Multimedia Technologies:

Concepts, Methodologies,
Tools, and Applications

Syed Mahbubur Rahman

Minnesota State University, Mankato, USA

INFORMATION SCIENCE REFERENCE

Hershey • New York
Acquisitions Editor: Kristin Klinger
Development Editor: Kristin Roth
Senior Managing Editor: Jennifer Neidig
Managing Editor: Jamie Snavely
Typesetter: Michael Brehm, Jeff Ash, Carole Coulson, Elizabeth Duke, Chris Hrobak, Sean Woznicki
Cover Design: Lisa Tosheff
Printed at: Yurchak Printing Inc.

Published in the United States of America by

Information Science Reference (an imprint of IGI Global)
701 E. Chocolate Avenue, Suite 200
Hershey PA 17033
Tel: 717-533-8845
Fax: 717-533-8661
E-mail: [email protected]
Web site: http://www.igi-global.com/reference

and in the United Kingdom by

Information Science Reference (an imprint of IGI Global)
3 Henrietta Street
Covent Garden
London WC2E 8LU
Tel: 44 20 7240 0856
Fax: 44 20 7379 0609
Web site: http://www.eurospanbookstore.com

Copyright © 2008 by IGI Global. All rights reserved. No part of this publication may be reproduced, stored or distributed in any form or by
any means, electronic or mechanical, including photocopying, without written permission from the publisher.

Product or company names used in this set are for identification purposes only. Inclusion of the names of the products or companies does not
indicate a claim of ownership by IGI Global of the trademark or registered trademark.

Library of Congress Cataloging-in-Publication Data

Multimedia technologies : concepts, methodologies, tools, and applications / Syed Mahbubur Rahman, editor.
p. cm.
Includes bibliographical references and index.
Summary: "This book offers an in-depth explanation of multimedia technologies within their many specific application areas as well as
presenting developing trends for the future"--Provided by publisher.
ISBN 978-1-59904-953-3 (hardcover) -- ISBN 978-1-59904-954-0 (ebook)
1. Multimedia systems. 2. Multimedia communications. I. Syed, Mahbubur Rahman, 1952-
QA76.575.M5218 2008
006.7--dc22
2008021157

If a library purchased a print copy of this publication, please go to http://www.igi-global.com/agreement for information on activating
the library's complimentary electronic access to this publication.
 1409

Chapter 7.3
QoS Adaptation in
Multimedia Multicast
Conference Applications for
E-Learning Services
Sérgio Deusdado
Instituto Politécnico de Bragança, Portugal

Paulo Carvalho
Universidade do Minho Braga, Portugal

ABSTRACT cally according to the resources availability can

be considered a quality factor.Tolerant real-time
The evolution of the World Wide Web service has applications, such as video conferences, are in
incorporated new distributed multimedia confer- WKH IURQWOLQH WR EHQH¿W IURP 4R6 DGDSWDWLRQ
ence applications, powering a new generation of However, not all include adaptive technology
e-learning development and allowing improved able to provide both end-system and network
interactivity and prohuman relations. Groupware quality awareness. Adaptation, in these cases,
applications are increasingly representative in the can be achieved by introducing a multiplatform
Internet home applications market, however, the middleware layer responsible for tutoring the
Quality of Service (QoS) provided by the network applications’ resources (enabling adjudication
is still a limitation impairing their performance. or limitation) based on the available processing
Such applications have found, in multicast tech- and networking capabilities. Congregating these
QRORJ\ DQ DOO\ FRQWULEXWLQJ IRU WKHLU HI¿FLHQW technological contributions, an adaptive platform
implementation and scalability. Additionally, has been developed integrating public domain
considering QoS as a design goal at the application multicast tools, applied to a Web-based distance
level becomes crucial for groupware development, learning system. The system is user-centered (e-
enabling QoS productivity to applications. The student), aiming at good pedagogical practices and
applications’ ability to adapt themselves dynami- proactive usability for multimedia and network

Copyright © 2008, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
 QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

resources. The services provided, including QoS adaptation in mind; they need to employ built-in
adapted interactive multimedia multicast confer- mechanisms that allow them to probe the condi-
ences (MMC), are fully integrated and transpar- tions of the network environment and alter their
ent to end-users. QoS adaptation, when treated transmission characteristics accordingly (Miras,
systematically in tolerant real-time applications, 2002). Self-adaptive applications, in the sense of
denotes advantages in group scalability and QoS proactive behavior for transmission of continuous
sustainability in heterogeneous and unpredictable media in multiparty applications, are a well-ac-
environments such as the Internet. cepted solution due to the correct integration of
new services in today’s Internet (Deusdado, 2002;
Li, Xu, Naharstedt, & Liu, 1998).
INTRODUCTION E-learning, as a component of ÀH[LEOHOHDUQ-
ing, encompasses a wide set of applications and
Technology has been a strong catalyst for edu- processes which use available electronic media
cational innovation and improvement, especially to deliver vocational education and training. It
when the World Wide Web is involved. The next includes computer-based learning, Web-based
generation Internet needs technological support to learning, virtual classrooms and digital collabo-
accommodate promising new applications, such ration (Eklund, Kay, & Lunch, 2003). Our work
as interactive real-time multimedia distribution. aims to integrate interactive multimedia e-learn-
Predictable bandwidth availability and capacity ing applications in a proactive fashion taking
solvency imply QoS management to regulate into account the available network resources and
resources in heterogeneous environments. Ac- QoS sustainability. In this way, our motivation is
tually, increasing the network capacity through to offer improved learning experience based on
advanced network and media technology is not ultimate technology with QoS warranties.
per se DXELTXLWRXVDQGGH¿QLWLYHVROXWLRQWRRYHU- The system architecture proposed in this ar-
come the network capacity problem. Historically, ticle includes an adaptive module based on Java
the users have always managed to consume the applets and embedded Javascript, responsible
entire system capacity soon after it was enlarged for assessing the existing operating conditions,
(Ferguson & Huston, 1998). IP Multicasting E\FROOHFWLQJPHWULFVUHÀHFWLQJWKHFOLHQW¶VHQG
techniques (Deering, 1998; Kosiur, 1998; Moshin, system performance (e-student’s host), the cur-
Wong, & Bhutt, 2001; Thaler & Handley, 2000) rent network conditions and relevant multicast
are attractive solutions for this capacity shortage group characteristics. The collected data are
problem as bandwidth consumption is reduced subsequently computed weighting parameters
when network resources are shared. On the other such as the available bandwidth at the client side,
hand, the QoS support (Moshin, Wong, & Bhutt, the round-trip time between the client and the
 VKRXOGEHLQD¿UVWLQVWDQFHLQKHUHQWWR e-learning server, the client’s current CPU load
applications in order to integrate conveniently and free memory. The obtained results are used
enhanced real-time multimedia applications in the for proper multicast applications scheduling and
present Internet, barely QoS aware and increas- parameterization in a transparent way.
ingly heterogeneous.
With the advent of wireless and mobile
networks, heterogeneity is likely to subsist; en- MOTIVATIONS
visioned applications should merge QoS adapta-
tion and multicast in a proactive utilization of Basically, e-learning services are used to pro-
resources. Applications should be designed with mote connections between people (e-students)

1410
QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

Figure 1. QoS tolerance for generic audio and video applications (Miras, 2002)

packet
loss

desired acceptable

6-7%

2-3%

delay
150ms 200ms 300ms 400ms 8-10seg

and training resources (Steeples & Jones, 2002). source conditions and the applications’ inability
E-learning research is wide and growing in im- to assess available conditions and adjust internal
portance, especially in higher education. Several parameters before conference initiation. Without
institutions are developing interactive Web-based regulation, Real-time Transport Protocol (RTP)
learning systems, integrating rich media stream- WUDI¿FÀRRGVWKHQHWZRUNFDSDFLW\LQVHQVLWLYHO\
ing which may compromise network performance. forcing network congestion in certain cases or in-
The design of e-learning systems should consider hibiting better performance. A coherent behavior
QoS as mandatory for successful learning expe- RIDQDSSOLFDWLRQZLWKRXWDGDSWDWLRQLVGLI¿FXOW
riences, selecting the appropriate technologies in today’s Internet.
and applications, and regulating proactively the Public domain multicast applications used in
information and communications technology this article, vic (McCanne & Jacobson, 1995), rat
(ICT) resources utilization (Allison, Ruddle, (Hardman, Kirstein, Sasse, Handley & Watson,
McKechan & Michaelson, 2001). 1995) and Java Media Framework (JMF) (JMF
The Multicast Backbone (MBone) is a network 2.0, 1999) were designed with no QoS “sensors,”
overlaying the global Internet designed to support so the communication dynamics is not automati-
multipoint applications. MBone tools comprise a cally interdependent of end-systems or network
collection of audio, video and whiteboard applica- conditions. Effectively, such applications allow
tions that use Internet multicast protocols to enable preparameterization to adjust critical parameters
multiway communications (point-to-multipoint such as throughput, number of frames per second,
and multipoint-to-multipoint), satisfying most video and audio encoding formats and so forth.
of the needs of group communication, such as Adaptation, in these cases, can be obtained by
e-learning services. Using these applications by introducing a multiplatform middleware layer
common e-students drives recurrently to poor responsible for tutoring the applications’ resources
QoS satisfaction due to the heterogeneity of re- (adjudication or limitation) based on the available

1411
 QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

processing and networking capabilities (Miras, with heterogeneous equipment when they transmit
2002). audio and video to the group, as it happens in a
Common interactive real-time applications conventional classroom. If a client (e-student)
are fault-tolerant but suffer from QoS constraints; wants to interact and multicast video then the
low-latency requirements and reliability are cumu- system’s architecture will be integrated with fair
lative to achieve conference success. The diagram adjustments attending both to the connection to
in Figure 1 attempts to illustrate the QoS toler- the server (e-tutor) network and to the hardware
ance, in terms of delay and packet loss, for generic processing capabilities. Client’s adaptation should
interactive audio and video applications. not depend on the other group members because
The main motivation of this article is to pro- they are transient, and consequently stability of
vide adaptive behavior to applications used on transmissions could be very poor.
both sides of multimedia conferencing, focusing As the involved applications are characterized
essentially on multicast members that initiate by an intensive use of host and network resources,
audio and/or video transmission. The underlying the purpose of the middleware platform is to
idea is to launch automatically MMC applications DFKLHYHE\FRPSXWDWLRQRQDVFDOHRI¿YHGLI-
with proper audio and video codecs, bandwidth ferentiated modes, the proper integration of new
allocation inference and other parameters that multicasting members. Within this thematic, it
affect sustainability and scalability during an e- means implementing an adaptive learners’ par-
learning session. Our emphasis is on the concept ticipation in e-learning sessions by starting MMC
of “interactive e-learning services,” relegating the applications transparently, with their functionality
concept of “e-learning course” to a secondary goal, optimized for the current operating conditions.
which will be considered in future work. To clarify these aspects Figure 2 illustrates
Most prominent related work on friendly mul- the system architecture. As shown, three applets,
timedia transmission over the Internet, based on operating sequentially and interdependently,
a combination of system and network QoS feed- are responsible for monitoring and assessing
back implementing equation-based adaptation is QoS conditions, inferring, announcing and/or
summarized in Bouras and Gkamas (2003) and editing computed adaptation parameters. The
Vandalore, Feng Jain, and Fahmy (2001). process culminates with the initiation of MMC
applications, depending on the host and network
SUR¿OHVDQGFRYHULQJHYHQWXDOHQGXVHUH[SOLFLW
SYSTEM’S ARCHITECTURE requirements.
Audio and video encoding formats, frame
For multicast video distribution to heterogeneous rate and other quality metrics may be chosen ac-
users in an e-learning session, we assume that a cording to the resources’ availability, providing
class server (e-tutor’s system) should be distributed coherent, friendly and fair participation in the
and platform independent, considering inclusively network load balance. After monitoring sustain-
multitutoring. Thus, a class server should connect able network QoS with repeated measurements
to an e-learning server (Web server) and be sub- during approximately 15 seconds before media
mitted to adaptation as a regular new sender. The transmission, the Round-Trip Time (RTT) and
QoS requirements for the class server, operating bandwidth are calculated using a moving average.
in a centralized fashion, may justify the need of In addition, system’s status variables, such as pro-
layered multicast (Johanson & Lie, 2002; Liu, Li & cessor load, free memory, processor performance
Zhang, 2004), enhancing the service’s adaptation. and so forth, are acquired, taking advantage of
However, this work aims at integrating e-students operating system facilities.

1412
QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

Figure 2. System’s architecture

E-learning System - Services

Multicast multimedia conference applications

Rat Vic
: ›
Application’s Application’s Client’s QoS
QoS profile QoS profile profile

Middleware Applet
Multicast applications adaptator (audio and video)
LiveConnect
Applet
HTML
Forms LiveConnect Parameter visualization (edition is still possible)

Applet
LiveConnect
Available resources monitor

Operating System

Properties and available processing resources

Client’s System

Available bandwidth and elapsed round-trip time

Multicast Capable Network

Information Adaptation

The system is multiplatform as the included accepting user preferences. To achieve this goal,
applets differentiate the most popular operative applets and HTML forms interchange data using
systems (Windows and Unix), invoking appropri- Sun’s Liveconnect technology.
ate inner services to obtain instant measures for
the processor’s load and free memory. The col-
lected data constitutes another input to compute ADAPTIVE QoS FRAMEWORK
an adaptation index. Different compilations were
produced for common browsers. In the proposed framework, QoS management is
All of the adaptation process is transparent, performed individually for each new conference
however, regarding the experimental nature of member and occurs before the transmission’s
this work, each phase allows interaction with start, such that MMC applications are launched
the user, providing technical information or even adaptively facing the previous QoS sensing period

1413
 QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

7DEOHYLFDQGUDW4R6SDUDPHWHUVXVHGWRDGMXVWDSSOLFDWLRQV¶SUR¿OH

Rat -f format Indicates audio encoding format: l16, pcm, dvi, gsm and lpc

Vic -B kbps Sets the maximum bandwidth slider (kbps)

On a color-mapped display, uses the algorithm indicated by dither (e.g., ed,
Vic -c dither
gray, od, quantize) to convert to the available color palette
Vic -f format Indicates the video encoding format: h261, h263, jpeg, nv, ...

Vic -F fps Sets the maximum frame rate (fps)

Table 2. Set of parameters for different QoS adaptation modes

ADAPTATION MAXIMUM FRAME VIDEO AUDIO
COLOR
MODE BANDWIDTH RATE CODEC CODEC
5 1 Mbps 30 fps H.261 Yes L16

4 512 kbps 25 fps H.261 Yes PCM

3 256 kbps 20 fps H.261 Yes DVI

2 128 kbps 15 fps H.263 Yes GSM

1 64 kbps 10 fps H.263 No LPC

conditions. QoS variability during the conference video quality (Bolot, Crépin, & Garcia 1995). The
is not used to dynamically readapt the applica- success of video conferencing communication
tions. If an e-student experiences lack of QoS also depends on factors such as received frames
while conferencing, the membership process per second, image quality, resolution, size and
should be restarted. Corroborating this practice, illumination.
MMC applications, especially vic, are not stable For this work, the representative parameters of
enough. In fact, if some critical adjustments are vic and rat used to modulate QoS are presented
PDGHRQWKHÀ\WKHUHVXOWLVRIWHQWKHFROODSVHRI in Table 1.
the application. Nevertheless, dynamic adaptation The values for these parameters, deriving from
is currently a subject of study within group com- a mathematical expression that generates an adap-
munication applications (Layaida & Hagimonte, tation mode based on the sustainable QoS level,
2002; Tusch, Böszörményi, Goldschmidt, Hell- compose a set of adjusting directives determining
wagner & Schojer, 2004). the applications’ behavior. Each adaptation mode
Considering the applications’ specificity indexes the respective set of adjustments, which
DQGW\SHRIWUDI¿FJHQHUDWHGDGDSWDELOLW\RQO\ will then be passed to the application. Since QoS
includes interactive audio (rat) and video (vic) scale varies from mode 1 to 5, when the obtained
applications and services. The heuristics regard- result is under or over this range it will be assigned
ing the choice of applications’ QoS parameters to the nearest limit. Equation (1) determines the
HPHUJHGIURPH[SHULPHQWDOUHVXOWVDQGVFLHQWL¿F adaptation mode to be applied:
references in this matter (Wu, Hou, Zhu, Zhang,
& Peha, 2001). For instance, video conference M = (int) (B/(RTT/2) + FM/P) *K
users typically require better audio quality than (1)

1414
QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

where 1, E-LEARNING SERVICES AND

M = QoS adaptation Mode (Table 2); FEATURES
B = Bandwidth (kbps); RTT = Round-Trip Time
(ms); The developed distance e-learning system pres-
FM = Free Memory (MB); P = Processor load ents numerous features providing distinct service
(%); levels, such as:
K = 1/50 - constant to scale the result (1 to 5).
 9LUWXDODFDGHP\:HEEDVHGZLWKUH¿QHG
For vic (version 2.8), the video encoding usability, integrating authentication and
formats H.261 (ITU-T H.261, 1993) and H.263 services for the e-learning community.
(ITU-T H.263, 1998) were those who revealed 2. Registration, authentication and mainte-
best performance for e-learning purposes, lead- nance of educational agents.
ing to low loss ratios and high reliability. H.263 3. Multicast sessions maintenance and schedul-
is especially appropriate for low bandwidth ing.
environments.

Figure 3. E-learning system screenshots and MMC’s adaptation HTML forms based on available QoS

1415
 QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

4. Access to asynchronous material such as Certain processes for assessing hardware

video on demand, slide presentations and performance require user’s explicit authorization,
other multimedia resources. allowing extended security privileges to applets in
5. Interactive multimedia multicast confer- order to perform system’s inspection and collect-
ences with QoS adaptation. ing substantial data used by subsequent applets
6. Other multicast tools for shared work- RIWKHFRQWUROSDWK7KHVHFXULW\FHUWL¿FDWHVXVHG
space. LQWKLVZRUNDUHQRWSURYLGHGE\RI¿FLDOHQWLWLHV
7. Discussion spaces such as forum and mul- but generated by applet compilation tools for
ticast chat room. testing.
Although the adaptation process is totally
This information system incorporates online transparent, effectively, the users may edit QoS
databases structuring courses, students, tutors and parameters suggested by the system. If editing
sessions’ data. These resources were developed occurs, correctness and validation are assured
using MySQL/PHP. A Web site congregating all by embedded Javascript code for parsing pur-
developed application component prototypes is poses.
available at www.esa.ipb.pt/multicast.

Figure 4. BW needs for each QoS mode

100%
90%
CPU Occupation .

80%
70%
Pentium II 333 MHz 192 MB RAM
60%
50% Pentium III 700 MHz 256 MB RAM
40%
Pentium IV 1700 MHz 256 MB RAM
30%
20%
10%
0%
1 2 3 4 5

Adaptation Modes

Figure 5. CPU needs for each QoS mode

1416
 QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

an IP network to preserve their operational in- the Multicast Monitor (www.multicastmonitor.

tegrity. Beyond routing improvements, QoS in a com) to collect and handle the resulting data.
multimedia conference needs primarily to deal The e-learning system, more concretely the
with several sources with different characteristics, tested prototype showed good performance indica-
VKLIWLQJ ODUJH DPRXQW RI WUDI¿F FRPSHWLQJ IRU tors that validate the architecture model proposed.
network capacity. MMC applications may easily %HFDXVH YLGHR WUDI¿F LV TXDQWLWDWLYHO\ PRUH
absorb all network resources and the subjective representative of resource consumption, it was
quality sensed by users, would remain poor if analyzed preferentially. Figures 4 and 5 exhibit
the available resources are used indiscriminately. the levels of resource consumption for each QoS
As mentioned earlier, the adaptation purpose, mode considered. Here, the overhead introduced
with e-learning in mind, was to integrate MMC by middleware to prepare applications is marginal
applications with QoS conscience, preserving as it occurs before transmission time.
resources in order to maintain conference quality QoS adaptation, when treated systematically
and improve scalability. in tolerant real-time applications, denotes advan-
In order to test the framework, different tages in group scalability and QoS sustainability
scenarios were simulated and the corresponding in heterogeneous and unpredictable environments
UHVRXUFHV¶FRQVXPSWLRQYHUL¿HGFRQVLGHULQJWKH such as the Internet and Mbone. Figures 6, 7 and
QoS limitations associated with each QoS mode 8 illustrate a comparison between two simulated
GH¿QHGLQ7DEOH)RUYLGHRFRQIHUHQFLQJUHJDUG- VHVVLRQV WKH ¿UVW ZLWKRXW 4R6 DGDSWDWLRQ DQG
ing e-learning purposes, it is widely accepted that the second including adaptation managed by the
reference values correspond to “Maximum qual- developed middleware layer. The results show that
ity, few action scene.” Bandwidth consumption scalability increased, but equally important is the
in vic default mode is 128 kbps. When adaptation IDFWWKDWDSSOLFDWLRQVPD\EHQH¿WIURPUHVRXUFH
is requested, the different adaptation modes use availability that does not occur when using the
the values charted in Figure 4. For instance, the GHIDXOW DSSOLFDWLRQV FRQ¿JXUDWLRQ :KHQ WKH
best quality mode consumes around 400 kbps, available resources decrease, the system allocates
allowing better image and motion. them to critical parameters. For instance, while
Different equipment was also tested in order to the frame rate should not be below 10 fps, the
YDOLGDWHWKHUDQNRIWKHGH¿QHGDGDSWDWLRQPRGHV image quality may be poor or monochromatic if
We observed that modern high performance equip- the contents are correctly perceived.
ment tends to be neutral; in this case adaptation Limiting bandwidth to applications, not only
ZLOOEHLQÀXHQFHGRYHUDOOE\QHWZRUNFRQGLWLRQV with explicit parameterization but also choosing
but with mobile computation in mind, PDAs and the right encoding format for e-learning sessions,
cellular phones, CPU performance should not be DOORZVHI¿FLHQWUHVRXUFHXWLOL]DWLRQDQGSURDFWLYH
relegated. usability, avoiding network overload and conges-
The experimental results were obtained tion. If the network load remains high, it is easier
varying the number of new multicast members to recover if adaptation is used.
transmitting voice and video, considering that The experience with Mbone showed that e-
all the multicast group members are multimedia learning groups tend to be small, usually less than
receivers and transmitters capable using Any 20 members. Effectively, e-learning communities,
Source Multicast (ASM) technology via vic and as in traditional training methods, need a tutor,
rat applications. Due to the limited number of who is mainly an educational agent and not neces-
multicast monitoring tools publicly available, we sarily a learning technologist. Indeed, questions
use embedded applications resource meters and related with communication technologies will

1418
QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

not constitute pedagogical limitation if intelligent group scalability, QoS sustainability and proactive
QoS management autonomy is provided natively resource utilization.
or by middleware to applications. Comprising multiple sources (even unauthor-
(OHDUQLQJFRQIHUHQFLQJVSHFL¿FLW\UHTXLUHV ized ones), ASM involves high complexity and
appropriate video encoding formats able to achieve may compromise the success of e-learning confer-
low loss ratio and fast recovery from congestion. ences. Future work includes the use of Source-
We compared H.261 and Motion JPEG (ITU- 6SHFL¿F 0XOWLFDVW 660  +ROEURRN  &DLQ
JPEG, 1992) performance in experimental ses- 2004) in order to overcome this limitation. The
sions, using a modest PC (PIII 0.7GHz - 256 MB development of new multilayer video encoding
RAM). The results were penalizing for IRUPDWVFRXOGDOVRLQFUHDVHWKHÀH[LELOLW\ZKHQ
MJPEG, where loss was about 30%, in op- using QoS adaptation. When cumulative layers
position to 1% for H.261. are transmitted avoiding redundancy, using dif-
ferent SSM groups or channels, adaptation can
be performed in a transparent way in order to
CONCLUSION AND DISCUSSION DFKLHYHHI¿FLHQWUHVRXUFHVXWLOL]DWLRQ -RKDQVRQ
& Lie, 2002; Liu, Li & Zhang, 2004).
The goal of this article was, in one topic, to foster
“ecological” practices in the Internet when using
MMC’s applications in e-learning services. The REFERENCES
proposed system integrates public domain mul-
ticast applications for synchronous media com- Allison, C., Ruddle, A., McKechan, D., &
munication, being supervised by a middleware Michaelson, R. (2001). The architecture of a
based QoS management framework, intending framework for building distributed learning
to preserve the QoS of critical parameters for environments in advanced learning technolo-
HOHDUQLQJVHVVLRQ¶VVSHFL¿FLW\ gies. In Proceedings of the IEEE International
As main contributions, this work: Conference on Advanced Learning Technologies,
Madison,WI: IEEE Press.
1. Provides an integrated e-learning envi-
Bolot, J., Crépin, H., & Garcia, A. (1995). Analysis
ronment based on interactive multimedia
of audio packet loss in the Internet. LNCS, 1018,
services with proactive QoS;
pp. 154-165. France: INRIA.
2. Improves the usability of MMC applications;
and Bouras, C., & Gkamas, A. (2003). Multimedia
3. Allows the development of end-to-end QoS- transmission with adaptive QoS based on real
aware multimedia conferences, coordinat- time protocols. International Journal of Com-
ing resources from network, end-system munications Systems, 16, 225-248
processing equipment and applications.
Deering, S. (1998). Multicast routing in internet-
works and extended LANs. In Proceedings of
Middleware adaptation is a solution that suits
ACM SIGCOMM (pp 55-64).
the present state of Internet and the requirements
of new multimedia distributed applications. We Deusdado, S. (2002). Integração Adaptativa de
use a middleware layer to manage QoS adapta- Aplicações Multicast para Conferência Multi-
tion in interactive audio and video applications média. Unpublished master thesis, Universidade
coordinating resource demand, monitoring and do Minho.
adjudication. Substantive results were obtained in

1419
 QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

Eklund, J., Kay, M., & Lynch, H. (2003). E-learn- Li, B., Xu, D., Naharstedt, K., & Liu, J. (1998).
ing: Emerging issues and key trends. Australian End-to-end QoS support for adaptive applica-
Flexible Learning Framework, Australian Na- tions over the Internet. Unpubluished manuscript,
tional Training Authority. University of Illinois at Urbana-Champaign.
Ferguson, P., & Huston, G. (1998). Delivering Liu, J., Li, B., & Zhang, Y. (2004). An end-to-end
QoS on the Internet and in corporate networks. adaptation protocol for layered video multicast
New York: John Wiley & Sons. using optimal rate allocation. IEEE Transactions
On Multimedia, 6(1), 87-102.
Hardman, V., Kirstein, P., Sasse, A., Handley, M.,
& Watson, A. (1995). RAT, Robust Audio Tool. 0F&DQQH6 -DFREVRQ9  YLF$ÀH[-
Retrieved July 27, 2006, from http://www-mice. ible framework for packet video. In Proceedings
cs.ucl.ac.uk/multimedia/software/rat/ of the ACM Multimedia Conference. Retrieved
July 27, 2006, from http://www-mice.cs.ucl.
Holbrook, H., & Cain, B. (2004). 6RXUFHVSHFL¿F
ac.uk/multimedia/software/vic/
multicast for IP. Unpublished manuscript. Interent
Draft, IETF, <draft-ietf-ssm-arch-07.txt>. Miras, D. (2002). A survey of network QoS needs
of advanced Internet applications. Unpublished
ITU-JPEG. (1992). JPEG Standard, Information
manuscript, University College London.
Technology- Digital Compression and Coding of
Continuous-Tone Still Images- Requirements and Mohsin, M., Wong, W., & Bhutt, Y. (2001). Sup-
Guidelines, Recommendation T.81, ITU. SRUWIRUUHDOWLPHWUDI¿FLQWKH,QWHUQHWDQG4R6
issues. Unpublished manuscript, University of
ITU-T H.261. (1993). ITU-T Recommendation
Texas at Dallas.
H.261, Video CODEC For Audiovisual Services
At p x 64 kbits. Steeples, C., & Jones, C. (2002) Networked
learning: Perspectives and issues. London, UK:
ITU-T H.263. (1998). ITU-T Recommendation
Springer.
H.263, Video Coding For Low Bitrate Commu-
nication. Thaler, D., & Handley, M. (2000). On the aggre-
gatability of multicast forwarding state. Tel-Aviv,
JMF 2.0 (1999). JMF- Java TM Media Framework
Israel: IEEE INFOCOM.
API Guide, Sun Microsystems, JMF 2.0 FCS.
Tusch, R., Böszörményi, L., Goldschmidt, B.,
Johanson, M., & Lie, A. (2002). Layered encod-
Hellwagner, H., & Schojer, P. (2004). Offensive
ing and transmission of video in heterogeneous
and defensive adaptation in distributed multimedia
environments (Tech. Rep.), Sweden: Department
systems. ComSIS, 1(1), 49-77.
of Computer Engineering, Chalmers University
of Technology. Vandalore, B., Feng, W., Jain, R., & Fahmy, S.
(2001). A survey of application layer techniques
Kosiur, D. (1998). IP multicasting: The complete
for adaptive streaming of multimedia. Real Time
guide to interactive corporate networks. New
Imaging, 7(3), 221-235.
York: John Wiley & Sons.
Wu, D., Hou, Y., Zhu, W., Zhang, Y., & Peha,
Layaida, O., & Hagimonte, D. (2002). Dynamic
J. (2001). Streaming video over the Internet:
adaptation in distributed multimedia applica-
Approaches and directions. IEEE Transactions
tions (Tech. Rep.) INRIA Saint Ismier, France:
on Circuits System Video Technology,11(3)
Rhône-Alpes Research Unit,
282–300.

1420
QoS Adaptation in Multimedia Multicast Conference Applications for E-Learning Services

ENDNOTE

1
RTT and/or P values will be, if necessary,
assigned to 1 to avoid division by zero. To
prevent incongruence, the maximum band-
width allowed cannot exceed the detected
value (B), otherwise the computed mode will
suffer cyclic decrements while the excess
remains and M>1.

This work was previously published in International Journal of Distance Education Technologies, Vol. 4, Issue 4, edited by
S.-K. Chang & T. K. Shih, pp. 56-68, copyright 2006 by IGI Publishing, formerly known as Idea Group Publishing (an imprint
of IGI Global).

1421