A new Peer to Peer Caching Techniques for Transmission The Video over Wireless Ad Hoc Network

Shared by: ijcsis

Categories

Technology > Emerging Technologies

Stats

views:: 124
posted:: 7/11/2010
language:: English
pages:: 6

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 3, 2010

A new Peer to Peer Caching Techniques for
Transmistting the Video over Wireless Ad Hoc
Network
Saleh Ali K. Al-Omari, Putra Sumari, Mohammad Azmi Q. Al-Madi
School of Computer Science
Universiti Sains Malaysia
Penang, Malaysia
salehalomari2005@yahoo.com, putras@cs.usm.my, almadi_moh@yahoo.com

The P2P caching can reduce the loading of cluster head and
Abstract—In our life today’s the wireless networks becomes so shorten the cache distance and increase hit ratio. Video is
significant due to the advancing in science and technology. It is
encoded into three layers and broadcasts all layers to the
also playing an important role in many applications in our daily
lives especially in widespread Ad Hoc network. Ad hoc networks
cluster head via the internet/backbone [20]. Through radius-
are the wireless mobile networks to create a communication based binary search algorithm, it can evenly distribute clients
between the nodes or devices without using any existing fixed to caching which layers of the video. By using a Peer to Peer
network infrastructure. Currently multimedia services are caching system through Radius Based Binary Search
applied to ad hoc networks where the problem of the mobility Algorithm (RBBSA) and mobile management, the mobile
constantly makes an interruption of the video packets nodes can request nearby nodes for lost video clips through
transmission. This paper proposes a new peer to peer caching P2P model. Mobile management able to predicate the time
techniques using radius based binary search algorithm (RBBSA) when mobile nodes leave cluster by predication model, this
and mobile management in order to solve the problem at the
can inform members to help cache the videos from the cluster
same time proposed the cluster head to provide Video services to
other nodes or devices. In turn to transmit video packages head. In order to solve the problems which can cause an
through ad hoc network, layered scalable video coding are interruption of the video packets transmission in mobility, a
adopted where the video is encoded into a three layers by layered new Peer to Peer caching system are proposed in this paper.
scalable video coding. The result for this new technique that the Furthermore, the node acting as a client, server or cluster head
hit ratio of the peer to peer can increase the catching and reduce provider are able to transmit the video to another nodes and
the cache distance, where the cluster head for the node this leads to extending the range of the services as well as to
mechanism can provide the Video to another nodes which the construct lesser load on the main cluster head or server. An
node acting as server. It also extends the range of the services as important network (Ad Hoc Network) to transmit the video
well to make lesser load on the main cluster head or server.
packet from node to another within the transmission range will
be discussed in detail in section 2.

Keywords-wireless ad hoc networks, RBBSA, ad hoc network, The subsequent sections are organized as follows. Section 2
peer to peer (P2P), random caching, mobile management. described the architecture for Peer to Peer caching system,
while in Section 3 introduces the performance evaluation with
the result of our simulation for our new peer to peer caching
I. INTRODUCTION system. Finally, the conclusion is presented in Section 4.
In recent years, the popularity of the Peer to Peer (P2P)
system for media streaming become so significant which it II. THE ARCHITECTURE FOR PEER TO PEER CACHING
been studied more extensively in recent years by a lot of SYSTEM
number of authors. The P2P system have offered many
advantages where it extensively developed in excess of the The proposed architecture consist three components as
Internet such as [1][2][3][4][17][18][19]. The basic building shown in Figure 1, Central Video Services Provider (CVSP),
block for the Peer to peer streaming system is the nodes. The cluster head and clients.
nodes acting as a server or client and at the same time form an
overlay and simultaneously send and receive the media
streaming data between each other’s.

240 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 3, 2010
.have to be done between all the nodes. This case Called Peer
level Multi Hopping and that is the main building block for Ad
Hoc Network.
In the past few years, the people became realized to use all
the technology so widely and the people’s future living
environments are emerging, based on information resource
provided by the connections of different communication
networks for clients also we have seen a rapid expansion in the
field of Mobile Computing because the proliferation not
expensive, widely available wireless devices .A new small
devices such as personal communication like laptops, PDAs,
cell phone, handhelds, and also there are a lot of traditional
home appliances such as a digital cameras, cooking ovens,
refrigerators, with computing and communicating powers
attached. Expand this area to became a fully pervasive and so
widely. With all of this, the technologies must be formed the
good and new standard of pervasive computing, that including
the new standards, tools, services, devices, protocols and a new
Figure 1. Our architecture for peer to peer caching system architectures. In this days the internet users in ad hoc network
through increase in the use of its advantage is that not involve
Central Video Services Provider (CVSP) is a Server or cluster any connection link and the wiring needed to save space, and
of servers as a main entity for providing VOD services to the building low cost, and improve the use, and can be used in
end clients and it has storage of video services. It provides mobile phone, because of these advantage local wireless
video service to the end client through cluster head and network architecture readily. And also beads in these
assumed that the network is cluster-based, it also proposed to advantages the wireless network can be used in the local area
use scatter of cluster head for covering wider area. Every network terminal part of the wireless [6].
cluster head receives the packet from the server and forwards it B. Wireless Mobile Network Systems
to the client. The cluster head is the main component of the
peer to peer caching system (P2PCS) which it took over the
The past decade the Mobile Network is the only one much
roles of the managing node mobility and peer to peer caching important computational techniques to support computing and
system. The Clients are the mobile hand-held devices such as widespread, also advances in both software techniques and the
PDA and Cell Phone. Every cluster head get the service (video hardware techniques have resulted in mobile hosts and wireless
packets) through the backbone directly from the CVSP and networking common and miscellaneous. Now we will discuss
then forwards it too the clients through the internet backbone. about to distinct approaches very important to enabling Mobile
The server allows the video contents to be encoded into wireless Network or IEEE 802.11 to make a communication
multiple layers by layered scalable video coding, one of the between each other [7, 8, and 9]. Firstly, infrastructure wireless
compressed layered is the base layer also other compressed networks also called Cellular where the communication
layers are the enhancements layer 1 and enhancements layer 2, between nodes must through the Base Station. Basically the
or as known as layer 1 (LR1) and layer 2 (LR2). Cellular supported by a wired fixed infrastructure and the
communication only occurs within the single hop. Secondly,
A. Wireless Ad Hoc Networks infrastructure-less wireless networks is knows as mobile
Mobile Ad Hoc Network (MANETs) is a collection of two network commonly known as Ad Hoc Networks, unlike the
or more devices or nodes or terminals with wireless cellular nodes the ad hoc network dynamically from a
communications and networking capability that communicate temporary network without nay support of any network
with each other without the aid of any centralized administrator infrastructure based wireless LANs [10].
also the wireless nodes that can dynamically form a network to
exchange information without using any existing fixed network The communication between the nodes is the network is in a
infrastructure. and all the nodes in a wireless ad hoc network P2P style. The source node will forward a packet of data
act as a router and host as well as the network topology is in
directly to the destination node if both nodes in within the
dynamically, because the connectivity between the nodes may
transmission range. Otherwise, the intermediate node is use
vary with time due to some of the node departures and new
node arrivals. The special features of MANETs bring this forward the message towards the destination node if the source
technology great opportunity together with severe challenges node is unable to send the message directly as well as each
[5]. All the nodes are responsible to organize themselves node work as end hosts and as router because of the mobility of
dynamically the communication between the each other and to the node and the unfixed network topology. In figure 2
provide the necessary network functionality in the absence of Illustrations the difference between two types of wireless
fixed infrastructure or we can call it ventral administration, It networks, for the cellular and Ad Hoc networks.
implies that maintenance, routing also management, and so on

241 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 3, 2010
E. Peer to Peer Caching Management.
As we discussed, the Central Video Services Provider
(CVSP) allows video content to encode into several layers
(base, enhancements layers) and broadcasts all layers videos to
all clusters heads through the backbone, where all clusters
heads broadcast all layers videos to their through separate
(a) (b)
channels [14][15]. Clients will face interruption during
Figure 2. Illustrations different types of wireless network (a) infrastructure receiving video packet when their move far away or quit the
wireless networks also called Cellular (b) Ad Hoc Networks cluster. A new Peer to Peer caching technique called radius
based binary search algorithm are proposed in order to solve
C. Layered Scalable Video Coding above problem. A cluster head assigns the members of cluster
These sections briefly summarize the concept of scalable to cache one of the bit-streams, including base layer,
video coding. Generally; each video is encoded into one enhancements layers LR1, LR2.
compressed bit-stream, which is referred to as non-scalable
video coding. The layered scalable coding in contras is to The cluster head will response to the request with some
encode video into multiple layers including a base layer, one or information’s such as cached node ID and position. When a
more enhancement layers. The base layer can be independently mobile node joins a cluster, it can request cluster head for lost
decoded and provides the version of least equality, while its video clips. Through responding message, the mobile node can
combination enhancement layers can only be decoded together request for the cached video clips from the nearby clients
with the base layer to provide a better quality. In detail, the where the nearby clients can send cached video clips to mobile
layered scalable video coding can be classified into three types, nodes using unicast protocol. Clients are used to cache the
(1) SNR Scalability, (2) Spatial Scalability and (3) Temporal video clips for reducing the cluster head loading, but the cluster
Scalability owing to spatial scalability which been used in our head will be overload when many clients move and request a
video encoder. The following discussion in the subsection video clips at the same time. Besides decreasing cache
below focuses on Spatial Scalability [11]. distance, caching video clips by nearby clients also can reduce
the server during loading, as well as the nodes or device can be
D. Spatial Scalability cluster head entity and have enough resource to provide the
Spatial scalability is using the same characteristic of video video services to other nodes or devices. In figure 1
but different in spatial resolution or size. For example: A illustrations the procedure whereby PDA1 also working as a
spatial scalability video encoder compress a video which has cluster head and provide the services to PDA2 and cell phone.
solution 352x288(CIF) into two layers at the same frame rate, Due to the characteristic of wireless Ad Hoc Network the
but different in spatial resolutions. The base layer is encoded at nodes organize themselves dynamically the communication
a lower spatial resolution, and therefore to reconstructed base- between each other and provide the necessary network
layer picture resolution which has only 176 x 144 (QCIF). In functionality in absence of fixed infrastructure.
other words, it is the highest qualities (352 x 288) that combine F. Probability Analysis
with all the layers [12]. The characteristics of spatial scalability
encoding in Table 1 described the bitrates of single layer. It is For the probability in the cluster, we assumed that the
compared between both base and enhancement layers of spatial arriving nodes get base layer with probability Pprob, get LR1
scalability encoding which been used the CIF format for layer with probability Rprob, and get LR2 layer with probability
enhancement layer and the QCIF format for base layer [13]. (1- Pprob - Rprob). It is apparent that to getting the base layer,
LR1 and LR2 layer are independent events. Consequently at
TABLE I. COMPARISON OF AVERAGE BITRATE OF SINGLE LAYER TO the same time we assume in the equation 1 the probability of
THOSE OF TWO LAYERS OF SPATIAL SCALABLE ENCODING AT DIFFERENT getting base layer, LR1 and LR2 is S. also the probability of S
QUANTIZATION VALUES. is Pprob * Rprob * ( 1 – Pprob - Rprob ). in additional we assuming
Bit rate Bit rate Ratio of Base that the number of members in the cluster head is “C”, the
Quantization Index (Mbps) Ratio(%) Layer Bitrate to number of nodes that cache base layer is Nnodes, the number of
Total
Enhance Base nodes that cache LR1 layer is LR1 and then the number of
Layer Laye nodes that cache LR2 layer is ( C-Nnodes-LR1 ). After our
(LR1,LR2) r assumption above, we obtain equations 2. Transposing terms
Single 13 0.87 100 NA and simplifying, in additional the equations 3 indicates the
Spatial 13 4 1.66 191 66.9% probability of w in proportion to the result of Nnodes * LR1 *
8 1.35 155 46.9% LR2.
13 0.96 110 31.4%
24 0.87 100 15.9% S = Pprob ∗ Rprob ∗ (1 − Pprob − Rprob) (1)

242 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 3, 2010
⎛ N nodes ⎞ ⎛ LR1⎞ ⎛ C− Nnodes− LR1⎞
S=⎜ ⎟×⎜ ⎟ ×⎜ ⎟ (2)
⎝ C ⎠ ⎝ C ⎠ ⎝ C ⎠

S = N nodes ∗ LR1 ∗ LR2 (3)
Where describe these equations in the Underneath.
S Probability of base layer, LR1 and LR2
Pprob Probability in the base layer.
Rprob Probability in the LR1 and the Probability of LR2 is (1- Pprob -
Rprob ).
C Number of members in the cluster head.
Nnodes Number of nodes that cache using base layer, the number of
node chacing in LR1 is LR1, and in LR2 it’s ( C-Nnode-LR1 ).
LR1, LR2 It’s the enhancements layers.
Figure 3. shown the Cluster head Using RBBSA

Also describe the number of base layer, and the enhancement H. The Mobile Management (MM)
layers LR1, LR2 versus the probability as we shown in Table 2. Ad Hoc Network to be provided by Mobility is a fully
wireless, can be any mobile network infrastructure without a
base station, all the nodes can be any link, each node at the
TABLE II. ILLUSTRATIONS THE PROBABILITY VS. WITH THE BASE LAYER, same time take router work with the Mobile IP completely
RL1 AND RL2 different levels of Mobility. And the Mobility becomes an
important issue in an ad hoc network. How cluster head
The The Results of
The number of
number of number of manage the mobility and join of members will be discuss in
(C)
96 12 12 LR1 *
(C *13824 LR2) this section. Equations number 4 shows the prediction of model
(RL1) (RL2)
84 24 12 24192 [16] which is introduced the mobility and join of members. The
72 48 12 41472 information of Global Positioning System (GPS) is introduced
into predication model; to get the Dt, the equations 4 indicates
60 36 24 51840
which it can predicate how long a node has left its cluster.
48 36 36 62208 Through the prediction model, a mobile node is able to detect
40 40 40 6400 the poor connection with the cluster. When the mobile nodes
detect a poor connection from the cluster, mobile nodes will
directly inform the cluster head in advance. The cluster head
There are 120 nodes with different rates to distribute numbers will forward a message to nearby cluster head when it receives
of members among caching base layer, enhancement LR1 and a message from the cluster head, so that the members of nearby
LR2. These layers will substitute the numbers of three different cluster can help cache videos. Through this way, the members
layers into the equation number 3 from Table 2. From this also can request for lost video clips from nearby cluster
equation the number of getting base layer, LR1 dan LR2. and members when the mobile node is moving to the nearby cluster
we can know that the numbers of getting base layer, LR1, LR2 and in the same time the when the cluster head provide the
layers distribute more averagely, the probability of “S “ is
video packet to the node within the transmission range and
higher.
some of the these nodes may have the enough resources, and
also as we mentioned earlier we use the Ad Hoc Network that
means the nodes can contact with each other and move freely.
G. Radius Based Binary Search Algorithm (RBBSA)
in this case the node is an MANET node which can have
According to the results in previous section, the proposed enough resources to provide Video services to other device
radius-based binary search algorithm cluster head can evenly with his transmission range. In figure 1 Illustrations his
distribute one of base, LR1 and LR2 layered video to clients for approach whereby PDA1 get the services from the head cluster
caching. also is working as a cluster head and providing services to
PDA2 and cell ph1.
Figure 3 is a flow chart which shows and describes how a
cluster head uses the radius-based binary search algorithm.
This flow chart also shows the position and distance between − (ab + cd )+ ⎛ 2 2⎞ 2
nodes and nodes by GPS system and at the same time assume ⎜ a + c ⎟Tx − (ab − cd )2
⎝ ⎠
that clients are built-in GPS system. Dt = (4 )
(a 2 + c 2 )

243 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 3, 2010
layered scalable video coding due to the BLR which it can be
( ) ⎛( independently decode and provide the version of the least
⎜ v1COSθ 1−v2COSθ 2 ) +
− v1COSθ −v 2COSθ * x1− x 2 + v1Sinθ * v 2Sinθ * y1− y 2 + 2
1 2 1 2
⎝ quality. In the same time, the probability of getting the base
= layer and the LR1 it’s almost 70%. It means the Radius Based
Binary Search Algorithm (RBBSA) can provide the better
quality of the video in Ad Hoc Network and better transmit
(v1Sinθ1 * θ
v2Sin (
)2 )Tx2 − v1COS1 − v2COS 2
θ θ * x1−x2 − θ
1
θ
v1Sin * v2Sin * y1− y2
2
)
2 video packet within a short distance. Also shown in figure 5
the Radius Based Binary Search Algorithm (RBBSA) get
2 2
(a + c ) cached the Video clips within almost 1.5 hops. In contrast the
random caching algorithm is within almost 2.5. In random,
caching algorithm has more than one hop caching distance
(( θ θ
v1COS 1−v2COS 2 )2 + (v1Sinθ1 * v2Sinθ2 )2 ) than the Based Binary Search Algorithm.

Where a is v1COS θ 1− v 2 COS θ 2 , b is x1 − x 2 , c is a
v 1 Sin θ 1 * v 2 Sin θ 2 and d is y1 − y 2 .

III. THE PERFORMANCE EVALUATIONS
We evaluated the performance of our system through a
simulation study and present the result of the study in this
section.

A. Simulation Model
Table 3 summarize of the input parameters for our simulation.
TABLE III. ILLUSTRATIONS THE PROBABILITY VS. WITH THE BASE Figure 4. peer to peer caching hit ratios
LAYER, RL1 AND RL2

The Parameters
TABLE IV. Notation Default Value

Clients Transmission Range d 10m

Number of nods Nnods 100
Number of the broadcast k 5
channels
Video Size in Mbyte Vsize 150

Clients arrival rate Carrival {2,4,5,8,10}
Session S 1000

B. The simulation result Figure 5. Illustrations the caching distance
The simulation results are shown in this section in the
IV. CONCLUSIONS
figures 4, present the probability of getting caching video and
we present three situations for the mobile nodes. The first We proposed a new peer to peer (P2P) caching technique
mobile nodes getting Base Layer, and the second nodes getting using the radius based binary search algorithm (RBBSA) and
the base layer and the LR1 layer, while the last mobile nodes mobile management. The mobile management can predicate
getting the three layers (Base Layer + LR1 +LR2). the time when mobile nodes leave the cluster head by
predication model, and after that the cluster head uses the
The Radius Based Binary Search Algorithm (RBBSA) get radius based binary search algorithm (RBBSA) to assign
the probability of caching video is generally almost 10% more members to cache video packets. Furthermore, when the
than the Random caching algorithm. This will explain in mobile nodes have enough resources the nodes can provide the
Figures 4 where the probability of getting Base Layer is 93%. services to the others. By the result of our simulations, it is
In this study, the base layer is the most important of the

244 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 3, 2010
appear that RBBSA and mobile management can promote
Peer to peer caching hit ratio and shorten the cache distance.
IEEE Internatio- nal Conference on Mobile Data Management (MDM),
ACKNOWLEDGMENT pp. 212–223, Berkeley, CA,January 2004.
Special thank and recognition go to my advisor, Associate [15] R. Cucchiara, M. Piccardi, A. Prati. “Neighbor cache prefetching for
Professor. Dr. Putra Sumari, who guided me through this multimedia image and video processing,” in Press on IEEE Transactions
on Multimedia, 2004.
research, inspired and motivated me. Last but not least, the
[16] W. Su and M. Gerla, “IPv6 Flow Handoff in Ad- Hoc Wireless
researchers would like to thank the University Sains Malaysia Networks Using Mobility Prediction,” Proceedings of IEEE
USM for supporting this research. GLOBECOM’ 99, Rio de Janeiro, Brazil, Dec. 1999, pp. 271-27.
[17] Zhang X,Liu J,Li B, and YUM TP (2005) coolStream/DONet : A data
driven overlay network for efficinalt live media streaming. INFOCOM
REFERENCES 2005. 24th Annual Joint Conference of the IEEE Computer and
Communication Societies, vol. 3, Marech, PP. 2102-2111.
[18] Padmanabhan VN, Wang HJ, And Chou PA (2004) Resilient Peer to
[1] Guo Y, Suh K, and Towsley D (2003) A peer to peer on Demand Peer Streaming. ICNP ’03: Proceeding of the 11th IEEE international
Streaming Service and it’s performance evaluation. ICME ’03: conference on Networks Protocols and multimedia,Atlanta, GA,
Proceeding of the 2003 international conference on multimedia and November, PP. 16 -27.
expo, Washington, DC.pp.649-652.
[19] Ye S and Makedon F (2004) Collaboration-aware peer to peer media
[2] Hefeed M, Habib A, Rotev R,Xu D, and Bhargava B (2003) PROMISE: streaming. MULTIMEDIA ’04: Proceeding of the 12th annual ACM
Peer to Peer media Streaming using collectCast.MULTIMEDIA ’03: international conference on multimedia, NY. Pp. 412-415.
Proceeding of the eleventh ACM international conference on
multimedia, Berkeley, CA, November,pp.45-54. [20] M. Tran and W. Tavanapong, “Overlay caching schemes for overlay
networks,” in Proc. SPIE/ACM Multimedia Computing and Networking
[3] Lu Y and Sebastien V (2007) Peer to peer media streaming application Conf., pp. 150-161.
survey.UBICOMM ’07 :international conference on Mobile Ubiquitous
Computing,Sysytem,Services and technologies. pp. 139-148.
[4] Saleh A. AlOmari, Putra Sumari and Sadik A. AlTaweel (2010) Peer to AUTHORS PROFILE
Peer Media Streaming System: A /V Conferencing System Using
Custom Protocol. IEEE Conference FutuerTech 2010. 5th International Saleh Ali Khattab Al-Omari has obtained his
Conference of Future Information Technology, Korea. Bachelor degree in Computer Science from Jerash
[5] http://www.comp.brad.ac.uk/~sburuha1/wirelessadhoc.htm University, Jordan in 2004-2005 and Master degree in
[6] In conclusion, health, free wireless network multi-path redundant Computer Science from Universiti Sains Malaysia,
mechanisms, in June 2008. Penang ,Malaysia in 2007. Currently, He is a PhD
candidate at the School of Computer Science,
[7] IEEE Computer Society LAN MAN Standards Committee, Wireless
University Sains Malaysia, Penang, His main research
LAN medium access control (MAC)
interest now on Video on Demand (VoD) over Heterogeneous Mobile Ad Hoc
and physical layer (PHY) specifications, IEEE standard 802.11, 1997.
The Institute of Electrical and Electronics Engineers, New York, NY, Networks (MANETs).
1997.
[8] IEEE Computer Society. IEEE standard for information technology
telecommunications and information exchange between systems – local Putra Sumari is currently working as Assistant
and metropolitan networks – specific requirements – part 11: Wireless Professor of School of Computer Science, Universiti
LAN medium access control (MAC) and physical layer (PHY) Sains Malaysia, Penang in 2009. Assoc. Prof Dr. Putra
specifications: Higher speed physical layer (PHY) extension in the 2.4 received his MSc and PhD in 1997 and 2000 from
GHz band, 1999. Liverpool University, England. Currently, he is a
lecturer at the School of Computer Science, Universiti
[9] Saleh A. AlOmari, Putra Sumari .”An overview of Mobile Ad Hoc
Sains Malaysia, Penang. He is the head of the
Networks For The Existing Protocols and Applications” The
Multimedia Computing Research Group, School of
International Journal on Applications of Graph Theory in Wireless Ad
hoc Networks and Sensor Networks (GRAPH-HOC),Vol.1, No.1, March Computer Science, USM. He is Member of ACM and IEEE, Program
2010. Committee and reviewer of several International Conference on Information
and Communication Technology (ICT), Committee of Malaysian ISO
[10] Xuefei Li, “Multipath Routing and QoS Provisioning in Mobile Ad hoc Standard Working Group on Software Engineering Practice, Chairman of
Networks”, Department of Electronic Engineering, Queen Mary, Industrial Training Program School of Computer Science USM, and Advisor
University of London, April 2000. of Master in Multimedia Education Program, UPSI, and Perak.
[11] D. Wu, Y. Hou, and Y.-Q. Zhang, “Scalable video coding and transport
over broad-band wireless networks,” Proc. IEEE, vol. 89, pp. 6–20, Jan. Mohammad Azmi Qasem Al-Madi has obtained his
2001.
Bachelor Degree in Computer Information System
[12] W. Li, “Overview of fine granularity scalability in MPEG-4 video (CIS) from Al-Zaytoonah Private University of Jordan
standard, ”IEEE Trans. Circuits Syst. Video Technol., vol. 11, pp. 301– in 2005 and his Masters degree in Computer Sciences
317. from Universiti Sains Malaysia, USM in 2009.
[13] TMN3.1 Currently, he is a PhD candidate at the School of
[14] Tran, Duc A., Le, Minh, Hua, Kien A. “MobiVoD: A Video-on-Demand Computer Sciences in USM. His main research
System Design for Mobile Ad Hoc Networks”. In: Proceedings of the interests are in Computer Networks, IPv6, Mobile IPv6 and 4G.

245 http://sites.google.com/site/ijcsis/
ISSN 1947-5500