International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No.

1, February 2012
DOI : 10.5121/ijwmn.2012.4110 139



PERFORMANCE ANALYSIS FOR BANDWIDTH
ALLOCATION IN IEEE 802.16 BROADBAND
WIRELESS NETWORKS USING BMAP QUEUEING
Said EL KAFHALI Abdelali EL BOUCHTI Mohamed HANINI and Abdelkrim HAQIQ
Computer, Networks, Mobility and Modeling laboratory
e-NGN research group, Africa and Middle East
FST, Hassan 1st University, Settat, Morocco
{kafhalisaid, a.elbouchti, haninimohamed, ahaqiq}@gmail.com

ABSTRACT
This paper presents a performance analysis for the bandwidth allocation in IEEE 802.16 broadband
wireless access (BWA) networks considering the packet-level quality-of-service (QoS) constraints.
Adaptive Modulation and Coding (AMC) rate based on IEEE 802.16 standard is used to adjust the
transmission rate adaptively in each frame time according to channel quality in order to obtain multi-
user diversity gain. To model the arrival process and the traffic source we use the Batch Markov Arrival
Process (BMAP), which enables more realistic and more accurate traffic modelling. We determine
analytically different performance parameters, such as average queue length, packet dropping
probability, queue throughput and average packet delay. Finally, the analytical results are validated
numerically.
KEYWORDS
IEEE 802.16; Quality of Service; Bandwidth Allocation; Performance Parameters; BMAP Process;
OFDMA; Queueing Theory; Adaptive Modulation and Coding.
1. INTRODUCTION
1.1. Reference system
IEEE 802.16 standard networks accommodate the increasing user demand to enable pervasive,
high-speed mobile internet access to a very large coverage area. Worldwide Interoperability for
Microwave Access (WiMAX), first standardised in 2004 [1] known as IEEE 802.16, can
provide broadband communications over wireless for various types of multimedia traffic, such
as video streaming, VoIP, FTP etc.
WiMAX presents a very challenging multiuser communication problem [12] many users in
the same geographic area will require high on-demand data rates in a finite bandwidth, with low
latency. Multiple access techniques allow different users to share the available bandwidth by
allotting each user some fraction of the total system resources. Due to the diverse nature of
anticipated WiMAX traffic, and the challenging aspects of the system deployment (mobility,
neighboring cells, and high required bandwidth efficiency), the multiple access problems are
quite complicated in WiMAX.
The IEEE 802.16 standard defines two types of operating mode for sharing the wireless
medium: Point-to-Multipoint (PMP) and Mesh. The PMP mode adopts a cellular architecture,
in this mode subscriber stations are scattered in the cellule around a central base station. There
are two directions: Downlink (from BS to SS) and Uplink (from SS to BS). Transmissions from
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
140



SSs are directed to and coordinated by the BS. On the other hand, in Mesh mode, the nodes are
organized ad hoc and scheduling is distributed among them.
The WiMAX standard [1] defines the physical layer specifications and the Medium Access
Control (MAC) signaling mechanisms. IEEE 802.16 uses two types of the modulation systems:
OFDM (Orthogonal Frequency Division Multiple) and OFDMA (Orthogonal Frequency
Division Multiple Access). OFDMA [12], extended OFDM, to accommodate many users in
the same channel at the same time, and it has been adopted as the physical layer transmission
technology for IEEE 802.16 based broadband wireless networks.
1.2. Related works
In order to promise the quality of real-time traffic and allow more transmission opportunity for
other traffic types, an Adaptive Bandwidth Allocation model (ABA) for multiple traffic classes
in IEEE 802.16 worldwide interoperability for microwave access networks was studied in [17].
The aim of work in [28] is to show how to exploit adaptive bandwidth allocation to increase
system utilization (for the system administrator) with controlled QoS degradation (for the
users). Instead of only focusing on bandwidth utilization or blocking/dropping probability, two
new user-perceived QoS metrics, degradation ratio and upgrade/degrade frequency, are
proposed. A Markov model is then provided to derive these QoS metrics. Using this model,
authors evaluate the effects of adaptive bandwidth allocation on user-perceived QoS and show
the existence of trade-offs between system performance and user-perceived QoS. Mathematical
tools were used in [7, 8 and 9] to study performances parameters of both the connection-level
and the packet-level for a model using two Connection Admission Control (CAC) schemes
considered at a subscriber station in a single-cell IEEE 802.16 environment in which the base
station allocates sub-channels to the subscriber stations in its coverage area.
For wireless mobile networks, the problem of providing packet-level QoS was studied quite
extensively in the literature. A scheduling mechanism for downlink transmission was proposed
in [32] to provide delay guarantee. In [15], authors proposed two credits based scheduling
schemes which can efficiently serve real time burst traffic with reduced latency. The effect of
the proposed schemes on latency, bandwidth utilization and throughput for real time burst flows
is compared with Round Robin scheduling scheme. In [4], the proposed intergraded model can
be applied to IEEE.16e. This model supports quality of service for request mechanism and data
transmission in the uplink phase in the presence of channel noise; the authors calculate the
performance parameters for single and multichannel wireless networks, like the requests
throughput, data throughput and the requests acceptance probability and data acceptance
probability. In [33], a dynamic fair resource allocation scheme was proposed to support real-
time and non-real-time traffic in cellular CDMA networks. In [34], authors considered a data
transmission system over a wireless channel, where packets are queued at the transmitter. Using
A Markov approximation, they studied the statistics of the packet dropping process due to
buffer overflow under automatic repeat request (ARQ) based error control scheme.
In [27], the authors consider a point-to-point wireless transmission where link layer ARQ is
used to counteract channel impairments. They presented an analytical model framework to
compute link-layer packet delivery delay statistics as a function of the packet error rate. An
adaptive cross-layer scheduler was proposed in [14] for multiclass data services in wireless
networks. The proposed scheduler uses the queuing information as well as it takes the physical
layer parameters into account so that the required QoS performances can be achieved. The
capacity of TDMA and CDMA-based broadband cellular wireless systems was derived in [30]
under constrained packet-level QoS.
In [13], an analytical model is proposed to study the impacts of the channel access parameters,
bandwidth configuration and piggyback policy on the performance. The impacts of physical
burst profile and non-saturated traffic have also been taken into account. It is observed by
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
141



simulations that the bandwidth utilization can be improved if the bandwidth for random channel
access can be properly configured according to the channel access parameters, piggyback
policy and network traffic. Besides, there isnt a single set of configurations that is always the
best for all the network scenarios.
The authors in [5] present a pipeline approach to grant bandwidth at the BS of an IEEE 802.16
FDD network with half-duplex SSs. Based on this, they proposed a grant allocation algorithm,
namely, the Half-Duplex Allocation (HDA) algorithm, which always produces a feasible grant
allocation provided that the sufficient conditions are met. Although there have been several
proposals for QoS scheduling frameworks and algorithms in IEEE 802.16 BWA networks in
the literature [24, 31], they mainly focus on the QoS architecture and scheduling algorithm in a
base station to satisfy diverse QoS requirements, rather than bandwidth request algorithm in a
subscriber station.
A previous researcher in an attempt to address bandwidth allocation in IEEE 802.16 was
reported by the authors in [10]. They considered a similar model in OFDMA based-WiMAX
but they modeled packet-level by MMPP process and they compared various QoS measures.
Since the introduction of Batch Markovian Arrival Process (BMAP) by Lucantoni [11], the
researchers [16, 21, and 29] prove that BMAP enables more realistic and more accurate traffic
modeling; it can also capture dependency in traffic processes and outperforms MMPP and
Poisson traffic models.
Since the incoming traffic in IEEE 802.16 has a self-similarity and a bursting nature causing
correlation in inter-arrival times -which influences the performance of the system- we are
motivated for using BMAP which can model such traffic correlation.
1.3. Aims of the paper
In this paper, we present a performance analysis for bandwidth allocation in IEEE 802.16
broadband wireless access networks considering the packet-level quality-of-service (QoS).
Adaptive modulation and coding (AMC) rate based on IEEE 802.16 standard is used to adjust
the transmission rate adaptively in each frame time according to channel quality in order to
obtain multi-user diversity gain. A queueing analytical model is developed based on a Discrete-
Time Markov Chain (DTMC) which captures the system dynamics in terms of the number of
packets in the queue. We assume that the arrival process is modelled by the Batch Markov
Arrival Process (BMAP) as the traffic source. Based on this model, various performance
parameters such as average queue length, packet dropping probability due to lack of buffer
space, the queue throughput, and the average queueing delay are obtained. Finally, the
analytical results are validated by numerical results.
1.4. Organisation of the paper
The rest of the paper is organized as follows: In Section 2, we briefly introduce QoS
architecture of IEEE 802.16 networks. Section 3 presents Modulation and Coding Schemes for
IEEE 802.16. Section 4 describes the system model. The formulation of the analytical model is
presented in Section 5. In section 6, different performance parameters are analytically
determined. Section 7 states numerical results. Finally, section 8 gives a conclusion of this
paper.
2. QOS ARCHITECTURE OF IEEE 802.16 NETWORKS
In this paper, we consider a point-to-point wireless mode (PMP) of IEEE 802.16, where a base
station (BS) serves a set of subscriber stations (SSs). The Uplink and the downlink are served
in the separate region of physical layer (OFDMA/TDD) frame .the downlink channel is in
broadcast mode, but an SS is only required to process data which are addressed to itself. In the
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
142



uplink sub-frame, on the other hand, the SSs transmit data to the BS in a Time Division
Multiple Access (TDMA) manner. Downlink and uplink sub-frames are duplexed using one of
the following techniques: (Frequency Division Duplex FDD), where downlink and uplink sub-
frames occur simultaneously on separate frequencies, and Time Division Duplex (TDD), where
downlink and uplink sub-frames occur at different times and usually share the same frequency.
SSs can be either full duplex or half-duplex.
IEEE 802.16e uses a connection-oriented medium access control (MAC) protocol which
provides a mechanism for the SSs to request bandwidth to the BS. IEEE 802.16e MAC supports
two classes of SS: grant per connection (GPC) and grant per SS (GPSS). In the case of GPC,
bandwidth is granted to a connection individually. In contrast, for GPSS, a portion of the
available bandwidth is granted to each of the SSs and each SS is responsible for allocating
bandwidth among the corresponding connections.
The lengths of the downlink and uplink sub-frames for each SS are determined by the BS and
broadcast to the SSs through downlink and uplink map messages (UL-MAP and DL-MAP) at
the beginning of each frame. Therefore, each SS knows when and how long to receive from and
transmit data to the BS. In the uplink direction, each SS can request bandwidth to the BS by
using BW-request packets.
WiMAX is associated with the IEEE 802.16 standard [1, 2, and 3], which defines five classes of
traffic flows representing different types of services in the following order: Unsolicited Grant
Service (UGS), Extended Real Time Polling Service (ertPS), Real Time Polling Service (rtPS),
Non-Real Time Polling Service (nrtPS), and Best Effort Service (BE). Each class has its QoS
mechanisms at the Media Access Control (MAC) layer to support the various applications.
UGS is designed to support real-time service flows that generate fixed-size data packets on a
periodic basis, such as VoIP without silence suppression. ertPS supports real-time applications
which generate variable-sized data packets periodically that require guaranteed data rate and
delay with silence suppression. rtPS supports real-time service flows that generate variable data
packets size on a periodic basis. nrtPS supports delay-tolerant data streams which are more
bursty in nature, such as FTP, in general, the nrtPS can tolerate longer delays and is insensitive
to delay jitter, but requires a minimum throughput. BE supports traffic with no QoS
requirements, such as email, and therefore may be handled on a resource-available basis.
3. MODULATION AND CODING SCHEMES FOR IEEE 802.16
Adaptive modulation and coding scheme (AMC) is supported in the WiMAX networks. The
basic idea of AMC is to maximize the data rates by adjusting the transmission parameters
according to the fluctuations in the channel.
The channel quality is determined by the instantaneous received Signal-to-Noise Ratio (SNR)
in each time slot. We assume that the channel is stationary over the transmission frame time.
Lower data rates are achieved by using Modulation Level and rate error correcting
corresponding to Rate 0 ID = (BPSK and 1/2). The higher data rates are achieved by using
Modulation Level and rate error correcting corresponding to Rate 6 ID = (64QAM and 3/4). In
all, there are 52 different possible configurations of modulation order and coding types and
rates [12], although most implementations of WiMAX will offer only a fraction of these. Table
1 lists these schemes represented by different rate IDs for IEEE 802.16 WiMAX Networks.
In an OFDMA system [12], each user will be allocated a block of subcarriers, each of which
will have a different set of SNR. Therefore, care needs to be paid to which constellation/coding
set is chosen based on the varying SNR across the subcarriers.
To determine the mode of transmission (i.e., modulation level and coding rate), an estimated
value of SNR at the receiver is used. In this case, the SNR at the receiver is divided into 1 N +
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
143



nonoverlapping intervals (i.e., 7 N = in WiMAX) by thresholds ( {0,1,..., })
n
n N where
0 1 1
...
N+
< < = . The subchannel is said to be in state n (i.e., rate ID n = will be used)
if
1 n n

+
< . To avoid possible transmission error, no packet is transmitted when
0
< .
Note that, these thresholds correspond to the required SNR specified in the WiMAX standard,
SNR, Signal-to-Noise Ratio [18, 23, and 25].
That is,
0 1
6.4, 9.4,..., 24.4
N
= = = (as shown in Table 1).
Table 1: IEEE 802.16 Profiles.
Rate ID Modulation Level
(Coding)
Information
Bits/Symbol
Required
SNR (db)
0 BPSK (1/2) 0.5 6.4
1 QPSK (1/2) 1 9.4
2 QPSK (3/4) 1.5 11.2
3 16QAM (1/2) 2 16.4
4 16QAM (3/4) 3 18.2
5 64QAM (2/3) 4 22.7
6 64QAM (3/4) 4.5 24.4
4. MODEL DESCRIPTION
4.1. Arrival Process Traffic
The BMAP has received considerable interest during the last few years. It was first introduced
by Neuts [11] as the versatile Markovian point Process. It generalizes Markovian Arrival
Process (MAP) introduced by Lucantoni et al. [20].
To capture the arrival process traffic, we use a BMAP process [6]. The arrivals in the BMAP is
directed by the irreducible continuous time Markov chain CTMC with a finite state space {0, 1,
, S}. Sojourn time of the CTMC in the state s has exponential distribution with parameter
s
.
After time expires, with probability
0
( , ') p s s the chain jumps into the state s without generation
of packets and with probability ( , ')
k
p s s the chain jumps into the state s and a batch consisting of
k packets is generated, 1 k . The introduced probabilities satisfy conditions:
0
( , ) 0, p s s = the
sum of the probabilities of all outgoing transitions has to be equal to 1,

' '
' '
1 0 0
'
( , ) ( , ) 1, 0 .
S S
k k
k s s
s s
p s s p s s s S

= = =

+ =

(1)
The BMAP is a two dimensional Markov process { ( ), ( )} A t J t on the state space
{( , ) / 0, 0 } i j i j S with infinitesimal generator given by:

0 1 2 3
0 1 2
0 1
0
D D D D
0 D D D
0 0 D D
0 0 0 D
| |
|
|
| =
|
|
|
\
K
K
K
K
M M M M O
(2)
where the matrices
0 '
D [ ], 0 S, 0 ' S
ss
D s s = has negative diagonal elements and non
negative off diagonal elements given by:
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
144



'
'
,
( , '), '
s
ss
s
s s
D
p s s s s

(3)
The matrices , 0
k
D k > are defined by:
, '
D [ ], 0 S, 0 ' S, 0
k k ss
D s s k = > (4)
where:
, '
( , '), 0 S, 0 ' S, 0.
k ss s k
D p s s s s k = > (5)
The matrix Ddefined by.
0
D D
k
k

=
=

, is an irreducible infinitesimal generator. We also
assume that
0
D
k
D , which ensures that arrivals will occur.
The variable ( ) A t counts the number of arrivals during [0, [ t and the variable ( ) J t represents
the phase of the arrivals process.

The steady-state probability vector
BMAP
of the CTMC with generator D can be calculated as
usual:
.D 0, . 1.
BMAP BMAP
e

= = (6)
Where 0

and e are row and column vectors consisting of zeros and units, respectively.
The mean steady-state arrival rate generated by the BMAP is:

1
D .
BMAP BMAP k
k
k e

=
=

(7)
More detail and results concerning this Process can be found in [19] for instance.
The probability ( , )
a s
f T of 0,1,..., a A = incoming packets, with Adenoting the maximum
packets number, that arrive with mean rate
s
within a time slot interval T is given by:
( )
( , )
!
s
T a
s
a s
e T
f T
a

= (8)
It is also essential for the condition
( )

!
s
T a
s
a A
e T
er s
a

+
=
<

always to stand, with er expressing

a sufficiently small number.
Note that the probability that aPoison arrivals with average rate
s
occur during an interval T
is given by the S S diagonal matrix
a
which is defined as:
1
2
( , )
( , )
=
( , )
a
a
a
a s
f T
f T
f T

(
(
(
(
(

O
(9)
4.2. System Model
We consider an infrastructure-based wireless access network, where connections are established
between a base station (BS) and multiple subscribers stations (SSs) through a TDMA/TDD
access mode using single carrier air-interface (as shown in Figure 1). Each subscriber station
serves multiple connections. For each connection a separate queue in SS with size X packets
is used for buffering the packets from higher layers. In particular, for one connection, there is a
queue for uplink and another queue for downlink transmissions from the SS and the BS,
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
145



respectively. We consider an SS of type GPC. Therefore, through the SS a certain amount of
bandwidth is reserved for each connection during bandwidth allocation.


Figure 1: System model
5. FORMULATION OF THE ANALYTICAL MODEL
5.1. State Space
The state of the queue is observed at the beginning of each frame. We assume that connection i
is allocated with
i
b units of bandwidth and a packets arriving during frame period f will not be
transmitted until frame period 1 f + at the earliest. The state space of the queue can be defined
as follows:
{( , ); 0 ,1 }, E x s x X s S =
(10)
where x , s represent, the number of packets in the queue, the state (phase) of an irreducible
continuous time Markov chain of BMAP arrival process respectively.
5.2. Transition Matrix for the Queue
The transition matrix M for the queue can be expressed as follows:
0,0 0,1 0,
,0 ,1 , ,
, ,



M



A
D D D D D D A
X A X A D X A X
m m m
m m m m
m m
+

=
L
M M O O
L L L
O O O O O O
L L
O O O O M
, ,

X X D X X
m m

(
(
(
(
(
(
(
(
(
(
(

L L
(11)
The rows of matrix M represent the number of packets in the queue and element
'
, x x
m inside
this matrix denotes the transition probability for the case when the number of packets in the
queue changes from x in the current frame to
'
x in the next frame. Also, the maximum
number of packets that can enter into or depart from the queue within a frame time is
represented with A and D respectively.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
146



The maximum number of transmitted packets within a time slot is given by
'
min( , ) D x D = .
Hence, if k represents the number of the successfully transmitted packets. The probability that
k packets will be transmitted in a timeslot is obtained by the matrix
k
T which is defined as
follows:
'
'
'
(1 ) ,
(1 ) ,
k x k
k
x
j x j
j U
x
k D
k
T
x
k D
j

=
| |
<
|
\
=

| |

=
|

(12)
where is the probability that a packets is successfully transmitted.
The elements in the matrix M can be obtained as follows:
, x x u a k
k a u
m T

=
=

(13)
, x x v a k
a k v
m T
+
=
=

(14)
, x x a k
k a
m T
=
=

(15)
for
'
1, 2,..., u D = and 1, 2,..., v A = where,
'
{0,1,..., } k D and {0,1,..., } a A represent the number of
departed packets and the number of packets arrivals, respectively.
With
, x x u
m

,
, x x v
m
+
and
, x x
m
we represent the probability that the number of packets in the queue
increases by u, decreases by v, and does not change, respectively.
The remaining rows { 1, 2,..., } x X A X A X = + + of the matrix M, include the occurrence where
some packets would be dropped due to lack of queue space. We calculate the probabilities as
follows:
'
, ,
for
A
x x v x x a
a v
m m x v X
+ +
=
= +

(16)
Which express that no packet has been dropped and occur when more incoming packets to an
already fully queue are dropped. Additionally, the last element of the main diagonal of M is
given by:
' '
, , ,
1
for
A
x x x x x x a
a
m m m x X
+
=
= + =

(17)
where
'
, x x
m is obtained for the case without any packets dropping.
Equations (16) and (17) indicate the case that the queue will be full if the number of incoming
packets is greater than the available space in the queue. In other words, the transition
probability to the state that the queue is full can be calculated as the sum of all the probabilities
that make the number of packets in queue equal to or larger than the queue size X .
6. PERFORMANCE PARAMETERS
The performance parameters are analytically calculated using the steady state probability of the
system. The vector of these probabilities is obtained by solving the system .M = and
.1 1 = , where1 is a column matrix of ones.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
147



The steady-state probabilities, denoted by ( , ) x s for the state that there are {0,1,..., } x X
packets in the queue, can be extracted from matrix as follows:
[ ]
( , ) , s 1,...,
X S
x s S

= =
(18)
The matrix contains the steady state probabilities corresponding to the number of packets in
the queue and the state (phase) of an irreducible continuous time Markov chain of BMAP
arrival process. Using the steady state probabilities, the various performance measures can be
obtained.
6.1. Average Queue Length
The average number of packets in the transmission queue is obtained as follows:
0 1
( ) ( , )
X S
x s
X E x x x s
= =
= =

(19)
6.2. Packet Dropping Probability
In order to compute the packet dropping probability (
drop
p ), we firstly obtain the number of
dropped packets per time slot, obtained using the average number of dropped packets per frame.
If x is the number of packets in the queue and this number increases by n, the number of
dropped packets
drop
X is:
] [ ,
( ). ( )
drop X n
X n X x x
+
= where
1
( )
0
A
if x A
x
otherwise

=

(20)
The average number of dropped packets per frame is obtained as follows:
, ,
0 1 1 1
( )
[ ] .( ( )). ( , )
drop
drop
X S A S
x x n s l
x s n X x l
X E X
m n X x x s
+
= = = + =
=
| |
=
|
\

(21)
where the term
, ,
1
[ ]
S
x x n s l
l
m
+
=
| |
|
\

in Equation (21) indicates the total probability that the number

of packets in the queue increases by n at every arrival phase. The probability
, x x n
m
+
is used
rather than the probability of packet arrival, because the packet transmission in the same frame
is considered.
After calculating the average number of dropped packets per frame, we can obtain the
probability that an incoming packet is dropped as follows:
drop
drop
BMAP
X
p

=
(22)
where
BMAP
is the mean steady state arrival rate generated by the BMAP (as obtained from
(7)).
6.3.Queue throughput
It measures the number of packets transmitted in one frame and can be obtained from:
(1 )
BMAP drop
p =
(23)


International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
148



6.4. Average Packet Delay
The average packet delay is defined as the number of frames that a packet waits in the queue
since its arrival before it is transmitted.
We use Littles formula [22] to obtain average packets delay as follows:
(1 )
BMAP drop
X X
D
P
= =

(24)
where is the throughput and X is the average queue length.
7. NUMERICAL RESULTS
In the next, performance parameters are numerically evaluated, using Matlab software.
7.1. Parameter Setting
We consider the system model depicted in section 4. Adaptive Modulation and Coding (AMC)
is used in which the modulation level and the coding rate are increased if the channel quality
permits. Table 1 lists these schemes represented by different rate IDs for IEEE 802.16.
The maximum number of packets that can be transmitted in one frame period is 150 packets per
frame.
The queue size is assumed to be 150 packets (i.e. 150 X = ).
For simplicity of computation we assume the maximum size of the batch to be 2, and the
matrices governing the state transitions of the BMAP are giving as follows:
0 1 2
1 1 1 1 1
2
2 4 2 2 4
, ,
1 1 1 1 1
1
8 4 8 4 4
D D D
| | | | | |

| | |
= = = | | |
| | |

| | |
\ \ \

Its sojourn times at each state are assumed to be exponentially distributed with rates
0
2.0 = ,
1
1.0 = respectively.
The performance parameters are measured respectively under different amount of allocated
bandwidthb , under different channel qualities with constant traffic intensity, and under
different traffic intensities with channel SNR in the range of rate 0 IDn = .
In this work, bandwidth b is defined (as in [10]) as the number of packets that can be
transmitted in one frame using rate 0 IDn = .
7.2. Results and Discussion
We first examine the impact of traffic intensity on bandwidth allocation. Variations in
throughput with traffic intensity are shown in Figure 2. When the traffic intensity increases, the
throughput increases until it becomes saturated. At this point (e.g., 3.0), the arriving packets
cannot be transmitted faster than the transmission rate that the channel quality allows.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
149




Figure 2: Throughput under traffic intensity.

Figure 3: Average queue length under traffic intensity.
.
Figure 4: Average delay under traffic intensity.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
150




Figure 5: packets dropping probability under traffic intensity.
Average queue length, average delay and packets dropping probability increase as the traffic
intensity increases (Figures 3, 4 and 5). Therefore, we can say that the increase of the
parameters previously mentioned is linked with the traffic intensity. On the other hand, those
parameters decrease as the channel quality improves (Figures 6, 7 and 8).

Figure 6: Average queue length under different channel qualities.

Figure 7: Average delay under different channel qualities.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
151




Figure 8: Packet dropping probability under different channel qualities.

Figure 9: Throughput under different channel qualities.
When the channel quality improves, the transmitter can utilize higher modulation level and
code rate to increase throughput (Figure 9). Note that if transmission rate is high enough to
accommodate most of the arrival traffic, increased amount of allocated bandwidth or better
channel quality will not impact the queue throughput since all the packets can be transmitted
within a few frames. Moreover, different amount of allocated bandwidth results in different
throughput.
8. CONCLUSION
In this paper, a queuing analytical model based on a Discrete-Time Markov Chain (DTMC) has
been presented to analyze the packet-level performance in IEEE 802.16 broadband wireless
access networks considering adaptive modulation and coding at the OFDMA physical layer. In
the considered WiMAX system model, a base station serves multiple subscriber stations, and
the base station allocates a certain number of sub-channels for each subscriber station.
To model the arrival process and the traffic sources we use the Batch Markov Arrival Process
(BMAP), which enables more realistic and more accurate traffic modelling.
Using this queuing model, the impact of different traffic sources and the impact of
channel quality on QoS parameters, such as average queue length, packet dropping
probability, queue throughput and average packet delay, are analytically studied.
Finally, the analytical results are validated numerically.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
152



REFERENCES
[1] IEEE 802.16 WG, IEEE standard for local and metropolitan area networks part 16: Air interface for
fixed broadband wireless access systems, IEEE 802.16 Standard, June 2004.
[2] IEEE 802.16 WG, IEEE standard for local and metropolitan area networks part 16: Air interface for
fixed and mobile broadband wireless access systems, Amendment 2, IEEE 802.16 Standard,
December 2005.
[3] Mobile WiMAX Part I: A Technical Overview and Performance Evaluation, August, 2006.
[4] Abdelsalam Amer and Fayez Gebali, General Model for Single and Multiple Channels WLANs
with Quality of Service Support, International Journal of Wireless & Mobile Networks (IJWMN),
Vol 1, No 2, November 2009.
[5] Andrea Bacioccola, Claudio Cicconetti, Alessandro Erta, Luciano Lenzini, and Enzo Mingozzi,
Alessandro Erta, Luciano Lenzini, and Enzo Mingozzi, Bandwidth Allocation with Half-Duplex
Stations in IEEE 802.16 Wireless Networks, IEEE transactions on mobile computing, vol. 6, no. 12,
december 2007.
[6] B. Baynat, S. Doirieux, G. Nogueira, M. Maqbool, and M. Coupechoux, An efcient analytical
model for wimax networks with multiple trafc proles, in Proc. of ACM/IET/ICST IWPAWN,
September 2008.
[7] Abdelali EL BOUCHTI, Abdelkrim HAQIQ and Said EL KAFHALI, Analysis of Quality of
Service Performances of Connection Admission Control Mechanisms in OFDMA IEEE 802.16
Network using BMAP Queuing, International Journal of Computer Science Issues (IJCSI), Vol. 9,
Issue 1, No 2, ISSN (Online): 1694-0814, pp. 302-310, January 2012.
[8] Abdelali EL BOUCHTI, Said EL KAFHALI, and Abdelkrim HAQIQ Performance Modeling and
Analysis of Connection Admission Control in OFDMA based WiMAX System with MMPP
Queueing World of Computer Science and Information Technology Journal (WCSIT), Vol. 1, No.
4, pp. 148-156 , 2011.
[9] Abdelali EL BOUCHTI, Said EL KAFHALI, and Abdelkrim HAQIQ Performance Analysis of
Connection Admission Control Scheme in IEEE 802.16 OFDMA Networks International Journal of
Computer Science and Information Security (IJCSIS), Vol. 9, No. 3, pp. 45-51 March 2011.
[10] M. Fathi, H. Taheri ,Queuing analysis for dynamic bandwidth allocation in IEEE 802.16 standard,
3rd IEEE International symposium on wireless pervasive computing , 7-9 May 2008.
[11] M.F.Neuts,Aversatile Markovian Point Process, J.Appl.Prob, 16:764-779, 1979.
[12] Jeffrey G. Andrews, Orthogonal Frequency Division Multiple Access (OFDMA), book chapter,
July 29, 2006.
[13] Jianhua He, Kun Yang, Ken Guild, and Hsiao-Hwa Chen, On Bandwidth Request Mechanism with
Piggyback in Fixed IEEE 802.16 Networks, IEEE transactions on wireless communications, vol. 7,
no. 12, december 2008.
[14] K.B. Johnsson and D.C. Cox, An Adaptive Cross-Layer Scheduler for Improved QoS Support of
Multiclass Data Services on Wireless Systems, IEEE J. Selected Areas in Comm., vol. 23, no. 2, pp.
334- 343, Feb. 2005.
[15] C.Kalyana Chakravarthy and Prof. P.V.G.D. Prasad Reddy , Selfless Distributed Credit Based
Scheduling For Improved QoS In IEEE 802.16 WBA Networks, International Journal of Wireless
& Mobile Networks (IJWMN), Vol 1, No 2, pp. 118-125, November 2009.
[16] A. Klemm, C. Lindemann, and M. Lohmann, "Traffic Modeling of IP Networks Using the Batch
Markovian Arrival Process", in Proc. Computer Performance Evaluation / TOOLS, 2002, pp.92-
110.
[17] T.-L. Sheu K.-C. Huang, Adaptive bandwidth allocation model for multiple traffic classes in IEEE
802.16 worldwide interoperability for microwave access networks, The Institution of Engineering
and Technology, Vol. 5, Iss. 1, pp. 9098, 2011.
[18] Q.Liu, S. Zhou, and G. B. Giannakis, Queuing with adaptive modulation and coding over wireless
links: cross-layer analysis and design, IEEE Transactions on Wireless Communications, vol. 4, no.
2, pp. 11421153, May 2005.
[19] D.M. Lucantoni, The BMAP/G/1 queue : A tutorial, In L.Donatiello and R.Nelson, editors,
Performance Evaluation of Computer and Communications Systems, pages 330-358, lectures Notes
in Computer Science 729, Springer Verlag, 1993.
[20] D.M.Lucantoni, K.S.Meier-Hellstern, and M.F.Neuts,A single server queue with server vacations
and a class of non-renewal arrival processes, Adv.Appl.Prob,22: 676-705,1990.
[21] D. M. Lucantoni, New Results on the Single Server Queue with a Batch Markovian Arrival Process,
Comm. in Statistics: Stochastic Models 7, 1-46, 1991.
International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
153



[22] R. Nelson, Probability, stochastic process, and queueing theory, Springer-Verlag, third printing,
2000.
[23] D. Niyato and E. Hossain, Delay-based admission control using fuzzy logic for OFDMA broadband
wireless networks, in Proc. IEEE ICC06, June 2006.
[24] D. Niyato, E. Hossain, Queue-Aware Uplink Bandwidth Allocation and Rate Control for Polling
Service in IEEE 802.16 Broadband Wireless Networks, IEEE trans. on Mobile Computing, Vol. 5,
No. 6, June 2006.
[25] D. Niyato and E. Hossain, "Connection admission control in OFDMA-based WiMAX networks:
Performance modeling and analysis," invited chapter in WiMax/MobileFi: Advanced Research and
Technology, (Ed. Y. Xiao), Auerbach Publications, CRC Press, December 2007.
[26] D. Niyato, E. Hossain, QoS-aware bandwidth allocation and admission control in IEEE 802.16
broad band wireless access networks: A non-cooperative game theoretic approach, Comut. Netw.
(2007), doi:10.1016/jcomnet.2007.01.031.
[27] M. Rossi and M. Zorzi, Analysis and Heuristics for the Characterization of Selective Repeat ARQ
Delay Statistics over Wireless Channels, IEEE Trans. Vehicular Technology, vol. 52, no. 5, pp.
1365-1377, Sept. 2003.
[28] C.T. Chou and K.G. Shin, Analysis of Adaptive Bandwidth Allocation in Wireless Networks with
Multilevel Degradable Quality of Service, IEEE Trans. Mobile Computing, vol. 3, no. 1, pp. 5-17,
Jan.-Mar. 2004.
[29] Takine, T. and Takahashi, Y. 1998. On the relationship between queue lengths at a random instant
and at a departure in the stationary queue with BMAP arrivals. Stoch. Mod. 14 601610.
[30] S.V. Krishnamurthy, A.S. Acampora, and M. Zorzi, On the Radio Capacity of TDMA and CDMA
for Broadband Wireless Packet Communications, IEEE Trans. Vehicular Technology, vol. 52, no.
1, pp. 60-70, Jan. 2003.
[31] K. Wongthavarawat and A. Ganz, Packet scheduling for QoS support in IEEE 802.16 broadband
wireless access systems, International Journal of Communication systems, vol. 16, pp. 8196, 2003.
[32] D. Wu and R. Negi, Downlink Scheduling in a Cellular Network for Quality-of-Service
Assurance, IEEE Trans. Vehicular Technology, vol. 53, no. 5, pp. 1547-1557, Sept. 2004.
[33] L. Xu, X. Shen, and J.W. Mark, Fair Resource Allocation with Guaranteed Statistical QoS for
Multimedia Traffic in Wideband CDMA Cellular Network, IEEE Trans. Mobile Computing, vol. 4,
no. 2, pp. 166-177, Mar.-Apr. 2005.
[34] M. Zorzi, Packet Dropping Statistics of a Data-Link Protocol for Wireless Local Communications,
IEEE Trans. Vehicular Technology, vol. 52, no. 1, pp. 71-79, Jan. 2003.


Authors
Said EL KAFHALI received the B.Sc. degree in Computer Sciences from the
University of Sidi Mohamed Ben Abdellah, Faculty of Sciences Dhar El- Mahraz,
Fez, Morocco, in 2005, and a M.Sc. degree in Mathematical and Computer
engineering from Hassan 1
st
University, Faculty of Sciences and Techniques (FSTS),
Settat, Morocco, in 2009. He has been working as professor of Computer Sciences in
high school since 2006, Settat, Morocco. Currently, he is working toward his Ph.D. at
FSTS. His current research interests performance evaluation, analysis and simulation
of Quality of Service in mobile networks.


Abdelali EL BOUCHTI received the B.Sc. degree in Applied Mathematics from the
University of Hassan 2
nd
, Faculty of Sciences Ain chock, Casablanca, Morocco, in
2007, and M.Sc. degree in Mathematical and Computer engineering from the Hassan
1
st
University, Faculty of Sciences and Techniques (FSTS), Settat, Morocco, in 2009.
Currently, he is working toward his Ph.D. at FSTS. His current research interests
include performance evaluation and control of telecommunication networks, stochastic
control, networking games, reliability and performance assessment of computer and
communication systems.


International Journal of Wireless & Mobile Networks (IJWMN) Vol. 4, No. 1, February 2012
154



Mohamed HANINI is currently pursuing his PhD. Degree in the Department of
Mathematics and Computer at Faculty of Sciences and Techniques (FSTS), Settat,
Morocco. He is member of e-ngn research group. His main research areas are:
Quality of Service in mobile networks, network performance evaluation.



Dr. Abdelkrim HAQIQ has a High Study Degree (DES) and a PhD (Doctorat
d'Etat) both in Applied Mathematics from the University of Mohamed V, Agdal,
Faculty of Sciences, Rabat, Morocco. Since September 1995 he has been working as
a Professor at the department of Mathematics and Computer at the faculty of
Sciences and Techniques, Settat, Morocco. He is the director of Computer, Networks,
Mobility and Modeling laboratory and a general secretary of e-NGN research group,
Moroccan section. He was the chair of the second international conference on Next
Generation Networks and Services, held in Marrakech, Morocco 8 - 10 July 2010. He
is also a TPC member and a reviewer for many international conferences.
Professor Haqiq' interests lie in the area of applied stochastic processes, stochastic control, queueing
theory and their application for modeling/simulation and performance analysis of computer
communication networks.
From January 98 to December 98 he had a Post-Doctoral Research appointment at the department of
systems and computers engineering at Carleton University in Canada. He also has held visiting positions
at the High National School of Telecommunications of Paris, the universities of Dijon and Versailles St-
Quentin-en-Yvelines in France, the University of Ottawa in Canada and the FUCAM in Belgium.