In the absence of a shared multiplexed network, the design space for multisite conferencing is quite limited. There are two ways you could set up a multisite conference. Firstly, have multiple CODECs at each site, and multiple circuits, one from each site to all the others. This would involve n*(n-1) circuits in all, and n CODECs at each site to decode the incoming video and audio. Alternatively one could design and use a special purpose Multi-point Control Unit, which mixes audio signals, and chooses which video signal from which site is propagated to all the others.
With this latter approach, each site has a single CODEC, and makes a call to the MCU site. The MCU has a limit on the number of inbound calls that it can take, and in any case, needs at least n circuits, one per site. Typically, MCUs operate 4 to 6 CODECs/calls. To build a conference with more than this many sites, you have multiple MCUs, and there is a protocol between the MCUs, so that one build a hierarchy of them (a tree). Essentially, this must reproduce some of the funcitonality of routing in the multicast distribution schemes of the mbone. However, currently, such systems are manually configured.
To select which site's video is seen at all the others, a floor control protocol is required, since the CODEC for circuit based video can usually only decode one signal. FLoor control protocols distribute information through a control channel to the conference sites concerning who is speaking, who wishes to speak, and who can select who speaks next. They may be triggered by who is speaking, simply by automatic detection of audio activity above some threshold level for some minimum threshold period, or on a chairman approach, i.e. through human intervention. Thus ISDN based conferencing systems must include a data channel between the terminals to carry this control protocol, and have a user interface to allow users to exercise this protocol. The diagram 7.1 illustrates the use of an MCU to link up 3 sites for a circuit based conference.
Should greater than basic rate ISDN be needed, it can be combined via a device that synchronises a set of parallel ISDN channels together, as shown in figure 7.2. This was once known as a ``bonding'' box, and is neccessary because seperate channels are not guareanteed to be routed over the same path through a circuit switched digital telephone network, so may incur different delays. Thus if data is striped over these different channels, it is necessary to calibrate the delay offset of the bits at the receiver end so that the origianl data order can be recovered.
Next: Distributed Multisite Circuit BasedConferencing Up: ITU Model H.320/T.Gcc Previous: ITU Model H.320/T.Gcc Jon CROWCROFT