TELKOM
NIKA Indonesia
n
Journal of
Electrical En
gineering
Vol. 12, No. 9, September
2014, pp. 66
4
4
~ 665
0
DOI: 10.115
9
1
/telkomni
ka.
v
12i9.640
0
6644
Re
cei
v
ed
Jun
e
19, 2014; Revi
sed
Jul
y
8, 2014; Accept
ed Jul
y
25, 2
014
Performance Analysis of Extended AODV with
IEEE802.11e HCCA to Support QoS in Hybrid Network
Shalini Sing
h*
1
, Rajeev
T
r
ipathi
2
1
Departme
n
t of Electronics a
n
d
Commu
nicati
on Eng
i
n
eeri
n
g
,
Graphic Era U
n
iversit
y
, De
hr
adu
n, India
2
Departme
n
t of Electronics a
n
d
Commu
nicati
on Eng
i
n
eeri
n
g
,
Motilal N
ehr
u Natio
nal Institu
t
e
of
T
e
chnol
o
g
y
, All
a
h
a
b
ad, India
*Corres
p
o
ndi
n
g
author, e-ma
i
l
: shali
n
i1
11.ra
j
@
gmai
l.com
A
b
st
r
a
ct
T
he i
n
tegr
ation
of th
e fixe
d w
i
red
netw
o
rk
a
nd w
i
re
less
mobil
e
ad
h
o
c n
e
tw
ork can
be
use
d
t
o
eli
m
i
nate
de
ad
z
o
n
e
s i
n
th
e
w
i
reless
netw
o
rk, and
can
a
l
so b
e
us
ed
to
extend
the
cov
e
rag
e
of w
i
re
le
ss
netw
o
rks.
T
he integrati
on of
w
i
red and w
i
reless n
e
tw
orks al
so know
n
as hybri
d
net
w
o
rks, is gaini
ng
pop
ular
ity d
u
e
to its
usefu
l
n
e
ss a
n
d
practi
cal
use
.
Rea
l
t
i
me
app
licati
o
n
s
in
hy
brid
n
e
tw
ork nee
d s
o
me
suitab
le qu
ality
of service. T
h
e qua
lity threshol
ds are i
m
po
sed on
p
a
ra
me
ters
lik
e del
ay,
jitter,
p
a
cket l
o
s
s
and thro
ugh
pu
t.
T
h
is paper
utili
z
e
s the Ex
tende
d AODV
routing prot
o
c
ol for commu
nicati
on betw
e
en
MANET
an
d fi
xed w
i
re
d
net
w
o
rk and
IEE
E
80
2.11
e M
A
C functi
on
H
C
F
Co
ntroll
ed
Ch
ann
el
Acc
e
ss
(HCCA) to su
p
port qu
ality of
service i
n
hy
br
id n
e
tw
ork. The perfor
m
ance
of extend
ed A
O
DV, w
i
th HCCA
(IEEE 802.1
1
e
)
and w
i
tho
u
t
HCCA (IEEE8
02.11)
is co
mp
ared
usin
g si
mulati
on for r
eal
time v
o
ic
e ove
r
IP
traffic. T
he ext
ensiv
e set
of
simulati
ons
sh
ow
s t
hat exte
nde
d AODV
w
i
th H
CCA
pro
v
ides
the
dras
ti
c
reducti
on in j
i
tter compar
e to w
i
thout HCCA.
Ke
y
w
ords
: qu
ality of service,
MANET, HCCA, extende
d AODV, hybrid n
e
tw
ork
Copy
right
©
2014 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
A variety of approa
che
s
have be
en p
r
opo
se
d to
p
r
ovide
wirele
ss i
n
ternet a
c
cess to
mobile ad ho
c netwo
rks (MANETs). Since the integ
r
ation of wire
d and wirele
ss n
e
two
r
ks
is
gainin
g
pop
ularity due to its useful
ne
ss and practi
ca
l use. The in
terco
nne
ction
of fixed wire
d
netwo
rk with
MANET is a
c
hieved by
int
r
odu
cin
g
an
Internet
Gate
way that p
r
o
v
ides the
link to
external h
o
st
s. Thu
s
, a ga
teway a
c
ts a
s
a b
r
idg
e
be
tween
a MA
NET an
d the
Internet an
d
all
comm
uni
cati
on bet
wee
n
the two
networks m
u
st
pa
ss th
ro
ugh
g
a
teway. Thi
s
pape
r u
s
e
d
the
modified version of AODV routing p
r
oto
c
ol whi
c
h is
kn
own a
s
exten
ded AODV, to route pa
cke
t
s
not only
withi
n
a m
obile
a
d
ho
c n
e
two
r
k, but
also
to
a fixed
wire
d net
work [1]
.
Although, t
h
e
Internet Engi
neeri
ng T
a
sk Force
(IETF
)
ha
s p
r
o
p
o
s
ed several ro
uting p
r
oto
c
o
l
s for
MANE
Ts,
su
ch as
A
d
h
o
c On-
D
e
m
a
nd Dist
a
n
c
e
V
e
ct
or
(A
O
D
V) [2], Dyna
mic So
urce
Routin
g (DS
R
) [3],
Optimize
d Li
nk State
Rou
t
ing Proto
c
ol
(OLS
R)
[4].
Ho
wever, all these
protoco
l
s were
d
e
si
g
ned
for com
m
uni
cation within a
n
autonom
ou
s MANET.
T
hese routin
g
proto
c
ol
s are
not suitable
for
integratio
n of MANET and f
i
xed wire
d ne
twork or hyb
r
i
d
netwo
rk.
If Quality of service (Q
oS) is provided to
the hybrid net
wo
rk, then it is more
advantag
eou
s to real time
application
s
[5]. The Qual
i
t
y of service (QoS) [6] of wirele
ss Inte
rn
et
acce
ss th
at coul
d be off
e
red to
MANET node
s
in
hybrid n
e
twork
strongly
depe
nd
s on
the
selection of MAC
schem
e
s.
Th
e IEEE 802.11 [7],
the WLAN l
ega
cy standard
ca
nnot provide
QoS
sup
port
for multim
edi
a ap
plication
s
. Thu
s
, con
s
i
dera
b
le
re
se
a
r
ch
effort
s
ha
ve bee
n
ca
rri
e
d
out to enhance QoS support
for IEEE
802.11 [8]. Among them, IEEE 802.11e [9, 10] is the
upcoming Q
o
S enhan
ced
stand
ard p
r
o
posed by t
he IEEE worki
n
g grou
p. In 802.11e a ne
w
MAC layer functio
n
calle
d the hybrid
coordi
natio
n function (HCF
) that d
eals
with bo
th
conte
n
tion-ba
sed and co
ntention-f
r
ee
acce
ss
m
e
cha
n
ism
s
a
n
d
provid
es
prio
ritize
d a
nd
para
m
eteri
z
e
d
QoS. In pa
rticula
r
two n
e
w MA
C
fun
c
tion
s are ad
ded to the
pre-existin
g
on
es:
the Enhan
ce
d Dist
ribute
d
Cha
nnel A
c
cess (E
DCA
) [11] and the
HCF controlled
Chan
nel Access
(HCCA
) [12]. EDCA improves
the ma
ndatory an
d conte
n
tion-ba
sed
Distri
but
ed Co
ordi
nat
ion
Functio
n
(DCF) by introdu
cing t
r
affic p
r
io
ritizati
on. HCCA
enha
nces th
e optimal
Point
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Perform
a
nce
Analysis of E
x
tended AODV with
IEEE8
02.11e
HCCA to Support
…
(Shalini Si
ngh)
6645
Coo
r
din
a
tion Functio
n
(PCF)
p
o
lling sch
e
me with
a
parameteriz
e
d traffic
c
l
as
s
i
fic
a
tion [13]. This
pape
r analy
s
es the pe
rfo
r
man
c
e of Extended
AODV with 8
02.11e
HCCA to support in
Integration of
MANET with fixed wired n
e
tworks [
14]. The simul
a
tio
n
is ca
rrie
d
out by using NS2
(2.29
)
for real
time voice over IP traffic.
T
he simul
a
tio
n
results prov
e that extended AODV wit
h
IEEE802.11e
HCCA is ben
eficial
for real
time applicati
on.
The re
st of the pape
r is organi
ze
d as follo
ws: In section 1.1
the routing proto
c
ol
extended A
O
DV is descri
bed. In
section 1.2 the I
EEE
802.11e MAC protocol illust
rated.
In
se
ction 2 si
mulation sce
nario
with p
a
rmete
r
s
val
ues a
r
e expl
ained. Sectio
n 3 provid
es the
perfo
rman
ce
meri
cs u
s
e
d
for propo
se
d
simulatio
n
scenari
o
. Secti
on 4 is d
e
tail
ed de
scriptio
n of
simulatio
n
re
sults. Se
ction
5 presents
concl
u
si
on of pape
r.
1.1. Extend
ed
AODV
Extended AO
DV routin
g protocol, which
is al
so known as AO
DV+,
extends the
widely
use
d
Ad ho
c On-Dema
n
d
Distan
ce Ve
ctor (A
ODV
)
routing
proto
c
ol to ro
ute
packet
s
between
the wirel
e
ss a
d
hoc n
e
two
r
k and the
wired Internet, throu
gh a gat
eway [1, 14].
Whe
n
a mobi
le node wi
sh
to commu
nicate with
a fixed wired nod
e as a de
stin
ation; the
mobile n
ode
need
s to be
g
i
n route
discovery proc
ess if it does
n
o
t find a ro
ute towa
rd
s fixed
wire
d no
de i
n
its routing
table.
Ro
ute
discovery p
r
oce
s
s is i
n
itiated by broa
dca
s
ting
RREQ
messag
e a
s
i
n
conventio
n
a
l AO
DV rout
ing p
r
ot
o
c
ol.
Whe
n
a
RRE
Q
me
ssage
i
s
re
ceived
by an
interme
d
iate
mobile n
ode,
an inte
rmedi
ate mobile
n
ode
sen
d
a
RREP b
a
ck t
o
the o
r
igin
ator of
the RREQ if i
t
has
route to
wards th
e wi
red de
sti
natio
n. But in that ca
se, the
sou
r
ce
wo
uld thi
n
k
that the d
e
st
ination i
s
a
mobile
nod
e
that can
be
rea
c
h
ed via
the inte
rme
d
iate n
ode. I
t
is
importa
nt tha
t
the so
urce
kno
w
s that t
he de
st
inatio
n is
a fixed
node
and
no
t a mobil
e
n
ode,
becau
se th
ese a
r
e
som
e
times processed diffe
rent
ly. But in
exten
ded A
O
DV, t
h
is
problem
has
been
solved
by preventing
the intermed
iate node to
sen
d
a RREP back to the originato
r
of the
RREQ
if the
destin
a
tion i
s
a wi
re
d n
ode.
Instea
d,
the
i
n
terme
d
iate
n
ode
upd
ates i
t
s routing
tabl
e
and re
broad
casts the recei
v
ed RREQ m
e
ssag
e. To det
ermin
e
wh
e
t
her the de
stination is a wi
red
node
o
r
n
o
t, an
interm
ed
iate no
de
co
nsult
s
its rou
t
ing table. If
the next
hop
add
re
ss of t
h
e
destin
a
tion is a default ro
ute (see Ta
b
l
e 1), the
de
stination i
s
a
wire
d nod
e. Otherwise, the
destin
a
tion is a mobile no
d
e
or a g
a
teway. Since neit
her the fixed
node n
o
r the
mobile no
de
s in
the MANET
can reply to th
e RREQ, it is reb
r
oa
dcast
ed until it
s T
T
L value
re
a
c
he
s
ze
ro.
When
the timer of the RREQ ex
pire
s, a ne
w RREQ m
e
ss
age is b
r
o
a
d
c
asted
with a l
a
rge
r
TTL val
ue.
Ho
wever, si
n
c
e the fixed n
ode cann
ot re
ceive
the RREQ messag
e (no matter h
o
w
larg
e the T
T
L
value is) the
source will never receive the RREP me
ssage it is wait
ing for. After a network-wi
d
e
sea
r
ch witho
u
t any RREP
,
the wirele
ss station as
su
mes that the
destin
a
ti
on is a wire
d stati
on
and send
s its data packets to the gat
eway, which in turn forwa
r
d
s
them to the de
stination. As
an
alternative a
ppro
a
ch to waiting for a
network
-wid
e sea
r
ch, the gateway could re
sp
ond
to
incomi
ng RREQs on b
ehal
f of wired stat
ions o
n
the Internet.
Table 1. The
Routin
g Tabl
e of Mobile Node
Destination Addr
ess
Next H
op Addr
e
s
s
Fixed no
de
Default
Default Gate
w
a
y
Gate
w
a
y IMN
1.2.
IEEE 802.11e MAC Pro
t
o
c
ol
The IEEE 802.11e
compensates for the lack
of QoS and real-ti
m
e
support of the IEEE
802.11
b stan
dard
by intro
duci
ng two n
e
w fun
c
ti
on
s: the Enhan
ce
d Dist
ributed
Cha
nnel Access
(EDCA) an
d the HCF
Cont
rolled
Ch
ann
el Access
(HCCA
) [15]. EDCA i
s
a di
stributed
sche
me
so
it can be use
d
in both infrast
r
u
c
ture
and ad
h
o
c netwo
rks. Ho
wever,
it
can
not provide a
n
y
Quality of Service (Q
oS) g
uara
n
tee
s
; o
n
ly serv
i
c
e dif
f
erentiation.
On
the
oth
e
r hand, HCCA can
provide
QoS
guarantee
s t
h
rou
gh
re
so
u
r
ce
reservatio
n but it
is a
centrali
zed
an
d mo
re
co
mp
lex
scheme,
whi
c
h is u
s
eful in
infrast
r
u
c
ture
netwo
rks o
n
l
y
. The contro
lled ch
ann
el
acce
ss
refe
rred
to as HCF
co
ntrolled
chan
nel a
c
ce
ss (HCCA
). Th
e
co
ntrolled
chan
nel a
c
ce
ss is
a polli
ng-ba
sed
scheme
en
h
anced from
point coo
r
din
a
tion fun
c
ti
on
(PCF
) of 8
0
2
.11.
The HCCA
m
e
chan
ism
use
s
a Q
o
S-awa
r
e
central
i
zed
coo
r
din
a
t
or [16,
17] called hybri
d
coordi
nato
r
(HC), an
d ope
rates
unde
r so
me rules that are different from
the point coo
r
dinato
r
(P
C)
of the PCF.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
44 – 665
0
6646
The IEEE 802.11e standard
can deliver multimedia
streams
with respect of their QoS
and
real
-time
(i.e. timing
constraints ex
pre
s
sed
in
te
rms of
flows deadli
nes) re
quire
ment
s
[18
,
19].
1.3.
IEEE 802.11e HCCA
HCCA has
been pr
oposed in IEEE 802.11e to prov
i
de
parameterized QoS support
in
the centrali
zed p
o
lling m
e
ch
ani
sm of
PCF
(Point
Co
ordi
natio
n Fun
c
tion
) [
20, 21]. EDCA is
basi
c
ally a
n
i
m
prove
d
me
chani
sm for
DCF, and
HCCA is ba
sically an imp
r
oved
mech
ani
sm f
o
r
PCF. O
ne m
a
in n
e
w feat
ure
of
HCF i
s
the
con
c
e
p
t
of tran
smi
s
sion
op
portu
nity (TXOP) [
22],
whi
c
h refe
rs t
o
a time dura
t
ion durin
g which a Qu
ality of Service Station (QST
A) is allowed
to
transmit a bu
rst of data frames. T
h
e
s
e
bound
ed ti
m
e
interval
s were introdu
ce
d to solve th
e
probl
em
with
unkno
wn tran
smissio
n
time
s of
polle
d
stations in P
C
F
.
HCCA TX
O
P
is
cal
c
ulate
d
according to
TSPEC (T
raffic Specificati
on)
sent
by each
QSTA, then use the
CF-Poll f
r
am
e to
transmit to each QS
TA. Tran
smi
ssi
on
different tr
af
f
i
c cla
s
se
s c
a
l
l
ed t
r
af
f
i
c st
r
eams
(TS
s
)
are
introdu
ce
d in
HCCA. TS
PEC (T
raffic Specifi
c
atio
n) i
s
a
dded
inform
ation
element i
n
I
EEE
802.11e standard. The Qo
S request frame includes a Traffi
c Specification (T
SPEC) elem
ent
(se
e
T
able
2
)
that b
r
ing
s
t
he info
rmatio
n to n
o
ti
fy the
re
quirement
s of
the traffic stream
(TS
)
[23
,24]. This si
mple scheduler
uses the mandatory set of TSPEC
parameters to generate a
s
c
hedule. A
TSPEC
desc
ribes the QoS charac
teris
t
ics
of
a t
r
af
fic
s
t
ream (TS) by
s
p
ec
ifying
para
m
eters such a
s
Mea
n
Data Rate, Service Inte
rva
l
(SI), Delay Bound, No
minal SDU
size etc.
Each TS
ca
n either
be
uni-di
r
e
c
tiona
l or bi-dir
ecti
onal (or b
o
th
of them), correspon
ds t
o
a
s
p
ec
ific
s
e
rvic
e level identified by the values
of the Traffic
S
pec
ific
ation
(TSPEC) prot
oc
ol
para
m
eters. I
n
order to
co
ntrol the
dela
y
, the
maximum value
of a
TXOP is bou
nded
by a val
u
e
calle
d TXOP
Limit, whi
c
h i
s
d
e
termi
ned
by the
Qualit
y of Service
Acce
ss Poi
n
t (QAP). A
Q
S
TA
can transmit multiple fram
es
within its
TXOP
allocation. A QSTA never all
o
we
d exce
eding t
h
e
TXOP limit imposed by the QAP, including interf
ram
e
spa
c
e
s
and
ackn
owl
edg
e
m
ents. Thi
s
new
feature al
so tend
s to provi
de time-b
ase
d
fair
ne
ss b
e
t
ween QSTA
s. The Servi
c
e Interval (SI),
whi
c
h is th
e
time interval
betwe
en two su
cce
ssive
polls of the
node, an
d the tran
smi
s
si
on
oppo
rtunity (TXOP) whi
c
h
is the node
transmiss
io
n duratio
n, based on the m
ean ap
plicati
on
data rate
s of its TSs.
Table 2. TSPEC Element Fields
TS Info
Nominal
MSDU Size
Max
.
MSDU Size
Min. Service
Interval
Max
.
Service
Interval
Inactivity
Interval
Mean Data
Rate
Peak Data
Rate
Max. Burst
Size
Dela
y
Bound
Min. Ph
y
s
ical
Rate
Surplus Band
w
i
d
t
h
Allow
ance
2. Simulation Setup
The
network sim
u
lator n
s
-2
(ve
r
sion
2.29)
is u
s
e
d
to
evaluat
e the
pe
rformance
of
extended AODV with IEEE 802.11e
(with HCCA
) and IEEE 802.11 (w
ithout HCCA) [25], for t
he
integratio
n of
MANET and
fixed wired
netwo
rk. T
h
e
main re
ason
for usi
ng n
s
-2.29 is that t
he
extended AODV and IEEE 802.11e
HC
CA both are compatible
with this version.
The
simul
a
te
d sce
nari
o
sh
own
in Fi
gu
re
1, an
d
scena
rio p
a
ramete
rs a
r
e
given i
n
Table
3.The co
nsi
d
ered
simul
a
tion scen
ario f
o
r hybri
d
net
work con
s
ist
of 18 mobile
node
s in MA
NET
domain, a ga
teway, 3 FTP serve
r
and
1-7 real
time
CBR Source
s in fi
xed wired domai
n. The
topology a
r
ea
is 1000
m x 1000m i
s
take
n for this
sim
u
lation sce
n
a
r
io. All MANET domain n
o
d
e
s
can
comm
uni
cate directly with gate
w
ay
and its dire
ct transmi
ssi
o
n
rang
e is 2
50 meters. T
h
e
simulatio
n
ra
n for 10
0 second
s and th
e
first 20
se
co
nds
are
co
nsi
dere
d
a
s
warm up time. T
he
interes
t
is to
s
t
udy the s
t
eady
state
be
h
a
viour
of the
prop
osed
net
work
scen
ari
o
of Fig
u
re 1.
To
achi
eve this
obje
c
tive, so
me tests
con
c
lud
e
s that
th
e first 20
se
cond
s are the transi
ent stat
e of
the net
work
d
u
ring
which t
he
conn
ectio
n
s
are
set up
, so that th
e fi
rst 2
0
se
con
d
s
a
r
e i
gno
red
in
simulatio
n
s. I
n
o
r
de
r to
evaluate th
e eff
e
ctivene
ss of
HCCA, thi
s
scena
rio
use
s
real
time
CBR
traffic. Th
e
re
al time
CB
R
traffic i
s
m
o
d
e
lled
acco
rdi
ng to
VoIP st
ream
ba
se
d
on G.7
1
1
voi
c
e
cod
e
c g
ene
rating 160 byt
e
s eve
r
y 20 ms, re
sulting
in 64 kb
ps d
a
ta rate. Thi
s
kind of traffic is
given high
er
prio
rity over the othe
r traffic
used in the
prop
osed scenari
o
. The a
nother
sele
ct
ed
appli
c
ation i
s
FTP, which
repre
s
e
n
ts a
bulk d
a
ta
tra
n
sfer
of large
size, sendi
n
g
TCP se
gm
ents
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Perform
a
nce
Analysis of E
x
tended AODV with
IEEE8
02.11e
HCCA to Support
…
(Shalini Si
ngh)
6647
equal to 102
4
bytes. FTP is given
low pri
o
rity and this appli
c
ation h
a
s always so
mething to se
nd
and ru
ns th
ro
ugho
ut the whole sim
u
lati
on. Theref
ore
,
when CB
R
sou
r
ce begi
n
s
se
ndin
g
da
ta
,
then FTP has to stop its tran
smi
ssi
on that time
and give the prio
rity to CBR traffi
c for
transmissio
n.
Figure 1. Simulation Sce
n
a
r
io
For
simul
a
tio
n
, 1 to 7 n
o
d
e
s from
wire
d dom
ain inv
o
lved in VoI
P
comm
uni
cation an
d
remai
n
ing th
ree
serve
r
s fo
r FTP. T
he p
u
rpo
s
e
of
se
l
e
cting
the va
rying num
ber
of real
time V
o
IP
call
s wa
s ma
de to demon
strate and an
al
yzed the abilit
y of IEEE 802.11e HCCA.
The VoIP me
ssage
s a
r
e e
n
ca
psulated i
n
UDP
/IP packets
while th
e FTP me
ssa
ges
are
encap
sulate
d
in T
C
P/IP packets. At th
e net
work
l
a
yer, the
exten
ded AO
DV i
s
used a
s
ad
hoc
routing
proto
c
ol
with rea
c
tive gateway
discove
r
y to acce
ss the
fixed netwo
rk or vice versa
through
gateway. At the
MAC a
nd
physical
layer I
EEE 802.11e
HCCA
or IEEE 802.11 wi
thout
HCCA
are
u
s
ed fo
r the
ev
aluation
an
d
comp
ari
s
o
n
p
u
rpo
s
e.
To
p
r
ovide
the
qu
ality of service
(QoS
) to the
hybrid
network
, t
he IEEE 802.11e s
t
andard is us
ed
for
s
i
mulation.
The MAC layer
para
m
eters use
d
in sim
u
lation
a
r
e given
in
T
a
ble 4. Th
e
traffic sou
r
ce
s a
r
e
sta
r
ted
simultan
eou
sl
y and the
si
mulation
re
su
lts are p
r
e
s
e
n
ted
when
th
ere i
s
no m
o
bility. This p
a
per
simulate
s a
scen
ario
whe
r
e u
s
e
r
s sit i
n
a
cafe,
univ
e
rsity
cam
p
u
s
, confe
r
en
ce
hall, ai
rp
ort,
or
railway statio
n and a
c
cess the In
ternet usin
g their la
ptops.
Table 3. Simulation Para
meters
Parameter Value
Topolog
y area
1000m x
1000m
Mobile Nodes
18
Gate
w
a
y 1
Number of
CBR
sour
ces
1-7 variable
Number of
FTP
sour
ces
3
CBR packet size
and rate
160 b
y
t
e
s,
64kbps
FTP packet size
1024 b
y
tes
Transmission range
250 m
Simulation Time
100 sec.
Warm up time
20 sec
Table 4. MAC Simulation Param
e
ters
Parameter Value
SlotTime 20µs
SIFS 10
µs
Preamble Length
144 bit
PLCP Heade
r
Length
48 bits
PLCP Data Rat
e
1 Mbps
Data Rate
11Mbps
Basic Rate
1 Mbps
CWMIN
31
CWMAX 1023
Max SDU Size
2132
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
44 – 665
0
6648
3. Performan
ce Metric
s
The follo
wing
metrics a
r
e
use
d
in
simul
a
ti
on
for purp
o
se of
evalua
tion
and pe
rforma
nce
comp
ari
s
o
n
of extended AODV with HCCA and wit
h
o
u
t HCCA in hybrid network sce
nari
o
sh
o
w
n
in Figure 1.
Packet Deliv
ery Ratio
(PDR) is
cal
c
ul
ated as
th
e numbe
r of d
a
ta packet
s
receive
d
at
the destin
a
tio
n
divided by the numb
e
r of
data packet
s
generated at
the sou
r
ce.
Average E
n
d
–to- En
d De
lay: End –to- End Delay i
s
calculated
as the
time
whe
n
a
packet i
s
received at the
destin
a
tion m
i
nus th
e
time
whe
n
the p
a
cket was
g
enerated at t
he
sou
r
c
e
.
Jitter is
cal
c
ul
ated as the v
a
rian
ce of the
end- to- e
nd
delay.
Thro
ugh
put is cal
c
ulate
d
as the num
b
e
r of
data bits re
ceived at the destinatio
n divided
by the time the con
s
id
ere
d
traffic type (VoIP, FTP).
4. Simulation Resul
t
s
This sectio
n
pre
s
ent
s the
results of inte
grat
ion
of MANET with int
e
rnet mo
del scenari
o
whi
c
h is obtai
ned thro
ugh
simulatio
n
s.
The per
fo
rma
n
ce an
alyze
d
the benefits of IEEE 802.11e
(HCCA
) over
the IEEE 802.11 wi
thout HCCA in the consi
dered si
mulation scenario.
4.1.
Av
erage End to End Dela
y
The Figure
2 and Figure
3
shows average end-
to-
end delay pl
ot for IEEE 802.11 and
802.11e HCCA MAC and proves that
the average end- to-
end delay
varies significantly in IEE
E
802.11
MAC
comp
ared to
802.11
e (HCCA)
MAC. T
h
is
i
s
b
e
cau
s
e
of the
re
se
rvation to
acce
ss
the medium i
n
IEEE 802.
11e
(HCCA)
MAC for transmission
of tr
affic stream (TS),
based on
QoS requi
re
ments. The
Service Interval (SI)
value and Tran
smi
ssi
on Opp
o
rt
unity (TXOP) are
pred
efined
b
y
Quality of
servi
c
e A
c
ce
ss Point
(Q
AP) for th
e
real tim
e
pe
riodic traffic
and
cont
e
n
t
i
on f
r
e
e
mediu
m
a
c
ce
ss f
o
r V
o
I
P
call
s.
The
r
ef
ore, allo
catio
n
of medi
um
perio
dically to a
spe
c
ific traffi
c
stre
am
by QAP p
r
ovid
es
almo
st
consta
nt
en
d-t
o
-en
d
delay as de
sired. An
intere
sting o
b
s
ervatio
n
that
need
s to be
focu
sed i
s
th
e sh
arp i
n
cre
a
se
of avera
ge en
d- to
- e
nd
delay a
s
n
u
m
ber
of CB
R
source
s
in
crea
se
s. Thi
s
i
s
b
e
ca
use that
fe
w
s
t
a
t
io
ns
ar
e
ou
ts
ide
the
transmissio
n
rang
e of b
o
th tran
smitters and
re
ce
iv
ers.
B
u
t
wi
t
h
inc
r
ea
se i
n
CB
R
s
our
ce
s,
average end- to- en
d delay is very less in the case
of IEEE 802.11e (with HCCA) comparing to
IEEE 802.11 (without HCCA).
Figure 2. Packet Del
a
y Plot without HCCA
Figure 3. Packet Del
a
y Plot with HCCA
4.2. Jitter
Table 5 i
s
for the com
pari
s
on of jitter fo
r IEEE 802.11 (i.e. Without
HCCA)
and 802.11e
HCCA MAC,
whe
n
the nu
mber of
CBR sou
r
ces i
s
varying fro
m
1
to 7. This Ta
ble 5 sho
w
s t
hat
as the CB
R sources are in
creasi
ng, the jitter is quite low an
d almost constant wi
th IEEE 802.11e
HCCA MA
C
compare to the IEEE
802.11 MA
C. The result
show
s that the IEEE 802.11e
HCCA
is abl
e to pro
v
ide QoS to
high p
r
iority traffic even d
u
ring high t
r
affic loa
d
. In co
n
t
rast to thi
s
, the
jitter variation pattern i
s
random and
qui
te hi
gh
with the IEEE 802.
11 and
does
not provide QoS
to the
hybrid
network. Sin
c
e th
e jitter i
s
relate
d
with en
d to
en
d
delay, th
eref
ore, th
e
simil
a
r
0
100
0
200
0
300
0
400
0
0
0.
0
2
0.
0
4
0.
0
6
0.
0
8
0.
1
0.
1
2
0.
1
4
Sequ
enc
e
N
u
m
b
e
r
(n
)
P
a
ck
e
t
R
e
ce
i
v
e
d
D
e
l
a
y(
s
)
0
1000
2000
3000
4000
0
0.
02
0.
04
0.
06
0.
08
0.
1
0.
12
0.
14
Sequen
c
e
N
u
m
ber
(
n
)
P
a
ck
e
t
R
e
ce
i
v
e
d
D
e
l
a
y(
s
)
Evaluation Warning : The document was created with Spire.PDF for Python.
TELKOM
NIKA
ISSN:
2302-4
046
Perform
a
nce
Analysis of E
x
tended AODV with
IEEE8
02.11e
HCCA to Support
…
(Shalini Si
ngh)
6649
reason is here to get low ji
tter value
wit
h
IEEE 802.11e
HCCA MA
C and
i
s
able to provide the
QoS to the co
nsid
ere
d
hybrid netwo
rk
scenari
o
.
Table 5. Jitter
J
i
tter
(10
-
4
s
2
)
No. of CBR
Sour
ces
CBR
Without HCCA
With HCCA
1 2.7999
0.0016
4 4.9612
4.9688
5.0256
5.0707
0.0035
0.0035
0.0035
0.0015
7 3.1377
3.1872
3.1851
3.2014
3.1919
3.1434
3.3024
0.0034
0.0036
0.0035
0.0036
0.0033
0.0015
0.0036
4.3. Packet
Deliv
er
y
Ratio
Table 6
shows that, the pack
et delivery
ratio decreases
with IEEE 802.11 MA
C
for CBR
is u
s
e
d
a
nd
packet l
o
ss
b
e
com
e
hi
ghe
r a
s
n
u
mb
er
of
CB
R
s
o
u
r
ce
s in
cr
ea
se
s. On
the
other
hand, the packet loss is negligible with I
EEE
802.11e MAC (with HCCA
) fo
r CB
R. It is observed
from Ta
ble
6
that a ve
ry h
i
gh pa
cket d
e
livery ra
tio fo
r both
with
a
nd
without
HCCA i
n
FTP,
as
the numb
e
r o
f
CBR so
urce
s incre
a
ses.
This i
s
due t
o
the relia
ble
delivery se
rv
ice p
r
ovided
by
TCP. FTP use con
n
e
c
tion
– oriente
d
TCP,
which retra
n
smits
dro
p
p
ed pa
ckets.
Table 6. Packet Delivery Ratio
Packet delivery
r
a
tio %
No. of CBR
Sour
ces
CBR FTP
Without HCCA
With HCCA
Without HCCA
With HCCA
1 96.57
99.93
96.53
96.96
4 91.12
99.92
95.50
96.22
7 86.35
99.89
94.64
95.87
4.4. Through
put
Table 7 is for through
put a
nalysi
s
of the co
n
s
ide
r
ed simulation sce
nario. And it is found
that the throughput is sli
ghtly better
with IEEE 802.11e HCC
A
compare to the IEEE 802.11
(without HCCA). Table 7 proves
for
the case of
throughput wit
hout
HCCA, as the more number
of traffic stre
ams a
r
e cont
endin
g
for m
edium a
c
cess, the highe
r
the prob
abilit
y of collision
s
and
ret
r
a
n
smi
s
sio
n
s
re
sult
ing
lowe
r t
h
rou
g
hput
.
A
l
t
hou
gh,
a
s
t
h
e t
r
af
f
i
c
st
re
am
s in
c
r
ea
se
s,
t
h
e
throug
hput is almost co
nsistent
and ve
ry close to the actu
al value in ca
se o
f
IEEE 802.11e
HC
CA.
Table 7. Th
ro
ughp
ut
Throug
hput
(kbp
s)
No. of CBR
Sour
ces
CBR FTP
Without HCCA
With HCCA
Without HCCA
With HCCA
1 61.82
63.97
3154.94
3202.15
4 233.33
255.86
2709.50
2910.82
7 386.96
447.65
2313.11
2619.70
5. Conclusio
n
This pa
per ev
aluate
s
the ability of ext
ended AODV wi
th IEEE 802.11e HCCA a
nd IEEE
802.11
witho
u
t HCCA to hybrid net
wo
rks. Simula
tion re
sults
sh
ows that the perform
an
ce
is
degraded with
increasi
ng
number
of CBR sources
when IEEE 8
02.11
(without HCCA) is used to
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 23
02-4
046
TELKOM
NI
KA
Vol. 12, No. 9, September 20
14: 66
44 – 665
0
6650
the hybri
d
network
scena
ri
o, whereas I
EEE 802.11e (with
HCCA
) successfully
providi
ng QoS
sup
port in terms of low a
n
d
cont
rolled
e
nd-to
-en
d
del
ay and jitter, the req
u
ire
d
throu
ghp
ut, and
negligibl
e
pa
cket loss to re
al time applications.
Referen
ces
[1]
A Hami
di
an. A
Stud
y
of Inter
net C
onn
ectivit
y
for
Mob
i
l
e
A
d
Hoc
N
e
t
w
ork
s
in
NS 2”. M
a
ster’s thes
is.
Dep
a
rtment of Commun
i
cati
o
n
S
y
stems, Lu
nd Instit
ute of T
e
chnolog
y, L
und U
n
ivers
i
t
y
.
Januar
y 2
0
0
3
.
[2]
C Perki
n
s E
M
, Beld
ing-
Ro
yer, S
D
a
s.
Ad
h
o
c On-D
eman
d D
i
stan
ce Vector
(A
ODV) Ro
uting
.
Experi
m
ental RF
C
356
1.
[3]
DB Johnson, DA Maltz, Y
Hu,
JG Jetcheva.
T
he
Dyna
mic S
ource
Ro
utin
g
Protocol
for M
obil
e
A
d
H
o
c
Networks (DSR)
. IE
T
F
Internet Draft. April 2003. Work in pr
ogress.
[4]
T
Clausen, P J
a
cqu
e
t, A Lao
uiti, P Min
e
t, P Muhl
ethal
er, A Qa
yyum,
L Vi
enn
ot.
Op
ti
m
i
z
e
d
Li
n
k
State
Routi
ng Protoc
ol.
Experim
ent
al RF
C 36
26.
[5]
Ali Hamidian,
Ulf K¨orner.
D
i
stribute
d
R
e
s
e
rvatio
n-bas
ed
QoS i
n
A
d
Hoc N
e
tw
orks w
i
th Interne
t
Access Connectivity.
In proceedi
ng of: T
e
letraffic Congr
ess IT
C 21. 2009.
[6] Lei
Chen
,
Wen
d
i B. Hei
n
ze
lman
, QoS-Aw
are Routi
ng B
a
s
ed o
n
Ban
d
w
i
d
t
h Estimati
on f
o
r Mobi
le A
d
Hoc Networks. IEEE Journal
On Se
lected Ar
eas In Comm
unications.
20
05
; 23(3).
[7]
Q Ni, L Romdhani, T
T
u
rletti.
A Survey of QoS Enh
ance
m
ents for IEEE 802.11 Wire
less
LAN. Journ
a
l
of W
i
reless Co
mmu
n
icati
ons
and Mo
bil
e
Co
mp
utin
g, W
iley
.
2004; 4.
[8]
Dmitri D Perk
i
n
s, Herma
n D
Hug
hes.
A s
u
rvey o
n
q
ual
i
t
y-of-service s
upp
ort for mo
bile
ad
hoc
netw
o
rks. W
i
reless Co
mmu
n
i
c
ations a
nd M
obil
e
Co
mput
in
g Speci
a
l Issu
e: Mobil
e
Ad Hoc Netw
orkin
g
Rese
arch, T
r
ends an
d App
lic
ations.
20
02; 2
(
5): 503–
51
3.
[9]
A Grilo, M Ma
cedo, M N
u
n
e
s
.
A scheduling algorithm
f
o
r
QoS su
pport
in IEEE 8
02.11e networks.
IEEE Wireless Comm
unications.
2003; 3
6–4
3.
[10]
S Man
gol
d, S
Choi,
GR H
i
ert
z
, et al.
A
n
aly
s
is of IEEE
802.11e for Qo
S
Support in Wireless
LANs
.
IEEE Wireless Comm
unications.
2003; 1
0
(6)
.
[11]
Hamid
i
a
n
, U K¨orner.
An E
nha
nce
m
e
n
t to the IEEE 802.11
e EDCA
Providi
ng Qo
S Guarantees.
Telecommunic
a
tion System
s.
2006; 3
1
.
[12]
L Palo
pol
i, R
Lo Cig
no A. Col
o
mbo.
Co
ntrol
and Opti
mis
a
ti
on of HCCA 8
02.11
e Access
Schedu
lin
g.
Procee
din
g
s o
f
the 46th IEEE Confere
n
ce
on De
cis
i
o
n
a
nd Co
ntrol Ne
w
Orleans, LA
, USA. 2007;
121
4.
[13]
Anna
Li
na
R
u
scelli,
Gabri
e
l
e
C
e
cchetti,
Ange
lo
Alifa
n
o
,
Giusep
pe
Li
pari. E
n
h
ance
m
ent of Q
o
S
supp
ort of HCCA sched
ulers
using ED
CA
function in I
EEE 802.1
1e
net
w
o
rks.
Jour
nal Ad H
o
c
Networks.
201
2; 10(2): 14
7-1
61.
[14]
A Ham
i
di
an,
U K¨or
ner, A
Nilsso
n.
Perfor
ma
nce
of Inte
rnet Access
S
o
luti
ons
in
Mo
bile
Ad
H
o
c
Netw
orks. Lecture Notes i
n
C
o
mputer Sci
e
n
c
e (LNCS).
20
05; 342
7.
[15]
Jing-R
o
n
g
Hs
i
eh, T
s
ern-Hue
i
Le
e.
Dat
a
rate estim
a
tion
algorithm
for
IEEE 802.11e HCCA scheduler.
Internatio
na
l Journ
a
l of Perva
s
ive Co
mputin
g and C
o
mmu
n
icati
ons.
20
08
3(3):243 - 25
6
.
[16]
YS Chen, YW
Lee, GH Hu
ang.
Mod
i
fie
d
HCCA
mech
a
n
is
m for impr
ovin
g QoS support in IEE
E
802.1
1
e
netw
o
rk.
Procee
din
g
s
of the
Intern
ation
a
l
Co
nfer
enc
e
on
Mo
bil
e
T
e
chnol
og
y,
App
licati
ons,
and S
y
stems, ACM
. 2008.
[17]
D Sk
y
r
i
ano
gl
o
u
, N Passas, AK Salkintzis,
ARROW
:
an efficient traffic sched
uli
ng al
gor
i
t
hm for IEEE
802.1
1e H
CCA
. IEEE
Transac
tion on Wire
les
s
Communic
a
ti
ons.
200
6; 5: 3558
–3
567.
[18]
G.Cecchetti, AL Rusce
lli. Re
al-T
ime Suppo
rt
for HCCA Function i
n
IEEE 802.1
1e
Net
w
orks: A
Performanc
e E
v
alu
a
tion, S
e
c
u
rit
y
an
d C
o
m
m
unic
a
tion
Net
w
o
r
ks.
Spec
ia
l
Issue o
n
Security for QoS
Assured Wireless Networks.
2010.
[19] G
Cecchetti
, AL Rusc
el
li.
P
e
rfo
r
m
a
n
c
e e
v
alu
a
t
i
o
n
o
f
re
al
-
t
i
m
e s
c
h
e
d
u
l
e
r
s
fo
r
H
C
C
A
fu
nc
ti
o
n
in
IEEE
802.1
1e w
i
rel
e
ss netw
o
rks.
Proc. ACM Q2SW
inet, Vancou
ver, Cana
da. 2
008.
[20] Qiang
Ni.
P
e
r
f
orma
nce
Ana
l
ysis an
d E
n
h
ance
m
entsf
or
IEEE 802.1
1
e
Wireless
Net
w
orks.
IEEE
Net
w
ork. 20
05;
19(4): 21-2
7
.
[21]
Hua L
i
an
g, F
eng Z
eng.
A R
e
searc
h
on H
CCA Mech
ani
sm of W
i
reles
s
LAN Access
. Journal o
f
Networks.
201
2; 7(5).
[22]
Aphir
a
k Jans
a
ng, Ana
n
Pho
n
pho
em.
Adjust
abl
e T
X
OP me
chan
is
m for supporti
ng vi
deo
transmissio
n
in IEEE 802.1
1
e
HCCA. EUR
ASIP Journal
o
n
Wireless C
o
mmu
n
icati
ons
and N
e
tw
orkin
g
.
2011.
[23]
Higuc
hi Y, F
o
rond
a A, Oh
ta
C, Yoshim
oto M, Okada
Y.
Delay Gu
a
r
antee
an
d S
e
rvice Interv
al
Optimi
z
a
t
i
o
n
f
o
r HC
CA i
n
IEEE802.
1
1e W
L
A
N
s.
W
i
relss C
o
m
m
unic
a
tions
a
nd
Net
w
ork
i
n
g
Confer
ence. 2
007
.
[24] Liu
H
u
i.
A N
o
vel QoS R
outi
ng Al
gorithm
i
n
W
i
reless M
e
sh N
e
t
w
orks.
T
E
LKOMNIKA Indon
esi
a
n
Journ
a
l of Elec
trical Eng
i
ne
eri
ng. 201
3;
11(3)
: 1652-1
6
6
4
.
[25]
F
eng Z
e
n
g
, L
a
n
Yao,
Hig
an
g
Che
n
. Impact
of T
opolog
y
a
nd T
r
affic on Interferenc
e a
n
d
Ro
utin
g in
IEEE 8021
1 Wireless Mesh
Net
w
o
r
k.
T
E
LKOMNIKA Indones
ian Jo
urn
a
l of
Electrica
l
Engin
eeri
n
g
.
201
2; 10(4): 79
8-80
5.
Evaluation Warning : The document was created with Spire.PDF for Python.