
o
de
in
to the multicas
t group
Step 1
:
N
←P
S
Step 2
:
Calculate
N
→P
S
P
S
←
P
S
∗G
Step 3
:
N
,
,
,
N
Step 4
:
Calculate
N
→P
S
,
P
S
,
←
P
S
∗P
S
N
→P
S
,
P
S
,
←
P
S
∗P
S
N
→S
N
,
,
S
N
,
,
←
P
S
∗S
N
,
Step 5
:
N
c
bro
a
d
casts i
n
term
ed
iate k
e
y to
N
a
, N
b
N
,
N
N
,
N
Step 6
:
Calculate
N
→S
N
,
,
S
N
,
,
←
P
S
∗P
S
∗P
S
N
→S
N
,
,
S
N
,
,
←
P
S
∗P
S
∗P
S
The a
b
ove
al
g
o
ri
t
h
m
sho
w
s
joi
n
i
n
g
of
ne
w
n
ode
i
n
t
o
t
h
e
m
u
l
t
i
cast
gr
o
u
p
.
S
u
pp
ose,
i
f
a
n
ode
N
c
wan
t
s to
j
o
i
n
in
a g
r
ou
p. First it sen
d
s
jo
in requ
es
t m
e
ssa
g
e
to
group
man
a
g
e
r.
Grou
p
m
a
n
a
g
e
r will g
r
an
t
perm
i
ssi
on t
o
new
n
o
d
e
N
c
.
no
de
N
c
selects one
private
key
P
S
a
n
d
gene
r
a
t
e
p
ubl
i
c
key
P
S
. Gro
up
man
a
g
e
r sen
d
s all
th
e in
termed
iate k
e
ys to
n
e
w
n
o
d
e
N
c
. After
receiving all
the keys from
GM, N
c
ac
ts as a
new
group m
a
nage
r and calc
u
late the
ne
w
keys by usi
ng
the interm
ediate keys whic
h
are receive
d
by old
GM
. Fi
nal
l
y
i
t
b
r
oa
dcast
t
h
e
key
s
i
n
t
o
rem
a
i
n
i
n
g
g
r
ou
p m
e
m
b
ers.
Th
ose
pe
rf
orm
com
put
at
i
ons
o
n
rec
e
i
v
ed
key
s
a
n
d
ge
ner
a
t
e
a ne
w
gr
o
u
p
key
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
18
57
–
1
865
1
862
Algorithm
:
Joining
of n n
e
w nodes in
to th
e multicast
gr
oup
Ro
und
i
∈
0
,
n
2
N
∗
|
∈
,
⋀
|
∈
,
,∗
…….
N
N
∗
………
………
.
N
In a l
a
r
g
e g
r
ou
p, i
f
a
no
de wa
nt
s j
o
i
n
i
n
t
h
at
gr
ou
p
pre
v
i
o
u
s
no
de se
nd
s a
l
l
t
h
e i
n
t
e
rm
ed
i
a
t
e
key
s
t
o
joi
n
i
n
g n
ode a
nd
br
oad cast
t
o
rem
a
i
n
i
n
g
gro
u
p
m
e
m
b
ers. The a
b
o
v
e
equat
i
o
ns re
prese
n
t
s
sen
d
i
ng a
nd
b
r
o
a
d
castin
g in ro
und
i.
Al
g
o
ri
thm
:
L
e
avi
n
g
the
n
o
d
e
fr
om
the m
u
l
t
i
c
ast
gr
ou
p
These t
w
o
t
o
pol
ogi
cal
pr
ot
ocol
s
are
di
f
f
e
r i
n
re
du
n
d
a
n
cy
o
f
t
h
e
p
a
t
h
s
bet
w
ee
n
so
urce
an
d
dest
i
n
at
i
o
n.
W
h
ereas t
r
ee
-
b
as
ed m
u
l
t
i
cast pr
ot
oc
ol
s [
15]
p
r
ovi
de o
n
l
y
a si
ngl
e
pat
h
an
d
m
e
sh-ba
s
ed
pr
ot
oc
ol
s
p
r
ov
id
e m
u
ltip
le p
a
th
s b
e
t
w
een
sou
r
ce no
de to
d
e
stin
ation
no
des. In
mu
lticastin
g
,
robu
stn
e
ss and
rel
i
ab
ilit
y
bot
h are
very
im
port
a
nt
a
nd
t
h
ese pa
ram
e
t
e
rs are
hi
g
h
i
n
m
e
sh base
d ne
t
w
o
r
ks
. Ex
am
pl
es o
f
m
e
sh base
d
pr
ot
oc
ol
s a
r
e [
16]
P
U
M
A
a
n
d
ODM
R
,
C
A
M
P
.
Alg
o
rithm
:
Step 1
:
N
N
Step 2
:
N
←n
e
w
P
S
Step 3
:
Calculates
N
→P
S
P
S
←n
e
w
P
S
∗G
Step4 :
N
br
o
a
d
castsintermedia
t
e
k
e
y
v
aluest
o
allgr
o
u
pno
des
Step5 :
Nod
e
sgener
a
t
e
sgr
o
u
pk
e
yusing
theirP
The a
b
ove
al
g
o
ri
t
h
m
di
scuss
e
s h
o
w
t
h
e
n
o
d
e
l
eaves
fr
om
the m
u
l
t
i
cast
grou
p.
N
o
de
N
wan
t
to
leave
fro
m
a
g
r
oup
First it sen
d
s
qu
it requ
est (QuitREQ) to
group
m
a
n
a
g
e
r N
gm
. Gr
o
up m
a
nag
e
r gra
n
t
pe
rm
issi
on
and c
h
a
nges
hi
s pri
v
at
e key
.
Next
,
Gr
o
up m
a
nage
r cal
cul
a
t
e
s pu
bl
i
c
key
,
i
n
t
e
rm
edi
a
t
e
key
s
and b
r
oadc
ast
to
al
l
gr
ou
p m
e
m
b
er
.
Gr
ou
p m
e
m
b
ers gene
rat
e
g
r
o
u
p
key
by
usi
n
g t
h
ei
r
pri
v
at
e key
s
.
If Gr
ou
p m
a
n
ager
w
a
n
t
s
t
o
l
e
ave
Step 1
:
N
pr
e
v
iousN
Step
2
:
Call leav
in
g
al
g
o
rith
m
Special case
:
Sup
p
o
s
e, if
g
r
o
u
p
m
a
n
a
g
e
r
w
a
n
t
s to
leave f
r
o
m
a g
r
oup it sen
d
s
qu
it r
e
qu
est to
pre
v
i
o
us
gr
o
u
p
m
a
nager
.
Ne
w G
r
o
u
p
m
a
nager
gra
n
t
p
e
rm
i
ssi
on a
nd c
h
a
nge
s hi
s
pri
v
at
e key
.
Ne
xt
, P
r
ese
n
t
gr
o
u
p
m
a
nager
cal
cul
a
t
e
s
bot
h
pu
bl
i
c
key
a
n
d
i
n
t
e
rm
edi
a
t
e
key
s
w
h
i
c
h
br
oa
dcast
t
o
al
l
gr
o
u
p
m
e
m
b
ers i
n
a
g
r
ou
p. Fi
n
a
lly, all th
egroup
me
m
b
ers
will g
e
n
e
rate
g
r
ou
p key u
s
ing
t
h
eir
p
r
i
v
ate k
e
ys.
5.
3.
Securing P
U
MA multic
ast
routing pr
otocol using E
C
GDH
PUMA is a receiver initiated appr
oach and a
l
so establishes a
m
e
sh network to comm
unicate a
m
ong
th
e
g
r
o
u
p
.
The attack
er will exp
l
o
it th
e
weakn
e
ss of
PUMA (a
n
ode can jo
in in
a group
withou
t an
y
constraint) to launc
h MITM a
ttack [17]. T
h
e
attacker
has
a
chancet
oe
nterintothe
g
r
o
u
p
an
dm
ay
drop
or
al
t
e
r t
h
e
p
ack
ets withou
t forward
i
n
g
to
th
eir n
e
ighb
oring
no
d
e
s. We p
r
op
osed
Ellip
tic cu
rv
e g
r
ou
p
d
i
ffie-hell
m
a
n
secu
r
ity
m
ech
an
ism to
p
r
o
t
ect
th
e
m
u
lticast
c
o
mm
u
n
i
catio
n
s
. I
n
th
is pr
oposed
wo
rk
, in
a g
r
ou
p
on
e node w
ill
be el
ect
ed as a gro
u
p
co
nt
r
o
l
l
e
r usi
n
g EC
G
DH [
1
8]
m
echani
s
m
,
whi
c
h i
s
di
scusse
d i
n
sect
i
on V.
We
al
so
p
r
op
o
s
ed
securely j
o
in
i
n
g and leav
ing
algorith
m
s
fo
r au
th
enticatio
n
of
m
o
bile n
o
d
e
s i
n
a
gr
oup
.Mor
eo
v
e
r, all
th
e nod
es in
a
g
r
ou
p
will commu
n
i
cate with
a si
n
g
l
e
g
r
ou
p
k
e
y
wh
ich
is g
e
n
e
rated
by all th
e g
r
o
up n
o
d
e
s.
Thi
s
pr
o
p
o
s
ed
EC
GD
H p
r
o
v
i
d
es bac
k
wa
r
d
and f
o
rwa
r
d se
crecy
whe
n
t
h
e no
des wa
nt
to j
o
i
n
o
r
l
eave
from
th
e g
r
ou
p.In
back
w
a
rd
secr
ecy, n
e
w
nod
e can
no
t o
b
t
ai
n past
com
m
uni
cat
i
on.
Whi
l
ei
n fo
rwa
r
d secrecy
leaving
node
cannot access
prese
n
t gr
oup comm
unication.
Hence
,
in both the ca
ses past and
prese
n
t
i
n
f
o
rm
at
i
on cann
o
t
be o
b
t
a
i
n
ed by
t
h
e n
ode
s. On t
h
e
ot
he
r
han
d
, i
f
any
o
f
t
h
e no
de m
i
sbeha
v
es
una
ut
ho
ri
zed
fu
nct
i
o
ns l
i
k
e
not
fo
r
w
ar
di
n
g
t
h
e
dat
a
pac
k
et
s t
o
t
h
e
nei
g
hb
o
r
i
n
g n
o
d
e a
n
d
gi
vi
n
g
fal
s
e
i
n
f
o
rm
at
i
on s
u
ch
as
sh
ortest
h
o
p
cou
n
t
. In
su
ch
cases th
e
group
co
n
t
ro
ller
will ob
serv
es and
d
i
scard
e
d
fro
m
th
e m
u
lt
i cast g
r
o
u
p
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
S
ecure Mu
lticast Rou
tin
g Pro
t
o
c
o
l
i
n
MAN
E
Ts Us
ing
Efficien
t
ECGDH Al
g
o
r
it
h
m
(Gop
i
Arepa
lli
)
1
863
6.
SIMULATION RESULTS
In
t
h
is p
a
p
e
r, we
co
m
p
are
PUMA with
4
param
e
te
rs: Th
ro
ugh
pu
t,
Packet d
e
liv
ery
fractio
n
,
C
o
n
t
rol
ove
rhead, T
o
ta
l ove
rhead wit
h
respect
to
Num
b
er of nodes
i
n
a gr
o
u
p
.
I
n
Fi
gu
re 2 sh
o
w
s
t
h
e gra
p
h of
p
acket
del
i
v
ery
fract
i
o
n
V
s
num
ber
o
f
no
des
i
n
a
g
r
o
u
p
f
r
o
m
5
to
10
0 resp
ectiv
ely u
n
d
e
r leg
iti
m
a
te,
m
a
n
in
th
e
middle attack
and after providing secu
rity ag
ain
s
t attack
situ
atio
n
s
. Legiti
mate situ
ati
o
n
s
g
i
v
e
s
h
i
gh
p
d
f
com
p
are t
o
re
m
a
i
n
i
ng si
t
u
at
i
ons
. B
u
t
ot
her
si
de m
a
n i
n
t
h
e
m
i
ddl
e at
t
a
ck de
gra
d
es t
h
e per
f
o
r
m
a
nce o
f
p
d
f
s
o
fo
r de
fen
d
in
g
this attack we
pr
op
ose a sec
u
rity
m
echanis
m
called ECGDH e
xplai
ned
in Section
V.
After
pr
o
v
i
d
i
n
g t
h
e
securi
t
y
t
o
P
U
M
A
r
o
ut
i
n
g
p
r
ot
ocol
,
t
h
e
pe
rf
orm
a
nce o
f
Packet
del
i
v
e
r
y
has i
n
crea
se
d a
n
d
g
i
v
e
n
b
e
tter
resu
lts co
m
p
ared to
m
a
n
in
t
h
e
mid
d
l
e attack
situ
atio
n
.
Fi
gu
re 3 s
h
ow
s t
h
e gra
p
h o
f
PUM
A
ro
ut
i
n
g
pr
ot
oc
ol
s t
h
r
o
ug
h
put
Vs n
u
m
ber of n
o
d
e
s
i
n
a gr
o
u
p
fro
m
5
to
1
00 with
ou
t attack
,
with
m
a
n
i
n
th
e m
i
d
d
l
e a
ttack
an
d
i
n
secu
rity. If th
roug
hpu
t is h
i
gh
th
en
it
indicates that
maxim
u
m
packets deli
ver t
o
receiver.
As
us
ually throughput is hi
ghe
r i
n
norm
al situatio
ns
but
i
t
i
s
degrade
d
whe
n
t
h
e at
t
a
c
k
has
ha
ppe
ne
d. EC
GD
H i
m
pr
o
v
es t
h
e t
h
ro
ug
h
put
per
f
o
r
m
a
nce even at
t
ack has
h
a
pp
en
ed
o
r
no
t.
ECGDH
secu
rity
m
ech
an
ism p
r
ov
id
es
better th
roug
hpu
t in
attack
scen
ari
o
.Th
r
ou
gh
pu
t is
d
i
rectly p
r
opo
rtio
n
a
l to
PDF.
If
PDF in
creases th
rou
ghp
u
t
also
in
creases.
Fi
gu
re
4 d
e
pi
ct
s t
h
e c
ont
r
o
l
o
v
er
hea
d
V
s
n
u
m
ber of
no
des
i
n
a g
r
ou
p.C
o
n
t
rol
o
v
e
r
hea
d
h
a
s i
n
crease
d
whe
n
we p
r
o
v
i
de secu
ri
t
y
m
echani
s
m
t
o
our
pr
ot
oc
ol
com
p
ared
to
legiti
mate an
d
no
rm
al situ
atio
n
s
.
In
no
rm
al
sit
u
at
ion c
o
nt
rol
ove
rhea
d m
a
i
n
t
a
i
n
s bet
t
e
r res
u
lts com
p
ared to
attack s
cenari
o
. Cont
rol ove
r
head
in
creases ev
en
th
ou
gh
we
prov
id
e t
h
e secu
ri
ty. Th
e m
a
in
ai
m
o
f
secu
rity
is con
t
ro
lling
t
h
e attack
s and
traffi
c
fl
o
w
s
o
c
ont
rol
o
v
er
hea
d
has i
n
crease
d
i
n
a
f
t
e
r
pr
ovi
di
n
g
se
curi
t
y
.
Fi
gu
re 5
depi
c
t
s abo
u
t
t
o
t
a
l
ove
r
h
ead
Vs n
u
m
b
er of n
o
d
e
s
i
n
a gro
u
p
.
C
ont
r
o
l
o
v
er
h
ead di
rect
l
y
pr
o
p
o
r
t
i
onal
t
o
t
o
t
a
l
o
v
er
he
ad.
I
f
co
nt
r
o
l
o
v
er
hea
d
has
i
n
creased th
en
to
tal ov
erh
e
a
d
also increa
se
s. Li
ke
cont
rol ove
r
he
ad, total overhead also
hi
gher in after sec
u
rit
y
com
p
ared to
leg
iti
m
a
te
an
d
attack
scen
ari
o
. Bu
t
i
n
co
nt
rast
t
o
c
ont
rol
o
v
er
hea
d
, t
o
t
a
l
o
v
er
he
ad
d
ecrease
s
when
num
b
er of no
d
e
s in
a gr
oup
in
cr
eases.
Fi
gu
re 2.
Pac
k
et
del
i
v
ery
frac
t
i
on of
P
U
M
A
W
i
t
h
an
d
W
ithou
t
M
ITM
A
ttacK
Vs
Secu
re
PUM
A
Fi
gu
re 3.
Th
r
o
ug
h
put
o
f
P
U
M
A
W
i
t
h
a
n
d
W
i
t
h
o
u
t
MI
TM A
ttacK
V
s
Secur
e
PUMA
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE
Vo
l. 6
,
N
o
. 4
,
Au
gu
st 2
016
:
18
57
–
1
865
1
864
7.
CO
NCL
USI
O
N
Mo
b
ile ad
ho
c
n
e
two
r
k
is an in
frastru
c
ture less n
e
t
w
or
k
th
at
h
a
s no
t
r
u
s
ted au
thority. In
su
ch typ
e
of
net
w
or
ks,
gr
o
up c
o
m
m
uni
cat
i
on i
s
one
of t
h
e i
m
p
o
rt
a
n
t
com
m
u
n
i
cat
i
ons f
o
r
vari
o
u
s ap
pl
i
cat
i
ons.
M
u
l
t
i
cast
i
ng
M
echani
s
m
can be a
p
pl
i
e
d t
o
achi
e
ve t
h
i
s
g
r
o
u
p
com
m
uni
cat
i
on.
In t
h
i
s
pape
r,
we
used
PUM
A
ro
ut
i
n
g pr
ot
oc
ol
whi
c
h pr
o
v
i
des bet
t
e
r res
u
l
t
s
com
p
ared
t
o
ot
her m
e
sh pr
ot
oc
ol
s.
H
o
we
ve
r, Thi
s
ro
ut
i
n
g
p
r
o
t
o
c
o
l
s su
ffers
fro
m
m
a
n
in
th
e m
i
d
d
l
e attack
.
To d
e
fen
d
th
is attack
, we propo
sed a
no
v
e
l m
e
th
od
Ellip
tic
C
u
r
v
e G
r
ou
p
Di
ffi
e
Hel
l
m
a
n (EC
G
D
H
)
.
Fi
nal
l
y
, we c
o
m
p
ared P
U
M
A
r
out
i
n
g
p
r
ot
ocol
un
der
no
rm
al
situ
atio
n
,
und
er attack
scen
ario
an
d d
e
fen
d
i
ng
with
EC
GDH secu
rity.
REFERE
NC
ES
[1]
R. Vaishampay
an and J. J.
Garcia-Lun
a-Acev
es, “Protocol for uni
fied multicasting through
announcements
(PUMA),
”
in
Proceed
ings of the IEEE Internati
onal Conferen
ce
onMobile Ad- Hoc and Sensor Systems
(
M
ASS
’04)
, 2004.
[2]
IEEE, “IEEE Standard 1363-200
0: Stand
a
rd specifi
cations for
pu
blic key
cr
y
p
tog
r
aph
y
,” IEEE, 2
000.
[3]
E.
Babu
,
et al.
,
“An Implementation and Performance Evalua
tion Study
of AODV,
MAODV,
R
AODV in Mobil
e
Ad hoc Networks,” vol. 4
,
pp
. 69
1-695, 2013
.
[4]
E. B
a
bu,
et al.
, “
A
n Im
plem
entati
on Anal
y
s
is
and
Evalu
a
tion
S
t
udy of
DS
R with I
n
act
ive DoS
Att
ack
in M
obil
e
A
d
hoc Networks,”
vol. 2
,
pp
. 501-5
07, 2013
.
[5]
F
.
Blak
e,
et al.
,
“Advances Ellip
tic Curves in
Cr
y
p
togr
aph
y
,” C
a
m-bridge Univer
sity
Press, 2005
.
[6]
J. Liu and J
.
Li,
“A better improvement on the
in
tegra
t
ed diff
ie-h
ellm
an-ds
a
ke
y
agreem
entpro
toc
o
l,”
In
ternationa
l
Journal of Netw
ork Security
, vol/issue: 11(2)
, pp
. 114–117, 2010.
[7]
E.
Babu
,
et al.
,
“Inspired Pseudo Biotic DNA based Cr
y
p
tograp
hic Mech
anism against Adap
tiv
e Cr
y
p
tograph
i
c
Attacks
,
”
In
tern
ational Journal of
Network
Secu
rity
, vol/issue: 1
8
(2), pp
. 291-30
3, 2016
.
[8]
E.
B
a
bu,
et al.
, “Light-Weighted
DNA-Base
d Cry
p
tographic
M
echanism Against Chosen Cipher Text A
ttacks,
”
Advanced Comp
uting and
Systems for Security,
Springer India, p
p
. 123-144
, 201
6.
[9]
Y.
Wang,
et al.
, “The Perform
ance of E
llip
tic Curve Based
GroupDiffie-Hel
l
m
a
n Protocols
for Secure Group
Communication over Ad Hoc
Networks,”
IEEE Internationa
l Conference on Communication
, vol. 5, pp. 224
3
–
2248, 2006
.
[10]
E. B
a
bu,
et al.
, “Light-Weighted DNA Based
H
y
br
id Cry
p
tographic Mechanism
Against Chosen Cipher Text
Attacks
,
”
Int
e
rnational Journal of
Information Processing
and Indexed
With
arXiv
, Indian Citation Index
,
201
5
.
ISSN-0973-821.
[11]
M.
J.
Moy
e
r,
et al.
,
“
A
S
u
rve
y
o
f
S
ecuri
t
y
Is
s
u
es
in M
u
l
tic
as
t Co
m
m
unication”,
I
EEE
Network
, p
p
. 12–23
, 1999
.
[12]
O. S. Badarneh
and M. Kadoch,
“
R
eview Arti
cle Mult
icast R
outing Protocols
in Mobile Ad Hoc Networks:
AComparative
Survey
and Tax
onomy
,
”
EU
RASIP Journal on Wireless Communications and
Networking
, 2009.
Article ID76404
7, 42
pa
ges doi: 10.1155/2009/76
4047.
[13]
S.
Kumar,
et al.
, “An Empirical Critique of On-
D
emand Routing
Protocols against Rushing Attack in MANET,”
International Jo
urnal of
Electr
ical and Computer Engin
eering
, vo
l/issue: 5
(
5), 201
5.
[14]
A.
Gopi,
et al.
,
“Designing an A
dversarial Mode
l Against R
e
active and
Proactive Routing Proto
c
ols in MANETS:
A Com
p
arative P
e
rform
ance S
t
ud
y,”
Internatio
nal Journal of Electr
ica
l
and Computer Engineering
, vol/issue:
5(5), 2015
.
Fi
gu
re 4.
R
o
ut
i
n
g
O
v
e
r
hea
d
o
f
P
U
M
A
W
i
t
h
and
W
i
t
h
out MIT
M
AttacK
Vs
Secure PUMA
Fi
gu
re 5.
Tot
a
l
o
v
er
hea
d
of P
U
M
A
Wi
t
h
an
d
W
i
t
h
out MIT
M
AttacK
Vs
Secure PUMA
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
S
ecure Mu
lticast Rou
tin
g Pro
t
o
c
o
l
i
n
MAN
E
Ts Us
ing
Efficien
t
ECGDH Al
g
o
r
it
h
m
(Gop
i
Arepa
lli
)
1
865
[15]
H.
Deng,
et al
.,
“Routing security
in wireless ad
hocnetworks,”
I
EEE Commun. Mag
., vol/issue:
40(10), pp. 70–7
5,
2002.
[16]
P. Sinha,
et
al
.
, “CEDAR: aCoreExtraction
Distribut
ed Ad ho
c Ro
uting
algorithm,”
IEEE INFOC
O
M’99
, 1999.
[17]
E.
B
a
bu,
et al.
, “An Implementation and Perfo
rmance Evaluation of Passive
DoS Attack on AODV Routing
Protocol in
Mobile Ad ho
c Networks,”
International
Journal o
f
Emerging
Trends
&
Technology in Computer
Scien
c
e
, vol/issue: 2(4)
, pp
. 124-
129, 2013
.
[18]
E.
Babu
,
et a
l
.
, “Efficien
t DNA-Based Cr
y
p
to
graphic Mechan
ism
to Defend and Detect Blackhole Attack
in
MANETs,” in
Proceedings of International Conf
er
ence on ICT for Sustainable Developmen
t
, Springer Singapor
e,
pp. 695-706
, 20
16.
Evaluation Warning : The document was created with Spire.PDF for Python.