![]() |
|||
![]()
|
![]() |
![]() Click Here! |
![]() |
HOW DESKTOP-TO-DESKTOP VIDEO AND AUDIO CONNECTIONS OPERATE IN IP NETWORKS Given the large number of freeware and shareware solutions for videoconferencing on IP networks and proposed standards, and the rapidity of new developments on the IP landscape, it is difficult to make generalizations about the manner in which desktop-to-desktop connections are negotiated, maintained, and torn down in all user applications. This section explores some of the approaches different products use to enable videoconferencing and collaboration over IP networks. Negotiating/Establishing Desktop Connections Some applications have relied primarily on Internet (and later, intranet) servers for negotiating desktop connections over LANs and WANs. White Pine Softwares Reflector software, for example, supports unicast, broadcast, and multicast sessions. Unicasting and multicasting are achieved by specifying a reflector as the destination and sharing (publishing) the appropriate IP address of the reflector in question with other conference participants. To control unwanted participants, the reflector lets network managers issue passwords for different conferences. In addition, a roster of conference participants (one for a unicast, more for multicasts) is published dynamically to all participating desktops. Several freeware and shareware programs are available to initiate and administer online conferences over IP networks. Confman is one such tool for conference initiation and monitoring that employs certain Mbone toolsvat (for audio data), vic and nv (for video transmission), and wb (for whiteboarding) on the Internet. Confman does not handle multimedia data, but helps the user to plan, set up, and control a conference by letting the network manager choose meeting participants (by IP address), the start time, and the tools (and codecs) the session members need to run on their end-points. Conferences can be held in two different modes:
Microsoft and Intels Internet telephony products (and subsequent IP-based videoconferencing offerings) use the standard User Locator Service (ULS) to negotiate/establish calls between desktop videoconferencing users on IP networks. In June 1996, ULS was submitted to the IETF for consideration to become a standard and incorporated with LDAP (lightweight directory addressing protocol). Ideally, all intranet and Internet service providers will have standards-compliant directories. For users not on a corporate network, the ISPs directory will automatically associate the E-mail address with a person. Every time users connect to the Internet, the network service can then pick up the IP address and initiate a local call to the end-point nearest the recipient. In a corporate environment, where there is a Novell network with ULS and computer telephony integration, simply having a desktop computer turned on and connected to the network should suffice to identify the end-point for any incoming video and audio (or audio-only) sessions. Maintaining/Modulating a Videoconference Session As long as the sockets between end-points, or between the end-point and server, are open, the IP session is maintained. Another way of expressing this is that within a conference, the virtual circuits are always present. Some applications use a single circuit for all audio, video, and data; other applications use a separate circuit for each media type. Having a separate socket for each permits higher error recovery and, therefore, fewer chances for problems to occur during a live conference.
|
![]() |
|
Use of this site is subject certain Terms & Conditions. Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details. |