TELKOM
NIKA
, Vol.14, No
.1, March 2
0
1
6
, pp. 72~8
1
ISSN: 1693-6
930,
accredited
A
by DIKTI, De
cree No: 58/DIK
T
I/Kep/2013
DOI
:
10.12928/TELKOMNIKA.v14i1.1355
72
Re
cei
v
ed
No
vem
ber 8, 20
15; Re
vised
Febr
uary 7, 2
016; Accepte
d
February 2
2
, 2016
Improving the Channel Utilization of Basic Safety
Messag
e in VANETs with Superposition Coded
Modulation
Xiang Li
1
, Chao Wang
1
*, Fuqiang Liu
1
, Xiaobo Yin
2
1
School of Elec
tronics an
d Informatio
n
Engi
n
eeri
ng, T
ongji Univers
i
t
y
, Sh
a
ngh
ai, 20
180
4,
Chin
a
2
Colle
ge of Ec
onom
ics and F
i
nanc
e, Hua
q
ia
o Univ
ersit
y
, Q
uanz
ho
u, F
u
jia
n, 3620
21, Ch
i
n
a
*Corres
p
o
ndi
n
g
author, em
ail
:
lixi
a
n
g_2
77
@
163.com
A
b
st
r
a
ct
In this pa
per, w
e
pro
pose
a br
oadc
ast sche
m
e to effectively
utili
z
e
t
he scar
c
e an
d shar
ed
w
i
reles
s
me
di
um for v
ehic
u
lar a
d
-ho
c
netw
o
rks (VANET
s).
By u
s
ing su
perp
o
si
tion cod
ed
modu
latio
n
(SC
M),
infor
m
ati
on e
l
e
m
e
n
ts that co
mpr
i
se
b
a
sic s
a
fety mess
ag
e
s
(BSMs) w
e
re deliv
ere
d
w
i
th different servi
c
e
qua
lities
deter
mi
ne
d by re
al-
t
ime traffic s
i
tu
ations.
T
h
e
op
tima
l p
o
w
e
r all
o
catio
n
strateg
y
and
ach
i
eva
b
l
e
perfor
m
ance gain of the proposed
me
thod
were first theoretically
anal
y
z
e
d
. To apply the proposed
method
into practice, the
h
i
erarc
h
ic
al qua
dratur
e
a
m
pl
it
ud
e
modu
latio
n
tec
h
nol
ogy w
a
s t
hen
e
m
p
l
oy
ed
to
imple
m
ent the
prop
osed SCM
-
base
d
broa
dc
ast scheme for
VANET
s. T
o
eval
uate the p
e
rf
ormanc
e of the
prop
osed
met
hod
in
re
al-ti
m
e traffic e
n
viro
nments,
a joint traffic-comm
unicat
i
on s
i
mul
a
tion w
a
s furt
h
e
r
cond
ucted. Re
sults agre
e
th
at t
he prop
ose
d
method
exte
nds the co
v
e
r
age of the BS
M broadc
ast whil
e
ma
inta
ini
ng a
n
accepta
b
l
e
co
mmu
n
icati
on r
e
lia
bi
lit
y to me
et the requir
e
ment of drivin
g safety.
Ke
y
w
ords
:
Vehic
u
lar
ad
hoc netw
o
rks,
Superi
m
pos
e
d
c
ode
d mo
d
u
lati
on,
Hi
erar
chical qu
adrat
ure
amplit
ude mo
d
u
lati
on
Copy
right
©
2016 Un
ive
r
sita
s Ah
mad
Dah
l
an
. All rig
h
t
s r
ese
rved
.
1. Introduc
tion
As a
promi
s
i
ng
wirel
e
ss communi
catio
n
technol
ogy
for futu
re
int
e
lligent tran
sportation
sy
st
em
s,
v
ehicul
a
r ad
-ho
c
net
w
o
r
ks
(
V
A
N
E
T
s
)
[1] have attracted signifi
ca
nt attention. By
providin
g hi
g
h
spee
d an
d l
o
w-l
a
ten
c
y d
a
ta lin
ks amo
ng vehi
cle
s
,
VANETs
ena
ble a
broad
range
of appli
c
atio
n
s
to
enh
an
ce
driving
safet
y
and t
r
an
sp
ortation
effici
ency [2]. VA
NETs have
b
een
standardi
zed
by IEEE 802.11p [3]. Ho
wever, given t
he high
mobi
lity
of vehicles and the tim
e
-
varying req
u
irements of qu
ality-
of-service (QoS), 80
2
.
11p-b
a
sed
vehicular com
m
unication
s
still
suffer f
r
om
a
few
ope
n challen
ges.
O
ne of th
e m
o
st imp
o
rta
n
t pro
b
lem
s
i
s
mediu
m
a
ccess
con
g
e
s
tion [4
]. To en
able
versatile a
ppli
c
ations,
movin
g
vehi
cle
s
in
VANETs
nee
d to p
e
ri
odica
lly
broa
dcast th
e
i
r o
w
n
statu
s
informatio
n, i.e., the
so-call
ed b
a
si
c
safe
ty message
(BSM), whi
c
h
is
comp
osed of
several d
a
ta
elements
(DEs), e.g
., position, accel
e
ration, hea
di
ng, and velo
city.
The BSM p
a
c
ket ha
s a
typical l
ength
of 300
–400
by
tes, is disseminated
at
a data
rate
o
f
6
Mbps, h
a
s
si
ngle-hop
cov
e
rag
e
of 30
0
–500 m, a
n
d
has a
me
ssage rate of 1
–10 Hz [5]. T
h
e
band
width of
the sha
r
e
d
wi
rele
ss m
ediu
m
is lim
ited,
and gu
arante
e
ing the requi
red d
e
livery rate
and tra
n
smi
s
sion d
e
lay of the BSM dissemination of
all vehicle
s
is difficult, particula
rly in den
se
traffic scen
ari
o
s. Thi
s
prob
lem sig
n
ifica
n
tly redu
ce
s the validity of VANETs in ai
ding d
r
iving. In
the cu
rrent p
aper,
we
aim
to red
u
ce the
band
widt
h
re
quire
ment of
BSM broa
dca
s
t by imp
r
ovi
ng
cha
nnel utilization.
Several stu
d
i
e
s have a
ddressed thi
s
problem
in the
context of VANETs a
nd ha
ve used
media
-
a
c
cess-cont
rol (MA
C
) meth
od
s
and info
rmati
on co
mpressi
on metho
d
s i
n
the appli
c
at
ion
layer. On
on
e han
d, give
n that the
ran
dom
a
c
ce
ss
scheme
empl
oyed by the
MAC laye
r of
the
IEEE 802.11p standard i
s
the mai
n
inducement of packet colli
sions and m
edium access
con
g
e
s
tion [6
, 7], numero
u
s stu
d
ie
s h
a
ve focu
sed
on improving
chan
nel utili
zation by no
vel
acce
ss
sche
mes [4]. In [8], instead of
the rand
om
media a
c
ce
ss, a time-slot
t
ed-ba
se
d M
A
C
scheme
was
prop
osed to
redu
ce p
a
cket
colli
sion
s.
B
y
dividing the
entire
se
rvice pe
riod i
n
to
the
colli
sion
dete
c
tion
and
colli
sion
avoid
a
n
c
e ph
ases, a content-b
ased MA
C scheme
wa
s propo
se
d
in [9] to improve the
reliability of BSM delivery.
On the other hand,
consid
eri
ng t
hat the BSM
is
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Im
proving the
Chan
nel Utili
zation of Basi
c Safety Me
ssag
e in VANETs with…
(Xiang Li)
73
comp
osed of
seve
ral in
dividual
DEs
an
d that di
ffere
nt appli
c
ation
s
would
broa
dca
s
t the
sa
me
DEs, several method
s were
propo
se
d
to
co
mpress
t
he data
to tra
n
smit in th
e a
pplication lay
e
r.
The me
ssag
e com
p
o
s
itio
n and
codin
g
sch
eme in [
10] red
u
ced the ban
dwi
d
th
requi
rem
ent
b
y
mergi
ng the
packet
s
broa
dca
s
ted in
different
VANE
T
appli
c
ation
s
and
by eliminating redun
dant
DEs. On the
basi
s
of the analysi
s
of
real-t
ime tra
ffic environm
ents o
r
ch
an
nel statu
s
, the
adaptive b
r
o
adcast
rate
control m
e
tho
d
s p
r
op
osed
in [11, 12] im
proved
ch
an
nel utilization
by
redu
cin
g
the
freque
ncy of
BSM dissemi
nation.
Mo
re
over, ada
ptive po
wer
co
ntrol
contri
bute
d
a
cla
s
s of u
s
ef
ul metho
d
s [
13] to in
crea
se th
e reu
s
a
b
ility of band
width
re
sou
r
ce by ad
aptively
adju
s
ting th
e
cove
rag
e
of
sin
g
le-hop
transmi
ss
ion, and network codi
ng wa
s a
l
so employe
d
to
improve the
reliability of BSM dissemi
nation
[14-16
]. Although significa
nt improveme
n
ts o
n
cha
nnel utilization
have be
en
a
c
hiev
e
d
by the aforem
entione
d works,
the
differe
nce
s
in th
e Q
o
S
requi
rem
ents of ea
ch
DE
of a BSM h
a
v
e not be
en
con
s
id
ere
d
a
nd
can
be fu
rther e
nha
nce
d
by
applying a
d
vanced mod
u
l
a
tion and
cod
i
ng sche
mes
to
the physical layer of tra
d
itional VANETs.
To the
be
st
of our kno
w
l
edge, thi
s
i
ssue h
a
s
not
b
een
rep
o
rte
d
in the
conte
x
t of VANETs by
previou
s
works.
In this pape
r,
we propo
se
a novel sche
me for BSM broa
dcast in
VANETs to improve
cha
nnel
utilization. By usi
n
g
supe
rpo
s
iti
on
cod
ed m
o
dulation
(SCM) [17], the
p
r
opo
se
d met
hod
extends the
coverag
e
of BSM while mai
n
taining a
c
ce
ptable comm
unication reli
ability for driving
safety appli
c
ations. Con
s
i
derin
g the real-time tr
affi
c situatio
n, the pro
p
o
s
ed
method is f
i
rst
theoreti
c
ally
optimize
d
an
d the ac
hieva
b
le pe
rform
a
nce g
a
in i
s
a
nalyze
d
. To a
pply the pro
p
o
se
d
method i
n
to p
r
acti
ce, hi
erarchi
c
al
quad
ra
ture a
m
p
litud
e mod
u
lation
(HQAM) [1
8]
is em
ployed
to
impleme
n
t ou
r SCM
-
b
a
sed
bro
a
d
c
a
s
t scheme. Altho
u
gh
HQAM h
a
s
b
een
used
in digital vid
e
o
broa
dcastin
g
[19], its u
s
a
ge in
VANE
T
context is
different from
the fo
rme
r
becau
se
of the
relatively sm
all si
ngle
-
ho
p-tra
n
smi
s
sio
n
covera
ge
and time
-varying QoS
re
quire
ment
s.
The
results
given
by a joint traf
fic-commu
nication si
mulati
on in
real
-tim
e traffic e
n
vironment
s a
g
ree
that the
ch
a
nnel
utilizatio
n of tradition
al VANET
s
can b
e
signifi
cantly im
prov
ed by
u
s
ing
the
prop
osed met
hod.
The re
maind
e
r of this pa
p
e
r is o
r
ga
nized as follo
ws. Section 2 d
e
scrib
e
s the
probl
em
discu
s
sed in
this p
ape
r.
Section
3 p
r
opo
se
s ou
r
SCM-b
a
sed
BSM broa
dcast sch
e
me
and
analyzes the
achi
evable
p
e
rform
a
n
c
e
g
a
in. Sectio
n 4
sho
w
s
the perfo
rman
ce
evaluation
of our
method by joi
n
t traffic-com
m
unication si
mulation. Section 5 con
c
lu
des.
2. Descrip
tio
n
of the Pro
b
lem
Different
fro
m
othe
r
wirel
e
ss
comm
uni
cation
net
works,
every
DE
of a
BSM p
a
cket in
VANETs ha
s its own spe
c
ific QoS req
u
irem
ent, wh
i
c
h is
signifi
cantly determi
ned by real
-ti
m
e
traffic scena
ri
os. Figu
re 1
sho
w
s three
typical scen
a
r
ios of vehi
cular commu
n
i
cation
s. Given
that all th
e n
e
i
ghbo
ring
veh
i
cle
s
of
vehi
cle v1
are
di
sta
n
t eno
ugh,
broad
ca
sting
DEs that i
ndi
ca
te
a sm
all adj
u
s
tment o
n
its statu
s
(e.g., its a
c
cele
ration an
d the
status of
its
sign
al light
s),
is
unne
ce
ssary.
Other DE
s, su
ch a
s
its position
an
d velocity, shoul
d be tran
smitted with every
effort to e
n
a
b
le b
o
th e
a
rl
y alerting
an
d othe
r
appli
c
ation
s
. T
h
e
opp
osite
sit
uation
occu
rs to
vehicle
v2, wherei
n all
DE
s
sho
u
ld b
e
delivere
d
with a
high
reli
a
b
ility and lo
w delay to
pe
rform
colli
sion avoi
ding. Fo
r veh
i
cle v3, seve
ral DEs
(such
as the p
o
siti
on and vel
o
city) in its BSM
packet shoul
d be re
ceive
d
as reli
ably as possible
by
all neighb
ors within its cove
rage; othe
r DEs
only need to
be delivered t
o
its nea
rby n
e
ighb
ors. In
other word
s, from the view
of aiding d
r
iving
safety, the QoS req
u
ire
m
ent of each DE is
st
ro
ngly determi
ned by the
distrib
u
tion o
f
the
broa
dcaste
r’
s neig
hbo
rs.
Ho
wever,
all
DEs that
co
mpri
se
a BS
M are d
e
livered
with the
same
QoS in tradi
tional schem
es de
spite t
heir
spe
c
if
ic requi
reme
nts. This
situa
t
ion re
sults i
n
inefficient cha
nnel utilization.
Specifying a
different QoS
for each
DE
of a
BSM is
theoretically feasible but
is not
feasibl
e
in
p
r
actice b
e
cau
s
e
of the
hig
h
implem
entati
on
com
p
lexity. Acco
rdin
g
to their expe
cted
coverage,
all
DE
s that
co
mpri
se
a BS
M can
be
div
i
ded i
n
to t
w
o
se
gme
n
ts:
(1) Se
g-1: the
DE
segm
ent th
at
is o
n
ly e
s
sent
ial to
neigh
bo
rs withi
n
a relatively s
m
all
s
a
fety
-c
r
i
tical c
o
verage
(CC
)
durin
g
colli
si
on avoid
a
n
c
e and
in
clud
es th
e a
c
cel
e
ration
an
d t
u
rn
sig
nal; (2) Se
g-2: th
e DE
segm
ent that
is requi
red
by all
vehicl
es
within the
bro
adcaste
r’s full cove
rage
(F
C)
and
prima
r
ily
inclu
d
e
s
the
identificatio
n, po
sition,
and
hea
di
ng.
The
s
e
DE
s are
critical
not only fo
r the
neigh
bors wit
h
in the CC to
avoid colli
sio
n
s, but
also for the nei
ghb
ors
out of the CC to cond
u
c
t
early ale
r
ting
and othe
r a
pplication
s
. More
over, fr
o
m
the view o
f
improving d
r
iving safety,
two
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 14, No. 1, March 2
016 : 72 – 81
74
requi
rem
ents for the
reliab
ility of BSM disseminat
io
n
can
be
define
d
. On
one
ha
nd, to avoid
th
e
con
s
id
ere
d
collision
s
, the
Level-1
req
u
i
r
eme
n
t is def
ined a
s
the reliable d
e
live
r
y of the enti
r
e
BSM to all
n
e
ighb
ors
with
in CC. O
n
th
e othe
r
hand
, conveyin
g t
he Seg
-
2
to
the nei
ghb
ori
ng
vehicle
s
i
n
F
C
a
s
reli
ably
as
po
ssi
ble
e
x
presse
s the
Level-2
requi
reme
nt, whi
c
h is imp
o
rtant
to
enabl
e othe
r appli
c
ation
s
. The Level
-1 requireme
nt must be
gua
ranteed, whe
r
eas the
Leve
l
-2
requi
rem
ent should b
e
met as mu
ch a
s
p
o
ssible.
v1
v2
v3
Figure 1. Typical vehi
cula
r comm
uni
cati
on scen
ario
s
To provid
e
different Qo
S guara
n
tee
s
to the
Seg-1
a
nd
Seg-2
of a B
S
M, SCM
techn
o
logy is employed i
n
this study.
SCM
is a spatial multip
lexing tran
smissi
on sch
e
m
e
whe
r
ein the
BSM symbol to broad
ca
st (den
oted by
s
x
) is a linear superpo
sition i
n
the powe
r
domain of th
e
Seg-1
an
d
Seg-2
sym
bol (de
noted
by
1
x
and
2
x
) with rate
s
1
R
and
2
R
(bits/
s/Hz). T
he BSM symbol at the tr
an
smitter can b
e
expre
s
sed
as follo
ws:
2
12
1
s
x
Px
(
)
Px
,
(1)
Whe
r
e
is
a config
ura
b
le para
m
eter. Without
lo
ss
of gene
rality, the pro
pag
a
t
ion ch
annel
is
assume
d
to b
e
a Rayleig
h
fading cha
n
n
e
l
ch
aracte
rized
by
0
n
h(
d
)
~
C
N
(
,
d
)
, w
h
er
e
d
is
the
distan
ce bet
wee
n
the bro
adcaste
r and
receive
r
an
d
n
is the cha
nnel fading fa
ctor dete
r
min
ed
by the pro
p
a
gation e
n
viro
nments. Afte
r propa
gating
throug
h the
wirel
e
ss
ch
annel, the B
S
M
symbol at the
receive
r
si
de
read
s as foll
ows:
2
12
1
s
x
h(
d
)
P
x
h(
d
)
(
)
P
x
w
,
(2)
Whe
r
e
w
denot
es the
additiv
e
noi
se. By condu
cting
su
ccessive inte
rferen
ce
can
c
ellation (SI
C
)
[20],
Seg-2
(
2
x
) is
de
code
d first
and
then
su
btra
cted fro
m
the
re
ceived
sy
mbol, who
s
e
remai
nde
r i
s
use
d
to
de
co
de
Seg-1
(
1
x
).
For th
e recei
v
ers
clo
s
e
to
the b
r
oa
dca
s
ter,
given th
e
sho
r
t propa
g
a
tion di
stan
ce and
small
cha
nnel
attenuation, the
entire BSM
can be
de
cod
ed.
Otherwise, if the receivers are far from
the broad
ca
ster, they wo
uld be able t
o
decode
Seg-2
even thou
gh i
t
would b
e
dif
f
icult to de
co
de the
w
hole
BSM. Notably, the delivery
ratio of both t
he
BSM within CC an
d
Seg-2
within F
C
are
affected
sign
ificantly by the value of
. By optimizing
the po
we
r
al
locatio
n
a
c
co
rding
to
re
al-time traffic si
tuation, we
expect to
m
eet the
Le
ve
l-2
requi
rem
ent
as fa
r
as th
e b
r
oad
ca
ste
r
coul
d at th
e conditio
n
of gua
rant
eei
ng the
Le
vel
-
1
requi
rem
ent.
Before we p
r
oceed, seve
ral
a
s
sumpti
ons sh
ould
be
ma
de.
Fi
rst,
the radi
u
s
of CC
(den
oted by
CC
r
) is treate
d
to be known. Specifying a
pr
oper
radi
us f
o
r CC i
s
imp
o
rtant for th
e
broa
dcaste
r b
u
t is not the focu
s of this study.
Second,
the broad
ca
ster is a
ssu
me
d to be able to
learn the
po
sition
s of all its neighb
ors within
its
CC, from th
eir previou
s
BSM broad
cast.
More
over, co
nsid
erin
g that the optim
al powe
r
allo
cati
on (spe
cified by
) is affected by the real-
time distrib
u
tion of the nei
ghbo
rs,
we a
s
sume that th
e
Level-1
requirem
ent will
be met provided
that the deliv
ery ratio i
s
g
u
a
rante
ed fo
r the farthe
st n
e
ighb
orin
g ve
hicle
within th
e bro
a
d
c
a
s
te
r’s
CC. W
e
also
let
ma
x
d
be the distan
ce bet
ween the broa
dc
a
s
ter a
nd
su
ch a neig
h
bor. Finally,
con
s
id
erin
g that the Shan
non capa
city is ze
ro
in Rayleigh fading
chann
el, the outage proba
bi
lity
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Im
proving the
Chan
nel Utili
zation of Basi
c Safety Me
ssag
e in VANETs with…
(Xiang Li)
75
is empl
oyed
as a
perfo
rm
ance indi
cato
r to evaluate
wheth
e
r the
Le
v
e
l-
1
or
Le
ve
l-
2
req
u
irem
ent
is sat
i
sf
ied.
3.
Proposed SCM-b
ased Broad
cas
t
Scheme
In this
se
ctio
n, we
prese
n
t our novel
BSM broad
ca
st sch
e
me
ba
sed
on
the SCM
techn
o
logy.
The optim
al
syst
em
parameters a
r
e
theoreti
c
a
lly
derived
an
d the a
c
hiev
able
perfo
rman
ce
gain i
s
the
n
a
nalyze
d
. Mo
reover, by
sp
e
c
ifying that
b
o
th
Seg-1
and
Seg-2
sym
bols
are mod
u
late
d with a 4-Q
A
M sch
eme, the prop
osed
method is carri
ed into practice throu
g
h
a
4/16-HQAM constell
ation.
3.1. Optimal Transmitting Po
w
e
r
Allocation
Firs
t, to meet the
Le
vel-1
requi
rem
ent, the total tran
smitting po
wer of a BSM
must b
e
large e
nou
gh,
rega
rdle
ss m
a
tter wh
ether
the SCM
is u
s
ed o
r
not. F
o
r presentatio
n simpli
city, we
let
21
R
(R)
E
be the minimal si
gn
al-to-noi
se ra
ti
o (SNR) to
achieve th
e
data rate
R
.
Acco
rdi
ng to Shanno
n’s th
eore
m
s,
at a prop
agatio
n distan
ce of
d
,
the outage p
r
obability of a
BSM reads
12
2
12
1
t
(R
R
)
ou
t
s
t
(R
R
)
P
(
d)
P
r
h
(
d)
e
E
E
,
(3)
Whe
r
e
t
is th
e SNR at the transmitter
side, and
2
h(
d
)
is the co
rre
sp
on
ding ch
ann
el
gain.
For the p
u
rp
ose
of meeti
ng the
Le
vel
-
1
requirement, the outage pr
obability of the entire B
S
M
sho
u
ld b
e
bel
ow a
given th
reshold
th
s
P
. Wit
h
in the
CC, g
i
ven that
out
'
t
h
s
s
P(
d
)
P
hol
ds p
r
ovide
d
that
out
'
t
h
s
max
s
P(
d
)
P
, we obtain:
12
12
2
1
tm
i
n
nn
max
m
th
t
x
h
a
ss
(R
R
)
(R
R
)
dl
o
g
(
d
P
)
P
EE
,
(4)
Whe
r
e
2
1
th
th
ss
PP
log
(
)
is ap
plied, whi
c
h
allows u
s
to con
s
id
er that
th
s
P
is typically a small
enou
gh value
,
and
mi
n
is the minimal SNR that should
b
e
satisfie
d at the tran
smitter to enabl
e
safety-related
application
s
.
Secon
d
, for
a
given tra
n
sm
itting power,
an up
pe
r bo
u
nd of
sho
u
ld
be o
b
tained.
Seg-
2
expresse
s t
he same
Qo
S requi
rem
e
n
t
as
Seg-1
i
n
CC but i
s
ex
pecte
d to be
delivere
d
wit
h
a
large
r
cove
ra
ge tha
n
the
latter. He
nce,
the outa
ge
probability
sho
u
ld b
e
smalle
r than
that of
the
latter at the same
ch
ann
el con
d
ition.
For an
SIC receiver, according to E
quation
(2),
the
receiving SNR pair fo
r the
Seg-1
a
nd
Seg-2
sym
bol
s at the distan
ce of
d
read
s as follows:
22
22
12
22
1
1
t
rr
t
t
()
|
h
(
d
)
|
(
,
d)
,
(
,
d
)
|
h
(
d)
|
,
|h
(
d
)
|
.
(5)
With the
sam
e
ch
ann
el ga
in
2
|h
(
d
)
|
, we let
22
1
1
rr
(,
d
)
(
Pr
P
R
)(
,
d
)
R
)
r(
.
Acco
rdi
ngly, we obtai
n
1
12
ma
x
(R
)
(R
R
)
E
E
,
(6)
W
h
er
e
ma
x
is th
e maximal
all
o
wa
ble val
u
e
for th
e p
a
ra
meter
. If no
ne of
the
bro
adcaste
r’s
neigh
bors l
o
cates i
n
it
s
CC, then
the
required
data
rate
1
R
for the
Seg-1
is
z
e
ro, and thus
, we
obtain
0
ma
x
.
More
over,
should b
e
larg
e enou
gh to
guarantee th
e outage p
r
o
bability of the entire
BSM within CC. For the SIC re
ceive
r
, the outage p
r
o
bability of BSM read
s.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 14, No. 1, March 2
016 : 72 – 81
76
22
2
2
1
1
21
ou
t
sr
r
r
Seg
e
rror
S
e
g
e
r
r
o
r
P
(
,
d
)
P
r(
d
)
(
R
)
P
r(
d
)
(
R
)
,
(
d
)
(
R
)
EE
E
.
(7)
Acco
rdi
ng to Equation (4)
and (6
),
out
s
P(
,
d
)
can
be re
written a
s
follows:
11
ou
t
sr
P
(
,d
)
P
r
(
,d
)
(
R
)
E
.
(8)
The
Le
vel
-
1
requireme
nt is met provided
that
out
th
s
s
P(
,
d
)
P
hold
s
. We can then
obtain
1
min
nt
h
ma
x
s
t
(R
)
dP
E
,
(9)
Whi
c
h is the l
o
we
r bou
nda
ry of
.
Given that the smalle
r
is, the larger p
o
we
r the
Seg-2
ca
n be all
o
cate
d, we b
e
lieve
that
*
mi
n
is only the optimal
to meet the
Level
-2
req
u
irem
ent. No
tably, the optimal
config
uratio
n
of
is si
gni
ficantly determined by re
al-t
ime traffic sce
na
rios.
Acco
rdi
ng to
Equation (9),
the smalle
r th
e distan
ce b
e
t
ween t
he b
r
oad
ca
ster an
d its nea
re
st neigh
bor
with
in
its CC, the le
ss tran
smitting power is all
o
cate
d to its
Seg-1
.
3.2. Achiev
a
b
le Performa
nce Gain
In
this su
bse
c
tion, we ana
lyze
the
a
c
hi
evable
gain
o
n
the
delivery
rate
of the
Seg-2
in
FC. With
the
optimal
*
, according to Equation (5), the
maximal deliv
ery probability of the
Seg-
2
in logarith
m
ic form read
s.
01
2
2
2
02
12
2
n
d*
,d
B
tm
i
n
m
i
n
k(
R
R
)
(
R
)
d
P(
d
)
d
B
,
k
l
o
g
e
(R
R
)
(R
)
EE
EE
.
(10)
With tra
d
ition
a
l metho
d
s (without th
e u
s
e
of
SCM
)
, su
ch a
p
r
ob
a
b
ility
can be easily
o
b
tain
ed
from Equatio
n (3), that is,
01
2
2
d
,d
B
n
t
k(
R
R
)
P(
d
)
d
B
d
E
.
(11)
Therefore,
at
the sam
e
chann
el con
d
i
t
ion
an
d S
N
R, the
a
c
hie
v
able g
a
in
o
n
the
delivery
probability of the
Seg-2
ca
n be expre
ssed in loga
rith
mic form a
s
:
12
12
2
0
2
21
2
2
0
n
tm
i
n
d
tt
m
i
n
t
(R
R
)
(R
R
)
(
R
)
(
)
kd
G(
d
)
d
B
(R
)
(
R
R
)
(
R
)
(
)
EE
E
EE
E
.
(12)
We al
so let
22
d*
d
,d
B
,
d
B
P(
d
)
P(
d
)
, and thus, t
he gain o
n
the cove
rag
e
of the
Seg-2
can b
e
achi
eved a
s
:
12
2
2
2
1
tm
i
n
m
i
n
c
n
t
(R
R
)
(
R
)
d
G
(R
)
d
EE
E
.
(13)
Acco
rdi
ng
to Equation (12
)
and (13
)
,
b
o
th
2
d
G(
d
)
and
2
c
G
are affected by
the real
-time
traffic enviro
n
ment. Un
de
r the ca
se
o
f
that
all neighbo
ring veh
i
cle
s
of the broa
dcaste
r
are
outsid
e
of it
s
CC,
co
nsi
deri
ng that th
e
S
eg-1
of the
B
S
M to broad
cast
can
be
ca
ncell
ed from t
he
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Im
proving the
Chan
nel Utili
zation of Basi
c Safety Me
ssag
e in VANETs with…
(Xiang Li)
77
transmitted si
gnal, the m
a
ximal gain
for
both
the
deliv
ery proba
bility and
covera
ge of the
Se
g-
2
can b
e
achie
v
ed, that is:
0
12
21
2
2
2
2
n
d
c
n
,m
a
x
,m
a
x
t
kd
(R
R
)
G(
d
)
(
R
R
)
(
R
)
d
B
,
G
(R
)
.
(14)
3.3. HQAM-b
ased BSM
Broadca
s
t
To apply the
prop
osed m
e
thod into p
r
actice, we p
r
ese
n
t a 4/16
-HQAM-b
a
se
d BSM
broa
dcast
scheme by spe
c
ifying both t
he
Seg-1
and
Seg-2
symb
ols a
s
squa
re
4-QAM
symbols.
The optimal
powe
r
allo
cation betwee
n
the
Seg-1
and
Seg-2
symbols a
n
d
the availa
ble
perfo
rman
ce
gain at su
ch
a schem
e is
analyzed in
succe
ssi
on. Fi
gure
2 sh
ows the con
s
tella
tion
of our m
e
tho
d
, whi
c
h is
o
b
tained
by using the G
r
ay
mappin
g
. Thi
s
con
s
tellatio
n
ca
n be trea
ted
as a line
a
r superpo
sition
of two indep
ende
nt squ
a
re 4-QAM
co
nstellatio
n
s:
L1
and
L2
. The
former i
s
d
edi
cated
to the
d
e
livery of
Seg-
1
,
whe
r
ea
s t
he latte
r i
s
d
e
voted to the
d
e
livery of
Seg-
2
. In the
re
st of this pap
er,
we
assu
me that
all t
he
symbol
s
are
tran
smitt
ed e
qually li
kely.
More
over, we
assume th
at both
Seg-1
and
Seg-2
are
prote
c
ted b
y
erro
r-co
rrecting co
de
s, a
nd
that they can
be decode
d
at the re
ceiv
er provided t
hat their bit error rate (B
ER) is b
e
lo
w the
corre
s
p
ondin
g
threshold
s
,
1
th
b
P
and
2
th
b
P
. The propo
sed sch
e
me is equival
ent to the con
v
entional
squ
a
re
16
-Q
AM wh
en
0
15
/
, and thu
s
,
we o
n
ly co
nsid
er the
ca
se
s wh
en
0
. For
pre
s
entatio
n
simpli
city, we
let
1
4
erfc
2
(,
x
)
/
(
x
/
)
I
and
2
1
, where
erfc(
)
is
the compl
e
m
entary erro
r functio
n
.
L1
L
a
y
e
r,
fo
r
t
h
e
Se
g
-
1
L
2
La
y
e
r
,
f
o
r
t
h
e
Se
g
-
2
00
01
11
10
01
00
10
11
00
0
0
00
1
0
00
0
1
00
1
1
11
0
0
11
0
1
11
1
1
11
1
0
01
0
0
01
1
0
01
0
1
01
1
1
10
1
0
10
0
0
10
1
1
10
0
1
H
Q
A
M
c
o
n
s
t
e
lla
tion
+=
Figure 2. Con
s
tellation
conf
iguratio
n of the HQAM
-ba
s
ed bro
a
d
c
a
s
t in VANETs
To obtain
a prop
er
to maximize the
chann
el utiliza
t
ion, the que
stion rega
rdi
ng ho
w
the BERs of
Seg-1
an
d
S
eg-2
a
r
e
det
ermin
ed
by the p
a
ra
meter
in thi
s
situa
t
ion shoul
d b
e
answe
red first. Following t
he metho
d
pr
opo
sed in [2
1
], the BER pair of the
Seg-1
and
Seg-2
c
an
be obtain
ed a
s
follows:
22
2
1
22
2
22
2
be
ttt
be
tt
P(
,
d
)
,
h
(
d
)
,
h
(
d
)
,
h
(
d
)
P(
,
d
)
,
h
(
d
)
,
h
(
d
)
II
I
II
(
1
5
)
W
h
er
e
0
tb
E
/N
is the
ratio
of the
e
nergy
pe
r bit
to noi
se
po
we
r
spe
c
tral
de
n
s
ity (al
s
o
call
ed
SNR fo
r pre
s
entatio
n sim
p
licity in the
remai
nde
r o
f
this pap
er) at the tran
smitter sid
e
. The
outage p
r
o
b
a
b
ility in this st
udy is defin
e
d
as the
prob
ability that the BER is a
b
o
v
e the pre
defi
ned
threshold. Co
rre
sp
ondi
ngly, the outage p
r
oba
bility pair of
Seg-1
and
Seg-2
reads
as
follows
:
11
1
2
2
2
ou
t
b
e
t
h
out
be
t
h
bb
P
(
,
d
)P
r
P
(
,
d
)
P
,
P
(
,
d
)P
r
P
(
,
d
)
P
.
(16)
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 14, No. 1, March 2
016 : 72 – 81
78
With the
sa
me co
nsi
deration a
s
in
Subse
c
tion
3
.
1,
2
out
P(
,
d
)
shoul
d be far
small
e
r than
1
out
P(
,
d
)
within the CC, and the latter can thu
s
b
e
treated as
an effective approxim
ation of the
outage p
r
ob
a
b
ility of the e
n
tire BSM. That is,
2
1
2
2e
r
f
c
2
1
th
out
b
s
n
t
(P
)
P
(
,
d
)
exp
d
,
(17)
whe
r
e
only t
he le
adin
g
it
em (th
e
first
item)
of the
1
be
P(
,
d
)
is co
ns
id
er
e
d
b
e
c
a
us
e th
e
receiving SNR (
2
t
h(
d
)
) is larg
e e
noug
h within
CC.
To me
et the
Level-1
requi
reme
nt, the o
u
tage
pro
bab
ility of the B
S
M mu
st be
belo
w
a
given thre
sh
old
th
s
P
within CC, which will
be met provided that
out
th
s
max
s
P(
,
d
)
P
. T
hen, we
obtain:
1
1
2
e
rfc
2
th
b
mi
n
nt
h
ma
x
s
t
(P
)
dP
,
(18)
Whe
r
e
2
1
th
t
h
s
s
log
(
P
)
P
is
ap
plied, with
th
e fact that
th
s
P
i
n
con
s
ide
r
ati
on
shoul
d b
e
small
enou
gh. Thu
s
, the optima
l
configu
r
atio
n for
read
s
0
*
mi
n
mi
n
(
,
)
. Compa
r
ed
with the
conve
n
tional
squ
a
re
16
-Q
AM, who
s
e B
E
R can b
e
ob
tained by a
p
p
l
ying
0
to Equa
tion (1
7),
the achi
evabl
e gain on
2
be
P
at the SNR
regi
me
t
read
s:
22
2
2
22
13
5
32
2
55
5
22
tt
t
be
**
*
*
tt
,h
(
d
)
,
h
(
d
)
,
h
(
d
)
G(
d
)
,h
(
d
)
,
h
(
d
)
II
I
II
(19)
In this
section, we propose
an SCM
-
based BSM broadcast
scheme. Both the reliability
and
cove
rag
e
of BSM
de
livery in the
prop
osed
scheme are sig
n
ificantly
imp
r
oved co
mpa
r
ed
with tho
s
e
in
the tra
d
ition
a
l VANET
scheme. A
c
cordingly, the
ch
annel
utilizati
on i
s
e
nhan
ced.
Given that th
e di
stan
ce
b
e
twee
n the
b
r
oad
ca
ste
r
a
nd its ne
are
s
t neigh
bor wi
thin the
CC i
s
signifi
cantly d
e
termin
ed by
the real
-time
traffi
c scen
a
r
io, obtaini
ng
the clo
s
ed f
o
rm exp
r
e
ssi
on
for the avera
ge perfo
rma
n
c
e gai
n of the prop
osed
method in re
al worl
d is di
fficult. Thus, we
analyze the
perfo
rman
ce
of ou
r m
e
th
od in
Se
ction 4 th
ro
ugh
a joint t
r
affic-co
mmuni
cati
on
s
i
mulation.
4. Simulation Resul
t
s an
d Discus
s
io
n
To evaluate
the perfo
rma
n
ce of the propo
s
ed BSM
broad
ca
st schem
e in re
al traffic
scena
rio, a
si
mulation
com
b
ining th
e traf
fic with
com
m
unication syst
em
is co
ndu
cted.
The
trace
data of vehicl
es are gene
rated by
usin
g the traffic simulator SUMO [22]. These data a
r
e then
importe
d into
the
commu
nicatio
n
si
m
u
lation, which is
de
sign
e
d
ba
se
on t
he IT++
sig
nal
pro
c
e
ssi
ng li
bra
r
y [23]. Di
fferent vehicl
e den
sity values
and
CC
radiu
s
a
r
e
co
nsid
ere
d
in o
u
r
simulatio
n
, and the detaile
d config
uratio
n of this simul
a
tion is de
scri
bed in Tabl
e 1.
Figure 3
sho
w
s th
e BER
of
Seg-1
and
Seg-2
at different SNRs an
d
re
gime
s.
A
smal
l
dec
r
eases
Seg-1
delive
r
y ratio
within
CC, but
hel
ps to
imp
r
ov
e the
S
eg-2
coverage. Th
e
receiving S
N
R in
CC i
s
ty
pically la
rge
enou
gh, an
d
thus, the
Le
ve
l-
1
requirem
ent still could be
met by usi
n
g a
small
. For exa
m
ple,
althoug
h th
e BER of th
e entire BSM deteri
o
rates
signifi
cantly with
03
.
, the BER could still
be bel
ow the given
th
s
P
provi
ded that th
e
SNR i
n
the CC is a
b
o
ve 18 dB. Such a pe
rformance is ty
pically accepta
b
le whe
n
an
error-corre
c
ting
cod
e
is
used.
Comp
ari
ng t
h
is p
r
op
osed
met
hod with
the
co
nventi
onal squ
a
re 16-QA
M
sho
w
ed
that the requi
red S
N
R thre
shol
d i
s
redu
ced
from
14
dB to 10
dB,
and
a
signifi
cant
cove
rag
e
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Im
proving the
Chan
nel Utili
zation of Basi
c Safety Me
ssag
e in VANETs with…
(Xiang Li)
79
enha
ncement
on the
Seg-2
is a
c
hi
eved.
More
over, at
the same
S
N
R regime, th
e
delivery
rate
of
Seg-2
i
s
also improve
d
by the pro
p
o
s
ed
method.
Table 1. Simulation config
uration
Traffic
Simulation
Road net
work: gr
id-like netw
o
rk
The numbe
r of cr
ossroads in both directions: 5
The length of h
o
r
i
zontal and vertical streets: 500 m
e
ters
Average densit
y
of vehicles: 1.5 or
3.5 vehicles per 100-mete
r roa
d
Length of simulation time: 1800 seconds
Updating interval
of vehicles lo
cation: 0.1 second
Communication
simulation
SNR at tra
n
smitter:
115
t
dB
Fading factor of t
he channel:
37
6
n.
Length of
Seg
-
1
and
Seg-2
bit
-
str
eam: 256 bits
BER threshold:
12
00
2
th
th
bb
PP
.
O
u
tage proba
bilit
y
threshold:
00
1
th
s
P.
Radius of the CC
:
100
or
15
0
CC
rm
m
Figure 3. BER of the prop
ose
d
method
at different
an
d SNR regim
e
s
(a)
35
,
1
0
0
CC
.r
m
(b)
35
,
1
5
0
CC
.r
m
(c
)
15
,
1
5
0
CC
.r
m
Figure 4. Outage proba
bility of our method with
different traffic den
sity and CC radiu
s
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 16
93-6
930
TELKOM
NIKA
Vol. 14, No. 1, March 2
016 : 72 – 81
80
The a
c
hi
evab
le pe
rform
a
n
c
e g
a
in i
s
al
so ev
al
uated
in ou
r
simul
a
tion. Figu
re
4 sh
ows
the outage
probability
of t
he BSM
and its
Seg-2
wi
th differe
nt traffic de
nsity
values an
d
CC
radiu
s
, whe
r
e
the optimal
is ad
opted. F
o
r a la
rg
e CC radiu
s
, mo
re
tran
smitting
power
sho
u
ld
be assign
ed to
Seg-1
to meet the
Level
-1
req
u
ireme
n
t. Hence, th
e performan
ce gain on
Seg-2
is degraded. Com
p
aring
Figure 4(
a) and Figure 4(b)
illustrates the
followi
ng concl
u
sion.
For
sparse-traffic
scenarios, gi
ven that
the
probability that a broadcas
ter
has no neighbor
withi
n
its
CC i
s
la
rge
a
nd that the
S
eg-1
ca
n be
can
c
ell
ed fro
m
the tran
sm
itted BSM in su
ch a
ca
se,
a
large
pe
rform
ance g
a
in
ca
n be
achieve
d
. Com
p
a
r
in
g
Figu
re 4
(
b
)
and Fi
gu
re 4
(
c) vali
date
s
t
h
is
con
c
lu
sio
n
. More
over, a
sha
r
p in
crea
se in BSM
o
u
tage o
c
curs at the boun
d of the CC.
This
finding i
s
al
so due
to that
the
Seg-1
in
BSMs a
r
e
ca
ncell
ed
whe
n
no n
e
igh
bor
is p
r
e
s
ent
wit
h
in
the broa
dcast
e
r’s
CC.
5. Conclusio
n
In this pap
er,
we propo
se
a novel broa
dca
s
t sche
m
e
to improve
the cha
nnel
utilization
of BSM dissemination
in
VANETs. S
pecifi
c
ally,
from the view of trans
p
ort
a
tion safety, al
l
informatio
n e
l
ements incl
u
ded in
a BSM are
divide
d into two
separated
seg
m
ents. Th
ese
segm
ents a
r
e then co
ncu
rre
ntly
transmitted by using the SCM
techn
o
logy with different QoS
guarantee
s d
e
termin
ed by
the real-time
traffic sce
na
rio. The opti
m
al power all
o
catio
n
strate
gy
that maximizes the
cove
ra
ge an
d delive
r
y rate of
BS
M is theo
reti
cally derived
a
t
the con
d
itio
n o
f
meeting
the requi
rem
ent of
drivin
g sa
fety.
Moreov
er, to
evalua
te ou
r m
e
tho
d
in
re
al traffic
environ
ment
s, we con
d
u
c
t a joint traffic-comm
uni
cati
on sim
u
lation
. The re
sults
of this simul
a
tion
indicate that the pro
p
o
s
ed
method
can
consi
der
ably improve the
coverag
e
and
reliability of BSM
delivery com
pare
d
with t
r
a
d
itional VANET schem
e.
The p
r
op
ose
d
method
is n
o
t cont
radi
cto
r
y to
traditional
scheme
s
th
at i
m
prove
chan
nel utili
zation
in MA
C
or
a
pplication l
a
yer, b
u
t rather the
prop
osed met
hod can be u
s
ed tog
e
ther
with the existi
ng method
s.
Ackn
o
w
l
e
dg
ements
This
wo
rk
was p
a
rtly su
pporte
d by the Key Prog
ram of
Natio
nal Natural
Scien
c
e
Found
ation o
f
China
(NO.
6133
100
9)
and the
Fun
damental
Re
sea
r
ch Fu
nd
s for th
e Ce
ntral
Universitie
s
, Tongji University (NO. 17
0
9219
004
).
Referen
ces
[1]
Karagiannis G, Altintas O,
Ekici E, et al.
Vehicul
a
r Net
w
o
r
ki
ng: A Surve
y
an
d T
u
toria
l
on
Req
u
irem
ents, Architectures,
Cha
lle
ng
es, S
t
andar
ds a
nd
Soluti
ons.
IEE
E
Co
mmu
n
icat
ions S
u
rvey
s
and T
u
tori
als
. 201
1; 13(4): 58
4-61
6.
[2]
Papa
dimitratos
P, La
F
o
rtel
le
A, Evenss
e
n
K,
et al. V
e
h
i
c
u
lar
Comm
uni
cation
S
y
stem
s: Enab
lin
g
T
e
chnolog
ies,
Appl
icatio
ns, a
nd F
u
ture
Outl
ook on Inte
lli
g
ent T
r
ansportation.
IEEE Comm
unications
Maga
z
i
ne
. 2
0
0
9
; 47(11): 8
4
-9
5.
[3]
Amendm
ent to IEEE 802.11:
Wireless Acce
ss in
Vehic
u
lar Enviro
nments.
IEEE Std. 802.11. 201
0.
[4]
W
u
X, Subr
amani
an S, et al. Vehic
u
lar C
o
m
m
unic
a
tions U
s
ing DS
RC: C
hall
e
n
ges, Enh
ancem
ents
,
and Evo
l
uti
on.
IEEE Transactions on Selected
Areas in Comm
unications
.
201
3; 31(9): 39
9-40
8.
[5]
Vehic
l
e S
a
fety C
o
mmu
nic
a
tions—A
ppl
ic
ations VS
C-A
.
F
i
nal R
e
p
o
rt. U.S. Department o
f
T
r
ansportation,
Nation
al Hi
gh
w
a
y
T
r
affic Safet
y
Admi
nistrat
i
on, W
a
shi
ngto
n
, DC, USA. 2011.
[6]
Shuh
aimi
NI, Heria
n
s
y
a
h
H,
Juha
na T
,
et al. Performa
nce An
al
ysis
for Un
iform an
d Bin
o
mia
l
Distributi
on o
n
Contenti
on W
i
nd
o
w
us
i
ng D
S
RC an
d W
i
F
i
Direct Stand
a
r
d.
Internatio
n
a
l Jour
nal
of
Electrical and Co
mp
uter
Engi
neer
ing
(IJECE
)
. 2015; 5(6): 145
2-14
57.
[7]
Kumar J. Br
oa
dcastin
g
T
r
affic Lo
ad
Perform
ance
An
al
ysis
of 80
2.11
MAC
in
Mob
ile
Ad
h
o
c N
e
t
w
orks
(MANET
)
Using Ran
dom W
a
yp
oi
nt Mode
l (RW
M).
Internationa
l Jour
nal
of Informati
on
& Netw
ork
Security (IJINS).
2012; 1(3): 2
23-2
27.
[8]
Subram
ani
an
S, W
e
rner M, Liu S, et al.
C
ong
estio
n
cont
rol for veh
i
cul
a
r safety: Synchron
ous a
n
d
asynchr
ono
us MAC
alg
o
rith
ms
. Proceedi
ngs
of the ninth A
C
M intern
ation
a
l
w
o
rks
hop
o
n
Vehic
u
l
a
r
inter-n
et
w
o
rki
n
g, s
y
stems, an
d app
licati
ons.
201
2: 63-7
2
.
[9]
Laso
w
s
k
i R, St
rassber
ger M.
A multi-ch
an
ne
l be
aco
n
in
g se
rvice for c
o
ll
isi
on av
oi
danc
e i
n
veh
i
cu
la
r
ad-h
o
c netw
o
rks
. IEEE Vehicular T
e
chnol
o
g
y
C
onfer
ence
(V
T
C
Fall). S
an Francisc
o
, CA, USA.
201
1: 1-5.
[10]
Robi
nso
n
CL,
Cav
ene
y
D,
et a
l
. Efficie
n
t
Messa
ge
Comp
ositio
n a
nd Cod
i
n
g
for
Co
oper
ative
Vehic
u
lar Saf
e
ty Ap
plic
ations.
IEEE Transaction on Ve
hic
u
la
r Technol
ogy
. 200
7; 56(6): 32
44-3
255.
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
1693-6
930
Im
proving the
Chan
nel Utili
zation of Basi
c Safety Me
ssag
e in VANETs with…
(Xiang Li)
81
[11]
Bansa
l
G, Ke
n
n
e
y
JB,
Ro
hrs
CE. LIMERIC
:
A li
ne
ar mes
s
age
rate
cont
rol
alg
o
rithm
for ve
hicu
la
r
DSRC s
y
stems
.
IEEE
Transac
tions on Ve
hic
u
lar Technology
. 2013; 62(
9): 418
2-41
97.
[12]
Panic
hpa
pi
boo
n S, Pattara-At
i
kom W
.
A Rev
i
e
w
of Informat
i
on
Dissem
i
n
a
tion Prot
ocols f
o
r Vehic
u
la
r
Ad Hoc Net
w
or
ks.
IEEE Comm
unic
ations S
u
rveys and Tutorials
. 2
012; 1
4
(3): 784-
79
8.
[13]
Ra
w
a
t
DB, Po
pescu
DC, Y
a
n G, et a
l
. E
nha
ncin
g VA
N
E
T
Performance b
y
J
o
i
n
t A
daptati
o
n
of
T
r
ansmission
Po
w
e
r
an
d C
o
ntentio
n W
i
n
d
o
w
Siz
e
.
IEEE
T
r
ansacti
ons
on P
a
ral
l
el
a
n
d
Distri
bute
d
System
s
. 20
11
; 22(9): 152
8-1
535.
[14]
Hassa
nab
ad
i B, Valaee S.
Relia
bl
e Peri
odic Safet
y
M
e
ssag
e
Broa
d
c
asting i
n
VANET
s Using
Net
w
ork C
odi
n
g
.
IEEE Transactions on Wireless Comm
unic
ations
. 20
14; 1
3
(3): 128
4-1
2
9
7
.
[15]
Sahu
PK, Hafi
d A, Ch
erkao
u
i
S.
Inter stree
t
interfere
n
ce
cance
l
atio
n i
n
urba
n ve
hicu
la
r netw
o
rks
usin
g netw
o
rk codi
ng
. IEEE Globa
l Comm
unic
a
tions C
o
n
f
erence (GLO
BECOM 2014)
. Austin,
T
X
.
201
4: 374-
379.
[16]
W
u
C, C
h
e
n
X, Ji Y,
et al.
Efficient
Br
oa
dca
s
ting
in
VANE
T
s Using D
y
n
a
m
ic Back
bon
e
and
Net
w
o
r
k
Codi
ng.
IEEE Transactions on Wireless Comm
unications.
201
5; 14(1
1
): 6057-
607
1.
[17]
Hoe
her PA, W
o
T
.
Superp
o
s
i
tion
Mod
u
l
a
tio
n
: M
y
t
h
s a
n
d
F
a
cts.
IEEE Communic
a
tions
Maga
z
i
ne
.
201
1; 49(1
2
): 110-1
16.
[18]
Chang SH, Rim M, Cosman PC, et
al.
Opti
mized
Un
equ
al Err
o
r P
r
otection
Usi
n
g Multi
p
le
xed
Hierarc
hic
a
l Modu
latio
n
.
IEEE Transaction
on Infor
m
atio
n
Theory
. 201
2; 58(9): 58
16-
58
40.
[19]
Alaj
el KM, Xian
g W
,
W
ang Y. W
ang. Une
q
u
a
l
e
rror prot
ectio
n
scheme
bas
e
d
hier
archic
al 1
6
-QAM for
3-D vid
eo trans
mission.
IEEE Transactions on Cons
um
er Electronics
. 20
12
; 58(3): 731-7
3
8
.
[20]
Hu Z
,
Jiang H,
Li H, et al. A
Lo
w
-
C
o
mp
le
xi
t
y
Dec
o
d
i
ng Al
gorithm for Co
ded H
i
erarc
h
ic
al Mod
u
lati
o
n
in Sin
g
le Freq
u
enc
y Net
w
o
r
ks
.
IEEE
Transac
tion on Broadc
ast
. 2014; 60(
2
)
: 302-31
2.
[21]
Vitthala
dev
uni
PK, Aloui
ni MS
. A Recursive
Al
gorit
hm for the Exact BER
C
o
mputati
on of
Genera
lize
d
Hierarc
hic
a
l QAM Constel
l
ati
ons.
IEEE Transaction on Inform
ation Theory
.
2003
;
49(
1): 297-3
07.
[22]
Griggs W
M
, Ordon
ez-Hurta
d
o
RH, et al
.
A Large-Sc
ale
SUMO-Based
Emulatio
n Platform.
IEEE
T
r
ansactio
n
s o
n
Intelli
ge
nt T
r
ansp
o
rtatio
n Systems
. 20
15; 16(6):1-
10.
[23]
Rodríg
uez VI, Sanchez J. Empo
w
e
rin
g
So
ft
w
a
re Ra
di
o: IT
++
AS A GNU Radio
Out-Of-T
r
ee
Implementation.
IEEE Latin Am
er
ica Trans
actions
. 201
4; 1
2
(2): 269-
27
6.
Evaluation Warning : The document was created with Spire.PDF for Python.