LEVERAGE
LEarn from Video Extensive Real Atm Gigabit Experiment

LEVERAGE News No 1, September 1996

Welcome | Co-ordinator | Pedagogical | InterACTS | Acronym key


Technophile

Juha Leppanen, VTT Information Technology in Finland Juha Leppanen from VTT Information Technology in Finland explains how videoconferencing is being integrated and used in the LEVERAGE project.

Videoconferencing was selected as the technical means to arrange realistic face to face interaction between users (in our case students and tutors in Cambridge, Paris and Madrid) working at remote workstations. VTT is building the LEVERAGE videoconferencing system in collaboration with Telecom Finland Ltd and Bitfield Oy.

Each workstation is a standard PC running Windows95. Additional hardware is needed for viewing the video in the application window and coding and decoding the videoconferencing video and audio. The videoconferencing has to work over the experimental LEVERAGE ATM-network using generic IP-sockets with TCP or UDP protocols.

The system uses H261 video streams and G711 audio streams (both ITU-T standards) as the quality required by the LEVERAGE application is satisfied by these standards. Quality is ensured by broadband data transmission over an optical network. In theory the video transmission can handle up to a maximum of 2 Mbits/s.

At the time of designing and implementing the first trial system no other usable standards were available for videoconferencing over packet-switched networks. In future the standard H323 which defines how video codecs can use any IP network to talk to each other will be considered.

The most straightforward mode of videoconferencing is that of point-to-point conferencing, i.e. a pair of workstations talking only to each other. However, the LEVERAGE project also requires groups of up to six users at remote workstations to be able to communicate as a group through multipoint videoconferencing.

Each workstation generates its own video and audio streams. In multipoint conferencing the streams can be copied to everyone else in the group by using the multicasting available in IP networks. As each user needs to be able to hear and see all other users, the audio signals from all the workstations need to be summed and the video signals from the workstations need to be combined into a split screen video window.

Processing the videoconferencing data and delivering the processed result to every workstation in the group requires an additional Server MCU (Multipoint Control Unit) with enough capacity to achieve that result. T-series standards by ITU-T are used as the basis for developing the MCU.

Different modes of videoconferencing used in the LEVERAGE project


The figure above shows how the LEVERAGE network can be used in the two modes:
  1. in point-to-point videoconferencing two workstations are logically connected to each other. This can serve as private conferencing but also as fall-back conferencing, because no additional server is needed.
  2. in multipoint videoconferencing workstations are logically connected to the server. Group videoconferencing and other collaborative work software can use the various services provided by the server.

We expect that the experience gained with LEVERAGE will contribute to the future development of high-quality videoconferencing for use in broadband-based learning and teaching systems.

About LEVERAGE | Conferences | Deliverables | Partners | Press Archive | Related Projects

LEVERAGE News 1 - Sept 96 | 2 - April 97 | 3 - Sept 97 | 4 - July 98 | 5 - Feb 99


LEVERAGE home page

Last updated 1st June 1999
E-mail: leverage@cilt.org.uk