Int. J. Communications, Network and System Sciences, 2009, 2, 759-763
doi:10.4236/ijcns.2009.28088 blished Online November 2009 (http://www.SciRP.org/journal/ijcns/).
Pu
A MAC Scheme with QoS Guarantee for MANETs
Yanbin YANG1,2, Yulin WEI2
1School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China
2Department of Communications, Guangdong Vocational Technology College of Posts & Telecom, Guangzhou, China
Received July 30, 2009; revised August 31, 2009; accepted September 25, 2009
Abstract
IEEE 802.11 distributed coordination function (DCF) can alleviate the collision and hidden station problem,
but it doesn’t differentiate traffic categories (TC). Therefore, it can’t provide sufficient QoS support for dif-
ferent traffic categories. Recently, a new contention-based enhanced distributed channel access (EDCA)
scheme was proposed which provides a probabilistic QoS support. In this paper, an adaptive EDCA scheme
with QoS guarantee for mobile ad hoc networks (MANETs) is proposed. In this scheme, the EDCA scheme
and the token bucket algorithm (TBA) are combined to adjust the contention window (CW). Our scheme
provides the traffic differentiation.
Keywords: Mobile Ad Hoc Network, Medium Access Control, Distributed Coordination Function, En-
hanced Distributed Channel Access, Contention Window
1. Introduction
Mobile ad hoc network (MANET) becomes popular be-
cause of its low cost and easy deployment. The medium
access control (MAC) protocol for this wireless commu-
nication system employs a mandatory contention-based
channel access function called Distributed Coordination
Function (DCF) [1], which is based on the Carrier Sense
Multiple Access (CSMA) mechanism. Mobile stations
deliver MAC Service Data Units (MSDUs) after detect-
ing that there is no other transmission on the same wire-
less medium. However, if two stations detect that the
channel is free at the same time, a collision occurs. The
IEEE 802.11 standard defines a Collision Avoidance
(CA) mechanism to reduce the probability of such colli-
sions. As a part of CA, before starting a transmission a
station performs a backoff operation. It has to keep sens-
ing the channel for an additional random time after de-
tecting the channel as being idle for a minimum duration
called DCF Interframe Space (DIFS). Only if the channel
remains idle for this additional random time period, the
station is allowed to initiate the transmission. The dura-
tion of this random time is determined as a multiple of a
slot. Each station maintains a so-called Contention Win-
dow (CW), which is used to determine the number of
slots a station has to wait before transmission.
To avoid the hidden station problem inherent in
CSMA, IEEE 802.11 defines a Request to Send/Clear to
Send (RTS/CTS) scheme, which can be used optionally.
Before transmitting data frames, a station has the option
to transmit a short RTS frame, followed by the CTS
transmission by the receiving station. The RTS and CTS
frames include the information of how long it does take
to transmit the next data frame, i.e., the first fragment,
and the corresponding ACK response. Thus, other sta-
tions close to the transmitting station and hidden stations
close to the receiving station will not start any transmis-
sions; their timer called Network Allocation Vector
(NAV) is set. RTS/CTS helps to protect long data frames
against hidden stations. With fragmentation, multiple
ACKs are transmitted, whereas with RTS/CTS the
MSDU can be efficiently transmitted in a single data
frame. Between two consecutive frames in the sequence
of RTS, CTS, data, and ACK frames, a Short Interframe
Space (SIFS) offers transceivers time to turn around.
Because RTS/CTS handshake may introduce some delay
especially in case of wireless networks, [2] proposed a
method to improve overall throughput of the network
where selected RTS packets are dropped based on a node
sequence number.
802.11 DCF reduces collision and hidden terminal
problem, but it doesn’t differentiate TC. Therefore it
can’t provide sufficient QoS support for different traffic
categories. Recently, a new contention-based EDCA
scheme was proposed which provides a probabilistic
QoS support. [3] presented a new protocol, called dis-

Y. B. YANG ET AL. 760
tributed end-to-end allocation of time slots for real-time
traffic (DARE). It allocates and uses periodic time slots
for QoS-demanding applications. DARE reserves these
time slots in a fully distributed way, schedules the real-
time data packets, repairs broken reservations, and dis-
seminates the reservation information to potential inter-
ferers using a piggyback technique. It works well for
high traffic load network but poorly for low traffic load
network. [4] presented a distributed MAC scheme called
opportunistic synchronous array method (SAM) for a
large network of wireless routers. [5] presented a novel
tone-based contention resolution mechanism that exploits
space-time uncertainty and high latency to detect colli-
sions and count contenders, achieving good throughput
across all offered loads for underwater acoustic sensor
networks.
In this paper, an adaptive EDCA scheme with QoS
guarantee for MANETs is proposed. In this scheme, the
EDCA algorithm and the token bucket algorithm (TBA)
are combined to adjust the CW. Our scheme provides the
traffic differentiation.
2. Token Bucket Algorithm
The number of bytes in the bucket (bucket length) and
the occupancy of the transmission buffer (queue length)
as input parameters are employed in the token bucket
algorithm [6] (see Figure 1).
The bucket size (bsize) determines the accepted
burstiness of source. The bucket length (blen) represents
the resources that the user can use to transmit packets.
The bucket limit (blim) represents the maximum packet
length allowed. The queue size (qsize) represents the
source capacity. The queue length (qlen) denotes the
willingness of a station to transmit packets.
This algorithm computes a value which is used to
scale the CW values defined in IEEE 802.11. It is de-
scribed as follows:
If (overload) then p=(1+Δ4)p
Else if (qlen=0) then p=(1+Δ1)p
Else if (blen<blim) then p=(1+Δ2)p
Else p=(1-Δ3)p
P=min{p,1}
Figure 1. Brief description of token bucket algorithm.
CW=p*CW802.11
where Δ1=0.025, Δ4=0.25, 21
lim
lim
bblen
b

 ,
3
lim
lim
blen b
bsize b

1



 . A solution to determine the over-
load is described as follows:
If (av_nr_coll>c) then overload=true
Av_nr_coll=(1-t)*num_coll+av_nr_coll
where av_nr_coll is the average collision number; c is a
constant with value in [0,8] and is always set to 4; t=0.25;
num_coll is the collision number after transmission.
3. Adaptive EDCA
EDCA [7] is designed to provide prioritized QoS by en-
hancing the contention-based DCF. It provides differen-
tiated, distributed access to the wireless medium for QoS
stations (QSTAs) using 8 different user priorities (UPs).
Before entering the MAC layer, each data packet re-
ceived from the higher layer is assigned a specific user
priority value. How to tag a priority value for each
packet is an implementation issue. The EDCA scheme
defines four different first-in first-out (FIFO) queues
called access categories (ACs) that provide QoS support
for the delivery of traffic with UPs at the QSTAs. Each
data packet from the higher layer along with a specific
user priority value is mapped into a corresponding AC
according to Table 1. Different kinds of applications (e.g.,
background traffic, best effort traffic, video traffic, and
voice traffic) can be casted into different ACs. For each
AC, an enhanced version of the DCF, called enhanced
distributed channel access function (EDCAF), contends
for TXOPs using a set of EDCA parameters from the
EDCA Parameter Set Element or from the default values
of the parameters when no EDCA Parameter Set Element
is received from the QAP of the QBSS with which the
QSTA is associated. [8] proposed an effective and simple
model for two-level collision to demonstrate that EDCA
Table 1. User priority to access category mappings.
User priority
(UP) Access category (AC) Designation
1 0 Background
2 0 Background
0 1 Best effort
3 1 Best effort
4 2 Video
5 2 Video
6 3 Voice
7 3 Voice

Y. B. YANG ET AL. 761
form a backoff operation with increased CW values. [10]
proposed an analytical model to calculate the EDCA
parameter set that ideally achieves a predetermined utili-
zation ratio between uplink and downlink flows.
Transmission opportunity (TXOP) is defined in IEEE
802.11e as the interval of time when a particular QSTA
has the right to initiate transmissions. There are two
modes of EDCA TXOP defined, the initiation of the
EDCA TXOP and the multiple frame transmission within
an EDCA TXOP. An initiation of the TXOP occurs
when the EDCA rules permit access to the medium. A
multiple frame transmission within the TXOP occurs
when an EDCAF retains the right to access the medium
following the completion of a frame exchange sequence,
such as on receipt of an ACK frame. The TXOP limit
duration values are advertised by the QAP in the EDCA
Parameter Set Information Element in Beacon frames.
During an EDCA TXOP, a STA is allowed to transmit
multiple MAC protocol data units (MPDUs) from the
same AC with a SIFS time gap between an ACK and the
subsequent frame transmission. A TXOP limit value of 0
indicates that a single MPDU may be transmitted for
each TXOP. This is also referred to as contention free
burst (CFB).
Figure 2. Implementation model.
mechanism, which adopts priority classification and
two-level collision, can not only preferentially guarantee
the throughput of voice and video, but also greatly de-
crease the average delay of the system.
Figure 2 shows the implementation model with four
transmission queues, where each AC behaves like a vir-
tual station: It contends for access to the medium and
independently starts its backoff procedure after sensing
the medium idle for at least AIFS period. In EDCA a
new type of IFS is introduced, the arbitrary IFS (AIFS),
in place of DIFS in DCF. Each AIFS is an IFS interval
with arbitrary length as follows: We propose an adaptive EDCA scheme for the con-
tention period (CP). In our scheme, we combine the
EDCA scheme and the token bucket algorithm (TBA) to
adjust the CW. In the adjustment of the CW, there are
additional aspects that have to be taken into account: 1)
We do not want the CW to increase above the values
used by the Best Effort terminals, since this would lead
to a worse performance than Best Effort. For backward
compatibility, the CW for Best Effort should be the one
defined by the 802.11 standard; 2) If the low transmis-
sion rate of the application is the reason for transmitting
below the desired rate, then the CW should obviously not
be decreased; 3) When estimating the transmission rate,
it would be desirable to control the allowed burstiness of
the source; 4) CWs should not be allowed to decrease in
such a way that they negatively influence the overall
performance of the network. After considering all the
above issues, we describe the adaptive EDCA scheme as
follows.
AIFS[AC] = SIFS + AIFSN[AC] × slot time
where AIFSN[AC] is called the arbitration IFS number
and is determined by the AC and the physical settings.
The timing relationship of EDCA is shown in Figure 3.
The AC with the smallest AIFS has the highest priority.
The values of AIFS [AC], CWmin [AC], and CWmax
[AC], which are referred to as the EDCA parameters, are
announced by the AP via beacon frames. The purpose of
using different contention parameters for different
queues is to give a low-priority class a longer waiting
time than a high-priority class, so the high-priority class
is likely to access the medium earlier than the
low-priority class. [9] presented the effect of different
values of AIFS on MAC access delay. An internal colli-
sion occurs when more than one AC finishes the backoff
procedure at the same time. In such a case, a virtual col-
lision handler in every QSTA allows only the high-
est-priority AC to transmit frames, and the others per-
Figure 3. The timing relationship of EDCA.

Y. B. YANG ET AL. 762
3.1. Successful Transmission
Step 1. Compute the collision rate
([
(_ []
j
j
curr
j
E collisionsp
fE datasentp 
])
)
.
Step 2. Introduce exponentially weighted moving av-
erage (EWMA) to compute the average collision rate
1
(1) **
j
jj
avgcurr avg
f
ff 

 ,
where α is between 0.2 and 0.25.
Step 3. Compute p by using the TBA
If (overload) then p=(1+Δ4)p
Else if (qlen=0) then p=(1+Δ1)p
Else if (blen<blim) then p=(1+Δ2)p
Else p=(1-Δ3)p
P=min{p,1},
where Δ1=0.025, Δ4=0.25, 2
lim
lim
bblen
b

 
1
,
31
lim
lim
blen b
bsize b

 
.
Step 4. Use the multiply factor (MF)
MF[i]=min(p,0.8).
Step 5. After successful transmission, CW[i] is set to
CWnew[i]=max(CWmin[i],CWold[i]*MF[i]).
3.2. After Collision
Step 1. Compute the collision rate
([
(_ []
j
j
curr
j
E collisionsp
fE datasentp ])
)
.
Step 2. Introduce exponentially weighted moving av-
erage (EWMA) to compute the average collision rate
1
(1) **
j
jj
avgcurr avg
f
ff 

 ,
where α is between 0.2 and 0.25.
Step 3. Compute p by using TBA
If (overload) then p=(1+Δ4)p
Else if (qlen=0) then p=(1+Δ1)p
Else if (blen<blim) then p=(1+Δ2)p
Else p=(1-Δ3)p
P=min{p,1}
If (overload) then p=(1+Δ4)p
If (j
avg
f
c) then overload=true,
where c=0.5, Δ1=0.025, Δ4=0.25, 21
lim
lim
bblen
b

  ,
31
lim
lim
blen b
bsize b

 
.
Step 4. Use the multiply factor (MF)
MF[i]=min(p,0.8).
Step 5. After transmission failure, CW[i] is set ac-
cording to each TC’s PF[i] to guarantee that high priority
TC has low PF[i] value
CWtemp[i]=min(CWmax[i],CWold[i]*MF[i])
CWnew[i]=max(CWold[i],CWtemp[i]*MF[i]).
4. Simulation Results
In this section we evaluate the performance of EDCF (IEEE
802.11e) [11], AEDCF and our scheme (Adaptive EDCA).
AEDCF [12] is another improved scheme of EDCA. It
also improves the throughput through changing CW.
The simulation is with respect to two scenarios,
namely Scenario 1 and Scenario 2. Scenario 1 is a light
traffic load environment. Scenario 2 is a high traffic load
environment. The parameters CWmin, CWmax, AIFS
are set to 5, 100, 5 in the scenario 1 and 30, 500, 15 in
the scenario 2. In our scheme p, α, bsize, and c are set to
0.5, 0.25, 10, 0.5, respectively. Through Figure 4 we find
that there is a slump of EDCF and AEDCF in the 10 sta-
tions because of collision among these stations. But our
scheme (Adaptive EDCA) keep good throughput by us-
ing TBA.
Through Figure 5 we find that the throughput of
AEDCF is 4.9% higher than EDCF in the high traffic
load environment. But the throughput of our scheme
(Adaptive EDCA) is 11.4% higher than EDCA.
Figure 4. Scenario 1.
Figure 5. Scenario 2.

Y. B. YANG ET AL. 763
5. Conclusions
The real-time and multimedia services are more and
more popular. We find that providing service differentia-
tion and high throughput is a new challenge in MANETs.
The MAC protocol of the IEEE 802.11e is changed in
order to improve its effectiveness. The EDCA scheme
and the TBA are combined to adjust the CW. Our
scheme provides the traffic differentiation. The results
show the good performance of our scheme under both
light and high traffic loads.
6. References
[1] IEEE Std, “802.11-1999, Part 11: Wireless LAN Medium
Access Control (MAC) and Physical Layer (PHY) Speci-
fications,” Reference number ISO/IEC 8802-11:1999(E),
IEEE Std. 802.11, 1999 Edition.
[2] P. V. Krishna and Iyengar, “Sequencing technique: An
enhancement to 802.11 medium access control to im-
prove the performance of wireless networks,” N.Ch.S.N.,
International Journal Communication Networks and Dis-
tributed Systems, Vol. 1, No. 1, pp. 52–70, 2008.
[3] E. Carlson, C. Prehofer, C. Bettstetter, et al, “A distrib-
uted end-to-end reservation protocol for IEEE 802.11-
based wireless mesh networks,” IEEE Journal on Select-
ed Areas in Communications, Vol. 24, No. 11, November
2006, pp. 2018–2027.
[4] B. Zhao and Y. B. Hua, “A distributed medium access
control scheme for a large network of wireless routers,”
IEEE Transactions on Wireless Communications, Vol. 7,
No. 5, pp. 1614–1622, May 2008.
[5] A. A. Syed, W. Ye, and J. Heidemann, “T-Lohi: A new
class of MAC protocols for underwater acoustic sensor
networks,” IEEE INFOCOM’08, The 27th Conference on
Computer Communications, pp. 231–235, 13-18 April
2008.
[6] A. Banchs and X. Perez, “Providing throughput guaran-
tees in IEEE 802.11 wireless LAN,” Wireless Communi-
cations and Networking Conference, IEEE WCNC’02,
Vol. 1, pp. 130–138, 17-21 March 2002.
[7] S. Sehrawat, R. P. Bora, and D. Harihar, “Performance
analysis of QoS supported by enhanced distributed chan-
nel access (EDCA) mechanism in IEEE 802.11e,”
IAENG International Journal of Computer Science,
IJCS_33_1_6, Vol. 33, No. 1, February 2007.
[8] J. R. Yan, S. Y. Zhang, H. Long, and Y. F. Sun, “An
analytical model for EDCA mechanism,” Journal of Elec-
tronics & Information Technology, Vol. 30, No. 4, April
2008.
[9] X. Y. Zhou, H. F. W, L. M. Sun, et al., “MAC access
delay analysis of EDCA mechanism of wireless LANs,”
Journal of Software, Vol. 19, No. 8, pp. 2127−2139, 2008.
http://www.jos.org.cn/1000-9825/19/2127.htm.
[10] F. Keceli, I. Inan, and E. Ayanoglu, “Weighted fair up-
link/downlink access provisioning in IEEE 802.11e
WLANs,” ICC’08, IEEE International Conference on
Communications, pp. 2473−2479, 19-23 May 2008.
[11] S. H. Choi, J. del Prado, N. S. Shankar, et al., “IEEE
802.11 e contention-based channel access (EDCF) per-
formance evaluation,” ICC’03, IEEE International Con-
ference on Communications, pp. 1151−1156, 11-14 May
2003.
[12] L. Romdhani, Q. Ni, T. Turletti, “Adaptive EDCF: En-
hanced service differentiation for IEEE 802.11 wireless
ad-hoc networks,” IEEE Wireless Communications and
Networking, Conference, Vol. 2, pp. 16−20, March 2003.
C opyright © 2009 SciRes. IJCNS