John Reister
Chief Architect,
IPTV
Q. Almost everyone is talking about IPTV these days. What do you feel is missing from the dialog?
A. Although a lot of clever people are adding their opinions to the debate, there's not enough appreciation of the QoS (quality of service) needed to make IPTV a reality. If access networks don't have sufficient video processing intelligence, service quality becomes "optimized" around the lowest common denominator.
Q. What do you mean by "lowest common denominator"?
A. Well, as many of us have experienced personally, the signal quality in DSL-based networks degrades rapidly with distance: the farther you are from the DSLAM, the lower the bit rate. Telecom providers want to reach as many subscribers as possible, but must reduce the bit rates of video as low as possible to get the best coverage. The video quality for everyone within the coverage area is driven by the subscribers on the perimeter — the lowest common denominator.
Q. Couldn't this problem be solved with higher bit rates and managing congestion?
A. Well, of course, if there's unlimited capacity in the access network, all the problems go away. In reality, though, the access network is where bandwidth is scarcest. Access devices can help by prioritizing video, but that doesn't solve the problem. What happens when only 8 Mbps is available to a deliver a 9 Mbps video stream? Even if the video is prioritized over other data and voice services, the quality may be terrible, with frozen images or very disturbing deterioration. Now consider the effects of adding admission control. What happens? The subscriber's set-top box is denied the stream and the TV screen goes black.
Q. How do you solve the lowest common denominator?
A. The solution to the lowest common denominator problem is to have more video smarts at the edge of the network. The current batch of IPTV network designs keep the video functions centralized at super headends, or locate it at video hub offices. Deploying video-aware networking platforms at the edge of the network will give telecom providers the ability to meet per-subscriber QoS requirements.
Q. What results from having control of per-subscriber QoS requirements?
A. We anticipate that service providers will be able to provide video quality that is at least equivalent to that of the cable industry, while extending coverage areas by 20% and more. This translates into more customers, fewer complaints and improvements in customer service costs.
Q. What can you reveal about BigBand Network's IPTV solution?
A. BigBand Networks has a long track record of introducing technologies that improved video networking functionality, both in the core of the network and at its edge. Switched broadcast is chief among these, but the list also includes zoned ad splicing, program-level protection switching and rate shaping. Our IPTV solution will leverage our video processing and routing experience, and enable benefits such as targeted addressable advertising. We're continuing to apply our innovative culture to develop exciting new IPTV opportunities.