Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
V
o
l.
6, N
o
. 2
,
A
p
r
il
201
6, p
p
.
73
5
~
74
2
I
S
SN
: 208
8-8
7
0
8
,
D
O
I
:
10.115
91
/ij
ece.v6
i
2.9
038
7
35
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
Sufficien
t Authen
tication for Energy Consu
m
ption in Wireless
Sensor Networks
Ri
az
Sh
ai
k*
,
L
o
kesh
K
a
n
a
gal
a*
, He
ma
Gopi
n
a
th S
u
k
a
v
a
si
*
* Department of
Computer Scien
ce
a
nd
Engineering, KL University
, India
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Sep 15, 2015
Rev
i
sed
No
v
22
, 20
15
Accepted Dec 13, 2015
Given the
under
s
tanding of
the
prospectiv
e W
S
N program
s and becaus
e
of
source restrictio
ns, key
management em
erges
a
s
a com
p
lica
t
ed
problem
for
W
S
Ns
. One of t
h
e m
a
in is
s
u
es
when
develop
i
n
g
a key
ma
nag
e
ment scheme
is the s
y
stem scalability
.
Indeed,
the method shou
ld assistance
a h
uge number
of nodes to allo
w a larg
e rang
e
implemen
tation
of the s
y
s
t
em. I
n
this pap
e
r
we im
plem
ente
d a perform
anc
e
trad
e-off res
e
arch of power
intake vs
.
Quality
of
Solutions obtain in stabil
it
y,
s
u
i
t
abil
it
y,
and
s
ecuri
t
y
for
redundancy
con
t
rol of cluster
e
d
heter
ogen
e
ous wireless i
ndicator sy
stems
using m
u
ltipath
routing to r
e
sponse custom
er concerns. We
urbanized
a
novel prob
ability
sty
l
e to
evalu
a
te th
e best red
undancy
stage in terms of
direction r
e
dund
ancy
(m
p
) and
resource redund
an
cy
(m
s
), as
w
e
ll
as
the b
e
s
t
att
ack iden
tifi
cat
ion configurat
io
ns
w
ith regards t
o
the number of
voters (m)
and the attack
incantation
in
terval
under
which the life-time of a
heterog
e
neous
wireless sensor networ
k is optim
ized whil
e f
u
lfill
ing the
st
a
b
i
l
ity
,
ti
me
l
i
ne
ss a
nd
protection specifications of question processing
appli
cat
ions
in t
h
e exis
ten
ce of
untrus
t
worth
y
w
i
rel
e
s
s
com
m
uni
cat
ion and
harmful nodes. Lastly
, we used our
analy
s
is outcomes to the sty
l
e of
powerful redund
ancy
management criter
ia to r
e
cognize and implement the
best design par
a
meter configu
r
ations at
play
back
in reaction to
environment
changes
to
exten
d
the
program
l
i
f
e
-tim
e.
Keyword:
Energy c
o
nservation
Key
m
a
nagem
e
nt
Netwo
r
k
scalab
ility
Reliab
ilit
y
Secure c
o
nnect
ivity cove
rage
Security
W
i
rel
e
ss se
ns
o
r
net
w
or
ks
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
:
Hem
a
Gopi
nat
h
S
u
kava
si
Depa
rt
m
e
nt
of
C
o
m
put
er Sci
e
nce a
n
d
E
ngi
n
eeri
n
g,
KL
U
n
i
v
ersi
t
y
Gree
n
fi
el
ds,
V
a
dde
swa
r
am
, Gu
nt
u
r
Di
st
ri
ct
, A
n
dh
ra P
r
a
d
e
s
h,
I
ndi
a
Mo
b
ile: +91
-
85
004
918
97
Em
a
il: su
k
a
v
a
sig
o
p
i
@g
m
a
i
l
.co
m
1.
INTRODUCTION
Nowa
days,
W
i
reless indicat
or syste
m
s (W
S
N
s) a
r
e pr
ogressiv
e
ly u
s
ed
i
n
cru
c
ial prog
ram
s
with
in
several areas
suc
h
as army,
m
e
dical and commercial areas. Key
m
a
nagem
e
nt is an
area roc
k
for
m
a
ny
security servic
es such as pri
v
acy and
verification wh
ich
are requ
ired
to
p
r
o
t
ected
emails
in
WSNs, th
e
orga
nization of
protected
hyperlinks
bet
w
ee
n
nodes is t
h
en a
com
p
licated problem
in
WSNs [1].
Because
of
sou
r
ce rest
ri
ct
i
ons
, sy
m
m
e
t
r
ical
key
orga
ni
zat
i
on i
s
one
of t
h
e m
o
st
app
r
op
ri
at
e par
a
di
gm
s for o
b
t
ai
ni
ng
transactions i
n
W
S
Ns. On the othe
r
ha
nd,
because
of
t
h
e
lack
of facilities in
WSNs,
we
have
usual
l
y no
rel
i
a
bl
e t
h
i
r
d
part
y
w
h
i
c
h c
a
n feat
ure
pai
r
wi
se key
i
m
port
a
nt
fact
o
r
s
t
o
nea
r
by
no
d
e
s, t
h
at
i
s
w
h
y
m
o
st
cur
r
ent
al
t
e
r
n
a
t
i
v
es are
de
pe
ndi
ng
o
n
key
pre
-
di
st
ri
but
i
o
n.
O
v
er
t
h
e l
a
st
seve
ral
y
ear
s, a
vari
et
y
of
st
u
d
y
han
d
l
e
d
sy
m
m
et
ri
cal
key
pre
-
di
st
ri
but
i
o
n
pr
obl
em
for
WS
Ns a
nd m
a
ny
al
t
e
rnat
i
v
es
ha
ve bee
n
s
u
gge
st
ed i
n
th
e literary work
s [2
]. Nev
e
rt
h
e
less, in
m
o
st
cu
rren
t
alte
rnativ
es, th
e styl
e o
f
k
e
y j
e
welry (b
lo
cks o
f
k
e
ys) is
h
i
gh
ly relev
a
nt to
th
e system
s
i
ze, th
ese
altern
ativ
es eith
er exp
e
rien
ce fro
m
lo
w scalab
ility (n
u
m
b
e
r of
r
e
in
for
ced
nodes)
, or
b
r
eak
d
o
wn
o
t
h
e
r
eff
i
cien
cy
an
alytics su
ch
as pro
t
ected
con
n
e
ctio
n
,
sto
r
ag
e
sp
ace
ex
p
e
n
s
e an
d
resilien
c
y in
th
e case o
f
hug
e syste
m
s. In
th
is p
e
rform
;
o
u
r
ai
m
is to
d
eal with
th
e scalab
ility
p
r
ob
lem with
ou
t d
e
grad
ing
th
e o
t
h
e
r system efficien
cy a
n
alytics [3
]. Fo
r th
is pu
rpo
s
e, we fo
cu
s on
th
e style
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
73
5 – 7
4
2
73
6
of a
plan whic
h
guara
n
tees a
n
e
x
cellent
secure
cove
ra
ge of extensi
v
e syste
m
s with a
low key stora
g
e spac
e
ex
p
e
n
s
e an
d
an
ex
cellen
t
syste
m
resilien
c
y.
To
th
is end
,
we m
a
k
e
u
s
e,
for th
e fi
rst time, o
f
th
e
u
n
i
t
a
l style
conce
p
t
f
o
r ef
f
i
ci
ent
W
S
N ke
y
pre-
di
st
ri
b
u
t
i
on [
1
]
.
I
ndee
d
,
we recom
m
end an i
n
noce
n
t
appl
y
i
n
g
fr
om
uni
t
a
l
st
y
l
e t
o
key
p
r
e-di
st
ri
but
i
o
n
and
we
sh
o
w
t
h
r
o
ug
h sy
st
em
atic research
th
at it allo
ws ach
i
ev
ing
a v
e
ry g
r
eat
scalab
ility [4
]. Howev
e
r, t
h
i
s
in
no
cen
t
app
l
yin
g
d
o
e
s no
t assuran
ce a h
i
gh
er
k
e
y discu
ssing
po
ssib
ility.
There
f
ore,
we recom
m
end an
im
prove
d u
n
i
t
a
l
-
base
d key
p
r
e-
di
st
ri
b
u
t
i
o
n pl
an t
h
at
p
r
ese
r
ves a
n
excel
l
e
nt
key
d
i
scu
ssi
ng
p
o
s
sib
ility wh
ile im
p
r
o
v
i
ng
t
h
e syste
m
scalab
ili
ty [3
].
Fig
u
r
e
1
.
W
i
r
e
l
e
ss
sensor
n
e
t
w
or
k
Ar
ch
itectu
r
e
W
i
t
h
the aim of
decreasi
n
g ene
r
gy inta
ke
wh
ile tak
i
n
g
t
h
e algo
ri
th
mic co
m
p
lex
n
e
ss in
to
con
s
i
d
erat
i
o
n
,
we rec
o
m
m
e
nd a no
vel
st
rat
e
gy
t
h
at
di
vi
de
s t
h
e uni
que i
n
fo
rm
ati
on i
n
t
o
several
pac
k
et
s
suc
h
that each
node in the
pr
ogra
m
will forwa
r
d
only little
sub pac
k
a
g
es. The
breaki
n
g process is ac
hieve
d
applying the
China Rest Theorem
(
CRT) criteria, whic
h is recognized
by
an easy
fl
i
p
di
vi
si
on
b
e
t
w
een
i
n
t
e
gers
[4]
.
T
h
e d
r
ai
n n
o
d
e,
once al
l
su
b packet
s
(cal
l
e
d C
R
T
com
ponent
s
)
are o
b
t
a
i
n
ed
pr
ope
rl
y
,
wi
l
l
recom
b
i
n
e t
h
e
m
,
t
hus re
b
u
i
l
d
i
n
g t
h
e
uni
qu
e conce
p
t
.
T
h
e
brea
ki
n
g
pr
oc
ess i
s
especi
al
l
y
bene
fi
ci
al
fo
r t
h
ose
forwa
r
ding nodes that a
r
e
m
o
re seek tha
n
ot
hers
due
t
o
their place within
t
h
e program
.
Regarding t
h
e
com
p
l
e
xness, i
n
t
h
e s
u
g
g
est
e
d st
rat
e
gy
, al
m
o
st
al
l
no
des f
unct
i
o
n as i
n
a cl
assi
cal
sendi
ng c
r
i
t
e
ri
o
n
an
d,
wi
t
h
the exem
ption
of the
drai
n,
a few low-co
m
p
lex
m
a
th
e
m
atic
s fu
n
c
tion
s
are
n
eed
ed
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISS
N
:
2088-8708
S
u
fficien
t
Au
then
tica
tion
fo
r
En
erg
y
C
o
n
s
ump
tion
i
n
Wireless S
e
n
s
o
r
Net
w
o
r
ks (Hema
Gop
i
na
th Su
kava
si)
73
7
Fig
u
r
e
2
.
En
erg
y
conv
er
si
o
n
ev
en
ts in
w
i
r
e
l
e
ss sensor
n
e
t
w
or
ks
M
o
re
part
i
c
ul
a
r
l
y
, we a
n
al
y
ze t
h
e o
p
t
i
m
a
l
am
ount
of
red
u
n
d
a
n
cy
t
h
r
o
u
gh
w
h
i
c
h i
n
f
o
rm
ati
on are
di
rect
ed t
o
a
rem
o
t
e
drai
n i
n
t
h
e e
x
i
s
t
e
nc
e of
unt
rust
w
o
rt
hy
an
d m
a
li
ci
ous
no
des,
so t
h
at
t
h
e
qu
est
i
o
n
ach
iev
e
m
e
n
t
s p
o
s
sib
ility is
max
i
m
i
zed
wh
ile i
m
p
r
o
v
i
ng
th
e HWSN life-ti
m
e
[5
]. Bu
t so
m
e
o
b
s
tacles
i
n
t
e
rr
upt
t
h
e
sy
st
em
s perf
orm
a
nce di
ffe
rent
l
y
such as t
h
e i
m
provi
n
g
bu
n
d
l
e
wai
t
,
t
h
us
di
ffi
c
u
l
t
fo
r re
or
deri
ng
the packages
, Marketing is
not effe
ctive
l
y handled
al
so Stream
ing issue in
lo
w d
a
ta tran
sfer u
s
ag
e
i
n
f
o
rm
at
i
on t
h
us
red
u
ces
pe
r
f
o
r
m
a
nce. To
be a
b
l
e
t
o
get
ove
r t
h
e
di
sad
v
ant
a
ges
of
t
h
e f
o
rm
erl
y
sug
g
est
e
d
pr
o
g
ram
,
we appl
y
t
h
e ne
w i
d
ea i
n
t
h
i
s
d
o
c
u
m
e
nt
. In o
u
r
sug
g
est
e
d p
r
o
g
r
am
, t
h
e best
possi
bl
e get
i
n
t
ouc
h
wi
t
h
ra
nge a
n
d i
n
t
e
ract
i
o
n
t
echni
q
u
e
wer
e
m
i
xt
ure t
o
im
pl
em
ent
t
h
e Het
e
ro
ge
neo
u
s
W
i
rel
e
ss
S
e
ns
or
Netwo
r
k
s
lifestyle in
n
a
ture [3
].
In
HWSN, th
e
in
tra-clu
s
ter
arran
g
i
n
g
an
d
in
ter clu
s
ter
m
u
l
ti-h
o
p
red
i
recting
o
f
fer to
take adv
a
n
t
ag
e th
e n
e
t
w
ork
lifetim
e.
And
it is re
garded as a
hierarchal HWSN
wi
th CH node
s such a
s
ex
cellen
t
en
ergy an
d
prov
id
i
n
g
o
u
t
cap
a
b
ilities th
an reg
u
l
ar
SNs
[4
]. Ou
r su
gg
ested
strateg
y
g
i
v
e
s so
lu
tio
n to
com
e
up wi
t
h
as o
p
t
i
m
i
zati
on p
r
obl
em
s t
o
bal
a
nce e
n
er
gy
i
n
t
a
ke acr
oss
al
l
node
s i
n
t
h
e ent
i
r
e het
e
ro
gene
o
u
s
sens
or
syste
m
s [6]. Though i
n
this
docum
e
nt,
we s
u
gg
est
two-tier HWSN with
th
e objectiv
e of cap
italizin
g
o
n
p
r
og
ram
li
fe-tim
e wh
ile
satisfyin
g
en
erg
y
co
n
t
ro
l and
cov
e
rag
e
goals. Th
ey id
entified
th
e m
a
x
i
m
u
m
d
e
nsity ratio
o
f
th
e two
tier’s
n
o
d
e
s to
i
n
crea
se the
program
life-tim
e [6].
Th
e
relax
o
f
th
e
d
o
c
u
m
en
t will b
e
org
a
n
i
zed
as
fo
llows: In
section
2
,
we see abo
u
t
th
e relev
a
nt
per
f
o
r
m
s
of t
h
e doc
um
ent
.
In
sect
i
on 3
,
we t
a
l
k
ab
out
t
h
e s
u
g
g
est
e
d t
ech
n
i
que.
The al
go
r
i
t
h
m
s
and si
m
u
l
a
t
o
r
are
prove
n i
n
t
h
e a
r
ea
4 a
n
d 5.
2.
RELATED WORK
We resea
r
ch a desi
g
n
o
f
a wi
r
e
l
e
ss ad hoc n
e
t
w
o
r
k w
h
e
r
e n
ode
s
m
a
t
c
h i
n
i
rrel
e
va
nt
t
o
t
h
e reso
urce
–
targ
et sets [7
]. Th
ese wi
reless n
o
d
e
s are tacit to
b
e
m
o
b
ile for th
e interactio
n
system
s. Th
e p
e
rspectiv
e
conce
p
t offe
rs with the lifestyle and de
velopment of
techni
que
s of limited places wh
ose
crossi
ng poi
nts
have
sp
ecified m
a
th
e
m
atical q
u
a
lities. Officially,
A t-d
e
si
g
n
(,
b, r,
k
,
)d
efi
n
ed as
fo
llows : Giv
e
n a lim
i
t
ed
set X
of fact
ors
(ele
ments),
we
bui
ld a
group of
b s
u
bsets
of
X, known a
s
pre
v
ents
, s
u
ch
tha
t
each pre
v
e
n
t has a
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
73
5 – 7
4
2
73
8
dim
e
nsion k, each fact
or is
include
d in
r pre
v
ents a
nd each t factors are include
d
togethe
r
in e
x
actly
pre
v
e
n
ts. For e
x
am
ple, the symme
trical
Healthy
Im
perfect Block De
sig
n
(SBIB
D
)
pr
o
v
ided a
b
ove is a
(b,
r,
k,
) st
y
l
e,
whe
r
e b
= m
2
+ m
+ 1
,
r =
k
=
m
+ 1 an
d =
1
.
Ty
pi
cal
l
y
ke
y
su
bm
i
ssi
on m
a
y
not
rel
i
e
s
o
n
key
su
b
m
issio
n
resid
e
n
ce, i
n
th
is we sp
lit d
i
fferen
t
p
r
ev
en
ts
. Each
prev
en
t contain
s
m
+ 1
facto
r
s and
each
facto
r
i
s
i
n
cl
ude
d i
n
r
= m
2
preve
n
t
s
. Each
co
u
p
l
e
of
fact
o
r
s i
s
i
n
cl
ude
d i
n
e
x
act
l
y
one
pre
v
e
n
t
t
oget
h
e
r
.
We si
gni
fy
t
h
e Uni
t
a
l
by
2
– desi
g
n
(m
3
+ 1,m
2
(
m
2
−
m + 1
)
,
m
2
,
m +
1
,
1
)
o
r
b
y
(
m
3
+ 1
,
m
+
1
,
1
)
style
f
o
r
conven
i
en
ce
bene
fi
t
[
1
]
.
1
)
St
orag
e spa
c
e ov
erhea
d:
Wh
en using
th
e
su
gg
ested
innocen
t un
ital centered
ed
itio
n related
a
u
n
ital
of
pu
rc
hase m
,
each n
ode i
s
p
r
e-l
o
a
d
ed
wi
t
h
one
key
ban
d
cor
r
es
po
n
d
i
n
g t
o
one
pre
v
e
n
t
from
t
h
e st
y
l
e,
hence
,
each
node is pre
-
loa
d
e
d
with
(m
+ 1) disjoint
im
portant factors
.
T
h
e st
orage s
p
a
ce neede
d
to s
h
op
im
port
a
nt
fact
o
r
s i
s
t
h
en
l
× (
m
+ 1) w
h
ere
l
i
s
t
h
e
key
di
m
e
nsi
o
n.
2
)
Sy
stem
Sca
l
ability
:
Fro
m
d
e
v
e
lop
m
en
t, th
e co
un
t of
p
o
ssib
le k
e
y j
e
welry wh
en
using
th
e inn
o
c
en
t
u
n
ital
cent
e
re
d
pl
an i
s
n =
m
2
× (m
3+1
)
(m
+1) =
m
2
× (
m
2
−
m
+
1), t
h
is is the
n
t
h
e m
o
st of
rei
n
forced nodes.
3) Immediate
Pr
otec
te
d Connection Cover
a
ge:
W
h
en u
s
i
n
g
th
e
pr
im
ar
y
u
n
ital app
l
ying
,
w
e
k
now
that
each key is us
ed in exactly m
2
key jewelry am
ong the
m
2
× (
m
2
−
m
+ 1) p
o
ssi
bl
e
key
jewel
r
y
.
L
e
t
us
con
s
i
d
er t
w
o n
ode
s u an
d v a
r
bi
t
r
a
r
i
l
y
chos
en. T
h
e n
ode
u i
s
pre
-
l
o
a
d
ed
wi
t
h
a key
ba
nd
KR
u
of m
+ 1
di
ffe
re
nt
i
m
port
a
nt
fact
o
r
s.
Ea
ch
of t
h
em
i
s
incl
u
d
ed
i
n
m
2
−
1
ot
he
r
key
j
e
wel
r
y
am
ong
t
h
e p
o
ssi
bl
e m
2
× (m
2
−
m +
1
)
−
1
one
s.
Un
derst
a
ndi
ng t
h
at
t
w
o co
u
p
l
e
of i
m
port
a
nt
fa
ct
ors
hap
p
e
n
s t
oget
h
er i
n
exa
c
t
l
y
one
pre
v
e
n
t
,
w
e
di
sco
v
er t
h
at
pre
v
ent
s
c
ont
ai
ni
ng t
w
o
di
ff
erent
i
m
port
a
n
t
fact
ors o
f
K
R
u are abs
o
l
u
t
e
l
y
di
sj
oi
nt
.
He
nce
,
eac
h
no
de
st
o
c
ks e
x
act
l
y
o
n
e
key
wi
t
h
(m
+ 1
)
×
(m
2
−
1)
no
des
am
ong t
h
e m
2
(m
2
−
m +
1)
−
1
o
t
h
e
r possib
l
e nod
es, then
th
e po
ssi
b
ility p
c
of
d
i
scussin
g
a typ
i
cal key can
b
e
m
easu
r
ed
as fo
llows:
2
(1
)
(
1
)
2(
2
1
)
1
mx
m
Pc
mm
m
Th
e assessm
en
t o
f
t
h
is inn
o
c
en
t rem
e
d
y
rev
e
als clearly th
at th
e prim
ary ap
p
l
yin
g
fro
m
u
n
itals to
k
e
y
p
r
e-d
i
stri
b
u
t
i
o
n
g
i
v
e
s a
h
i
gher system
scal
ab
ility wh
ich
gets to
O(k4
).
Mo
reo
v
e
r,
g
i
ven
a system
d
i
men
s
io
n
n, t
h
i
s
i
n
noce
n
t
pl
an al
l
o
ws t
o
dec
r
ease t
h
e
key
ban
d
di
m
e
nsi
o
n u
p
t
o
p
4
n. H
o
we
ve
r, t
h
i
s
i
n
n
o
ce
nt
re
m
e
dy
o
u
t
co
m
e
s a lo
w k
e
y
d
i
scu
s
sin
g
po
ssib
ility wh
ich
tend
s to O(1k
).
To
b
e
ab
le to
en
h
a
n
c
e th
e k
e
y
d
i
scu
ssi
ng
p
o
s
sib
ility wh
ile k
eep
ing
an
ex
cellen
t
scalab
ility en
h
a
n
c
emen
t, we recommen
d
in
th
e n
e
x
t
area an
im
p
r
o
v
e
d
scal
abl
e
an
d
e
f
fect
i
v
e uni
t
a
l
-
b
a
sed key
pre
-
di
st
ri
but
i
o
n f
o
r WS
Ns.
3.
BACK
G
R
O
U
ND
AP
PR
OA
CH
A ne
w
uni
t
a
l
-
base
d key
p
r
e
-
di
st
ri
b
u
t
i
on
pl
an f
o
r
WS
Ns.
To b
e
abl
e
t
o
enha
nce t
h
e
ke
y
di
scussi
n
g
p
o
s
sib
ility wh
ile k
eep
ing
great syste
m
scala
b
ility, we recommen
d
to
b
u
i
l
d
th
e un
ital sty
l
e p
r
ev
en
ts and
pre-
load eac
h
node
with a
va
riety of preve
n
ts
sel
ected
in a particular way
[1].
Key Pre-dis
t
ribution:
B
e
f
o
re
t
h
e i
m
pl
em
entat
i
on
pha
se,
w
e
pr
o
d
u
ce
pre
v
ent
s
of m
pu
rc
hase
u
n
i
t
a
l
st
y
l
e, where e
ach bl
ock c
o
r
r
e
sp
on
ds t
o
a
key
set
.
W
e
pre-lo
ad
th
en
each
no
de with t ab
so
lu
tely disj
o
i
n
t
pre
v
e
n
ts where
t is a
m
e
thod
param
e
ter that we will talk
about later in t
h
is area. In lem
m
a 1,
we illustra
te the
con
d
i
t
i
on
of l
i
f
est
y
l
e
of such
t
absol
u
t
e
l
y
di
sj
oi
nt
pre
v
e
n
t
s
am
ong t
h
e u
n
i
t
a
l
prevent
s
. I
n
t
h
e basi
c ap
pr
oac
h
each node
is pre-loa
d
ed with only
one
unital pre
v
e
n
t and
we show t
h
at ea
ch
two
nodes s
h
are at m
o
st one key.
As
opposed to this,
pre-lo
adi
n
g each two
nodes with t disj
oi
nt unita
l blocks m
eans that
each t
w
o nodes talk
about
betwee
n
zero and t
2
important
f
actors
sin
ce each
two
u
n
itals
p
r
ev
en
ts talk
ab
ou
t at
m
o
st o
n
e
fact
or.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISS
N
:
2088-8708
S
u
fficien
t
Au
then
tica
tion
fo
r
En
erg
y
C
o
n
s
ump
tion
i
n
Wireless S
e
n
s
o
r
Net
w
o
r
ks (Hema
Gop
i
na
th Su
kava
si)
73
9
Fi
gu
re
3.
Pr
o
p
o
se
d arc
h
i
t
ect
u
r
e i
n
wi
rel
e
ss
s
e
ns
or a
p
pl
i
cat
i
ons
After the im
ple
m
entation phase, each two others
wh
o live nea
r
by ret
u
rn the ide
n
tifi
e
rs of their
im
port
a
nt
fact
ors i
n
or
der t
o
fi
gu
re o
u
t
t
h
e t
y
pi
cal
im
port
a
nt
fact
or
s. I
f
t
w
o
near
by
no
d
e
s t
a
l
k
abo
u
t
o
n
e o
r
m
o
re im
port
a
n
t
fact
ors
,
we
re
com
m
e
nd c
o
m
put
i
n
g t
h
e pai
r
-wi
s
e
key
as t
h
e has
h
o
f
al
l
t
h
ei
r t
y
pi
cal
im
port
a
nt
factors concat
enated t
o
eac
h othe
r [2]. T
h
e use
d
hash
operate m
a
y be SHA-1
for
e
x
am
ple. This
strategy
in
creases th
e syste
m
resilien
c
y sin
ce t
h
e en
em
y h
a
s to
barg
ai
n
m
o
re
ov
erlap
im
p
o
r
tan
t
fact
o
r
s to crack a
pr
ot
ect
ed we
b
l
i
nk.
Ot
her
w
i
s
e, wh
en
nei
g
h
b
o
r
s d
o
not
t
a
l
k
ab
o
u
t
any
ke
y
,
t
h
ey
sho
u
l
d
di
sco
v
er a
pr
ot
ect
ed
di
rect
i
o
n co
nsi
s
t
i
ng
of s
u
bse
que
nt
p
r
ot
ect
ed hy
perl
i
n
ks
[
8
]
.
The
si
g
n
i
f
i
cant
be
nefi
t
s
o
f
t
h
i
s
st
rat
e
gy
are t
h
e
en
h
a
n
cem
en
t o
f
th
e k
e
y d
i
scussin
g
p
o
s
sib
ility. As we will p
r
ov
e in
n
e
x
t
su
bsectio
n, th
i
s
strateg
y
allo
ws to
accom
p
lish a
higher protecte
d
c
o
nnection
protection si
nce
each node
is pre
-
loa
d
ed
wit
h
t disjoint pre
v
ents
.
Mo
reo
v
e
r, t
h
is strateg
y
g
i
v
e
s ex
cellen
t
syst
e
m
resilie
n
c
y th
rou
g
h
the b
l
en
d p
a
i
r
-wise key i
m
p
o
r
tan
t
facto
r
s
whi
c
h s
u
pp
ort
s
p
r
ot
ect
ed
hy
p
e
rl
i
nks
.
4.
PROP
OSE
D
APP
R
O
A
CH
In t
h
i
s
o
u
r
s
u
gge
st
ed
pr
o
g
r
a
m
,
t
h
e best
pos
si
bl
e c
ont
a
c
t
ran
g
e a
n
d
i
n
t
e
ract
i
o
n t
e
c
hni
que
we
re
mix
t
u
r
e to
imp
l
em
en
t th
e Heterog
e
n
e
ou
s
W
i
reless Ind
i
cato
r System
s l
i
festyle in
n
a
tu
re
[4
].
In HWSN, the
in
tra-cluster arrang
ing
an
d
i
n
ter clu
s
ter m
u
lti-h
o
p
red
i
r
ecti
n
g
o
f
fer t
o
take ad
v
a
n
t
ag
e t
h
e n
e
two
r
k
lifeti
m
e.
A
n
d
it is r
e
g
a
r
d
ed
as a h
i
erar
ch
al HW
SN w
ith
CH
nodes su
ch
as excellen
t
en
erg
y
an
d
p
r
ov
id
ing
ou
t
cap
ab
ilities
th
an
regu
lar SNs
[9
].
4.
1.
Prop
osed
Al
g
o
ri
thm
f
o
r Pr
ocessi
n
g
E
n
er
gy
Ser
v
i
c
i
n
g i
n
A
ppl
i
c
a
t
i
o
n
Devel
o
pme
n
t
CH Execution
Get next event
If event is T
D
timer then
determine radio range to maintain CH connectivity
determine optimal T
IDS
, m, m
s
, m
p
b
y table lookup based on the current
estimated density, CH radio range and compromise rate
notify SNs within the cluster of the new optimal settings of T
IDS
and m
else if event is query arrival then
trigger multipath routing using m
s
and m
p
else if event is T
clustering
timer then
perform clustering
else if event is T
IDS
timer then
For each neighbor CH
If selected as a voter then
Execute voting based intrusion detection
else // event is datapacket arrival
follow multipath routing protocol design to route
Our sugg
ested strateg
y
g
i
v
e
s so
lu
tion
to
come u
p
with
as a
m
a
rk
etin
g
p
r
ob
lem
s
to
b
a
lan
ce en
erg
y
in
take
acros
s al
l
no
de
s i
n
t
h
e e
n
t
i
r
e
het
e
r
oge
ne
ous
sens
or
net
w
o
r
ks.
Th
o
u
g
h
i
n
t
h
i
s
d
o
c
u
m
e
nt
, we
su
g
g
est
t
w
o-t
i
e
r
H
W
SN
wi
t
h
t
h
e
ob
ject
i
v
e
o
f
t
a
ke a
d
vant
a
g
e
on
ne
t
w
ork
life-tim
e wh
ile satisfyin
g en
erg
y
co
n
t
ro
l and
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
73
5 – 7
4
2
74
0
co
v
e
rag
e
go
als [5
]. Th
ey id
entified
th
e
m
a
x
i
m
u
m d
e
n
s
ity ratio
o
f
th
e two
tier’s no
d
e
s t
o
in
crease th
e
p
r
o
g
ram
life-ti
m
e
.
5.
RESULTS
A
N
D
DI
SC
US
S
I
ON
Mu
ltip
ath
red
i
rectin
g
is reg
a
rd
ed
an
effective p
r
o
cedure for mistak
e an
d
attack
p
a
tien
c
e to
i
m
p
r
ov
e
d
e
tails d
i
stributio
n
in
WSNs
[10
]
. The essen
ce is th
at
th
e p
r
o
b
a
b
ility o
f
at least o
n
e
directio
n
attain
in
g
t
h
e
drai
n node
or
base
place im
proves as
we
ha
ve m
o
re rout
es doing
data del
i
very.
Wh
ile
m
o
st before analysis
targ
eted on u
s
in
g
m
u
ltip
ath
red
i
recti
n
g to
en
h
a
n
c
e stab
ility, so
m
e
in
terest h
a
s b
e
en
co
m
p
en
sated
t
o
u
s
i
ng
m
u
ltipath re
directing to
acce
pt inside
r attac
k
s [11]. T
h
ese
analysis, howev
er, m
o
stly ignored t
h
e tra
d
eoffs
'
b
e
tween
QoS ob
tain
v
s
.
power
in
tak
e
wh
ich
can
n
e
g
a
tively red
u
c
e th
e
program
life-ti
me.
Fi
gu
re
4.
Pac
k
et
Loss
wi
t
h
re
spect to tim
e
analysis
User I
n
ter
f
ac
e:
In
t
h
i
s
c
o
m
ponent
,
we
have
t
o
m
a
ke t
h
e
use
r
i
n
t
e
rface
f
o
r
est
a
bl
i
s
hi
n
g
t
h
e
relations
hip
be
tween t
h
e em
ailer and t
h
e rec
e
iver. He
re th
e
cust
om
er ha
s t
o
get
rea
d
y
t
h
e
det
a
i
l
s
t
h
at
ha
ve t
o
send
to th
e p
a
rticu
l
ar lo
cation. Fo
r
eve
r
y
dea
l
, use
r
i
n
t
e
rfac
e
i
s
t
h
e
p
r
i
m
ar
y
aspect
fo
r
de
vel
o
pi
n
g
c
o
nn
ect
i
o
n
b
e
tween
th
e emailer an
d
th
e recip
i
en
t.
After d
e
v
e
lop
i
ng
t
h
e connection, the em
aile
r ha
s to get
ready
fo
r the
d
e
tails, wh
ich
h
e
wan
t
s to
d
e
l
i
v
e
r t
o
th
e p
a
rticu
l
ar l
o
catio
n.
Calcul
ate the direction fe
ature:
I
n
t
h
i
s
co
m
ponent
, t
h
e d
i
rect
i
on f
unct
i
on
has t
o
m
easure by
l
o
a
d
ad
ap
ter to redu
ce t
h
e
d
i
rectio
n wait and
bu
nd
le
wait, th
us m
i
nim
i
zing
su
fficient
dif
f
i
cult fo
r
reo
r
d
e
rin
g
the
p
ack
ets at th
e lo
cation
.
Th
i
s
d
e
tails
h
a
s t
o
d
e
liv
er to
t
h
e traffic
sp
littin
g
elem
en
t an
d d
i
rectio
n selecto
r
ele
m
ent. The path calculatin
g is in accorda
n
ce with the
fill
controlling se
rver known
as cell breathing serve
r
,
whic
h successfully looks
for t
h
e
direction
function by
usi
n
g m
u
ltipath int
e
raction.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
S
u
fficien
t
Au
then
tica
tion
fo
r
En
erg
y
C
o
n
s
ump
tion
i
n
Wireless S
e
n
s
o
r
Net
w
o
r
ks (Hema
Gop
i
na
th Su
kava
si)
74
1
Fi
gu
re
5.
Pac
k
et
s effi
ci
ency
i
n
real
t
i
m
e
dat
a
t
r
ansm
i
ssi
on
of
wi
rel
e
ss
se
n
s
or
net
w
o
r
ks
Splittin
g
the p
a
cke
t
s:
Accord
ing
to
th
e d
i
rectio
n
d
e
tails th
at is sen
t
b
y
th
e lo
ad
ad
ap
ter, th
e bun
d
l
e
will b
e
sp
itted to
d
e
liv
er acro
ss th
e
p
a
th
Selectio
n
.
Th
e sp
itted
p
a
ck
ag
es are sen
t
to
th
e d
i
rectio
n
selecto
r
com
pone
nt, ac
cording to the
direction
details, the
path
selector elem
ent will select the direction a
n
d
delive
r
the pac
k
et through th
e program
. W
e
desc
ribe the following criteri
a for
packet loss
in data accuracy of
wi
rel
e
ss se
ns
or
net
w
o
r
ks
.
We no
tice th
at
we pro
v
i
d
e
th
e co
mm
o
n
syste
m
scalab
ilit
y (nu
m
b
e
r o
f
n
o
d
e
s) wh
en
usin
g
UKP*
pl
an. O
n
t
h
e o
t
her han
d
, we est
i
m
a
t
e
t
h
e
com
m
on
pr
ot
ect
ed
di
rect
i
o
n d
u
rat
i
o
n depe
n
d
i
ng on
m
odel
s
. W
e
rel
a
t
e
i
n
t
h
ese
m
odel
s
t
o
t
h
e out
c
o
m
e
s gi
ven i
n
or
der
t
o
creat
e a l
i
n
es i
m
pl
em
entat
i
on m
odel
whi
c
h
gua
ra
nt
ees t
h
e sy
st
em
phy
si
cal
connect
i
o
n
and
pr
ot
ect
i
o
n
.
Num
e
ri
cal
out
com
e
s sho
w
t
h
at
t
h
e uni
t
a
l
-
base
d
k
e
y pre-d
i
stri
bu
tio
n p
l
an
UKP*
im
p
r
ov
es t
h
e system
scal
ab
ility o
v
e
r the SBIB
D-KP an
d th
e
Trad
e-KP
p
l
an
whi
l
e
keepi
n
g
great
pr
ot
ect
ed
co
nnect
i
o
n
pr
ot
ect
i
on.
F
o
r e
x
am
pl
e, t
h
e sy
st
em
hi
ghest
p
o
ssi
bl
e
di
m
e
nsi
on i
s
im
pro
v
ed
by
a fact
or
of 3
and
4.
8 w
h
en
t
h
e key
ri
ng
di
m
e
nsi
on i
s
sim
i
l
a
r t
o
68 a
nd
14
0 res
p
ec
t
i
v
el
y
com
p
ared t
o
t
h
e SBIB
D-KP plan.
In a
ddition, we m
a
in
t
a
i
n
a hi
ghe
r c
o
nne
ct
i
on
o
v
er
0
.
6
3
w
h
i
c
h
g
u
a
r
ant
ees a
lo
w
r
e
g
u
l
ar pro
t
ected
d
i
r
ectio
n dur
atio
n which
do
es no
t sur
p
ass 1.37
.
6.
CO
NCL
USI
O
N
We re
veal
ed t
h
at
a basi
c ap
p
l
y
i
ng fr
om
uni
t
a
l
s
t
o
key
pre
-
di
st
ri
b
u
t
i
on al
l
o
ws ac
hi
evi
n
g
a
very
g
r
eat
syste
m
scalab
i
lity
wh
ile o
f
fering
a lo
w d
i
rect secu
re conn
ectio
n
p
r
o
t
ectio
n
.
We sugg
ested
th
en
an
efficien
t
scalab
le un
ital-b
a
sed
k
e
y pre-d
i
stri
b
u
tion
sch
e
m
e
o
ffe
ri
ng
great system scalab
ility an
d
ex
cellen
t
p
r
o
t
ected
co
nn
ection
p
r
otectio
n
.
We talk
abou
t th
e so
lu
tio
n
p
a
ram
e
te
r and
we recommen
d
su
fficien
t
prin
cip
l
es
o
f
feri
ng
a v
e
ry ex
cellen
t
trad
e-off
b
e
tween
n
e
two
r
k scalab
ility
an
d
pro
t
ected
con
n
ection
.
We
p
e
rform
e
d
syste
m
atic
researc
h
an
d m
odel
s
t
o
eval
uat
e
ou
r new
rem
e
dy
t
o
curre
nt
ones
,
t
h
e
resul
t
s
reveal
ed t
h
at
ou
r ap
pr
oac
h
pr
o
v
i
d
es a
n
e
x
cel
l
e
nt
p
r
ot
ect
ed co
ve
rage
of
ext
e
ns
iv
e
syste
m
s with
a lo
w k
e
y
st
orage expe
nse and
a
n
excellent syste
m
resiliency.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJEC
E V
o
l
.
6, No
. 2, A
p
ri
l
20
16
:
73
5 – 7
4
2
74
2
REFERE
NC
ES
[1]
“A
Highly
Scalable Key
Pre-distributi
on Scheme for Wireless Se
nsor Networ
ks”, b
y
Wa
lid
Bechkit ‡
,
Yaci
ne
Challal ‡
,
Abdelmadjid Bouabd
allah ‡
and
Vahid
Tarokh
,
in
IE
E
E
Tr
ans
act
ions
o
n
Wireless Communications, Vol:
12 No: 2
Year
2
013.
[2]
S. Ruj, A. Nay
a
k, and I
.
St
ojmenovic. Fully
secure
pairwis
e
and
triple key
d
i
stribu
tion in wir
e
less sensor networks
using combinato
r
ial designs. In
I
EEE
INFOCOM
, pag
e
s 326–330
, 2011.
[3]
W
.
Bechkit
,
Y.
Challa
l,
and A.
Bouabdallah. A new scalab
le k
e
y
pr
e-distribu
tio
n scheme for wsn. In
IEEE ICC
C
N
,
pages 1–7, 2012
.
[4]
J.
Zhang and V.
Varadhar
a
jan.
Wireless sensor network key
mana
g
e
ment survey
and
tax
onomy
.
Journal of
Network and
Co
mputer Applications
, 33(2):63
–
75, 2010
.
[5]
B. M
aal
a, Y. Ch
all
a
l,
and A. Bo
uabdal
l
ah. H
e
ro:
Hierarch
cal key management pr
otocol
for h
e
tero
geneous wsn. In
IFIP WS
AN
, pag
e
s 125–136, 200
8.
[6]
B. Umakanth an
d J. Damodhar “Detection of
Energ
y
dr
ain
i
ng attack using EWMA in Wireless Ad Hoc Sensor
Networks”,
In
ter
national Journal
of Engine
ering
Trends and Tech
nology (
I
JETT)
, Volume 4 Issue
8, August 2013
.
[7]
G.
Acs,
L.
Butty
a
n,
and I. Vajda,
“Provably
S
ecur
e
On
-Demand Source Routing
in
Mobile Ad Ho
c
Networks,”IEEE Trans. Mobi
le Computing,
vol. 5,
no
.
11
, pp
. 15
33-1546, Nov. 2
006.
[8]
Chaudhari H.C.
and Kadam L.
U “Security
in Ad Hoc Networks”
-
Intern
ational Journal of Networking Volume 1,
Issue 1, 2011
, pp
- 04-16.
[9]
S. Ruj and B. Ro
y
.
Key
pred
istr
ibution using co
mbinator
ial designs for grid-group depl
o
y
men
t
scheme in wireless
sensor networks.
ACM Transactions on
Sensor N
e
tworks
, 6
:
4
:
1–
4: 28, Januar
y
2
010.
[10]
T. Aur
a
, “Dos-R
esistant Au
thentication with
Client Pu
zzles,”Proc. Int’
l Workshop Security
Proto
c
ols, 2001.
[11]
J. Bellardo and S. Savage,
“802.11 Denial-of-Service Attacks: R
eal Vuln
er
abilities and Practic
al Solutions,”Proc.
12th Conf. USENIX Security
, 2
003.
[12]
T. Choi, H.B.
Achar
y
a, and M.
G. Gouda. Th
e best key
i
ng pr
otoc
ol for senso
r
networks. In
IEEE WOWMO
M
,
pages 1–6, 2011
.
[13]
M. Doddavenkatappa, M.C. Chan, and
A.
L. Ananda. A du
al-r
adio framework
for
mac proto
c
ol
implementation
in
wireless sensor
networks. In
IEEE ICC
, pag
e
s 1–
6, 2011
.
[14]
Detection fo
r Wireless Sensor Networks”- SenS
y
s
’09, November 4
–
6, 2009, Berkeley
, CA, USA.
[15]
J.
W.
Bos,
D.
A.
Osvik,
and D.
St
efan, “
F
as
t Im
plem
entations
of AES
on Various
P
l
atform
s
,
” Cr
y
p
tolog
y
ePrintsArchive,
Report 2009
/ 50
1,
http://eprin
t.iacr.org
, 2009
.
[16]
D. Berns
t
ein
an
d P
.
S
c
hwabe, “
N
ew AES
S
o
ft
ware S
p
eed
Re
c
o
rds,”Proc. Nin
t
h Int’l Conf. Cr
ypto
log
y
in Indi
a:
Progress in Cr
y
p
tolog
y
(INDOCRYPT), 2008.
Evaluation Warning : The document was created with Spire.PDF for Python.