Abstract:
Video conferencing is one of the most demanding services that a future network infrastructure may have to support. This is because of the combination of a need for interactivity and a high bandwidth demand. Video conferencing might not demand as much interactivity as virtual reality or as much bandwidth as a hard disk backup, but still enough to make any network analysts pull his hair out by the roots trying to support both at the same time. The same difficulties applies inside the computer, where the same amount of data have to be moved multiple times between different devices within the same overall time constraints as the network. Thus, video conferencing pose quite a challenge for both network and computer designers and in this report It is pointed out to push the limits of both. The end users of a video conferencing system are humans and the system should be designed to help the users to conduct a meeting where the participants are distributed in different locations. Transfer effects and association play an important role in the user's decision to adopt a new medium, so for a video conferencing system to succeed it should provide an audio and video quality comparable to that of other services offered in similar environments. The practical implications of the theories presented in this report is demonstrated in the design and implementation of a video conferencing system. using standard computers interconnected by an IP-based network that offer better than broadcast TV resolution and still maintain a good interactivity. A test with the prototype seems to support earlier observations that; even if rework bit rates increases rapidly, the computer design of today has problems to catch up. The designer of systems providing high-bandwidth, networked, real-time, interactive multimedia services like video conferencing has to be aware of this trend
