RESOURCE ALLOCATION IN WIMAX MESH NETWORKS STEPHEN ATAMBIRE NSOH Bachelor of Science, Kwame Nkrumah University of Science and Technology, 2009 A Thesis Submitted to the School of Graduate Studies of the University of Lethbridge in Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE Department of Mathematics and Computer Science University of Lethbridge LETHBRIDGE, ALBERTA, CANADA c Stephen Atambire Nsoh, 2012
85
Embed
RESOURCE ALLOCATION IN WIMAX MESH NETWORKS STEPHEN ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
RESOURCE ALLOCATION IN WIMAX MESH NETWORKS
STEPHEN ATAMBIRE NSOHBachelor of Science, Kwame Nkrumah University of Science and Technology, 2009
A ThesisSubmitted to the School of Graduate Studies
of the University of Lethbridgein Partial Fulfillment of the
Requirements for the Degree
MASTER OF SCIENCE
Department of Mathematics and Computer ScienceUniversity of Lethbridge
t← 1 // t is timeslotwhile exists any demand( j) > 0 for any SS j do
A← set of links that have data to transmitLu←�s←one-hop link with data and whose children have least trafficA← A−{s}if Channel Assignment(s) is true then
Lu← Lu∪{s}A← A− children(s)
end ifif demand(one-hop links)< 2 then
l← argmax demand(i)i∈A :hopcount(i)=2
A← A−{l}if Channel Assignment(l) is true then
Lu← Lu∪{l}A← A− children(l)
end ifend ifwhile A 6=� do
l← argmax p(i)i∈A
A← A−{l}if Channel Assignment(l) is true then
Lu← Lu∪{l}A← A− children(l)
end ifend whileadjust demands for each i ∈ Lut← t +1
end while
43
If l transmits on c and this causes n links to remove c from their channel list, we refer to n
as the interference degree of c.
Definition 3 Interference degree (In) of a channel c with respect to a link l is the number
of links that are blocked from transmitting on c as a result of l transmitting on c.
To assign a channel to a link, the goal is to choose the channel that has the least interference
degree In. The channel assignment algorithm is presented in Algorithm 4.
Algorithm 4 Channel AssignmentInput: l // link to be assigned channelOutput: channel assignment
if Cl 6=� thenc(l)← argmin In(ch)ch∈Cl
Ci←Ci−{ch}∀ i ∈ S(l)return true
elsereturn false
end if
(1,2)
(1,2)
(1,2)
(1,2)
d
ab
c
f
g
e
(a)
1
(1,2)
(1,2)
(2)
(b)
(1)
(1,2)
2
1
(c)
(1)
(1,2)
2
1
(d)
Figure 4.2: Example channel assignment
Consider the diagram shown in Figure 4.2 with the dotted lines indicating the interfer-
ence between links. Assume we want to assign channels to the links d,b,g and f and that
there are only two channels available. At the start, each link has in its link set two channels
as indicated on Figure 4.2(a). When link d is considered, we have to choose between chan-
nel 1 or two. Since both channels have the same interference degree (1), we will assign
44
channel 1 to link d. As a result, we remove channel 1 from the channel set of link b as
indicated on Figure 4.2(b). Next, we consider link b and since there is only one channel
available, channel 2 will be assigned. We then remove channel 2 from the channel set of
link f as shown in Figure 4.2(c). The next link to be considered is link g which still has two
channels in its channel set as shown in Figure 4.2(d). At this point, assigning channel 1 to
link g will result in the channel set of link f being empty. However, since the interference
degree of channel 2 is the least( zero), we will assign channel 2 to link g. This will allow
link f to be assigned channel 1.
4.6 Experiments
In this section, we evaluate our scheduling framework through extensive experiments using
random as well regular square grid graphs. The experiment set up is the same as described
in section 3.5.1. In each experiment, we compute routes using ILMR and then perform
scheduling using throughput aware scheduling (TS), maximum degree first select (MDFS)
and nearest select (NS) [17]. We also record a trivial lower bound which is the total num-
ber of packets in the network. We compare the schemes using metrics such as length of
schedule and channel utilization ratio.
4.6.1 Length of Schedule
We compare the length of schedule produced by the different scheduling schemes with
limited and unlimited number of channels. Figure 4.3 shows the results obtained using
just one channel. The results indicate that TS recorded the least length of schedule and
outperformed MDFS and NS in all instances. With a small number of nodes, TS produced
the same results as the bound. As the number of nodes increases, TS produced results about
45
95% of the lower bound in the regular graphs and about 92% in the random graphs while
MDFS and NS achieved approximately 90% and 87% respectively in the regular graphs.
For random graphs, MDFS achieved about 88% while NS achieved 87% of the bound
For the case of unlimited number of channels, there was no much difference in the
performance recorded by the different algorithms. TS still produced the best results and
was about 98% close to the bound.
4.6.2 Channel Utilization Ratio
We record the channel utilization ratio on different graphs for the different algorithms as
shown in Figure 4.5. The results follow the same trend. TS outperformed MDFS and NS
in all instances. MDFS slightly outperformed NS when the number of nodes was higher.
46
200
400
600
800
1000
1200
1400
60 80 100 120 140 160 180 200 220 240 260
Leng
th o
f Sch
edul
e
Number of nodes
TSMDFS
NSLower Bound
(a) regular graphs with 1 channel
150
200
250
300
350
400
450
50 60 70 80 90 100
Leng
th o
f Sch
edul
e
Number of nodes
TSMDFS
NSLower Bound
(b) random graphs with 1 channel
Figure 4.3: Limited number of channels
47
150
200
250
300
350
400
50 60 70 80 90 100
Leng
th o
f Sch
edul
e
Number of nodes
TSMDFS
NSLower Bound
Figure 4.4: Random graphs with Unlimited number of channels
0.075
0.08
0.085
0.09
0.095
0.1
0.105
0.11
0.115
0.12
50 60 70 80 90 100
CU
R
Number of nodes
TSMDFS
NS
Figure 4.5: Number of nodes vs CUR
48
Chapter 5
Quality of Service
5.1 Introduction
In this section, we present a model for another important design challenge in WiMAX
mesh networks known as QoS provisioning. In recent times, there has been a tremendous
increase in the types of applications in internet protocol (IP) networks. Besides the tra-
ditional file transfer, email and web browsing, multimedia applications are also becoming
increasingly popular. These applications send large amounts of audio and video streams
with variable bandwidth and delay requirements. For example, an email application may
not need any guarantee except reliable delivery of the message. A VoIP application on the
other hand will require low latency (delay) while a video streaming application may toler-
ate long delay but require relatively high bandwidth. Accommodating all these will require
QoS guarantees. QoS provisioning in mesh networks is thus a very important ingredient
towards the vision of globally connected heterogeneous networks.
QoS provisioning is a challenging task faced by the research community. Wireless
networks are generally less efficient and unpredictable compared to wired networks, which
makes QoS provisioning a bigger challenge for wireless communication [4]. For instance,
the wireless medium is often characterised by limited bandwidth and high packet error rate
which together limit the capacity of the network to provide QoS guarantees.
5.2 WiMAX QoS Specification
There is no formal definition for QoS although several standards have proposed different
definitions. In the field of telephony, the International Telecommunication Union (ITU) de-
49
fines QoS as a set of quality requirements on the collective behavior of one or more objects.
This definition lists six primary components: support, operability, accessibility, retainabil-
ity, integrity and security [31]. The ITU again defines QoS in the field of data networking
as the probability of the telecommunication network meeting a given traffic contract. QoS
provisioning encompasses providing Quality of Service to end users in terms of several
generic parameters [31]. In WiMAX, such parameters include throughput, average delay,
average jitter and packet loss.
1. Throughput is a measure of the data rate (bits per second) generated by the applica-
tion.
T P =
n
∑i=1
PacketSizei
PAn−PS0(5.1)
Equation 5.1 shows the calculation of throughput T P, where PacketSizei is the packet
size of the ith packet reaching the destination, PS0 is the time when the first packet
left the source and PAn is the time when the last packet arrived.
2. Average delay or latency is the time taken by packets to travel from source node
to destination node. The principal sources of delay are source processing delay,
propagation delay and destination processing delay. The calculation of delay is show
in equation 5.2 where PAi is the packet arrival time of the ith packet, PSi is the packet
start time and n is the total number of packets.
AverageDelay =∑
i(PAi−PSi)
n(5.2)
3. Jitter is the variation in the delay introduced by the components along the commu-
nication path [31]. It is the variation in time between packets’ arrival. Jitter gives
a measure of the consistency and stability of a network. Equation 5.3 shows the
50
calculation of jitter.
AverageJitter =
n
∑i=1
((PAi+1−PSi+1)− (PAi−PSi))
n−1(5.3)
4. Packet loss affects the perceived quality of the application. Several causes of packet
loss or corruption would be bit errors in an erroneous wireless network or insuffi-
cient buffers due to network confestion when the channel becomes overloaded [31].
Equation 5.4 shows the calculation of packet loss.
PacketLoss = ∑LostPacketSizei
∑PacketSize j×100 (5.4)
5. Traffic Priority- this parameter specifies the priority assigned to a service flow. Given
two service flows identical in all QoS parameters besides priority, the higher priority
service flow should be given lower delay and higher buffering preference [2].
Among the numerous proposals brought forward by WiMAX, perhaps the most attrac-
tive is its ability to deliver Asynchronous Transfer Mode (ATM) like connection oriented
QoS guarantees in broadband wireless networks. This is due to the increase in the demands
for multimedia content delivered wirelessly. WiMAX is designed to support a wide range
of applications which require different levels of quality of service. To accommodate these
applications, the IEEE 802.16 standard [2] defines four different classes of traffic: unso-
licited grant service (UGS), real-time polling service (rtPS), non-real-time polling service
(nrtPS) and best effort service (BE).
• UGS
This service class is designed to support real-time applications that generate fixed
data packets on periodic basis, such as T1/E1 and VoIP without silence suppression
51
[2]. The BS allocates fixed number of slots to UGS at periodic intervals regardless
of current estimation backlog. UGS requests are granted bandwidth without polling
or contention. These requests have the most stringent QoS requirements.
• rtPS
rtPS class supports real-time applications that generate variable size data packets on a
periodic basis, such as moving pictures experts group (MPEG) streaming video. The
BS provides periodic dedicated request opportunities for SSs to meet the applications
real-time demands [32]. Unlike UGS, rtPS connections have to always notify the BS
of their current bandwidth requirements.
• nrtPS
nrtPS class is designed to support delay tolerant data stream and consists of vari-
able sized data packets which require a minimum data rate. An example of nrtPS
application is FTP.
• BE
This service class is designed to support data streams for which no minimum service
level is required and therefore may be handled on a space-available basis, such as
HTTP. The SS is allowed to use contention request opportunities as well as unicast
request opportunities for BE service requests.
In addition to the definition of the different service classes, the standard also specifies QoS
parameters for each service class as illustrated on Table 5.1.
52
Table 5.1: WiMAX QoS Parameters
Service ClassQos parameter
Minimum Rate Maximum Rate Latency Jitter PriorityUGS X X XrtPS X X X XnrtPS X X XBE X X
5.3 Related Work
Several works [16, 17, 14, 18] have been proposed for centralized scheduling and channel
assignment in WiMAX networks. While [16, 17, 14] consider single transceiver with mul-
tiple channels, [18] consider multi-transceiver and multiple channels. The goal of these
algorithms is to minimize the length of schedule. None of these however takes into con-
sideration the different WiMAX service classes. Several generalized schedulers like Fair
path← path of request (links arranged in reverse order)packets← minimum bandwidth of requestwhile packets > 0 do
temp← pathfor each l in temp do
compute feasible slot intervalFind an available slot from feasible slot intervalif no available slot then
return falseend if
end forend whilereturn true
64
5.6 Experiments
5.6.1 Introduction
In this section, we perform extensive simulation to evaluate the algorithms proposed in this
chapter. We compare our routing schemes to the routing scheme proposed in [23] which we
label as IR. We compare the algorithms using several metrics like throughput, acceptance
rate and packet drop rate.
5.6.2 Simulation Setup
We used a custom simulator written in C++ to evaluate our proposed scheme. We per-
formed the experiments using a 25 node regular grid graph. The arrival of requests fol-
lows a Poisson distribution with mean arrival rate of λ. For each request, the source node
and service class are uniformly distributed. The life time of requests is exponentially dis-
tributed with mean life time of 1000 frames. The bandwidth request size of rtPS and nrtPS
is uniformly distributed between the minimum and maximum bandwidth and its duration
follows an exponential distribution with mean of 20 frames. The simulation was performed
for 5000 frames and we assume the length of each frame to be 10ms. We use 50 subchan-
nels and 200 slots in the data subframe. The QoS parameters of the different service classes
are shown in Table 5.2
Table 5.2: QoS Parameters of different service classesService Class Minimum Bandwidth Maximum Bandwidth Delay Jitter
(in slots) (in slots) (in slots) (in slots)UGS 20−30 20−30 200 10rtPS 15−20 35−45 250nrtPS 15−20 35−45BE 100
65
40
50
60
70
80
90
100
0.05 0.06 0.07 0.08 0.09 0.1 0.11
Acc
epta
nce
ratio
(%
)
Arrival rate (requests/frame)
IRILR
ILMR
Figure 5.3: Acceptance ratio vs Arrival rate of requests
5.6.3 Acceptance Ratio
We investigate the acceptance ratio of the different routing schemes. For an incoming re-
quest, we compute the route based on the three routing schemes before it goes through call
admission. We define the acceptance ratio as the ratio of the number of accepted requests to
the total number of requests. Since BE requests do not go through call admission, we define
the acceptance ratio only in terms of UGS, rtPS and nrtPS requests. Figure 5.6.3 illustrates
the results obtained. The results indicate that ILMR produces the best acceptance ratio fol-
lowed by ILR. IR does not ensure load sharing leading to congestion on some links in the
network. As a result, more requests are rejected and this accounts for its poor performance.
As expected, the results indicate that load sharing metrics are very useful in QoS provision-
ing when the network is congested. By using session based routing, the acceptance ratio
was increased by 15%.
66
60000
70000
80000
90000
100000
110000
120000
130000
0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12
Thr
ough
put (
slot
s)
Arrival rate (requests/frame)
IRILR
ILMR
Figure 5.4: Throughput vs Arrival rate of requests
5.6.4 Throughput
We define throughput as the total number of packets received at the BS during the sim-
ulation time. Since ILMR accepts more requests, it produces the highest throughput fol-
lowed by ILR. IR produced the least throughput as indicated on Figure 5.4. The results
also indicate that load sharing has the potential to increase the efficiency of the use of net-
work resources. ILMR increased the network throughput by approximately 25% while ILR
achieved a 15% increase.
5.6.5 Packet Drop Ratio
In this section, we investigate the packet drop rate of the different routing schemes. We
define the packet drop rate as the ratio of the number of rtPS packets that are dropped to
the total number of rtPS packets. An rtPS packet is dropped if it fails to reach the BS by
its deadline. Figure 5.5 illustrates the results for varying arrival rates. As the arrival rate
67
3
4
5
6
7
8
9
0.05 0.06 0.07 0.08 0.09 0.1 0.11 0.12
Per
cent
age
drop
(%
)
Arrival rate (requests/frame)
IRILR
ILMR
Figure 5.5: Packet drop vs Arrival rate of requests
increases, the packet drop increases for all the routing schemes. The results indicate that
ILMR and ILR outperformed IR for the different arrival rates with ILMR producing the
least packet drop. ILR and ILMR are able to allocate more extra slots to rtPS requests
resulting in fewer packets missing their deadlines. By using session based routing, the
percentage of packets dropped was approximately halved when the network was congested.
68
Chapter 6
Conclusion And Future Work
In this thesis, we study the resource allocation problem in WiMAX mesh networks. We pro-
pose a joint routing, centralized scheduling and channel assignment scheme for WiMAX
mesh.
We present routing schemes that use a metric combining interference and load sharing.
We depart from classical tree based routing by constructing session based routes and we
quantify the gains when QoS guarantees are considered. We also propose a fast and effec-
tive heuristic algorithm for link scheduling. Our scheduling aims to find a shortest length
schedule for all data to reach the BS by keeping the BS busy in each timeslot. We present a
simple channel assignment algorithm inspired by a constraint programming heuristic which
proves effective in maximizing the number of concurrent transmissions. We compare our
scheme to a simple combinatorial bound through extensive simulations. Results from our
experiments indicate that our scheme improves the network performance. Our session
based routing and link scheduling produce results close to 90% of the trivial lower bound
while our tree based routing achieved about 85% of the bound.
We also investigate the impact of routing, link scheduling, channel allocation and CAC
on QoS provisioning in WiMax mesh networks. While several works consider these prob-
lems in isolation, we provide a comprehensive framework that considers all. For routing,
we consider two schemes that incorporate interference and load sharing. One scheme con-
structs routing tree and the other assigns a path for each request without constraining the
union of all routes to form a tree. We propose a link scheduling that is also used as CAC
for new requests and considers all classes of service. We provide simulations results which
indicate that load sharing metrics are indeed useful in QoS provisioning when the network
is congested. Our session based routing scheme provided significant improvement in net-
69
work performance. By using the session based routing, the acceptance ratio increased by
approximately 15% while the percentage of packets dropped was almost halved.
Future work on WiMAX resource allocation can be driven in several directions. This
thesis performs routing and scheduling separately which are sub optimal. In our future
work, we will like to consider a scheme that performs both routing and scheduling at the
same time. We will also like to make some extension to our QoS provisioning. As we
considered unlimited buffer sizes in this work, we will study QoS and buffer management
in our future work. We will also consider a CAC scheme that performs back tracking and
changes slot assignment of accepted requests in order to find a feasible assignment for a
new request.
70
Bibliography
[1] I. Akyildiz and X. Wang, “A survey on wireless mesh networks,” CommunicationsMagazine, IEEE, vol. 43, no. 9, pp. S23–S30, 2005.
[2] I. . W. Group et al., “Ieee standard for local and metropolitan area networks, part 16:Air interface for fixed broadband wireless access systems,” IEEE Standard, pp. 802–16, 2004.
[3] G. Narlikar, G. Wilfong, and L. Zhang, “Designing multihop wireless backhaul net-works with delay guarantees,” Wireless Networks, vol. 16, no. 1, pp. 237–254, 2010.
[4] Y. Zhang, Wimax network planning and optimization. Auerbach Publications, 2009.
[5] I. Akyildiz, X. Wang, and W. Wang, “Wireless mesh networks: a survey,” Computernetworks, vol. 47, no. 4, pp. 445–487, 2005.
[6] X. Li, Wireless Ad Hoc and Sensor Networks: Theory and Applications. CambridgeUniversity Press, 2008.
[7] R. Bruno, M. Conti, and E. Gregori, “Mesh networks: commodity multihop ad hocnetworks,” Communications Magazine, IEEE, vol. 43, no. 3, pp. 123–131, 2005.
[8] D. Ghosh, A. Gupta, and P. Mohapatra, “Scheduling in multihop wimax networks,”ACM SIGMOBILE Mobile Computing and Communications Review, vol. 12, no. 2,pp. 1–11, 2008.
[9] B. Kumar and P. Gupta, “Scheduling algorithms in a wimax network,” in AdvancedComputing & Communication Technologies (ACCT), 2012 Second International Con-ference on, pp. 457–462, IEEE, 2012.
[10] P. Djukic and S. Valaee, “Centralized scheduling algorithms for 802.16 mesh net-works,” Wimax/MobileFi: Advanced research and technology, 2007.
[11] J. He, K. Yang, K. Guild, and H. Chen, “Application of ieee 802.16 mesh networksas the backhaul of multihop cellular networks,” Communications Magazine, IEEE,vol. 45, no. 9, pp. 82–90, 2007.
[12] W. Stallings, Wireless communications and networking. Prentice Hall, 2002.
[13] S. Deb, V. Mhatre, and V. Ramaiyan, “Wimax relay networks: opportunistic schedul-ing to exploit multiuser diversity and frequency selectivity,” in Proceedings of the14th ACM international conference on Mobile computing and networking, pp. 163–174, ACM, 2008.
71
[14] P. Djukic and S. Valaee, “Delay aware link scheduling for multi-hop TDMA wirelessnetworks,” IEEE/ACM Transactions on Networking (TON), vol. 17, no. 3, pp. 870–883, 2009.
[15] Y. Qassem, A. Al-Hemyari, C. Ng, N. Noordin, and M. Rasid, “Review of networkrouting in ieee 802.16 wimax mesh networks,” Australian Journal of Basic and Ap-plied Sciences, vol. 3, no. 4, pp. 3980–3996, 2009.
[16] Y. Tang, Y. Yao, and X. Lin, “A joint centralized scheduling and channel assignmentscheme in wimax mesh networks,” in Proceedings of the 2009 International Con-ference on Wireless Communications and Mobile Computing: Connecting the WorldWirelessly, pp. 552–556, ACM, 2009.
[17] P. Du, W. Jia, L. Huang, and W. Lu, “Centralized scheduling and channel assignmentin multi-channel single-transceiver wimax mesh network,” in Wireless Communica-tions and Networking Conference, 2007. WCNC 2007. IEEE, pp. 1734–1739, IEEE,2007.
[18] J. Xiao, N. Xiong, and A. Vasilakos, “Centralized scheduling and channel assignmentscheme in third generation router based wimax mesh network,” in Proceedings of the2009 International Conference on Wireless Communications and Mobile Computing:Connecting the World Wirelessly, pp. 721–725, ACM, 2009.
[19] L. Chen, Y. Tseng, D. Wang, and J. Wu, “Exploiting spectral reuse in resource alloca-tion, scheduling, and routing for ieee 802.16 mesh networks,” in Vehicular TechnologyConference, 2007. VTC-2007 Fall. 2007 IEEE 66th, pp. 1608–1612, IEEE, 2007.
[20] Y. Cao, Z. Liu, and Y. Yang, “A centralized scheduling algorithm based on multi-path routing in wimax mesh network,” in Wireless Communications, Networking andMobile Computing, 2006. WiCOM 2006. International Conference on, pp. 1–4, IEEE,2006.
[21] J. Tao, F. Liu, Z. Zeng, and Z. Lin, “Throughput enhancement in wimax mesh net-works using concurrent transmission,” in Wireless Communications, Networking andMobile Computing, 2005. Proceedings. 2005 International Conference on, vol. 2,pp. 871–874, IEEE, 2005.
[22] F. Jin, A. Arora, J. Hwang, and H. Choi, “Routing and packet scheduling in wimaxmesh networks,” in Broadband Communications, Networks and Systems, 2007.BROADNETS 2007. Fourth International Conference on, pp. 574–582, IEEE, 2007.
[23] H. Wei, S. Ganguly, R. Izmailov, and Z. Haas, “Interference-aware ieee 802.16 wimaxmesh networks,” in Vehicular Technology Conference, 2005. VTC 2005-Spring. 2005IEEE 61st, vol. 5, pp. 3102–3106, Ieee, 2005.
72
[24] A. Raniwala, K. Gopalan, and T. Chiueh, “Centralized channel assignment and rout-ing algorithms for multi-channel wireless mesh networks,” ACM SIGMOBILE MobileComputing and Communications Review, vol. 8, no. 2, pp. 50–65, 2004.
[25] L. Fu and Z. Cao, “Joint optimization of routing and scheduling for higher throughputin ieee 802.16 mesh networks,” in Wireless, Mobile and Multimedia Networks, 2006IET International Conference on, pp. 1–4, IET, 2006.
[26] T. Chung, H. Fu, and H. Lee, “Dcls: A fair and efficient scheduling with bandwidthguarantee in wimax mesh networks,” Journal of Information Science and Engineer-ing, vol. 26, no. 4, pp. 1429–1442, 2010.
[27] S. Nazari and H. Beigy, “A new distributed uplink packet scheduling algorithm inwimax newtorks,” in Future Computer and Communication (ICFCC), 2010 2nd In-ternational Conference on, vol. 2, pp. V2–232, IEEE, 2010.
[28] P. Kong, H. Wang, Y. Ge, C. Ang, J. Pathmasuntharam, W. Su, M. Zhou, andH. Harada, “Distributed adaptive time slot allocation for wimax based maritime wire-less mesh networks,” in Wireless Communications and Networking Conference, 2009.WCNC 2009. IEEE, pp. 1–6, IEEE, 2009.
[29] B. Makarevitch, “Distributed scheduling for wimax mesh network,” in Personal, In-door and Mobile Radio Communications, 2006 IEEE 17th International Symposiumon, pp. 1–5, IEEE, 2006.
[30] A. Morell, G. Seco-Granados, and J. Vicario, “Distributed algorithm for uplinkscheduling in wimax networks,” in Broadband Communications, Networks and Sys-tems, 2008. BROADNETS 2008. 5th International Conference on, pp. 257–264, IEEE,2008.
[31] R. Talwalkar and M. Ilyas, “Analysis of quality of service (qos) in wimax networks,”in Networks, 2008. ICON 2008. 16th IEEE International Conference on, pp. 1–8,IEEE, 2008.
[32] H. Wang, W. Li, and D. Agrawal, “Dynamic admission control and qos for 802.16wireless man,” in Wireless Telecommunications Symposium,2005, pp. 60–66, IEEE,2005.
[33] S. Lu, V. Bharghavan, and R. Srikant, “Fair scheduling in wireless packet networks,”IEEE/ACM Transactions on Networking (TON), vol. 7, no. 4, pp. 473–489, 1999.
[34] K. Wongthavarawat and A. Ganz, “Packet scheduling for qos support in ieee 802.16broadband wireless access systems,” International Journal of Communication Sys-tems, vol. 16, no. 1, pp. 81–96, 2003.
73
[35] J. Chen, W. Jiao, and Q. Guo, “Providing integrated qos control for ieee 802.16 broad-band wireless access systems,” in IEEE Vehicular Technology Conference, vol. 62,p. 1254, IEEE; 1999, 2005.
[36] G. Chu, D. Wang, and S. Mei, “A qos architecture for the mac protocol of ieee 802.16bwa system,” in Communications, Circuits and Systems and West Sino Expositions,IEEE 2002 International Conference on, vol. 1, pp. 435–439, IEEE, 2002.
[37] C. Cicconetti, A. Erta, L. Lenzini, and E. Mingozzi, “Performance evaluation of theieee 802.16 mac for qos support,” Mobile Computing, IEEE Transactions on, vol. 6,no. 1, pp. 26–38, 2007.
[38] D. Cho, J. Song, M. Kim, and K. Han, “Performance analysis of the ieee 802.16wireless metropolitan area network,” in Distributed Frameworks for Multimedia Ap-plications, 2005. DFMA’05. First International Conference on, pp. 130–136, IEEE,2005.
[39] C. Cicconetti, L. Lenzini, E. Mingozzi, and C. Eklund, “Quality of service support inieee 802.16 networks,” Network, IEEE, vol. 20, no. 2, pp. 50–55, 2006.
[40] J. Chen, W. Jiao, and H. Wang, “A service flow management strategy for ieee 802.16broadband wireless access systems in tdd mode,” in Communications, 2005. ICC2005. 2005 IEEE International Conference on, vol. 5, pp. 3422–3426, Ieee, 2005.
[41] H. Safa, H. Artail, M. Karam, R. Soudah, and S. Khayat, “New scheduling architec-ture for ieee 802.16 wireless metropolitan area network,” in Computer Systems andApplications, 2007. AICCSA’07. IEEE/ACS International Conference on, pp. 203–210, IEEE, 2007.
[42] P. Ting, C. Yu, N. Chilamkurti, W. Tung-Hsien, and C. Shieh, “A proposed red-basedscheduling scheme for qos in wimax networks,” in Wireless Pervasive Computing,2009. ISWPC 2009. 4th International Symposium on, pp. 1–5, IEEE, 2009.
[43] P. Goyal and A. Sahoo, “A scheduling and call admission control algorithm for wimaxmesh network with strict qos guarantee,” in Communication Systems and Networks(COMSNETS), 2010 Second International Conference on, pp. 1–10, IEEE, 2010.
[44] L. You, M. Song, and J. Song, “Multi-service scheduling mechanism with qos guar-antee for centralized scheduling of wimax mesh mode,” in Global Mobile Congress2009, pp. 1–6, IEEE, 2009.
[45] D. Ghosh, A. Gupta, and P. Mohapatra, “Admission control and interference-awarescheduling in multi-hop wimax networks,” in Mobile Adhoc and Sensor Systems,2007. MASS 2007. IEEE Internatonal Conference on, pp. 1–9, IEEE, 2007.
74
[46] X. Cheng, P. Mohapatra, S. Lee, and S. Banerjee, “Maria: Interference-aware admis-sion control and qos routing in wireless mesh networks,” in Communications, 2008.ICC’08. IEEE International Conference on, pp. 2865–2870, IEEE, 2008.
[47] A. Sayenko, O. Alanen, J. Karhula, and T. Hamalainen, “Ensuring the qos require-ments in 802.16 scheduling,” in Proceedings of the 9th ACM international symposiumon Modeling analysis and simulation of wireless and mobile systems, pp. 108–117,ACM, 2006.
75
Appendix A
Glossary
ATM: Asynchronous Transfer ModeBE: Best EffortBFS: Breadth First SearchBS: Base StationBPSK: Binary Phase Shift KeyingBWA: Broadband Wireless AccessCAC: CAll Admission ControlCUR: Channel Utilization RatioDSL: Digital subscriber LineFDM: Frequency Division MultiplexingFTP: File Transfer ProtocolGPC: Grant Per ConnectionGPSS: Grant Per Subscriber StationHTTP: Hyper Text Transfer ProtocolIEEE: Institute of Electrical and Electronic EngineeringIP: Internet ProtocolITU: International Telecommunication UnionLOS: Line of SightMAC: Medium Access ControlMPEG: Moving Pictures Experts GroupMSH-CSCH: Mesh Centralized SchedulingMSH-DSCH: Mesh Decentralized SchedulingMSH-NCFG: Mesh Network ConfigurationMSH-NENT: Mesh Network EntrynrtPS: Non-real Time Polling ServiceOFDM: Orthogonal Frequency Division MultiplexingOFDMA: Orthogonal Frequency Division Multiple AccessPDU: Protocol Data UnitPHY: Physical (Layer)QAM: Quadrature Amplitude ModulationQPSK: Quadrature Phase Shift KeyingRS: Relay StationrtPS: Real Time Polling ServiceSDU: Service Data UnitSINR: Signal to Interference plus Noise RatioSOFDM: Scalable Orthogonal Frequency Division MultiplexingSS: Subscriber StationTDMA: Time Division Multiple Access
76
UDG: Unit Disk GraphUGS: Uninterrupted Grant ServiceVoIP: Voice Over Internet ProtocolWMN: Wireless Mesh NetworkWMAN: Wireless Metropolitan Area NetworkWiMAX: World Interoperatability for Microwave Access