Cable's Spectrum Preservationist
June 26, 2006
Time Warner Cable's chief technology officer Mike LaJoie has big plans for his company's hybrid fiber coax plant, but thanks to switched digital, these plans won't take a huge bite out of his bandwidth.
By M.C. Antil
Some 15 years ago, a Time Warner engineer named Louis Williamson developed an engineering concept that would eventually be known as hybrid fiber coax. And while fiber optic technology continues to creep closer to the home, there is industry consensus that, regardless of technological advances, HFC will remain a viable engineering strategy for cable operators for years.
Time Warner Cable continues to serve as a technology bellwether for operators, having been first to market with a branded high-speed data service (Road Runner) and commercially viable DVRs. Its latest development is the rollout of switched digital video, which will help make more efficient use of HFC spectrum and enable the company to keep adding digital services. We asked Time Warner Cable's chief technology officer Mike LaJoie about the benefits of switched digital, among other subjects, on the eve of the SCTE Cable-Tec Expo in Denver.
How has Time Warner Cable's faith in hybrid fiber coax been validated over the years?
Mike LaJoie: When we first rolled out HFC architecture we simply used it to offer a whole lot more television channels--though, obviously, with a lot better quality. The first real switched product we launched was high-speed data, and that was Road Runner in '95 or '96. Back then we had a few tens of thousands of customers, there were a few hundred thousand websites and we were offering speeds of 1 Mbps--which was unheard of at the time. And all that was accomplished with one 6 MHz channel.
HFC CHAMPION: Cable's plant is paying dividends through the expanded application of spectrum-conserving digital switching technology, says Mike LaJoie.
Today, we have over 5 million customers, speeds as high as 10 Mbps, and there are hundreds of millions of websites. But we still only use one 6 MHz channel.
And the reason for this is that HFC architecture was designed to be segmentable, and was designed to accommodate a switched infrastructure, so I can add more subscribers, increase the speed of the product and introduce more and more sources of product when I combine switching with HFC.
But that's with data. How does that apply to television?
LaJoie: In a television scenario, if I combine HFC with switching I can offer as much programming as I want--much like websites--to as many customers as I want. And I can offer standard-definition television at 3.5 Mbps, or I can increase the speed of the television and offer hi-def at 12, 13, 15 Mbps--all without using more spectrum.
The first switched video product we had was VOD. When we first set it up we did it to accommodate 100 or so movies. We didn't set it up to simulate an actual broadcast experience, such as allowing instant channel changes, but with switched digital video we can now do that.
Your Start Over product in Columbia, S.C., is a fully switched system that records 70-plus cable channels for playback in the head-end, and allows consumers to access archived content. What advantage does that hold for you as an operator?
LaJoie: Start Over is actually a skinnied down version of our Mystro product that was developed by Joe Collins and Jim Chiddix. And its advantage is this: When you broadcast a full channel, everyone connected to your plant is receiving that channel, even if they're actually watching it. That's wasteful, in terms of spectrum. And our research told us that of the 200 or so digital networks, only 70 or 80 might be watched in any one neighborhood. That means that there are up to 130 channels worth of bits just spilling out at the end of your system and creating a wasteful mess.
With switching technology, just like with Road Runner and VOD, you can monitor what people in a neighborhood are watching and allow multiple homes watching the same channel to share the same stream. And if a person tunes to a channel that isn't currently being transmitted to that neighborhood you can just take that channel off the fiber ring and switch it on to that particular node, a process that takes a matter of milliseconds.
What that means is that we can deliver a multichannel broadcast experience without actually broadcasting, which gives us about 60% efficiency on the spectrum. And, just like we were able to go from a few thousand websites to hundreds of millions without adding spectrum, we don't have to add spectrum if we want to go from 200 channels to 500 channels.
In essence, by combining HFC with switching capability we've given ourselves limitless capacity to offer as many channels--be they standard-def, hi-def, MPEG 2, MPEG 4, MPEG whatever--as we want.
Given the fluid nature of technology, how do you manage the development and roll out of new products?
LaJoie: We are constantly trying to look out and imagine the product profile that we're going to have to support in three to five years. And we've said we'll probably have to support up to 75 hi-def channels, we'll have to support continuing increases in speed for data, we'll have to continue to support the analog channels, because they're not going away in five years, and we'll have to continue to support standard def. And to get there, we determined that broadcast will have to diminish and switching will have to grow.
We also try to anticipate product and consumer demand without getting too far ahead of it. Because the danger in getting too far ahead of demand or technology is that you have a bunch of sunk capital that nobody can use and from which you can't make a penny. If money were no object and if we didn't have a financial responsibility to our investors, that would be one thing. But that's not the case, so we have to be financially prudent when we look at technology.
I noticed you didn't mention telephone in your product profile.
LaJoie: Telephone is just another application and, frankly, it's one that doesn't even represent 3% of the traffic on our high-speed data platform.
At the National Show this year we heard about downloadable conditional access. On one panel you called it a "sleeper." Why?
LaJoie: Basically, conditional access is a way to secure video content to ensure only those authorized to receive certain video signals have access to them. And it's a great system. In fact, our conditional access systems have never been cracked.
Now, downloadable conditional access works by taking that concept and downloading the security algorithms directly into the set-top box. This allows the box to decode the encrypted content, regardless of what system it's being used on.
What makes downloadable conditional access such a sleeper is that it overrides what we call the duopoly of conditional access systems--meaning you have to choose one or the other. Today, if you choose one system you basically have to use it on your entire network, and it's very difficult to undo that choice.
Downloadable conditional access will allow MSOs much greater flexibility by allowing them to utilize a number of encryption systems. It will also free us from having to use CableCARDs. And once the consumer electronics people adopt downloadable conditional access they won't have to include CableCARD slots in their devices. That will not only decrease costs, but increase security as well.
We're also hearing a lot about IP multimedia subsystems (IMS). What role will IMS play in the convergence of cable products?
LaJoie: The IP multimedia subsystem is a set of specifications that, frankly, are still moving around a little. The concept was born out of a consortium of telephone companies who wanted to accomplished fixed local convergence, so that a wireline carrier and a wireless carrier could converge their products. In other words, you could receive calls on both your wired or wireless phone, you could share your contact database--all sorts of things.
But the truth is IMS allows you to converge products across any series of diverse platforms, be it twisted pair, fiber, coax or whatever.
The key is IMS separates things into three layers: a physical layer, a services layer and an applications layer. And through such abstracts, it allows services to be developed that most applications will need--services like authorization, accounting, authentication, security, content management and so on. Knowing that such things will be universally available, application developers will no longer have to worry about them. They'll be there and be operable from both a management and a network perspective.
What this means is that we will be able to expand on the concept of bundling. We already know that customers like bundled products; we'll now be able to more easily bundle features of those products across platforms. We know customers like it when they see caller ID on their TV set. We know they want to be able to set their DVRs using their wireless phones or their desktops.
The long and the short of it is, when you have a subsystem that is both network aware and applications aware, it makes it a lot easier to build these apps because you don't have to build all the support services into each one. You simply use the services already being used by your current apps.
We have a lab trial going on right now, and it shows a lot of promise. It's still a little nascent, but when you look at all the companies that have ascribed to IMS technology and have begun building products for it, and all the companies that have stated they want to start utilizing it, I really think it has a good chance of working itself out and finding its way into our production numbers.
Time Warner Cable
By the Numbers
Homes passed: 19.7 millionBasic customers: 11 million
Digital customers: 5.6 million
Residential HSD customers: 5.2 million
Commercial HSD customers: 216,000
Digital phone customers: 1.4 million
Source: Time Warner Cablehttp://www.cableworld.com/cgi/cw/sho...themsotime.htm