Digital Services Drive Network Evolution

By Mitch Shapiro and Don Gall

In January the Society of Cable Telecommunications Engineers (www.scte.org) held its 2001 Emerging Technologies (ET) conference in San Jose, California. The charter of this annual SCTE conference, which typically draws over 1,000 attendees, is to look ahead to technical and business trends just over the horizon.

Besides providing a good look at the current thinking in technology, the ET conference is a great place to catch up with old friends and the movers and shakers in the cable industry's technical community.

The central question addressed at this year's conference was how best to support the flood of digital services that are just now starting to have an impact on HFC networks. Digital services are not new to the industry; in fact their roots go back over thirty years. What has changed is that they are moving from being small time incremental businesses that in many cases cost more to deploy than the cash flow they generated, to major contributors to the bottom line. Reflecting this change was the fact that there was not a single paper presented at the conference focused on analog services.

In addition to generating an impressive alphabet soup of acronyms (a requirement of any engineering conference), the ET sessions highlighted the fact that the cable industry is wholeheartedly embracing a digital future with increased emphasis on interactive content such as multimedia (video, voice, music, etc) on demand and high-speed data services.

This is good news for the fiber optic industry for at least two reasons. First, everything eventually has to achieve equilibrium. Cable operators will have to balance the added complexity of the digital services with the need to provide better reliability so they can compete with the satellite, wireless, and enhanced telephone-centric providers vying for parts of the same pie. This feat can be accomplished by either adding highly trained staff and automated telemetry equipment to the entire network or by deploying fiber optics deeper into their networks. The additional fiber optics will lower operational costs in the middle of the network and allows resources to be shifted to the network ends. Operational costs for the new services will then be in line with the added complexity.

This shift of resources, while inevitable, will not happen overnight. Several of the larger operators, like AOL/Time Warner and Cox, for example, are very close to completing large and expensive HFC network upgrade programs. As a result, they will be reluctant to spend more on infrastructure until they establish the incremental revenue-producing businesses these upgrades have enabled and competition or new applications push them into the next upgrade cycle.

At the same time, there are other operators that got a late start in their upgrade programs or have recently acquired lots of cable systems that need to be upgraded. These companies will continue to spend heavily over the next several years to add fiber and electronics to enable their networks to deliver today's new digital services. As a whole, the industry can be expected to spend several billion additional dollars to bring their networks up to the new status quo.

Looking beyond this investment to deliver today's digital services we find a second piece of good news for the fiber optics industry. As new lines of interactive digital business begin to develop, they will generate new content. And if the cable industry has had any constant in its history, it's that "new content fuels consumer demand which in turn drives network infrastructure."

As interactive content evolves, so will competition to deliver it to the consumer. Today's HFC networks are configured to deliver mostly broadcast-style services with some incremental data and interactive multimedia. With the help of new technologies from innovative companies and some creative thinking, HFC networks will be able to deliver tens of megabits per second of unique content per customer.

As discussed in earlier columns, one such enabling technology has been developed by Advent Networks (www.adventnetworks.com) of Austin, Texas (we remind readers that Pangrac & Associates has a close business relationship with Advent Networks). Advent's system can provision dedicated downstream bandwidth in increments of five Mbps up to 40 Mbps or the maximum RF channel capacity on request. The upstream bandwidth can be symmetrical for networks configured for the bandwidth, but Advent's standard product delivers up to 8 Mbps in 500 KBs increments that work well in standard cable-industry HFC networks.

Another technically feasible (but, in our view, less practical) approach is based on a concept from AT&T Labs-Research, which would couple several RF frequency assignments together in an HFC network to deliver 100 Mbs Ethernet services. Readers can learn more about this approach in a paper entitled "Fast Channel: A Higher-Speed Cable Data Service" by Sheryl Woodward of AT&T Labs-Research.

We believe that even these data rates will not be enough in the 5-10 year timeframe if some of the predictions about content come to pass. Several industry sources have suggested that future needs for subscriber-based bandwidth will approach 155 MBs.

At bandwidth levels in this range the case for FTTH networks begins to make more sense. We get asked then why not just build FTTH networks today? The answer is that: 1) FTTH networks are still relatively expensive to build; 2) existing networks not only handle existing content, but can also accommodate near-term content growth and; 3) there isn't a network medium out there--including FTTH--that will last forever. In many urban environments you have to constantly spend operational dollars reconfiguring and repairing "THE" network--no matter what type it is.

Today's risk-averse network operators are likely to respond to "the FTTH question" with a question of their own: why spend a significant premium on a network you may need to upgrade before you get to use its benefits? Given the uncertainty and risk related to future service and technology developments--and today's conservative capital markets-- the management of today's network operators would have a tough time justifying the added investment required for a jump to FTTH.

Bottom line, what we saw and heard at the SCTE Emerging Technologies conference was good news for the optical equipment industry. At the same time, it was also very grounded in the realities of business, where typically its unwise to "put the cart before the horse." One lesson of the dot-com debacle is that you can build a business on a dream, but unless you have something to sell--and at a price the market will bear--don't expect to have any customers. In our market economy, the optimum driver of network evolution is expanding demand for content and services, and healthy competition aimed at profitably satisfying it.