Priority Based Congestion Control for Multimedia Traffic In 3G Networks

Report as inappropriate Your report has been received

Description

IJCSIS is an open access publishing venue for research in general computer science and information security. Target Audience: IT academics, university IT faculties; industry IT departments; government departments; the mobile industry and computing industry. Coverage includes: security infrastructures, network security: Internet security, content protection, cryptography, steganography and formal methods in information security; computer science, computer applications, multimedia systems, software, information systems, intelligent systems, web services, data mining, wireless communication, networking and technologies, innovation technology and management. The average paper acceptance rate for IJCSIS issues is kept at 25-30% with an aim to provide selective research work of quality in the areas of computer science and engineering. Thanks for your contributions in September 2010 issue and we are grateful to the experienced team of reviewers for providing valuable comments.

Shared by: ijcsis

Stats

views:: 278
posted:: 10/10/2010
language:: English
pages:: 7

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Priority Based Congestion Control for
Multimedia Traffic In 3G Networks
Neetu Sharma1, Amit Sharma2 , V.S Rathore3, Durgesh Kumar Mishra4
123
Department of Computer Engineering, Rajasthan, India
13
Rajasthan College of Engineering for women, Rajasthan, India
2
Shri Balagi College of Engineering & Technology, Rajasthan, India
4
Acropolis Institute of Technology and Research, Indore, MP, India
neetucom10@gmail.com, amitit_04@rediffmail.com
drvsrathore@rcew.ac.in,drdurgeshmishra@gmail.com

ABSTRACT- There is a growing demand for efficient TCP friendly rate control (TFRC) and Adaptive increase
multimedia streaming applications over the Internet and next multiplicative decrease (AIMD) used in networks. These algorithms
generation mobile networks. Multimedia streaming services are used for multimedia traffic but not much effective in packet loss.
receiving considerable interest in the mobile network business. TCP is the dominant transport protocol in the Internet, and the
As communication technology is being developed, the user current stability of the Internet depends on its end-to-end congestion
demand for multimedia services raises. The third generation control, which uses an Additive Increase Multiplicative Decrease
(3G) mobile systems are designed to further enhance the (AIMD) algorithm. End-to-end congestion control of best-effort
communication by providing high data rates of the order of 2 traffic is required to avoid the congestion collapse of the global
Mbps. High Speed Downlink Packet Access (HSDPA) is an Internet [11]. While TCP congestion control is appropriate for
enhancement to 3G networks that supports data rates of several applications such as bulk data transfer, some real-time applications
Mbit/s, making it suitable for applications like multimedia, in (that is, where the data is being played out in real-time) find halving
addition to traditional services like voice call. Services like the sending rate in response to a single congestion indication to be
person-to-person two way video calls or one way video calls, aim unnecessarily severe, For providing a better congestion control with
to improve person-to-person communication. Entertainment higher data rates a new effective scheme is used. Congestion control
services like gaming, video streaming of a movie, movie trailers is an important issue in both wired and wireless streaming
or video clips are also supported in 3G. Many more of such applications. Multimedia applications should use some form of
services are possible due to the augmented data rates supported congestion control, both in wired and cellular networks, in order to
by the 3G networks and because of the support for Quality of adapt the sending rate to the available bandwidth. Today’s Internet
Service (QoS) differentiation in order to efficiently deliver stability is due to TCP and its congestion control algorithm. TCP
required quality for different types of services. represents a very efficient transport protocol in general and is
suitable for data transfer. However, it has been argued [13] that TCP
This paper present congestion control schemes that are suitable
is unsuitable for video streaming because strict delay and jitter
for multimedia flows. The problem is that packet losses, during
requirements of video streaming are not respected by TCP.
bad radio conditions in 3G, not only degrade the multimedia
Moreover, some TCP retransmissions are unnecessary for video
quality, but render the current congestion control algorithms as
when data may miss the arrival deadline and become obsolete. This
inefficient. This paper proposed a solution that integrated the
has led researchers to look for alternative options. Most of the work
congestion control schemes with a priority based multimedia
related to congestion control for video flows has either emulated TCP
packets to increase the speed of multimedia data and reduce the
or has used the TCP model. The well-known TCP-Friendly Rate
packet loss that is developed due to congestion in networks
Control (TFRC) congestion control consists in an equation based rate
Key words: UMTS, CN, BS, TFMCC, UTRAN, RNC
control mechanism [13][14][15], designed to keep a relatively steady
sending rate while still being responsive to congestion. When used
I. INTRODUCTION
over wireless links, TFRC and TCP cannot distinguish between the

The emerging multimedia application requires a fresh approach for wireless losses and the congestion losses. They both may suffer from

congestion control. A widely popular congestion control schemes are the link underutilization if the connection traverses a wireless link.

167 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

This is because they consider dropped packets as a sure sign of UMTS networking architecture is organized in two domains. The
congestion and reduce the ending rate significantly. The inability to user equipment (UE) and the public land mobile network (PLMN).
identify a wireless loss followed by unnecessary reduction in sending The UE is used by the subscriber to access the UMTS services.
rate results in link underutilization. PLAN is further divided into two land-based infrastructures
A. UMTS Introduction
(i) UTRAN (UMTS terrestrial radio– access network)
Universal Mobile Telecommunications System (UMTS) is a third-
(ii) CN (core network).
generation (3G), wireless cellular network that uses Wideband Code
Division Multiple Access (WCDMA) as its radio interface The UTRAN handless all radio-related functionalities and the CN is
technology. UMTS offers higher data rates with respect to older 2G responsible for maintaining subscribes data and for connections.
and 2.5G networks and, with the Release 5 version, is evolving into UTRAN contain two types of nodes Radio network controller (RNC)
an all-IP, wireless network. The increased bandwidth provided by and Node B. Node B is the base station and provides radio coverage
UMTS allows for the deployment of a wide range of services, like to one or more cells. Node B connected to UE via Uu interface and to
voice, data and multimedia streaming services. In wireless networks, the RNC via Iub interface. Uu is a radio interface based on the
congestion control, alone, may not be enough to ensure good quality wideband code division multiple access (WCDMA) technology [7].
of multimedia streaming and efficient utilization of the network.
The CN consist by two types of general packet radio service support
Packet losses due to the high bit error rate not only degrade the
nodes (GSNs). That is gateway GSN (GGSN) and serving GSN
multimedia quality, but render the current congestion control
(SGSN). SGSN provide the routing functionality. It manages a group
algorithms as inefficient: these algorithms back-off on every packet
of RNCs and interacts with the home location register (HLR). HLR
loss even when there is no congestion. We integrate the congestion
permanently store the subscriber data. SGSN connected to GGSN via
control schemes with an adaptive retransmission scheme in order to
the Gn interface. RNC connect to SGSN via Iu interface. Through the
selectively retransmit some lost multimedia packets. Fig.1 shows the
GGSN the UNTS network connect to external packet data network
transmission of multimedia data over a wireless channel.
like the internet.

Fig. 1 Transmission of Multimedia data Fig.2 General UMTS Network

C. 3G/UMTS Problems
B. GENERAL UMTS NETWORK:
 Problems due to the use of IP
o IP doesn’t support real time streaming
UMTS, the successor of GSM, is evolving toward a future wireless
requirements
all-IP network. In this paper we present how it supports real-time IP
o Overhead due to packet header
multimedia services, as these services are expected to drive the
 Problems due to radio conditions
adoption of wireless all-IP networks.
o Scarce and time varying bandwidth
o Congestion, wireless losses & large delay

168 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

D. UMTS QoS Classes control profiles that may be used with the DCCP transport protocol
UMTS defines four QoS classes [2] and the classified traffic gets the [10]; TFRC may also be implemented by UDP-based applications
treatment inside the UMTS network according to its class. The four wishing to perform congestion control. This paper presents a
QoS classes are: simulation study of TFRC over UMTS networks supporting

• Conversational class: The traffic from the applications like person- HSDPA. Since we are interested in video streaming applications, we
to-person video or voice call is classified into conversational class. evaluate the performance of TFRC in terms of rate stability over
The delay and jitter requirements for this type of traffic are very different time scales, and compare it with that of TCP. Several
strict. This is because on the both end points there is a human scenarios of MAC-layer scheduling, radio conditions and background
expecting the delivery of the voice and/or video data in very short traffic are considered.
time after it is sent.
This paper proposed a more reliable algorithm that provides
• Streaming class: Video on Demand (VoD) falls under this class. congestion control for different multimedia classes. Priority assigned
The delay requirements are there but are not as strict as the to each of the packet according to multimedia classes. So whenever
conversational class. the congestion occurs in the network the lowest priority packets are
dropped. If overall loss rate for lower priority packets is not very
• Interactive class: The interactive traffic like interactive e-mail or
high, then we can safely assume that the congestion loss rate for the
web browsing falls under this category. Though there is still some
highest priority packets will be insignificant. In such a case, the loss
delay requirement, it is less strict than the conversational and
of highest priority packets will be mainly due to wireless errors.
streaming classes. Moreover, since the traffic mostly pertains to data
Thus, it is to be expected that, in general, there is a good correlation
applications, the bit error rate should be very low.
between wireless packet loss rate and the total loss rate of highest
• Background class: This class is the most insensitive to delay. It priority packets.
includes the traffic from background applications like background
email and SMS. Though, the bit error rate, like the Interactive class,
2. THE PROPOSED SCHEME

should be very low. This paper provides a mechanism of congestion control for the

D Congestion Control for Multimedia data multimedia transmission over UMTS. We analyze TCP friendly
multicast congestion control (TFMCC) over UMTS and generalize it
TCP-Friendly Rate Control (TFRC) is an end-to-end congestion
to different multimedia classes [5][6]. We design a novel mechanism
control mechanism, whose goal is to provide rate control for unicast
for congestion control that is Content Sensitive TCP Friendly
flows in IP networks. The main feature of TFRC is its ability to
Multicast Congestion Control (CSTFMCC). We perform a little
smoothly adapt the sending rate of a flow to network conditions,
modification in UMTS network and the packet field. At various level
while competing for bandwidth with TCP flows in a relatively fair
of network we provide the control mechanism that prevents the
manner. TFRC was designed to offer a more stable sending rate than
network from the congestion. Multimedia Class I traffic includes
TCP on wired, best-effort networks, making it suitable for
video and audio traffic from users equipped with an adjustable rate.
applications like multimedia streaming. We evaluate the performance
Class II traffic includes non-real time data traffic such as e-mail, file
of TFRC, compare it with that of TCP and new TFRC for different
transfer and web browsing traffic. These two classes contain different
multimedia classes, under different scenarios of MAC-layer
multimedia traffic that is more delay sensitive or less delay sensitive.
scheduling, radio conditions and background traffic.
So class I traffic support the real time applications and more delay
TFRC [4][10] is an end-to-end congestion control mechanism sensitive. Due to congestion, if any loss of the packet or the delay
suitable for applications with constraints on rate stability, like voice between the packets can reduce the quality of received video/audio.
or streaming media. It has been designed to adapt the sending rate of Whether in class II traffic, if congestion occur it is acceptable to
a flow in a smooth manner, while trying to fairly share the available buffer non-real time data at a network node or at the user station and
bandwidth with competing TCP flows. TFRC is an Internet standard transmit them at a slower rate. In a large multicast group, there will
[4], and it has been adopted at the IETF as one of the congestion usually be at least one receiver that has experienced a recent packet

169 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

loss. If the congestion control mechanisms require that the sender If (incoming request for higher priority packets)
If (there is a free channel) then
reduces its sending rate in response to each loss, as in TCP, then
allocate the free channel
there is little potential for the construction of scalable multicast else
If (lower priority packets)
congestion control.
put in a buffer
If (there is free channel again)
In wireless communication systems like UMTS, the packet loss may allocate the free channel to lower
not mean network congestion. The quality of wireless link may be priority packets
else
degrading due to signal fading. During a fading period, the bit error Ignore request
rate of wireless link may become very high but after that period the endif
else
wireless link is expected to recover. TFMCC uses a feed back Ignore request
scheme which allows the receiver to calculating the slowest endif
ignore request
transmission rate to always reach the sender. endif
endif
End.
A. Sender end: Fragmentation of data packets perform at the sender
end. The sender fragment the data packet with on bit of priority. Fig.3 Algorithm For proposed model

There are two parts of the data (i) packet header and (ii) payload. The
size of header part change by one bit shows the priority of packet. So Fig. 4 shows the flow chart for the proposed model. Flow chart
shows the arrival of packet and priority check by the routers at
there is only one bit modification perform in the size of data packet layer2. According to this priority the packets is being processed.
and it increases the speed of multimedia packets.
Packet Arrival

B. Multimedia Packet Size: Multimedia packet size depends on the High Check priority of
Packets?
Low

multimedia classes. The proposed scheme redesigns the multimedia Process the
higher priority Process the
packets lower priority
packets
packets. It increases the multimedia packet size by one bit. This bit
shows the priority of multimedia packets. The highest priority Any Free No Any Free Yes
Channel?
Channel?

packets serve first by the routers at the layer 2. So the size of the Yes No

Yes
Buffer full ?
packet is increased by one bit. No
Accept packet and Block (reject) Buffer the Accept packet and
assign channel the packet packet assign the channel

C. Routing Scenario: For fast transmission of multimedia
information the proposed scheme give the priority to all multimedia
Fig. 4 Flow Chart for the proposed model
packets. When a user want to send multimedia data the data framing
perform at the sender end. The sender constructs the frame with a
priority bit. This information stored in the header of the packet for Receiving End: At the receiving end defragmentation perform. The
priority access to the router. Sender sends the packets towards its receiving data packet reaches at the destination and multimedia
destination. Multimedia packets reach at the network. At layer two information is available for the user respectively
the router checks the destination address and priority bit of the
A. Simulation Platform
packet. If a higher priority packet arrives then router serves first to
the packet which contains a highest priority. This increases the speed The simulation that we use for this is EURANE (NS-2
of multimedia packets and decreases the congestion in the networks. Extension)[16]. Following fig. 3 shows the simulation topology to
increase the multimedia quality. This paper focuses on the problem
For implementing this scenario the changes perform in the size of
of evaluating the subjective video quality and presents the quality
packets and in routers. Following algorithm shows the scenario for
estimation tool that we employed. a performance evaluation study
routing the various packets according to priority. Fig. 3 shows the
done with the well known ns-2 network simulator
algorithm for the proposed model.

170 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

[6] Ljiljana Trajkovic and S. Jamaloddin Golestant”
Congestion Control for Multimedia Services” proc of
IEEE INFOCOM 1992

[7] Antonios Alexiou, Dimitrios Antonellis and Chritos Bouras
“Equation Based Congestion Control for Video
Transmission Over WCDMA Networks”proc of IEEE
Fig:3 Simulation Topology
AINA’06 Vienna, Austria, pp .445-450.
the video packet trace file is fed to the ns-2 simulator (compiled with
the EURANE extensions). This trace file serves as a traffic generator [8] Minghua Chen and Avideh Zakhor “Rate Control for
during the simulation. A simulation script allows defining the Streaming Video over Wireless” proc of IEEE
particular scenario under consideration (network topology, INFOCOM 2004
simulation parameters, and so on). When the simulation is run, an
[10] Kamal Deep Singh*, Árpád Huszák., David Ros§, César
output trace file is produced which contains delay- and loss-related
Viho* and Gábor Jeney.
information for every video packet sent by the (simulated) video
“Congestion Control and Adaptive Retransmission for
server.
Multimedia Streaming over Wireless Networks” The
Conclusion Second International Conference on Next Generation
Mobile Applications, Services, and Technologies
This paper represents to design a multicast group congestion control
[11] Kamal Deep Singh∗, Julio Orozco†, David Ros‡ and
mechanism over the UMTS networks. This mechanism is content
sensitive and optimized for multimedia traffic. This paper performs Gerardo Rubino “Improving perceived streaming-video

some changes in existing UMTS architecture and the packet field of quality in High Speed Downlink Packet Access” proc of

data packet. User equipment (UE) has some additional function to IEEE IEEE "GLOBECOM" 2008

detect the multimedia class type. The additional functionality of UE [12] S. Floyd and K. Fall. Promoting the Use of End-to-end

and the field of data packet has main target to remove the packet loss Congestion Control in the Internet. IEEE/ACM

and congestion. Transactions on Networking, Aug. 1999.
[13] R. Jain, K. Ramakrishnan, and D. Chiu. Congestion
Refrences
Avoidance in Computer Networks with a
[1] Holma. H. and Toskala, A. “WCDMA for UMTS: Radio Connectionless Network Layer. Tech. Rep. DEC-TR-
Access for Third Generation Mobile Communications” 506, Digital Equipment Corporation, August 1987.
rd
(3 Edition),Wiley,2004.
[14] S. Floyd, M. Handley, J. Padhye, and J. Widmer .Equation
[2] Antonios Alexiou, Dimitrios Antonellis and Chritos based congestion control for unicast applications., In
Bouras” A Study of Multicast Congestion Control for Proceedings of ACM SIGCOMM 2000, pages 43.56,
UMTS” proc Int. J. Commun. Sys. (2009) Stockholm, Aug. 2000.

[3] Miiighua Clien and Avitleh Zaklior”Rate Control for [15] S. Floyd, E. Kohler, and J. Padhye, .Profile for Datagram
Streaming Video over wireless “proc of IEEE Congestion Control Protocol (DCCP) Congestion Control
INFOCOM 2004 Hong Kong China ID 3: TCP-Friendly Rate Control (TFRC)., RFC 4342,
IETF, Mar. 2006.
[4] J. Widmer, R. Denda, M. Mauve “A Survey on TCP-
Friendly Congestion Control” IEEE Network,15(3), [16] M. Handley, S. Floyd, J. Padhye, and J. Widmer. TCP
May-June 2001 Friendly Rate Control (TFRC): Protocol Specification.
[5] Cui-Quing Yang and Alapati V.S. Reddy “A Taxonomy Internet Standards Track RFC 3448, IETF, Jan. 2003.
for congestion control algorithms in packet switching
[16] Ns-2 . Network Simulator,
networks”proc IEEE 1995
http:///www.isi.edu/nsnam/ns/index.html

171 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Authors Profile Architecture, Operating System Fundamentals,
DBMS & RDBMS (Oracle, DB2, SQL Server, MS-
Ms. Neetu Sharma, Reader Access, DBASE, etc.), Data Structures,
Programming Languages (C, C++, Java (J2SE,
J2ME, J2EE), VB, COBOL), Networking
Technologies (Data Communications, Internet &
Intranet, E-Commerce, Network Security,
Cryptology etc.), Software Engineering, System
Analysis & Design, Management Information
System, Decision Support System, Artificial
Intelligence, E-Governance, Computer Center
Management, UNIX, etc.. He is the member of
Biography: Mrs. Neetu Sharma obtained her renowned society like ISTE.
Engineering degree from University of Rajasthan
and Masters Degree from Rajasthan Vidyapeeth, Dr. Durgesh Kumar Mishra
Udiapur securing First division with honors in both. Professor (CSE) and Dean (R&D),
Currently she is pursuing Ph.D. (CSE) in Acropolis Institute of Technology and Research,
Congestion Control in 3G from Gyanvihar Indore, MP, India,
Ph - +91 9826047547, +91-731-4730038
University, Jaipur, India. She has been Reader and Email: durgeshmishra@ieee.org
HOD of the department of CSE at Rajasthan
College of Engineering for Women, Jaipur, India. Chairman IEEE Computer Society, Bombay Chapter
She has extensively worked in various field of Vice Chairman IEEE MP Subsection
Computer Engineering. She has published many
national papers in the reputed journals and
conferences. She is an author of the book 'System
Software Engineering' for B.Tech. students. She is
the member of renowned societies like IEEE, IEEE
computer society, ISTE and CSI also.

Dr. Vijay Rathore, Associate Professor

Biography: Dr. Durgesh Kumar Mishra has
received M.Tech. degree in Computer Science from
DAVV, Indore in 1994 and PhD degree in
Computer Engineering in 2008. Presently he is
working as Professor (CSE) and Dean (R&D) in
Acropolis Institute of Technology and Research,
Indore, MP, India. He is having around 21 Yrs of
teaching experience and more than 7 Yrs of
research experience. He has completed his research
Biography: Dr. Vijay Singh Rahore obtained his work with Dr. M. Chandwani, Director, IET-DAVV
MCA and Ph.D. (CSE) from University of Indore, MP, India in Secure Multi- Party
Rajasthan, India. He is an Associate Professor, Computation. He has published more than 60
Shree Karni College, Jaipur, India. He has more papers in refereed International/National Journal
than 10 years of industrial and teaching experience. and Conference including IEEE, ACM etc. He is a
His areas of interest are Computer Organization & Senior Member of IEEE, Chairman of IEEE

172 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 8, No. 6, September 2010

Computer Society, Bombay Chapter, India. Dr.
Mishra has delivered his tutorials in IEEE
International conferences in India as well as other
countries also. He is also the programme committee
member of several International conferences. He
visited and delivered his invited talk in Taiwan,
Bangladesh, Nepal, Malaysia, Bali-Indonesia,
Singapore, Sri Lanka, USA and UK etc in Secure
Multi-Party Computation of Information Security.
He is an author of one book also. He is also the
reviewer of tree International Journal of
Information Security. He is a Chief Editor of
Journal of Technology and Engineering Sciences.
He has been a consultant to industries and
Government organization like Sale tax and Labor
Department of Government of Madhya Pradesh,
India.

Mr. Amit Sharma, Assistant Professor

Biography: Mr. Amit sharma obtained his MCA
from University of Rajasthan and aboout to
complete his M.Tech.(CSE) from Rajasthan
Technical University. He is an Assistant Professor
in Sri Balaji College of Engineering & Technology,
Jaipur. He has more than 5 years of industrial and
teaching experience. His areas of interest are Open
Source, Networking, Advance Computer
Architecture and Information Security. He is the
member of renowned society like ISTE.

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