Video conferencing standards and protocols are necessary to define common means for video encapsulation and for session management. Encapsulation standards define how video and audio are captured, converted to digital format and transmitted between endpoints. Popular standards include ITU H.264 for video, G.711/G.722/G.729 for voice, and H.239/T.120 for data such as screen sharing or Web conferencing.
An emerging protocol, H.264 Scalable Video Coding (SVC), is based on the idea of slicing individual frames into layers, with each layer holding part of a single frame's image. In the event of excess congestion or delay, H.264 SVC sessions still provide high quality even if endpoints can exchange only a small percentage of layered frames.
In other encapsulation approaches (such as H.264 or MPEG-4), congestion means that entire frames are lost, leading to visible degradation of video quality. The standard is still evolving. Where there is agreement on coding/decoding, the actual transmission of H.264 SVC frames is still under development within the ITU. Video conferencing vendors including Radvision and Vidyo (and their partners) have already introduced H.264 SVC solutions, but until final video conferencing standards are established, interoperability among different H.264 SVC solutions isn't possible.
In addition, not all vendors rely on open video conferencing standards for encapsulation. Microsoft Office Communications Server, for example, uses RTVideo, a proprietary video conferencing codec based on its VC-1 video codec.
Video conferencing systems also vary in their ability to support different frame rates (e.g., 30 or 60 frames per second), screen sizes and HD video modes (e.g., 720p, 1080i, 1080p), regardless of codec. Polycom recently announced H.264 High Profile, a subset of the H.264 standard. High Profile was originally designed to support Blu-ray DVD video. Polycom claims High Profile cuts HD bandwidth requirements in half compared with standard H.264. Polycom also offers proprietary approaches for error correction.
The introduction of proprietary encapsulation approaches and the lack of full standardization of H.264 SVC are obstacles for buyers wishing to integrate different products from different vendors. Vendors continue to perceive competitive advantage toward offering a "better than standard" solution while supporting baseline interoperability for well-defined video conferencing standards. One possible long-term solution is H.265, now under development by the ITU, but standardization and solutions are at least one to two years away.
One key area in which video conferencing standards are lacking is multivendor telepresence. While a team at the University of New Hampshire recently proved that multivendor telepresence was a possibility, their efforts required a great deal of engineering. Currently, it's not feasible for most enterprises to support interoperability among telepresence systems from Cisco, HP, Polycom and Teliris, to name a few. To address the need for telepresence interoperability, Cisco recently announced that it would release its Telepresence Interoperability Protocol (TIP) to the market via a royalty-free license. This protocol is designed to enable telepresence sessions across multiple vendors.
Though it may be relatively easy to create a multivendor video conference if all vendors rely on open video conferencing standards for all aspects of the video session, the reality is that many vendors still use some proprietary features in their systems. Some vendors provide gateways to enable cross-system interoperability, translating between proprietary and open protocols. Nevertheless, even with the use of gateways, there are still limitations. Often, the need to transcode between different codecs will reduce video quality or add unacceptable latency.
This was first published in February 2011