network is quite complicated because it connects seven sites in four countries
and consisting of the following core components: · Four ATM LANs ·
Seven Ethernet based LANs · Four ATM Cross Connects · Eight static
IP routes · Seven multicast routers · Two 2 Mbps leased lines
· Ten International ATM links (virtual path) · One basic rate ISDN
link The following is the figure of logical network infrastructure: Figure 4 :
Logical Network Infrastructure. The ATM infrastructure that represents ATM
public network provider consists of a single ATM XC at each internationally
connected sites. These XCs are interconnected by permanent VPs (Virtual Paths).
The ATM CPE at each site based on one or more Fore System ASX-200 switches. It
is employed as logically separate ATM LANs besides as providing ATM access
between public network and routers at each site. The following is the figure of
ATM configuration. Figure 5 : ATM Configuration. The IP configuration consists
of routers at each connected sites being connected by Permanent Virtual Channel
(PVC) running over VPs. The routing function at each site performed either by
dedicated hardware router or by workstations running routing daemon software.
Routing of multicast IP packets (used for multimedia conferencing applications)
is not fully supported by most current IP routers, therefore, routing performed
by multicast routing daemon (mrouteds) running on workstations. The mrouteds are
interconnected by unicast IP tunnels, which can be used to be routed via routers
together with all other unicast traffic. The IP tunnels between mrouted at
internationally connected sites used the second sets of VPs. This supports
partition of multicast traffic from other unicast traffic and thus enables
provision of more deterministic Quality of Service (QoS) for multimedia
conferencing application. For external infrastructure, the aim is to provide
international ATM links between IP routers at the customer sites. Parallel VPs
are used between each pair of sites; one for multicast routing and another one
for unicast routing. Figure 6 : The network configuration Reflecting the
contemporary trends in multimedia and information services, all software
communication is over IP, including management system traffic. For the network
infrastructures that are conducted at a single site, the requirement its network
is fairly simple, requiring simply Ethernet connection to support IP
communication between PCs and workstations. If the system includes the
management of connections over IP switches, then the network infrastructure
would include both a representative public network ATM cross connect and
customer premises network ATM work-group switch (a FORE systems ASX200). These
are connected and configured with multiple VPs to emulate a network with a
larger number of nodes. IP routing functions in this network are provided by the
SPARC workstations with ATM interface cards performing IP forwarding. The
following is the network configuration of this kind of network: Figure 7 :
Network configuration For this network configuration, the TES Customer is able
to request the set-up of a new connection to the TES provider. The TES provider
then requested the VPN provider to do likewise. The VPN provider made a request
to the Public Network Provider and Customer Premises Network Provider to ensure
that the end-to-end IP/ATM connection was in place for the TES Customer. This is
the goal for the configuration scenario. One of the most important on an ATM
network level management system is to provide end-to-end connectivity across
constituent ATM network element, and so support the connectivity provisioning
with fault management and quality of service features. Challenged by these
requirements, a system that is able to set up ATM Virtual Paths and to correlate
faulty conditions, determining how these fault effect the connectivity for each
end user has been built. The following is the Network infrastructure of this
system: Figure 8 : Network infrastructure The figure shows that all the network
equipment is connected to one Ethernet hub, that is, the hub that acts as a
backbone for one Public Network domain and two Customer Premises Networks. In
reality, this hub could be partitioned into a number of internets that are
inter-connected by routers, also known as the Internet. For the network that is
required to operate over six sites in four different countries, would require a
much more comprehensive network infrastructure. This infrastructure consisted of
an ATM VP service, leased lines, and the internal ATM and IP network
infrastructure. The following is the example of this network infrastructure :
Figure 9 : Network infrastructure 4.4 Performance of Tele-educational Service
4.4.1 Courses There were two courses, both aimed at students with above average
prior knowledge of computing and/or computer networks. The first, an
introduction to SQL, was a self-study course, consisting mainly of modules of
written text with assessments based on these. The second course, an introduction
to ATM, was led by a tutor and involved varied methods of delivery, including
lecture/seminar, individual study and group work. Students were therefore
expected to interact both with one another and with the tutor. This course, too,
included assessment modules. Both of the courses were offered over a three-day
period and students were expected to participate for three half days. Within
this time, those taking the SQL course was able to pace their own study. On the
ATM course, the students? use of the different resources was timetabled and
directed by the tutor. Time was divided between events, such as lectures, at
which all students were expected to be present, and study time, during which
they would work through a series of modules, with assessment associated with
each one. 4.4.2 Students There were 16 students on the more interactive of the
two courses, the Introduction to ATM, and a similar number on the self-paced
study course, An Introduction to SQL. All the students appeared to be
experienced computer users. This has to be accepted as necessary in a trial such
as this , which takes place in the context of a research project which uses
leading edge technology, some of it is still being tested. The prototypical
nature of parts of the system may make unusual demands on the students, such as
imposing unexpected delays. Having students who appreciate the difficulties may
well be important. Having said this, it appeared that although they were
knowledgeable about computers, these students were not experts in networked
multimedia technology, and did need some initial training in the use of the
software. This was given prior to the start of the course. The courses were
clearly directed at this target group, as their titles suggest. The students
also stated that they had a genuine wish to learn the subjects being offered and
that this was a major motivating factor. They were also paid for their
participation, which may have helped improve their persistence when there were
technical hitches. 4.4.3 System The system used for the ATM course is described
here. Those taking the SQL course used only those parts suited to self study.
There are three main elements: audio, and video communications channels support
a Tele-education system built on a web-browser base, but with considerable
functionality added. The audio tool, rat, allows participants to receive and
transmit audio, to identify who is speaking, control the volume of incoming and
outgoing audio streams. Since this tool was developed as a research platform,
there are many extra features which the average end-user is not likely to use in
an application such as this one, for example, the facility to change the audio
encoding scheme. The tool?s basic functionality is easy to learn and use. The
video tool, vic, also offers functionality suitable for its use as a platform
for research into networked video. For the non-expert, however, the most
important features are that multiple users can send and receive video
simultaneously and that they can control some features of both display and
capture/transmission (image size and frame rate are two examples). Video images
can be displayed at various sizes from thumbnail image to CIF. Enlarging images
does, however, involve creating a new window for each one. Students access the
Tele-education system via a web browser and navigate within it using hypertext
links, buttons and active areas of images. Initial access is password protected
and the system supports the notion of groups and hence, presumably of multiple
classes and tutorial groups. The interface is based on the metaphor of an
educational institution, a building divided into rooms whose function most
students will be able to predict from their real-life experience of education:
classroom, tea room, hall, office, library and seminar room. Users are presented
with an aerial view of the layout, in which the rooms are labeled. They gain
access to a room by clicking on the appropriate part of this image. The
resulting window sometimes maintains the metaphor but is more often mainly
textual – a list of hypertext links, for example. Once "in" a room,
students have access to the resources they need for the part of the course they
are taking. As might be assumed from the description, the system is intended to
support a mixed mode of course delivery, including lectures, group discussions
and assignments, individual study, assessment with feedback. The existence of
the office implies that students can also access relevant course administrative
information. The Hall and tea rooms suggest that the intention is also to
support less formal, social interactions. 4.4.4 Positive Findings The courses
both seemed to be appropriate for the target group. Students reported that they
believed they had learned a considerable amount and felt they would retain the
important points. The pacing of the study also seemed successful. The tutor
clearly had a sense that this was a real class in a real institution and made
considerable efforts to generate a relaxed and positive atmosphere. Use of
students? names, and greeting them as soon as they logged in, contributed to
this. This is no mean achievement, given the constraints. The tutor tended to
refer to the environment as if it were a real place, arranging with students,
for example, to "meet in the tea room" or telling them to "go to
the library". Whether the students shared this perception is less clear.
This may be due to the short time available to become familiar with it. It would
be interesting to see whether the environment would become more "real"
to the students over a longer course. The room-based structure therefore seems
to have been successful. The metaphor seems to have been well chosen, since
students seemed to have appropriate expectations of each "room". None
of them appeared to have difficulty navigating between different rooms.
Observation did show that some students had to scroll up and down repeatedly,
however, when they were working on individual study texts. This seemed
particularly to be the case where they found the material more difficult. Again,
there was no sign that they were unsure of where to go or had difficulty in
navigation. In terms of course delivery, the trial showed that students
experienced considerable variety in the ATM course (inevitably less so in the
SQL course). Not only this, but the tutor seemed able to exploit the flexibility
of the system and to direct the student to alternative areas of study from what
had been planned originally, if necessary. One of the problems with distance
education is that such flexibility can be harder to achieve than in a
face-to-face situation, so this is promising and an interesting result of having
different applications integrated in this way. It also has a pragmatic use:
given technical problems in one area, it was possible to shift students to
another activity quite easily. Interactivity, both structured and casual was
potentially considerable. The shared whiteboard used for group work was
perceived by students as a good feature. It seemed, however, that they did not
all realize at first that they could write and draw on it. Perhaps this should
be pointed out in the introductory sessions, or the whiteboard should be
accompanied by a short explanatory note. It would also be fair to say that this
was not a long enough trial to assess usability of this part of the system. In
the limited time it was also not easy for students to establish relationships.
The system and the way the tutor used it did encourage students to get to know
one another since, for example, one of the first activities for students was to
upload their CVs and pictures and to browse through those of other students. The
level of concentration appeared to be high. Naturally, as in a classroom, there
were moments when students? attention moved away from the subject of study but
these were not frequent. Interestingly, they usually stayed at the workstation
but moved to another activity such as reading e-mail. The students observed
"live" appeared to maintain concentration despite considerable
background noise and other potential distractions. This is not a surprise, since
other computer-based teaching and learning trials have drawn similar conclusions
- but it is another promising feature. At best, the material with which the
students were engaged appeared well designed for delivery on a computer screen.
The information was "packaged" into manageable chunks and was visually
stimulating. Diagrams, colour and animation were used effectively, and the
layout was clear and appealing. As the next section suggests, however, not all
of the written material was so suitable for this method of presentation.
Feedback was given to students both by the tutor, during discussions (for the
ATM course), and as a result of assessments done at the end of each module.
Students appeared to take these assessments seriously and were observed to
return to the relevant part of the notes when unsure or when they had given an
incorrect answer. The scope of this evaluation did not extend to assessing the
course design or the assessment methods, but it is worth mentioning that the
regular assessment seems to have been a successful feature of the course.
Awareness of other students is something that is hard to achieve in distance
education. Interestingly, with the audio channel left open during private study
periods, it appeared that students experienced something similar to working in a
library with other students around them. They were able to hear conversations
and could have asked questions if they needed to. The potential disadvantage is
that the additional background noise might interfere with concentration. It
would probably be worth investigating whether the availability or otherwise of
the audio channel makes a difference to students. 4.5 Operation and Management
of Tele-education Service A vital element of any service is the reliability,
configurability and administration of that service. In order to ensure success
of an educational service from both the participants? and tutor?s
perspectives, the delivered service must be well managed and monitored. It is
crucially important to realize the software and procedures necessary to manage
and deliver Tele-educational services over broadband networks. Four basic
principles for successful teaching in a virtual classroom environment have been
identified as · media richness, · interaction, · timely
responsiveness and · organization of materials. Media richness and
interaction mechanisms can be satisfied by the educational services described
earlier. The organization of course materials and the insurance of timely
response by systems, participants and tutors are goals of the management
service. During the delivery of a course, there is a significant mass of
material presented to participants as well as a high degree of interactive
responses amongst participants. Unless this mass of materials is organized and
interaction controlled, participants can become confused and disillusioned.
Proper maintenance and management of the dissemination of material must be put
in place to provide an effective learning environment. Segregation of material,
both between and within course modules should also be supported. The strategy of