Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
g
in
eeri
ng
(IJ
E
C
E)
Vo
l.
9
, No
.
5
,
Octo
ber
201
9
, pp.
3531
~3
541
IS
S
N: 20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v9
i
5
.
pp3531
-
35
41
3531
Journ
al h
om
e
page
:
http:
//
ia
es
core
.c
om/
journa
ls
/i
ndex.
ph
p/IJECE
Modelin
g and an
alys
i
s of IEE
E
1609.4 MA
C in the
p
re
s
ence
of
error
-
p
rone chan
nels
Ak
r
am A. Al
mohamme
di
1
, Nor
K
. No
or
din
2
, A. Sali
3
, F
az
ir
ulhi
sy
am
H
as
him
4
,
Wah
e
b
A.
Jab
ba
r
5
,
Sabri
S
ae
ed
6
1
,2,3,4
Depa
rtment
of
Com
pute
r and
Com
m
unic
at
ion
Engi
n
ee
ring
,
U
nive
rs
it
i
Putra
Malay
si
a
,
Malay
si
a
2
,3,4
W
ire
le
ss
and
Photonic
s Net
w
orks L
abor
a
torat
or
y
,
Unive
rsiti Putra
Ma
lay
sia
,
Malay
s
ia
5
Facul
t
y
of Engi
nee
ring
T
ec
hnol
o
g
y
,
Univer
si
ti
Malay
s
ia Paha
n
g,
Mal
a
y
s
ia
6
Depa
rtment of
Com
m
unic
at
ion
S
y
stems
Engi
n
e
eri
ng
,
T
ai
z
Univ
ersity
,
T
ai
z
-
Yem
en
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Ju
l
1
9
, 2
01
8
Re
vised
A
pr
4
,
201
9
Accepte
d
Apr
15
, 201
9
Vehic
ul
ar
Ad
H
oc
Networks
(V
AN
ET
s)
have
b
e
en
dev
el
oped
to
improve
the
safe
t
y
,
comfort
and
eff
ic
i
ency
o
f
drivi
ng
on
th
e
roa
d.
The
IE
EE
1609.
4
is
a
standa
rd
intend
ed
to
support
m
ult
i
-
cha
nn
el
i
n
VA
NETs.
The
se
cha
nne
ls
inc
lud
e
one
control
cha
nnel
for
safe
t
y
applic
at
ion
s
and
six
servic
e
cha
nne
ls
for
servic
e
appli
ca
t
ions.
How
ever,
the
r
e
is
sti
ll
no
comprehe
nsi
ve
anal
y
s
is
for
the
av
era
g
e
del
a
y
and
s
y
st
em
throughput
of
IEE
E
1609
.
4
MA
C
in
VA
NETs
conside
ring
err
or
-
pro
ne
cha
nn
el
und
er
non
-
saturate
d
condi
ti
ons.
In
thi
s pa
per
,
we
propose
an
anal
y
t
ical
m
odel
s ba
sed
on
1
-
D a
nd
2
-
D Markov
cha
in
to
evalua
t
e
th
e
pe
rform
anc
e
ana
l
y
sis
o
f
I
EE
E
1609.
4
-
b
ase
d
M
AC
i
n
the
pr
ese
nce
of
err
or
-
prone
cha
n
nel
s.
B
eside
s,
fr
ee
z
ing
of
the
b
a
ck
-
off
t
imer
is
ta
ken
int
o
con
sidera
t
ion
to
p
ro
vide
a
n
a
cc
ur
at
e
esti
m
at
ion
of
acce
ss
to
th
e
cha
nne
l
.
Th
e
sim
ula
ti
on
resul
t
s
have
bee
n
c
arr
ie
d
out
to
v
al
id
at
e
th
e
ana
l
y
t
ical
resul
ts
of
o
ur
m
odel
.
The
result
s
show
tha
t
the
pe
rform
anc
e
of
our
m
odel
outpe
rfor
m
s
the
exi
sting
m
odel
in
te
rm
s
of
pac
ke
t
delive
r
y
ra
ti
o
and
ave
rag
e
dela
y
of
safe
t
y
pac
k
et
s
o
ver
CCH,
and
s
y
stem
throughput
of
servic
e
pac
ke
ts ove
r
SC
Hs
.
Ke
yw
or
d
s
:
B
ack
-
off freezi
ng tim
er
Error
-
pron
e
cha
nn
el
s
IEEE 1
609.4
Perfo
rm
ance an
al
ysi
s
V
A
NETs
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
:
Akram
A
. Alm
oh
am
m
edi
Dep
a
rtm
ent o
f C
om
pu
te
r
an
d C
omm
un
ic
at
io
n
Syst
em
s En
gi
ne
erin
g,
Un
i
ver
sit
i P
utr
a Ma
la
ysi
a,
43300 Se
rd
a
ng
, Mal
ay
sia
.
Em
a
il
:
akr
am
a
2810@
gm
ai
l.c
om
1.
INTROD
U
CTION
Re
centl
y,
with
inc
reasin
g
the
po
pu
la
ti
on,
the
num
ber
of
reg
ist
ere
d
veh
ic
le
s
has
dr
am
at
ic
ally
increase
d
ov
e
r
a
ll
the
world,
a
nd
t
his
le
ads
t
o
a
high
rate
of
traff
ic
acci
de
nt
s
on
the
ro
a
ds
.
In
orde
r
to
pr
e
ven
t
su
c
h
acci
den
ts
,
an
I
ntell
igent
Tran
s
portat
io
n
Syst
em
s
(I
TSs)
are
nee
de
d.
I
TSs
ty
pica
ll
y
has
the
abili
ty
to
i
m
pr
ove
the
qu
al
it
y,
eff
ect
ive
ness
a
nd
sa
fety
of
t
he
fu
t
ur
e
tr
ans
portat
ion
syst
e
m
s.
Howe
ve
r,
VANET
s
are
the
key
com
po
ne
nt
of
ITSs
w
hich
integ
rate
w
irel
ess
netw
orks
into
ve
hicle
s.
V
AN
E
Ts
suppo
rt
three
ty
pes
of
com
m
un
ic
at
ion
s
incl
ud
i
ng
V
ehicl
e
-
to
-
Ve
hi
cl
e
(V2V),
Ve
hicle
-
to
-
I
nfrast
ru
ct
ur
e
(V2
I)
,
and
Hy
br
id
Ve
hicular
(HV)
c
omm
un
ic
at
ion
s.
T
he
a
pp
li
cat
io
ns
of
VANETs
are
di
vid
ed
into
tw
o
cat
egories,
sa
f
et
y
app
li
cat
ions
an
d
serv
ic
e
ap
plica
ti
on
s.
Saf
et
y
app
li
cat
io
ns
a
re
us
e
d
t
o
noti
fy
dr
i
ver
s
ab
ou
t
the
crit
ic
al
s
it
uation
i
n
a
dvance
.
On
t
he
ot
her
hand,
ser
vice
app
li
c
at
io
ns
ar
e
us
e
d
f
or
im
pro
ving
dri
ving
com
fo
rt
a
nd
the
eff
ic
ie
nc
y
of
trans
portat
ion.
As
a
re
su
lt
,
s
afety
app
li
cat
ion
s
are
delay
-
sensiti
ve
an
d
hav
e
a
hi
gh
e
r
pr
i
or
it
y,
w
hile
serv
ic
e
app
li
cat
io
ns
ar
e
thr
oughput
-
s
ensiti
ve
a
nd
ha
ve
lo
wer
pri
ori
ty
.
The
US
Fe
der
al
C
omm
un
ic
at
ion
s
Com
m
issi
on
(F
CC
)
has
al
lo
cat
ed
a
fr
eq
ue
ncy
band
of
5.
9
G
Hz
in
a
total
ban
dwidt
h
of
75
M
Hz
unde
r
DS
RC
to
suppo
rt
7
-
c
ha
nn
el
s
ea
ch
of
w
hich
is
10
MHz
wi
de
an
d
the
guar
d
ba
nd
is
5
MHz
wi
de.
These
cha
nn
el
s
are
functi
onal
ly
div
ide
d
int
o
one
co
ntr
ol
cha
nn
el
(CCH
_178),
an
d
up
to
six
are
se
rv
ic
e
c
ha
nn
el
s
(S
C
Hs
)
[
1
-
3]
.
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
20
19
:
3531
-
35
4
1
35
32
The
CC
H
is
e
xc
lusively
us
e
d
to
broa
dcast
sa
fety
-
crit
ic
al
ap
plica
ti
on
s
a
nd
regular
tra
ff
ic
s
,
w
hile
the
six
oth
e
r
channels,
SC
Hs,
are
ded
ic
at
ed
to
trans
f
er
serv
ic
e
da
ta
app
li
cat
ions.
W
ire
le
ss
Access
in
Ve
hicular
En
vironm
ent
(W
A
VE
)
has
been
desi
gn
e
d
for
V
ANETs
based
on
the
IEEE
802.1
1p
an
d
I
EEE
1609.
x
sta
nd
a
rds
fam
i
ly
.
The
IE
EE
802.1
1p
sta
nda
rd
s
uppo
rts
both
the
ph
ysi
cal
(P
H
Y
)
an
d
m
e
diu
m
acce
ss
con
t
ro
l
(MAC)
la
ye
rs
of
D
SRC
.
T
he
I
EEE
1609
.4
is
t
he
sta
ndar
d
(legacy)
intende
d
t
o
s
uppo
rt
m
ulti
-
chan
nel
op
e
rati
on
in
V
AN
E
Ts
[
1
-
3].
The
re
peati
ng
synch
ronizat
io
n
inter
vals
(S
I
)
a
m
on
g
the
ch
ann
el
s
to
tra
nsm
it
the
pack
et
s
a
re
10
0
ms
,
an
d
eac
h
SI
is
e
ve
nly
div
ide
d
int
o
50
ms
for
CC
H
I
nter
val
(C
CHI)
a
nd
5
0
ms
SC
H
In
te
r
val
(S
CH
I
)
as s
how
n
i
n
F
ig
ure
1
[
1
-
5].
Figure
1. Mult
i
-
ch
a
nnel
MAC
operati
on i
n V
AN
E
Ts
-
IEEE
1609.
4
Fo
r
SC
Hs
re
s
erv
at
io
n
i
n
IE
EE
16
09.4,
ve
hicle
(
prov
i
de
r)
i
niti
al
ly
br
oa
dcasts
WAVE
Se
r
vice
Adve
rtisem
ent
(
WSA)
pac
ke
t
on
the
CC
H
durin
g
C
C
HI
to
a
dverti
se
it
s
ser
vice
a
nd
s
el
ect
an
a
ppr
opriat
e
SCH
f
or
this
s
erv
ic
e.
If
ot
her
ve
hicle
s
(
us
er
s)
intere
st
to
join
t
he
se
rv
ic
e
,
w
hich
is
ad
ve
rtise
d
by
a
p
r
ov
i
der,
will
rep
ly
by
Re
qu
est
f
or
Se
rv
ic
e
(RF
S).
T
hen,
the
pr
ov
i
der
rep
li
es
the
us
ers
by
Ac
knowl
ed
gm
ent
(A
C
K)
ei
ther
f
or
acce
ptance
or
re
j
e
ct
ion
.
Ty
pical
ly
,
pack
et
s
coll
isi
on
m
ay
occu
r
if
m
or
e
than
one
ve
hicle
sta
rts
transm
itti
ng
a
pack
et
sim
ultan
eo
us
ly
within
the
sa
m
e
t
i
m
e
slot.
W
he
rea
s,
transm
issi
on
err
or
m
ay
m
a
nifest
du
e
t
o
the
co
m
plex
con
diti
on
o
f
a
wirele
ss
cha
nn
el
in
VANETs
s
uc
h
as
path
loss
,
therm
al
no
ise
,
channel
fad
i
ng, or
i
nter
fer
e
nce
from
o
ther radi
o resources
.
The
pr
i
nciple
a
naly
sis
of
IEE
E
802.1
1
Distr
ibu
te
d
Co
ordin
at
ion
F
unct
ion
(D
CF
)
was
int
r
oduce
d
by
Bi
anch
i
[
6].
Bi
anch
i
propo
sed
2
-
D
M
a
r
kov
C
hain
m
od
el
t
o
a
naly
ze
the
perf
orm
ance
of
M
AC
DC
F
m
echan
ism
by
com
pu
ti
ng
t
he
thr
oughput,
assum
ing
sat
urat
ed
traf
fic
an
d
er
ror
f
ree
c
ha
nn
el
.
Seve
ral
works
su
c
h
as
[7
-
9]
f
ollow
e
d
Bi
anc
hi’s
m
od
el
by
analy
zi
ng
the
t
hro
ughput
an
d
delay
of
IEE
E
802.1
1
DCF
under
sat
ur
at
ed
tra
ff
i
c
ta
kin
g
into
c
on
si
der
at
io
n
f
r
a
m
e
retry
l
i
m
i
t
s,
error
-
pr
on
e
channels
an
d
f
reezin
g
of
the
back
-
off
tim
er.
Un
li
ke
sat
ur
at
e
d
tr
aff
ic
,
the
anal
yt
ic
al
m
od
el
fo
r
the
th
rou
ghpu
t
an
d
delay
of
IE
EE
802.1
1
DC
F
perform
ance
un
de
r
no
n
-
sat
urat
ed
tr
aff
ic
wa
s
stud
ie
d
by
[
10
-
11
]
.
F
or
th
e
broad
cast
an
al
ysi
s
in
VA
N
ETs
,
sever
al
wor
ks
su
c
h
as
[
12
-
14]
hav
e
e
valuat
ed
the
sat
ur
at
i
on
pe
rfor
m
ance
of
sa
fety
m
e
ssages
broa
dca
st
in
VANETs
to
c
al
culat
e
the
t
hro
ughput
a
nd
delay
f
or
e
m
erg
ency
a
nd
r
ou
ti
ne
m
ess
ages.
Dif
fer
e
nt
fro
m
broa
dcast
analy
sis
in
VA
NE
Ts,
the
un
ic
ast
per
f
or
m
ance
analy
sis
of
IEE
E
802.11p
in
the
pr
e
sence
of
hid
de
n
te
rm
inals
unde
r
both
sat
ur
a
te
d
an
d
unsat
ur
at
e
d
traf
fics
was
prese
nted
in
[
15]
.
Th
e
analy
ti
cal
mo
del
of
VANETs
incl
ud
i
ng
bo
t
h
broad
ca
st
and
unic
ast
analy
sis
fo
r
sa
fety
and
ser
vice
app
l
ic
at
ion
s,
res
pe
ct
ively
,
base
d
on
Ma
r
kov
c
hain
m
od
el
we
re
int
rodu
ce
d
i
n
[
16]
.
The
a
utho
rs
i
n
[
16]
analy
ze
d
the
pe
rfor
m
ance
of
IEEE
80
2.11p
base
d
MAC
unde
r
no
n
-
sat
urat
ed
traff
ic
an
d
erro
r
fr
ee
c
ha
nn
el
s
.
The
pe
rfor
m
ance
m
etr
ic
s
of
delay
,
pac
ket
de
li
ver
y
rati
o
a
nd
syst
em
thro
ughput
were
in
vestigat
ed
in
[
16
]
.
The
aut
hors
in
[
17
-
18
]
offe
re
d
the
analy
ti
cal
stud
y
of
the
IE
EE
16
09.4
M
AC
in
V
ANE
Ts
un
der
no
n
-
sat
ur
at
ed
c
ondi
ti
on
.
T
he
pe
rfor
m
ance
m
et
rics
of
delay
,
pack
et
deliv
ery
rati
o
and
s
yst
e
m
through
pu
t
we
re
stu
died
in
[17
-
18]
.
Howe
ver,
fr
ee
zi
ng
of
the b
ac
k
-
off
ti
m
er and
er
r
or
-
pro
ne
c
hannels
w
e
re
no
t t
a
ken into c
onside
ra
ti
on
in
[1
7].
Our
m
od
el
is
e
xtensi
on
of
the
existi
ng
m
od
el
in
[1
7].
Fr
eez
ing
of
the
bac
k
-
off
ti
m
er
with
the
M/
M/
1
qu
e
ue
an
d
e
rro
r
-
pro
ne
c
hanne
ls
are
ta
ken
int
o
c
onside
red
in
our
m
od
el
.
Takin
g
t
hese
e
lem
ents
into
a
ccoun
t
will
pr
ovide
a
n
accu
rate
est
im
at
ion
of
acce
ss
to
the
cha
nn
el
and
al
so
a
void
the
overesti
m
at
ion
of
the
s
yst
e
m
thr
oughput.
H
ow
e
ve
r,
t
his
pa
per
f
ocuses
on
a
naly
zi
ng
tw
o
ty
pes
of
tra
f
fics;
safety
a
nd
se
rv
ic
e
t
raffic
s
with
higher
a
nd
lo
wer
pr
i
or
it
ie
s
resp
ect
ively
.
1
-
D
a
nd
2
-
D
Ma
rkov
c
hain
are
em
plo
ye
d
to
m
od
el
the
back
-
off
proce
dures
f
or
each
traff
ic
ty
pe
in
the
pr
e
sence
of
erro
r
-
pro
ne
cha
nne
ls
und
er
no
n
-
s
at
ur
at
ed
c
ondi
ti
on
s
.
Gau
s
sia
n
wirel
ess
er
ror
c
ha
nnel
is
ad
op
te
d
i
n
this
m
od
el
,
in
wh
ic
h
a
c
ons
ta
nt
cha
nnel
bi
t
error
rate
(BER)
is
su
pp
os
e
d
to
be
identifie
d
in
a
dv
a
nce
a
nd
ea
ch
bit
has
t
he
s
a
m
e
bit
error
pro
ba
bili
ty
.
The
perf
or
m
ance
m
et
ric
PD
R,
av
e
ra
ge
delay
of
safety
app
li
cat
ion
s
a
nd
syst
em
through
pu
t
of
ser
vi
ce
app
li
cat
ions
are
inv
est
igat
ed
in
this pa
per to
ev
al
uate the
perf
or
m
ance an
al
y
sis of the
I
EE
E
1609.4.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c &
C
om
p
En
g
IS
S
N: 20
88
-
8708
Mo
deling a
nd
analysis
of I
E
E
E 16
09.4 MAC
in
the
pre
senc
e o
f err
or
-
pro
ne
.
..
(
Akra
m
A
. A
lmoh
amme
di)
3533
2.
ANALYTI
C
A
L MO
DEL
2.1.
Multi
-
priorit
y
tr
an
smissi
on
mod
el
This
analy
ti
cal
m
od
el
con
sid
ers
the
I
EEE
1609.
4
sta
ndar
d
sp
eci
ficat
ion
s
fo
r
the
safety
and
se
rv
i
c
e
a
pp
li
cat
io
ns
as
disp
la
ye
d
in
Fig
ure
1.
T
he
r
e
are
n
ve
hicle
s
in
the
netw
ork
c
onte
ndin
g
to
acce
ss
th
e
c
hannel
base
d
on
the
E
DCA
schem
e.
All
ve
hicle
s
ar
e
in
t
he
tra
ns
m
issi
on
ra
ng
e
of
each
ot
her
a
nd
the
re
a
re
no
hidde
n
te
rm
inals
in
the
syst
e
m
.
Ho
w
ever,
acco
rd
i
ng
to
the
IE
EE
1609.
4
sta
ndar
d,
safety
an
d
WSAs
pac
kets
are
on
ly
transm
itted
over
CC
H
durin
g
CC
HI
as
sh
own
in
Fi
g
ure
1.
I
f
safety
or
WSAs
pac
kets
arr
ive
at
MAC
la
ye
r
durin
g
t
he
SC
HI
at
rate
,
t
hey
ha
ve
t
o
queue
at
M
AC
la
ye
r
buf
fer
wait
ing
f
or
t
he
s
ub
seq
uen
t
CC
HI
to
be
transm
itted.
T
hu
s
,
under
he
avy
traf
fic,
m
any
pac
kets
will
be
que
ue
d
at
MAC
la
ye
r
buf
fer
wa
it
ing
f
or
transm
issi
on
at
the
be
ginnin
g
of
the
ne
xt
C
CHI.
T
his
will
increa
se
c
olli
sio
n
a
nd
delay
of
tran
sm
i
tt
ed
safety
pack
et
s
,
an
d
de
crease
the
PDR
and
acco
rd
i
ng
ly
the
pe
rform
ance
of
V
A
NETs
will
be
degra
ded.
T
he
crit
ic
al
so
luti
on
t
o
ac
hieve
a
reli
a
ble
dissem
inati
on
of
safety
pa
cket
on
the
road
is
t
o
m
it
ig
at
e
the
c
onflic
t
whe
n
acce
ssing
t
h
e
CC
H.
Th
us
,
in
order
to
m
itigate
the
colli
sion
pro
ba
bili
ty
o
ver
CC
H
,
the
con
si
der
e
d
a
pp
l
ic
at
io
n
la
ye
r
has
to
sc
hedule
t
he
generate
d
pac
kets
to
ar
rive
at
M
AC
la
ye
r
with
Po
iss
on
m
ann
e
r
by
delay
in
g
a
tim
e
of
SCH
I
(
50
ms
)
.
Me
a
ning
that
there
will
be
t
w
o
qu
e
ue
s
with
the
sa
m
e
arr
ival
rat
e
duri
ng
the
CC
HI.
The
total
of
tw
o
in
dep
e
ndent
Po
iss
on
proces
s
with
rate
is
2
.
The
re
by,
the
pac
ket
ar
riv
al
rate
fo
r
sa
fet
y
and
WSAs
tra
ff
ic
durin
g
t
he
CC
HI
a
re
de
no
te
d
by
an
d
,
resp
ect
ively
,
a
nd
they
f
ollo
w
the
Po
iss
on
distrib
ution.
Let
(
)
be
the
ra
nd
om
pr
oces
s
re
pr
ese
ntin
g
th
e
back
-
off
ti
m
er
value
(
0
,
1
,
2
,
…
,
−
1
)
at
tim
e
slot
,
w
hile
The
back
-
off
sta
te
process
is
denoted
by
(
).
,
a
nd
are
the
transm
issio
n
fail
ur
e
pro
ba
bi
li
ty
and
t
he
prob
a
bili
ty
of
at
le
ast
on
e
pac
ket
in
the
buff
e
r
for
a
sa
f
et
y
app
li
cat
io
n,
r
e
sp
ect
ively
.
T
he
sta
te
transiti
on
dia
gra
m
o
f
the
1
-
D M
arkov
c
hain for the sa
fety
ap
pli
cat
io
ns
pr
oc
ess is shown i
n
Fig
ure
2. Th
e n
on
-
nu
ll
tra
ns
it
io
n pro
bab
il
it
ie
s are
g
ive
n by:
{
(
|
0
)
=
/
,
0
≤
≤
−
1
(
|
)
=
,
,
1
≤
≤
−
1
(
|
+
1
)
=
1
−
,
,
0
≤
≤
−
2
(1)
Her
e
,
the
non
-
nu
ll
tra
ns
it
ion
pro
bab
il
it
ie
s
de
sc
ribe
the
una
vaila
bili
ty
of
s
afety
pack
et
s
t
ran
sm
issi
on
in the b
uffer,
he
nce c
hangi
ng
the stat
ion i
nto i
dle
(
)
sta
te
after
s
uccess
fu
l t
ra
ns
m
issi
on
is as
foll
ow
s;
{
(
|
0
)
=
1
−
(
|
)
=
1
−
(
|
)
=
/
,
0
≤
≤
−
1
(2)
.
Figure
2. Ma
r
kov chai
n
m
od
e
l of safety
a
pp
l
ic
at
ion
s
Let
,
=
→
∞
P
{
(
)
=
}
f
or
∈
(
0
,
−
1
)
be
the
sta
ti
on
ary
distrib
ution
of
the
1
-
D
Ma
rkov
chai
n,
w
her
e
is
the
con
te
ntion
wi
ndow
of
safet
y
process
.
F
rom
the
Ma
rko
v
chain
m
od
el
in
Fig
ure
2, we
c
an get:
=
(
1
−
)
,
0
+
(
1
−
)
(3)
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
20
19
:
3531
-
35
4
1
3534
=
1
−
,
0
(4)
,
=
−
1
1
−
,
,
0
1
≤
≤
−
1
(5)
Ther
e
f
or
e,
by
us
in
g
th
e
no
r
m
al
iz
ation
c
onditi
on
1
=
+
∑
,
−
1
=
0
f
or
sta
ti
on
a
ry
distri
buti
on
,
we
ca
n deri
ve
,
0
as foll
ows
:
,
0
=
2
(
1
−
,
)
2
(
1
−
,
)
+
(
−
1
)
(6)
Let
be
the
t
ra
ns
m
issi
on
pro
ba
bili
ty
of
sa
fety
app
li
cat
io
ns
that
a
ve
hicle
can
tra
ns
m
it
in
a
ra
ndom
chosen
tim
e slot. T
he v
e
hicle
can o
nly t
ran
s
m
it
w
hen
t
he b
ack
-
off
ti
m
e cou
nte
r
is ze
ro (
,
0
).
=
,
0
=
2
(
1
−
,
)
2
(
1
−
,
)
+
(
−
1
)
(7)
To
a
naly
ze
,
le
t
(
)
an
d
(
)
be
the
ra
ndom
va
riables
repre
sentin
g
the
ba
ck
-
off
sta
ge
(
0
,
1
,
2
,
…
,
)
an
d
the
val
ue
of
the
bac
k
-
off
ti
m
er
(
0
,
1
,
2
,
…
,
,
−
1
)
for
a
gi
ven
sta
ti
on
at
ti
m
e
slo
t
,
resp
ect
ively
.
T
ypic
al
ly
,
the
m
axim
u
m
value
of
the
bac
k
-
off
tim
er
reli
es
on
the
back
-
off
sta
ge;
the
reb
y,
these
rand
om
v
ar
ia
bl
es are
not i
nde
pende
nt.
,
=
{
2
i
,
0
,
≤
′
2
′
,
0
,
>
′
(8)
w
he
re
,
0
is
the
init
ia
l
con
te
ntio
n
wi
ndow
siz
e,
,
0
=
(
+
1
)
,
an
d
′
is
the
m
axi
m
u
m
nu
m
ber
of
tria
ls
befor
e
t
he
pac
ket
is
dro
pped
acc
ord
ing
to
,
′
=
2
′
,
0
=
(
,
+
1
)
.
T
he
va
lue
of
′
is
assum
ed
to
be
5.
T
he
m
axi
m
u
m
value
of
bac
k
-
off
sta
ges
is
de
no
te
d
by
.
T
he
tra
ns
m
issi
on
fail
ur
e
pr
obabili
ty
,
is
const
ant
a
nd
in
dep
e
ndent
in
t
his
a
naly
sis.
S
o,
t
he
t
wo
-
dim
ensio
nal
(
(
)
,
(
)
)
pro
cesses
are
a
naly
zed
her
e
with
a
di
screte
-
tim
e
Ma
rko
v
c
hain
at
w
hich
t
he
c
ha
nn
el
sta
te
ch
ang
e
s,
as
sho
wn
in
Fig
ure
3. The
s
ta
te
o
f
t
his
pro
cess is
denoted
b
y
(
,
).
T
hu
s
, the
non
-
nu
ll
tra
nsi
ti
on
prob
a
bili
ti
es are g
i
ve
n b
y
{
(
,
|
,
+
1
)
=
1
−
,
,
0
≤
≤
,
−
2
,
0
≤
≤
(
,
|
,
)
=
,
,
1
≤
≤
,
−
1
,
0
≤
≤
(
,
|
−
1
,
0
)
=
,
/
,
,
0
≤
≤
,
−
1
,
1
≤
≤
(
0
,
|
,
0
)
=
(
1
−
,
)
/
,
0
,
0
≤
≤
,
0
−
1
,
0
≤
≤
(
0
,
|
,
0
)
=
1
/
,
0
,
0
≤
≤
,
0
−
1
,
(9)
Her
e
,
the
no
n
-
nu
ll
tra
ns
it
ion
pro
bab
il
it
ie
s
descr
ibe
the
una
vaila
bili
ty
of
pa
cket
transm
issi
on
s
i
n
the
buff
e
r w
hich
is
r
e
directed
into
idle sta
te
(
)
after a s
uccess
f
ul tran
sm
issi
on
.
{
(
|
,
0
)
=
(
1
−
,
)
(
1
−
)
,
0
≤
≤
−
1
(
|
,
0
)
=
1
−
(
|
)
=
1
−
(
0
,
|
)
=
/
,
0
,
0
≤
≤
,
0
−
1
(10)
Let
,
,
=
→
∞
{
(
)
=
,
(
)
=
}
be
the
sta
ti
on
ary
di
stribu
ti
on
of
t
he
Ma
r
kov
c
hain
,
wh
e
re
∈
(
0
,
)
,
∈
(
0
,
,
−
1
)
. F
irst,
note that
:
,
−
1
,
0
.
,
=
,
,
0
→
,
,
0
=
,
.
,
0
,
0
0
<
i
≤
m
(11)
,
,
0
=
,
,
−
1
,
0
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c &
C
om
p
En
g
IS
S
N: 20
88
-
8708
Mo
deling a
nd
analysis
of I
E
E
E 16
09.4 MAC
in
the
pre
senc
e o
f err
or
-
pro
ne
.
..
(
Akra
m
A
. A
lmoh
amme
di)
3535
Du
e
to
the
cha
in
re
gu
la
riti
es,
for
eac
h
k
∈
(
1
,
,
−
1
)
,
the
sta
ti
on
ary
distrib
ution
of
idle
an
d
bac
k
-
off
sta
te
s
o
f
servic
e p
ac
kets a
r
e d
e
no
te
d by
and
,
,
an
d
cal
c
ulate
d
as
foll
ows:
,
,
=
,
−
,
(
1
−
,
)
{
(
1
−
,
)
∑
,
,
0
+
,
,
0
+
=
0
−
1
=
0
,
.
,
−
1
,
0
0
<
≤
(12)
o
r
,
,
=
,
−
,
1
(
1
−
,
)
,
,
0
0
≤
≤
,
1
≤
≤
,
−
1
(13)
=
(
1
−
)
(
1
−
)
∑
,
,
0
+
(
1
−
)
,
,
0
+
(
1
−
)
−
1
=
0
(14)
Figure
3. Ma
r
kov chai
n
m
od
e
l of ser
vice a
ppli
cat
ion
s
By
us
in
g
the
norm
al
iz
ation
c
onditi
on
1
=
∑
=
0
∑
,
,
,
−
1
=
0
+
for
st
at
ion
a
ry d
ist
rib
ution,
th
e
n
,
0
,
0
is gi
ven b
y
,
0
,
0
=
{
2
(
1
−
,
)
(
1
−
,
)
(
1
−
2
,
)
£
,
≤
′
2
(
1
−
,
)
(
1
−
,
)
(
1
−
2
,
)
¥
,
>
′
(15)
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
9
, N
o.
5
,
Oct
ober
20
19
:
3531
-
35
4
1
3536
w
he
re:
£
=
(
1
−
2
,
)
(
1
−
2
,
)
(
1
−
,
+
1
)
+
,
0
(
1
−
,
)
(
1
−
(
2
,
)
+
1
)
+
2
(
1
−
,
)
(
1
−
,
)
(
1
−
2
,
)
(
1
−
)
(16)
a
nd:
¥
=
(
1
−
2
,
)
(
1
−
2
,
)
(
1
−
,
+
1
)
+
,
0
(
1
−
)
(
1
−
(
2
,
)
′
+
1
)
+
2
′
,
0
,
′
+
1
(
1
−
,
−
′
)
(
1
−
2
,
)
+
2
(
1
−
,
)
(
1
−
,
)
(
1
−
2
,
)
(17)
Let
be
the
t
ra
ns
m
issi
on
pro
bab
il
it
y
of
ser
vice
ap
plica
ti
ons
t
hat
a
veh
ic
le
can
tra
ns
m
it
a
ser
vice
pack
et
in
a
ra
ndom
ch
os
e
n
ti
m
e slot. Th
e ve
hicle
can
only
tran
sm
it
the pac
ket when
t
he
b
ack
-
off
tim
e
counter
is zero
(
,
i
,
0
) rega
r
dless
of the
ba
ck
-
off
sta
ge.
=
∑
,
,
0
=
0
=
,
0
,
0
1
−
,
+
1
1
−
,
(18)
2.2.
Failure
and c
olli
sion
p
ro
babi
li
ties
The
tra
ns
m
issio
n
fail
ur
e
pro
bab
il
it
ie
s
,
a
nd
,
of
s
afety
and
se
r
vi
ce
pack
et
s
a
r
e
resp
ect
i
vely
der
i
ved as
fo
ll
ow
s:
{
,
=
1
−
(
1
−
,
)
(
1
−
,
)
,
=
1
−
(
1
−
,
)
(
1
−
,
)
(19)
w
he
re
,
an
d
,
de
no
te
t
he pr
oba
bili
ty
o
f
f
ram
e error f
or safety
and se
rv
ic
e
pa
ckets,
res
pecti
vely
.
{
,
=
1
−
(
1
−
,
_
)
,
=
1
−
(
1
−
,
_
)
(
1
−
,
_
)
(
1
−
,
_
)
(20)
wh
e
re
,
_
,
,
_
,
,
_
,
a
nd
,
_
denote
the
F
ram
e
Error
Ra
te
(F
E
Rs)
for
safety
a
nd
WSA/R
FS/AC
K
fr
am
es
err
or
,
resp
ect
iv
el
y.
The
pro
ba
bili
ty
of
these
er
rors
can
be
c
ompu
te
d
from
bit
err
or
pro
bab
il
it
y (i.e
. BER)
as
fo
ll
ow
s
[9]:
{
,
_
=
1
−
(
1
−
)
,
_
=
1
−
(
1
−
)
,
_
=
1
−
(
1
−
)
,
_
=
1
−
(
1
−
)
(21)
w
he
re
,
,
a
nd
repre
sents
the
siz
e
of
safety
and
W
S
A/RF
S/ACK
fr
am
es
res
pecti
vely
.
The pr
obabili
ti
es of c
olli
sion
for
sa
fety
,
an
d serv
ic
e
,
pac
ket
s ar
e
res
pecti
ve
ly
d
efine
d
as
foll
ows:
{
,
=
1
−
(
1
−
)
−
1
(
1
−
)
,
=
1
−
(
1
−
)
−
1
(
1
−
)
(22)
t
hen
,
{
,
=
1
−
(
1
−
)
−
1
(
1
−
)
(
1
−
,
)
,
=
1
−
(
1
−
)
−
1
(
1
−
)
(
1
−
,
)
(23)
Fr
om
(
7)
,
(
18)
,
a
nd
(
23),
we
can
s
olv
e
t
he
tw
o
unknow
n
va
riable
s,
,
,
,
a
nd
,
by
us
i
ng
nu
m
erical
te
chn
iq
ues
in
orde
r
to
cal
culat
e
t
he
tra
ns
m
issi
o
n
an
d
fail
ur
e
pro
bab
il
it
ie
s
fo
r
sa
fety
an
d
serv
ic
e
app
li
cat
io
ns
re
sp
ect
ively
.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c &
C
om
p
En
g
IS
S
N: 20
88
-
8708
Mo
deling a
nd
analysis
of I
E
E
E 16
09.4 MAC
in
the
pre
senc
e o
f err
or
-
pro
ne
.
..
(
Akra
m
A
. A
lmoh
amme
di)
3537
2.3.
Ti
me
analysis
fo
r
safet
y
and
WSA t
r
an
smi
ssion
In
e
ver
y
tim
e
s
lot
durin
g
the
con
te
ntio
n
-
ba
sed
MAC
sche
m
e
(CCH),
the
sta
te
of
the
chan
nel
co
ul
d
be
i
dle
,
,
suc
ces
sfu
l
tra
ns
m
issio
n,
c
olli
sion
tr
ansm
issi
on
or
fail
ur
e
tra
ns
m
i
ssion
du
e
t
o
t
he
fram
e
err
or
.
Th
us
, t
he pr
ob
abili
ti
es o
f
c
ha
nn
el
sta
te
s ar
e
expresse
d b
y (
24).
{
,
=
(
1
−
)
(
1
−
)
,
=
(
1
−
)
−
1
(
1
−
)
(
1
−
,
_
)
,
_
=
(
1
−
)
−
1
(
1
−
)
,
_
,
=
(
1
−
)
−
1
(
1
−
)
(
1
−
,
_
)
(
1
−
,
_
)
(
1
−
,
_
)
,
_
=
(
1
−
)
−
1
(
1
−
)
,
_
,
_
=
(
1
−
)
−
1
(
1
−
)
(
1
−
,
_
)
,
_
,
_
=
(
1
−
)
−
1
(
1
−
)
(
1
−
,
_
)
(
1
−
,
)
,
_
,
=
1
−
,
=
1
−
(
1
−
)
(
1
−
)
,
=
(
1
−
)
(
1
−
(
1
−
)
−
(
1
−
)
−
1
)
,
=
(
1
−
)
(
1
−
(
1
−
)
−
(
1
−
)
−
1
)
,
=
,
−
,
−
,
−
,
−
,
(24)
w
he
re
,
a
nd
,
re
pr
es
ent
t
he
su
cces
sfu
l
tra
ns
m
issio
n
pro
bab
il
it
ie
s
of
sa
fety
an
d
ser
vice
pac
kets
,
resp
ect
ively
.
,
_
,
,
,
,
_
a
nd
,
_
de
no
te
t
he
uns
uccess
fu
l
tra
nsm
issi
on
pro
bab
il
it
ie
s
of
sa
fety
an
d
W
SA
/R
FS/AC
K
pack
et
s
due
t
o
the
fr
am
es
error.
T
he
busy
c
hannel
pro
bab
i
li
ty
is
denoted
by
,
.
Th
e
transm
issi
on
pro
bab
il
it
y
of
a
safety
pac
ket
with
c
olli
sion
cause
d
by
sa
fe
ty
pack
et
s
only
is
d
en
oted
by
,
.
The
tra
ns
m
issio
n
pro
ba
bili
ty
of
a
ser
vice
pa
cket
with
c
olli
sion
ca
us
e
d
by
serv
ic
e
pac
ket
s
on
ly
is
denot
ed
by
,
.
T
he
t
r
ansm
issi
on
prob
a
bili
ty
of
a
pac
ket
with
colli
sion
cau
s
ed
by
safety
and
serv
ic
e
pac
kets
.
Now,
le
t
,
deno
te
s
the
idle
slo
t
durati
on.
,
and
,
rep
re
sent
the
su
ccess
fu
l
tra
ns
m
issi
on
tim
e
of
safety
an
d
se
rv
ic
e
pa
ckets,
res
pecti
vely
.
,
an
d
,
de
no
te
t
he
tra
nsm
issi
on
tim
e
of
safety
an
d
serv
ic
e
pac
ket
s
with
colli
s
ion,
resp
e
ct
ively
.
,
_
,
,
_
,
,
_
,
an
d
,
_
de
no
te
the
un
s
ucces
sf
ul
tr
ansm
issi
on
ti
m
es
of
sa
fety
an
d
WAS/R
FS/
A
CK
pac
kets
du
e
to
f
ram
es
error.
,
,
,
and
are
the
ti
m
e
du
rati
on
t
o
tra
ns
m
it
saf
et
y,
and
WSA
/R
FS/ACK
pa
ckets
res
pecti
ve
ly
.
,
,
and
are
the
ti
m
e
durati
on
of
SI
F
S,
DIFS
a
nd
EI
FS
res
pe
ct
ively
.
is
the
pro
pag
at
io
n
de
la
y.
is
the
idle
tim
e slot. Thus
, th
e t
ran
sm
issio
n t
i
m
e w
it
h
a
doptio
n broa
dc
ast
and
un
ic
ast
m
echan
ism
s is cal
culat
ed by
{
,
=
=
,
=
,
=
,
_
=
+
+
,
=
+
+
+
3
+
2
+
,
=
+
+
,
_
=
+
+
,
_
=
+
+
+
2
+
,
_
=
,
(25)
Howe
ver,
ass
um
ing
that
the
serv
ic
e
data
pa
cket
siz
e
is
c
on
sta
nt,
the
n
t
he
ti
m
e
slot
du
rati
on
to
transm
it
a ser
vi
ce data
pack
et
over
the
SCH
base
d on the c
on
te
ntion
-
f
ree
MAC schem
e is ex
pr
e
ssed
b
y
{
_
,
ℎ
=
ℎ
+
+
+
2
+
+
_
,
ℎ
=
ℎ
+
+
+
_
,
ℎ
=
,
_
(26)
w
he
re
_
,
ℎ
de
no
te
s
the
s
ucc
ess
f
ul
transm
issi
on
t
i
m
e
of
se
rv
ic
e
data
pac
ket
over
SCH
.
_
,
ℎ
is
t
he
un
s
ucces
sf
ul
transm
issi
on
ti
m
e
of
a
se
r
vice
data
pack
et
over
SCH
due
to
fr
am
e
error.
ℎ
=
ℎ
+
ℎ
is
the
pac
ket
head
e
r.
=
/
,
w
here
represe
nts
t
he
payl
oa
d
of
the
se
r
vice
data
pac
ket
,
and
R
is
t
he
transm
issi
on
da
ta
rate.
,
,
a
nd
are
P
HY
-
la
ye
r
de
pende
nt,
an
d
the
f
ram
e
transm
issi
on
s
in
the
un
it
of
Or
t
hogonal
Fr
e
q
ue
ncy
Di
visio
n
Mult
ip
l
exin
g
(
OFDM
)
sym
bo
ls
are
giv
en
by (2
7)
[
19
]
:
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
20
19
:
3531
-
35
4
1
3538
{
=
⌈
+
+
⌉
=
⌈
+
+
⌉
=
⌈
+
+
⌉
=
⌈
+
+
⌉
=
⌈
+
+
⌉
(27)
w
he
re
,
,
,
a
nd
de
no
te
t
he
durati
on
of
a
transm
issi
on
sy
m
bo
l,
OF
D
M
PHY
la
ye
r
serv
ic
e
fiel
d
siz
e,
OFDM
P
HY
la
ye
r
ta
il
fiel
ds
siz
e,
an
d
t
he
nu
m
ber
of
e
nc
oded
bites
pe
r
one
sym
bo
l.
I
n
order
t
o
fig
ur
e
ou
t t
he s
yst
e
m
thr
oug
hput a
nd
delay
, t
he
durati
on
of the log
ic
al
ti
m
e slots
per
sta
te
in
the ch
a
nne
l
is require
d
a
nd
g
ive
n by:
=
,
,
+
,
,
+
,
,
+
,
,
+
,
,
+
,
_
,
_
+
,
_
,
_
+
,
_
,
_
+
,
_
,
_
+
,
ma
x
(
,
,
,
)
(28)
In
t
his
m
od
el
,
the
P
oisson
distr
ib
ution
m
od
el
is
as
su
m
ed,
i
n
w
hich
the
inter
a
rr
i
va
l
tim
e
i
s
expo
nen
ti
al
ly
distrib
uted.
T
he
n,
f
r
om
the
a
ver
a
ge
du
rati
on
of
t
he
lo
gical
tim
e
slot
,
t
he
loa
d
eq
uatio
n
of
qu
e
ue p
roba
bili
ty
,
a
nd
for
sa
fety
an
d ser
vice a
ppli
cat
ion
s
is
g
i
ven res
pecti
vel
y by [
10]
:
{
=
1
−
−
2
=
1
−
−
2
(29)
The
pack
et
del
ivery
rati
o
(PDR)
of
t
he
saf
et
y
app
li
cat
ion
is
der
i
ved
a
s
the
pro
ba
bili
ty
of
ha
ving
a
su
ccess
fu
l
tra
nsm
issi
on
du
rin
g
a
giv
e
n
ti
m
e
slot
over
t
he
a
ver
a
ge
nu
m
ber
of
ve
hicle
s
tra
ns
m
itti
ng
pac
ke
ts
in
a g
e
ner
ic
ti
m
e
slot;
=
,
=
(
1
−
)
−
1
(
1
−
)
(30)
The
a
ver
a
ge
ti
m
e
slot
of
a
s
afety
pac
ket
to
exec
ute
the
ba
ck
-
off
is
giv
e
n
by
(
−
1
)
2
.
T
her
e
by,
the
aver
a
ge
total
s
erv
ic
e
ti
m
e
[
]
of
a
safety
pac
ke
t,
wh
ic
h
e
xperi
ences
the
a
verage
bac
k
-
off
durati
on,
ca
n
be
est
i
m
at
ed
by
=
[
]
=
(
−
1
)
2
∗
(31)
In
t
his
m
od
el
,
each
veh
ic
le
is
m
od
el
le
d
as
a
n
M/
M/
1
que
ue
with
a
n
in
fin
it
ive
buff
e
r
siz
e,
ser
vice
rate
,
a
nd
t
he
pack
et
ar
rival
rate
2
.
In
1609.
4
sta
ndar
d,
th
e
CC
HI
a
nd
S
CHI
ha
ve
the
sam
e
du
rati
on
(50
m
s
fo
r
eac
h
inter
val)
,
the
reb
y,
t
he
ave
ra
ge
ar
rive
d
pac
kets
are
e
qual
.
Howe
ver,
safe
t
y
pack
et
s
w
hi
ch
are
gen
e
rated
d
uri
ng
SC
HI h
a
ve
to d
el
ay
b
y
ℎ
.
Th
eref
or
e
,
t
he
av
erag
e
d
el
ay
o
f
saf
et
y packets
[
]
incl
ud
i
ng
qu
e
uing a
nd tr
ansm
issi
on
d
el
ay
s is expres
se
d by
[
]
=
2
(
1
−
2
)
+
,
+
ℎ
2
(32)
Howe
ver,
the
r
et
ry
lim
it
is
con
side
red
i
n
the
m
od
el
analy
sis
of
the
WSAs
pack
et
s
in
or
de
r
to
m
ee
t
the
I
EEE
802.
11p
sp
eci
ficat
ion
s
.
Th
us
,
the
m
axi
m
u
m
back
-
off
sta
ge
f
or
the
WSA
packet
to
be
tr
ansm
it
te
d
in
this
m
od
el
is
de
no
te
d
by
m
.
If
the
WSA
pac
ket
faces
m
colli
sion
s
in
t
he
pr
e
vious
sta
ge
s,
a
nd
there
f
ore
thi
s
pack
et
will
be
dro
pp
e
d
if
it
exp
e
riences
an
oth
e
r
colli
sio
n.
The
n,
the
W
SA
pack
et
dro
p
pr
ob
a
bili
ty
,
expresse
d by
,
=
,
+
1
(33)
Fo
r
se
rv
ic
e
ap
plica
ti
on
s,
wh
e
n
the
ve
hicle
s
su
ccess
fu
ll
y
ex
change
the
WSA
pac
kets
ov
e
r
CC
H,
they
will
tun
e
to
th
e
sel
ect
ed
SCH
duri
ng
SC
H
I
to
trans
fer
se
rv
ic
e
data.
T
he
m
axi
m
u
m
time
that
veh
ic
le
s
us
e
to
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Ele
c &
C
om
p
En
g
IS
S
N: 20
88
-
8708
Mo
deling a
nd
analysis
of I
E
E
E 16
09.4 MAC
in
the
pre
senc
e o
f err
or
-
pro
ne
.
..
(
Akra
m
A
. A
lmoh
amme
di)
3539
exch
a
nge
se
rv
i
ce
data
is
one
SCHI
(
50
m
s).
In
t
his
m
od
el
,
the
SCH
I
is
di
vid
ed
i
nto
Q
transm
issi
on
slots.
In
it
ia
ll
y,
veh
ic
le
s
hav
e
t
o
e
xc
hange
WSA
pack
et
s
over
CC
H
duri
ng
C
CHI
to
ch
oose
su
it
able
tra
nsm
issi
on
slot
f
or
t
heir
serv
ic
e
data
tr
ansm
issi
on
ov
er
SC
Hs.
T
hu
s,
the
a
ve
rag
e
num
ber
of
WSA
pack
et
s
,
su
ccess
fu
ll
y t
r
ansm
itted over t
he
CC
H
du
rin
g
the
CC
H
I
is
cal
culat
ed by:
,
=
ℎ
×
,
(34)
Accor
ding
to
the
1609.
4
sta
nd
a
r
d,
the
num
ber
of
ser
vice
channel
is
six,
th
us
,
th
e
m
axi
m
u
m
transm
issi
on
s
lots
can
be
ut
il
iz
ed
is
6Q
.
Finall
y,
the
aggre
gate
thr
ough
pu
t
of
t
he
serv
ic
e
pac
ke
ts
,
is evaluate
d by co
ns
ide
rin
g
th
e num
ber
of se
le
ct
ed
tran
sm
is
sion sl
ots,
and
thu
s
, it can
b
e
est
i
m
at
ed
by [17
]
,
=
min
[
,
,
6
]
(35)
3.
MO
DEL V
AL
IDA
TI
ON
We
us
e
M
ATL
AB
to
car
ry
out
the
nu
m
erical
resu
lt
s,
wh
il
e
the
extensi
ve
s
i
m
ulati
on
s
are
cond
ucte
d
to
validat
e
the
pro
po
se
d
a
naly
ti
cal
m
od
el
us
ing
NS
-
2.3
4.
T
he
sim
ulati
on
scenari
o
inclu
de
s
100
ve
hicle
s
with
a
GP
S
a
nd
a
si
ng
le
-
ra
dio
WA
VE
c
omm
un
icati
on
de
vice.
T
he
s
peed
of
ve
h
ic
le
s
is
60
km
/h.
The
value
of
bit
error
rate
f
or
the
cha
nn
el
c
onditi
on
is
as
s
um
ed
to
be
10
−
5
,
wh
ic
h
is
one
of
t
he
m
os
t
aff
ect
ed
a
nd
sensiti
ve
valu
es
f
or
t
he
c
ha
nn
el
BER
i
n
a
com
par
at
ive
ly
no
isy
,
c
hannel
fa
ding
a
nd
unreli
able
wireless
env
i
ro
nm
ent
[
20
-
21
]
.
The
t
ypic
al
par
am
et
ers
values
f
or
bo
t
h
a
naly
ti
cal
m
od
el
an
d
sim
ulati
on
s
are
su
m
m
arized
i
n
Table
1
as
ob
ta
ine
d
f
r
om
[1
7].
T
he
safety
pack
et
arr
ival
rate
is
var
ie
d
up
to
10
0
pa
cket
s/sec
ond
(
pp
s)
,
wh
il
e
the
ser
vice
pa
cket
ar
rival
ra
te
and
t
he
num
ber
of
ve
hi
cl
es
n
are
fix
ed
,
=
50
a
nd
=
30
ℎ
.
Per
form
ance
m
et
rics
su
ch
a
s
aver
a
ge
d
el
ay
and
P
DR
of
safety
pack
et
s,
as w
el
l as t
he n
et
work thro
ughput
of se
rv
ic
e
pack
et
s
are
i
nv
est
igate
d
in
thi
s secti
on.
Table
1.
Pa
ram
et
ers
v
al
ues
Para
m
eters
v
alu
e
Para
m
eters
v
alu
e
Nu
m
b
e
r
o
f
CCH
1
δ
1
µs
Nu
m
b
e
r
o
f
SCHs
6
1
6
bits
Tr
an
s
m
iss
io
n
data
rate
f
o
r
each ch
an
n
el
6
M
b
p
s
6
bits
MAC head
er
2
7
2
bits
10
−
5
PHY head
er
1
2
8
bits
(µs)
4
Service pack
et siz
e
8
0
0
0
bits
24
Saf
ety
pack
et size
8
0
0
bits
8
W
SA
8
0
0
bits
,
0
16
RFS
1
6
0
bits
Sh
o
rt
retry li
m
it
5
ACK
1
1
2
bits
Lon
g
r
etry li
m
it
7
SIFS
1
6
µs
Nu
m
b
e
r
o
f
tr
an
s
m
i
ss
io
n
slo
ts
Q
6
Ti
m
e
slo
t σ
9
µ
s
Frequ
en
cy
5
.9 GHz
DIFS
3
4
µs
10
-
1
0
0
p
p
s
Figure
4
dis
play
s
the
perfor
m
ance
of
th
e
IEEE
16
09.4
MAC
in
V
ANETs
under
vari
ou
s
safety
pack
et
ar
rival
rate
.
In
fa
ct
,
t
he
val
ue
of
c
olli
sion
pro
ba
bili
ty
increases
wi
th
inc
reasin
g
the
pac
ket
a
rr
iv
al
rate
or
t
he
num
ber
of
ve
hicle
s
in
t
he
netw
ork.
The
rea
son
for
t
his
ca
n
be
at
trib
uted
t
o
the
fact
that
in
t
he
heav
y
pac
ket
arr
ival
rate
(tr
aff
ic
loa
d),
m
or
e
broa
dcasti
ng
pac
kets
ar
e
exc
hange
d,
le
ading
to
a
high
e
r
colli
sion
pro
ba
bili
ty
.
In
a
dd
it
ion
,
due
to
t
he
la
ck
of
ACK
and
ex
pone
nt
ia
l
back
-
off
m
echan
ism
s
in
s
afet
y
pack
et
s
br
oad
c
ast
,
veh
ic
le
s
with
broa
dcast
m
od
e
at
tem
pt
s
to
transm
i
t
a
pack
et
after
the
la
st
pack
et
bein
g
broa
dcasted
co
ns
ta
ntly
,
w
hich
resu
lt
s
in
high
er
colli
sion
pr
obabili
ty
.
Accor
ding
to
these
f
act
s,
we
obser
ve
the
PD
R
of
sa
fety
pack
et
s
i
n
Fi
g
ure
4(b)
a
nd
the
thr
ough
pu
t
of
se
r
vice
pa
ckets
in
Fi
g
ure
4(d)
dec
reas
e
with
increasin
g
the
pack
et
a
rr
ival
rat
in
the
ne
tw
ork.
We
al
so
no
ti
ce
in
Fi
g
ure
4(
c
)
that
the
W
S
A
pac
kets
dro
p
pro
bab
il
it
y
sig
nif
ic
antly
inc
r
eases
wit
h
inc
reasin
g
sa
fety
pac
ket
a
rr
iva
l
rate.
Fig
ure
4(
a
)
s
hows
th
at
the
aver
a
ge
delay
of
sa
fety
pac
kets
inc
reases
li
near
ly
with
increa
sin
g
pa
cket
ar
rival
ra
te
.
T
his
obse
rv
at
io
n
exp
la
in
s
t
hat
the
a
ver
a
ge
s
erv
ic
e
tim
e
[
]
=
is
directl
y
propo
rtion
al
to
the
pac
ket
ar
r
ival
rate.
Ex
plici
tl
y,
as
the
pack
et
a
rr
i
va
l
rate
inc
reas
es,
the
pac
kets
queue
an
d
ser
vice
tim
e
incre
ase
as
well
an
d
this
certai
nly
le
ad
to
longer
delay
fo
r
pa
ckets
to
be
tran
sm
i
tted.
I
n
ad
diti
on,
with
m
or
e
safety
pa
ckets
bein
g
broa
dcasted,
t
he
bac
k
-
off
ti
m
e
occu
rs
m
or
e
fr
e
quently
and
t
he
overal
l
back
-
off
ti
m
e
increases
f
or
each
veh
ic
le
i
n
the
net
work due t
o m
or
e fr
e
qu
e
nt
fr
eezi
ng
of the
back
-
off
ti
m
er.
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
20
19
:
3531
-
35
4
1
3540
Figure
4. Per
f
orm
ance an
al
ysi
s of the
I
E
EE
1609.
4
,
=
50
a
nd
=
30
ℎ
Figure
4
cl
ea
rly
sh
ows
that
the
res
ults
of
th
e
propose
d
m
od
el
outpe
rfor
m
the
res
ults
of
the
existi
ng
m
od
el
,
al
tho
ug
h
the
BER
is
consi
der
e
d
in
t
he
pro
posed
m
od
el
.
T
his
is
due
to
the
f
reez
ing
of
the
bac
k
-
off
tim
er
m
echan
ism
th
at
is
ta
ken
into
co
ns
i
derat
ion
in
the
pr
opos
e
d
m
od
el
.
This
m
echan
i
sm
al
ways
keep
s
the
veh
ic
le
s aw
a
re
o
f
the c
hannel
stat
us
to
app
ly
the f
reezin
g of t
he
back
-
off
ti
m
er w
he
n
ther
e is a colli
sion
in
the
channel
to
re
duce
the
colli
sion
pro
bab
il
it
y,
especial
ly
when
the
pac
ket
arr
ival
rate
inc
reases
in
the
ne
twork
.
In
a
dd
it
io
n,
th
e
si
m
ulati
on
resu
lt
s
in
Figure
4
are
cl
os
e
to
the
analy
ti
cal
resu
lt
s,
w
hich
va
li
date
the
accuracy
of the
pro
pose
d
m
od
el
.
4.
CONCL
US
I
O
N
This
pa
per
ha
s
propose
d
an
al
yt
ic
al
m
od
el
s
based
on
1
-
D
an
d
2
-
D
M
arko
v
chain
for
safety
an
d
serv
ic
e
a
ppli
cat
ion
s,
res
pecti
ve
ly
,
to
eval
uate
the
pe
rfo
rm
ance
of
the
IEEE
1609.
4
M
AC
in
in
t
he
pr
ese
nc
e
of
error
pron
e
c
ha
nn
el
s
.
The
r
esult
disp
la
ys
the
eff
ect
of
th
e
pack
et
ar
riva
l
rate
on
the
ne
twork
perfor
m
ance
.
The
net
work
pe
rfor
m
ance
de
gr
a
des
with
i
nc
reasin
g
the
sa
fety
pack
et
arri
val
rate.
The
s
tud
y
al
so
sho
w
s
that
the
res
ults
of
the
pr
opos
e
d
m
od
el
ou
t
perf
orm
the
resu
lt
s
of
the
e
xisti
ng
m
od
el
in
te
rm
s
of
a
ve
rag
e
del
ay
an
d
PD
R
of safety
pack
et
s
over
C
CH,
a
nd
ne
tw
ork
thro
ughput
of ser
vice
packet
s over
SCHs
.
REFERE
NCE
S
[1]
“
IEE
E
Stand
ard
for
W
ire
le
ss
Acc
ess
in
Veh
ic
u
la
r
Env
ironment
s
(W
AV
E)
-
Mult
i
-
cha
nn
el
Oper
a
ti
on,
”
IE
EE
Std
1609.
4
-
2010,
(R
evi
sion o
f
IE
EE
Std 1609.
4
-
2006
),
2011
.
[2]
A.
A.
Alm
ohamm
edi
,
e
t
a
l
.
,
“
An
Adapti
v
e
Mult
i
-
Channe
l
As
signm
ent
and
Coord
ina
ti
on
Scheme
for
IEEE
802
.
11
P/1609.4
in
Vehi
cul
ar
Ad
-
Hoc
N
et
works
,
”
IEEE Access
,
vol
.
6
,
p
p.
2781
-
2802
,
2
018.
[3]
A.
A.
Alm
oham
m
edi
,
et
al
.
,
“
An
ac
cur
at
e
p
erf
o
rm
anc
e
anal
y
s
is
of
h
y
brid
eff
i
cient
and
r
el
i
abl
e
MA
C
protoc
ol
i
n
VA
NET
under
non
-
satura
t
ed
co
ndit
ions,
”
Inte
rn
ati
onal
Journal
of
El
e
ct
ri
cal
an
d
Computer
Engi
nee
ring
,
vol.
7,
p
p
.
999
,
2017
.
[4]
W
.
A.
Jabba
r,
et al
.
,
“
MEQS
A
-
O
LSRv2:
A Mult
ic
riteria
-
B
ase
d
Hy
brid
Mult
ipa
th P
rotoc
ol
for
Energ
y
-
E
fficie
nt
an
d
QoS
-
Aw
are
Data
Rou
ti
ng
in
M
AN
ET
-
W
SN
Co
nver
gence
Scen
ari
os
of
IoT
,
”
I
EE
E
Acce
ss
,
vo
l.
6
,
pp.
76546
-
76572,
2018
.
[5]
W
.
A.
Jabba
r,
e
t
al
.
,
“
Design
a
nd
Im
ple
m
ent
at
i
on
of
IoT
-
Base
d
A
utomati
on
Sy
stem
for
Sm
art
Hom
e
,”
in
201
8
Inte
rnational
Sy
mpos
ium
on
Networ
ks,
Compute
rs
and
Comm
uni
cat
ions (
ISNCC)
,
pp
.
1
-
6
,
2018
.
[6]
G.
Bia
n
chi
,
“
Perform
anc
e
an
aly
s
is of
the
IEEE
8
02.
11
dis
tri
bu
te
d
coor
dination
fu
nct
ion
,
”
I
EE
E
Jo
urnal
on
select
e
d
areas i
n
commu
nic
ati
ons
,
vol
.
1
8,
pp
.
535
-
547
,
2000.
[7]
H.
W
u,
et
al
.
,
“
Perform
anc
e
of
rel
i
abl
e
tra
n
sport
protoc
o
l
over
IEEE
802
.
11
wire
l
ess
LAN:
ana
l
y
s
is
a
nd
enha
nc
ement,
”
in
INFOCOM
2002.
Tw
ent
y
-
F
irst
Annual
Jo
int
Confe
ren
ce
of
the
IEEE
Computer
and
Comm
unic
ati
ons Soc
i
et
i
es.
Proceedi
ngs.
I
EEE
,
pp
.
599
-
607
,
2002
.
[8]
F.
Dane
shgar
an,
et
al
.
,
“
Saturati
on
throughput
a
naly
s
is
of
I
EEE
802.
11
in
th
e
pr
ese
nce
of
non
id
ea
l
tra
nsm
ission
cha
nne
l and
c
apture
ef
fects,
”
IE
E
E
transact
ions o
n
Comm
unic
ati
o
ns,
vol. 56, 2008
.
Evaluation Warning : The document was created with Spire.PDF for Python.