/
11.13.2007
Originally featured on BroadcastEngineering.com
Delivering the Quality of Experience viewers expect is critical to IPTV success, says Brix Networks CTO

If IPTV providers wish to compete successfully with satellite and cable operators, they must deliver the same user experience TV viewers have come to expect of those media.

According to Brix Networks VP of product management and CTO Kaynam Hedayat, so far they are generally failing to hit that mark. Facing problems in their core networks and particularly in the last mile and in the homes of consumers, IPTV operators have a formidable task ahead of them. However, Hedayat says there are sound approaches available to tackling the problem of assuring Quality of Experience.

IPTV Update: Could you put the overall issue of network quality into perspective when it comes to IPTV networks?

Kaynam Hedayat: The requirements for the quality of networks have reached a new level in terms of IPTV. If you look at the evolution of services on the Internet and over IP networks, we first started with data. Data was e-mail, and then transitioned into Web and transactions. There were some requirements in terms of quality of service, but ,in general, it was something that fixed itself. It was driven by technology, and really not people.

For example, a Web page would show up in 10 seconds. Over time, with broadband, better servers and the evolution of technology, we experienced better quality. Pages would come up faster, and our personal experience with the Web became much more friendly.

Then came VoIP. And with it, demands all of a sudden were higher. Now, small packet loss or small latency would impact user experience and conversation quality. In that way, VoIP put a lot of pressure on providers to manage their networks better and pay a lot of attention to quality of service. At the same time, VoIP, by its nature, is a bit resilient to quality impact on a network. For example, if you and I are in a conversation and one of my sentences drops because of packet loss, you can come back and say, “What did you say? I didn’t hear you.” And I would repeat what I said. Also with VoIP, we were so conditioned at the time that particular technology was emerging by our experiences with the mobile market that we were sometimes willing to trade quality for lower-cost calls.

Interestingly enough, the demands for quality are getting higher and higher at the same time the mobile market — with the advantage of mobility — trained us to accept some level of bad quality. So people just psychologically were willing to accept some impairments. Again, I don’t want to diminish the importance of quality of service to VoIP, but the pressure on the providers was not as high as some people had expected.

Now with IPTV, the name of the game has changed. There are no more excuses, and operators know that. For example, if while watching the Red Sox in the recently concluded World Series, I missed an important play because there was huge packet loss followed by a froze frame on the screen, I wouldn’t have called Verizon and said, “Please unfreeze that frame for me.” I would have probably called Verizon and canceled my service.

IPTV Update: What about network demands?

Kaynam Hedayat: With IPTV, the demands on the network are much higher. In VoIP, you are talking about 64kb/s for a voice session. In IPTV, we are talking about 2Mb/s to 10Mb/s for one channel.

Then there is the whole personal experience. People are going out and purchasing high-definition TVs for thousands of dollars. They are expecting crisp pictures on their screen without any pixelizaton for any impairment. If they experience anything bad, they simply change their service provider.

With Telco TV and providers coming to the game, the stakes are extremely high. They’ve spent a lot of money and they know they have to win customers, and they have to win based on quality.

IPTV Update: So, in your view are IPTV roll-outs meeting the quality of service and user experience customers expect?

Kaynam Hedayat: As a technology vendor, I wish my answer was “yes.” But as someone who has experienced IPTV firsthand and has also talked to a lot of providers, I have to say they have not met these goals.

The reasons are multiple. I believe all providers underestimated the challenges as to the demands on the network, both in terms of bandwidth and the quality of their network. Second, I believe a lot of providers underestimated the quality of the network of the last mile. I am mainly talking about providers that are using DSL for the last mile.

With our experience, right now we are seeing 10 to 20 percent of the problems happening in the core of the network, and 80 percent happening in the last mile and the home of the customer.

So, in short, I believe they have underestimated the challenges and now they are hurrying to address them. I believe they can address the core because it is under their control, and in the core network, they have the option of throwing routers and bandwidth at the problem.

In the last mile, the problems are more challenging. You are at the mercy of the environment. You are challenged by the customer’s home network. I believe after the technology challenges of rolling out the network, the next challenge will be monitoring and understanding and managing the customer end to end — basically having the ability to look into what the customer is experiencing. It’s what we call the Quality of Experience (QoE). After QoE, there is the need of having enough information to turn around and solve those problems very quickly. The key is detection of QoE problems and the ability to turn around and correct problems in a matter of minutes instead of hours.

IPTV Update: What approaches are there to monitoring that Quality of Experience?

Kaynam Hedayat: The good news is a lot of solutions are evolving. It all starts with monitoring in the core and extending it to the edge and customer network. Approach in the core and edge is monitoring of all channels and VoD streams along test channels that are injected into the network and are monitored at various locations.

In the home, the approach is deployment of algorithms and agents on the set-top box to measure QoE. The bad news is that we are still early in the evolution phase of these algorithms and agents that address the quality of experience. Standards are emerging that will allow those set-top boxes to report back into the core of the network, which will enable providers to use that information to effectively manage their business.

There are also technologies evolving that will make a network more resilient at the application layer — for example, retransmission of packets or enabling set-top boxes to provide feedback into the core of the network to be able to adapt to the network.

In monitoring QoE, one of the challenges providers have is drawing a distinction between lab tools and operational tools. What I have found in the industry is that a lot of providers try to force the tool that they have for engineering their solution into the operational network to find limitations.

For example, you purchased a $10,000 lab tool to measure the Quality of Experience in the lab in an environment to make sure your architecture is working properly. But, we find that using a $10,000 lab tool in a customer’s home doesn’t scale. Therefore, you have to rely on standards and your set-top box vendors to work with your operational monitoring tools, such as what Brix Networks offers, to get access to that information in a very scalable and seamless manner without the extra cost of hardware or deploying the hardware in the customer’s home.

Other advancements that are evolving are real-time dashboards that show IPTV providers how the service is performing across the network and the customer base. Basically, one of the challenges providers have today is the sheer number of paths they have for points of presence for distribution of television. You are talking thousands of these locations where television gets distributed to their customers.

The big question for them at the point of distribution is whether the TV core is good enough. A lot of providers today are doing that with human resources. They are deploying people. I will call them channel surfers who sit at a path and basically watch TV. If there is a problem, they pick up the home and call the NOC and say there is a problem with HBO that day.

There are technologies available to automate this process, and part of that automation is real-time dashboards that show across all of the network, across all of the locations, how the Quality of Experience is performing.

IPTV Update: How does video on demand play into the equation for monitoring Quality of Experience?

Kaynam Hedayat: With IPTV broadcast, the demands on the network are fairly fixed. You know how many channels you have. You know how many are HD and how many are SD and you can provision the bandwidth and — at least in terms of infrastructure capability — you can make sure you are going to be OK.

With video on demand, that’s no longer true because now your customers on demand are asking your network to deliver a certain amount of bandwidth, and that puts on more pressure in terms of quality of experience. That again raises the stakes as far as visibility. You definitely need to have visibility for Quality of Experience both in terms of the content and visibility. Was my video on demand available for the customer? How long did it take for that video on demand to get to my customer? And during that movie they were watching, were any impairments happening?

IPTV Update: Could you tell me a little more about the Brix Networks solutions to monitoring Quality of Experience?

Kaynam Hedayat: Our vision for monitoring is centered on Quality of Experience, with rich performance metrics at three layers — infrastructure, content and service — for troubleshooting and network management. When you look at an IPTV network in terms of monitoring, there are really three parts of the network that you want to have access to.

One is the headend. That is extremely important because what is sent into an IP network, if it is bad, is going to be replicated to thousands or even millions of customers, based on the multicast fabric. I mentioned earlier that the core, we believe, has 10 to 20 percent of the problem, but the fact of the matter is one of those problems can affect millions of customers. Our solution enables providers to have full visibility across all of their streams, be they broadcast or video on demand. Therefore, at the point of injection, you know that what you are feeding the network so you can guarantee yourself that the quality is as good as you expected.

Our solution also offers monitoring at the point of presence for the distribution layer. Again, if you think of the network as a cloud, we can ensure that what comes into the cloud is good and also what is going out of the cloud is good. That’s a powerful tool for segmentation of the network. You can immediately detect whether the core is affecting any quality across your service.

Lastly, our solution also offers software agents that can be deployed on set-top boxes and enable providers to proactively monitor the quality of experience across all of the customer base.

Tell Us What You Think!
IPTVU invites response from our readers. Please submit your comments to editor@broadcastengineering.com. We'll follow up with your comments in an upcoming issue.



Comments
Post New Comment
If you are already a member, or would like to receive email alerts as new comments are
made, please login or register.

Enter the code shown above:

(Note: If you cannot read the numbers in the above
image, reload the page to generate a new one.)

No Comments Found





D. Pagan Communications /  Wednesday 4:03PM
APPEAR TV DEMONSTRATES MODULAR HEADEND TECHNOLOGIES AT ABTA 2014
Wall Street Communications /  Wednesday 3:46PM
Dejero Launches LIVE+ Carrier System

 
Featured Articles
Discover TV Technology