INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS Int. J. Commun. Syst. (2012) Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/dac.2392

Adaptive scheduling mechanism for IPTV over WiMAX IEEE 802.16j networks Mohamed-el-Amine Brahmia* ,† , Abdelhafid Abouaissa and Pascal Lorenz Laboratories MIPS-GRTC, University of Haute Alsace, 34, Rue de Grillenbreit, 68000 - Colmar, France

SUMMARY A WiMAX technology is a very promising Broadband Wireless Access technology that is able to transmit different service types. This latter can have different constraints such as traffic rate, maximum latency, and tolerated jitter. The IEEE 802.16 Medium Access Control specifies five types of QoS classes: UGS, rtPS, ertPS, nrtPS, and BE. However, the IEEE 802.16 standard does not specify the scheduling algorithm to be used. Operators have the choice among many existing scheduling techniques. Also, they can propose their own scheduling algorithms. In this paper, we propose a scheduling strategy (Adaptive Weighted Round Robin, AWRR) for various Internet Protocol Television (IPTV) services traffic over 802.16j networks. Our scheme adapts dynamically the scheduler operation to according queue load and quality of service constraints. In particular, the proposed mechanism gives more priority to high definition television and standard definition television traffic by using two schedulers. The proposed scheduling algorithm has been simulated using the QualNet network simulator. The experimental results show that our scheduler schemes AWRR have a better performance than the traditional scheduling techniques for rtPS traffic, which allows ensuring QoS requirements for IPTV application. Copyright © 2012 John Wiley & Sons, Ltd. Received 11 March 2012; Revised 1 May 2012; Accepted 29 May 2012 KEY WORDS:

IEEE 802.16j; WiMAX; scheduling; WRR; QoS; IPTV

1. INTRODUCTION Institute of Electrical and Electronics Engineers (IEEE) 802.16 [1] is an emerging wireless technology for deploying broadband Wireless Metropolitan Area Network. It is one of the promising wireless technologies for the next-generation network [2]. 806.16 is initially developed for a pointto-multipoint single hop communication. Multi-hop relaying has recently received a greater attention in the IEEE 802.16j [3] Mobile Multihop Relay (MMR) task group. IEEE 802.16j aims to develop relay mode based on IEEE 802.16e to provide user throughput enhancement, coverage extension, and capacity enhancement through introducing the Relay Stations (RS) [4]. Internet Protocol Television (IPTV) is a technology that enables users to transmit and receive TV program data through IP-based wired and wireless networks [5]. The advances in broadband internet access and coding video technologies have made it possible to IPTV to become one of the rapidly developed sections of the internet protocol TV system [6]. With the recent release of IEEE 802.16d/e/j (WiMAX), Broadband Wireless Access is envisioned to further extend IPTV services into a new application scenario with wireless and mobility dimensions. WiMAX Multihop Relay technology is the adequate technology to provide IPTV services for heterogeneous user requests or their devices, because it supports QoS based multicasting functionality [7]. This progress has enabled providers to offer various IPTV services. However, IPTV application is expected to

*Correspondence to: Mohamed-el-Amine Brahmia, Laboratories MIPS-GRTC, University of Haute Alsace, 34, Rue de Grillenbreit, 68000 - Colmar, France. † E-mail: [email protected] Copyright © 2012 John Wiley & Sons, Ltd.

M.-E.-A. BRAHMIA, A. ABOUAISSA AND P. LORENZ

provide high definition television (HD-TV), standard definition television (SD-TV), Web television (Web-TV), and mobile television (Mobile-TV). The IEEE 802.16 standard supports many traffic types, that is, data, video, and voice [8]. Also, it has been developed to support stringent QoS requirements of various applications but not specified in terms of how the diverse QoS requirement can be achieved. Many optimization criteria can be considered for scheduling algorithms such as the total maximum data rate, fairness, and operator revenue optimization [9]. The IEEE 802.16 standard does not define any slot allocation criterion or scheduling algorithm for any type of service [3]. A scheduling module is necessary to provide QoS for each Mobile Station (MS). In this paper, we propose a scheduling algorithm for IPTV over IEEE 802.16j networks. Our algorithm is based on Weighted Round Robin (WRR) scheduling architecture. We have focused on IPTV application that belongs to real-time Polling Service (rtPS) class. In order to support QoS constraints and users heterogeneous devices, we divide this class video stream into four subclasses. Each subclass represents a video stream service (HD-TV, SD-TV, Web-TV, and Mobile-TV). The aim of the proposed algorithm is to support QoS constraints (in particular, HD-TV and SD-TV real-time traffic) by using two schedulers, and assigning a dynamic weighted value to each service queue. The remainder of this text is organized as follows. 802.16 MAC basic theories are described in Section 2, whereas Section 3 provides a survey on the related work. Section 4 highlights problem motivation and gives details about the proposed scheduling scheme. In Section 5, we present simulations and results showing the proposed scheme. Finally, conclusions and future works are stated in Section 6.

2. BASIC THEORY The 802.16 MAC is connection-oriented. The Base Station (BS) assigns a unique Connection IDs in order for each Uplink (UL) and Downlink (DL) transmission [10]. At the Convergence Sub-layer (CS), the classifier maps traffic flows to the connections with distinct scheduling services. Hence, it is anticipated that multiple schedulers will be implemented to address heterogeneous demands of subscribers. The IEEE 802.16e-2005 is the standard for IEEE 802.16j mobile relay networks. It is fully compatible with IEEE 802.16e MSs but the IEEE 802.16e BS needs to be modified in order to support relay operations. The systems consist of one Multi-hop Relay Base Station (MR-BS), several Relay Station (RSs), and several MSs. Two different modes of resource allocation and scheduling are specified in the standard [3] for MSs. Those modes are: the centralized scheduling mode and the distributed scheduling mode [11]. In the former mode, the bandwidth/resource allocation for all nodes (including RSs and MSs) is determined at the MR-BS. In the latter mode, a part of the bandwidth/resource allocation for RSs and MSs is determined at RSs, whereas another part is determined at the MR-BS. Two different relay modes are defined in the IEEE 802.16j standard, that is, the transparent mode and the non-transparent mode. Table I shows the differences between these two modes. Thew success of WiMAX lies in its comprehensive supports for a variety of dominant broadband services in a suite of QoS scheduling types [12]. The IEEE 802.16j WiMAX standard offers five categories for the traffic prioritization: Table I. Comparisons between the transparent and non-transparent relay mode [8].

Scheduling Number of hops Performance Coverage extension Cost / Complexity Forward framing info Copyright © 2012 John Wiley & Sons, Ltd.

Transparent mode

Non-transparent mode

Centralized 2 High No Low No

Centralized / Distributed 2 or more Low Yes High Yes Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

ADAPTIVE SCHEDULING MECHANISM FOR IPTV

 Unsolicited grant service (UGS): such as T1/E1 transport. It requires reserved traffic rate,

maximum latency, and tolerated jitter.  Extended real-time polling service (ertPS): such as voice of IP (VoIP). It is built on the effi-

ciency of both UGS and rtPS, reduces overhead and access delay of rtPS, and improves uplink resource utilization of the UGS. As a subcategory of rtPS, ertPS has the same QoS parameters as rtPS in the scheduler design.  Real-time polling service (rtPS): such as MPEG audio/video streaming and video conferencing. It supports variable bit rate traffic via minimum reserved and maximum sustained traffic rates and requires tolerable stringent latency constraints.  Non-real-time polling service (nrtPS): delay tolerant streams with variable-sized packets, for which only minimum reserved and maximum sustained traffic rates are required, such as File Transfer Protocol (FTP).  Best effort (BE) service: such as HTTP and email. BE services are handled on a space available basis and do not require tight latency/jitter constraints, with upper limited bandwidth consumption via maximum sustained traffic rate. 3. RELATED WORK Scheduling algorithms are the most important aspect in WiMAX for radio resources efficient use. A scheduling algorithm should take into account the WiMAX QoS classes and service requirements. We can distinguish between two types of scheduling algorithms. The first type includes well-known schedulers, and the second one represents the schedulers specifically proposed for WiMAX networks [13]. In this subsection, we discuss the design of two known scheduling algorithms related to our work (WRR and Strict Priority).  The WRR scheduling is designed to better handle servers with different processing capaci-

ties. It assigns static weight to each queue, and the bandwidth is then allocated according to these static weights [14]. Moreover, higher priority queues get more connections than the lower weight queues [15]. WRR ensures that all service classes have access to network bandwidth to avoid queue starvation problem. Although, the algorithm will not provide good performance in the presence of variable size packets.  Strict Priority (SP) is a simple scheduling algorithm that serves all the highest priority queue until it is empty and then moves to the next highest priority queue. In the lower classes of service, this mechanism could cause bandwidth starvation [16]. Kwak et al. [17] proposed a modified dynamic weighted round robin (MDWRR) cell scheduling algorithm, the aim of this algorithm is to guarantee the delay property of real-time traffic and also efficiently transmits non-real-time traffic. The proposed scheduling idea is a variation of the dynamic weighted round robin (DWRR) algorithm by using a threshold to prevent the cell loss of non-real-time traffic that is due to the cell transmission procedure based on delay priority. Although the MDWRR scheduling algorithm is more complex than the conventional DWRR and WRR techniques, taking into account delay priority minimizes cell delay and decreases the required size of the temporary buffer. Mardini et al. [15] proposed a Modified Weighted Round Robin (MWRR) scheduler in order to decrease the average end-to-end delay and improve the average throughput. The proposed scheduling technique has been compared with well-known scheduling techniques (WRR, SP, and Weighted Fair Queuing). The MWRR tends to be a reliable scheduling algorithm in avoiding the problems in WRR that causes a starvation and unnecessary delay for lower services class. The goal was to reduce the average delay and increase the average throughput, especially to the lower classes by increasing the size of overall packets that WRR should serve. Rashwan et al. [16] focused mainly on studying some scheduling algorithms such as Weighted Fair Queuing (WFQ), Round Robin, WRR and SP, through analyzing and evaluating the performance of each scheduler to support the different QoS classes. Authors have compared the behavior of the SP, Round Robin, WRR, and WFQ scheduling algorithms in WiMAX networks. Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

M.-E.-A. BRAHMIA, A. ABOUAISSA AND P. LORENZ

Simulation results have verified that the SP scheduler had the highest throughput and minimum delay for high QoS classes. However, it caused bandwidth starvation for the BE and the nrtPS classes. The average end-to-end delay in the SP had large value for the rtPS traffic. The RR scheduler had better performance for low QoS classes on the expense of the high QoS classes. Both WFQ and WRR can control the performance of each class by assigning different weight to each queue. Hou et al. [18] proposed a cooperative multicast scheduling scheme for IPTV service over IEEE 802.16 networks. Authors proposed a multicast scheduling algorithm, using downlink cooperative transmission to achieve high throughput not only for all multicast groups but also for each group member. 4. PROPOSED SCHEDULING ALGORITHM In this section, we detail our proposed scheduling scheme, which is a new variant of WRR scheduling technique called Adaptive Weighted Round Robin (AWRR). We propose a centralized scheduling algorithm for IEEE 802.16(j) relay WiMAX network operating in the transparent mode. The WRR scheduler is designed to serve various service classes, that is, real-time, interactive, and file transfer. It is assigned to a queue that is specifically dedicated to that service class. Each queue is serviced in a round robin order depending on the assigned weight. A queue with the highest weight takes the priority to get first a connection [15, 19]. Moreover, we have chosen WRR scheduling technique to avoid low classes starvation problem. In addition, WRR is easy to implement in Qualnet simulator. Our idea is to use two schedulers, that is, input scheduler and output scheduler. The former has two buffers. The first one is High Priority (HP) buffer that consists of UGS, ertPS, and rtPS flows. The rtPS flow is only concerned when HD-TV or SD-TV video stream in IPTV application. The second is Low Priority (LP) buffer, which contains all the other flows, that is, rtPS_web-TV, rtPS_mobile-TV, nrtPS, and BE. The objective of the first scheduler (input scheduler) is to give priority to video streams with superior quality (HD and SD), which has more constraints in terms of bandwidth and latency. The goals of the second scheduler (output scheduler) are to control data flows and manage the service classes, that is, UGS, ertPS, rtPS, nrtPS, and BE. Figure 1 depicted the architecture of proposed scheduler. For both input and output schedulers, we apply AWRR algorithm to adjust service classes weights. We also set a threshold for each class of service that will be used to trigger dynamically weights adjustment if a threshold is exceeded. However, to respect QoS constraints and to give priority to real-time traffic, we assume that HP queues thresholds are lower than LP queues thresholds. 4.1. Input scheduler design The main feature of the adaptive WRR algorithm is to regulate the weights of high and LP queues according to HP traffic load and buffer size. Weights regulation of each queue makes it capable to guarantee the minimum throughput of best-effort traffic under severe congestion. It also gives priority to HP traffic treatment. Figure 2 shows the design architecture of output scheduler.

Figure 1. Proposed scheduler architecture. Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

ADAPTIVE SCHEDULING MECHANISM FOR IPTV

Figure 2. Design architecture of input scheduler.

4.1.1. Adjusting the weights of each class queue (input scheduler). To adjust the weight of both classes with considering traffic loads, we have used two thresholds and lengths of queues. Each weight of the class queues is defined as follows: WH D WHi C

NPH  TH NPH

WL D 1  WH ,

(1)

(2)

where WH is the weight of the HP class and WL is the weight of the LP class. TH is the threshold of the HP class queue and NPH (number of packets) is the buffer occupancy of the HP class queue. The buffer threshold (TH ) is used to start weights adjustment when buffer occupancy exceeds the fixed threshold. WHi is the initial weight of the HP queue that allows the AWRR scheduler to calculate WH weight. 4.1.2. Dynamic weights calculation algorithm. We calculate the weight dynamically with algorithm in the succeeding text.

Variables definition: TH : TL : NPH : NPL :

threshold of HP buffer. threshold of LP buffer. buffer occupancy of the HP class queue. buffer occupancy of the LP class queue.

4.2. Output scheduler design The output scheduler allows the management of all flows and use adaptive WRR algorithm to regulate the weights of all classes. Weights calculation must respect the constraints of each service class and avoid starvation problem for low class priority. Figure 3 shows the design architecture of output scheduler. Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

M.-E.-A. BRAHMIA, A. ABOUAISSA AND P. LORENZ

Figure 3. Design architecture of output scheduler.

4.2.1. Adjusting the weights of each class queue (output scheduler). Weights adjustment is an important function that allows a better queues management. The constraint of each class weight is determined as follows: Wi > 0 m X

(3)

Wi D 1,

(4)

i D1

where m is the number of buffers and Wi is the weight of the i th buffer. However, we suppose that Wi D fi .k/ D ai k C bi ai D

(5)

NPi , Ti

(6)

where NPi is the buffer occupancy in the i th buffer, and Ti is the threshold for this buffer. bi is a predefined constant whose objective is to favorite high classes priority. .4/ H)

m X

fi .k/ D 1 H)

i D1

m X .ai k C bi / D 1.

(7)

i D1

For our solution, we will use five buffers (one buffer per service class). From Equation (7), we can solve the system of equation by calculating the variable k. .7/ H) w1 C w2 C w3 C w4 C w5 D 1

(8)

.8/ H) .a1 k C b1 / C .a2 k C b2 / C .a3 k C b3 / C .a4 k C b4 / C .a5 k C b5 / D 1

(9)

.9/ H) k D

1  .b1 C b2 C b3 C b4 C b5 / . a1 C a2 C a3 C a4 C a5

(10)

After calculating the variable k, we can calculate the weight of each buffer (class of service). Otherwise, the values of bi are arbitrarily prefixed to favorite HP traffic classes. In our case, we have five service classes queues, so we assume that: (b1 D ˛1 for (USG), b2 D ˛2 for (ErtPS), b3 D ˛3 for (rtPS), b4 D ˛4 for (nrtPS), and b5 D ˛5 for (BE)). The choice of bi value depends on the application to implement with the follows conditions:

Copyright © 2012 John Wiley & Sons, Ltd.

0 6 bi < 1

(11)

bi > bi C1

(12) Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

ADAPTIVE SCHEDULING MECHANISM FOR IPTV

4.2.2. Algorithm for dynamic weights calculation. We design the algorithm to calculate the weight dynamically. The weights are always positive to guarantee a minimum throughput for LP traffic. We also ensure the privilege of the HP classes by using the constant bi , which is important for UGS and rtPS classes. The algorithm is shown in the succeeding text.

The algorithm of Weights Adjustment has been explained in section 4.2.1. 5. PERFORMANCE EVALUATION This section discuss the details of the simulation environment and the results of simulation. We evaluate the performance of the AWRR scheduling algorithm in terms of end-to-end delay and throughput. The simulations are performed using QualNet version 5.0.2 paid version. Table II recapitulates the simulation configurations. Using the IP protocol’s precedence field, applications can direct traffic to a specific service class specified in WiMAX. In our case, the mapping between precedence values and service classes is the flows: (UGS=7, ertPS=4, rtPS=3, nrtPS=2, and BE=0). We evaluate and compare the performance of our proposed scheduling scheme performance (AWRR) with SP and WRR scheduling algorithms. The three different scheduling algorithms performed a comparison in five of QoS classes and with different traffic loads. We also evaluate the three schedulers algorithm behavior by varying the number of IPTV channels. The average throughput for QoS classes is shown in Figure 4. The UGS and ertPS traffic have the largest throughput value with SP scheduler. However, rtPS traffic has the best throughput with AWRR scheduler because we use two schedulers. In fact, input scheduler enhances rtPS (IPTV video stream) performance by prioritizing HD and SD IPTV services. Figure 4 shows that WRR schedulers have the best performance for low classes in terms of throughput. It also shows that

Table II. Parameter simulations. Parameter Antenna model Channel frequency Relay type Scheduling mode Number of IPTV channels Precedence values Radio type Scheduling algorithms evaluated

Value Omni directional 2.4 GHz Transparent mode Centralized 50 0, 2, 3, 4, and 7 802. 16 Radio SP, WRR, AWRR

SP, Strict Priority; WRR, Weighted Round Robin; AWRR, Adaptive Weighted Round Robin. Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

M.-E.-A. BRAHMIA, A. ABOUAISSA AND P. LORENZ

Average Throughput (Mbps)

2.5

2

1.5 SP WRR

1

AWRR

0.5

0 UGS

ertps

rtps

nrtps

BE

QoS classes

Figure 4. Average throughput versus classes of QoS.

Average End-to-End Delay (sec)

1.2 1 0.8 0.6

SP WRR

0.4

AWRR

0.2 0 UGS

ertps

rtps

nrtps

BE

QoS classes

Figure 5. Average end-to-end delay versus classes of QoS.

AWRR scheduler average throughput has the best performance for all QoS classes because AWRR adjust dynamically queue weights according to class priority and buffer occupancy. Figure 5 represents the average end-to-end delay of the three scheduling algorithms for all QoS classes. It shows that our AWRR scheduler performs the best performance in ertPs, rtPS, and nrtPS classes. For UGS class, SP has the best end-to-end delay because all higher priority queues processes are always completely served before the lower priority queues. Unlike WRR, which uses fixed weights for each service classes, AWRR gives good results because it adjusts dynamically queue weight of each QoS classes and gives priority to real-time traffic by using the input scheduler. Figures 6 and 7 represent average end-to-end delay and average throughput for rtPS traffic, respectively. However, as we have said earlier, rtPS class contains IPTV video stream content. To evaluate rtPS class traffic performance, we change the number of IPTV channels requested by users. The latter can request any IPTV channels with one of the different existed qualities (HD-TV, SD-TV, Web-TV and Mobile-TV) in a random manner. Figures 6 and 7 show that AWRR scheduler has the highest throughput and minimum delay regardless of number of requested IPTV channels. This due to our scheduler (AWRR) mechanism use two schedulers to reduce end-to-end delay for HD-TV and SD-TV IPTV video stream, and we adapt rtPS queue weight automatically when threshold is exceeded. In Figure 8, we compare the number of IPTV channel that could be provided in a limited time for the various IPTV services (60 minutes, in our case). However, to better study the impact of Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

ADAPTIVE SCHEDULING MECHANISM FOR IPTV

Figure 6. Average end-to-end delay of rtPS class and different number of IPTV channels.

Figure 7. Average throughput of rtPS class and different number of IPTV channels.

The number of IPTV channels

10 9 8 7 6 SP

5

WRR

4

AWRR

3 2 1 0 HD-TV

SD-TV

Web-TV

Mobile-TV

IPTV Service

Figure 8. Average number of IPTV channels for different services.

our scheduling mechanism, we send more IPTV video streams (channels) with HD and SD quality. Our simulation shows that the number of HD-TV and SD-TV channel that is provided by our proposed scheduling mechanism (AWRR) is greater than SP and WRR schedulers. This means that the Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

M.-E.-A. BRAHMIA, A. ABOUAISSA AND P. LORENZ

efficiency of AWRR is better in terms of packets treatment time. This is because of our scheduler algorithm design tends to serve more packets from rtPS class, especially HD and SD IPTV services. Because when we classify HD and SD packets in HP buffer in input scheduler, we ensure a fast treatment at the output scheduler level. 6. CONCLUSION In this work, we have presented the design and simulation of AWRR scheduling algorithm for IEEE 802.16j networks. We have also provided performance measurements using the QualNet 5.0.2 simulator. The proposed scheduler scheme allows giving priority to the service classes, which have more constraint without having the problem of starvation and unnecessary delay for lower service classes. The simulation experiments show that our proposed scheduler, AWRR significantly outperforms the SP and WRR algorithms in terms of reducing end-to-end delay and increasing throughput for rtPS traffic. Overall, our proposed scheduler has a good performance for IPTV application, with adjusting dynamically the weight of each queue and giving priority to HD-TV and SD-TV video stream through the use of the input scheduler. In future work, we will plan to adapt the proposed scheduler with distributed scheduling mode by using non-transparent relay stations. We will also propose a connection admission control mechanism to efficiently manage the resources among the existed and new flows. ACKNOWLEDGEMENT

This research project is funded by France Telecom-Orange R&D and MIPS-GRTC laboratory at University of Haute Alsace.

REFERENCES 1. IEEE-Std. IEEE 802.16-2004: air interface for fixed broadband wireless access systems. IEEE Standard for Local and Metropolitan Area Networks 2004. 2. Kim HJ. IEEE 802.16/WiMAX security. Technical Report, Dept. of Electrical and Computer Engineering, Stevens Institute of Technology, Hoboken, New Jersey. Available from: http://www.ibluemojo.com/contents/IEEE%20802. 16%20Security.pdf. 3. IEEE-Std. 802.16j-2009 part 16: Air interface for fixed and mobile broadband wireless access systems, multihop relay specification. IEEE Standard for Local and Metropolitan Area Networks 2009. 4. Genc V, Murphy S, Murphy J. Performance analysis of transparent relays in 802.16j mmr networks. In Proceedings of International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks, Berlin, Germany, 2008; 273–281. 5. Lai YX, Lai CF, Hu CC, Chao HC, Huang YM. A personalized mobile iptv system with seamless video reconstruction algorithm in cloud networks. International Journal of Communication Systems October 2011; 24(10):1375–1387. DOI: 10.1002/dac.1254. http://dx.doi.org/10.1002/dac.1254. 6. She J, Hou F, Ho P, Xie L. IPTV over WiMAX: key success factors, challenges, and solutions. Advances in Mobile Multimedia, IEEE Communications Magazine 2007; 45:87–93. 7. Brahmia M, Abouaissa A, Lorenz P. Optimal bandwidth consumption for IPTV services over WiMAX multihop relay networks. In Proceedings of the Eighth Advanced International Conference on Telecommunications, Stuttgart, Germany, May 2012; 40–45. 8. de Moraes LFM, Maciel PDJ. Analysis and evaluation of a new mac protocol for broadband wireless access. In IEEE Wirelesscom 2005 - International Conference on Wireless Networks, Communications, and Mobile Computing, Maui, Hawaii, USA, 2005; 107–112. 9. Li Y, Yang Y, Zhou L, Wei A, Cao C. Qos-aware fair packet scheduling in IEEE 802.16 wireless mesh networks. International Journal of Communication Systems June 2010; 23:901–917. DOI: 10.1002/dac.v23:6/7. Available from: http://dx.doi.org/10.1002/dac.v23:6/7. 10. Retnasothie FE, Ozdemir MK, Yucek T, Celebi H, Zhang J, Muththaiah R. Wireless IPTV over WiMAX: challenges and applications. In Proceedings of the IEEE Annual Wireless and Microwave Technology Conference, WAMICON ’06, Clearwater Beach, FL, vol. 45, December 2006; 1–5. 11. Wang SY, Chen HY, Chuang SW. Nctuns tool for IEEE 802.16 j mobile WiMAX relay network simulations. In Computer Science Research and Technology. Nova Science Publishers, 2009; 1–31. 12. Du H, Liu J, Liang J. Downlink scheduling for multimedia multicast/broadcast over mobile WiMAX: connectionoriented multistate adaptation. In Wireless Commun, Vol. 16. IEEE Press: Piscataway, NJ, USA, August 2009; 72–79. Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac

ADAPTIVE SCHEDULING MECHANISM FOR IPTV

13. Belghith A, Nuaymi L. Comparison of wimax scheduling algorithms and proposals for the rtps qos class. In Proceedings of the 14th European Wireless Conference 2008. IEEE: Prague, Czech Republic, 2008; 1–6, DOI: 10.1109/EW.2008.4623857. 14. Khan AS, Sattar AR, Mustafa T, Ahmad S. Performance evaluation and enhancement of uplink scheduling algorithms in point to multipoint WiMAX networks. European Journal of Scientific Research (EuroJournals) 2010; 42:491–506. 15. Mardini W, Alfool MA. Modified WRR scheduling algorithm for WiMAX networks. Journal of Network Protocols and Algorithms 2011; 3:24–43. USA. 16. Rashwan AH, MElBadawy H, Ali HH. Comparative assessments for different WiMAX scheduling algorithms. In Proceedings of the World Congress on Engineering and Computer Science 2009, San Francisco, USA, October 2009; 362–366. 17. Kwak JY, Nam JS, Kim DH. A modified dynamic weighted round robin cell scheduling algorithm. ETRI Journal October 2002; 24:360–372. DOI: http://dx.doi.org/10.4218/etrij.02.0102.0504. 18. Hou F, Cai LX, Ho PH, Shen X, Zhang J. A cooperative multicast scheduling scheme for multimedia services in IEEE 802.16 networks. In Trans. Wireless. Comm., Vol. 8. Institute of Electrical and Electronics Engineers Inc., The: Beijing, China, 2009; 1508–1519, DOI: 10.1109/TWC.2009.080417. 19. Ali NA, Dhrona P, Hassanein H. A performance study of uplink scheduling algorithms in point-to-multipoint WiMAX networks. In Comput. Commun., Vol. 32. Butterworth-Heinemann: Newton, MA, USA, 2009; 511–521, DOI: 10.1016/j.comcom.2008.09.015.

AUTHORS’ BIOGRAPHIES

Mohamed-el-Amine Brahmia is a PhD student at the University of Haute Alsace in France. He received his master’s degree in computer science from the University of Versailles (France) in 2008 and an engineer’s degree in computer science from the University of 08 Mai 1945-Guelma (Algeria) in 2006. His research focuses on multicast, routing, path selection, admission control, and scheduling design in Multihop Relay WiMAX Networks. He is also a part-time teacher in Network & Telecom department at IUT Colmar.

Abdelhafid Abouaissa is an associate professor at the University of Haute-Alsace, in Colmar, France. He received a bachelor’s degree from the Technical University of Wroclaw, Poland, in 1995, and a master’s degree from Franche-Comté University of Besançon, France, in 1996. He obtained his PhD at the Technical University of Belfort, France, in January 2000. His interests include multimedia synchronization, group communication systems, QoS routing in ad-hoc, Mesh networks, sensor networks, MPLS, DiffServ, and QoS management.

Pascal Lorenz received his master’s degree (1990) and PhD (1994) from the University of Nancy, France. Between 1990 and 1995, he was a research engineer at WorldFIP Europe and at Alcatel-Alsthom. He is a professor at the University of Haute-Alsace, France, since 1995. His research interests include QoS, wireless networks, and high-speed networks. He is the author/co-author of 3 books, 3 patents, and 200 international publications in refereed journals and conferences. He was the technical editor of the IEEE Communications Magazine Editorial Board (2000-2006), chair of Vertical Issues in Communication Systems Technical Committee Cluster (2008-2009), chair of the Communications Systems Integration and Modeling Technical Committee (2003-2009), and chair of the Communications Software Technical Committee (2008-2010). He has served as co-program chair of IEEE WCNC 2012 and ICC 2004, tutorial chair of VTC 2013 Spring and WCNC 2010, track chair of PIMRC 2012, symposium co-chair at Globecom 2011-2007 and ICC 2010-2008. He has served as co-guest editor for special issues of IEEE Communications Magazine, Networks Magazine, Wireless Communications Magazine, Telecommunications Systems, and LNCS. He is senior member of the IEEE and member of many international program committees. He has organized many conferences, chaired several technical sessions, and gave tutorials at major international conferences. Copyright © 2012 John Wiley & Sons, Ltd.

Int. J. Commun. Syst. (2012) DOI: 10.1002/dac