Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
Vol.
5, No. 6, Decem
ber
2015, pp. 1452~
1
457
I
S
SN
: 208
8-8
7
0
8
1
452
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
Performance Analysis for Uniform
and Binomial Distribution
on Contention Window using DS
RC and Wi-Fi Direct Standard
Nur
a
in Iz
z
a
ti Shuh
aimi*
,
**
,
Heri
an
sy
ah
*,
T
u
tu
n Juh
a
n
a
*
,
Adi
t
Kur
n
i
a
w
a
n
*
*
Sekolah
Teknik Elektro
dan
Inf
o
rma
tika, Institu
t
Teknologi Ban
dung,
Jl. Gan
e
sa No.10 B
a
ndung
, Indonesia
**
Fakulti Kejur
u
teraan
Elek
trik
, Universiti
Tekn
ologi MARA, 40
450 Shah Alam,
Malay
s
ia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
J
u
l 18, 2015
Rev
i
sed
Au
g
28
, 20
15
Accepted
Sep 13, 2015
In this
p
a
per
,
w
e
pres
en
t
a m
a
t
h
em
atic
al
ana
l
ys
is
on the
perfo
rm
ance and
behavior of un
iformly
distribu
ted and non-unifor
mly
distribu
ted backoff
timer based on
binomial
algor
ithm b
y
using
two standards which is
th
e
conventional DSRC and the latest
Wi-Fi Direct. DSRC is a
well-known
techno
log
y
bein
g consider
ed
as the
most promising wireless standard
in
VANET. On the other hand,
as the late
st wire
le
ss ne
twor
king standard, th
e
potential of Wi-Fi Direct techn
o
log
y
should be concern
e
d. We evalu
a
tes
these standards
using uniform and
binomial distribution fo
r contention
window under m
a
them
atical m
odelling
in ord
e
r to an
al
y
z
e t
h
e aver
ag
e
throughput and
collision probabilit
y
perfo
rm
an
ce. Th
e results
show that
binomial distrib
u
tion in Wi-Fi Direct
stand
a
rd
is 7.05% and 97.13% better
than uniform
distribution
,
in t
e
rm
s of average throughput and collision
probabili
t
y
, r
e
s
p
ect
ivel
y.
Thus
we can def
e
r th
at W
i
-F
i Direct
is
feas
ible
to
be us
ed as
an alterna
tive s
t
and
a
r
d
s
i
nce it has
been cons
ider
ed as
potentia
l
competitor
of D
S
RC in VANET.
Keyword:
Bin
o
m
ial alg
o
r
ith
m
DSRC
U
n
i
f
or
m
an
d
no
n-u
n
i
for
m
l
y
Di
st
ri
b
u
t
e
d bac
k
o
f
f
t
i
m
e
r
VA
NET
W
i
-Fi direct
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
:
Nurain Izzati Sh
uh
aim
i
,
Seko
lah
Tekn
i
k
Elek
tr
o d
a
n In
fo
r
m
atik
a,
In
stitu
t Tekn
o
l
o
g
i
Band
ung
,
Jl. G
a
n
e
sa
No
.
1
0
Bandu
ng
, In
don
esia.
Em
a
il: nuraini
zzati_shuhaim
i
@
yahoo.com
1.
INTRODUCTION
In
rece
nt
y
ear
s, a g
r
owi
ng
num
ber
of m
obi
l
e
a
ppl
i
cat
i
ons
has
bee
n
i
n
t
r
o
d
u
ced t
h
r
o
u
g
h
b
r
oad
ado
p
t
i
o
n
of
wi
rel
e
ss
net
w
or
k.
Part
i
c
ul
arl
y
i
n
ve
hi
cul
a
r
com
m
uni
cat
i
on sy
st
em
, vehi
cl
e c
a
n act
s as
a m
obi
l
e
no
de t
h
at
a
b
l
e
t
o
se
nd
dat
a
t
o
t
h
e
nei
g
h
b
o
u
r
i
ng
ve
hi
cl
e o
r
roa
d
si
de
uni
t
,
and
vi
ce
ve
rsa.
A
ve
hi
cul
a
r a
d
-
h
oc
net
w
or
k (
V
A
N
ET) co
nsi
s
t
s
of
vehi
cl
e-t
o
-ve
h
i
c
l
e
and vehi
cl
e-t
o
-i
nf
rast
r
u
ct
ure com
m
uni
cat
i
on base
d o
n
l
o
cal
area net
w
ork technology. T
h
e transm
ission of i
n
form
ation
on
vehic
u
lar node incl
udes
a variety of s
e
rvices
ran
g
in
g
fr
om
road
safety
,
tra
ffic e
fficiency
an
d i
n
f
o
ta
in
men
t
app
licatio
ns [1
]. In term
s
of stan
d
a
rd,
IEEE
80
2.
1
1
p
o
r
al
s
o
kn
o
w
n a
s
WAVE
ha
s bee
n
defi
ned a
s
t
h
e
com
m
uni
cat
i
on
pr
ot
oc
ol
s o
f
phy
si
cal
a
nd m
e
di
um
access cont
rol
(MAC) layer for ve
hic
u
la
r environm
ents. The operating freque
ncy is fixe
d in De
dicated Short
R
a
nge
C
o
m
m
uni
cat
i
o
n
(D
S
R
C
)
ba
n
d
s
p
a
nni
ng
f
r
om
5.
85
t
o
5.
9
2
5
GHz
, i
n
w
h
i
c
h se
ve
n c
h
an
nel
s
are
allocated whe
r
e 1 c
h
annel s
p
ecified
as con
t
ro
l ch
an
nel (CCH)
wh
ile th
e
rem
a
i
n
i
ng
6 c
h
an
nel
s
ar
e
set
as
s
e
r
v
ic
e
ch
an
nels
(
S
CHs
)
[2
].
In
acco
r
d
a
n
c
e with
th
e in
itial d
e
sign
of 80
2.11
wireless n
e
two
r
k
i
ng
stan
d
a
rd
s, 802
.1
1p
allows
vehic
u
lar
node
s to
access t
h
e m
e
diu
m
by
using Ca
rri
er
Sense
Multipl
e
Access
with Collision Avoida
nce
(C
SM
A/
C
A
) [
3
]
whi
c
h im
pl
em
ent
e
d by
En
h
a
nced
Di
st
ri
b
u
t
i
on C
o
or
di
nat
i
on F
u
nct
i
on
(E
DC
F) t
h
at
ba
se
d o
n
DCF.
With
EDCF m
ech
an
is
m, a v
e
h
i
cu
lar
no
d
e
is allowe
d
to
start p
a
ck
et
tran
sm
issio
n
on
ly if th
e ch
ann
e
l is
idle for
Distributed Inter Fra
m
e Space
(DIFS) duration
before the bac
k
off
proce
d
ure is perm
itted to resum
e
.
A t
r
a
n
sm
i
t
t
i
ng st
at
i
on m
u
st
m
a
ke s
u
re t
h
at
t
h
e m
e
di
um
i
s
i
d
le fo
r t
h
is requ
ired
duratio
n
b
e
fo
re atte
m
p
t to
do
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Perf
or
ma
nce A
nal
ysi
s
f
o
r
U
n
i
f
o
rm
a
n
d
Bi
n
o
m
i
a
l
Di
st
ri
but
i
o
n
o
n
C
o
nt
ent
i
on…
(
N
u
r
ai
n
I
zzat
i
S
h
u
h
a
i
m
i
)
1
453
th
e tran
sm
itt
in
g
pro
cess [4
].
Sin
ce th
ere is
a p
o
ssi
b
ility o
f
statio
n
s
cho
o
s
in
g
th
e sam
e
b
ack
off
n
u
m
b
e
r, th
ey
may send the data in sam
e
slot whic
h leads to prob
a
b
ility of colliding with each
other. That is why a
u
n
i
form
l
y
d
i
strib
u
t
ed rand
o
m
b
ackoff
nu
m
b
er is
d
e
sign
ed
in ord
e
r to allev
i
ate th
is prob
abilit
y o
f
co
llision
.
Figure
1. Basic access m
e
thod
As in
Figu
re 1, after sensing
th
at th
e
m
e
d
i
u
m
is b
u
s
y, th
en
th
e statio
n
sh
all d
e
fer for
an
o
t
h
e
r
free
DIF
S
t
i
m
e
and
sel
ect
s a uni
f
o
rm
random
backo
ff
n
u
m
b
er fr
om
cont
ent
i
on
wi
n
d
o
w
(C
W)
.
Usual
l
y
t
h
e st
at
i
o
n
will doubles the curre
nt C
W
so t
h
at it can selects
anot
her
backoff
until reaching t
h
e maxim
u
m
size of C
W
[5]
.
The
eq
uat
i
o
n
o
f
t
h
e
back
of
f
peri
od
i
s
de
t
e
rm
i
n
ed
by
:
2
∗
wh
er
e
n
is num
ber of
retry
(1)
Ho
we
ver
,
t
h
e
m
a
i
n
charact
er
i
s
t
i
c
s i
n
vehi
c
u
l
a
r e
nvi
ro
nm
ent
i
n
whi
c
h
no
des t
y
pi
cal
l
y
m
ove o
n
a
ro
ad
with
h
i
gh
er m
o
b
ility a
n
d freq
u
e
n
t
n
e
twork top
o
l
o
gy ch
an
g
e
d
y
n
a
micall
y
th
an
t
h
o
s
e in o
t
h
e
r
ad
-ho
c
net
w
or
k sce
n
a
r
i
o
s l
e
a
d
t
o
s
o
m
e
i
ssues an
d c
h
al
l
e
nge
s i
n
net
w
or
k
des
i
gn,
espe
ci
al
l
y
o
n
M
A
C
l
a
y
e
r [
6
]
.
Ai
m
i
ng t
o
i
m
pro
v
e t
h
e M
A
C
per
f
o
r
m
a
nce i
n
VA
NET
,
we
propose a m
a
the
m
atical analys
is by im
ple
m
enting
a no
n-u
n
i
for
m
l
y
d
i
str
i
bu
ted back
of
f tim
er
b
a
sed
on
b
i
no
mial d
i
str
i
bu
tio
n fo
r
con
t
en
tion w
i
ndo
w wh
ich
can
redu
ce th
e collisio
n
p
r
o
b
a
b
i
lity
wh
ile in
crease th
e p
e
rfo
rm
an
ce o
f
th
rou
ghp
u
t
. Ap
art fro
m
u
s
in
g
th
e
con
v
e
n
t
i
onal
DSR
C
as com
m
uni
cat
i
on pr
ot
oc
ol
of
p
h
y
s
i
cal
and M
A
C
l
a
y
e
r for
ve
hi
cul
a
r e
nvi
r
o
n
m
ent
,
we
will u
s
e
Wi-Fi Direct
stan
d
a
rd
as an
altern
ate.
Altho
ugh
we
kn
ow th
at
DRSC
prov
id
es a
reliab
l
e
com
m
uni
cat
i
on
bet
w
ee
n v
e
h
i
cl
e-t
o
-v
ehi
c
l
e
,
pr
o
b
l
e
m
concerni
ng
o
n
t
h
e
cost
o
f
DSR
C
har
d
ware
bec
o
m
e
a
con
s
t
r
ai
nt
. B
e
c
a
use
of t
h
e
fac
t
t
h
at
DSR
C
r
e
qui
res a de
di
cat
ed ha
rd
war
e
, we ca
n ass
u
m
e
t
h
at
i
t
cannot
be
co
nsid
ered
as
th
e b
e
st so
l
u
tio
n
to
warn
p
e
d
e
strians or cyclists, sin
ce it
may req
u
i
re the
m
to
carry DSRC
equi
pm
ent all the tim
e [7], [8]. Because
of this s
h
or
t
c
om
ing, we
c
h
oose
Wi-Fi
Direct standa
rd as a
n
altern
ativ
e for
v
e
h
i
cle-t
o
-v
eh
icle co
mm
u
n
i
c
a
tio
n
sin
c
e it h
a
s
b
een con
s
i
d
ered
as
po
tential co
m
p
etito
r o
t
her
th
an
DSRC stan
d
a
rd
.
Th
e
rem
a
in
d
e
r of th
is p
a
p
e
r
is d
i
v
i
d
e
d in
to th
e
fo
llowing sectio
n
s
,
wh
ere Section
2 discu
sses t
h
e
ove
r
v
i
e
w
of
rel
a
t
e
d w
o
r
k
s
,
Se
ct
i
on
3
descri
b
e
s o
n
m
a
t
h
em
at
i
cal
anal
y
s
i
s
, and
Sect
i
o
n
4
pr
esent
s
on
res
u
l
t
and
di
scussi
o
n
.
Wh
i
l
e
t
h
e co
ncl
u
si
on
an
d
f
u
t
u
re
wo
rk
i
s
gi
ve
n i
n
Sect
i
o
n
5 a
n
d
6,
res
p
ect
i
v
el
y
.
2.
RELATED WORKS
A lo
t
o
f
stud
ies h
a
v
e
b
een
d
o
n
e
t
h
at sub
s
tan
tially ai
min
g
to
im
p
r
ov
e
th
e MAC
p
e
rfo
rm
an
ce in
vehi
c
u
l
a
r
net
w
or
ks s
u
ch
as b
y
m
odi
fy
i
ng t
h
e con
v
e
n
t
i
o
n
a
l
B
i
nary
Ex
p
o
n
e
nt
i
a
l
B
acko
ff
(B
EB
) al
g
o
ri
t
h
m
.
In
th
ese stud
ies, th
ey m
a
d
e
a p
r
o
p
o
s
al wh
eth
e
r to
ad
ju
st th
e i
n
itial v
a
lu
e o
f
C
W
b
a
sed
on
t
h
e esti
m
a
ted
d
e
n
s
ity
vehi
cl
es or
m
odi
fy
t
h
e
m
e
tho
d
of sel
ect
i
ng bac
k
of
f
i
n
t
e
ger wi
t
h
a d
e
t
e
rm
i
n
i
s
t
i
c
fu
nct
i
o
n
,
f
i
n
st
e
a
d of
a
ran
d
o
m
vari
abl
e
t
echni
que
. F
o
r i
n
st
ance
, t
h
e aut
h
o
r
s i
n
[
9
]
,
[1
0]
ex
pl
ai
n
e
d t
h
at
w
h
en
t
h
e
net
w
or
k c
o
nge
st
i
o
n
b
eco
m
e
s sev
e
re, th
e size of co
n
t
en
tio
n
windo
w
will d
e
p
e
nd
s stro
ng
ly on
th
e nu
m
b
er o
f
co
m
p
etin
g
v
e
hicles
al
on
gsi
d
e
wi
t
h
t
h
e
num
ber
of
t
r
ansm
i
ssi
on re
t
r
i
e
s.
Whi
l
e
i
n
[1
1]
, t
h
e a
u
t
h
o
r
s m
a
de a
pr
o
p
o
sal
by
a
ppl
y
i
n
g
t
h
e
schem
e
of collision alleviati
o
n. Here
, instead of
a
d
justi
ng the c
onte
n
tion wi
ndow size according to the
num
ber of com
p
eting node
s,
they propos
ed the adjusting
of conten
ti
on
window size accord
ing to the
average
num
ber of
ret
r
i
e
s.
Ot
he
rwi
s
e,
t
h
e
aut
h
or
s i
n
[
1
2
]
pr
op
ose
d
a si
m
p
l
e
m
odi
fi
cat
i
on
on
B
E
B
a
l
go
ri
t
h
m
whi
c
h ca
n
begi
n
with a relatively large value for cont
ention window, afte
r that divide it
by two each time when a rea
l
-tim
e
safet
y
m
e
ssage on t
h
e c
ont
r
o
l
cha
nnel
ex
pi
res,
part
i
c
ula
r
ly foc
u
sed t
h
e delivery of
safety
m
e
ssages. Per
cont
ra, t
h
e
val
u
e f
o
r l
a
rge c
o
nt
ent
i
o
n wi
n
d
o
w
i
n
t
h
ei
r en
ha
nced
p
r
ot
ocol
r
e
m
a
i
n
s un
de
fi
ned
.
Al
s
o
, a
g
r
ou
p
of
researc
h
er
s ha
ve com
e
out
w
i
t
h
one
-t
o
-
o
n
e
fu
nct
i
o
n m
a
ppi
ng
fr
om
severa
l
uni
q
u
e
refe
re
nce t
o
a sl
ot
n
u
m
b
er
in contention
window i
n
order t
o
inc
r
ease the cha
n
ces
of successful
reception proba
b
ility on the c
ont
rol
ch
ann
e
l.
For
a v
e
h
i
cu
lar node, th
e
un
iqu
e
r
e
f
e
r
e
n
c
e in
t
h
is ap
pro
ach is r
e
lev
a
n
t
to its ow
n g
e
og
r
a
p
h
i
cal
lo
catio
n
[
1
3
]
,
w
h
ile th
e study b
y
[
1
4
]
is on
its MA
C add
r
ess
of
w
i
r
e
l
e
ss in
ter
f
ace.
A
s
in
[15
]
, the n
o
d
e
s
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 :
1452 –
1457
1
454
gra
n
t
e
d t
o
c
h
o
o
se i
n
ad
vanc
e
a free back
of
f
val
u
e fo
r t
h
e
next
pac
k
et
wa
i
t
i
ng fo
r t
r
ans
m
i
ssi
on t
o
t
h
e up
dat
e
d
backoff by a
pplying T
r
ansm
it and Rese
rve (TAR) cha
n
nel
access sc
hem
e
.
Als
o
, t
h
e
node
s nee
d
t
o
piggy
b
ac
k
the
chosen bac
k
off value
i
n
c
u
rrent
p
ack
et t
h
at read
y to
tran
sm
it. W
ith
out
consi
d
eri
ng t
h
e increasing
num
b
e
r
of ve
hicular nodes, the
range
of
TAR
cycle can
rising ra
pidl
y whic
h
leads i
n
to
un
bou
nd
ed d
e
lay.
In
a
nut
s
h
el
l
,
we ca
n c
o
ncl
u
de t
h
at
si
ze
of
cont
e
n
tio
n
wind
ow is th
e m
a
jo
r fact
o
r
affectin
g
t
h
e M
A
C
per
f
o
r
m
a
nce i
n
vehi
c
u
l
a
r
net
w
o
r
k
.
T
h
us i
n
t
h
i
s
resea
r
ch
,
we
pr
op
ose a
m
odi
fi
ed bac
k
of
f al
g
o
ri
t
h
m
t
h
at
can
decrease t
h
e pos
sibility of continuous collisions
am
ong the com
p
eting
vehicles
by generating
a non-
uni
fo
rm
ly
di
st
ri
but
ed
bac
k
of
f
t
i
m
e
r based
o
n
bi
n
o
m
i
al
di
st
ri
but
i
o
n.
3.
MAT
H
EM
AT
ICAL
A
N
A
LYSIS
R
e
ferri
ng t
o
w
e
l
l
-
kn
o
w
n B
i
a
n
chi
m
odel
fo
r
m
odel
l
i
ng t
h
e
C
S
M
A
/
C
A m
e
chani
s
m
prese
n
t
e
d i
n
[1
6]
,
we calcu
late th
e th
rou
ghp
u
t
an
d
p
r
ob
ab
ility o
f
a p
a
ck
et tran
sm
issio
n
failu
re du
e to
collisio
n
in
bo
th
DSRC
an
d
W
i
-
F
i D
i
rect stan
d
a
r
d
.
In
co
nv
en
tio
n
a
l
un
if
or
m
d
i
str
i
b
u
tion
,
t
h
e thr
oug
hpu
t,
which
is th
e av
erag
e
in
fo
rm
atio
n
tran
sm
it
ted
in
a
slo
t
ti
m
e
o
v
e
r the av
erag
e du
ratio
n
of a sl
o
t
time can
b
e
calcu
lated
as fo
llows:
=
(2)
Whe
r
e
D
is t
h
e ave
r
a
g
e pa
cket payloa
d
size,
1
is th
e prob
ab
ility th
at
a slo
t
is id
le,
is the proba
b
ility that a slot
transm
its data successfully, and
1
1
is th
e prob
ab
ility
th
at a slo
t
is i
n
a co
llisio
n state. Th
e tran
smissio
n
pro
b
a
b
ility,
p
can be calculated
with minim
u
m
value
of
cont
e
n
t
i
on wi
n
d
o
w
as
. Fu
rthe
r,
refer t
o
idle
slot tim
e
,
is th
e ti
m
e
sp
en
t to
tran
sm
i
t
a pac
k
et s
u
cce
ssfully a
n
d
is th
e tim
e sp
en
t du
ri
n
g
p
a
ck
et co
llisio
n
.
As fo
r
b
i
no
m
i
al d
i
strib
u
tion
fun
c
tion
,
th
e
no
rm
alized
th
rou
ghp
u
t
is th
e sa
m
e
as in
Eq
uatio
n
(2
)
bu
t
the di
ffe
rence
will be
on the
calcula
tion
for proba
b
ility of idle, success a
n
d collision.
Noted t
h
at
proba
b
ilit
y
m
a
ss
fu
nct
i
o
n fo
r bi
n
o
m
i
al
dist
ri
but
i
o
n [
17]
i
s
gi
ve
n by
:
:
,
1
(3)
Whe
r
e
n
is
num
b
e
r o
f
t
r
ial,
x
is num
b
er of success
in
n
t
r
ial and
p
is
proba
b
ility of success i
n
si
ngl
e trial.
Th
us base
d
on
E
quat
i
o
n
3,
we can
det
e
r
m
i
n
ed
1
as
p
r
ob
ab
ility th
at
th
e slo
t
is i
n
i
d
le state,
1
is th
e
prob
abilit
y th
at a tran
sm
issio
n
o
c
cu
rring
o
n
th
e
ch
ann
e
l is
su
ccessfu
l
, an
d
1
1
1
is th
e
p
r
ob
ab
il
ity o
f
co
llisio
n
tran
sm
issio
n
.
4.
RESULT A
N
D
AN
ALY
S
IS
We
per
f
o
r
m
e
d an a
n
al
y
s
i
s
f
o
r
b
o
t
h
u
n
i
f
or
m
and bi
n
o
m
i
al
di
st
ri
b
u
t
i
o
n
usi
n
g t
h
e
par
a
m
e
t
e
rs f
o
r
DSRC a
n
d
Wi-Fi Dire
ct standa
rd as
sum
m
arized in
Ta
ble 1, i
n
orde
r to analyse t
h
e
perform
a
nce of
th
ro
ugh
pu
t and
p
r
ob
ab
ility of co
llisio
n
.
Fo
r th
e an
alysis,
we
d
ecid
e
d
t
o
u
s
e CSM
A
/CA m
ech
an
ism
p
r
o
t
o
c
o
l
for tran
sm
i
ttin
g
in
stead
of
RTS/CTS for
tw
o
reason
s.
First, th
is pro
t
o
c
o
l
w
a
stes ban
d
w
i
d
t
h
du
e
to
long
waitin
g
p
e
riod
an
d
freq
u
e
n
tly ex
ch
ang
i
ng
con
t
ro
l p
ack
ets. Secon
d
l
y, RTS/CTS will cau
ses
m
o
re co
m
p
licated
and a
d
ditive interfere
nce sce
n
arios
due
to c
o
llisions am
ong cont
rol
a
nd dat
a
m
e
ssages
[18]. Furt
her, we
fixed
th
e con
t
en
tion wind
ow at 31 ti
m
e
slo
t
s, as i
t
is th
e v
a
lu
e
u
s
ed
in
alm
o
st
p
a
p
e
rs in
th
e literatu
re wh
ile th
e
cont
e
ndi
ng
n
o
d
es a
r
e
vari
e
d
f
r
om
1 t
o
2
0
.
Tabl
e 1. Si
m
u
lat
i
on param
e
t
e
r
Para
m
e
ter
DSRC
Wi-Fi
Di
rect
Data rate
6 M
bps
6 M
bps
ACK
192 bits
192 bits
DI
FS 58
μ
s 34
μ
s
SIFS
32
μ
s 16
μ
s
Pr
opagation delay
2
μ
s 1
μ
s
M
A
C header
256 bits
256 bits
PHY header
192 bits
192 bits
Slot tim
e
13
μ
s 9
μ
s
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Perf
or
ma
nce A
nal
ysi
s
f
o
r
U
n
i
f
o
rm
a
n
d
Bi
n
o
m
i
a
l
Di
st
ri
but
i
o
n
o
n
C
o
nt
ent
i
on…
(
N
u
r
ai
n
I
zzat
i
S
h
u
h
a
i
m
i
)
1
455
From
Fi
gu
re 2
,
we can
ob
ser
v
e
th
at th
e av
erag
e th
roug
hpu
t o
f
u
n
i
form
d
i
strib
u
tion
in
W
i
-Fi Di
rect
st
anda
rd
i
s
bet
t
er t
h
an
D
S
R
C
st
anda
r
d
by
3
9
.
2
%.
Her
e
,
w
e
can say
t
h
at
by
usi
ng
u
n
i
f
o
r
m
di
st
ri
but
i
o
n
,
t
h
e
p
r
ob
ab
ility o
f
id
le and
su
ccess in
bo
th
stand
a
rd
s seem
s t
o
redu
ce as the nu
m
b
er of
no
d
e
s are in
creasin
g
,
resu
lting
o
n
t
h
e d
e
grad
ation
of av
erag
e
th
ro
ugh
pu
t.
In term
s o
f
stan
d
a
rd
,
W
i
-Fi
Direct h
a
s a
b
e
tter
per
f
o
r
m
a
nce si
nce t
i
m
e
t
a
ken
of
,
and
reco
rd
ed
m
u
ch
lo
wer
v
a
lu
e
wh
ich
will cau
ses
a
g
r
eater th
rou
g
h
p
u
t
.
It fo
llo
ws th
e th
eo
retical stu
d
y
in
p
r
ev
iou
s
stud
y i
n
wh
ich th
e throug
hpu
t is inv
e
rsely
p
r
op
or
tio
n
a
l t
o
th
e av
er
ag
e
dur
atio
n of
a
slo
t
ti
m
e
[
1
9
]
.
Fi
gu
re 2.
Th
r
o
ug
h
put
f
o
r di
f
f
e
rent
n
odes
u
s
i
n
g
uni
fo
rm
di
st
ri
but
i
o
n
Fi
gu
re 3.
Th
r
o
ug
h
put
f
o
r di
f
f
e
rent
n
odes
u
s
i
n
g
bi
n
o
m
i
al
di
st
ri
but
i
o
n
While i
n
Fi
gure 3,
we ca
n se
e that the
a
v
erage
t
h
ro
ugh
pu
t
of
b
i
no
m
i
al d
i
str
i
b
u
tion in
W
i
-
F
i
D
i
r
e
ct
st
anda
rd
i
s
b
e
t
t
e
r t
h
a
n
DSR
C
st
anda
rd
by
3
9
.
1
3
%
.
As
we c
a
n see
fr
om
t
h
e gra
p
h,
w
h
en
t
h
e c
ont
e
ndi
ng
no
de
s
are va
ri
ed f
r
o
m
1 t
o
7 no
d
e
s, t
h
e av
erage th
roug
hpu
t is lo
wer
p
a
rtic
ularly whe
n
the node
s are
sm
a
ll.
Co
nv
ersely, wh
en
t
h
e con
t
end
i
ng
nod
es are v
a
ried
un
til 2
0
nod
es, th
e t
h
roug
hpu
t u
s
ing b
o
t
h
stand
a
rd
s see
m
s
to
b
e
stab
le and
grad
u
a
lly redu
ced.
Here,
we can
ob
serv
e th
at b
y
using
b
i
n
o
m
ial d
i
strib
u
tio
n
,
th
e prob
ab
ility
of s
u
ccess i
n
both sta
nda
rds s
e
e
m
s to reduce
as the nu
m
b
er of c
o
m
p
eting
nodes a
r
e sm
all. On t
h
e contrary,
th
e prob
ab
ility
o
f
su
ccess
g
e
ttin
g
h
i
gh
er as t
h
e
n
u
m
b
e
r
o
f
no
d
e
s are in
creasin
g
. Th
er
efore fro
m
th
e resu
l
t
, we
can sum
m
ari
z
e t
h
at
bi
nom
i
a
l
di
st
ri
b
u
t
i
on i
s
sui
t
a
bl
e t
o
use
com
p
ared t
o
uni
fo
rm
di
st
ri
but
i
on i
n
heavy
l
o
ad
traffic con
d
ition
.
Fi
gu
re 4.
Th
r
o
ug
h
put
f
o
r di
f
f
e
rent
n
odes
u
s
i
n
g
b
o
t
h
d
i
str
i
bu
tio
n
Fig
u
re
5
.
Co
llisio
n prob
ab
ility for
d
i
fferen
t
n
o
d
e
s
u
s
ing
bo
th
d
i
stribu
tio
n
Fig
u
r
e
4
sh
ow
s th
e co
m
p
ar
ison
o
f
thro
ug
hpu
t e
fficiency between both uniform
a
nd
binom
i
a
l
di
st
ri
b
u
t
i
on i
n
DSR
C
an
d
W
i
-Fi
Di
rect
st
an
dar
d
.
Here,
we
can concl
u
de that W
i
-Fi Direct can works
better
t
h
an
DSR
C
st
a
nda
r
d
, es
peci
al
l
y
by
gene
rat
i
ng
a n
o
n
-
u
ni
fo
rm
ly
di
st
ri
but
e
d
b
ack
of
f t
i
m
er bas
e
d
o
n
bi
n
o
m
i
al
di
st
ri
b
u
t
i
o
n
,
w
h
ere
bi
n
o
m
i
al
di
st
ri
b
u
t
i
on i
n
W
i
-Fi
Di
r
ect
i
s
7.
05%
bet
t
e
r
t
h
an u
n
i
f
orm
di
st
ri
b
u
t
i
on i
n
W
i
-F
i
Di
rect
.
Fr
om
the
resul
t
,
we c
a
n see
t
h
at
bi
n
o
m
i
al
di
st
ri
b
u
t
i
on
ha
vi
n
g
sl
i
g
ht
l
y
bet
t
e
r
per
f
o
rm
ance t
h
a
n
uni
fo
rm
d
i
stribu
tio
n. Th
is is du
e to
t
h
e tran
sm
issio
n
d
a
ta
rate
t
h
a
t
we set
at
6 M
bps
fo
r
bot
h
st
anda
rd
s. It
i
s
kn
o
w
n
t
h
at
W
i
-Fi
Di
r
ect
offe
rs fast
e
r
dat
a
rat
e
o
v
e
r
t
w
o
-
way
area
cove
rage
, w
h
i
c
h i
s
up t
o
25
0 M
b
ps com
p
ared t
o
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 :
1452 –
1457
1
456
DSR
C
t
h
at
o
f
f
e
rs
onl
y
6-
2
7
M
bps
. I
n
t
h
i
s
s
t
udy
,
we c
h
oo
s
e
t
h
e m
i
nim
u
m
dat
a
rat
e
at
6
M
bps
i
n
o
r
de
r
t
o
see
t
h
e fai
r
di
ffe
r
e
nce
bet
w
ee
n
bot
h st
a
nda
rd
s
.
W
e
al
so
can summ
arize that
W
i
-Fi
Dire
ct can
ha
ve a
good
t
h
r
o
u
g
h
p
u
t
at
hi
g
h
c
ont
e
ndi
n
g
no
des
o
n
l
y
i
f
we set
t
h
e t
r
a
n
sm
i
ssi
on dat
a
rat
e
at
hi
g
h
er
num
ber,
w
h
i
c
h
i
s
u
p
to
250
M
b
p
s
.
Afterward, Figu
re
5
sho
w
s the co
llisio
n
p
r
ob
ab
ility
b
e
tween
b
o
t
h
un
ifo
r
m
an
d
b
i
no
m
i
a
l
d
i
stribu
tion
i
n
DSR
C
an
d
W
i
-Fi
Di
rect
st
anda
rd
. He
re,
we can
ob
ser
v
e t
h
at
bi
nom
ial
di
st
ri
but
i
o
n
i
n
b
o
t
h
st
an
da
rds i
s
9
7
.23
%
b
e
tter th
an
u
n
i
form
d
i
strib
u
tion
.
We
can
in
fer
th
at wh
en
th
e n
u
m
b
e
r o
f
n
o
d
e
s
i
n
crease,
th
e
co
llisio
n
probability will also in
crease
.
Noticeably the collisi
on
probability will al
ways grow
up with
the
inc
r
ea
sing
i
n
n
u
m
b
e
r
o
f
d
a
t
a
flows. Fu
rt
h
e
r, t
h
e co
llisio
n prob
ab
ility o
f
b
i
no
m
i
a
l
d
i
strib
u
tion
is m
u
ch
lo
wer t
h
an unifo
rm
di
st
ri
b
u
t
i
on
be
cause n
o
d
es c
a
n sel
ect
t
h
ei
r
back
o
ff i
n
t
e
ge
r i
n
o
r
de
r t
o
a
voi
d o
n
ch
o
o
si
ng t
h
e sam
e
backo
ff
integer as the
othe
r nodes
do in conve
ntional schem
e
. It
m
eans, each node can e
n
ter diffe
re
nt cha
nnels
,
resu
lting
on
red
u
c
i
n
g
th
e
po
ssib
ility o
f
co
llisio
n
.
Th
us we can
con
c
lu
d
e
th
at p
e
rfo
r
m
a
n
ce o
f
b
i
n
o
m
ia
l
di
st
ri
b
u
t
i
o
n
o
n
cont
e
n
t
i
o
n wi
nd
o
w
pe
rf
o
r
m
s
bet
t
e
r t
h
a
n
u
n
i
f
o
r
m
di
st
ri
but
i
on i
n
b
o
t
h
DS
R
C
and
Wi
-Fi
Di
rect
st
anda
rd
.
5.
CO
NCL
USI
O
N
Thi
s
pape
r
pre
s
ent
e
d t
h
e a
n
a
l
y
s
i
s
of t
w
o
d
i
ffere
nt
di
st
ri
b
u
t
i
o
n
w
h
i
c
h i
s
uni
fo
rm
di
st
ri
but
i
o
n a
n
d
no
n
-
u
n
i
f
orm
l
y
di
st
ri
b
u
t
e
d
bac
k
o
f
f
t
i
m
e
r base
d
on
bi
n
o
m
i
al
al
go
ri
t
h
m
for c
ont
e
n
t
i
o
n
wi
n
d
o
w
by
usi
n
g
D
S
R
C
an
d
Wi-Fi Di
rect stan
d
a
rd
. In
a nu
tsh
e
ll, the si
m
u
latio
n
resu
lts sho
w
th
at
b
i
n
o
m
ial d
i
strib
u
tion
ou
tp
erfo
rm
s
th
e con
v
e
n
tional u
n
i
fo
rm
d
i
stribu
tio
n
i
n
term
s
o
f
av
erag
e
th
ro
ugh
pu
t and co
llisio
n
p
r
o
b
ab
ility. No
ted
t
h
at th
e
ex
p
l
an
ation
on th
ese
resu
lts
hav
e
b
een presen
ted
as in
pr
evio
u
s
sectio
n. B
y
lo
ok
ing
at
b
o
th
stand
a
rd
s, we can
concl
ude
t
h
at
W
i
-Fi
Di
rect
has a
bet
t
e
r
per
f
o
rm
ance than DSRC since it s
hows th
at it h
a
s
a go
od
achi
e
vem
e
nt
. For
a co
ncl
u
si
on
, we
k
n
o
w t
h
at
DSR
C
has
bee
n
co
nsi
d
er
ed as t
h
e m
o
st
pr
om
i
s
i
ng wi
rel
e
ss
standa
rd
but
there are several circum
sta
n
ces that
s
h
o
u
l
d
be em
pha
si
zed. I
n
st
ea
d
of
rel
y
i
ng
o
n
t
h
i
s
con
v
e
n
t
i
onal
s
t
anda
rd t
h
at
h
a
s been
wi
del
y
used i
n
ve
hi
cul
a
r net
w
o
r
k,
t
h
e ext
r
a ad
v
a
nt
age
of u
s
i
n
g
W
i
-
F
i
Direct where i
t
is known as
a peer-to-pee
r
standa
rd
t
h
at allows
W
i
-Fi devices, ca
n connect
directly a
m
ong
each othe
r without the nee
d
for a wire
less hots
p
ot. Thus we can say that W
i
-Fi Di
rect as the latest wireles
s
net
w
or
ki
n
g
st
a
nda
r
d
ca
n
ove
r
t
ake D
S
R
C
pe
r
f
o
r
m
a
nce i
n
t
h
e nea
r
f
u
t
u
re.
Fo
r fu
ture
wo
rk
,
we
will ev
al
u
a
te and
stud
y m
o
re on
th
e beh
a
v
i
o
u
r
o
f
b
i
n
o
m
ial b
acko
f
f algo
rith
m
,
in
add
itio
n we
can
test th
e proto
c
o
l
p
e
rform
a
n
ce
with
real larg
e scale of test b
e
d exp
e
rim
e
n
t
using
NS-2
.
REFERE
NC
ES
[1]
N. M. Sadek
,
et
al
., “A Robust Multi-RAT VANET/LTE for
Mixe
d Contro
l and
Enter
t
ainment
Traffic,”
Journa
l o
f
Transpo
rtation Technologies
, V
o
l. 5
,
pp
. 113-12
1, 2015
.
[2]
K.
A.
Hafeez,
et al
.,
“Optimizing the Control Channel Interv
al of
DSRC for
Vehicu
lar Saf
e
ty
Applications,”
Globecom Workshop on Wireless Networki
ng and Control for
Unmanned Autonomous
Vehicle
s
, pp. 1481-1486,
2014.
[3]
V. D. Khairnar
1 and K. Kote
c
h
a, “
P
erform
anc
e
of
Vehic
l
e-
to-
V
ehicl
e
Com
m
u
nica
tion using I
EEE 802
.11p in
Vehicular Ad-h
oc Network En
vironment,”
International Journ
a
l of Network
S
ecurity and App
lications (
I
JNSA)
,
Vol. 5
,
No. 2, 20
13.
[4]
G.
Shanti,
et
al
., “Performance
Analy
s
is of I
E
EE 802.
11e under
Binomial Back
off Algorithm,”
IEEE In
ternatio
nal
Advance Compu
ting Conference
(
I
ACC
)
, pp. 267
-
270, 2014
.
[5]
V. Tiwari
, “
P
erf
o
rm
ance Enh
a
nc
em
ent in VANET with Ad
mission Control and
Contenti
on Win
dow Adjustment,”
Master of
Applied Science Thesis
, University
of
V
i
ctoria, Can
a
da,
2012.
[6]
W. Alasm
a
r
y
and W. Zhuang
,
“The Mobilit
y
I
m
pact in
I
EEE
802.11p Infrastr
u
cture
less Vehi
cular N
e
tworks,”
Elsevier Ad
Hoc
Networks Journ
a
l
, vo
l. 10, no. 2,
pp. 222-230, 20
12.
[7]
C. Satish, “
I
nter
Vehicular Com
m
unication fo
r Collision Avoidance using W
i
-Fi Direct
,”
Master Thesis
, Rochester
Institute of
Tech
nolog
y
,
New
York United
Stat
es, 2014.
[8]
A. Pri
y
ank
a
, “In
c
reased Persist
e
nce of Wi-Fi Direct
Networks for Sm
artphone-b
ased Collision Avoidance,”
Master
The
s
is
, Rochest
e
r Institu
te o
f
T
e
chnolog
y
,
New
York United
Stat
es, 2014.
[9]
C. Chr
y
sostomou,
et
al
.,
“
A
ppl
ying Adapt
i
ve Q
o
S
-
aware M
e
diu
m
Acce
ss Contr
o
l in
Priorit
y
Ba
sed Vehicu
lar
Ad
Hoc Networks,”
Pr
oceed
ing on
1
6
th
IEEE S
y
mposium on Computers and Communications
, pp. 741-
747, 2011
.
[10]
R. Reind
e
rs,
et al
., “Contention
Window Analy
s
is
for Beacon
i
n
g
in VANETs,”
Proceeding on
7
th
Internat
ion
a
l
Wireless Communication
and Mo
bile Computing
Conference
, pp.
1481-1487, 201
1.
[11]
D.
J.
Deng,
et al
., “A Collisio
n Alleviation S
c
heme for IEEE 802.11p VANETs,”
Springer Wireless Personal
Communications
, Vol. 56
, No. 3,
pp. 371-383
, 20
11.
[12]
R. Stan
ica,
et
al
., “Enhan
cements
of IEEE 802.11
p Protocol
for A
ccess Contro
l on
VANET Contro
l Chann
e
l,”
IE
EE
International Co
nference of Com
m
unication
, pp
.
1–5, 2011
.
[13]
A.
Ahmad,
et
al
., “
A
New Adapted B
ack-off
Schem
e
for B
r
oadcast
i
ng on
IEEE 1609
.4 C
ontrol Chann
e
l
in
VANET,
”
Proceeding on
Ad
Hoc Networking Wo
rkshop
, pp. 9-15, 2012.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Perf
or
ma
nce A
nal
ysi
s
f
o
r
U
n
i
f
o
rm
a
n
d
Bi
n
o
m
i
a
l
Di
st
ri
but
i
o
n
o
n
C
o
nt
ent
i
on…
(
N
u
r
ai
n
I
zzat
i
S
h
u
h
a
i
m
i
)
1
457
[14]
K.J. Song,
et al
., “Distributed
Periodic Access
Scheme (DPAS) for the Periodi
c
Sa
fe
ty
Me
ssa
ge
s in the
IEEE
802.11p WAVE,”
3
rd
Proceed
in
g International
Conference
on
Communications and Mobile Co
mputing
, pp. 46
5-
468, 2011
.
[15]
I. Khoufi,
et al
., “
T
AR Channel Acces
s
M
e
c
h
anis
m
for
VANET Safet
y
-Cri
tic
al Situat
ions,
”
Pr
oceed
ing on
Intell
igen
t and
A
d
vanced
Syst
em
, 2012.
[16]
G. Bian
chi, “Performance Anal
y
s
is of
the I
E
EE 802.11
Distributed Coordin
a
tion Function
,
”
Journal on Selected
Areas in Commu
nications
, pp
. 53
5–547, 2000
.
[17]
J. Wroughton and T. Cole,
“Distinguishing between
Bino
mial, H
y
perg
eo
metric and
Negativ
e Binomial
Distributions,”
Journal of S
t
atistics Educa
tion
, V
o
l. 21
, N
o
. 1
,
20
13.
[18]
C. Liu
,
“An Innovativ
e MultiV
ariab
l
e Contro
l
Fram
ew
ork for Effective Wir
e
le
ss Resource Managem
e
nt,”
Ph
D
The
s
is
, University
of Alberta, C
a
nada, 2012.
[19]
J. Sheng, “Performance Modelli
ng of IEEE 802.11 Ad Hoc Networ
ks
under Time-Var
y
i
ng Channel,”
PhD T
h
esis
,
Rensselaer
Poly
technic Institute
Tro
y
, N
e
w York, 2008
.
BIBLIOGRAPHIES
OF AUTHO
R
S
Nur
a
in I
zzati Shu
haimi
is a
PhD candidate
in Teleco
mmunication Eng
i
neer
ing at B
a
ndung
Institute of T
ech
nolog
y
,
Indonesi
a. She receiv
ed
her M.Eng degr
ee in El
ectical E
ngineer
ing of
Telecommunication from Natio
nal University
of
Malay
s
ia in
2011. Her r
e
s
earch
interests
includ
e data n
e
tworking, d
a
ta communicatio
n, vehicular ad-hoc network
and network
sim
u
lation.
Heriansy
ah
is a Master’s Deg
r
ee Studen
t
of
Te
lecommunication Engin
eering
at Bandung
Institute of Technolog
y
,
Indonesi
a. He has received his Bach
el
or of Techno
l
o
g
y
degree in
Electrical
Engin
eering from Lampung University
, Indonesia in
2011. His research interests
includ
e veh
i
cu
lar ad-hoc network, com
puter n
e
twork and d
a
ta communication.
Evaluation Warning : The document was created with Spire.PDF for Python.