Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
C
om
put
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
8
, No
.
6
,
Decem
ber
201
8
, p
p.
4684
~
4692
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
8
i
6
.
pp
4684
-
46
92
4684
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
FSDA: F
ra
m
ewo
rk for S
ecur
e Data Ag
gregation in
Wirele
s
s
Senso
r Netwo
rk
for Enh
ancing
Key Ma
nagemen
t
Jyoti
M
e
tan
1
, K
.
N
.
N
arashi
nha M
urthy
2
1
Depa
rtment of
Com
pute
r
Scie
n
ce
&
Enggi
ne
ering,
ACS
Coll
ege of
Eng
ineeri
ng
,
India
2
Facul
t
y
of Engi
nee
ring
,
Chr
ist U
nive
rsit
y
,
Indi
a
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Ma
r 7
, 2
01
7
Re
vised
Ju
l
14
,
201
8
Accepte
d
J
ul
28
, 2
01
8
An
eff
ective
ke
y
m
ana
g
ement
play
s
a
cru
c
ia
l
role
in
impos
ing
a
resi
li
en
t
sec
urity
t
ec
hn
iq
ue
in
W
ire
le
ss
Sensor
Network
(W
SN
).
After
rev
ie
wing
the
exi
sting
appr
oa
che
s
of
ke
y
m
ana
gement
,
it
i
s
conf
irmed
th
at
exi
sting
appr
oac
hs
do
es
not
offe
r
good
c
over
age
on
all
p
ote
ntial
se
cur
i
t
y
bre
ac
h
es
in
W
SN
.
W
it
h
W
SN
bei
ng
esse
nti
al
par
t
of
In
te
rne
t
-
of
-
Thi
ngs
(IoT
),
th
e
exi
sting
appr
oa
che
s
of
ke
y
m
ana
gement
c
an
def
ini
t
ely
not
a
ddre
ss
such
sec
urity
bre
ac
he
s.
The
r
efo
re
,
this
pape
r
int
rodu
c
es
a
Fram
ework
for
Secur
e
Data
Ag
gre
ga
tion
(FS
D
A)
tha
t
h
y
bridizes
t
he
publi
c
k
e
y
enc
r
y
pti
on
m
ec
hani
sm
in
or
der
to
obtain
a
n
ovel
k
e
y
m
ana
g
e
m
ent
s
y
st
em.
Th
e
proposed
s
y
stem
does
not
ta
rge
t
an
y
spec
i
fic
at
t
ac
ks
but
is
widely
app
li
c
ab
le
for
both
int
ern
al
and
exte
rna
l
at
t
ac
ks
in
W
SN
owing
to
i
ts
design
princ
ip
le
.
Th
e
stud
y
outc
om
e
exhi
b
its
tha
t
proposed
FS
DA
offe
rs
hi
ghl
y
r
educed
co
m
puta
ti
ona
l
burde
n,
m
ini
m
al
del
a
y
,
le
ss
ene
rg
y
consu
m
pti
on,
and
highe
r
data
tra
nsm
ission
per
fora
nce
in
cont
r
ast
to
fre
qu
ency
used
enc
r
y
pt
ion
sche
m
es
in
W
SN
.
Ke
yw
or
d:
Encr
y
ption
Key m
anag
em
ent
Secu
rity
W
i
reless
Senso
r
N
et
w
ork
Copyright
©
201
8
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
:
Jyoti
Met
an
,
Dep
a
rtm
ent o
f C
om
pu
te
r
Scie
nce & E
nggin
e
erin
g,
ACS c
ollege
of e
ng
i
neer
in
g,
Ba
ng
al
or
e,
In
di
a
.
Em
a
il
:
j
yotim
e
ta
n@gm
ail.co
m
1.
INTROD
U
CTION
A
wireless
network
is
al
way
s
shro
uded
by
diff
e
re
nt
f
or
m
s
of
net
wor
king
chall
en
ges
t
ha
t
no
t
only
aff
ect
s
c
omm
u
nicat
ion
proce
ss
ab
ou
t
al
s
o
e
qu
al
ly
aff
e
ct
s
the
sec
ur
it
y
fea
tures
[1
]
.
F
r
om
diff
ere
nt
f
orm
s
of
wireless
netw
orks,
Wireless
Sensor
Netw
ork
(
WSN
)
is
on
e
of
the
m
os
t
su
ccessf
ul
de
p
loym
ents
in
com
m
ercial
m
ark
et
.
A
se
ns
or
y
node
assist
s
in
pe
r
form
ing
data
a
ggreg
at
i
on
f
ro
m
the
e
nvir
on
m
ent
w
he
re
it
is
com
plete
ly
exp
ose
d
to
s
warm
of
at
ta
cker
s.
Til
l
la
st
decad
e,
the
re
has
be
en
va
rio
us
stud
ie
d
associat
ed
with
at
ta
cks
[2
]
an
d
secur
it
y
so
luti
on
s
[3
]
,
[
4]
but
no
ne
of
them
are
cl
aim
ed
t
o
be
10
0%
res
il
ie
nt
against
al
l
the
at
ta
cks.
Ma
jor
it
y
of
the
e
xisti
ng
a
ppro
ac
he
s
towa
rds
sec
uri
ty
in
W
S
N
a
re
m
ai
nly
cryptographic
in
na
ture
wh
e
reas
the
re
al
so
exists
stud
ie
s
th
at
are
non
-
c
rypt
ogr
aph
ic
in
natu
r
e
e.g
.
[5
]
,
[6
]
.
The
c
rypto
grap
hic
appr
oach
es
m
ai
nly
deals
w
it
h
key
m
ana
gem
ent
syst
e
m
fo
ll
ow
e
d
by
it
erati
ve
encr
ypti
ons
us
i
ng
ei
ther
sy
m
m
e
tric
or
asy
m
m
e
tric
keys
wh
il
e
non
-
cr
yptogra
ph
ic
ap
proac
hes
deals
with
obser
vati
on
of
ce
rtai
n
f
or
m
of
sign
ific
a
nt
be
ha
viour
of
node
s
fo
ll
owe
d
by
form
ulati
on
of
ru
le
set
to
off
er
infe
re
nce
to
su
c
h
be
ha
viour
i
n
te
rm
s o
f
m
alicio
us o
r re
gu
la
r
patte
rn.
In
la
st
5
ye
ar
s,
th
ere
ha
s
bee
n
var
i
ou
s
form
s
of
im
pr
ovem
e
nt
in
WSN
w
he
re
heter
ogen
e
it
y
is
fu
rther
stud
ie
d
in
ord
er
to
m
ake
it
well
pr
e
pa
red
to
be
us
e
d
i
n
reconfi
gura
ble
netw
orks
li
ke
In
te
rn
et
-
of
-
T
hings
(IoT)
[
7].
I
oT
is
com
plete
a
new
c
on
ce
pt
to
desig
n
a
sm
art
ci
ty
and
cal
ls
fo
r
m
ai
nly
integrati
ng
WSN
with
pervasive
en
vi
ronm
ent
li
ke
cl
oud
c
om
pu
ti
ng
[
8].
H
oweve
r,
the
biggest
se
cur
it
y
co
nce
rn
in
this
re
gards
are
i)
the
at
ta
cks
stu
died
in
WSN
are
ver
y
dif
fere
nt
f
r
om
that
in
cl
oud
e
nvir
onm
ent,
w
hich
has
m
os
t
pote
ntial
to
induce
colla
te
r
al
network
da
m
age,
ii
)
the
translat
io
n
m
ec
han
ism
of
co
nt
ro
l
m
essage
(g
ene
rated
fro
m
qu
ery
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
FS
DA: Fr
am
e
work fo
r
Secu
r
e Data
Ag
gr
e
gati
on in W
ire
le
ss S
e
nsor Net
w
or
k f
or
E
nhanc
ing
... (
Jyoti
M
et
an
)
4685
syst
e
m
)
is
qu
it
e
chall
en
ging
t
o
be
reali
zed
if
hete
roge
neou
s
W
S
N
is
i
ntegr
at
e
d
with
cl
oud
(at
pr
ese
nt
Io
T
is
i
m
ple
m
ented
ei
ther
in
lo
w
scal
e
network
or
in
ho
m
og
e
nous
netw
ork),
ii
i)
identific
at
ion
of
at
ta
cks
from
ei
ther
side
is
qu
it
e
di
ff
ic
ult
a
nd
has
good
c
han
c
e
of
bypassi
ng
an
y
firew
al
l
syst
e
m
if
the
sec
uri
ty
pr
ot
oco
ls
does
n’
t
hav
e
wide
co
nsi
der
at
io
n
of
it
s
env
ir
onm
ent
al
par
am
et
ers,
and
iv
)
cost
ef
f
ect
iveness
is
not
e
m
ph
asi
zed
in
IoT
nodes
as
m
ajo
rity
of
t
he
I
oT
nodes
do
ha
ve
fair
po
ssibil
it
ie
s
of
r
eso
ur
ces
w
he
n
dem
and
ed
(
un
li
ke
conve
ntion
al
WSN).
T
her
e
a
re
al
so
va
rio
us
stud
ie
s
on
I
oT
that
discu
ses
a
bout
sec
ur
it
y
im
pr
ov
em
ent
but
ve
ry
le
ss
work
has
been
act
uall
y
carried
out
ti
ll
date
owin
g
to
the
novelty
of
the
te
chnolo
gy
[9
]
,
[
10]
.
W
it
h
ne
w
le
vels
of
feat
ures
bein
g
i
ncor
porated
withi
n
Io
T
there
is
on
e
thin
g
that
is
go
i
ng
to
be
al
ways
the
re
an
d
that
is
da
t
a
ag
gr
e
ga
ti
on
.
U
nlike
c
onve
ntio
nal
W
S
N,
I
oT
offers
data
a
ggre
gation
from
on
ly
reg
ist
ere
d
node
s
bu
t
with
new
proliferati
on
of
m
ob
il
e
no
des
it
is
ve
ry
li
kely
th
at
ad
op
ti
on
of
m
ob
il
e
nodes
will
be
le
ver
a
ge
d
for
perform
ing
d
y
nam
ic
d
at
a aggreg
at
io
n.
Hen
ce
,
an
e
ffec
ti
ve
key
m
a
nag
em
ent
sche
m
e
is
hig
hly
dem
and
s
in
th
is
.
Norm
al
l
y,
the
biggest
chall
enge
in
form
ing
a
no
vel
key
m
anag
em
e
nt
te
chn
iq
ue
is
to
sel
ect
the
p
ro
ces
s
of
gen
e
rati
on
of
key,
wh
i
c
h
has
to
be
m
oti
vated
f
ro
m
certai
n
existi
ng
e
ncr
y
ption
sc
he
m
e.
Un
f
or
t
un
at
el
y,
existi
ng
en
crypti
on
sche
m
es
are
too
sp
eci
fic
of
a
tt
acks
and
he
nce
their
ap
pl
ic
abili
ty
is
quit
e
nar
rowe
d
[
11
]
-
[
13
]
.
The
r
e
is
a
need
of
su
ch
desig
n
pr
i
ncipl
e
that
can b
e
e
qu
al
ly
ap
plica
ble
f
or
r
esi
sti
ng
intr
us
i
on
s
i
n
WSN. H
ence
, w
e
intr
oduce one
s
uch
so
luti
on
by
ha
rn
es
sin
g
the
po
te
ntial
featur
es
of
public
key
encr
ypti
on
syst
em
in
or
de
r
to
ge
ne
rate
a
li
gh
twei
ght
ci
pheri
ng
poli
cy
that
can
be
cl
ai
m
ed
f
or
secu
re
key
m
anag
em
ent
sc
hem
e
in
WSN.
We
al
s
o
show
that
it
is
feasible
fo
r
co
ntr
uct
a
robu
st
enc
ry
ption
sc
hem
e
t
hat
is
le
ss
it
era
ti
ve
and
m
or
e
pro
gr
essi
ve
wi
th
out
m
uch
dem
and
s
o
f
res
ources
for
it
s
ex
ecut
ion
.
Sect
io
n
1.1
discu
sses
a
bout
the
existi
ng
li
te
ratur
es
wh
e
r
e
diff
e
re
nt
te
chni
qu
es
are
disc
us
se
d
f
or
dete
ct
ion
schem
es
us
ed
in
pow
er
transm
issi
on
li
nes
f
ollowe
d
by
discuss
i
on
of
researc
h
pro
bl
e
m
s
in
Sect
io
n
1.2
a
nd
pr
opose
d
so
l
ution
in
1.3.
Sect
io
n
2
discusse
s
about
al
gorithm
i
m
p
lem
entat
ion
f
ol
lowed
by
dis
cussion
of
re
s
ult
analy
sis
in
Sect
ion
3.
Fi
nally
,
the
c
on
cl
us
ive
rem
ark
s ar
e
pr
ov
i
ded in
Sect
ion 4
.
1.1.
Back
ground
This
sect
io
n
updates
resea
rc
h
a
ppro
ac
hes
t
ow
a
r
d
s
stre
ngthin
g
key
m
an
agem
ent
fo
ll
owed
by
our
pr
i
or
in
vestiga
ti
on
[14].
The
w
ork
ca
rr
ie
d
out
by
W
a
ng
et
al
.
ha
s
presented
a
cl
ust
ering
ap
proac
h
for
i
m
pr
ovin
g
sec
ur
it
y
in
WSN
u
si
ng
a
ver
if
ic
at
ion
of
m
essage
[15]
.
Por
a
m
bag
e
et
al.
ha
ve
i
ntrod
uc
ed
a
n
authe
ntic
at
ion
sch
em
e
fo
r
im
pro
ving
key
m
anag
em
ent
on certi
ficat
es
[16
]
.
Stud
y o
n
m
ob
il
e
netw
orks
w
it
h
a
n
e
m
ph
asi
s
of
ke
y
m
anag
em
e
nt
was
car
ried
ou
t
by
Ka
ng
et
al.
[1
7].
T
he
auth
ors
ha
ve
us
e
d
key
s
har
i
ng
appr
oach
a
s
w
el
l
as
rek
ey
in
g
appr
oach
t
hat
is
cl
aim
ed
to
m
ai
n
ta
ined
bette
r
f
orward
-
bac
kwar
d
sec
recy.
Lee
et
al.
hav
e
prese
nt
ed
a
ty
pical
e
ncr
y
ption
sche
m
e
m
eant
fo
r
s
ecur
i
ng
ubiq
uitou
s
de
vices
[18]
.
Che
n
et
al.
ha
ve
pr
ese
nted
t
heir
key
m
anag
em
ent
schem
e
us
ing
sym
m
e
tric
encr
y
ption
a
ppro
ac
h
ap
plica
bl
e
on
hete
roge
neou
s
netw
ork
[
19
]
.
Pereira
et
al.
hav
e
in
vestiga
te
d
the
sec
ur
it
y
stren
gth
of
diff
e
re
nt
enc
ry
ption
te
ch
niqu
es
on
In
te
r
net
-
of
-
T
hi
ng
s
(IoT
)
[
20
]
.
Ado
ption
of
El
li
ptica
l
Curv
e Crypto
gr
a
phy
has
been
see
n
in w
ork
of
I
bra
him
and
Dal
kili
c
for
sec
ure
transm
issi
on
of
node
ta
gs
I
D
us
in
g
m
utu
al
a
uth
e
ntica
ti
on
proces
s
[21]
.
Sa
rk
a
r
an
d
Mukh
erj
ee
hav
e
discuss
e
d
their
key
P
re
di
stribu
ti
on
sc
hem
e
wh
ic
h
ha
s
bee
n
re
peat
edly
us
e
d
ev
e
n
in
past
with
fe
w
e
vid
e
nc
es
of
ben
c
hm
ark
in
g
[22]
.
Qi
et
al.
hav
e
im
ple
m
e
nted
a
com
pr
e
ssive
sen
sin
g
al
ong
with
blo
c
k
enc
ryptio
n
of
8
-
bit
integer
on
se
nsor
data
[
23
]
.
Wu
et
al.
have
presente
d
a
fr
am
ewo
r
k
des
ign
t
hat
pe
r
for
m
s
identific
at
i
on
of
at
ta
ck
s
us
in
g
vi
rtuali
zat
ion
an
d
softwa
re
de
fi
ned
netw
orks
[
24
]
.
Deng
et
al
.
hav
e
us
e
d
a
stochastic
ap
pr
oach
for
secu
rin
g
physi
cal
la
ye
rs
i
n
W
S
N
us
i
ng
m
ul
ti
ple
sink
appr
oach
[
25]
.
Um
ar
et
al
.
hav
e
us
e
d
a
cros
s
-
la
ye
r
base
d
ap
proac
h
that
al
lows
the
trust
fact
or
to
be
us
e
d
al
on
g
with
f
uzzy
lo
gic
i
m
ple
m
ent
at
ion
in
orde
r
to
offe
r
resou
rce sec
ur
i
ty
in W
S
N
[
26]
.
Near
ly
sim
i
la
r
appr
oach
on
physi
cal
la
ye
r
as
well
as
tru
s
-
base
d
ap
proac
h
of
sec
ur
it
y
ha
s
al
so
bee
n
carried
out
by
Zh
u
et
al
.
[
27
]
as
well
as
Qin
et
al
.
[
28
]
.
S
hi
n
et
al.
ha
ve
presented
a
r
oute
optim
iz
ation
-
bas
e
d
appr
oach
us
in
g
trust
fact
or
f
or
fa
ult
tolera
nt
i
m
ple
m
enta
ti
on
of
com
m
u
nicat
ion
sec
ur
i
ty
in
Io
T
[29]
.
Gua
n
and
Ge
hav
e
us
e
d
a
ra
ndom
m
od
el
ing
a
ppr
oach
us
in
g
pro
bab
il
it
y
schem
e
fo
r
resis
ing
j
am
m
ing
a
tt
ack
in
WSN
[
30
]
.
D
a
i
et
al
.
hav
e
pr
esented
a
ve
rif
ic
at
ion
m
e
tho
d
on
it
s
enc
od
i
ng
syst
e
m
fo
r
m
ini
m
iz
ing
th
e
cost
involve
d
in
se
cur
e
qu
e
ry
pro
cess
[31]
.
The
m
echan
ism
us
es
hash
i
ng
a
nd
sy
m
m
et
ric
en
crypti
on.
Al
-
T
urjm
an
et
al
.
ha
ve
pr
e
sented
a
key
a
ggrem
ent
strat
eg
y
hat
offe
rs
secur
e
com
m
un
ic
at
ion
us
i
ng
m
ob
il
e
sink
s
with
a
n
ai
d
of
el
li
ptica
l
curve
cry
ptogra
phy
[32]
.
A
f
ram
ewo
r
k
for
in
vestigat
ing
t
he
sec
ur
it
y
streng
t
h
of
harvester
node
is
d
esi
gned by V
o
et
al.
[
33
]
.
The
aut
hors
ha
ve
al
so
prese
nted
a
sche
duli
ng
ap
pro
ac
h
f
or
im
pr
ov
i
ng
t
he
secu
rity
upon
ph
ysi
cal
la
ye
r.
L
u
et
al.
ha
ve
pr
e
sente
d
a
disc
us
sio
n
of
va
rio
us
c
on
ven
ti
onal
e
ncry
ption
schem
es
us
e
d
i
n
WSN
[34]
.
Ther
e
is
var
i
ous
scal
e
of
sec
ur
it
y
appr
oaches
us
e
d
in
im
p
rovin
g
key
m
a
nag
em
ent
te
chn
iq
ues
in
rece
nt
tim
es
with
m
or
e
dom
inancy
of
usng
el
li
ptica
l
curve
c
rypt
os
yst
e
m
,
Secur
e
d
Hash
Algorith
m
(S
HA
)
,
Advanc
e
d
Encr
y
ption
St
and
a
r
d
(
AES),
et
c.
H
oweve
r
,
al
l
of
these
appr
oach
es
ar
e
al
so
featu
re
d
by
pitfal
ls
t
hat
are
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4684
-
4692
4686
require
d
to
be
add
re
ssed
in
order
to
ob
ta
i
n
suprem
e
s
ec
ur
it
y.
The
ne
xt
sect
ion
br
ie
f
s
about
su
c
h
pitfal
l
fo
ll
owe
d by pr
opos
e
d
s
olu
ti
on
for
a
ddressi
ng s
uc
h pit
fall
s.
1.2.
Ident
ific
at
i
on
of
Issues
The u
naddr
e
ss
ed
iss
ues
e
xplo
red after
r
e
vie
wing e
xisti
ng a
ppr
oach
es
are:
a.
Usage
of
com
plex
a
nd
highl
y
it
erati
ve
crypto
gr
a
phic
ap
proac
hes
e
nsure
s
higher
de
gr
e
e
of
sec
uri
ty
bu
t
do
e
sn’t em
ph
asi
ze o
n
it
s appl
ic
abili
ty
o
n
se
nsors
w
it
h c
on
st
raints
of r
es
our
ces.
b.
Ell
ipti
cal
Curv
e
Cryptogra
phy
of
fers
li
gh
tw
ei
gh
t
enc
ryptio
n
by
co
ntro
ll
in
g
the
m
ini
m
um
key
siz
e
bu
t
on the
oth
e
r
ha
nd it
also i
ncr
e
ases the ci
ph
e
r
ed
m
essage siz
e that res
ults in
co
m
plexity
.
c.
Existi
ng
ap
pro
aches
of
di
gital
sign
at
ur
es
do
esn’
t
discu
ss
t
he
c
os
t
of
ce
rtific
at
e
revocati
on
w
hich
is
no
t
on
ly
e
xp
e
ns
i
ve
aff
ai
r bu
t al
s
o offer
insec
ur
it
y o
f
it
s private
ke
ys.
d.
Usa
ge
of
dig
i
ta
l
sign
at
ur
es
has
higher
i
nvolv
em
ent
of
com
pu
ta
ti
onal
tim
e
that
cou
l
d
intr
oduc
e
sign
ific
a
nt am
ount
of n
et
wor
k delay
and
he
nce
no
t m
uch s
upportive
for e
m
erg
ency a
ppli
cat
ion
.
Ther
e
f
or
e,
the
sta
tem
ent
of
th
e
pro
blem
is
“
Con
str
uctin
g
a
uniq
ue
e
ncry
pt
ion
sc
he
me
usi
ng
public
key
crypt
ograph
y
th
at
c
ou
l
d
offer
li
ghtw
ei
gh
t
fe
atu
re
s
w
it
h
m
axim
um
cover
ag
e
of
s
ecurit
y
sta
ndards
i
n
wi
rel
ess en
vi
ronme
nt
of se
ns
ory
ap
plicati
on
.
” Th
e
n
e
xt sect
ion
outl
ines
propose
d
s
olu
ti
on.
1.3.
Prop
os
ed
S
olu
tion
This
pap
e
r
pr
e
sents
a
n
e
xten
ded
ve
rsion
of
our
previ
ous
inv
est
igati
on
[
35
]
t
ow
a
rds
a
novel
key
dep
l
oym
ent
st
rategy.
T
his
pa
per
furthe
r
optim
iz
es
the
secur
it
y
featu
re
by
hybri
dizin
g
the
po
te
ntial
s
of
el
li
ptica
l
cur
ve
crypt
ograph
y
and
di
gital
sign
at
ur
e.
Fig
ur
e
1
hi
gh
li
gh
ts
the
a
dopted
schem
e
of
pr
opose
d
syst
e
m
.
Figure
1
.
A
dopt
ed
sc
hem
a o
f pro
po
se
d
syst
e
m
The
a
bove
s
hown
sc
hem
e
is
m
ai
nly
intende
d
f
or
higher
de
gr
ee
of
pr
i
vacy
,
co
nf
i
den
ti
al
it
y,
as
well
as
data
integrity
by
hybr
i
dizin
g
app
r
oac
h.
T
he
pr
op
os
e
d
syst
e
m
con
siders
the
po
te
ntial
s
of
ge
ne
rati
ng
higher
degree
of
pr
i
va
te
keys
by
e
ll
ipti
cal
cur
ve
crypto
gr
a
phy
howe
ver
t
hey
are
hi
gh
e
r
in
nu
m
ber
that
cou
l
d
introd
uce
signi
ficant
a
m
ou
nt
of
com
pu
ta
ti
on
al
com
plexi
ty
in
low
pow
ered
se
ns
ors.
Hen
ce
,
the
pr
opos
e
d
syst
e
m
con
s
ide
rs
the
ref
e
re
nc
e
po
int
de
rive
d
fr
om
the
or
de
r
of
el
li
ptica
l
cur
ve
in
order
to
ens
ur
e
that
only
the
best
va
lue
of
pri
vate
c
ou
l
d
be
co
ns
ide
red
in
each
passes
.
T
he
nex
t
c
ontrib
ution
of
pro
posed
syst
em
is
t
hat
it
do
e
sn’t
us
e
c
onve
ntio
nal
digi
ta
l
sign
at
ure
a
s
it
is
e
xp
e
nsi
ve
in
te
rm
s
of
la
r
ge
scal
e
de
plo
ym
ent
ov
e
r
th
e
sens
or
s
.
Hen
ce
,
the
pro
po
s
ed
syst
em
hybri
dizes
bot
h
of
them
and
gen
e
rates
tw
o
al
gorithm
s
ie
.
1
st
al
gorithm
us
es
ra
ndom
a
ppr
oach
es
in
orde
r
to
generat
e
a
secur
it
y
to
ken
wh
ic
h
will
be
us
ed
f
or
ci
ph
e
rin
g
the
m
e
ssage
by
the
tra
ns
m
itti
ng
no
de
in
order
t
o
f
orward
it
to
the
receiv
er.
On
t
he
ot
he
r
ha
nd,
the
rec
ei
ver
node
will
us
e
public
key
crypto
gr
a
phy
as
w
el
l
as
second
al
go
rithm
in
or
de
r
to
pe
rform
validat
ion
of
the
receive
d
secur
it
y
tok
e
n.
A
su
c
ce
ssfu
l
id
entifi
ca
ti
on
of
sec
ur
it
y
tok
en
al
lo
ws
auth
or
iz
at
io
n
on
t
he
recei
ve
d
m
essage.
A
ny
fo
r
m
of
m
an
-
in
-
m
id
dle
at
ta
cks
wil
l
no
t
be
a
ble
t
o
dec
rypt
the
c
on
te
nt
of
the
m
essage
eve
nt
ually
hav
i
ng
possessi
n
of
sam
e
public
key.
He
nce,
th
e
novel
co
ntri
buti
on o
f
pro
pos
ed
syst
em
is
that
it
offer
s
bett
er
sec
ur
it
y
c
overag
e
A
l
g
o
r
i
t
h
m
f
o
r
G
e
n
e
r
a
t
i
n
g
S
e
c
u
r
i
t
y
T
o
k
e
n
A
l
g
o
r
i
t
h
m
f
o
r
V
a
l
i
d
a
t
i
n
g
t
h
e
S
e
c
u
r
i
t
y
T
o
k
e
n
E
l
l
i
p
t
i
c
a
l
C
u
r
v
e
C
r
y
p
t
o
g
r
a
p
h
y
D
i
g
i
t
a
l
S
i
g
n
a
t
u
r
e
H
y
b
r
i
d
i
z
e
T
r
a
n
s
m
i
t
t
e
r
R
e
c
e
i
v
e
r
U
s
e
A
l
g
o
-
1
U
s
e
A
l
g
o
-
2
M
a
l
i
c
i
o
u
s
C
o
d
e
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
FS
DA: Fr
am
e
work fo
r
Secu
r
e Data
Ag
gr
e
gati
on in W
ire
le
ss S
e
nsor Net
w
or
k f
or
E
nhanc
ing
... (
Jyoti
M
et
an
)
4687
from
m
axi
m
u
m
threats
in
WSN
us
in
g
a
li
gh
twei
gh
t
a
nd
hy
br
id
enc
ryptio
n
te
ch
ni
qu
e
.
T
he
nex
t
sect
io
n
discus
s
es a
bout
the alg
or
it
hm
i
m
ple
m
entat
io
n.
2.
ALGO
RITH
M
I
MPLEME
NTATIO
N
The
pro
posed
al
gorithm
pr
es
ents
a
no
vel
de
sign
of
dig
it
al
sign
at
ur
e
that
i
s
co
ns
tr
ucted
by
en
han
ci
ng
the str
uctu
re of
ell
ipti
cal
cu
rve crypt
ography
. Th
e
constr
uct
ion
of this
nove
l al
gorithm
resu
lt
s in ge
ner
at
ion
of
a
secu
rity
toke
n
that
will
be
furthe
r
s
ubje
ct
ed
to
valid
at
ion
process.
This
sect
io
n
will
discuss
a
bout
the
m
echan
ism
ado
pte
d
in
order
to
e
nh
a
nce
th
e
operati
ons
unde
rtake
n
by
e
ll
ipti
cal
cur
ve
crypto
gr
a
phy
with
a
pr
im
e
intenti
on
of
le
ve
ra
ging
data
integrity
,
pr
ivacy
,
as
we
ll
as
con
fide
ntial
it
y.
Fo
ll
ow
in
g
are
the
desc
r
ipti
on
of im
ple
m
ented
al
go
rithm
.
2.1.
Al
go
ri
thm
for Ge
nera
t
ing S
ec
urit
y
T
ok
en
In
orde
r
t
o
m
ai
ntain
a
bette
r
f
orm
of
co
nfi
de
ntial
it
y
of
the
data
as
w
el
l
as
node
’s
i
den
ti
ty
it
is
essenti
al
that
pro
po
se
d
syst
e
m
sh
ould
devel
op
s
uc
h
a
m
echan
ism
that
cou
l
d
dynam
icall
y
per
f
or
m
secure
gen
e
rati
on
of
dig
it
al
sig
natu
re.
He
nce,
t
he
pri
m
e
resp
on
sibil
it
y
of
the
pro
po
se
d
al
gor
it
h
m
is
to
ge
ne
rate
a
highly
dynam
i
c
and
sec
ur
e
t
ok
e
n
that
co
nsi
ste
ntly
al
te
rs
i
n
eve
ry
com
m
un
ic
at
io
n
proc
ess
as
well
as
is
al
so
li
gh
twei
ght
in
natu
re.
T
he
al
gorithm
ta
kes
th
e
input
of
O
u
(
uppe
r
lim
it
of
order),
a
(arbit
rar
y
v
al
ue
of
i
ntege
r
ty
pe)
that
after
processi
ng
r
esults
in
ge
nerat
ion
of
s
tok
(secu
rity
tok
e
n).
The
ste
ps
of
the
al
gorithm
are
as
fo
ll
ows:
Algori
th
m
for
Gener
at
in
g S
ecuri
ty Toke
n
Inpu
t
: O
u
(
uppe
r
li
m
it
o
f orde
r)
,
a
(arbit
rar
y
value o
f
inte
ge
r
ty
pe)
Out
p
ut
: s
t
ok
(s
ecur
it
y t
oken
)
St
ar
t
1.
i
nit O
u
,
2.
C
hoos
e
a
1
3.
C
om
pu
te
θ=
p
1
| O
u
|
4.
If
θ=
0
5.
Go to St
ep
-
2
6.
El
se
7.
C
om
pu
te
σ
1
σ(
b,
θ)
8.
Esti
m
a
te
α=β+
a
1
|O
u
|
9.
If
α
=
0
10.
Go to St
ep
-
3
11.
El
se
12.
Ob
ta
in
s
to
k
(θ
, α)
End
The
al
gorithm
sta
rts
by
i
niti
ating
uppe
r
li
m
i
t
of
or
der
O
u
ca
ptured
f
ro
m
the
el
li
ptica
l
curve
(Line
-
1)
.
The
e
xec
ution
of
t
he
al
gorith
m
beg
ins
by
tr
ansm
itti
ng
sen
so
r
no
de
init
ia
ti
ng
a
com
m
un
ic
at
ion
with
rec
ei
vin
g
sens
or
node.
I
n
this
proc
ess,
the
first
ste
p
is
to
perform
an
ar
bitrary
s
el
ect
ion
of
a
1
whose
value
r
ang
e
s
betwee
n
1
a
nd
(O
u
-
1)
(Line
-
2).
This
is
the
fi
rst
novelty
w
hich
reduces
co
m
pu
ta
ti
on
al
co
m
plexiti
es
associat
e
d
with
el
li
ptica
l
curve
c
ryptg
ra
ph
y
by
sel
ect
in
g
one
point
wit
hin
it
s
or
der
sc
op
e
a
nd
no
t
al
l
the
infi
nite
nu
m
ber
of
points
in
it
s
cu
rv
e
.
T
he
ne
xt
ste
p
of
im
ple
m
entat
ion
is
t
o
c
om
pu
te
θ
t
ha
t
will
be
re
quired
i
n
ge
ner
at
ion
of
secur
it
y
to
ke
n
at
the
e
nd
(
Line
-
3).Th
e
c
om
pu
ta
ti
on
of
θ
is
ca
rr
ie
d
ou
t
by
scal
a
r
pro
du
ct
of
posit
ion
al
inf
or
m
at
ion
p
1
and up
per lim
it of or
der in ell
ipti
cal
cu
r
ve O
u
(Line
-
3)
.
It
shou
l
d
be
know
n
that
(
p1,
q1)
re
presents
the
posit
ion
al
inf
or
m
at
ion
of
a
no
de
w
hos
e
e
m
pirical
value
is
c
on
si
de
red
t
o
be
e
quivale
nt
to
a
rb
it
r
ary
integer
val
ue
a1
a
nd
f
unc
ti
on
of
re
fer
e
nc
e
po
i
nt
f
(p
f
,
q
f
).
T
he
functi
on
of
re
fer
e
nce
point
is
con
si
der
e
d
to
li
e
within
the
el
li
ptica
l
c
urve
an
d
it
s
orde
r
is
co
ns
id
ered
is
m
axi
m
u
m
score
of
O
u
.
T
his m
echan
ism
con
trib
utes
to
no
vel
am
al
ga
m
ation
of n
ew
di
gital
sign
at
ure
as
well
as
el
li
piti
cal
cur
ve
crypt
ogra
ph
y.
The
ne
xt
pa
rt
of
im
ple
m
entat
ion
is
to
com
pu
te
an
e
ncr
y
ption
at
tribu
te
σ
app
li
ed
on
bea
con
(
or
c
on
t
ro
l
m
essage)
b
an
d
com
pu
te
d
va
riable
θ
(Li
ne
-
7)
.
It
can
be
al
so
note
d
that
unde
r
any
ci
rcu
m
stan
ces,
the
value
of
this
va
riabl
e
θ
is
con
side
r
ed
as
non
-
zer
o
nu
m
ber
(Line
-
4
a
nd
Li
ne
-
5).
This
process
is
f
ollow
e
d
by
ge
ne
r
at
ion
of
prel
i
m
inary
secu
rity
t
ok
e
n
α
by
ad
di
ng
up
a
ne
w
var
ia
ble
β
and
scal
ar
pro
du
ct
of ar
bi
trary inte
ge
r va
lue a
1
with
uppe
r
li
m
it
o
f orde
r
O
u
in eli
ptica
l
curve
crypt
ogr
aph
y.
We
perform
th
e
evaluati
on
of
ne
w
va
riable
β
as
pr
oduct
of
i)
va
riable
θ
obta
ined
f
ro
m
Line
-
3,
ii
)
a
n
arb
it
ra
ry
integ
er
[
1
(O
u
-
1)]
that
is
al
ways
c
on
si
der
e
d
t
o
be
it
s
pr
i
vate
ke
y,
an
d
ii
i)
σ
1
obta
ined
f
ro
m
Line
-
8.
We
al
so
e
nsure
that
the
em
piri
cal
value
of
th
e
pr
el
im
ina
ry
s
ecur
it
y
tok
e
n
α
is
al
ways
non
-
zero
a
nd
finall
y
the
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4684
-
4692
4688
al
gorithm
le
ad
s
to
sel
ect
ion
of
final
set
of
secur
it
y
tok
e
n
s
tok
acqu
ire
d
f
r
om
var
ia
ble
θ
ob
ta
ine
d
from
Line
-
3
and
va
riable
α
ob
ta
ine
d
from
Line
-
8.
A
cl
ose
r
lo
ok
i
nto
t
he
ab
ov
e
al
gorithm
ic
ste
ps
will
s
how
t
hat
it
hy
br
id
s
the
el
li
ptica
l
c
urve
cry
ptogra
ph
y
with
ty
pi
cal
sign
at
ure
i
n
or
der
t
o
ge
ner
at
e
a
li
ght
weig
ht
an
d
dynam
ic
secur
it
y
to
ken
that
is
require
d
to
m
ai
ntiai
n
higher
de
gr
ee
of
pr
i
vacy
as
well
as
co
nf
i
de
ntial
it
y.
At
the
sam
e
tim
e, the
al
gorithm
a
lso co
nt
ribu
te
s
to
m
ini
m
iz
at
ion
of t
he
c
om
pu
ta
ti
on
al
over
hea
d
as
wel
l.
2.2
.
Algori
t
h
m for
Va
li
d
at
i
ng
th
e
Securi
t
y
T
ok
en
The
exec
utio
n
of
this
al
gorith
m
cou
ld
only
beg
in
after
s
ucc
essfu
l
ge
ne
rati
on
o
f
sec
ur
it
y
t
ok
e
n
by
th
e
transm
itti
ng
sen
sor no
de.
Th
i
s g
ene
rati
on
se
cur
it
y t
ok
e
n
is t
hen
forwa
rd
e
d t
o
receivin
g
sensor
no
de
w
he
re th
e
the
la
tt
er
perf
or
m
s
validat
io
n.
T
he
in
put
to
this
al
go
rith
m
is
s
tok
(secu
re
to
ken)
a
nd
k
pub
(pu
blic
ke
y)
that
resu
lt
s
in
outc
om
e
of
V+
/
V
-
(
Su
cces
sf
ul/fai
le
d
validat
ion).
T
he
i
m
portant
ste
ps
of
the
al
gorithm
are
as
fo
ll
ows:
Algori
th
m
for
Vali
dating
t
h
e Securit
y
T
oken
Inpu
t
: s
tok
(sec
ur
e
to
ken), k
p
ub
(
public key)
Out
p
ut
:
V+ /
V
-
(Success
f
ul / f
ai
le
d vali
dat
ion)
St
ar
t
1.
If
k
pub
≠0
2.
If
k
pub
EC
3.
s
uccess
ful
1
st
sta
ge
of
va
li
dation
4.
En
d
5.
En
d
6.
If
(θ, α
)
Z
-
1
Z is intege
r
7.
C
om
pu
te
σ
1
σ(
b,
θ)
8.
C
om
pu
te
P
αf
-
β|O
u
|
9.
If
θ=
p
1
|O
u
|
10.
V
+
f
la
g
s
tok
as
valid
11.
el
se
12.
V
-
flag
s
tok
as
in
valid
13
. else I
f
14.
V
-
flag
s
tok
as
in
valid
15.
End
End
Be
fore
try
in
g
to
unde
rsta
nd
the
im
ple
m
entat
ion
sc
hem
e
of
t
he
ab
ove
validat
io
n
al
gorithm
,
it
is
essenti
al
to u
nd
erstan
d
one i
m
portant ass
um
ption
that a receiving
se
nsor node m
us
t hav
e a r
eplic
a o
r
acce
ss of
public
key
k
pub
of
tra
ns
m
itti
ng
sens
or
node
.
Otherw
ise
,
thi
s
validat
ion
ca
nnot
be
perfor
m
ed.
The
c
omplet
e
process
of
validat
ion
of
the
r
ecei
ved
sec
uri
ty
tok
en
by
the
receivin
g
se
nsor
node
is
ca
rri
ed
out
in
tw
o
sta
ges
viz.
pri
m
ary
st
age
an
d
sec
onda
ry
sta
ge.
I
n
th
e
pr
im
ary
sta
ge,
the
al
gorith
m
check
s
if
the
r
e
is
pr
e
sence o
f
non
-
zero
public
ke
y
(Line
-
1).
In
case
of
non
-
a
va
il
abili
ty
of
no
n
-
ze
ro
pu
blic
key,
the
com
m
un
ic
at
io
n
is
a
borte
d
instantl
y
sta
ti
n
g
that
it
s
extern
al
at
ta
ck
scenario
.
H
oweve
r,
if
it
is
valid
than
it
check
s
if
the
nu
m
erical
valu
e
of this
publi
c ke
y k
pub
act
ually
r
esi
de
s
within
the r
a
nges
of ell
ipti
cal
cu
rv
e
(Li
ne
-
2).
This
com
plete
s
the
pr
im
ary
validat
io
n
sta
ge
.
The
ne
xt
ste
p
of
the
al
go
rithm
ta
rg
et
s
to
pe
rfor
m
seco
nd
a
ry
validat
ion
of
obta
ined
sec
uri
ty
tok
en
s
tok
.
F
or
t
hi
s
pu
r
pose,
it
ensures
t
hat
bot
h
the
va
riables
θ
and
α
sh
ould
be
of
integer
ty
pe
a
s
well
as
their
scop
e
has
to
be
m
and
at
or
il
y
reside
withi
n
lowe
r
lim
it
of
1
an
d
higher
li
m
i
t
of
(O
u
-
1)
(Li
ne
-
6).
I
n
case
of
exp
l
or
at
io
n
of
non
-
intege
r
va
lue
ty
pe,
the
a
lgorit
hm
con
siders
i
t
equ
i
valen
t
t
o
e
aves
droppin
g
or
m
essage
ta
m
eper
in
g
a
nd
there
by
it
flag
s the
obta
ine
d
se
cur
it
y
to
ken
as
inv
al
i
d
(Line
-
14)
. Up
on con
firm
ing
that they
(
θ a
nd α) ar
eo
f
inte
ge
r
ty
pe
tha
n
the
al
gorithm
p
erfor
m
s co
m
pu
ta
t
ion
of
encr
y
ption
at
tr
ibu
te
σ
1
by
ap
plyi
ng
a
ny
f
orm
o
f
crypt
ogr
aph
ic
functi
on
on
t
he
co
ntr
ol
m
essage
b
a
nd
θ
.
It
sh
oul
d
be
note
d
that
the
i
m
ple
m
ented
functi
on
σ
(Lin
e
-
7)
is
si
m
il
ar
to
t
hat
us
e
d
in
previo
us
al
gorith
m
of
secur
it
y t
oke
n gen
e
rati
on.
The
ne
xt
valid
at
ion
ste
p
of
the
al
gorithm
c
al
ls
fo
r
com
puti
ng
the
a
sing
l
e
com
m
un
ic
at
i
on
vecto
r
of
po
sit
io
nal
inf
orm
ation
i.e.
P
,
wh
e
reP
=
(p
1
,
p
2
).
It
sho
uld
be
note
d
t
hat
po
sit
io
n
in
for
m
at
ion
of
tra
nsm
it
t
ing
and
receivi
ng
nodes
a
re
(
p
1
q
1
)
an
d
(
p
2
q
2
)
r
especti
vely
.
T
hi
s
com
pu
ta
ti
on
of
si
ng
le
c
omm
un
ic
at
ion
ve
ct
or
of
po
sit
io
nal
inf
orm
ation
P
is
em
pirically
fo
rm
ed
to
be
corr
esp
onding
to
α
f
-
β|O
u
|
(Line
-
8).
A
cl
os
e
r
lo
ok
int
o
this
em
pirical
form
ulati
on
wi
l
show
that
fir
st
com
po
ne
nt
i
s
a
scal
ar
pr
oduct
of
prel
i
m
inary
secu
rity
to
ken
α
and
f
un
ct
io
n
of
re
fer
e
nce
po
i
nt
f
(p
f
q
f
)
w
hile
the
sec
ond
c
om
po
ne
nt
co
rresp
onds
t
o
β
a
nd
uppe
r
li
m
i
t
of
ord
e
r
i.e.O
u
. T
he
em
pirical
v
al
ue o
f
β is co
ns
i
der
e
d
sam
e as p
rod
uct of v
a
riable
θ and an
a
rb
it
r
ary inte
ger
[
1
(
O
u
-
1)
]
that
is
al
ways
consi
der
e
d
to
be
it
s
pr
i
vate
ke
y.
The
fi
nal
ste
p
of
vali
datio
n
of
sec
ur
it
y
to
ken
is
ca
rr
ie
d
ou
t
by
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
FS
DA: Fr
am
e
work fo
r
Secu
r
e Data
Ag
gr
e
gati
on in W
ire
le
ss S
e
nsor Net
w
or
k f
or
E
nhanc
ing
... (
Jyoti
M
et
an
)
4689
check
i
ng
of
va
lue
of
t
he
va
riable
θ
is
eq
uiv
al
ent
to
p
1
|O
u
|
(Line
-
9).
I
f
the
le
ft
ha
nd
side
of
ex
pr
ession
exh
i
bited
in
L
ine
-
9
is
no
t
f
ound
eq
ui
valent
to
righ
t
hand
siz
e
than
the
al
go
rithm
con
firm
s
that
obta
ined
se
cur
it
y t
oke
n i
s h
ig
hly i
nval
id.
An
i
nteresti
ng
fact
about
this
al
gorithm
con
st
ru
ct
io
n
is
that
their
false
sta
teme
nt
pr
eci
sel
y
corres
ponds
to
the
at
ta
ck
sce
nar
i
o
w
hich
c
ou
l
d
be
ge
ner
at
ed
f
ro
m
any
node
.
He
nce,
th
e
al
gorithm
do
esn’t
al
low
the
r
ou
ti
ng
to
be
c
o
nfi
r
m
and
aborts
t
he
c
onnecti
on
on
ce
the
first
s
ta
ge
of
validat
ion
it
sel
f
fail
s.
Hen
ce
,
in
a
s
m
art
m
ann
er
,
the
al
gorithm
of
fer
s
secu
rity
to
i
ts
neigh
bo
rin
g
nodes
al
so
.
More
over
,
ow
i
ng
to
util
iz
at
ion
of
no
n
-
recursi
ve
a
ppr
oach,
t
he
al
gorithm
of
fe
rs
si
gn
ific
a
nt
ad
van
ta
ge
in
te
rm
s
of
com
m
un
ic
at
ion
ef
f
ic
ie
ncy
with
reduce
d
c
om
pu
ta
ti
on
al
bur
de
n
a
par
t
fro
m
it
s secu
rity
capab
li
ty
.
3.
RESU
LT
A
N
ALYSIS
This
sect
io
n
ou
tl
ines
t
he
ou
tc
om
es
ob
ta
ined
a
fter
im
ple
m
enting
the
pro
posed
FSDA
us
i
ng
MATLAB
.
For
this
pur
po
se
,
we
perform
si
m
ulati
on
stud
y
with
1000
sensors
bea
ri
ng
c
onfig
ur
at
i
on
s
of
MEM
SI
C
no
de
s.
T
he
sim
ulatio
n
a
rea
is
c
on
sidere
d
to
be
1100
x1300m
2
with
10
m
et
ers
of
tra
ns
m
issi
on
range
.
As
the
pro
pos
ed
stu
dy
intr
oduce
s
a
hy
br
i
d
ap
proac
h
with
el
li
ptica
l
cur
ve
c
rypto
grap
hy
as
well
as
dig
it
al
sign
at
ur
e
hen
c
e
it
is
antic
ip
at
ed
to
offe
r
li
gh
twei
ght
en
crypti
on
sche
m
e
fo
r
cl
ai
m
i
ng
a
n
ef
fecti
ve
ke
y
m
anag
em
ent
s
chem
e.
This
lig
ht
weig
ht
featur
e
ca
n
be
onl
y
pr
ove
n
if
the
al
go
rithm
of
fe
rs
le
ss
com
pu
ta
ti
on
al
bur
den
a
nd
e
quivale
ntly
m
ain
ta
ins
op
ti
m
al
com
m
un
ic
at
ion
perform
ance.
The
re
fore,
w
e
ch
oo
se
to
c
onside
r
al
gorithm
pr
oc
essing
ti
m
e,
end
-
to
-
en
d
de
la
y,
ener
gy
consum
ption
,
and
pac
ket
de
li
ver
y
rati
o
as
the
perform
ance
par
am
et
er.
The
stud
y
al
so
pe
r
f
or
m
s
co
m
par
at
ive
analy
sis
with
the
m
os
t
fr
equ
e
ntly
i
m
ple
m
ented
encr
y
ption sch
e
m
es o
f key
m
anag
em
ent.
The
ou
tc
om
es
cl
early
ind
ic
at
es
that
pro
pose
d
syst
em
offer
s
sig
nific
antly
bette
r
outc
om
es
in
com
par
ison
t
o
existi
ng
A
ES
or
S
HA.
Fro
m
the
Table
1,
it
can
be
seen
that
pro
po
s
ed
syst
em
of
fe
rs
appr
ox
im
at
ely
64
.
67%,
63.
12%,
4.9
4%,
and
60.02%
of
im
pr
ov
em
ent
with
res
pe
ct
to
ov
erall
energy
consum
ption
,
ov
e
rall
delay
,
pack
et
delive
r
y
rati
o,
a
nd
al
gorithm
pr
oce
ssing
tim
e.
O
wing
to
no
n
-
r
ecur
si
ve
base
d
operati
on,
FS
D
A
ex
hi
bits
l
ow
e
r
al
gorithm
pr
oces
s
ing
ti
m
e
Figu
r
e
2
an
d
it
of
fe
rs
en
hanced
se
cur
it
y
with
faster
res
pons
e
ti
m
e
with
inc
reasin
g
it
erati
on
s
.
T
his
al
so
offer
s
co
m
pl
i
m
entary
ben
efit
s
to
dela
y
factor,
wh
ic
h
is
fou
nd to be e
xtrem
ely less Fig
ur
e
3
.
Table
1
.
Su
m
m
ary o
f
Pe
rce
ntage of
I
m
pr
ove
m
ent
Techn
iq
u
e
Ov
erall
en
e
rgy
Co
n
su
m
p
tio
n
(
%)
Ov
erall
D
elay
(
%
)
Pack
et Delive
ry
Ratio
(
%)
Alg
o
rith
m
Pr
o
cess
in
g
Ti
m
e
(
%
)
AES
3
9
.28
5
1
.54
2
9
.73
4
5
.42
SHA
-
2
5
2
.08
4
4
.81
1
4
.77
4
7
.39
FSDA
2
6
.69
3
3
.23
4
9
.44
3
2
.79
Figure
2
.
Com
par
at
ive
ev
al
ua
ti
on
of alg
or
it
hm
p
ro
cessi
ng ti
m
e
The
pro
posed
syst
e
m
al
so
m
ake
us
e
of
fi
rs
t
order
ra
dio
e
nergy
m
od
el
that
esse
ntial
ly
com
pu
te
s
energy
dissipa
ti
on
in
or
der
t
o
fin
d
that
FS
DA
c
onsu
m
es
le
ss
energy
and
he
nce
offe
r
s
netw
ork
lo
ngevity
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4684
-
4692
4690
Figur
e
4.
Final
ly
,
the
num
ber
of
enc
ryptio
n
s
te
ps
are
not
m
assive
for
wh
ic
h
reas
on
m
or
e
num
ber
of
resou
rces
are
a
vaila
ble
for
lo
nger
du
rati
on
res
ulti
ng
in
an
ef
fec
ti
ve
res
ource
al
locat
ion
.
Thi
s
cause
s
sig
ni
ficant
i
m
pr
ovem
ent
in
ex
plorin
g
bette
r
c
omm
u
nicat
ion
c
h
a
nnel
with
utm
os
t
secur
it
y
Fi
gure
5.
The
tren
d
of
increasin
g
patt
ern
of
pac
ket
delivery
rati
o
ov
e
r
i
ncr
easi
ng
nu
m
ber
of
ne
ighborin
g
no
des
not
only
s
how
it
s
bette
r
scal
abili
ty
per
form
ance
bu
t
al
so
exh
i
bits
that
FSDA
offe
rs
non
-
re
pudiati
on
al
ong
wit
h
data
in
te
gr
it
y,p
rivac
y
and
c
onfide
nt
ia
li
t
y.
Hen
ce
,
app
li
cabil
it
y
of
F
DSA
is
m
or
e
f
or
a
ny
se
nsory
a
ppli
cat
ion
that
dem
and
s
lo
nger
te
rm
of
secur
it
y
su
r
veill
ance
ove
r
unce
rtai
n
com
m
un
ic
at
ion
as
it
offer
s
e
qual
resi
sti
vity
perform
ance to
m
axi
m
u
m
at
tack
s.
Figure
3
.
Com
par
at
ive
ev
al
ua
ti
on
of d
el
ay
Figure
4
.
Com
par
at
ive
ev
al
ua
ti
on
of ene
rg
y
consum
ption
Figure
5
.
Com
par
at
ive
ev
al
ua
ti
on
of p
ac
ket
delivery
rati
o
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
FS
DA: Fr
am
e
work fo
r
Secu
r
e Data
Ag
gr
e
gati
on in W
ire
le
ss S
e
nsor Net
w
or
k f
or
E
nhanc
ing
... (
Jyoti
M
et
an
)
4691
4.
CONCL
US
I
O
N
Secu
rity
is
one
of
the
m
os
t
chall
en
ging
pro
blem
s
in
WSN
ir
resp
ec
ti
ve
of
m
assive
am
ou
nt
of
researc
h
work
beng
ca
rr
ie
d
ou
t
ti
ll
date.
W
e
obse
rv
e
d
that
existi
ng
appr
oach
es
of
key
m
anag
e
m
ent
e
m
ph
asi
zes
on
sp
eci
fic
f
or
m
of
at
ta
cks
wh
i
ch
na
rro
ws
do
wn
th
e
ap
plica
bili
ty
of
key
m
anag
em
ent
when
th
e
at
ta
ck
scenari
o
is
changed.
A
t
the
sa
m
e
time,
we
fi
nd
t
hat
there
are
m
uch
pote
ntial
of
us
in
g
el
li
ptica
l
curv
e
cryptosyst
em
i
n
order
to
ge
ne
rate
pri
vate
ke
ys
but
the
pr
ocess
is
to
o
m
uch
rec
ursive
and
le
ads
to
in
creased
m
essage
siz
e.
At
the
sam
e
tim
e,
us
age
of
di
gital
sign
at
ure
is
no
t
t
oo
c
os
t
eff
ect
ive
owin
g
to
it
s
dep
e
nd
encies
on
c
erti
ficat
es.
Hen
c
e,
we
hy
br
idize
bo
t
h
el
li
ptica
l
cur
ve
crypto
syst
em
as
well
as
sign
at
ur
e
i
n
orde
r
t
o
const
ru
ct
a
no
vel
al
gorithm
.
The
stu
dy
outc
om
e
sh
ows
that
pro
pose
d
al
gorit
hm
of
fe
rs
sig
nifica
nt
data
integrity
,
co
nfi
den
ti
al
it
y,
an
d
pr
i
vacy
in
it
s
process
a
nd
is
fou
nd
t
o
off
er
su
it
a
ble
bal
ance
betwee
n
su
c
h
secur
it
y dem
an
ds
a
nd c
omm
u
nicat
ion
pe
rform
ance.
R
E
F
E
R
E
N
C
E
S
[1]
W
.
Os
te
rha
ge
,
W
ire
le
ss
Secur
ity
,
CRC Press,
2
016
[2]
T.
Ham
za,
G.
K
addoum,
A.
Me
ddeb
and
G.
Ma
ta
r
,
"A
Surve
y
on
Inte
l
li
g
ent
M
AC
Lay
er
Jam
m
ing
Atta
cks
an
d
Counte
rm
ea
sure
s
in
WSNs
,
"
201
6
IEE
E
84th
Ve
hic
ul
ar
Te
chno
l
og
y
Confer
enc
e
(VTC
-
Fall
),
Mo
ntre
a
l,
QC,
pp.
1
-
5,
2016
,
.
[3]
A.
Modirkhazen
i,
N.
Ithni
n
and
O.
Ibra
him,
"S
e
cur
e
Mult
ipath
Routi
ng
Protoco
ls
in
W
i
reless
Sensor
Networks:
A
Secur
ity
Surve
y
Anal
y
sis,"
201
0
Second
Inte
r
nat
ion
al
Confer
enc
e
on
Netwo
rk
Applic
ations,
Protocol
s
and
Servic
es,
Keda
h
,
pp.
228
-
233
,
20
10
[4]
Shashikal
a
and
C.
Kavit
ha
.
,
"A
surve
y
on
sec
ure
d
routi
ng
prot
ocol
s
for
wire
les
s
sensor
net
wor
k,
"
Com
puti
ng
Com
m
unic
at
ion
&
Networki
ng
T
ec
hnolog
ie
s
(IC
CCNT),
2012
T
hird
Inte
rn
at
ion
a
l
Confer
en
ce
on
,
Coim
bat
ore
,
pp
.
1
-
8,
2012
[5]
F.
Khedim,
N.
La
bra
ou
i
and
M
.
L
ehsa
ini,
"D
ishonest
re
comm
enda
ti
on
at
t
ac
ks
in
wire
l
ess
sensor
net
works
:
A
surve
y
,
" 2015
1
2th
In
t
ern
ationa
l
S
y
m
posium
on
Program
m
ing
and
S
y
stems
(ISP
S),
Algi
ers,
p
p.
1
-
10,
2015
[6]
H.
Yu,
Z
.
Shen,
C.
Miao
,
C.
Leung
and
D.
Ni
yat
o,
"A
Surve
y
of
Trust
and
R
e
puta
ti
on
Mana
g
ement
S
y
st
ems
in
W
ire
le
ss
Com
muni
cations,
"
in
P
roc
ee
d
ings o
f th
e
IE
EE,
vol
.
98
,
no.
10
,
pp
.
1755
-
1772,
Oct
.
2010
.
[7]
R.
Dou
and
G.
Nan,
"O
pti
m
iz
in
g
Sensor
Network
Covera
ge
and
Regi
onal
Conn
e
ct
ivit
y
in
Industr
ia
l
IoT
S
y
stems
,
"
in
IE
EE S
y
stems
Journal, vol.
11
,
no.
3,
pp.
1351
-
1360,
Sept
.
2017
.
[8]
J.
M.
W
il
li
ams
et
al
.
,
"W
ea
ving
t
he
W
ire
le
ss
W
eb:
To
ward
a
Low
-
Pow
er,
Dense
W
ire
le
ss
Sensor
Network
for
the
Industria
l
IoT
,
" in IE
E
E
Mi
cro
wa
ve
Maga
zi
n
e, vo
l.
18
,
no
.
7
,
pp
.
4
0
-
63,
Nov.
-
De
c.
2017.
[9]
K.
Yang,
D.
Bl
aa
uw
and
D.
Sy
lv
este
r
,
"H
ard
ware
Designs
for
Secur
ity
in
Ultra
-
Low
-
Pow
e
r
IoT
S
y
stems
:
An
Overvi
ew
and
S
urve
y
,
"
in
I
EE
E
Micro,
vo
l. 37, n
o.
6
,
pp
.
72
-
89
,
Novem
ber
/Dec
e
m
ber
2017.
[10]
L.
Chen
e
t
al.,
"
Robustness,
Secur
ity
and
Priva
c
y
in
Lo
cation
-
B
ase
d
Services
for
Future
IoT:
A
Surve
y
,
"
in
IEEE
Acc
ess,
vo
l. 5, p
p.
8956
-
8977
,
2
017.
[11]
Manjuna
th
,
B.
E.
,
and
P.
V.
R
ao.
"Ba
la
n
ci
ng
Tra
de
off
b
et
we
en
Dat
a
Secu
rity
and
Ene
rg
y
Model
for
W
ireless
Sensor Net
work.
" Int
ern
at
ion
al
J
ourna
l
of
E
le
c
trica
l
and
Com
puter E
ngin
ee
ring
(IJ
ECE
)
8
.
2
(2018)
:
1048
-
1055.
[12]
Singh,
Pooja,
a
nd
R.
K.
Chau
han.
"A
Surve
y
on
Com
par
isons
of
Cr
y
ptogr
aphi
c
Algor
it
h
m
s
U
sing
Cert
a
in
Para
m
et
ers
in
W
SN
.
" Int
ern
a
ti
on
al
Journa
l
of
Ele
ct
ri
ca
l
and
Com
pute
r Engineerin
g
7.
4
(2017):
22
32.
[13]
Sukavasi,
Hem
a
Gopinat
h
,
Lokes
h
Kana
gala,
an
d
Riaz
Shaik
.
"S
uffic
i
ent
Auth
en
ti
c
at
ion
for
Ene
r
g
y
Consum
pti
on
in
W
ire
l
ess Sensor
Networks."
In
te
rna
ti
ona
l
Journ
al
of
E
le
c
tri
c
al and Com
pute
r
En
gine
er
ing
6
.
2
(2
016):
735.
[14]
J.
Meta
n,
K
N
Nara
sim
ha
Murth
y
,
“
Group
Key
Mana
gement
T
ec
hniqu
e
base
d
on
Logi
c
-
Ke
y
Tre
e
in
th
e
Fiel
d
of
W
ire
le
ss
Sensor
Network”
,
Inte
r
nati
onal Journal
of
Comput
er
A
p
pli
cations
,
Vol
.
1
17,
No.12
,
Ma
y
2015
[15]
G.
W
ang,
D.
Ki
m
and
G.
Cho,
“
A
sec
ure
cl
ust
er
form
at
ion
sc
heme
in
wire
l
ess
sensor
net
works
,
”
Int
ern
a
ti
ona
l
Journal
of
D
istributed
Sensor N
e
tworks,
pp.
14,
2
012
[16]
P.
Pawani,
C.
Schm
it
t,
P.
Kum
ar,
A.
Gurtov,
an
d
M.
Ylia
nt
tila
,
"P
AuthKey
:
A
per
vasive
aut
he
nti
c
at
ion
proto
c
ol
and
ke
y
esta
bl
ishm
ent
sche
m
e
f
or
wire
le
ss
sens
or
net
works
in
di
stribut
ed
Io
T
ap
pli
c
at
ions,"
Inte
r
nat
ion
al
Journal
of
Distribut
ed
Sensor Net
works
,
vo
l
.
10
,
no
.
7
,
pp
.
35
7
-
430,
2014
[17]
S.
Kang,
C.
J
i,
and
M.
H
ong,
"S
ec
ur
e
c
o
ll
abor
at
iv
e
k
e
y
m
ana
gement
for
d
y
namic
g
roups
in
m
obile
net
works
,
"
Journ
al
of
Appli
ed
M
at
hemat
ic
s,
pp.
1
0,
2014
[18]
J.D.
Le
e
,
H.
J.
I
m
,
W
.
M.
Kang,
and
J.
H.
Park,
“
Ubi
-
RKE:
a
rhy
thm
ke
y
base
d
enc
r
y
pt
ion
sche
m
e
for
ubiqui
to
us
devi
c
es,
"
Math
e
m
at
ic
a
l
Problem
s in
Eng
ineeri
ng,
pp.
8,
2014
[19]
C
-
M.
Chen,
X.
Zhe
ng
and
T
-
Y.
W
u,
“
A
complete
hi
era
r
chi
c
al
ke
y
m
an
age
m
ent
sche
m
e
for
hetero
gene
ous
wire
le
s
s
sensor ne
tworks”,
The Sci
en
ti
f
ic World
Journal
,
pp.
13
,
2014
[20]
G.
C.
C.
F.
Pere
ira
,
R
ena
n
C
.
A.
Alves,
F.
L.
d
.
Silva
,
R
.
M.
Az
e
vedo,
B
.
C.
Alb
e
rti
ni
,
and
C
.
B.
Margi,
“
Resea
r
c
h
Artic
l
e
Perform
anc
e
Ev
al
ua
ti
on
o
f
Cr
y
ptogr
aphic
Algorit
hm
s ove
r
IoT
Plat
form
s a
nd
Opera
ti
ng
S
ystems
”,
Hindawi
Secur
ity
and
Co
m
m
unic
at
ion
Ne
tworks,
pp.
16,
2
017
[21]
A.
Ibra
him
and
G.
DalkJlJç,
“
Res
ea
rch
Art
ic
l
e
an
Advanc
ed
En
cr
y
p
ti
on
Stand
ar
d
Pow
ere
d
Mutual
Authentic
at
io
n
Protocol
Based
on
El
l
ipt
i
c
Curv
e
Cr
y
ptogra
ph
y
for
RF
ID,
Proven
on
W
ISP
”,
Hindawi
Journal
o
f
Sensors
,
pp.
1
0,
2017
[22]
P.
Sarka
r
and
S.
Mukherjee,
"
Secur
e
connect
e
d
sca
l
abl
e
com
bina
tori
al
KP
S
in
W
SN
:
Dete
r
m
ini
stic
m
erg
in
g,
loc
a
li
z
at
ion
,
"
38
th
Annua
l
IE
EE
Confe
renc
e
on
Local
Comput
er
Net
works
,
S
y
dne
y
,
NS
W
,
2013,
p
p.
622
-
629
.
[23]
J.
Qi,
X.
Hu,
Y
.
Ma
and
Y.
Su
n,
"A
Hy
brid
Secur
i
t
y
and
Com
pre
ss
ive
Sensi
ng
-
Based
Sensor
Data
Gathe
rin
g
Scheme,
" i
n
IE
E
E
A
ccess
,
vol
.
3
,
pp.
718
-
724,
20
15.
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.
8
, N
o.
6
,
Dece
m
ber
201
8
:
4684
-
4692
4692
[24]
J.
W
u,
K.
Ota
,
M.
Dong
and
C
.
Li,
"A
Hier
archic
a
l
Secur
ity
F
ramework
for
Defe
nding
Ag
ai
ns
t
Sophis
ti
c
at
ed
Atta
cks
on
W
ireless Sensor Ne
t
works
in
Sm
art
Cit
ie
s,"
in
IEEE Access
,
vol
.
4
,
p
p.
416
-
424
,
201
6.
[25]
Y.
Deng,
L.
W
a
ng,
M.
El
kashl
an,
A.
Nallana
t
han
and
R.
K.
Mall
ik
,
"P
h
y
sic
al
Lay
e
r
Secur
i
t
y
in
Thr
ee
-
Ti
e
r
W
ire
le
ss
Sensor
Networks:
A
S
toc
hastic
Geom
et
r
y
Appro
ac
h
,
"
in
IEEE
Tr
ans
act
ions
on
Information
Forensics
and
Sec
uri
ty
,
vol
.
11
,
no
.
6
,
pp
.
1
128
-
1138,
June
2016.
[26]
I.
A.
Um
ar,
Z.
M.
Hana
pi,
A.
Sali
and
Z
.
A.
Zul
kar
n
ai
n,
"Tr
uFiX:
A
Config
ura
ble
Trust
-
B
a
sed
Cross
-
Lay
e
r
Protocol
for
W
ir
el
ess Sensor Ne
t
works
,
" i
n
IE
EE
Ac
c
ess
,
vol
.
5
,
p
p.
2550
-
2562
,
2
017.
[27]
J.
Zhu,
Y.
Zou
and
B.
Zhe
ng
,
"P
h
y
sical
-
L
a
y
er
Secur
ity
and
Re
li
ability
C
h
allen
ges
for
Industrial
W
ire
le
ss
Sens
or
Networks,"
in
I
E
EE
Ac
c
ess
,
vo
l.
5,
pp
.
5313
-
532
0,
2017
.
[28]
D.
Qin,
S.
Yan
g,
S.
Ji
a,
Y.
Z
hang,
J
.
Ma
an
d
Q.
Ding
,
"Re
sea
rch
on
Trust
Sensing
Based
Secur
e
Routi
n
g
Mec
hani
sm
for W
ire
le
ss
Sensor
Network,
" i
n
IE
EE
Ac
c
ess
,
vo
l.
5,
pp
.
9599
-
960
9,
2017
.
[29]
D.
Shin,
V.
Sh
arma,
J.
Kim
,
S.
Kw
on
and
I.
You,
"S
ec
ure
and
Eff
i
cient
Pr
otoc
ol
for
Rout
e
Optimization
in
PM
IPv6
-
Based
Sm
art
Hom
e
IoT
Networks,"
in
I
EE
E
Acce
ss
,
vol
.
5
,
pp
.
11100
-
1
1117,
2017
.
[30]
Y.
Guan
and
X.
Ge,
"D
istri
bute
d
Secur
e
Esti
m
at
i
on
Over
W
ire
le
ss
Sensor
Networks Agai
nst Ra
ndom
Multi
cha
nne
l
Jam
m
ing
Atta
ck
s,"
in
I
EE
E
A
cce
ss
,
vol. 5, pp. 10
858
-
10870,
201
7.
[31]
H.
Dai,
M
.
W
a
ng,
X.
Yi
,
G.
Yang
and
J.
B
a
o,
"S
ec
ure
MA
X/MIN
Queri
es
in
Two
-
T
ie
red
W
ire
le
ss
Sensor
Networks,"
in
I
E
EE
Ac
c
e
ss
,
vo
l.
5,
pp
.
14478
-
14
489,
2017
.
[32]
F.
Al
-
Turj
m
an,
Y.
Kirsal
Eve
r
,
E.
Ev
er,
H.
X
.
Ngu
y
en
and
D.
B
.
David,
"S
ea
m
les
s
Key
Agr
ee
m
e
nt
Fram
ework
for
Mobile
-
Sink
in
IoT
Based
Clou
d
-
Cent
ri
c
Secur
ed
Public
Safe
t
y
Sensor
Networks,"
in
IEEE
A
cc
ess
,
vol
.
5,
pp.
24617
-
24631,
2
017.
[33]
V.
N.
Vo,
T
.
G.
Ngu
y
en
,
C.
So
-
In
and
D.
B.
Ha
,
"S
ec
re
c
y
Perfo
rm
anc
e
Anal
y
sis
of
Ene
rg
y
Harv
esti
ng
W
ire
l
ess
Sensor Net
works
W
it
h
a
Friendly
J
amm
er,
" i
n
I
E
EE
Ac
c
ess
,
vo
l.
5,
pp
.
25196
-
25
206,
2017
.
[34]
Yiqin
Lu
,
Jing
Zha
i
,
Ronghuan
Zhu,
and
Ji
ancheng
Qin,
“
Resea
rch
Article
Stud
y
of
W
ire
l
ess
Authent
i
catio
n
Cent
er
with
Mixed
En
cr
y
pti
on
in
W
SN
”,
Hindaw
i
Publishing
Cor
pora
ti
on
Journal
of
Sensors
,
pp.
7
,
2016
[35]
J.
Meta
n
and
K.
N.
N.
Murth
y
,
"Robus
t
and
sec
ure
ke
y
m
ana
ge
m
ent
in
W
SN
u
sing
arb
it
ra
r
y
k
e
y
-
dep
lo
y
m
ent
,
"
2015
Inte
rnatio
nal
Confe
ren
ce
on
Eme
rging
Re
search
in
El
e
ct
ronics,
Computer
Sci
en
ce
and
Technol
o
gy
(
ICER
ECT)
,
Mand
y
a
,
pp
.
246
-
25
0,
2015
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
J
y
oti
Met
an
has
rec
e
ive
d
B.
E.
fr
om
Pune
Univer
sit
y
,
Pune,
Indi
a
in
2002
and
M.T
ec
h
from
VTU,
Banga
lor
e,
Indi
a
in
2009.
She
joi
ned
Depa
rt
m
ent
of
Com
pu
te
r
Scie
n
ce
&
Engi
ne
eri
ng,
A
CS
Coll
ege
of
Enginee
ring
B
anga
lo
re
as
As
sistant
Profess
or
since
2012.
Her
rese
ar
ch
int
ere
st i
ncl
ud
es
Cr
y
ptogr
aph
y
,
W
ire
le
ss
Sensor Net
works
and
Secur
i
t
y
.
She
is
a Li
fe
Mem
ber
of the
Indi
an
Soci
e
t
y
for
Techni
ca
l
Ed
uca
t
ion
(ISTE
).
K
.
N
.
Nara
shinha
Murth
y
re
ceiv
ed
his
PhD
fro
m
Anaa
Univer
s
ity
,
Chenn
ai,
India
in
2013.
His
rese
arc
h
area
inc
lud
es
Im
age
proc
essing,
W
ire
le
ss
Sensor
Network,
Se
cur
ity
and
K
e
y
Mana
gm
ene
t.
At
pre
sent
he
is
working
as
a
Prof
essor
in
the
d
ep
art
m
ent
of
f
ac
ul
t
y
of
engi
ne
eri
ng
,
Christ
Univer
si
t
y
,
Ban
agl
or
e, I
n
dia
.
He is
hav
ing
m
ore
th
an
17
y
e
ars
of
t
eachi
ng
e
xper
ie
n
ce.
Evaluation Warning : The document was created with Spire.PDF for Python.