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.
3615
~
36
22
IS
S
N:
20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v
9
i
5
.
pp36
15
-
36
22
3615
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Enh
ancing netw
or
k lif
etim
e w
ith
an impr
oved MOD
-
L
E
ACH
Br
ijesh K
und
aliya, S
.
K.
H
ad
ia
Depa
rt
m
ent
o
f
E
le
c
troni
cs
and
C
om
m
unic
at
ion
E
ngine
er
ing,
Char
ota
r
Univ
ersity
o
f
Scie
n
ce a
nd
Technol
og
y
(CHA
RUSAT)
Univer
sit
y
,
Indi
a
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Feb
8
, 2
01
9
Re
vised
A
pr
8
,
201
9
Accepte
d
Apr
20
, 201
9
W
ire
le
ss
sensor
net
work
will
b
e
the
m
ost
domi
nat
ing
f
ie
ld
in
future
er
a.
The
re
ar
e
ce
r
tain
issues
which
wire
le
ss
sensor
net
work
suffers
from
.
The
m
ai
n
concern
with
wire
les
s
sensor
net
work
is
li
m
it
ed
en
erg
y
whi
ch
dire
c
tly
impact
on
net
work
li
fe
ti
m
e.
In
thi
s
pape
r
we
m
odify
the
cl
uste
r
sele
c
ti
on
pro
ce
d
ure
of
MO
DLE
ACH
.
MO
DLEACH
p
rotoc
ol
u
se
thre
shol
d
val
ue
for
se
le
c
ting
cl
uster
h
ea
d.
Once
a
c
luste
r
h
ea
d
is
sel
ec
t
ed,
i
t
retain
s
it
s
positi
on
unt
il
it
b
y
passes
th
e
th
r
eshold
li
m
it.
In
Basic
L
EACH,
it
does
no
t
use
an
y
thre
sh
old
val
u
e
but
i
t
ran
dom
l
y
sel
ec
ts
c
luste
r
h
ead
from
the
ava
i
la
bl
e
nod
es.
W
e
combine
th
e
proba
bi
li
sti
c
na
t
ure
of
LE
ACH
t
o
select
the
cl
uster
hea
d
and
thre
shold
base
sele
c
ti
on
of
c
luste
r
h
ea
d
of
MO
DLEACH.
W
e
al
so
app
l
y
p
roposed
m
odifi
c
at
ion
in
EAMM
H
protoc
ol
.
Our
m
ai
n
foc
us
is
on
the
enh
anc
ement
of
n
et
work
li
fe
ti
m
e
,
an
d
we
got
signifi
ca
n
t
improvem
ent
in
net
work l
ife
t
ime
.
Ke
yw
or
d
s
:
EAMM
H
LEAC
H
MODLE
AC
H
Netw
ork
li
fe ti
m
e
W
i
reless se
nso
r netw
ork
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
:
Brijes
h
Kun
daliy
a,
Dep
a
rtm
ent o
f
El
e
ct
roni
cs
and
Com
m
unic
at
ion
Engi
ne
eri
ng
,
Char
otar U
nive
rsity
o
f
Scien
ce an
d
Tec
hnol
og
y
(CH
ARU
S
AT)
U
niv
e
rsity
,
Chan
ga, G
ujar
at
, Ind
ia
.
Em
a
il
:
ku
ndal
iy
abr
ijes
h@y
ah
oo.co
m
1.
INTROD
U
CTION
In
wi
reless
s
ens
or
net
wor
ks,
the
root
of
m
axi
m
u
m
dissipati
on
of
energy
is
the
routing
of
inf
or
m
at
ion
[1]
.
Ther
e
are
ce
rtai
n
ways
are
dev
el
op
e
d
to
pro
vid
e
for
r
ou
t
ing
of
in
form
ation
wit
h
it
s
pros
an
d
cons.
I
n
ge
neral
,
the
ro
utin
g
pr
ot
oco
l
f
or
t
he
wireless
se
ns
or
net
works
cl
assifi
ed
in
four
m
ajo
r
cat
eg
or
ie
s:
1)
Data
cent
ric
Rou
ti
ng
al
go
rithm
2)
Hiera
rch
ic
al
Ro
utin
g
al
gorithm
3)
Geo
grap
hical
routin
g
in
for
m
at
ion
and
4)
Q
oS
ba
sed
r
ou
ti
ng
al
gorithm
[2
,
3].
In
the
data
cen
tric
al
go
rithm
,
routin
g
decisi
on
base
d
ei
ther
data
gen
e
rati
on
fro
m
the
senso
r
node
or
the
da
ta
de
m
and
e
d
by
the
sink
node
.
In
this
ty
pe
of
r
ou
ti
ng
pr
oto
c
ol,
the
data
ge
nerat
ed
by
the
no
de
is
m
or
e
im
portant
tha
n
th
e
node
it
sel
f.
The
m
ai
n
fo
cu
s
is
on
retrie
va
l
and
dissem
inati
on
of
in
f
or
m
at
ion
gen
e
rated
by
the
node.
N
or
m
al
ly
the
centric
ap
proac
h
com
es
with
flat
arch
it
ect
ure,
wh
e
re
each
node
of
netw
ork
play
s
equ
a
ll
y
i
m
po
rtant
ro
le
f
or
r
ou
ti
ng
of
inf
orm
at
ion
.
Go
s
sipin
g
[
4
-
7]
,
SPIN
[
5],
C
OU
C
AR
[6
]
,
CADR
[
7]
et
c.
are
the
fe
w
e
xam
ples
of
Da
ta
centric
al
gorithm
.
Figure
1
il
lustr
at
e
ver
y
basic
con
ce
pt
of
dat
a
centric
al
go
r
it
h
m
.
Her
e
cen
tre
node
ge
nerat
es
the
infor
m
at
ion
and
sen
d
a
dv
e
rtisem
ent
packet
to
the
net
w
ork
a
bout
ne
w
ly
gen
erate
d
in
form
ation
.
T
he
interest
ed
node
will
sen
d
the
re
que
st
pack
et
f
or
the
i
nfo
rm
ation.
The
ce
ntre
node
will
sen
d
the
inf
or
m
at
ion
to
inte
rested
node
on
ly
.
T
he
othe
r
way
c
ommun
ic
at
io
n
is
al
so
po
ssi
ble
w
her
e
si
nk
no
de
i
s
lookin
g
f
or
s
om
e
infor
m
at
ion
.
So
,
it
will
sen
d
inte
rest
propagati
on
to
the
entire
netw
ork.
The
no
de
wh
i
ch
has
t
he
require
d
in
form
at
i
on
wil
l
sen
d
the
data
t
o
the
sin
k
node
.
O
veral
l
in
t
his
ap
proac
h
e
it
her
s
ource
node
or
si
nk
no
de
init
ia
te
d
t
he
data
dissem
inati
on
process
.
In
Ge
ogra
ph
ic
al
routin
g
ap
pr
oach,
node
us
e
s
the
locat
io
n
i
nfor
m
at
ion
of
the
ot
her
node
for
routin
g
of
in
form
at
ion
.
The
no
de
will
send
the
data
to
near
ly
loca
te
d
node
in
th
e
directi
on
of
destinat
io
n.
F
or
the
locat
ion
in
for
m
at
ion
each
node
is
e
quip
pe
d
with
a
G
PS
or
any
ot
her
lo
cal
locat
ion
in
f
or
m
at
ion
al
gor
it
h
m
for
po
sit
io
ning
in
f
or
m
at
ion
of
node
.
It
is
f
urt
her
div
i
de
in
t
wo
sub
cat
eg
ori
es:
Un
ic
ast
r
ou
ti
ng
protoc
ol
and
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
:
3
6
1
5
-
3
6
2
2
3616
m
ul
ti
cast
ro
uting
protoc
ol.
I
n
unic
ast
routing
prot
oc
ol
pa
cket
sen
d
to
par
ti
cular
on
e
node,
w
her
e
as
in
m
ul
ti
cast
ro
uti
ng
prot
oc
ol
wil
l
send
the d
at
a to
m
ulti
ple
locat
ion
. G
PRS
[
8]
,
GFA
[
9],
GE
AR
[
10]
et
c.
ar
e
the
exam
ples
of
Geog
raphical
Rou
ti
ng
al
gori
thm
.
Figu
re
2
il
lustrate
the
basic
op
e
rati
on
f
or
Geog
r
aph
ic
al
inf
or
m
at
ion
w
her
e
no
de
us
es
locat
ion
in
for
m
at
ion
of
oth
e
r
no
de
of
the
netw
ork
t
o
f
or
ward
or
tra
nsm
it
the
data.
This
fi
gure
par
ti
c
ul
arly
ind
ic
at
es
the
GFA
al
gorith
m
wh
ere
the
e
ntire
area
of
the
netw
ork
div
ides
i
n
virtu
al
gr
i
d
t
o
s
i
m
plifie
d
locat
ion i
nfo
rm
at
ion
.
Figure
1
.
Ba
sic
operati
on
of da
ta
centric alg
ori
thm
Figure
2
.
Ba
sic
operati
on
of ge
ogra
ph
ic
al
routing o
pe
rati
on
The
QoS
protoco
l
fo
c
us
e
d
on
the
qual
it
y
of
li
nk
rathe
r
t
ha
n
the
distanc
e
or
ene
rg
y
c
onsu
m
ption.
Norm
al
l
y
in
WSN
the
Ene
r
gy
consum
ption
i
s
the
m
ai
n
concern
but,
in
se
ver
al
a
pp
li
cat
ion,
w
he
re
the
qu
al
it
y
of
data a
nd pr
om
pt d
el
ivery of d
at
a is
m
or
e im
po
rtant
li
ke m
ilit
ary app
li
cat
ion
or
m
edical
ap
plica
ti
on. In suc
h
a
sit
uation
QoS
prot
oco
l
is
pr
e
ferred
.
S
A
R
[11],
SPE
E
D
[
12]
,
MC
PF
[
13
]
et
c
.
a
r
e
the
exam
ple
of
Q
oS
protoc
ol.
I
n
Q
oS
prot
oco
l
m
ulti
ple
pat
h
a
re
create
d
f
ro
m
so
urce
t
o
destin
at
ion
s
o
t
hat
in
if
on
e
pa
th
fai
ls
due
to
so
m
e
un
pre
dicta
ble
reas
on
the
data
prom
ptly
m
ov
e
to
the
ot
her
path
.
Durin
g
the
path
f
or
m
at
ion
it
avo
i
ds
node
with
lo
w
e
nergy
lo
w
Qual
it
y
of
m
et
rics.
Th
e
Q
ualit
y
of
m
et
rics
m
ay
var
y
with
the
ap
plica
ti
on.
It
assu
red
ti
m
e
ly
delivery
of
data
a
t
cost
of
energy
an
d
re
so
urces
.
I
n
hie
rar
c
hical
routing
prot
oco
l,
cl
us
te
r
head
is
t
he
ke
y
el
e
m
ent
du
ri
ng
t
he
c
omm
u
nicat
ion
proce
ss.
It
is
the
ce
ntral
no
de
w
hi
ch,
c
onnect
th
e
cl
us
te
r
node
to the oth
er p
art
of
the
ne
twork
. S
el
ect
ion
of
cluste
r
he
ad
is t
he
cr
uci
al
p
ro
ces
s,
w
hi
ch
exte
ns
ively
aff
ect
the
netw
ork
li
f
e
tim
e.
Ther
e
are
num
ber
of
a
lgorit
hm
s
pr
op
os
e
d
f
or
the
cl
us
te
rin
g
-
base
d
com
m
un
ic
at
ion
li
ke
LEAC
H
[14],
PEGAS
IS
[
15
]
,
TEEN
[
16]
,
and
AP
TE
EN
[17]
et
c.
Gr
id
base
d
cl
us
te
rin
g
with
m
ob
il
e
sing
hav
i
ng
pr
e
de
fi
ned
pa
th
[
18
]
al
so
e
nh
a
nce
the
netw
ork
li
f
e
tim
e.
N
umber
of
optim
izati
on
te
c
hn
i
que
li
ke
Ar
ti
fici
al
Be
e
colo
ny
al
gorith
m
us
ed
f
or
e
ne
rg
y
ef
fici
ency
in
W
S
N
[
19]
or
f
uzzy
ap
proa
ch
is
ap
plied
f
or
the
cl
us
te
r
hea
d
sel
ect
ion
.
[20].
I
n
this
pap
e
r
w
e
com
par
e
an
d
analy
sed
the
netw
ork
li
feti
m
e
with
LEACH
[14]
and E
AMMH
[
21
]
a
nd MO
D
LEAC
H
[
22]
pro
t
oco
l
with
propose
d
m
od
ifi
cat
ion
in
them
.
Lo
w
Ene
rg
y
A
dap
ti
ve
Cl
ust
er
ing
Hierarc
hy
(LE
ACH
)
is
the
pr
im
e
pr
oto
c
ol
w
hich
gi
ves
the
idea
of
the
cl
us
te
r
-
bas
ed
c
omm
un
ic
at
ion
.
LE
AC
H
[14]
is
s
o
popula
r
t
hat
after
18
ye
ars
of
it
s
existe
nce
it
is
sti
ll
ho
l
ding
the
dom
inance
in
research
c
ommun
it
ie
s.
LE
ACH
[14]
has
nu
m
ero
us
s
ucces
so
rs
with
i
m
pr
ov
e
d
ver
si
on
sta
rtin
g
from
LEACH
to
Du
al
H
op
L
EAC
H
[23
]
.
LEAC
H
[14
]
op
e
rati
on
s
pl
it
s
in
to
tw
o
s
ta
ges.
The
fi
rst
sta
ge
is
known
as
s
et
up
ph
ase
,
w
her
e
t
he
cl
us
te
r
is
form
ed
an
d
cl
us
te
r
hea
d
is
sel
ect
ed.
Figure
3
il
lustrate
the
operati
on
of
le
a
ch
prot
oco
l
.
T
he
seco
nd
sta
ge
is
known
as
ste
ady
sta
te
stag
e,
wh
ic
h
inc
orp
or
at
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
En
hancin
g net
work li
fe
ti
me w
it
h
an im
prov
ed
M
OD
-
L
EA
CH
(
Brij
esh
K
undali
ya
)
3617
the
act
ual
data
transm
issi
on
.
In
a
set
up
phase,
entire
network
c
onsist
of
ra
ndom
ly
de
plo
ye
d
s
ens
or
node
i
s
bif
ur
cat
e
in
m
eagr
e
cel
l
known
a
s
cl
us
te
r
.
On
e
node
f
rom
the
cl
us
te
r
is
sel
ect
ed
as
c
luster
he
ad
w
hi
ch
will
work
as
bri
dge
node
bet
wee
n
sens
or
no
de
a
nd
the
ba
se
sta
ti
on.
T
he
sel
ect
i
on
of
cl
us
te
r
he
ad
is
de
pendi
ng
on
rand
om
m
ann
er as:
(
)
=
/
[
1
−
∗
(
(
1
)
]
⋴
(1)
=
0
ℎ
Wh
e
re:
n
=
ra
ndom
n
um
ber
b
et
wee
n 0 to 1
P = Pro
ba
bili
ty of cluste
r head
G= S
et
of
node
s which
where
no
t t
he
cl
ust
er
node
i
n pr
e
vious
rou
nd
Figure
3
.
Leac
h
a
rch
it
ect
ure
and ope
rati
on
Each
no
de
fro
m
the
cl
us
te
r
gets
the
cha
nc
e
to
beco
m
e
t
he
cl
us
te
r
hea
d.
O
nce
a
cl
ust
er
head
is
sel
ect
ed,
it
broad
ca
st
the
ad
ver
ti
sem
ent
m
essage
re
ga
rd
i
ng
it
s
hea
dship.
I
f
no
de
rec
ei
ved
m
or
e
th
an
one
adv
e
rtise
m
ent
m
essage,
it
w
il
l
sel
ect
cl
us
te
r
head
w
hos
e
a
dverti
sem
e
nt
m
essage
con
ta
in
hi
gh
e
r
sign
al
stren
gth
.
I
n
LE
ACH,
the
cl
ust
er
head
sel
ect
ion
is
base
d
on
pr
oba
bili
sti
c
m
ann
er
w
hic
h
giv
es
cha
nce
to
eac
h
node
of
a
cl
us
t
er
to
bec
om
e
cl
us
te
r
hea
d.
D
ur
i
ng
the
cl
us
t
er
head
sel
ect
ion
pr
ocess,
it
do
e
s
no
t
ta
ke
ener
gy
sta
tus
of
the
node
.
It
will
rando
m
ly
sel
ect
t
he
node.
Ene
rgy
Aw
a
re
m
ulti
hop
m
ulti
path
hiera
rch
ic
al
protoc
ol
(EA
MM
H)
[
21
]
is
al
so
a
cl
us
t
er
-
base
d
com
m
un
ic
at
ion
pro
tocol.
As
in
L
EACH
[
14]
,
it
s
op
e
rati
on
divi
des
in
two
pa
rts:
Set
up
phase
an
d
Data
tra
ns
m
iss
ion
phase.
I
niti
al
ly
,
the
de
ploy
ed
no
des
fin
d
it
s
neig
hbou
r
us
in
g
any n
ei
ghbour
disco
ver
y al
gorithm
. A
fter the
n
ei
gh
bour d
i
sco
ver
y, cl
us
te
rs
are
b
ei
ng cre
at
ed
an
d
cl
us
te
r
he
a
d
is
sel
ect
ed
fr
om
the
cl
us
te
r
no
de
s.
This
pro
cess
is
identic
al
to
the
LEAC
H
[
1
4]
prot
ocol
.
In
data
tra
nsm
issi
on
it
is
assum
ed
t
hat
al
l
the
no
de
s
had
a
data
to
sen
d,
so
par
t
ic
ular
tim
e
s
lot
is
a
ll
ocated
t
o
each
no
de
of
the
cl
us
te
r.
In
E
A
MM
H
a
routin
g
ta
ble
is
pres
erv
e
d
by
each
node
wh
ic
h
is
per
i
od
ic
al
ly
updated
.
N
ow
wh
e
n
nodes
get
da
ta
to
trans
fer
fro
m
it
s
neigh
bo
ur,
it
will
choos
e
the
opti
m
a
l
path
base
d
on
i
nfor
m
at
ion
ava
il
able
in
it
s rou
ti
ng ta
ble.
It u
se
s the
functi
on:
ℎ
=
(
ℎ
∗
)
(2)
Wh
e
re:
K
= C
on
sta
nt
E
avg
= Cu
rr
e
nt
p
at
h ave
ra
ge Ener
gy
h
= Mi
nim
u
m
hope
c
ount i
n
c
urren
t
path
t = Tra
ff
ic
i
n
c
urren pat
h
A
pat
h
with
the
highest
va
lue
of
h
is
sel
ect
ed
for
da
ta
delivery.
So
,
durin
g
th
e
routing
of
inf
or
m
at
ion
, E
AMMH sele
ct
energy ef
fici
en
t path.
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
:
3
6
1
5
-
3
6
2
2
3618
Pr
oble
m
sta
teme
nt
,
in
a
cl
us
te
r
-
base
d
c
om
m
un
ic
at
ion
,
cl
us
te
r
he
ad
play
a
key
r
ole
in
data
transm
issi
on
.
I
t
is
the
ce
ntra
l
point,
w
her
e
each
an
d
eve
ry
data
of
t
he
cl
us
te
r
is
rou
te
d
th
rou
gh.
Data
transm
issi
on
co
nsum
es
m
or
e energy tha
n
da
ta
p
r
ocessin
g.
Cl
us
te
r
hea
d
is
li
kely
to
con
s
um
e
m
or
e energy as it
need
to
t
ran
sm
it
the
data
fro
m
a
ll
the
node
s
to
base
sta
ti
on
a
nd
s
om
e
info
rm
at
ion
f
ro
m
base
sta
ti
on
t
o
eac
h
node
of
the
cl
ust
er.
If
the
sel
e
ct
ed
node
ha
s
low
e
nergy,
it
will
con
s
um
e
i
ts
ener
gy
to
tra
ns
fe
r
the
data
com
es
from
the
oth
er
nodes
of
cl
us
te
r
an
d
e
ven
t
ually
it
le
ads
to
de
a
d
no
de.
T
he
r
egio
n
of
that
node
is
c
ut
off
f
ro
m
the n
et
w
ork.
D
ur
i
ng
the
cl
us
t
er
sel
ect
io
n
it
i
s
ext
rem
ely
i
m
portant
t
hat w
e
sel
ect
a n
ode
w
it
h
higher
e
ne
rg
y
a
s
cl
us
te
r
hea
d.
As
disc
us
se
d
in
above
sect
ion,
LEAC
H
[
14
]
an
d
EAM
MH
[21]
hav
e
an
identic
al
cl
us
te
r
sel
ect
ion
pro
c
e
dure. Durin
g
t
he
cl
us
te
r hea
d
sel
ect
ion
,
they
d
id
no
t
c
on
si
de
r
ene
rg
y l
evel
o
f
th
e node
which is
sel
ect
ed
as
cl
ust
er
hea
d.
I
f
t
he
node
wit
h
lo
w
e
nergy
is
sel
ect
ed,
it
m
ay
resu
lt
in
dea
d
node
an
d
eve
nt
ually
a
blan
k
s
pot i
n n
et
work. T
hat
bl
ank s
po
t i
n
the
netw
ork wil
l l
ead to an
unacc
eptable
operati
on on net
w
ork
.
LEAC
H
[14]
nam
e
it
sel
f
a
l
egacy,
an
d
has
a
nu
m
ber
so
s
uccess
or.
MO
DLE
ACH
[
22]
is
m
od
ifie
d
le
ach
protoc
ol
.
It
a
ddres
ses
the
pro
blem
of
cl
us
te
r
hea
d
sel
ect
io
n
c
he
akin
g
t
he
e
ne
rg
y
le
vel
of
node.
In
MO
DLE
AC
H
[22]
auth
or
us
es
the
thres
hold
val
ue
to
sel
ect
the
cl
us
te
r
head
duri
ng
s
et
up
ph
a
se.
I
f
the
cl
us
te
r
head
ho
lds
ene
rg
y
a
bo
ve
the
t
hr
es
hold
val
ue,
it
will
con
ti
nu
e
as
cl
us
te
r
hea
d
f
or
nex
t
rou
nd.
It
m
eans
that
cl
us
te
r
he
ad
will
con
ti
nue
as
cl
us
te
r
he
ad
unti
l
it
s
en
erg
y
by
pass
th
e
threshold
value.
It
pe
rfor
m
s
m
or
e
eff
ic
ie
ntly
tha
n
the
L
EAC
H
[14].
But
pro
blem
with
this
appr
oach
is
the
no
de
wh
ic
h
is
the
cu
rr
e
nt
cl
us
te
r
head
will
rem
a
in
as
cl
us
te
r
he
ad
un
ti
ll
it
s
energy
by
passe
s
the
th
res
ho
l
d
value
.
T
his
w
il
l
create
a
pro
lem
of
un
e
ve
n
ene
r
gy
reducti
on
in
t
he
net
wor
k.
T
hat
m
a
y
resu
lt
in
vital
co
ns
e
qu
e
nce.
For
e
xam
ple,
durin
g
cl
us
te
r
head
sel
ect
io
n
process
in
fi
rst
ro
un
d
node
w
hich
is
ver
y
ne
ar
to
base
sta
ti
on
is
sel
ect
as
cl
us
te
r
hea
d.
N
ow
it
rem
ai
ns
as
cl
us
te
r
hea
d
un
ti
l
it
s
energy
rea
ched
bel
ow
t
he
thres
hold
value.
Once
it
s
energy
reac
he
d
bel
ow
thres
ho
l
d
valu
e
it
will
not
ge
t
change
to
be
a
cl
us
te
r
hea
d,
w
hich
m
ean
now
the
node
w
hich
is
fa
r
fro
m
the
base
sta
ti
on
w
il
l
be
a
cl
us
te
r
head.
N
ow
ne
w
cl
us
te
r
hea
d
nee
d
to
se
nd
the
data
fro
m
far
distance
wh
ic
h
increases
the e
nergy
co
nsum
ption
. T
his
phen
om
eno
n wil
l
le
ad
to
quick
r
e
duct
ion i
n
e
ne
rgy
o
f netw
ork
r
e
su
lt
s
in d
ea
d netw
or
ks
.
2.
PROP
OSE
D ALGO
RITH
M
As
we
discusse
d
in
ab
ove
se
ct
ion
LE
AC
H
[14]
an
d
E
AM
MH
[
21]
suffe
rs
f
r
om
the
early
energy
reduece
know
as b
la
nk
spot
pro
blem
. MODLEAC
H
[
22]
s
olv
e t
he pr
ob
le
m
o
f
early
en
er
gy r
e
du
ct
i
on
i
n n
od
e
,
bu
t
s
uffere
s
f
r
om
the
un
e
ve
n
en
er
gy
re
duct
ion
wh
ic
h
le
ads
to
dea
d
ne
twork
.
I
n
our
al
gorithm
we
address
these tw
o pro
bl
e
m
s b
y usin
g po
sit
ive
p
a
rt of
LEA
C
H [14]
and MO
DLE
A
CH [2
2]. T
he
e
arly
en
er
gy
dro
p wil
l
resu
lt
in
bla
nk
spot,
w
hich
can
be
av
oide
d
by
pro
vid
i
ng
the
th
res
hold
-
based
sel
ec
ti
on
of
cl
us
te
r
head.
The
un
e
ve
n
e
ne
rg
y
reducti
on
pro
blem
wh
ic
h
occ
urred
i
n
M
OD
L
EAC
H
[
22]
can
be
a
vo
i
ded
by
giv
in
g
equ
al
appo
rtu
nity
to
al
l
oth
er
node
s
for
beco
m
ing
a
cl
us
te
r
hea
d.
That
we
can
pr
ov
i
de,
us
in
g
prob
a
bili
sti
c
sel
ect
ion
of the cl
us
te
r h
ead a
fter eac
h r
ound
of comm
un
ic
at
on.
In
p
r
opose
d
a
lgorit
hm
we
took
th
e
pro
ba
bili
sti
c
sel
ection
proce
dure
from
LEACH
[
14]
an
d
Thes
ho
l
d
at
trib
ute o
f
MO
DLE
ACH
[
22]
, w
he
re each
no
de
wh
ic
h
co
ntains
the
m
or
e energy than
the th
r
esh
old
energy
(Et
h)
gets
eq
ual
appo
rtu
nity
in
cl
us
te
rh
ea
d
sel
ect
ion
by
app
li
ng
th
e
pro
ba
bili
sti
c
m
ann
e
r.
The
PS
EU
DO
cod
e
of
t
he
al
gorithm
dep
ic
te
d
as
belo
w.
S
liv
con
ta
ins
t
he
al
ive
nodes
of
ne
twork
,
S
ch
co
ntains
the
node
s
w
hic
h
are
el
e
gib
el
f
or
t
he
cl
us
te
r
he
ad,
a
nd
S
chf
is
the
final
sel
ect
ed
cl
us
te
r
hea
d.
First
li
ne
in
di
cat
es
tha
t
at
init
ia
l
sta
ge
al
l
the
nodes
are
al
ive
a
nd
cl
us
te
r
hea
d
is
ye
t
to
sel
e
ct
.
Line
num
ber
2
to
6
sho
w
s
that
it
checks
the
ene
rg
y
le
vel
of
node
an
d
if
the
node
e
nergy
is
gr
eat
er
th
an
th
e
threshold
val
ue
that
node
w
il
l
be
consi
der
e
d
as
cl
us
te
r
hea
d.
On
ce
we
fin
d
t
he
al
l
then
ode
wh
ic
h
a
re
el
egible
f
or
the
cl
us
te
r
he
ad
,
sel
ect
the
cl
us
te
r head
which
ha
ve
th
e
hi
gh
est
pro
bab
il
it
y, wh
ic
h i
s in
dicat
ed
i
n
la
st f
ew
li
nes
of the
cod
e
.
Prop
os
ed
A
l
gori
th
m
:
I
mpr
ov
e
d
MO
DLEACH
1: Init
ia
ll
y S
liv
=
S
,
S
ch
=
Ø, S
c
hf
=
Ø
2:
f
or
eac
h n
ode
i
do
3: calc
ulate
E
n
for
al
l t
he n
ode
4:
if
E
n
(
i
)
>
E
th
5:
S
ch
U
S
liv
(i
)
6:
End i
f
7:
End
f
or
8:
f
or
al
l t
he
nod
e
in S
ch
9: do se
le
ct
r
from
(
0
to
1)
10
: c
om
pu
te
T
(n) fo
r
al
l S
ch
11
:
S
chf
=
No
de
which
ha
ve
th
e h
ig
hest
pro
ba
bili
ty
12
:
e
nd
f
or
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
En
hancin
g net
work li
fe
ti
me w
it
h
an im
prov
ed
M
OD
-
L
EA
CH
(
Brij
esh
K
undali
ya
)
3619
By
do
in
g
s
o,
w
e
able
to
pro
vid
e
eq
ual
c
hang
e
to
eve
ry
no
de
to
be
cl
us
te
r
head,
ke
epi
ng
energy
le
vel
of
net
work
e
qual
ly
distrib
ut
ed.
This
unif
orm
ener
gy
re
duct
ion
th
rou
ghou
t
t
he
cl
us
te
r
,
will
a
vo
i
d
th
e
early
blan
k
s
pot an
d i
ncr
ease t
he ne
twork
li
fe tim
e
.
3.
E
X
PERI
MEN
TAL
RES
UL
TS
A
ND AN
A
LYSIS
We
c
om
par
e
the
Im
prov
e
d
ve
rsion
of
M
OL
EACH
with
M
OD
L
EAC
H
[
22]
,
E
AMMH
[
21
]
a
nd
wit
h
LEAC
H
[
14]
. We
us
e M
ATL
AB 20
15 as
pl
at
fo
rm
f
or the
si
m
ulati
on
. T
he
sp
eci
ficat
io
n i
s
show
n
i
n
T
a
ble
1
.
Table
1
.
Sim
ul
at
ion
p
aram
et
e
r
Para
m
eters
Valu
e
Grid Ar
ea
1
0
0
*
1
0
0
m
2
Maxi
m
u
m
T
rans
m
iss
io
n
Ran
g
e
1
0
0
m
ete
r
Maxi
m
u
m
I
n
itial
E
n
ergy
o
f
the No
d
e
0
.5 Jo
u
le
Pack
et Size
M
ax
i
m
u
m
1
0
0
0
By
ets
Data Agg
regatio
n
Energy
5*10
-
5
J/B
it/Signal
Thresh
o
ld
E
n
ergy
0
.5*
1
0
-
5
J
o
u
le
Energy Co
n
su
m
ed
b
y
tr
an
s
m
itter
3
5
*
1
0
-
9
J/B
it /
m
2
Energy Co
n
su
m
ed
b
y
Receiv
er
1
5
*
1
0
-
9
J/B
it /Packet
Energy Co
n
su
m
ed
b
y
A
m
p
lif
ier
1
0
*
1
0
-
12
J/B
it /
m
2
Figure
4
de
pict
the
si
m
ulati
on
net
work
ge
ne
rated
in
a
100*10
0
m
2
area.
We
com
par
e
the
al
li
ve
node
of
ne
tw
ork
aft
er
certai
n
r
ound
of
c
omm
un
ic
at
ion
.
Fig
ur
e
5
ind
ic
at
es
the
com
par
isi
on
be
tween
LE
AC
H
[14]
and
pro
posed
al
gorithm
IMP
-
MO
DLE
ACH
.
If
we
care
fu
ll
y
ob
se
rv
e
d
the
gr
a
ph
it
cl
earl
y
ind
ic
at
es
tha
t
in
a
first
few
rou
nd
LEAC
H
an
d
IMP
-
MO
DLE
ACH
perform
in
eq
u
al
eff
ic
ie
nt.
But
after
200
hu
ndress
r
ounds
energy
of
node
in
LEACH
sta
rts
decr
esi
ng
r
apidly.
The
res
on
beh
i
nd
this
is
LEACH
will
no
t
chec
k
thre
sh
ol
d
le
vel
an
d
ch
ose
rand
om
l
y
a
cl
us
te
r
he
ad
.
D
ue
to
i
ns
uffici
e
nt
ene
rg
y
this
cl
us
te
rh
ea
d
wi
ll
le
ad
to
dea
d
nod
e.
Wh
e
re
as
IMP
-
MO
DLECH
s
urpass
t
he
LE
ACH
i
n
ef
fici
ent
m
ann
er.
Fi
gure
6
com
pare
the
al
li
ve
no
de
of
EAMM
H
[21]
and
IMP
-
MO
DLE
ACH.
I
n
t
hat
IMP
-
MO
D
LEAC
H
gi
ves
the
bette
r
outp
ut.
Fi
gure
7
co
m
par
e
the
MOD
LE
A
CH
[
22
]
an
d
I
MP
-
MO
DLE
A
CH.
U
p
to
fir
st
few
hundre
ss
rou
n
ds
MO
DLE
ACH
a
nd
IMP
-
MODLE
AC
H
perform
equ
ll
y
well
.
As
we
know
that
MODLE
ACH
will
keep
no
de
as
cl
us
te
r
head
t
il
l
it
s
energy
goes
be
low
the
th
res
ho
l
d
le
vel.
Thi
s
le
ads
the
m
a
xim
u
m
ener
gy
dr
ai
n
ou
t
f
rom
a
c
luster
he
ad
an
d
even
t
ually
the
node
will
no
t
a
ble
to
b
ec
om
e
the
cl
us
te
r
hea
d
on
f
uture.
If
this
node
is
ne
ar
to
ba
se
sta
ti
on,
the
nex
t
cl
us
te
r
he
ad
is
far
f
ro
m
the
base
sta
ti
on
wh
ic
h
m
en
as
it
has
to
transm
it
to
a
lon
g
distance
.
Th
is
will
consum
e
m
or
e
energy
a
nd
le
ads
node
t
o
de
ad
node.
This
will
create
une
ven
ene
rg
y
reducti
on
in
the
ne
twor
k.
This
phen
om
enon
we
ca
n
obser
ve
f
ro
m
t
he
grap
h.
Fi
gure
8
we
com
par
e
d
the
dea
d
node
a
fter
a
certai
n
nu
m
ber
of
rou
nd
s
.
It
is
ver
y
cl
eare
that
IMP
-
MO
DLE
AC
H
will
incese
the
netw
ork
li
f
e
tim
e
in
sign
ific
ant
m
ann
er
whe
n we c
om
par
e it
w
it
h
LE
AC
H
[
14
]
.
EA
MM
H [21]
and M
ODLEAC
H
[
22]
.
Figure
4
.
Sim
ulati
on
n
et
w
ork
Figure
5
.
Acti
ve
nod
e
co
m
pari
sion
betwee
n LEAC
H
and IMP
-
LE
A
CH
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
:
3
6
1
5
-
3
6
2
2
3620
Figure
6
.
Acti
ve
nod
e
co
m
pari
so
n bet
ween E
AMMH
and IMP
-
MO
D
LEAC
H
Figure
7
.
Acti
ve
nod
e
co
m
pari
sion
betwee
n
MODLE
AC
H and IMP
-
MO
D
LEAC
H
Figure
8
.
Dea
d n
od
e
co
m
par
is
ion
of LE
ACH
, E
AMMH, M
OD
L
EAC
H
a
nd
IMP
-
M
OD
L
EACH
4.
CONCL
US
I
O
N
Pr
op
os
e
d
al
gor
it
h
m
co
m
bin
es
the
at
tribu
te
s
of
LE
ACH
an
d
MOD
LEAC
H.
W
e
use
thr
esh
old
bas
e
sel
ect
ion
of
cl
us
te
r
head
fro
m
MODLEA
C
H,
wh
ic
h
gi
ves
chan
ce
t
o
no
de
with
hi
gh
e
r
e
nergy
to
be
a
c
luster
head.
The
pro
ba
bili
sti
c
natur
e
of
cl
us
te
r
hea
d
sel
ect
ion
fro
m
LEACH
wil
l
ta
ke
care
of
e
ven
e
nergy
re
duct
ion
and
a
vo
i
d
the
early
blind
spo
t
in
netwo
r
k.
Pr
op
os
e
d
al
gor
i
th
m
increased
the
li
fe
t
i
m
e
of
the
net
wor
k
m
uch
m
or
e
than
the
LEAC
H
an
d
E
AMMH
al
gori
thm
.
It
al
so
of
fer
s
sig
nifica
nt
adv
a
ntage
over
MOD
LEAC
H
i
n
te
rm
of
li
fe
ti
m
e
of
the
net
work.
We
us
e
hard
t
hr
es
hold
va
lue
wh
ic
h
is
fixe
d
valu
e.
I
n
f
uture,
a
dap
ti
ve
thres
ho
l
d
valu
e
can
fo
r
the
c
luster
hea
d
sel
ect
ion
,
w
hich
will
def
init
el
y
i
m
pr
ove
the
ne
twork
li
fe
tim
e.
On
e
can
us
e
opti
m
i
ze
al
go
rithm
t
o
sel
ect
best
no
de
f
or
the
cl
us
te
r
head
wh
i
ch
will
again
im
pr
ov
e
the
ne
twork
li
fetim
e.
REFERE
NCE
S
[1]
B.
Kundaliy
a
an
d
S.
Hadia
,
"
A
Com
par
at
ive
An
aly
s
is
of
Optimiza
t
ion
Algorit
h
m
s
for
W
ire
le
ss
Sensor
Network
,
"
Inte
rnational
Jo
urnal
on
Fut
ure
Re
vo
lut
ion
in
Computer
Sci
en
ce
&
Com
municat
ion
Engi
n
ee
ri
ng
,
v
ol
.
3
,
no.
1
0,
Oct
2017.
[2]
D.
W
al
te
n
egus
and
P.
Chr
isti
an
,
Fundamen
tal
o
f
Wireless
Senso
r
Net
works
The
ory
and
Prac
ti
c
e
,
John
W
ile
y
&
Sons
,
Lt
d
,
Ch.
7
,
2017
.
[3]
J.
N.
Al
-
kaka
ri
,
A.
E.
Kala
m
,
"
Routi
ng
Te
ch
nique
s
in
wire
l
ess
sensor
net
w
ork:
A
surve
y
,
"
IEE
E
wireless
Coomunicat
ion
,
v
ol.
11
,
no
.
6
,
pp
.
6
-
28
,
De
c
2004
.
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
En
hancin
g net
work li
fe
ti
me w
it
h
an im
prov
ed
M
OD
-
L
EA
CH
(
Brij
esh
K
undali
ya
)
3621
[4]
S.
Hede
tni
emi
a
nd
A.
Li
estman
,
"
A
surve
y
of
goss
ipi
ng
and
broa
dca
st
ing
in
co
m
m
unic
at
ion
netw
orks
,"
Net
works
,
vol.
18
,
no
.
4
,
pp
.
319
–
349
,
1988
.
[5]
J.
Kulik,
W
.
R
.
Heinz
e
lman
and
H.
Bal
akr
ishna
n
,
"
Negotiati
on
base
d
protoc
o
ls
for
dissim
ina
ti
n
g
informati
on
in
wire
le
ss
sensor
net
works
,"
.
W
irl
ess sensor
net
wo
rks
,
vol
.
8
,
no
.
2/
3,
pp
.
169
-
185
,
2002
.
[6]
Y.
Yao
and
J.
Gehrke
,
"
The
C
ougar
appr
o
ac
h
to
in
-
n
et
work
quer
y
proc
essing
in
sensor
n
etw
orks
,"
SIGMO
D
Re
cord
,
vol
.
31
,
no.
3
,
pp
.
9
–
18
,
Sep
2002
.
[7]
M.
Chu,
H
.
Ha
uss
ec
ker
and
F.
Zha
o
,
"
Sca
la
bl
e
info
rm
ation
-
d
rive
n
sensor
qu
er
y
ing
and
rout
ing
for
ad
ho
c
het
ero
g
ene
ous
s
ensor
net
works
,"
Inte
rnationa
l
Journal
of
High
Pe
rform
ance
C
omputing
Appli
cat
ions
,
vol.
16
,
no.
3
,
pp
.
293
–
3
13,
2002
.
[8]
B.
Karp and
H.
T
.
Kung,
"
GP
SR
:
Gree
d
y
per
imeter sta
t
el
ess rout
in
g
for
wire
le
ss
ne
tworks
,"
Proc. Of
th
e
6th
Annua
l
Inte
rnational
co
nfe
renc
e
on
Mob
il
e
computi
ng
an
d
Net
work
ing
,
2
000.
[9]
Xu
Y.,
Heid
ema
nn
J.
and
Estri
n
D.
,
"
Geogr
aph
y
informed
en
erg
y
conse
rva
ti
on
f
or
ad
-
hoc
routi
n
g
,"
Proc
.
Of
the
7th
annual
Inte
r
nati
onal
Con
fe
re
nce
on
Mobi
le Computi
ng
and
N
et
working
,
2001
.
[10]
Yu
Y.,
Govindan
R.
,
and
Estri
n
D.
,
"
Geogra
phic
al
and
Energ
y
Aw
are
Routi
ng:
A
rec
ursive
dat
a
dissem
ina
ti
on
protoc
ol
f
or
wi
rel
ess
sensor
n
et
works
,"
Tech
nic
al
report
.
U
CL
A/
CSDT
R
010023
,
UCL
A
C
omputer
Scienc
e
Department
,
200
1.
[11]
K.
Sohrabi,
J.
Gao,
V.
Aila
wadh
i
and
G.
J.
Potti
e.
,
"
Protoco
ls
for
self
-
orga
nization
of
a
wire
le
ss
sensor
net
work
,"
IEE
E
Pe
rs
onal
Comm
unic
ati
ons
,
vol
.
7
,
no
.
5
,
pp
.
16
–
27
,
Oct
200
0.
[12]
T.
He
,
J.
A.
Sta
nkovic
,
C
.
Lu
a
nd
T.
Abdel
za
h
er
,
"
SP
EE
D:
A
stat
eless
protoc
o
l
for
re
al
-
t
ime
c
om
m
unic
at
ion
i
n
sensor
net
works
,"
In
Proceedi
n
gs
of
the
23rd
Inte
rnationa
l
Confe
renc
e
on
Distribute
d
Co
mputing
Syste
m
s
,
Providenc
e
,
RI
,
US
A,
pp.
46
–
55
,
Ma
y
2003
.
[13]
W
.
Ye,
J.
Hei
d
emann
and
D.
Estri
n
,
"
An
e
ner
g
y
-
eff
i
cient
MA
C
protoc
ol
for
wire
le
ss
sensor
net
works
,"
In
Proceedi
ngs
of
IE
EE INF
OC
OM’02
,
vol
.
3
,
p
p.
1567
–
1576
,
J
un
2002.
[14]
W
.
R.
He
inz
e
lman,
A
.
Chandr
ak
asa
n
and
H.
B
alakri
shnan
,
"
Ene
r
g
y
-
e
fficie
nt
com
m
unic
at
ion
prot
ocol
for
w
ire
l
ess
m
ic
ro
sensor
net
works
,
"
In
Proce
edi
ngs
of
th
e
I
EE
E
Hawaii
In
t
ernati
onal
Conf
ere
nce
on
S
ystem
Sci
ences
,
Ma
ui,
HI,
US
A,
pp.
11
0
,
Jan
2000
.
[15]
S.
Li
ndse
y
and
C.
S.
Ragha
v
endr
a,
"
PEGA
SIS
:
P
ower
eff
icient
g
at
her
ing
i
n
sensor
informati
on
s
y
stems
,
"
In
Proceedi
ngs
of
th
e
I
EE
E
A
erospace
Conf
ere
n
ce
,
Big
Sk
y
,
MT
,
US
A,
Mar
200
2.
[16]
A.
Manje
shw
ar
and
D.
P
.
Agra
wal
,
"
TE
EN:
A
protoc
o
l
for
en
hanc
ed
eff
ic
i
ency
in
wir
el
ess
se
nsor
net
works
,
"
In
Proceedi
ngs
of
th
e
1st
In
te
r
nati
onal
Wor
kshop
on
Paral
le
l
and
Distribut
e
d
Computing
Is
sues
in
Wire
le
s
s
Net
works and
M
obil
e
Computing
,
San
Franc
isco
,
US
A,
Apr 2001.
[17]
A.
Manje
shw
ar
and
D.
P.Agrawal
,
"
AP
TE
E
N:
A
hy
br
id
protoc
ol
for
eff
i
ci
en
t
routi
ng
a
nd
comprehe
nsive
informati
on
re
tr
ie
va
l
in
wire
le
s
s
sensor
net
works
,
"
In
Proce
e
dings
of
the
2
nd
Inte
rnationa
l
Workshop
on
Parall
el
and
Dist
ribute
d
Computi
ng
Iss
ues
in
Wirel
ess
Net
works
a
nd
Mobil
e
Computing
,
Ft
.
L
aude
r
dal
e
,
FL,
US
A,
Apr 2002.
[18]
T.
Kos
ti
war
an
,
M.
Krishnan
,
"
Grid
-
base
d
Cl
usteri
ng
wi
th
P
red
efi
n
ed
Pa
th
Mobili
t
y
fo
r
M
obil
e
Sink
Da
ta
Coll
ection
to
E
xte
nd
Network
Li
fe
ti
m
e
in
W
ire
le
ss
Sensor
Networks,
"
IETE
Techni
ca
l
Review
,
vol
.
29,
no
.
2
,
2012.
[19]
M.
Za
h
id,
S.
Na
vra
ti,
Q.
M.
Ijaz and
A.W.
Chang
,
"
H
y
br
id
Artif
icial
B
ee
Co
lon
y
Algorit
hm
for
an
Ene
rg
y
Eff
i
ci
en
t
Inte
rne
t
of
Things
base
d
on
W
ire
le
ss
Sensor
Network,
"
IET
E
Technical
Revie
w
,
v
ol
.
34
,
no.
sup1
:
TI
TR
Supplement
issu
e:
5G
W
ir
el
ess
with
Cognit
ive
Radi
o
and
Io
T
,
p
p.
39
-
51
,
2017
.
[20]
A.
K.
Singh
an
d
N.
Purohit,
S.
Verm
a
,
"
Fuzz
y
logi
c
b
ase
d
cl
u
steri
ng
in
w
ire
l
ess
sensor
net
works
:
A
surve
y
,
"
Inte
rnational
Jo
urnal
of El
e
ct
ro
nic
s
,
v
ol
.
100,
n
o.
1
,
2013
.
[21]
M.
R.
Mundada
,
V.
Cy
ril
R
aj
and
T.
Bhuvane
sw
a
ri
,
"
Ene
rg
y
Aw
a
re
Multi
-
Hop
Multi
-
Path
Hie
rar
c
hic
a
l
(EAMM
H)
Routi
ng
Protoc
ol
for
W
ir
el
ess
Sensor
Networ
ks,
"
European
Journal
of
Scie
nti
fic
Re
search
,
v
ol.
88,
n
o
.
4,
Oct
2012.
[22]
D.
Mahm
ood,
N.
Java
id
,
S.
M
ahmood,
S.
Qur
eshi3,
A.
M.
M
emon,
T.
Za
m
a
n5,
"
MO
DLEACH:
A
Vari
ant
of
LE
ACH
for
WS
Ns
,
"
Ei
ghth
Inter
nati
onal
Confer
enc
e
on
Broadband,
Wirel
ess
Computing,
Comm
unic
ati
on
and
Appl
ic
a
ti
ons
,
20
13.
[23]
S.
K.
Singh,
P.
Kum
ar
and
J.
P.
Singh
,
"
A
S
urve
y
on
Suc
essor
of
LE
ACH
Protocol
,"
IE
EE
Ac
ce
ss
,
v
ol
.
5
,
pp
.
4298
-
4328
,
2017
.
BIOGR
AP
H
I
ES
OF
A
UTH
ORS
Bri
jesh
L
Kun
dali
y
a
is
a
PhD
Schola
r
wi
th
th
e
Depa
rtment
of
E
le
c
troni
cs
and
C
om
m
unic
at
ion
Engi
ne
eri
ng
a
t
C
S
Pate
l
Instit
ute
of
T
ec
hnol
og
y
–
Charotar
Univer
sit
y
of
Scie
nc
e
an
d
Te
chno
log
y
,
Ch
anga
,
Anand
,
G
uja
ra
t,
Indi
a.
H
e
recei
ved
B
ac
h
el
or
of
Eng
ine
e
ring
degr
e
e
in
El
e
ct
roni
c
s
and
Comm
unic
at
io
n
Engi
nee
r
ing
from
Saura
shtra
Univer
sit
y
a
nd
Master
of
Engi
ne
eri
ng
d
e
gre
e
in
E
lectr
o
nic
s
and
Com
m
unic
at
ion
Eng
ine
er
ing
with
spec
i
al
i
za
t
ion
of
Com
m
unic
at
ion
S
y
stems
Engi
ne
eri
ng
from
Gujar
at
Univer
sit
y
.
He
is
cur
ren
tly
working
towar
ds
his
P
h.
D
degr
ee
at
Dep
art
m
e
nt
of
El
ectroni
c
s
and
Comm
un
ic
a
ti
on
Engi
n
eering,
Charotar
Univer
sit
y
of
Scie
n
ce
and
Tec
hnolog
y
.
His
c
urre
nt
rese
arc
h
int
er
est
lies
in
wire
le
ss
sensor
net
works
espe
cially
in
op
ti
m
izat
ion
in
wire
l
ess sensor ne
tworks.
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
:
3
6
1
5
-
3
6
2
2
3622
Sarman
K.
H
adia
is
an
Associ
ate
Profess
or
with
the
D
epa
rtment
of
E
le
c
troni
cs
and
Com
m
unic
at
ion
Engi
nee
r
ing
at
C
S
Pate
l
Instit
ute
of
Techno
log
y
–
Charo
ta
r
Univer
sit
y
of
Scie
nc
e
and
T
e
chnol
og
y
,
Ch
an
ga,
Anand,
Guj
ara
t
,
India.
His
cur
ren
t
r
ese
ar
ch
int
ere
sts
ar
e
in
W
ire
l
ess
Com
muni
cation
S
y
s
tem
s,
Networki
ng
and
Mi
cro
elec
tr
onic
s.
He
has
p
ubli
shed
sev
eral
pape
rs
in
nat
ion
al
/i
n
te
rn
at
ion
al
conf
ere
n
ce
s
and
int
ern
a
ti
ona
l
jo
urna
ls.
He
received
Bac
he
lor
of
Engi
ne
eri
ng
d
eg
ree
in
E
lectr
oni
c
s
and
Com
m
unic
ation
Eng
ine
e
ri
ng
from
Bhavn
a
gar
Univer
si
t
y
,
India
in
1997
and
Master
of
Engi
nee
r
ing
degr
ee
i
n
El
ec
tron
ic
s
an
d
Comm
unic
at
io
n
Engi
nee
ring
with
Specializa
tion
of
Com
m
unic
ation
S
y
st
ems
Engi
ne
eri
ng
fro
m
Gujar
at
Univ
ersity
,
India
in
2008.
He
re
ce
iv
e
d
Ph.D.
d
egr
e
e
i
n
Elec
tron
ic
s
an
d
Com
m
un
ic
at
io
n
Engi
n
ee
ring
fr
om
Charusa
t
.
Evaluation Warning : The document was created with Spire.PDF for Python.