IPTV is one of the most talked about topics in our industry. It is changing workflows, lowering costs and providing new opportunities. Yet the only difference between traditional television and IPTV is the underlying delivery mechanism. So why is IPTV considered such a powerful technology?
It is important to realize that a lot more than the delivery mechanism is changing. (See “Why IPTV thrives.”)
Defining IPTV
Before reading any further, you should be aware that the IPTV terminology is evolving. Some people use the term IPTV to refer to video over IP. Others use IPTV to refer to delivery of content to the home, while reserving video over IP for professional applications.
While both of these terms correctly refer to IPTV, using the abbreviation VoIP to mean video over IP should be avoided at all costs. VoIP is used in the telecom industry to indicate voice over IP.
Is IPTV all hype?
Some people in the industry believe that IPTV is a lot of hype and that it will never be successfully used for broadcast-quality video. I think they are wrong.
That said, if you try to send IPTV over a typical corporate network or over the Internet, you will fail. The cause of the failure will depend on what you are trying to do.
For example, if you are trying to FTP a large video file, the transfer will fail because FTP does not support the transfer of large files (files larger than 2GB), or because the traffic throttling mechanisms in FTP all but guarantee that a large transfer will fail in most situations. (See Figure 1.)
FTP exchanges files over any network that supports Transmission Control Protocol (TCP) over IP. TCP over IP frequently suffers from congestion collapse. If a link is congested, TCP will cut transmission rates by 50 percent until it finds a speed at which the transmission of packets is successful. In most algorithms, the time for TCP to reach full speed again can be much greater than the time it takes to throttle back. This is one of several reasons why TCP transfers may fail.
If you are trying to stream a live, high-bandwidth feed over the Internet, the characteristics of the network will likely result in numerous dropouts and perhaps in a complete failure of the link. But a video engineer experienced in IPTV knows not to use regular implementations of FTP for large video file transfers and that error correction will be required to transmit IPTV over the Internet.
Human error
Broadcasters have been using IP-based video transport for several years. Most of these broadcasters use specialized private networks, in many cases employing their own dark fiber-optic cables. IPTV has been used extensively to broadcast Olympic, football and baseball games.
Once the initial configuration is complete, IP-based contribution networks can operate for many days or weeks without a single error. The greatest source of errors in these networks comes from humans. People take equipment down for maintenance, expecting that end users will not notice the disruption as the network re-routes packets to their final destination. This may work for most IP-based traffic, but it definitely does not work for IPTV.
IPTV in the studio
IPTV has been deployed in the studio for some time now. Since the 1980s, graphics departments have been exchanging images over computer networks. Now, broadcasters include high-performance networks as a critical part of their broadcast core. File transfer of content between devices is common, and streaming real-time broadcast-quality video and audio over IP infrastructures is also frequently used.
One reason for broadcasters' success at employing IPTV in the studio is the broadcaster controls the entire network. This allows the network designer to ensure the network meets the demands of the on-air environment. Proper network design can make the difference between reliable IPTV transport and an unreliable network, which results in unhappy end users.
Core backbone capacity
It's also important to have adequate capacity when building a studio network. Broadcasters are used to working with nonblocking switches. A nonblocking switch never runs out of capacity. Analog and digital video routers can be used 100 percent and never miss a beat. This is because conventional cross-point routers switch inputs to outputs without sending the video over a common buss.
With IP switches and routers, however, data travels across a common backplane. This means that all port traffic is aggregated onto the backplane. For example, a five-port 100BASE-T switch would need a backplane capacity of 500Mb/s to be nonblocking.
Switch and router manufacturers may decide to employ lower bandwidth backplanes, particularly in large, inexpensive switches, as a way to reduce cost. The designers know most applications will not constantly require 100 percent capacity all of the time. However, IPTV is a demanding application.
When transferring a file or sending a streaming IPTV feed, there are few if any breaks in data transmission. If you are using a blocking switch or router, you may quickly run out of backplane bandwidth if IPTV applications are used on a large number of the devices ports. That's why it is imperative to know the backplane capacity of the devices in your core network. The way the network device handles queuing can also significantly affect IPTV.
It is essential to have adequate capacity on links between IPTV devices and network switches and routers. This is also required between core network devices. One of the best ways to achieve this is to aggregate bandwidth across multiple network interfaces, often called teaming. Teaming allows multiple Ethernet interfaces to act as a single, high-capacity interface. Of course, core network devices will need to be sized appropriately. In IPTV applications, this may mean spending extra money to purchase nonblocking devices.
Delivery to home
Some of the most exciting developments in IPTV are occurring in delivery to the home. Research and development investment has spurred major technology advancements in the areas of forward error correction (FEC), advanced network protocols, compression, and the commingling of IPTV, voice and data services all on one IP core network. Many of these developments will find their way into the professional broadcast space.
Much of the work on FEC and compression will directly affect broadcasters. For example, new FEC algorithms may allow broadcasters to achieve the required quality over what are now considered to be unreliable IP links. Compression technologies under development continue to push HD rates down to the sub 8Mb/s range, and it is likely that those rates will be cut in half in the near future.
The commingling of IPTV, voice and data in these networks is also good news for broadcasters. The technologies that allow different services to coexist on a single network means that there will be hardened commercial solutions, which will allow broadcasters to transport IPTV, voice and data on a common infrastructure between facilities.
Brad Gilmer is president of Gilmer & Associates, executive director of the Video Services Forum and executive director of the AAF Association.
Send questions and comments to:brad.gilmer@penton.com
Why IPTV thrives
- IP connectivity is available at a price that makes delivery of video economically viable.
- Switches, routers and other networking hardware have advanced to the point where it is possible to send multiple IPTV feeds through this equipment.
- Millions of consumer applications are driving the deployment of advanced networking protocols, which allow Internet backbone providers to manage the commingling of IPTV with data and voice traffic without unfairly impacting any of these services.
- Broadcasters have recognized the benefit of using self-routing packetized technology in their facilities.
