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.
9
, No
.
5
,
Octo
ber
201
9
, pp.
4176
~
41
83
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
9
i
5
.
pp
41
76
-
41
83
4176
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Random
routin
g schem
e with m
is
leading
d
ea
d
ends
Ch
itr
a Raj
arama
1
,
Jagadee
s
ha
N
ara
sim
hamurth
y Su
gat
oo
r
2
,
Yerri Sw
am
y
T
3
1
Dep
ar
t
m
ent
of
I
nform
at
ion
Sci
e
nce
and
Engi
ne
e
ring,
NIE
Insti
tu
te
of
T
ec
hno
log
y
,
Indi
a
2
Dep
art
m
ent
of
El
e
ct
roni
cs
an
d
Com
m
unic
at
ion
,
PES
Instit
u
te
o
f
Technol
og
y
M
ana
gem
ent,
Ind
i
a
3
Dep
art
m
ent
of
Com
pute
r
Scie
n
ce
and
Engi
ne
ering
,
KLE
Inst
it
ut
e
of Te
chnol
og
y
,
India
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
N
ov
28
, 201
8
Re
vised
A
pr
7
,
201
9
Accepte
d
Apr
1
9
, 201
9
A
new
m
et
hod
of
sink
loc
ation
sec
urity
in
a
W
ire
le
ss
Sensor
Network
is
proposed
.
In
the
proposed
sche
m
e,
al
l
t
he
node
addr
esses
are
e
ncr
y
p
te
d
an
d
an
a
tt
a
cke
r
ca
nn
ot
determ
ine
the
real
sink
addr
e
ss
b
y
c
apt
ur
ing
the
p
ac
ke
ts
and
an
aly
z
ing
i
t
s
cont
en
ts
for
th
e
fin
al
d
esti
na
ti
o
n.
Th
e
m
ai
n
con
tri
buti
on
of
our
proposed
me
thod
is
to
use
ran
dom
routi
ng
sche
m
e
with
m
isle
adi
ng
de
ad
ends.
Th
is pr
ovi
des
sec
uri
t
y
aga
i
nst t
raf
f
ic a
n
a
l
y
s
is a
t
ta
ck
.
Ke
yw
or
d
s
:
M
isl
eading
dea
d
e
nds
R
andom
r
ou
ti
ng
S
ink l
ocati
on s
ecur
it
y
T
raffic
ana
ly
sis at
ta
cks
Copyright
©
201
9
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Chit
ra Raj
a
ram
a
,
Dep
a
rtm
ent
of
Inform
at
ion
Sc
ie
nce and En
gi
neer
i
ng,
NI
E
Insti
tute o
f
Tec
hnology,
My
su
ru 57
0018, Ka
rn
at
a
ka,
I
nd
ia
.
Em
a
il
:
chitra
m
anu
el
@yah
oo.
co.
in
1.
INTROD
U
CTION
A
sin
k
is a crit
i
cal
n
od
e i
n
a W
i
reless Se
nsor Netw
ork,
as
it
co
ll
ect
s
all
t
he
data f
ro
m
the sen
sors and
act
s
as
a
gate
way
to
the
I
nter
net
an
d
ot
her
netw
orks.
Thus
we
s
ho
uld
prov
i
de
th
e
highest
de
gree
of
anonym
i
ty
and
secur
it
y
to
the
sink
[1
]
f
ro
m
t
he
ad
v
e
rsar
ie
s.
The
m
ai
n
te
c
hn
i
qu
e
s
ad
opte
d
by
the
a
dver
sari
e
s
are
traff
ic
an
al
ysi
s
attack
and
packet
tracing
attack
.
In
this
work
we
pro
po
se
a
com
pr
e
he
ns
ive
pr
ot
ect
ion
against
these
two
at
ta
cks.
Traffic
analy
sis
and
pac
ket
traci
ng
at
ta
cks
are
basical
ly
passiv
e
at
ta
cks
by
adv
e
rsa
ries
w
he
re
they
li
ste
n
to
the
tra
ff
ic
flo
w
a
nd
the
n
de
du
ce
the
di
recti
on
to
ward
s
the
si
nk.
He
re
the
adv
e
rsa
ry
us
es
pack
et
traci
ng
te
chn
i
qu
e
to
reach
the
sin
k.
To
c
ounter
this
at
ta
ck
we
pro
po
se
the
R
ando
m
Rou
ti
ng
Sc
he
m
e
with
Mi
sle
adin
g
D
ea
d
E
nds
(RRSM
DE).Sev
e
ral
w
orks
hav
e
bee
n
pr
e
sented
t
o
pro
vi
de
sin
k
locat
ion
sec
ur
i
ty
us
ing
ra
ndom
ro
utin
g
an
d
oth
e
r
m
e
tho
d
s
[2
-
5].
I
n
[
1],
th
e
authors
use
f
ake
sin
ks
away
fr
om
the
real
sin
k
t
o
m
isgu
ide
t
he
pac
ket
traci
ng
at
ta
cker
.
But
the
nu
m
ber
of
fa
ke
s
i
nk
s
a
nd
th
ei
r
l
ocati
ons
a
re
fixe
d
pr
i
or
i
an
d
the
at
ta
cker
can
identify
th
ese
fak
e
sin
ks
after
certai
n
tr
ia
ls.
In
Zo
ne
ba
sed
Sin
k
Lo
c
at
ion
Pr
ivacy
Ro
utin
g
Protoc
ol
[
2],
the
auth
or
s
ha
ve
us
e
d
zo
ne
pa
rtit
ion
in
g
of
the
W
S
N
a
nd
f
ake
sin
ks
an
d
a
real
sink
a
re
l
ocated
in
each
z
one
.
A
sou
rce
node
of
a
zo
ne
tra
ns
m
it
s
pack
et
s
to
it
s
ow
n
fake
sink
s
as
well
as
it
s
real
sin
k.
I
n
this
case
al
so,
the
nu
m
ber
of
fa
ke
sin
ks
per
zo
ne
is
lim
i
te
d
and
their
locat
io
ns
are
pre
-
dete
rm
ined.
Once the z
ones are id
e
ntifie
d
by the att
ack
er,
f
ur
t
her
dete
ct
ion
of f
al
se s
ink
s a
nd
t
he
re
al
sin
k
are easy
for
the
adversa
ry.
In
[1
]
a
nd
[
2],
the
fa
ke
pac
ket
in
j
ect
ion
s
do
not
occ
ur
at
ever
y
ho
p,
bu
t
occ
ur
s
at
sp
eci
fie
d
intersect
io
n
no
des.
On
t
he
c
on
t
rar
y,
in
our
propose
d
m
eth
od,
t
her
e
is
no
li
m
i
t
on
th
e
fak
e
destina
ti
on
s.
Fo
r
eac
h
trans
m
issi
on
,
the
f
ake
destinat
io
ns
will
be
dif
fer
e
nt.
W
e
al
s
o
pro
vid
e
fa
ke
br
an
chi
ng
at
ever
y
su
ccessi
ve
m
ain
path
node
s
s
o
that
the
rand
om
disp
ersi
on
is
ver
y
high.
I
n
our
pro
posed
schem
e,
pac
ke
t
data
secur
it
y
is
ac
hieve
d
by
en
crypti
ng
the
e
ntire
co
ntent
of
t
he
data
pa
c
ket
us
i
ng
pa
irwise
sec
ret
keys
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ra
ndom r
ou
ti
ng sc
he
me
wi
th
misle
adin
g dead e
nds (C
hitra
Raja
r
ama)
4177
In
[
3],
backb
one
flo
od
i
ng
is
adopted
a
par
t
f
ro
m
virtu
al
(f
a
ke)
sin
ks
.
I
n
our
schem
e,
sin
k
lo
cat
ion
sec
uri
ty
is
achieve
d
without
bac
kbone
flooding
res
ulti
ng
in
le
sser
ove
rh
ea
d.
Ra
ndom
iz
ed
Ro
utin
g
with
Hidden
A
ddres
s
is
us
e
d
by
N
ga
i
in
[
5].
He
use
s
ra
ndom
ro
ut
ing
paths
al
l
of
w
hich
sta
rt
f
r
om
the
so
urce.
This
will
endang
e
r
the
so
urce
loca
ti
on
secu
rity
wh
en
tra
ff
ic
ana
ly
sis
and
packet
traci
ng
at
ta
cks
are
m
ou
nte
d.
A
nother
dr
a
wb
ac
k
of
Ng
ai
’
s
m
e
t
hod
is
the
us
e
of
enc
rypte
d
data
pac
kets
instea
d
of
fa
ke
pack
et
s.
In
our
m
et
ho
d,
ra
ndom
br
a
nc
hing
occ
ur
s
at
e
ver
y
node
al
ong
t
he
m
ai
n
path
w
hich
pro
vid
es
bette
r
secu
rity
for
the
s
ource
w
hic
h
is
a
seco
nd
a
ry
ad
va
ntage.
Also,
the
us
e
of
fa
k
e
pack
et
s
for
the
fak
e
r
oute
s
wi
ll
increase
the
secur
it
y
of
real
data.
In
[
6],
Shu,
e
t
al
.,
hav
e
de
scribe
d
Ra
ndom
iz
ed
Mult
ipath
Deli
ver
y
to
achieve
si
nk
locat
ion
sec
ur
it
y.
The
basic
pri
nc
iple
of
S
hu’s
m
et
ho
d
is
t
he
secret
sh
a
rin
g
of
i
nfor
m
at
ion
base
d
on
t
he
well
know
n
S
ham
ir’s
al
gorithm
[
7
]
with
m
ulti
path
routing.
T
his
inform
at
ion
sp
li
tt
ing
an
d
it
s
su
bse
que
nt
r
ecov
e
ry
neces
sit
at
e
add
it
io
nal
ov
e
r
head.
In
our
s
chem
e,
info
rm
at
ion
sp
li
tt
ing
is
avo
ide
d
to
r
edu
ce
the
c
ompu
ta
ti
onal
ov
e
r
head.
Locati
on
Pr
i
va
cy
Rou
t
ing
(L
PR)
in
[
8
]
us
es
cl
os
e
an
d
far
t
her
neig
hbor
li
sts
to
m
anag
e
r
eal
and
fa
ke
pa
ckets.
This
i
nvolv
es
detai
le
d
locat
i
on
in
form
at
ion
of
al
l
the
neig
hbors
as
a
pr
e
-
requisi
te
.
Hen
c
e,
the
LPR
sch
e
m
e
is
com
pu
ta
ti
on
al
ly
ex
pe
ns
i
ve.
In our schem
e, th
e exact l
ocati
on in
f
or
m
at
ion
of n
ei
ghbors
is not re
quired
.
Our
pr
opos
e
d
sink
l
ocati
on
pr
i
vacy
sche
m
e
ov
erc
om
es
sever
al
disad
van
ta
ges
of
t
he
existi
ng
m
et
ho
ds.
O
ur
schem
e
us
es
non
-
re
petit
ive
ra
ndom
path
transm
issio
ns
with
m
is
le
ading
dea
d
end
s
.
The
descr
i
ptio
n
of
the
schem
e
is
giv
e
n
in
t
he
nex
t
sect
io
n.
A
rea
dy
to
use
al
gorithm
,
R
RSM
DE
is
pre
sented
in
Sect
ion
2.
T
he
sim
ulate
d
c
om
par
at
ive
stud
y
sh
ows
that
our
m
et
ho
d
pe
rfor
m
s
bette
r
t
han
oth
e
r
m
e
tho
ds
as
dem
on
strat
ed
in
Sect
io
n 3.
2.
BASI
C
S
CHE
ME
AND
W
O
RKING
RR
SMDE
is
ba
s
ic
al
ly
a
ran
dom
iz
ed
m
ulti
c
ast
routing.
Fa
ke
co
pies
of
t
he
ori
gi
nal
dat
a
pack
et
a
re
transm
itted
to
reach
diff
e
re
nt
rand
om
du
m
m
y
destinat
ion
s
w
hile
on
e
true
co
py
rea
ches
the
act
ua
l
final
destinat
io
n
w
hi
ch
is
norm
ally
the
sink
.
T
o
m
ake
m
at
te
r
clear
,
an
in
sta
n
c
e
of
the
m
ulti
-
paths
of
RR
S
MDE
is
sh
ow
n
in
Fig
ure
1.
The
s
horte
st
path
from
th
e
so
urce
to
the
sink
is
cal
le
d
the
m
ai
n
path.
I
n
Fig
ure
1,
the
path
li
st
[V
1,
V
2,
V3,
V
4,
V5
]
gi
ves
the
m
ai
n
path
wh
ic
h
is
sh
ow
n
in
the
bl
ack
bold
font.
Her
e
V
1
is
the
so
urce
an
d
V
5
is
the
sink
no
de.
I
n
th
is
pap
e
r,
sym
b
ol
V
j
re
pr
ese
nt
s
bo
t
h
the
no
de
nam
e
(iden
ti
t
y)
and
the
e
ncry
pte
d
node
a
ddres
s
of
node j
w
her
e
j
∈
{1
to N}.
Figure
1. RR
MSDE m
ulti
ca
st paths
The
no
des
al
on
g
the m
ai
n
path
are call
ed
the
ma
i
n
path
node
s
. In
our
sc
he
m
e, b
ranchi
ng
towa
rd
s t
he
rand
om
fak
e
destinat
ion
s
ta
ke
s
place
at
suc
cessi
ve
m
ai
n
path
no
des
in
cl
ud
in
g
the
final
destinat
io
n
node
.
In
Fig
ure
1,
m
ulti
cast
br
anc
hi
ng
occ
urs
al
on
g
the
m
ai
n
path
at
no
des
V
1
,
V
2
,
V
3
,
V
4
,
and
V
5
.
Sub
br
anch
e
s
are
s
how
n
in
blu
e.
I
n
our
schem
e,
a
t
pr
esent,
on
ly
one
sub
br
a
nc
h
or
iginate
s
at
each
br
a
nch
po
int.
The
num
ber
of
br
a
nc
h
po
i
nt
s
is
equ
al
to
t
he
num
ber
of
nodes
al
ong
t
he
m
a
in
path.
Th
e
nu
m
ber
of
s
ub
br
a
nc
hes
in
Fi
g
ure
1
is
5.
In
Fig
ure
1,
the
le
ng
t
hs
of
the
s
ub
br
a
nc
hes
ar
e
al
l
cho
sen
ra
ndom
ly
.
The
or
iginal
pack
et
is
se
nt
al
ong
the
m
a
in
path
a
nd
th
is
pack
et
is
c
al
le
d
the
m
ai
n
pack
et
.
The
m
ai
n
pack
et
he
ade
r
inf
or
m
at
ion
is
updated
at
eac
h
m
ai
n
path
node
as
it
tra
ve
ls
al
ong
t
he
m
ai
n
pat
h
a
nd
ul
tim
at
el
y
reaches
the
final
de
sti
natio
n.
At
each
bra
nch
point,
a
n
a
lt
ered
co
py
of
the
m
ai
n
pack
e
t,
cal
le
d
the
fa
ke
pa
cket,
is
c
r
eat
ed
and
it
is
se
nt
al
ong
the
sub
br
a
nc
h
f
or
it
s
travel
f
ur
t
her.
The
update
operati
on
of
the
m
ai
n
pack
et
a
nd
t
he
creati
on
of
fake
pack
et
s
are
de
scri
be
d
la
te
r.
In
RR
SMDE
,
the
sink
(the
f
inal
real
destin
at
ion
)
is
not
one
of
the
dea
d
en
ds.
A
bran
ch
path
sta
rts
f
r
om
the
sink
and
c
on
ti
nu
es
fu
rt
her.
This
i
s
to
con
f
use
the
at
ta
cker
furthe
r
.
U
24
U
23
U
22
U
21
V
1
sr
c
V
4
V
3
V
2
V
5
sin
k
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
201
9
:
4
1
7
6
-
4
1
8
3
4178
2.1.
Packe
t
he
ad
er
inf
ormatio
n for
th
e m
ain
p
at
h
The
ad
dr
es
s
inform
ation
store
d
in
the
encr
y
pted
f
or
m
in
th
e
m
ai
n
pack
et
head
e
r
is
sh
ow
n
in
Table 1
.
In
ge
ner
al
,
the
Mai
n
Pac
ket
at
sta
rt
is
cal
le
d
MP
1
an
d
that
a
t
k
th
m
ai
n
node
is
cal
le
d
MP
k
,
f
or k
= 1
t
o
L
w
he
re
L is the t
otal n
um
ber
of
node
s in
t
he
Ma
in
P
at
h.
T
he k
t
h
node
tra
ns
m
it
s M
P
k
to t
he
(
k+
1)
t
h node.
Table
1.
H
ea
de
r
fiel
ds i
n t
he m
ai
n
packet
MP
1
at
sou
rce
V
1
Presen
t Sou
rce Ad
d
ress
Nex
t Destin
atio
n
Ad
d
ress
No
d
e cou
n
t k
Main Path
List
Oth
ers
Initially
V
1
V
2
1
[
V
1
, V
2
,
…
,
V
L
]
--
-
--
-
--
The
sec
ond
row
gi
ves
the
c
orres
pondin
g
te
r
m
s
fr
om
the
Exam
ple
of
Fig
ur
e
1.
T
he
hea
der
c
onte
nt
of
the
m
a
in
packet
,
travell
ing
from
the
or
igina
l
so
urce
to
the
final
destinat
io
n,
c
hanges
as
t
he
pac
ket
goes
from
the
present
no
de
to
t
he
ne
xt
node.
I
n
ge
ne
r
al
,
the
Ma
in
Pa
cket
at
sta
rt
is
cal
le
d
MP
1
an
d
that
at
k
th
m
a
in
node
is
cal
le
d
MP
k
,
for
k
=
1
to
L
wh
e
re
L
is
the
total
nu
m
ber
of
nodes
i
n
the
Ma
in
Path.
T
he
k
th
node
tra
ns
m
it
s
MP
k
to
the
(
k+
1)
th
node
.
Wh
e
n
the
m
ai
n
pac
ket
reac
hes
V
2
,
The
prese
nt
a
ddress
is
up
dated
to
V
2
an
d
t
he
ne
xt
destinat
io
n
ad
dress
is
updated
fr
om
the
Ma
in
Path
List
and
k
is
increm
ented
by
1.
In
general,
w
hen
the
m
ai
n
pack
et
ar
rive
s
at
the
k
th
node
of
the
m
ai
n
path
the
val
ues
of
the
a
ddress
fiel
ds
an
d
k
va
lue
befo
re
an
d
after
updatin
g woul
d be as
sho
wn
in
Ta
ble 2.
Table
2.
H
ea
de
r
fiel
ds
of
t
he packet at
k
th
m
ai
n
pat
h node
Presen
t Sou
rce Ad
d
ress
Nex
t Destin
atio
n
Ad
d
ress
No
d
e cou
n
t k
Main Path
List
Oth
ers
Bef
o
re
u
p
d
ate
V
k
−1
V
k
k
−1
[V
1
,V
2
, ,
,
…,
V
L
]
--
-
--
-
Af
ter
u
p
d
ate
V
k
V
k
+1
k
[V
1
,V
2
, ,
,
…,
V
L
]
--
-
--
-
The
val
ues
of
Table
2
ho
l
d
true
for
k
=
2,
3,…,
L−
1.
N
ote
that
the
L
th
node
is
the
s
ink
a
nd
no
furthe
r
update
an
d
forw
a
r
din
g
at
the
sin
k,
beca
us
e
t
he
m
ai
n
path
r
outi
ng
is
over
.
Wh
e
n
t
he
m
ain
pack
e
t
arr
ive
s
at
the si
nk, it chec
ks
th
at
k
=
L and als
o
the
pr
e
sent s
ource a
ddress f
ie
ld v
al
ue V
k
=
L. In our sc
he
m
e, the
m
ai
n
pack
et
con
ta
in
s
the
address
of
th
e
nex
t
destin
at
ion
.
Th
us
we
us
e
sou
rc
e
ro
utin
g
to
send
the m
ai
n
data.
2.2.
Sub
branch
p
at
h
s
an
d
p
acke
ts
A
s
ub
br
a
nc
h
or
i
gin
at
es
fro
m
ever
y
m
a
in
path
no
de.
The
sub
branc
h
pa
th
sta
rtin
g
fro
m
V
k
is
cal
le
d
SB
k
for
k=
1
to
L
.
I
n
Fi
g
ure
1,
SB
2
=
[U
21
,
U
22
,
U
23
,
U
24
]
.
T
he
nodes
al
ong
the
s
ub
br
a
nc
h
paths
are
ra
ndom
.
The
pr
e
sent
source
no
de,
i
n
a
sub
br
a
nc
h
pat
h,
sel
ect
s
the
ne
xt
ho
p
destina
ti
on
rand
om
l
y
a
m
on
g
it
s
nei
gh
bo
rs
exclu
ding
the
m
ai
n
path
node
s
an
d
the
node
s
w
hich
hav
e
been
al
read
y
t
rav
e
rse
d
so
fa
r
.
Wh
en
there
a
re
no
neig
hbors
f
or
a
sub
br
a
nc
h
node,
the
tra
nsm
issi
on
aut
om
at
ic
ally
gets
te
rm
inate
d
eve
n
if
TTL
has
no
t
ye
t
reache
d
ze
ro.
The
s
ub
branc
h
pac
ket
wh
ic
h
is
a
fa
ke
pa
cket
co
ntains
the
Tim
e
To
Live
(T
TL)
wh
ic
h
determ
ines
the
le
ng
th
of
the
s
ub
br
a
nc
h
pat
h.
Af
te
r
each
s
ub
branc
h
hop,
TTL
gets
dec
r
e
m
ented
by
1.
Wh
e
n
the TTL
r
eac
he
s zer
o,
t
he pr
opagati
on is
ter
m
inate
d.
2.3.
Algori
th
m
for
t
he
ma
in
p
ath pr
opogat
i
on
Th
e
wor
king
of
the
m
ai
n
pat
h
prop
a
gatio
n
of
RR
SM
DE
i
s
re
pr
ese
nted
by
the
te
rm
RRSMDE
-
M
PP
and the al
gorithm
RR
SMDE
-
MPP is
descr
i
be
d
as
foll
ows.
Algori
th
m
RR
SMDE
-
MPP
.
I
nputs: T
he
s
hortest
path
[V
1
,
V
2
,…
,V
k,
…, V
L
]
.
-
Set
k
=
1.
//
Start at
V
1
.
-
C
reate t
he
m
a
in p
at
h pack
et
MP
k
,
-
Creat
e t
he fa
ke pack
et
an
d choose
it
s TTL
-
Start
sub
br
a
nc
h pro
pag
at
io
n st
arti
ng fro
m
this m
ai
n
path n
od
e
V
k
.
-
Se
nd the
m
ain
pack
et
t
o
t
he next m
ai
n
pat
h n
od
e
V
k+1
-
Re
cei
ve
t
he
pack
et
at
V
k+1
.
U
pd
at
e
t
he
P
r
esent
S
ource
Address
a
nd
the
Nex
t
Desti
nation
A
ddres
s
fiel
ds
of the m
ai
n
pa
cket
-
I
nc
rem
ent k
a
s,
k =
k+
1.
//
Ma
in p
ac
ket is
fu
ll
y u
pdat
ed
-
I
f k =
L g
o
t
o 9.
//
Fi
nal de
sti
nation
is
rea
ched
Else
go to 3.
-
Start t
he
s
ub
br
a
nc
h prop
a
ga
ti
on
starti
ng fro
m
this
m
ai
n
path n
od
e
V
L
.
-
O
ve
r.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ra
ndom r
ou
ti
ng sc
he
me
wi
th
misle
adin
g dead e
nds (C
hitra
Raja
r
ama)
4179
2.4.
Sub bra
nch
p
ropaga
tion
Sub
Bra
nc
h
P
r
op
a
gatio
n
(S
B
P)
is
a
ra
ndom
path
travel
with
fa
ke
data
pac
kets.
T
he
pur
po
se
is
to
conf
us
e
the
at
ta
cker
who
m
a
y
fo
ll
ow
the
pa
ckets
to
disco
ve
r
the
destinat
i
on
(sink
or
BS).
I
n
our
sc
hem
e,
on
e
su
b
br
a
nch
pr
opagati
on
sta
rts
from
each
m
a
in
pat
h
i
nclu
din
g
the
final
de
sti
nation.
T
here
are
L
se
par
at
e
sub
br
a
nc
h
paths w
her
e L is the num
ber
o
f
no
de
s alon
g
the m
ain
p
at
h. The
no
des
al
ong
a sub b
ra
nc
h
ra
ndom
p
at
h
are
so
sel
ect
ed
that
the
nodes
do
no
t
re
pe
at
(which
m
eans
no
lo
ops)
a
nd
the
path
e
xclu
des
the
no
des
of
th
e
m
ai
n
path
wh
i
ch
pro
vid
es
be
tt
er
secur
it
y
by
dr
awi
ng
the
at
te
ntion
away
fr
om
the
m
a
i
n
path
.
Wh
en
t
he
sub
br
a
nc
h path e
nds,
t
he fake
p
a
cket is
discar
de
d.
2.5.
RRSM
DE
e
xampl
e
The
WSN
la
yo
ut
is
sh
ow
n
in
Figure
2.
A
gr
i
d
base
d
de
plo
y
m
ent
is
us
ed
w
it
h
a
few
m
issin
g
nodes
at
rand
om
gr
id
po
i
nts.
Her
e
82
no
des
are
distrib
uted
in
a
10x10
gr
id.
Eac
h
gr
i
d
cel
l
is
100mx10
0m
.
The
com
m
un
ic
at
ion
ra
ng
e
of
each
node
is
s
et
to
15
0
m
et
e
rs.
A
node
can
hav
e
a
m
axi
m
u
m
of
8
nei
ghbor
s
al
ong
nort
h,
s
ou
t
h,
east
,
we
st,
north
-
east
,
so
ut
h
-
ea
st,
nor
th
-
west,
south
-
west.
I
n
this
e
xam
ple,
so
urc
e
node
is 1
1
a
nd the
final m
ai
n
destin
at
ion
(sink)
is
node 5
6.
Fig
ure
2. Ma
in
p
at
h an
d su
b
branc
h paths
for exam
ple 1
. Sr
c
=11 an
d
sin
k=
56
The
sho
rtest
(m
ai
n)
path
is:
[11,
12,
21,
29,
37,
38,
39,
48,
56]
.
The
le
ngth
of
the
shortes
t
path
SPL=
8.
T
he
total
n
um
ber
of
nodes
of
the
m
ai
n
path
=L=
9.
The
re
a
re
9
sub
branc
hes.
The
sub
branc
h
paths
f
or
on
e
t
rial
are
s
how
n
in
Fi
gure
2.
The
s
ub
br
a
nc
h
path
SB3
,
st
arti
ng
f
ro
m
node
21
do
es
no
t
exist
because
it
has
no
valid
nei
ghbo
rs.
F
or
the
sa
m
e
W
SN
la
yout
of
of
Fig
ur
e
2,
the
for
m
at
ion
of
path
s
wh
e
n
src=9 an
d si
nk
=
59 is s
how
n
i
n
Fi
gure
3. Fro
m
Figu
re
s
2
a
nd
3,
we
ca
n see
the
possible
pa
tt
ern
s
of
RR
S
ND
E
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
201
9
:
4
1
7
6
-
4
1
8
3
4180
Fig
ure
3. Ma
in
p
at
h an
d su
b b
ran
c
h paths
for
e
xam
ple 1
. Sr
c
=9
a
nd sin
k
=
59
3.
SIMULATI
O
N,
RES
ULTS
AND REL
AT
IVE PE
RFO
R
MAN
CE
3.1.
Av
er
age
p
ath
le
ng
t
h
(
APL
)
AP
L
is
m
easur
ed
in
te
rm
s
of
the
aver
a
ge
num
ber
of
hops
r
equ
i
red
to
reac
h
the
de
sti
nation
from
the
so
urce.
T
he
pa
cket
delivery
tim
e
m
ai
nly
dep
en
ds
on
the
AP
L.
Sm
al
le
r
the
AP
L
,
bette
r
is
the
per
f
orm
ance
.
In
RR
SM
DE,
t
he
tr
ue
pac
ket
s
fo
ll
ow
the
s
hortest
(m
ai
n)
pa
th
from
the
so
urce
to
t
he
de
sti
nation.
T
he
refor
e
the
AP
L
is
sa
m
e
as
the
Sh
ort
est
Path
Len
gth
(SPL
).
Bot
h
are
ex
pr
e
sse
d
in
te
rm
s
of
the
nu
m
ber
of
hops
.
T
her
e
f
or
e,
AP
L
(RRSM
D
E)=
SP
L
(
1)
In
P
ur
el
y
Ra
ndom
Pr
op
a
ga
ti
on
(P
R
P)
[
6]
and
L
ocati
on
P
rivacy
R
ou
ti
ng
(
LPR)
[8
]
,
the
r
oute
s
ar
e
rand
om
iz
ed.
I
n PRP, t
he APL
is give
n by,
AP
L
(P
RP
)
= a
ver
a
ge(TTL
)+
SPL
(2)
Her
e
, TTL
is t
he
le
ngth
of t
he
r
a
ndom
p
art
of the
path
in
e
ach trial
. I
n
LP
R, the a
ver
a
ge
le
ng
th
of
t
he ro
uting
path
is
g
i
ven by
[
8
],
AP
L
(LPR
)
= {
1/(
1−2*P
f)
}
*SPL
(3)
Her
e
, Pf is t
he pr
obabili
ty
o
f
s
el
ect
ing
the
n
e
xt no
de fu
rthe
r
aw
ay
from
the d
est
inati
on.
Takin
g
the
WSN
as
giv
e
n
in
t
he
E
xam
ple
of
Sect
ion
2.4,
t
he
A
PLs
a
re
c
al
culat
ed
f
or
di
ff
e
ren
t
S
PL
values
an
d
f
or
dif
fer
e
nt
m
eth
ods.
I
n
cal
cul
at
ing
AP
L
(P
R
P)
,
ave
rage(T
TL)
is
obta
ine
d
by
ta
ki
ng
10
0
tria
ls
with
TTL
(m
ax)
set
to
8.
In
ca
lc
ulati
ng
AP
L
(
LPR),
Pf
is
ta
ke
n
as
0.2.
T
he
var
ia
ti
on
of
A
PL
w
hich
re
presents
the p
ac
ket
delivery ti
m
e is sho
w
n
i
n
Fi
g
ure
4
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ra
ndom r
ou
ti
ng sc
he
me
wi
th
misle
adin
g dead e
nds (C
hitra
Raja
r
ama)
4181
Figure
4. A
verage
path
le
ngth
v
s
shortest
pat
h
le
ngth
3.2.
Ov
er
all Ener
gy
Consum
pt
io
n (OE
C)
The
Ov
e
rall
E
nergy
Co
nsum
ption
(
OEC)
f
or
a
gi
ven
tria
l
(to
se
nd
a
da
ta
pack
et
from
src
to
ds
t
)
dep
e
nds
on
the
total
le
ng
t
h
of
the
paths
ge
ne
rated
(
bo
t
h
m
ai
n
an
d
s
ub
branch
pat
hs
)
or
the
T
otal
N
umber
of
Hops (T
NH).
The
T
N
H valu
es for
alg
or
it
hm
s RRSM
DE,
PRP and
LPR
are
giv
e
n by,
TNH(RR
SMD
E)
=
SPL
+ s
um
o
f
the s
ub
br
anch pat
h
le
ng
t
hs
(
4)
TNH(PRP
)
=
S
PL+ a
ver
a
ge(T
TL)
(5)
TNH(LPR
) =
{1/(
1−2*P
f)
}
*SPL
(
6)
In PRP a
nd L
P
R m
e
tho
ds, in
each,
t
her
e
is
only
one
path pe
r
tria
l.
Var
ia
ti
on
of
T
NH
with
SPL
is
sh
own
i
n
Fig
ure
5.
I
n
the
case
of
RR
SM
DE,
the
T
N
H
value
is
ve
ry
la
rg
e
beca
us
e
of
m
ulti
ple
fak
e
path
s.
The
r
efore
the
ene
r
gy
consum
ption
is
m
or
e,
co
m
par
ed
to
th
e
oth
er
two
m
et
ho
ds.
Figure
5. Total
num
ber
of
paths
(TNH
) vs
s
hortest
path
3.3.
St
ren
gth
of re
cei
ver anony
mi
ty
protecti
on
I
n
e
va
l
ua
t
i
ng
t
he
s
t
r
e
ng
t
h
o
f
t
he
r
e
c
e
i
ve
r
a
no
ny
m
it
y
,
w
e
us
e
t
he
a
dv
e
r
s
a
r
y
m
od
e
l
a
s
i
n
[
8
]
.
O
ne
of
t
he
m
e
a
s
ur
e
s
t
o
re
pr
e
s
e
nt
t
he
st
r
e
ng
t
h
of
r
e
c
e
i
ve
r
a
no
ny
m
it
y
i
s
A
dv
e
r
s
a
r
y
’
s
A
t
t
a
c
k
Tim
e
(
A
A
T
)
w
hi
c
h
i
s
“m
e
a
s
ur
e
d
a
s
th
e
nu
m
be
r
of
m
ov
i
ng
s
t
e
ps
(fr
om
on
e
s
e
ns
or
l
oc
a
t
i
on
t
o
a
ne
i
gh
b
or
)
t
he
a
dv
e
r
s
a
r
y
ha
s
t
o
m
a
ke
be
f
o
r
e
he
r
e
a
c
he
s
t
he
r
e
c
e
i
ve
r
”
.
I
n
R
R
S
M
D
E
,
t
he
a
d
ve
r
s
a
r
y
ha
s
t
o
t
r
a
ve
r
s
e
e
a
c
h
s
ub
b
r
a
nc
h
t
w
i
c
e
,
o
nc
e
t
o
g
o
f
or
w
a
r
d (
m
i
s
s
th
e
r
e
a
l
r
e
c
e
i
ve
r
)
a
nd
t
he
n
t
o r
e
t
r
e
a
t
.
T
he
r
e
f
or
e
t
he
nu
m
be
r
o
f
s
t
e
ps
t
he
a
dv
e
r
s
a
r
y
ha
s
t
o
t
r
a
ve
r
s
e
be
f
o
r
e
r
e
a
c
hi
n
g
t
he
f
i
na
l
de
s
t
i
na
t
i
on
i
s
gi
ve
n
by
,
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
201
9
:
4
1
7
6
-
4
1
8
3
4182
AA
T
(RRSM
D
E)
=
SPL
+
2*su
m
o
f
the
s
ub
br
a
nc
h path le
ng
t
hs
(
7)
I
n
P
R
P
a
nd
L
P
R
,
A
A
T
i
s
s
a
m
e
a
s
t
he
A
ve
r
a
ge
P
a
t
h
L
e
n
gt
h
,
A
P
L
,
a
s
gi
ve
n
by
(
2)
a
nd
(
3
)
.
V
a
r
i
a
t
i
on
of
A
A
T
w
i
t
h
S
P
L
i
s
s
ho
w
n
i
n
F
i
gu
r
e
6.
T
he
r
e
s
ul
t
s
a
r
e
ob
t
a
i
ne
d
by
a
n
a
l
yt
i
c
al
c
al
c
ul
at
i
on
s
.
I
n
t
he
c
a
s
e
of
R
R
S
M
D
E
,
t
he
A
A
T
va
l
ue
i
s
v
e
r
y
l
a
r
ge
be
c
a
u
s
e
of
m
ul
ti
pl
e
f
a
ke
pa
t
hs
.
T
he
r
e
f
or
e
t
he
s
t
r
e
n
gt
h
o
f
a
n
on
y
m
it
y
is
ve
r
y
hi
gh
,
c
om
pa
r
e
d
t
o
t
he
ot
he
r
t
w
o
m
et
ho
ds
.
F
i
g
ure
6.
A
dv
e
r
s
a
r
y
’
s
a
t
t
a
c
k
t
im
e
(
A
A
T
)
vs
s
ho
r
t
e
s
t
pa
t
h
l
e
n
gt
h
4.
CONCL
US
I
O
N
A
ne
w
m
et
hod
of
r
an
dom
r
ou
ti
ng
sc
hem
e
with
m
isl
eadi
ng
dea
d
e
nd
s
is
pr
ese
nted
.
The
pack
e
t
traci
ng
at
ta
ck
e
r
will
be
utterl
y
confu
se
d
be
cause
of
m
ultip
le
ra
ndom
paths
w
hich
le
a
d
to
w
ron
g
dea
d
en
ds.
Fr
om
the r
el
at
i
ve per
form
ance p
lots
, it can
be seen
that RR
SMDE is m
uc
h bett
er th
a
n ot
her m
et
ho
ds
.
RE
FERE
NCE
S
[1]
L
.
Yao
,
e
t
al
.
,
“
Protecting
the
sin
k
loc
at
io
n
priva
c
y
in
wire
l
ess sensor ne
tworks,”
Springer
-
Verl
ag
L
ondon,
Li
m
it
ed
,
2012.
[2]
A.
R.
Ma
lviy
a
a
nd
B.
N.
J
agda
l
e,
“
Locat
ion
pri
vacy
of
m
ultipl
e
sink
using
zon
e
par
titioni
ng
app
roa
ch
in
W
SN
,
”
2015
Inte
rnation
al
Confe
renc
e
on
Appl
ie
d
and
Theoretical
Computing
and
Communi
cat
ion
Te
chn
ology
(
iCA
TccT),
Davange
re
,
pp.
449
-
454
,
2015
.
[3]
Pavit
ha
N
.
and
S.
N.
Shelke
,
“
Techni
ques
for
Protecting
Locat
ion
Privacy
of
Sourc
e
and
Sink
Node
Against
Global
Adversari
es
in
S
ensor,
”
Inte
rnat
i
onal
Journal
of
Re
search
(
IJR
)
,
v
ol
.
1,
Sep
2014.
[4]
C
.
George
and
D
.
Natha
nie
l
,
“
Protec
ti
ng
Loc
ation
Privacy
in
W
ire
le
ss
S
ensor
Networks
aga
inst
a
Local
Ea
vesdroppe
r
–
A
Surve
y
,
”
Int
ernati
onal Journal of
Computer
Ap
pli
cations
,
Oc
t
2
012.
[5]
E
.
Ngai
,
“
On
providi
ng
sink
ano
n
y
m
ity
for
sens
or
net
works
,
”
in
Proce
ed
ings
of
the
5th
In
te
rnati
onal
Confe
ren
c
e
on
Wireless
Comm
unic
ati
ons
a
nd
Mobile
Com
puti
ng:
Conn
ec
t
ing
th
e
World
Wirel
essly
–
IW
CMC’09.
ACM
,
pp.
269
-
273
,
2009
.
[6]
T.
Shu,
e
t
a
l.
,
“
Secur
e
d
ata
c
o
llection
in
wire
les
s
sensor
net
works
using
ran
do
m
iz
ed
dispersiv
e
rout
es,
”
IEEE
Tr
ans.
Mobil
e
C
omput
.
,
vol
/
issue:
9
(
7
)
,
pp.
941
-
9
54,
2010
.
[7]
D.
R.
Stinson
,
“
Cr
y
ptogr
aph
y
,
T
heor
y
and
Prac
tice
,”
CRC
Press
,
2006.
[8]
Y
.
Jian,
et
al.
,
“
Protec
ti
ng
Recei
ver
-
Locat
ion
Privacy
in
W
ire
le
ss
Sensor
N
et
works
,
”
IEEE
Comm
unic
ati
ons
Soci
e
ty
,
IE
EE IN
FOCOM 2007
p
roce
edi
ngs
,
pp
1
955
-
1963
,
2007
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N:
20
88
-
8708
Ra
ndom r
ou
ti
ng sc
he
me
wi
th
misle
adin
g dead e
nds (C
hitra
Raja
r
ama)
4183
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Ch
itr
a
Rajarama
rec
e
ive
d
he
r
B.
E.,
in
Com
pute
r
Scie
n
ce
&
Engi
ne
eri
ng
.
,
and
M.T
ec
h
.
,
in
Com
pute
r
Sc
ie
nc
e
&
Engi
n
e
eri
ng,
from
Visvesvara
y
a
T
ec
h
nologi
c
al
Univ
e
r
sit
y
,
B
el
gaum,
Karna
ta
k
a,
India
.
She
is
cur
r
ent
l
y
pursuing
PhD
under
Visvesvar
a
y
a
Technol
ogi
c
al
Univ
ersity
,
Bel
gaum,
Karna
ta
ka
,
India.
Her
are
a
of
intere
s
t
is
in
the
fie
ld
of
wire
le
ss
netw
orks.
She
has
guide
d
m
an
y
un
der
gra
dua
te
proj
ec
ts.
She
h
as
att
ende
d
m
an
y
n
at
i
onal
/i
n
te
rn
at
ion
a
l
conf
ere
n
ce
s
and
publi
shed
seve
ra
l
pape
rs
in
int
ern
a
ti
ona
l
jo
urna
ls.
At
pre
se
nt
she
is
Associ
at
e
Profess
or
in
the
depa
r
tment
of
Inform
at
ion
Scie
nc
e
&
Engi
ne
e
ring,
NIE
Instit
u
te
of
Te
chno
log
y
,
(
aff
iliated
to
Visvesvara
y
a
Technol
ogi
ca
l
Uni
ver
sit
y
)
M
y
suru,
Karna
t
aka,
Ind
i
a,
S.N.
Jagadeesha
recei
ved
his
B.
E.,
in
Elec
tro
nic
s
and
Com
m
unic
at
ion
Eng
ine
er
ing,
from
Univer
sit
y
B.
D.
T.
Coll
ege
of
Engi
nee
r
ing.,
Dava
nger
e
aff
il
iated
to
M
y
so
re
Univer
sit
y
,
Karna
ta
k
a,
Ind
i
a
in
1979,
M.
E.
from
India
n
Instit
ute
of
Sc
ie
nc
e
(IISC),
B
anga
lor
e,
Indi
a
spec
ializing
in
El
e
ct
ri
ca
l
Com
m
unic
at
ion
Enginee
ring
.
,
in
198
7
and
Ph.D.
in
El
e
ct
roni
cs
and
Com
pute
r
Engi
n
ee
ring
.
,
from
Univer
sit
y
of
Roo
rke
e,
Roorke
e,
I
ndia
in
1996.
He
is
an
IE
EE
m
ember.
His
res
ea
rch
in
te
r
est
inc
lude
s
Arra
y
Si
gnal
Proce
ss
ing,
W
ire
le
ss
Senso
r
Networks
and
Mobile
Com
m
unic
a
ti
ons.
He
h
as
publi
shed
an
d
pre
sent
ed
m
a
n
y
p
ape
rs
on
Adapti
ve
Arr
a
y
Signal
Proce
ss
i
ng
and
Dire
c
tion
-
of
-
Arriva
l
e
stim
at
ion.
Curr
ent
l
y
he
is
profe
ss
or
in
the
depa
rtment
of
E
le
c
troni
cs
and
Com
m
unic
at
ions
Engi
neering,
PES
Instit
ute
of
Technol
og
y
&
Mana
geme
nt
.
(A
ffil
i
at
ed
to
Visve
svara
y
a Technol
ogic
a
l
Univer
si
t
y
)
,
Shim
oga, K
a
rna
ta
k
a, I
ndia.
Yer
ri
S
w
am
y
T
rec
ei
v
ed
his
B.
E.,
in
Elec
tro
nic
s
and
Com
m
-
unic
a
ti
on
Enginee
ring
,
from
Gulbar
ga
Unive
risity
,
Gulb
arg
a
,
Karna
ta
k
a
,
Ind
ia
in
2000
.
MT
ec
h
in
N
et
work
and
Internet
Engi
ne
eri
ng,
fro
m
Visve
-
svara
y
a
Te
chno
logi
c
al
Univer
sit
y
,
Belgaum
,
Karna
ta
k
a.
at
J.
N.
N.
Coll
ege
of
Enginee
ring
,
Shim
oga,
Karna
t
aka
in
2005.
and
PhD
i
n
the
Facul
t
y
of
Com
pute
r
and
Inform
at
ion
Sci
enc
es
from
Visvesvara
y
a
Tech
nologi
c
al
niv
ersity
,
B
el
gaum,
K
arn
ataka
in
the
y
e
ar
2013.
He
is
an
ISTE
m
ember.
His
rese
ar
ch
intere
st
in
cl
udes
Antenn
a
Arra
y
Signa
l
Proce
ss
ing,
Sta
t
isti
cal
Sign
al
P
roc
essing,
D
et
e
ct
ion
and
Esti
m
at
ion
,
Cogni
ti
ve
rad
io
comm
-
unic
a
ti
on
,
LTE/MIMO
.
He
has
publ
ished
and
pre
se
nte
d
num
ber
of
pap
ers
in
nat
ion
al
/i
n
te
rn
ational
conf
ere
nc
es
and
journa
ls.
Cur
ren
tly
h
e
i
s
Profess
or
and
Hea
d,
in
the
depa
rtment
of
C
om
pute
r
Scie
n
c
e
&
Engi
n
ee
r
in
g,
KLE
Insti
tut
e
of
T
ec
hno
log
y
,
(Affil
i
at
ed
to
Visvesvara
y
a
Technol
ogi
ca
l
Uni
ver
sit
y
), Hubli
,
Karna
ta
k
a, I
ndi
a
.
Evaluation Warning : The document was created with Spire.PDF for Python.