Indonesi
an
Journa
l
of El
ect
ri
ca
l E
ng
inee
r
ing
an
d
Comp
ut
er
Scie
nce
Vo
l.
1
4
,
No.
1
,
A
pr
il
201
9
, p
p.
455
~
461
IS
S
N: 25
02
-
4752, DO
I: 10
.11
591/ijeecs
.v1
4
.i
1
.pp
455
-
461
455
Journ
al h
om
e
page
:
http:
//
ia
es
core.c
om/j
ourn
als/i
ndex.
ph
p/ij
eecs
Collisi
on
-
aware
cooper
ative MA
C p
ro
t
ocol d
esig
n f
or
m
obile a
d
-
hoc ne
tworks
Y.
Neer
aja
,
V
.
Su
m
alath
a
Depa
rtment
of
E
le
c
troni
cs
and
C
om
m
unia
ti
on
En
gine
er
i
ng,
JN
TUA,
Anant
apur
,
AP
,
India
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
J
un
1
9
, 2
018
Re
vised
N
ov
10, 2
018
Accepte
d
J
an
22
, 2
01
9
Mobile
ad
ho
c
n
et
works
are
d
esigne
d
to
m
ai
nt
ai
n
comm
unic
at
ion
among
the
inde
pend
ent
no
des
without
a
serve
r
or
b
ase
st
at
ion
.
High
eff
i
ci
en
t
MA
C
protoc
ol
t
ake
s
a
m
aj
or
role
to
m
ai
nta
in
co
ll
isi
on
fre
e,
bandwi
dth
eff
icient
comm
unic
at
ion
among
the
netw
orke
d
nod
es.
Coll
isions
among
the
nodes
provide
considerabl
e
re
du
ct
ion
in
th
e
p
erf
or
m
anc
e
of
the
net
work.
The
obje
c
ti
ve
of
th
e
work
is
to
provide
col
l
isio
n
awa
re
coop
er
at
iv
e
MA
C
protoc
ol
d
esign
b
y
m
odif
y
ing
th
e
bac
kof
f
to
a
spec
if
ic
r
ange.
It
i
m
prove
s
the
throughpu
t
and
re
duc
es
de
lay
with
le
ss
co
ll
i
sion
ra
t
e.
It
m
ini
m
iz
es
th
e
complexi
t
y
of
th
e
probl
em a
risin
g
in
m
emor
y
and
proc
essing.
Ke
yw
or
d
s
:
MAC
MANET
s
Netw
ork
Rou
ti
ng
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
:
Y.
Nee
raj
a
,
Dep
a
rtm
ent o
f El
ect
ro
nics
and C
omm
inica
tio
n E
ng
i
neer
i
ng,
JNT
UA,
An
at
a
puram
u,
An
at
a
pur, A
P, India
.
Em
a
il
:
neer
aja
y30@gm
ai
l.com
1.
INTROD
U
CTION
A
Mo
bile
ad
-
hoc
net
work
(M
AN
E
T)
is
a
co
ll
ect
ion
of
a
utono
m
ou
s
m
ob
il
e
nodes
c
onne
ct
ed
to
each
oth
e
r
with
ou
t
dep
e
ndin
g
on
s
pec
ific
fixe
d
in
fr
ast
r
uctu
re
[
1].
Each
node
in
the
MANE
T
c
an
act
as
a
rou
te
r
to
process
a
nd
tra
ns
fe
r
the
pac
ke
ts
to
the
de
sti
na
ti
on
,
eac
h
node
consi
dered
as
a
source
node
or
destinat
io
n
node
or
i
nterm
ediate
nod
e
t
o
f
orw
ard
t
he
pac
ket
s
to
the
destin
at
ion
node.
T
he
fr
e
quent
c
ha
ng
e
s
of
nodes
le
ad
to
li
nk
breaka
ges
and
m
ay
fail
the
par
ti
al
or
f
ull
networ
k
com
m
un
ic
at
ion
[2
]
.
MANET
s
are
distribu
te
d
in
natu
re
and
fo
ll
ows
m
ulti
-
hop
c
omm
un
ic
at
io
n
to
tra
ns
fe
r
the
data
pac
kets
to
t
he
destinat
io
n.
E
ve
ry
two
m
ob
il
e
nodes
in the n
et
work
can co
m
m
un
ic
at
e each o
the
r
i
f
they
are
in sa
m
e rad
io c
onne
ct
ivit
y rang
e.
Less i
m
ple
m
entat
ion
cost,
dynam
ic
topolo
gical
ch
ang
e
s
f
or
c
onne
ct
ion
m
ai
ntenan
ce
a
nd
qu
ic
k
dep
l
oym
ent
are
m
ajo
r
ad
va
ntag
e
s
of MA
NETs.
1.1.
Pr
ob
le
m
Set
The
m
ajor
c
halle
ng
es
inclu
de
s
la
ck
of
in
fras
tructu
re,
m
ob
il
it
y
and
dynam
i
c
nat
ur
e
an
d
la
ck
of
fi
xe
d
traff
ic
c
on
ce
nt
rati
on
points,
wireless
c
omm
un
ic
at
ion
an
d
batte
ry
c
onsu
m
ption
,
le
ss
processi
ng
powe
r,
secur
it
y
are
the
m
aj
or
chall
eng
e
s
and
lot
of
researc
h
is
currently
happe
ning
in
these
aspects.
MAC
la
ye
r
is
on
e
of
the
s
ub
la
ye
r
of
data
li
nk
la
ye
r
us
e
d
to
m
anag
e
the
colli
sion
prob
le
m
s
in
the
wire
d
an
d
wir
el
ess
netw
orks.
Ge
ne
rall
y,
the
MA
C
is
us
e
d
f
or
f
r
a
m
ing
,
flo
w
co
ntr
ol,
acce
ss
c
on
t
ro
l
a
nd
phy
sic
al
addressin
g
with
su
pp
or
t
of
ot
he
r
la
ye
rs
in
the
ref
ere
nce
m
od
el
.
Po
int
co
ord
inate
d
f
un
ct
io
n
(P
CF)
an
d
dis
tribu
te
d
co
ordi
nated
functi
on
(D
C
F
)
are
tw
o
MA
C
la
ye
r
wo
r
kin
g
m
echan
ism
s.
IEEE
802.1
1
PCF
is
centr
al
iz
ed
m
echan
ism
and
works
only
with
infr
ast
r
uct
ur
e
netw
orks
and
IEEE
80
2.11
DCF
is
distrib
uted
a
nd
w
orks
with
ad
-
ho
c
netw
orks
a
nd
al
so
for
in
frast
ru
ct
ure
net
works.
Cha
nn
el
colli
sion
s
a
re
m
ajo
r
pro
bl
e
m
s
in
the
wireless
netw
orks
i
nclu
ding
Ad
-
H
oc
net
work
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
1
4
, N
o.
1
,
A
pr
il
201
9
:
455
–
461
456
Con
te
ntion
ba
sed
protoc
ols
us
e
d
to
res
olve
the
prob
le
m
s
in
c
on
te
ntio
n;
ho
wev
e
r,
hi
dd
e
n
a
nd
expose
d
te
rm
i
nal
pro
blem
s
can
arise
i
n
he
te
rogen
e
ous
netw
orks
li
ke
m
ob
il
e
ad
-
hoc
netw
orks.
T
he
form
al
bin
a
ry
ex
pone
nt
bac
kof
f
al
gorithm
is
unsta
ble
with
lo
ng
wait
ing
queu
es
unde
r
s
om
e
assum
pt
ion
s.
Th
e
oth
e
r
existi
ng con
ge
sti
on
c
ontrol m
echan
ism
s w
or
ks
with lar
ge n
um
ber
of
retra
ns
m
issi
on
s.
1.2.
B
ackgro
und
In
a
d
hoc
net
works
the
com
m
un
ic
at
ion
ta
kes
place
am
on
g
the
no
des
di
rectl
y
with
in
the
range
or
thr
ough
the
interm
ediat
e
no
de
s.
Var
i
ou
s
M
AC
issues
ad
dresses
relat
ed
to
Qu
al
it
y
iss
ues,
sec
ur
it
y
i
ssu
es
,
node
fail
ures,
r
ou
ti
ng
distu
rb
a
nces
has
be
en
pro
po
se
d
by
m
any
au
thors.
I
n
this
pa
per,
the
s
urvey
s
con
ce
ntrate
d
on
desig
n
iss
ue
s
of
MAC
tha
t
gr
eat
ly
af
fec
ts
the
perform
ance
of
MAC
an
d
prov
i
de
a
ne
w
schem
e
called
coope
rati
ve
se
m
i
distribu
te
d
MAC
routin
g
protoc
ol.
T
he
auth
or
s
[3
]
a
distrib
uted
M
AC
for
wireless
pe
r
s
onal
area
netw
orks
,
w
hich
co
nc
entrates
on
be
acon
c
olli
sions
and
prov
e
s
that
the
i
m
pr
op
er
slot
cho
ic
es
,
le
ad
s
to
f
reque
nt
a
nd
rep
eat
e
d
c
olli
sion
s.
A
sel
f
l
earn
i
ng
colli
sion
av
oid
a
nce
s
chem
e
[4
]
pro
po
s
ed
for
wireless
ne
tworks
,
it
has
z
ero
ove
rh
ea
d
in
com
m
un
ic
at
i
on
a
nd
co
ns
i
de
rab
le
gro
wth
in
the
perform
a
nce
of
the n
et
wor
k.
The
M
AC
QoS
[5
]
,
[
6]
real
tim
e
app
li
cat
ion
s
are
disc
usse
d,
s
uch
as
pr
i
or
it
y
reall
oc
at
ion
s
a
nd
sche
du
li
ng.
T
he
authors
[
7]
di
scusse
d
ab
ou
t
the
cl
us
te
rin
g
s
chem
es
and
m
ulti
cast
ing
r
ou
t
ing
pr
o
toc
ol.
It
us
es
weig
ht
cost
f
unct
ion
by
co
nsi
der
in
g
po
wer
of
node,
a
ver
a
ge
net
work
e
ne
rg
y
co
nsum
ption
an
d
the
cl
us
te
r
head
is
sel
ect
ed
ba
sed
on
m
i
nim
u
m
weigh
t
value.
AR
AMA
[
8]
ant
r
outi
ng
al
gorithm
pr
op
os
ed
,
it
co
nsi
sts
of
two
update
m
echan
ism
s
cal
led
f
orwa
rd
pac
kets
f
or
r
ou
ti
ng
in
form
at
ion
and
bac
kwar
d
pack
et
s
f
or
up
datin
g
the ro
uting dat
a at t
he
in
div
i
dual
no
des.
It m
ini
m
iz
es the cong
e
sti
on probl
e
m
in
the
netw
ork
a
nd ide
ntif
ie
s the
sh
ort
est
po
s
sib
le
p
at
h
a
nd
reduces t
he
e
nd to
end
delay
in
tr
ansm
issi
on
s.
Coll
aborati
ve
cachin
g
needs
to
be
im
ple
m
e
nted
i
n
a
d
ho
c
netw
orks
f
or
node
m
ob
il
it
y.
A
nt
col
ony
base
d
routin
g
schem
e
[9
]
us
ed
to
ide
ntify
the
shortest
pa
ths
am
on
g
the
routes
us
in
g
ph
e
r
om
on
e
trai
ls.
The
path
stren
gth
is
cal
culat
ed
ba
sed
on
the
nu
m
ber
of
a
nts
f
ollows
that
pat
h.
A
nt
c
ol
on
y
op
ti
m
iz
ation
pro
posed
[10]
from
the
fo
ra
ging
behavio
r
of
a
nt
colo
nies
an
d
it
is
an
op
ti
m
a
l
so
luti
on
to
t
he
va
rio
us
N
P
-
Ha
r
d
pro
blem
s.
Mult
iple
An
t
col
on
y
op
ti
m
iz
ati
on
s
[
11
]
propose
d
f
or
a
d
hoc
r
outi
ng.
It
is
represente
d
the
ph
e
r
om
on
e
wit
h
var
i
ou
s
col
ors,
it
is
possi
ble
to
al
lo
w
form
i
ng
bette
r
pat
hs
dynam
ic
a
ll
y.
An
exte
ns
ive
s
tud
y
[12
]
-
[
17]
car
ried ou
t
on
IEEE
802.11
f
or
im
pro
ving the
p
e
r
form
ance o
f
th
e n
et
w
ork
.
The
rem
ai
nd
er
of
this
pa
per
is
or
ga
nized
as
fo
ll
ows.
Se
ct
ion
s
2
ha
ve
li
te
ratur
e
rev
i
ew
f
or
th
e
perform
ance
and
r
ou
ti
ng
asp
ect
s
that
m
otivate
d
to
t
he
des
ign
of
the
ne
w
routin
g
protoc
ol.
I
n
sect
ion
3,
we
pro
po
se
d
a
m
od
el
f
or
new
MAC
prot
oco
l.
S
i
m
ulati
on
res
ults
and
discu
ssion
s
are
a
ddres
sed
in
t
he
sect
ion
4.
Finall
y conc
lu
s
ion
s
are d
rawn
in secti
on
5.
1.3
.
O
bj
ec
tiv
es
The pa
per
obje
ct
ives
are
f
ollo
wing as
:
a.
To
im
pr
ove
th
e
netw
ork
li
fe
tim
e
and
to
e
lim
inate
the
possible
c
olli
sion
s
f
or
m
ulti
h
op
m
ob
il
e
ad
hoc
netw
orks.
b.
To w
ork wit
h
s
ensin
g
a
nd
bac
koff m
echan
ism
d
ur
in
g
the
c
o
ope
rati
ve beh
avio
ur.
c.
To
pro
vid
e
hi
gh
ef
fici
ent
MAC
protoc
ol
that
ta
kes
a
m
ajor
r
ole
to
m
ai
ntain
colli
sion
fr
ee
,
band
widt
h
eff
ic
ie
nt c
omm
un
ic
at
io
n
am
on
g t
he netw
ork
ed nodes
.
d.
To
pro
vid
e
c
ol
li
sion
awa
re
coope
rati
ve
M
AC
protoc
ol
de
sign
by
m
od
i
fyi
ng
the
bac
kof
f
to
a
s
pecif
ic
range.
e.
To
im
pr
ov
e
t
he
throu
ghput
and
reduces
del
ay
with
le
ss
co
ll
isi
on
rate
an
d
to
m
ini
m
iz
e
t
he
com
plexity
of
the pr
ob
le
m
arising in m
e
m
or
y and p
r
ocessi
ng.
2.
METHO
D
2.1
.
The
P
ro
po
sed
M
ac
Pr
otoc
ol D
esi
gn
To
im
pr
ov
e
t
he
netw
ork
li
f
e
tim
e
and
to
el
i
m
inate
the
po
ssi
ble
colli
sion
s
,
we
pres
ent
a
novel
CC
-
MAC
prot
oco
l
f
or
m
ulti
hop
m
ob
il
e
ad
ho
c
netw
orks
.
Wh
e
n
the
re
is
a
coopera
ti
ve
beh
a
vior
e
xists,
there
is
a
po
ssi
bili
ty
for
lo
w
rate
of
colli
sion
s
a
nd
higher
n
et
w
ork
eff
ic
ie
ncy
in
t
erm
s
of
netw
ork
li
fe
tim
e,
and
le
ss
ov
e
r
head. T
he pr
opos
e
d p
ro
t
oc
ol wo
rk
s
w
it
h seve
ral stages
li
ke
sensi
ng and
backo
ff
m
echan
ism
.
The
ba
sic
oper
at
ion
pr
opos
e
d
MAC
schem
e
is
fo
ll
owin
g
th
e
o
pe
rati
on
of
IEEE
802.1
1
DCF
[1
8
]
.
It
fo
ll
ows
the
sa
m
e
han
dsha
king
pr
ocedur
e
a
s
RTS,
CTS,
D
ATA
t
ransm
is
sion,
an
d
ac
know
le
dg
e
m
ent
(
ACK
)
as
descr
i
be
d
in
the
F
ig
ur
e
1.
Af
te
r
getti
ng
t
he
co
ntr
ol
pac
kets,
the
NAV
is
to
i
m
ply
th
e
tim
e
per
iod
durin
g
bu
sy
cha
nnel
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Coll
isi
on
-
awar
e co
op
e
ra
ti
ve
MAC
p
r
oto
c
ol
desig
n
f
or
mob
il
e a
d
-
hoc
net
works
(
Y.
Neer
aja
)
457
Alg
o
rith
m
1: MB
E
B Sch
e
m
e
Inp
u
ts: Mod
e,
n
,w
Ou
tp
u
t: Back
o
ff
cou
n
ter
at sp
ecif
ic stag
e
If
last tr
an
s
m
iss
io
n
was su
ccess
=
Bi =
Ran
d
o
m
(
C
W
)
* (Slot ti
m
e
)
Else if
collis
io
n
occu
rs
Bi =
Ran
d
o
m
(
C
W
)
* 2
(Slot ti
m
e)
Else
=
min
(
2
−
1
,
)
Back
o
ff
cou
n
ter
=
(
0
,
−
1
)
Sens
i
ng
:
Se
ns
i
ng
is
a
proce
ss
of
ta
king
a
ppr
ov
al
from
the
destinat
io
n
no
de
or
from
the
nex
t
hop
i
n
the co
m
m
un
ic
at
ion
. T
he
proc
e
ss takes
RTS a
nd CTS t
o
c
omm
un
ic
at
e w
it
h t
he
ot
her n
odes
.
Ba
ckoff
m
echan
ism
:
If
cha
nnel
is
identifie
d
as
idle
f
or
a
per
i
od
of
(D
ist
rib
uted
I
nter
fe
ren
ce
s
pace)
DI
F
S.
=
×
=
(
)
×
(1)
In
the
a
bove
e
qu
at
io
n
Co
nte
ntion
window
le
ng
t
h
is
cal
cul
at
ed
as
2
.
.
T
he
Ma
rkov
backo
ff
at
sen
der
m
od
e
is
sam
e
as
IEEE
802.1
1
bu
t
it
va
ries
at
the
rec
ei
ver
m
od
e
i
n
r
el
at
ion
with
th
e
Ma
rko
v
c
hai
n.
We
assum
ed
each
pack
et
ha
ve
eq
ual
pr
i
or
it
y
and
colli
sion
s
m
a
y
occu
r
at
the
sam
e
pr
obabili
ty
by
no
t
con
si
deri
ng
their
con
te
ntio
n
wind
ow.
Ra
ndom
con
te
ntion
wind
ow
from
(
1
)
determ
ined
by
a
rando
m
integer
is
fr
om
un
i
form
d
ist
ribu
ti
on, th
at
is in th
e ran
ge of
0 t
o
con
te
ntion
window m
axi
m
u
m
an
d
final b
ack
off
tim
e c
ou
l
d
be
the s
umm
at
ion
of Bi an
d p.
Stea
dy
sta
te
of
m
ark
ov
c
hain
sta
te
s
that
on
e
no
de
sta
ys
in
eac
h
sta
t
e
can
be
cal
c
ulate
d,
t
he
pro
bab
il
it
y t
hat a no
de
tra
ns
m
i
ts at
the
n
ext
slot is
giv
e
n
as:
=
∑
0
0
,
(2)
=
2
(
2
−
1
)
(
2
−
1
)
(
)
+
(
1
−
)
(
1
−
(
2
−
2
)
(3)
Fig
ure
1.
IEE
E
802.11
DCF
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
1
4
, N
o.
1
,
A
pr
il
201
9
:
455
–
461
458
Alg
o
rith
m
2.
MA
C at Send
er
Inp
u
t: HOL
f
ra
m
e
trans
m
iss
io
n
Ou
tp
u
t: Succes
sf
u
l
tr
an
s
m
iss
io
n
Tr
an
s
m
iss
io
n
bu
f
fer
>1
Sen
se th
e chan
n
el
f
o
r
id
len
ess
If
DI
FS=id
le
ti
m
e
Set the b
acko
f
f
co
u
n
ter
to
M
BEB
in
sen
d
er
m
o
d
e
If
chan
n
el is
id
le c
o
n
tin
u
o
u
sly
Co
u
n
t do
wn
the co
u
n
ter
v
alu
e
If
cou
n
td
o
wn
ti
m
e
r=0
Sen
d
the RTS con
t
rol p
acket
//co
m
p
l
ete han
d
sh
ak
in
g
app
roach
If
CTS
n
o
t r
ec
eiv
e
d
with in
+
+
Retran
s
m
it st
arts,
r
ep
eat the pro
cess
.
Else
Af
ter
2
S
IFS
the
so
u
rce
in
itiate
s co
o
p
erative tr
an
s
m
i
ss
io
n
.
W
ait f
o
r
A
CK/NAC
K
The so
u
rce wo
u
ld
perf
o
r
m
M
BEB
Res
et the cou
n
ter
Els
e
F
reez
e the b
acko
ff
.
In
t
he
al
gorith
m
2
the
sen
de
r
senses
t
he
ch
ann
el
for
tra
nsm
issi
on
of
dat
a,
If
t
he
m
edium
fo
un
d
f
ree,
it
perform
s
the
h
andsha
king
pr
ocess
an
d
proc
eed
with
the
ori
gin
al
data
pa
cket
transf
e
r.
I
f
idle
tim
e
rea
ches
t
o
the
DI
F
S
per
i
od
of
ti
m
e,
it
wait
s
for
MB
EB
bac
kof
f
ti
m
e
and
proceed
w
it
h
the
c
omm
u
nicat
ion
.
I
f
CT
S
not
receive
d
f
r
om
the
recei
ver,
re
transm
issi
on
at
tem
pts
sh
ould
m
ade
after
a
s
pecific
ti
m
e
in
te
rv
al
.
Finall
y
rese
t
the
co
un
te
r
aft
er
operati
on
c
om
ple
te
s
oth
er
wise
fr
eez
e
the
backoff.
T
he
al
gorithm
3
rep
rese
nts
the
operati
on
at
the r
ecei
ver
side that is
bas
ed on t
he
cl
ea
r t
o
se
n
d (CTS
).
Alg
o
rith
m
3: M
A
C at
Rec
eiv
er
Inp
u
t: Receiv
ed
or ov
erhead
f
ra
m
e
Ou
tp
u
t: ACK or N
ACK
If
RTS
r
eceived
Sen
d
CTS
co
n
trol
p
acket af
ter
SIF
S
If
collis
io
n
occu
rs
Perf
o
r
m
MBEB
at
th
e Receiver
m
o
d
e
Bro
ad
cast NACK
Else
Tr
an
s
m
it
ACK
In
t
he
ab
ove
pr
oces
s,
eac
h
us
e
r
has
data
pac
ke
ts
and
nee
ds
to
tra
ns
m
i
t
at
diff
e
ren
t
ra
ndom
tim
e
intervals.
Be
fore
tra
ns
m
i
tt
ing
the d
at
a, each
node
m
us
t
hav
e
to
c
hec
k
the
sta
tus of
th
e
cha
nn
el
f
or
i
dleness
.
T
he
m
arko
v
chain
is
u
se
d f
or the
bac
koff
window size
1
−
w
it
h
a
seq
ue
nc
e cal
culat
ion o
f
Ma
rkov chai
n of bac
kof
f
=
{
0
⁄
,
…
,
1
−
⁄
,
…
,
1
−
⁄
}
(4)
Nodes
ch
eck
t
heir
in
div
id
ual
backoff
tim
ers
and
ge
ne
rate
m
ark
ov
chain
backoff
as
lon
g
as
distri
bute
d
interfe
ren
ce
ra
ng
e
.
C
olli
sions
are
possi
ble
if
t
wo
or
m
or
e
no
des
c
ho
os
es
t
he
s
am
e
bac
kof
f
ti
m
er
,
it
is
el
i
m
inate
d
us
i
ng
c
oope
rati
ve
beh
a
vior
am
on
g
the
nodes.
Eac
h
no
de
in
the
ne
twork
m
ai
ntain
s
a
retransm
issi
on
value
for
the
ind
ivi
du
al
pac
ket.
T
he
retr
an
sm
issi
on
val
ue
is
increm
ented
eve
ry
tim
e
a
new
retransm
issi
on
occurs.
I
f
the
r
et
ran
sm
issi
on
counter
excee
ds
2,
it
is
due
t
o
co
ng
e
sti
on
in
the
net
wor
k.
A
t
this
sit
uation,
the
nodes
ide
ntify
the
c
urre
ntly
al
locat
ing
ca
pa
ci
ty
of
t
he
c
hann
el
a
nd
a
djust
acco
r
ding
to
t
hat
window.
3.
RESU
LT
S
AND DI
SCUS
S
ION
S
The
sim
ulati
on
s
perform
ed
usi
ng
ns
-
2
to
e
va
luate
the
perf
or
m
ance
of
t
he
propose
d
al
gorith
m
.
The
par
am
et
ers
des
cribe
d
in
T
a
bl
e
1
based
on
I
EEE
80
2.11
c
onfig
ur
at
io
n.
Nodes
a
re
scat
te
red
in
100
m
x
10
0m
area
with
15
0m
cov
erag
e
ra
diu
s
.
Ns
-
2
is
a
n
op
e
n
s
ource
si
m
ulator
w
hic
h
w
orks
ef
fecti
vely
to
pe
rform
the
si
m
ulati
on
s li
ke
creati
ng
netw
orks, cha
ngin
g node
be
hav
i
or
s
w
it
h
c
us
t
om
p
aram
et
ers.
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Coll
isi
on
-
awar
e co
op
e
ra
ti
ve
MAC
p
r
oto
c
ol
desig
n
f
or
mob
il
e a
d
-
hoc
net
works
(
Y.
Neer
aja
)
459
Table
1.
E
xper
i
m
ental
se
tup
Para
m
eters
Valu
e
Slo
t
Interval
2
0
m
i
cro sec
SIFS
1
0
m
i
cro sec
DIFS
5
0
m
i
cro sec
Slo
t I
n
terval
2
0
m
i
cro sec
Ch
an
n
el Bit Rate
2
M
b
p
s
Prea
m
b
le
leng
th
1
4
4
bits
PCLP head
er
4
8
bits
Pack
et leng
th
5
1
2
KB
CW
m
in
32
CW
m
ax
2048
Bit Rate
CBR
Ty
p
e
UDP
Coll
isi
on
Ra
te
is
a
rati
o
of
pac
kets
colli
ded
an
d
t
otal
num
ber
of
pa
ckets
recei
ve
d
s
uccess
fu
ll
y
at
destinat
io
ns
.
T
he
lo
wer
the
c
olli
sion
rate,
le
sser
the
c
olli
sion
s
,
the
sim
ulati
on
to
ok
130
seconds
t
o
cal
culat
e
the
c
olli
sion
r
at
e
and
are
s
how
n
in
F
i
gure
2.
W
e
t
ook
the
SIFS
an
d
DI
F
S
slot
inte
rv
al
s
as
10
µse
c
an
d
50
µse
c re
sp
ect
ively
.
Figure
2. Coll
isi
on
rate
vs
act
ive no
de nu
m
ber
Thro
ughput
is
cal
culat
ed
with
150
sec
onds
interval
with
resp
ect
to
the
channel
rate.
DCF
pe
rfor
m
ance
highly
aff
ect
s
if c
on
te
nd
i
ng stat
ion
s a
re m
ore.
T
he norm
al
i
zed thr
oughput
w
as
sho
wn in F
igure
3.
Fig
ure
3. N
orm
al
iz
ed
throu
g
hput
Figure
4
dem
on
stra
te
s
t
he
t
otal
offer
e
d
loa
d
agai
ns
t
c
on
tr
ol
over
head
c
om
par
ison
am
on
g
the
pr
oto
c
ol
s
su
c
h
as
tra
diti
on
al
IEE
E
802.1
1
with
D
UCHA,
IMCM
P
[
19
]
.
T
he
resu
lt
s
obta
ine
d
t
hroug
h
t
he
sim
ulati
on
0
10
20
30
40
50
60
70
80
0
10
20
30
40
50
60
70
80
90
1
0
0
Coll
isi
o
n
Rate (%)
A
c
tive N
o
d
e Numb
er
DCF
I
MCM
P
CCM
AC
0
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
0
.
9
0
10
20
30
40
50
60
70
80
90
1
0
0
No
rmali
zed
th
rough
p
u
t
A
c
tive node
n
u
mb
er
DCF
I
MCM
P
CCM
AC
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2502
-
4752
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci,
Vo
l.
1
4
, N
o.
1
,
A
pr
il
201
9
:
455
–
461
460
giv
e
n
fa
r
bette
r
res
ults,
D
U
CHA
is
pro
vi
ding
go
od
perform
ance
than
oth
er
s
.
Wh
e
r
eas
80
2.1
1
ha
ve
hi
gh
con
t
ro
l
over
he
ad
[20
]
,
[
21]
a
m
on
g
al
l com
par
ed
pro
t
oco
ls
.
Fig
ure
4. Total
off
e
red loa
d v
s
co
ntr
ol ove
rhead
4.
CONCL
US
I
O
N
In
this
pa
pe
r,
we
ha
ve
pro
pose
d
a
novel
colli
sion
awa
re
schem
e
wh
ic
h
fo
ll
ow
s
the
determ
inist
i
c
backo
ff
pe
rio
d
to
m
ini
m
iz
e
the
colli
sion
s
an
d
i
m
pr
ove
the
li
fetim
e
of
the
MANET
node
s
throu
gh
c
oope
rati
ve
retransm
issi
on
be
hav
i
or
.
Th
e
com
par
iso
n
re
su
lt
s
with
t
he
existi
ng
prot
oc
ols
gi
ven
sat
isfact
or
y
res
ults
in
the
colli
sion
rate,
thr
ough
pu
t
an
d
with
le
ss
ov
er
he
ad
.
It
shows
hig
h
im
pact
on
the
i
m
pr
ove
m
ent
of
the
ne
twork
li
fe ti
m
e b
y
m
i
nim
iz
ing
c
olli
sion
s
u
si
ng coo
per
at
ive
b
e
ha
vio
r
.
REFERE
NCE
S
[1]
Hongm
ei
,
D.,
W
ei
,
L
.
,
and
A
gra
wal,
D
.
P.,
Univer
sit
y
of
C
inc
inn
at
i
,
"Rout
ing
Secur
ity
in
W
ire
le
ss
Ad
Hoc
Networks",
IEEE
Comm
unic
at
io
n
magaz
ine
-
Tel
ec
omm
unic
at
ion
net
work
sec
uri
t
y,
O
ct
2002
.
[2]
Raz
z
aque,
M.
A.,
Sim
on,
D.,
and
Padd
y
,
N.
,
“
Cross
la
y
er
self
routi
ng
:
a
s
el
f
-
m
ana
ged
rou
ti
ng
appr
o
ac
h
f
or
MA
NETs”
,
In
proce
edi
ngs
of
the
4th
IE
EE
I
nte
rnational
Co
nfe
renc
e
on
Wi
rele
ss
and
Mobil
e
Computing
,
Net
working
and
Comm
unic
ati
ons,
pp
284
-
290
,
2
008.
[3]
Vishnevsk
y
,
V
.
M.,
e
t
a
l.
"S
t
ud
y
of
b
eaconi
ng
in
m
ult
ihop
wire
le
ss
PA
N
with
distri
bu
ted
cont
rol"
,
Mob
il
e
Computing,
I
EEE
Tr
ansacti
ons
on
7.
1
,
pp
.
113
-
126,
2008
.
[4]
Chen,
C
.
C
.
,
e
t
a
l.
"S
el
f
-
Learni
ng
Coll
ision
Avoid
anc
e
for
W
ir
el
es
s Net
works
",
IN
FOCOM
.
2006.
[5]
Li
,
M.
,
&
Balak
rishnan,
P.
,
"M
a
c
lay
e
r
admiss
io
n
cont
rol
and
pri
ority
re
-
al
lo
catio
n
for
handl
ing
q
os
guar
ant
e
es
in
non
-
coope
ra
ti
ve
wire
le
ss
l
ans",
Mobil
e
net
works
and
appl
ic
at
ion
s
10.
6,
pp
.
947
-
9
59,
2005
.
[6]
Tha
ch
a
y
ani,
M,
&
Nakke
era
n
,
R.
,
“
QO
S
Provisi
oning
through
Cla
ss
-
Based
Route
Sele
ction
for
OBS
Networks”,
Inte
rnational
Jo
urnal
of El
e
ct
ro
nic
s and
Comm
u
nic
ati
on
Eng
ine
e
ring
,
pp.
209
-
21
9,
2010
.
[7]
Hw
ang,
I.
S.,
a
nd
W
en,
H.
P.,
"Ene
rg
y
eff
ic
i
en
t
cl
uste
ring
t
ec
h
nique
for
m
ult
i
c
ast
routi
ng
pro
t
ocol
in
wir
el
ess
adhoc
n
et
works
",
IJ
CSNS
7.
8
,
pp
.
74
-
81
,
2007
.
[8]
Hus
sein,
O.,
Saada
wi,
T
.
,
“
Ant
Routi
ng
Algorit
hm
for
Mobile
Ad
hoc
Networks
(AR
AMA
)”
,
Inte
rnation
al
Pe
rform
ance
Co
m
puti
ng
and
Co
mm
unic
ati
ons C
onfe
renc
e
,
pp.
1
5
–
17,
2004
.
[9]
Joon,
W
.
,
B
y
ou
ng,
S.
C.
,
K
y
o
ung,
T.
P.
L
.,
and
Ju
,
J.,
“
Com
par
ison
bet
wee
n
Heterogen
eous
Ant
Colo
n
y
Optimiza
ti
o
n
Algo
rit
hm
and
Gene
tic
Algorit
hm
for
Global
Path
Planni
ng
of
Mobil
e
Robot”
,
In
IEEE
Inte
rnational
Symposium
on
I
ndustrial
Elec
tronic
s
,
pp
.
881
-
886
,
2011
.
[10]
Dorigo,
M.,
B
i
ra
ttari,
M
.
,
and
Stutz
le,
T
.
,
“
Ant
C
olon
y
O
pti
m
iz
ation:
Art
ifi
cial
Ants
as
a
Com
puta
ti
on
a
l
Inte
lligen
ce T
e
c
hnique
”
,
I
EE
E
C
omputati
onal
Int
el
li
g
ence
Magaz
ine
,
vol
.
1
,
no
.
4,
28
–
39,
2006
.
[11]
Kw
ang,
M.
S.,
and
W
eng,
H,
S.,
“
Ant
Colon
y
Optimiza
ti
o
n
fo
r
Routi
ng
and
L
oad
-
balanc
ing
:
Surve
y
and
N
e
w
Dire
ctions”,
IEEE
Tr
ansacti
ons
on
Syste
ms
,
Man
and
Cyb
erne
t
ic
s
,
vol
.
33,
no.
5,
p
p.
560
–
572,
200
3.
[12]
Ti
nnir
el
lo
,
I.
,
a
nd
Bia
nchi,
G.,
“
Ret
hin
king
the
IEE
E
802.
11
e
EDCA
per
form
anc
e
m
odel
in
g
m
et
hodolog
y
”
,
IEE
E
/A
CM
Tr
an
s.
Net
w
.
,
vo
l. 18
,
no.
2
,
pp
.
540
–
5
53,
Apr.
2010.
[13]
Cal
i
,
F.,
Conti
,
M.,
Grego
ri,
E.
,
“
IEE
E
802.
11
Protocol
:
Design
and
per
fo
rm
anc
e
ev
al
u
at
i
on
of
an
ada
p
tive
bac
koffm
ec
h
a
-
ni
sm
”,
IEEE
J
.
Sel.
Areas
Comm
un
.
,
vol. 18, no. 9,
pp.
1774
–
1786
,
Sep.
2000
.
0
0
.
1
0
.
2
0
.
3
0
.
4
0
.
5
0
.
6
0
.
7
0
.
8
0
.
9
1
0
0
.
3
0
.
6
0
.
9
1
.
2
1
.
5
1
.
8
2
.
1
2
.
4
No
rmali
zed
Cont
rol
o
verh
ea
d
Offered
Loa
d
in M
b
p
s
I
EEE 8
0
2
.
1
1
DU
CHA
I
MCM
P
CCM
AC
Evaluation Warning : The document was created with Spire.PDF for Python.
Ind
on
esi
a
n
J
E
le
c Eng &
Co
m
p
Sci
IS
S
N:
25
02
-
4752
Coll
isi
on
-
awar
e co
op
e
ra
ti
ve
MAC
p
r
oto
c
ol
desig
n
f
or
mob
il
e a
d
-
hoc
net
works
(
Y.
Neer
aja
)
461
[14]
Krishna,
P.
V.,
Misra,
S.,
Oba
i
dat
,
M.
S.
,
Sari
tha
,
V.
,
“
Virtual
bac
koff
al
gori
t
hm
:
An
enha
nc
ement
to
802.
1
1
m
edi
um
-
ac
ce
ss
cont
rol
to
improve
the
p
erf
orm
a
nce
of
wire
le
ss
net
works
,
”
IE
EE
Tr
ans.
Ve
h.
Tec
hnol
.
,
vo
l.
59
,
n
o.
3,
pp
.
1068
–
107
5,
2010
.
[15]
Tol
edo
,
A.
,
Vercaute
r
en,
T.,
W
an
g,
X.,
“
Adapti
ve
opti
m
iz
a
ti
on
of
IEE
E
802
.
11
D
CF
base
d
on
Ba
y
esi
an
est
imati
o
n
of
the num
ber
of
competi
ng
t
erminal
s”
,
I
EE
E
Tr
ans.
Mobile
Com
put
.
,
vol
.
5
,
no
.
9
,
pp.
1283
–
12
96,
Sep.
2006
.
[16]
Deng,
D.,
Ke,
C.
,
Chen,
H.,
&
H
uang,
Y.,
“
Contention
window
opti
m
iz
ation
for
IE
EE
802.
11
DCF
ac
c
ess
cont
rol”,
IEE
E
Tr
ans.Wi
rele
ss
Comm
un
.
,
v
ol.
7
,
no
.
12
,
pp
.
5129
–
5135,
De
c
.
2008
.
[17]
Tri
ved
i,
R.
,
“
Covera
ge
and
Con
nec
t
ivi
t
y
Iss
ue
i
n
W
ire
le
ss
Sensor
Networks”,
I
ndonesian
Journal
of
Elec
tri
ca
l
Engi
ne
ering
and
Informatic
s (
IJ
EE
I)
,
vol
.
1
,
no
.
2,
pp
.
49
-
52
,
20
13.
[18]
W
ire
le
ss
LAN
Medium Acc
ess
Control
(MA
C)
and
Ph
y
si
cal La
y
er
(PH
Y) Spec
,
IEEE
802
.
11
St
anda
rd, 1999.
[19]
Si
y
u,
D.
,
Cheng
du,
C.
,
“
Rese
arch of
improved
Multi
-
ch
annel
MA
C
Protocol i
n
MA
NET”
,
IEEE, 2
017.
[20]
Mahe
sw
ari
,
R.
,
&
Kira
n
Kum
ar,
T.
V.U.
,
“
Inte
rnational
Journal
of
El
e
ct
ronics
a
nd
Comm
unic
ation
Engi
nee
ring
”
,
Volum
e
6,
Num
ber
3,
pp.
199
-
2
04,
2013
.
[21]
Velmurugan,
V.
,
and
Marti
n
L
e
o
Manic
k
am,
J.
,
“
Chara
cterist
ic
Anal
y
sis
of
Eff
i
ci
en
t
and
Reliab
le
Bro
adc
ast
in
Vehic
ul
ar
Networks”,
Inte
rnatio
nal
Journal
of
El
ectronics
and
Comm
unic
ati
on
E
ngine
ering
.
vol.
6,
no.
3,
pp.
205
-
210,
2013
.
Evaluation Warning : The document was created with Spire.PDF for Python.