Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
Vol.
5, No. 6, Decem
ber
2015, pp. 1417~
1
423
I
S
SN
: 208
8-8
7
0
8
1
417
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
Perform
a
nce Comparis
on of Rou
t
ing P
r
ot
ocols in Bipartit
e
Wireless Sens
or Network
Devashis
h G
o
sain,
Itu
Sni
gdh
Departm
e
nt o
f
C
o
m
puter Scien
c
e, Bir
l
a
Institu
te o
f
Technolog
y
,
R
a
nchi
, Ind
i
a
Article Info
A
B
STRAC
T
Article histo
r
y:
Received Apr 12, 2015
Rev
i
sed
Ju
l 20
,
20
15
Accepted Aug 16, 2015
This pap
e
r ev
aluates
and r
a
nks the su
itab
ili
t
y
of rout
ing a
l
g
o
rithm
s
for
bipartite wirele
ss sensor networ
k topolog
y
.
Th
e network
consid
ered
in th
is
paper, consists o
f
an
irregu
lar
co
mbin
ation of f
i
x
e
d and mobile n
odes, which
leads to
constru
c
tion of
a bip
a
r
tite graph
am
ong them
. A wireless sensor
network is usually
constrain
e
d b
y
th
e en
erg
y
limitations and
processing
capab
ili
ties.
W
e
the
r
efore
,
co
nsider the
im
portant m
e
trics
for ana
l
y
s
is
nam
e
l
y
, c
a
rried
load, en
erg
y
co
ns
um
ption and the averag
e del
a
y incur
r
ed.
W
e
present
the
possibiliti
es of
e
m
plo
y
ing
the
ro
uting
algori
t
hm
s subjec
t to
the qu
ality
of
ser
v
ice required b
y
the
wir
e
less sens
or netw
orks applications.
Keyword:
A
v
er
ag
e
D
e
la
y
Bip
a
rtite
Carried
Loa
d
Ener
gy
C
ons
u
m
ed
Copyright ©
201
5 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
:
Deva
shi
s
h Gos
a
i
n
,
Depa
rt
m
e
nt
of
C
o
m
put
er Sci
e
nce,
Birla In
stitu
te
o
f
Techno
log
y
,
R
a
nchi
,
I
ndi
a
Em
a
il: g
o
s
ain
.
d
e
v
a
sh
ish
@
gmail.co
m
1.
INTRODUCTION
WSN’s
are
now all p
e
rv
asiv
e in
world
.
From
h
o
m
e
au
to
matio
n
system
to
critical b
o
iler
m
o
n
ito
rin
g
,
from
X-Box to
m
ilitary su
rve
illance, they are ubi
quitous
.
Recently, co
ns
idera
b
le am
ounts of re
searc
h
efforts
have e
n
a
b
led the actual im
ple
m
entation and
placem
e
n
t of se
ns
or networks tailore
d to the
uni
que
requ
irem
en
ts of certain sen
s
ing
an
d m
o
n
ito
ri
n
g
app
licatio
n
s
[1
].
An estab
lish
e
d well-fun
c
tion
i
n
g
sen
s
or
n
e
t
w
ork
h
a
s a fou
n
d
a
tion
,
bu
ilt on
, t
w
o fact
ors, a goo
d
co
mm
u
n
i
catio
n
p
r
o
t
oco
l
and
a r
o
b
u
st n
e
tw
or
k
top
o
l
ogy. Ex
ten
s
iv
e
rev
i
ew
of
ex
istin
g
co
mm
u
n
i
catio
n
protoc
ols ca
n
be found in [2],
[3] a
n
d for Se
nsor
placem
ent (or
network t
o
pol
ogy
) [4] ca
n
be
refe
rre
d.
Ob
ject
i
v
e
o
f
a
n
y
se
nso
r
t
o
po
l
ogy
i
s
t
o
i
n
cre
a
se t
h
e
co
vera
ge
wi
t
h
m
i
nim
i
zi
ng t
h
e c
o
st
.
To
i
n
c
r
ea
s
e
t
h
e co
vera
ge
v
a
ri
o
u
s sc
hem
e
s ha
ve bee
n
pr
op
ose
d
,
w
h
i
c
h
i
n
cl
ude
s bi
f
u
r
cat
ed net
w
o
r
k
of
fi
xe
d an
d
m
obi
l
e
no
des
,
l
i
k
e gri
d
de
pl
oy
m
e
nt
m
e
t
hod [
5
]
,
w
h
ere t
h
e
regi
o
n
i
s
di
vi
de
d i
n
t
o
gri
d
s. N
u
m
b
er
of
depl
oy
ed st
at
i
c
nodes
,
obstacle
s
and boundari
e
s
collectiv
el
y
deci
des
t
h
e
we
i
ght
of
eac
h
gri
d
.
The
g
r
i
d
wi
t
h
l
east
wei
g
ht
i
s
t
h
e
dest
i
n
at
i
o
n of
a
m
obi
l
e
sensor
. In [
6
]
Vo
r
o
n
o
i
p
o
l
y
go
n i
s
expl
oi
t
e
d t
o
fi
n
d
t
h
e num
ber of co
ve
rage
hol
es
(wi
t
h
t
h
ei
r
p
o
s
i
t
i
ons)
usi
n
g
a co
vera
ge e
n
hanci
n
g
al
g
o
ri
th
m
.
Si
milar to
th
is, in [7
]
au
tho
r
s pro
pose an
alg
o
rith
m
wh
ich
u
s
es
Delaun
ay trian
g
u
l
atio
n
to
d
e
term
in
e hol
es wi
t
h
t
h
e hel
p
of m
obi
l
e
and st
at
i
c
senso
r
no
des
.
Si
nce vari
e
g
at
ed net
w
or
k of
suc
h
ki
nd
ha
s m
a
ny
pract
i
cal
ap
pl
i
cat
i
ons;
we
f
o
c
u
s o
n
t
h
e
t
o
p
o
l
o
gy
wh
ich
is p
a
rtially stat
ic an
d
partially d
y
n
a
mic in
ou
r
research. Th
e wirel
e
ss sen
s
or n
e
t
w
orks th
at co
nstitu
te
m
obi
l
e
and fi
x
e
d se
ns
ors
us
u
a
l
l
y
co
m
p
rom
i
se bet
w
e
e
n c
o
st
and
co
ve
rag
e
. Al
s
o
, i
n
or
der t
o
ac
hi
eve
hi
g
h
cove
ra
ge, m
o
b
i
l
e
sensor
s m
a
y
requi
re t
o
m
ove
fr
om
dens
e areas t
o
sp
ar
se areas. T
h
e b
ackg
r
ou
n
d
st
re
ngt
h o
f
o
u
r an
alysis lies in
i
n
corp
oratin
g b
i
p
a
rtite
g
r
ap
hs in
t
o
su
ch
k
i
nd
o
f
top
o
l
og
y. Th
ere
h
a
s
b
een a l
o
t of
research
by
em
pl
oy
i
ng suc
h
g
r
ap
hs i
n
t
a
rget
t
r
acki
ng
appl
i
cat
i
o
ns. T
h
e net
w
o
r
k i
s
i
n
co
r
p
o
r
at
ed as
a set
Si
consi
s
t
of i
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE
Vol. 5, No. 6, D
ecem
ber
2015 :
1417 –
1423
1
418
sen
s
o
r
s an
d
set
Tj
co
nsist o
f
j targ
ets; with
th
e ob
j
ectiv
e to assig
n
sensors to
targ
ets o
p
t
i
m
all
y
(
m
ax
i
m
i
z
in
g
u
tility an
d
/
o
r
min
i
mizin
g
cost), su
bj
ect t
o
th
e im
p
o
s
ed
co
n
s
t
r
ain
t
s. Au
t
h
ors
v
i
ew th
is as a
b
i
p
a
rtite g
r
aph
,
whe
r
e an e
d
ge
e(Si, Tj
) corresponds to a
pairing
of se
ns
or Si with
targ
et Tj
with
th
e
ob
j
ective of assig
n
i
n
g
sen
s
o
r
s to
targets o
p
tim
a
lly (
m
ax
i
m
izin
g
u
tilit
y an
d
/
o
r
m
i
n
i
mizin
g
co
st), su
bj
ect to
th
e i
m
p
o
s
ed
con
s
train
t
s
[
8
]-[
10
].
Our work
is
differen
t
fro
m
t
h
e prev
i
o
u
s
wo
rk
,
b
ecau
s
e here we are p
a
rtitio
n
i
n
g
sen
s
ors in
to
t
w
o
di
sj
oi
nt
set
s
o
f
m
obi
l
e
and st
at
i
ona
ry
sen
s
o
r
s.
2.
MOTI
VATI
O
N
The m
o
t
i
v
at
i
on
behi
nd t
h
e i
d
ea i
s
t
h
at
, s
o
m
e
t
i
m
e
s i
t
becom
e
s im
possi
bl
e t
o
m
a
nual
l
y
depl
oy
sen
s
o
r
s
in
sites lik
e rou
g
h
m
o
un
tain
terrain
, d
e
n
s
e
fo
rests, cav
e,
battlefield
s
, and areas affected b
y
po
isono
us
g
a
ses.
The s
o
l
u
t
i
o
n i
s
scat
t
e
ri
ng
t
h
e
sens
ors
ra
nd
o
m
l
y
, wi
t
h
o
bvi
ous
d
r
a
w
bac
k
of
n
o
t
ha
vi
n
g
t
h
e
desi
re
d
pl
acem
e
nt
and c
o
vera
ge. In the recent
tim
e
s, researchers
ha
ve
enc
o
ura
g
ed
on m
i
xed se
ns
or
net
w
orks, in
whi
c
h the
stationary nodes and m
obile nodes
work
i
n
coherence to perf
orm
place
ment task. Such
placem
ents give
m
o
re
cove
ra
ge
a
n
d
r
o
b
u
st
ness wi
t
h
re
duce
d
n
u
m
b
er of n
ode
s.
Our article analyses this topology
construction in context to a bipar
tite graph. Essentially a bipartite
gra
p
h, al
so cal
l
e
d a bi
g
r
a
p
h
,
i
s
a set
of
gra
ph
ve
rt
i
ces de
com
posed i
n
t
o
t
w
o
di
sj
oi
nt
s
e
t
s
such t
h
at
n
o
t
w
o
gra
p
h ve
rtices within t
h
e sa
me set are adjacent [11]
. Our bipa
rtite
sensor network involves pa
rtitioning
sens
ors i
n
t
o
t
w
o di
s
j
oi
nt
set
s
of m
obi
l
e
an
d
st
at
i
onary
se
ns
ors
.
I
n
t
h
i
s
a
r
t
i
c
l
e
, we an
al
y
ze t
h
e pe
rf
orm
a
nce o
f
b
i
p
a
rtite n
e
twork
u
n
d
e
r three
classes o
f
rou
tin
g
algor
ith
m
s
,
n
a
m
e
l
y
DSR (o
n
d
e
m
a
n
d
)
[12
]
, OLSR (d
ist
a
n
ce
v
ector b
a
sed
, static) [13
]
,
an
d
FISHEYE (lin
k
state b
a
sed
)
[1
4
]
on
d
i
fferen
t
k
i
nd
s o
f
b
i
p
a
rtite sen
s
o
r
n
e
two
r
k
s
t
h
at
h
a
s
no
t b
e
en
ad
dressed
i
n
p
r
ev
iou
s
literatures till d
a
te.
3.
SYSTE
M
MO
DEL
We co
n
s
t
r
u
c
ted
a
b
i
p
a
rtite grap
h
B
(
V, E),
wh
ere
V =
S
⋃
M, where
S de
notes t
h
e set of static nodes
an
d M d
e
no
tes th
e set of m
o
bile n
o
d
e
s.
We
h
a
v
e
tak
e
n
th
e
d
i
stan
ce as
a co
st m
e
tric in
th
e con
s
tru
c
tio
n
o
f
t
h
e
b
i
p
a
rtite g
r
aph
.
3.
1. Ass
u
mp
ti
ons
1)
Sens
ors a
r
e l
o
cat
i
on awa
r
e ei
t
h
er o
b
t
a
i
n
ed
fr
om
Gl
obal
Posi
t
i
oni
ng Sy
st
em
(GPS)
or t
h
r
o
ug
h
l
o
cat
i
on di
sc
ov
ery
al
g
o
ri
t
h
m
s
.
2)
It
i
s
assum
e
d t
h
at
m
obi
l
e
sen
s
ors
are F
u
l
l
F
unct
i
o
nal
Devi
ce an
d St
at
i
ona
ry
Sens
o
r
s are
red
u
ce
d
fu
nct
i
o
nal
de
vi
ces.
3)
The m
obile/dynam
ic sensors
can easily m
ove and
ca
n
reac
h the
desi
red location efficiently and
accurately using the
m
obility
algo
rithm
for
dynamic sensors.
3
.
2
.
Mo
bility Algo
rithm
1)
For
eac
h m
obi
l
e
n
ode
2)
For
eac
h
st
at
i
c
sens
or n
o
d
e
3)
Calculate Euclidian
distance
from
itself
4)
Up
dat
e
no
de t
a
bl
e ent
r
y
5)
End F
o
r
6)
From
n
ode
t
a
bl
e sel
ect
t
h
e
no
de
wi
t
h
m
i
n. di
st
ance
7)
Move
towards
the node select
ed in step 5
8)
Latch
with
nod
e selected
in
step
5
an
d
t
o
all o
t
h
e
r static no
d
e
s wh
ich
are in
its co
mm
u
n
i
catio
n
ran
g
e
9)
Del
e
t
e
n
ode
e
n
t
r
i
e
s fr
om
i
t
s
node
t
a
bl
e
(o
f l
a
t
c
hed
n
ode
s f
r
o
m
st
ep 7)
10
)
Rep
eat step
6 to
9
u
n
til its nod
e tab
l
e
g
e
ts em
p
t
y
11
)
End F
o
r
12
)
Rep
eat step
s 1
to
11
m
u
ltip
le ti
m
e
s
3.
3. Step
s of C
o
ns
tructi
on
1)
Ad
d al
l
t
h
e
m
ovabl
e
n
ode
s i
n
t
o
set
M
.
2)
Add
all th
e statio
n
a
ry nod
es i
n
to
set S.
3)
∀
∈
and
∀
∈
i
f
m
obi
l
e
no
de m
can
re
ach st
at
i
c
no
de
s,
(t
he
di
st
a
n
c
e
bet
w
een
m
and
s
i
s
less than m
a
xim
u
m prescribe
d
distan
ce and the re
m
a
ining energy of
m
is above a certain thres
h
old E
T
) th
en
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISS
N
:
2088-8708
Perfo
r
man
ce C
o
mpa
r
ison
o
f
Rou
tin
g Pro
t
o
c
o
l
s in Bipa
rtite Wireless S
e
n
s
o
r
Netwo
r
k (Deva
sh
ish
Go
sa
in
)
1
419
ad
d
an
edg
e
e(m
,
s) in
to
th
e
b
i
p
a
rtite g
r
aph; th
e weig
h
t
of th
e ed
g
e
e(m
,
s) is re
presen
ted
as w(m
,
s),
den
o
t
es
t
h
e di
st
a
n
ce
be
t
w
een m
obi
l
e
no
de m
and
fi
x
e
d
no
de
s,
ot
he
rwi
s
e,
w
(
m
,
s)
= N
U
LL.
3.
4. M
o
del
Ge
nera
ted
Fig
u
re
1
.
An
i
n
itial d
e
p
l
o
y
m
e
n
t
of B
3,2
se
ns
ors in a
part
i
c
ul
a
r
fi
el
d
Fi
gu
re
2.
Fi
nal
de
pl
oy
m
e
nt
of
B
3,2
sens
ors
Fi
gu
re
1
de
pi
ct
s t
w
o cl
ass
e
s
of
se
nso
r
s;
st
at
i
c
and m
obi
l
e
w
h
ere
,
S
1
,
S2 a
n
d S
3
rep
r
esent
st
at
i
c
sens
ors
(n
o
n
e of t
h
em
i
s
i
n
range
of eac
h ot
her
,
com
m
uni
cat
i
on i
s
not
pos
si
bl
e). M
1
a
nd
M
2
rep
r
ese
n
t
m
obi
l
e
sens
ors
.
The dashe
d
arrows represe
n
t
the trajectories of M
1
and M2 (m
obile
sensors) at
a particular instance
of t
i
m
e. (Not
e
:
bot
h t
h
e se
n
s
ors a
r
e com
p
l
e
t
e
l
y
free t
o
m
ove arou
n
d
i
n
si
de t
h
e
peri
phe
ry
, t
h
ese a
r
r
o
ws
represen
t,
o
n
l
y
o
n
e
p
a
rticu
l
ar
p
o
s
sib
ility o
f
t
h
eir
d
i
rection
)
.
Fi
gu
re
2
de
pi
ct
s t
h
e sam
e
ne
t
w
o
r
k
b
u
t
aase
w
h
en
t
h
e
m
obi
l
e
n
o
d
e
m
o
v
e
s i
n
t
h
e c
o
m
m
uni
cat
i
on
rang
e of a p
a
rticu
l
ar static sen
s
or nod
e th
ereb
y in
itiati
n
g
d
a
ta tran
sfer. We see th
at M1
h
a
s m
o
v
e
d
alo
n
g
h
i
s
t
r
aject
o
r
y
an
d i
s
i
n
t
h
e t
r
ans
m
i
t
t
i
ng ran
g
e
of
bot
h S1 a
n
d S2
(re
prese
n
t
e
d by
sol
i
d
ar
ro
ws)
.
Al
s
o
,
M
2
has
m
oved al
o
ng
hi
s t
r
a
j
ect
ory
and i
s
i
n
t
h
e
t
r
ansm
i
t
t
i
ng ra
nge
o
f
S
3
(
r
e
p
rese
nt
ed
by
sol
i
d
a
r
r
o
ws
).
I
n
t
h
i
s
scenari
o
, S
1
,
S2 a
nd
S3 a
r
e
sensi
n
g t
h
ei
r
nei
g
hb
o
r
h
o
ods
. M
1
an
d M
2
com
e
s i
n
pro
x
i
m
i
t
y
wi
t
h
St
at
i
onar
y
sen
s
o
r
s an
d
collect th
e d
a
ta fro
m
th
e
m
. Late
r on
, th
ey will b
e
m
o
v
i
n
g
toward
s si
n
k
an
d
will tran
sfer t
h
e d
a
ta
(sen
sed b
y
t
h
em
se
lv
es and
collected
fro
m
st
atic sen
s
ors).
Sa
m
e
p
r
o
cess is
iterated
m
u
ltip
l
e
ti
m
e
s.
3.
5. Sn
apsh
o
t
of
Ac
tu
al
Si
m
u
l
a
ti
on of
B
3
,
2
Fig
u
re
3
shows th
at in
itially
m
o
b
ile sen
s
o
r
s 4
and
5
are
no
t in
rang
e of
an
y static sen
s
o
r
1
,
2
or
3
.
The
red
fl
a
g
s a
r
e ra
n
dom
way
poi
nt
s
whi
c
h i
ndi
cat
e t
h
e
nex
t
l
o
cat
i
o
n
o
f
m
obi
l
e
se
ns
ors
.
Fig
u
re
3
.
In
itial d
e
p
l
o
y
m
e
n
t
setu
p
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE
Vol. 5, No. 6, D
ecem
ber
2015 :
1417 –
1423
1
420
From
Fi
g
u
re
4
we
obs
er
ve t
h
a
t
fi
nal
l
y
, m
obi
l
e
n
ode
s
have
t
r
avel
l
e
d a
n
d
no
w t
h
ey
a
r
e i
n
r
a
nge
o
f
static n
o
d
e
s.
No
d
e
4 (m
o
b
ile)
is co
mm
u
n
i
cati
n
g with nod
es
1
an
d 2 (static). Sim
i
larly Nod
e
5
(m
o
b
ile) is
com
m
uni
cat
i
ng wi
t
h
no
de 3 (
s
t
a
t
i
c
)
Fi
gu
re
4.
A
f
t
e
r
m
ovem
e
nt
of
m
obi
l
e
nod
es
3.
6. N
o
t
a
ti
ons Used
Assem
b
l
y
of
S
1
,
S2
an
d M
1
i
s
nam
e
d as
B
2,1
,acro
n
y
m
o
f
Bi
p
a
rtite Graph
co
n
s
isting
t
w
o set o
f
nod
es
st
at
i
c
(S1 a
n
d
S2)
a
n
d
m
obi
l
e
(M
1
)
.
1)
Assem
b
l
y
of S3 a
nd M
2
i
s
nam
e
d as
B
1,1
,acron
ym
o
f
Bip
a
rtite Gr
aph
con
s
isting
two
set
of
no
des
st
at
i
c
(S
2)
an
d m
obi
l
e
(
M
2).
2)
Together these
assem
b
lies are known as
B
3,2
,acron
ym
o
f
Bip
a
rtite Gr
aph
co
nsistin
g
t
w
o set
of
no
des
st
at
i
c
(S
1,
S2
an
d
S3
) a
n
d
m
obi
l
e
(M
1
an
d M
2
).
3)
In
ge
ne
ral
,
B
m,n
represen
ts a Bi
p
a
rtite
grap
h of two
set of
nod
es, on
e
con
s
istin
g
o
f
m
f
i
x
e
d no
des
and
ot
her
co
nsi
s
t
i
ng
of
n
m
obi
l
e
no
des
.
4.
SIMULATION RESULT
Co
n
s
t
r
u
c
tion
of th
e v
a
ri
o
u
s
bip
a
rtite n
e
twork
scen
ar
io
s is i
m
p
l
e
m
en
ted
i
n
Qu
aln
e
t un
der th
e set o
f
gi
ve
n
e
xpe
ri
m
e
nt
s wi
t
h
f
o
l
l
o
wi
n
g
param
e
t
e
rs.
1)
B
m,1
grap
hs
w
h
i
c
h c
o
n
s
i
s
t
o
f
m
st
at
i
c
no
des
an
d
onl
y
1 m
ovi
n
g
no
de.
2)
B
m,2
gra
p
hs
w
h
i
c
h c
o
nsi
s
t
o
f
m
st
at
i
c
node
s
and
2
m
ovi
ng
no
de.
3)
B
m,3
grap
hs
w
h
i
c
h c
o
n
s
i
s
t
o
f
m
st
at
i
c
no
des
an
d
onl
y
3 m
ovi
n
g
no
de.
Table 1. Param
e
ter
Specifications
PARAMETERS
VALUES
Ar
ea
1500 x 1
500
m
Data Rate
2 Mbps
Radio T
y
pe
802.
11
b
Packet Reception
Model
PHY802.11b
Battery Model
Residual Life Esti
m
a
tor
En
erg
y
Mo
d
e
l
Mica Mo
tes
M
A
C Pr
opagation Delay
1 s
Application
Constant Bit Rate
Th
e m
a
in
in
terest lies in
ob
serv
i
n
g how the ro
u
ting
p
r
o
t
o
c
o
l
p
e
rfo
r
m
s
with
in
creasing
n
u
m
b
e
r
of
m
obile nodes
as they will re
qui
re extra power s
o
urce
for
m
o
tion.T
h
e three im
portant
param
e
ters of
WS
N,
C
a
rri
ed
Loa
d
,
Ene
r
gy
a
n
d
Del
a
y
ha
ve b
een
obse
r
ved
fo
r D
S
R
,
OL
SR
an
d F
I
S
H
EYE.
The
g
r
a
phs
are
an
alysed
for t
h
e resu
lts ob
tained
after sim
u
latio
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Perfo
r
man
ce C
o
mpa
r
ison
o
f
Rou
tin
g Pro
t
o
c
o
l
s in Bipa
rtite Wireless S
e
n
s
o
r
Netwo
r
k (Deva
sh
ish
Go
sa
in
)
1
421
4.1. Carried
L
o
ad
From
Fi
gure 5
(
a),
5(
b) an
d 5
(
c i
t
is
ev
id
en
t
th
at carried
lo
ad
in
th
e n
e
two
r
k
is
m
i
n
i
m
u
m fo
r DSR
fol
l
o
we
d
by
O
L
SR
an
d
FI
SH
EYE.
Pe
rf
orm
a
nce
of
D
S
R
c
a
n
be at
t
r
i
b
ut
e
d
t
o
i
t
s
ad
-h
oc
nat
u
re.
It
i
s
a
n
on
-
dem
a
nd pr
ot
o
c
ol
desi
g
n
e
d
t
o
m
i
nim
i
ze t
h
e ban
d
w
i
d
t
h
c
ons
um
ed by
cont
rol
pa
cket
s
i
n
ad h
o
c
wi
rel
e
s
s
net
w
or
ks
by
el
im
i
n
at
i
ng t
h
e p
e
ri
o
d
i
c
t
a
bl
e-u
pdat
e
m
e
ssa
g
e
s requ
ired
in
the tab
l
e-driv
en
ap
pro
ach
(lik
e
OLSR
and F
I
SHEYE) [15].
B1
1
B
2
1
B3
1
B
4
1
0
20
0
40
0
60
0
80
0
10
00
12
00
14
00
16
00
18
00
CA
RR
IE
D L
O
A
D
(
b
it
s
/
s
e
c
)
NO
DES
DSR
OL
S
R
FI
SH
E
Y
E
B3
2
B
4
2
B5
2
B
6
2
B8
2
85
0
90
0
95
0
100
0
105
0
110
0
115
0
120
0
125
0
130
0
135
0
140
0
145
0
150
0
155
0
160
0
165
0
CA
RR
I
E
D
L
O
A
D
(
b
i
t
s
/
se
c)
NO
D
E
S
DSR
OL
S
R
FI
S
H
EY
E
Fi
gu
re 5(a
)
.
B
m
,
1
gra
p
h
Fi
gu
re
5(
b
)
. B
m
,
2
gra
p
h
B
6
3B
7
3
B
8
3B
9
3
950
1
000
1
050
1
100
1
150
1
200
1
250
1
300
1
350
1
400
1
450
1
500
CAR
R
IE
D L
O
AD
(b
i
t
s/se
c)
NODE
S
DS
R
OL
S
R
F
I
SH
EY
E
Fi
gu
re 5(c
)
.
B
m,
3
gra
p
h
4.
2. E
n
erg
y
C
o
nsu
med
B1
1
B
2
1
B3
1
B
4
1
0.
000
0.
005
0.
010
0.
015
0.
020
0.
025
0.
030
0.
035
EN
ERGY (m
W
h
)
NO
D
E
S
DSR
OL
S
R
F
I
SH
EY
E
B
3
2
B
42
B5
2
B
6
2
B82
0.
01
0
0.
01
5
0.
02
0
0.
02
5
0.
03
0
0.
03
5
EN
ER
GY (
m
Wh)
NO
D
E
S
DS
R
OL
S
R
FI
SH
EY
E
Fi
gu
re 6(a
)
.
B
m
,
1
gra
p
h
Fi
gu
re 6(
b
)
.
B
m
,
2
gra
p
h
B6
3
B
7
3
B
8
3
B
9
3
0.
012
0.
014
0.
016
0.
018
0.
020
0.
022
0.
024
0.
026
EN
ER
GY (
m
Wh)
NO
DE
S
DS
R
OL
SR
FI
S
H
EY
E
Fi
gu
re 6(c
)
.
B
m,
3
gra
p
h
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE
Vol. 5, No. 6, D
ecem
ber
2015 :
1417 –
1423
1
422
On
obse
r
vi
n
g
Fi
gu
re 6
(
a)
, 6
(
b
)
an
d 6
(
c) i
t
i
s
cl
ear t
h
at
DSR
has m
i
nim
u
m
energy
r
e
qui
rem
e
nt
w
h
er
e
a
s
F
I
S
H
E
Y
E
h
a
s
ma
x
i
mu
m.
C
a
r
r
i
e
d
lo
ad
of a sen
s
or nod
e is d
i
rectly p
r
o
portio
n
a
l to
its en
erg
y
co
nsu
m
ed
. From
p
r
ev
iou
s
find
ing
s
it is estab
lish
e
d
th
at
for DSR rou
ting sch
e
m
e
carried
lo
ad
is m
i
n
i
m
u
m,
wh
ich
im
p
lies
en
erg
y
con
s
um
p
t
io
n
will also
b
e
m
i
n
i
m
u
m
fo
r th
e same. Si
n
ce,
FISHEYE and
OLSR are
proactive prot
ocols
,
signi
ficant am
ount
of energy will
be re
quire
d in
transm
itting and receiving
cont
rol
packets to m
a
i
n
tain th
e link state routing table by each node, this
leads
t
o
their high
energy requirem
ent.
4.
3. A
v
era
ge Del
a
y
Fi
gu
res
7(a
)
,
7
(
b
)
a
nd
7
(
c)
de
pi
ct
s t
h
at
D
S
R
ro
ut
i
n
g sc
hem
e
pr
o
duce
s
l
o
n
g
del
a
y
s
i
n
net
w
o
r
k
.
T
h
i
s
can be attribut
ed to the fact that
, in
lin
k
state ro
u
ting
algo
rith
m
s
(O
LSR and F
I
S
H
EYE), before transm
ission
of
dat
a
,
nei
g
h
b
o
r
t
a
bl
es o
f
a
l
l
t
h
e n
o
d
e
s a
r
e u
pdat
e
d at
t
h
e be
gi
n
n
i
n
g
on
l
y
. Next
h
o
p
s
e
l
ect
i
on t
a
kes
t
r
i
v
i
a
l
am
ount
of tim
e. But, i
n
DSR
every tim
e the data
packet i
s
receive
d
by t
h
e
node,
it is forwarde
d to the ne
xt
n
o
d
e
, if a rou
t
e to
the
d
e
stin
atio
n
is
kn
own
b
y
th
e
pr
esen
t
n
o
d
e
. Else, ro
ute d
i
scov
ery mech
an
ism
is in
itiated
by
t
h
at
n
ode
.
Thi
s
c
o
nsum
es co
nsi
d
era
b
l
e
am
ount
of
t
i
m
e. Thi
s
fact
o
r
ca
n
be at
t
r
i
b
ut
ed
t
o
DSR
’
s
p
o
o
r
perform
a
nce withrespectt
o
th
e a
v
er
a
g
e d
e
la
y
p
a
r
a
me
te
r
.
B
1
1
B21
B31
B
4
1
0
.
0010
0
.
0015
0
.
0020
0
.
0025
0
.
0030
0
.
0035
0
.
0040
NODES
DS
R
OL
S
R
FI
SHEY
E
AVE
R
AGE D
E
L
AY
(sec
ond
s
)
B32
B
42
B5
2
B
6
2
B82
0.
00
15
0.
00
20
0.
00
25
0.
00
30
0.
00
35
0.
00
40
0.
00
45
0.
00
50
0.
00
55
0.
00
60
0.
00
65
AVE
R
AGE D
E
L
AY
(sec
ond
s
)
NO
DE
S
DS
R
OL
S
R
FI
SH
EY
E
Fi
gu
re 7(a
)
.
B
m
,
3
gra
p
h
Fi
gu
re 7(
b
)
.
B
m
,
3
gra
p
h
B63
B
7
3
B
8
3
B
93
0.
00
15
0.
00
20
0.
00
25
0.
00
30
0.
00
35
0.
00
40
0.
00
45
AVERAG
E DELAY (s
econd
s
)
NO
DES
DSR
OL
S
R
F
I
SH
EY
E
Fi
gu
re 7(c
)
.
B
m,
3
gra
p
h
Tabl
e 2.
R
a
n
k
Tabl
e
PERF
ORMA
NC
E
PARAMETERS
RAN
K 1
RAN
K 2
RAN
K 3
Carried Load
(bits/sec)
DSR
OLSR
FISHE
Y
E
Energy Consu
m
ed
(
m
Wh)
DSR
OLSR
FISHE
Y
E
Average Del
a
y (se
c
)
FISHE
Y
E
OLSR
DSR
5. CO
N
C
L
U
S
I
ON
En
erg
y
and
d
e
lay are two
p
a
raph
ern
a
lia’s fo
r an
y
W
S
N. If th
e b
i
-p
artite n
e
twork
is estab
lish
e
d
in
d
i
fficu
lt-to-access terrain
s, wh
ere
h
u
m
an
in
terv
en
tio
n
is
in
feasi
b
le, to
pro
l
on
g
th
e
n
e
two
r
k
lifeti
m
e, DSR is
the
only option because of
least
en
e
r
gy re
qui
rem
e
nt.
Howeve
r, in surveillance applications a
n
d
disaster
m
a
nagem
e
nt
sy
st
em
, where del
i
v
ery
t
i
m
e
is of f
o
rem
o
st
i
m
port
a
nce, FI
SHE
Y
E/
O
L
SR
can be em
pl
oy
ed but
certain
ly DSR d
o
e
s no
t p
r
ove to
b
e
efficien
t. In
situ
ations wh
ere
W
S
N is restricted
to
b
e
o
p
e
rated
on
low
b
a
ndwid
th DSR rou
ting
sch
e
me is a b
e
tter
op
tio
n and
m
u
st be em
ployed for enforcing le
ast carrie
d
loa
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Perfo
r
man
ce C
o
mpa
r
ison
o
f
Rou
tin
g Pro
t
o
c
o
l
s in Bipa
rtite Wireless S
e
n
s
o
r
Netwo
r
k (Deva
sh
ish
Go
sa
in
)
1
423
REFERE
NC
ES
[1]
Ak
y
ild
iz, I
a
n F., and Mehmet Can Vu
ran, “Wireless sensor netwo
r
ks”,
Vol. 4
,
Joh
n
Wiley
& Sons, 2010.
[2]
De M
o
rais
Cord
eiro,
Carlos
, and
Dharm
a
P
r
akas
h Agrawal
,
“Ad hoc and
sensor n
e
tworks: th
eor
y
and app
lications
”,
2
nd
edition, World Scien
tific, 201
1.
[3]
Ro
y
e
r
,
Elizabeth M., and Ch
ai
-Keong Toh, "
A
review of cu
rrent routi
ng pr
otocols for ad h
o
c mobile wireless
networks",
Personal Communications, I
EEE
, V
o
l. 6
,
No
. 2
,
pp
.
46-55, 1999
.
[4]
P. Santi, “Topolog
y
Control in
Wireless Ad Ho
c and
Sensor Ne
twork”, John Wiley
& Sons, 2005
[5]
R. C. Luo and O
.
Chen, “Mobile
sens
or node deplo
y
ment and
as
y
n
chronous pow
er management fo
r wireless sensor
networks”,
IEEE Transactions on
Industrial
Electronics,
Vol. 59
,
No. 5, pp. 2377–
2385, 2012
.
[6]
A. Ghosh, “Estimating cover
a
ge
holes and enhancing cov
e
rage
in mixed sensor networks”,
In P
r
oceedings
of th
e
29th Annua
l IEEE International
Conference
on
Local Computer
Networks (
L
CN ’04)
,
pp. 68–76
,
2004.
[7]
Z. J
.
Zhang
,
J
.
S
.
F
u
, and H. C
.
Chao
, “
A
n ene
r
g
y
-
e
ffic
i
entmotion strateg
y
for
mobile
sensors in mixed wireless
sensor networks”,
In
ternational
Journal of Di
stributed S
e
nsor Networks
, Vol. 201
3, pp
. 12
, 2013
.
[8]
Rowaihy
,
Hosam,
et al
., "A sur
v
ey
of sensor selection sc
he
me
s in wire
le
ss se
nsor networks",
Defense and Secu
rity
Symposium. International So
ci
ety for Optics and
Photonics
, 2007
.
[9]
Deshpande, Amol, Samir Khul
ler, Azarakhsh
Malekian,
and
Mohammed To
ossi. "Energ
y
ef
ficient monitoring in
sensor networks",
Springer Algo
rithmica
, Vol. 5
9
, No. 1, pp. 94-
114, 2011
.
[10]
Khan, Usman A., and José MF Moura, "D
istrib
uted Kalm
an fil
t
ers in sensor
networks: Bipartite fusion graphs
",
Statistical Signal Processing, 200
7. SSP
'07
.
I
E
EE/SP 14th
Workshop on. IEEE,
20
07.
[11]
http:/
/m
athworld
.wolfram
.
c
o
m/
Bi
pa
rt
it
e
G
ra
ph.
h
tml
.
[12]
Johnson, D., Y.
Hu, and D. Maltz, “T
h
e
d
y
namic source rou
ting
protocol
(DSR) for mobile ad
hoc networks fo
r
IPv4”,
The
Inter
n
et Engineering Taskforce
, Vol.
260, pp
. 4728
, 2
007.
[13]
Jacquet
,
Philipp
e
, Paul Muhl
et
haler
,
Thom
as Clausen,
Anis
Laouit
i
, Am
ir
Qa
yyum
,
and
Lauren
t Vienno
t,
"Optimized link
state routing
pr
o
t
ocol for
ad hoc
networks",
In M
u
lti Topi
c Con
f
erence, 2001. IE
EE INMIC
2001
.
Technology for the 21st C
e
ntury.
Proceedings. IEEE In
ternationa
l,
pp
. 62-68
, 200
1.
[14]
P
e
i, Guang
y
u
,
M
a
rio Gerl
a,
an
d Ts
u-
Wei Ch
en, "Fishey
e
state routing: A
ro
uting sch
e
me fo
r ad ho
c wir
e
less
networks",
In C
o
mmunications,
2000. ICC 2000
. 2000 IE
EE International
Conference on,
Vol. 1,
pp. 70-74
, 2000
.
[15]
http:/
/en
.
wikiped
i
a.org
/
wi
ki/D
y
n
amic_Source_Ro
uting.
Evaluation Warning : The document was created with Spire.PDF for Python.