|
Current Issue: August 2007
Feature
U.S. telcos bet on video quality
Telcos are taking on cable rivals in the lucrative U.S. TV market. Success may hinge on the quality of their service
by Jim Barthold
As U.S. telcos move into TV markets, they must ensure the quality of their video service equals what is already being offered by entrenched cable competitors. Anything less and they risk failure.
The explosive growth of affordable and advanced home electronics makes this a tough challenge. High definition (HD) TVs are becoming commonplace. Many households have more than one, forcing providers to up their bandwidth capabilities.
Besides bearing responsibility for the quality of the proprietary network delivering bits to the side of the house, telcos now own the "second network" within the premises.
"Our demark has gone literally to the customer’s couch," says Carl Murray, strategic technologies director of SureWest Communications, a CLEC based in northern California. "How do we provide a quality level of service to that point?"
In the loop
In fact, SureWest does not have to worry much about last-mile loop issues because it runs fiber directly to the customer premises.
While other telcos have followed suit, some elect to save a little capex by running fiber from the CO through the network core to a demarcation point in the node or even at the curb. That creates the need for additional quality assurance in the connection from the DSLAM to the home.
The costs of these different approaches vary wildly. Late last year Verizon said the cost per home for its FiOS fiber-to-the-home (FTTH) service was less than US$1,000, and would fall to US$650 to US$700 by 2010. By contrast, AT&T’s U-Verse rollout, based on fiber-to-the-node (FTTN) technology, costs well below US$500 per home. Truck rolls to install each service generally cost between US$75 and US$100, depending on how much time the installer spends with the home network connection.
Vernon Reed, principal member of the technical staff at AT&T Labs, says AT&T adopted FTTN as part of a capex/opex tradeoff–copper in the last mile costs more to maintain–but AT&T does use FTTH in some greenfield deployments.
The difficulty with FTTN is ensuring the signal quality does not degrade when it leaves the fiber network and jumps on to copper, says Jeff Schmitz, vice president of the service assurance division at Spirent.
"There’s a fair bit of complication in the [DSLAM] cabinet," Schmitz says. "It’s much more expensive to deal with problems that might happen there."
For Verizon–which, like SureWest, has removed the last-mile loop by going directly to the customer premises–most of the opex accrues in the home.
"If the last 100 feet aren’t as robust as the core and edge networks, then our service is going to founder in the home," says Tushar Saxena, director of customer premises architecture at Verizon.
At home
In investing in the home network, Verizon has developed what Saxena calls a "low-end enterprise-class router. " This uses existing coaxial cable in the home to prioritize video over best-effort traffic.
"The business case is built around avoiding truck rolls," Saxena says. "Today we offer it for free, but as FiOS stabilizes we will probably expand it and be able to offer this as a [paid] service to our subscribers."
AT&T has made investments to improve home networking quality. "We have custom software on the set-top box and custom firmware on the residential gateway," says David Rackley, director of consumer product analysis at AT&T Labs. "Unless you have a custom solution, you can’t guarantee quality of service … you can’t deploy to millions."
Like Verizon’s solution, AT&T’s allows the operator to control the flow of traffic, prioritizing voice and video over best-effort traffic.
SureWest, meanwhile, has trained its home-network specialists to take extra pains when installing equipment. "All it takes is one bad coax connector, a staple going through the wire, and the user experience is going to suffer," Murray says. "It might not be noticed by a voice customer, but with video every little thing will show up, especially with HD."
Indeed, HD could represent a boom-or-bust scenario for telcos whose opportunity hinges on cable’s ongoing struggle with its own quality issues. "There’s nothing like HD to test your network," Murray says. "You find out where your problems are right away because it shows up. We’ve started to do our own homegrown [testing] where we monitor any device from a CPE or gateway or set-top box."
That is not always easy, because developments in monitoring have not kept pace with growth in home electronics gear. Peter Schweiger, business development manager for Agilent Technologies, which develops testing solutions, says cost is another barrier. "Everybody is trying to embed testing in their end devices: the set-top boxes, the routers and even client applications, but they all want that for US$10 to US$15," he says.
With millions of devices in circulation, anything more could cripple the business case.
Coax, copper, power lines or Ethernet
There are other ways to monitor the quality of a home network. The DSL Forum’s TR69 network monitoring tool is one. Others are being developed by vendors and providers of home-networking gear. None is really fully baked (see TR69 extends DSL Forum’s reach).
In the meantime, the infrastructure itself is critical. Most North American homes come pre-installed with coax and copper. Elsewhere in the world clean power lines run through homes. In a pinch, or in a greenfield environment, carriers can install category 5 Ethernet cable.
"The carriers need to choose a technology that will minimize opex inside the home," says Andrew Morton, general manager of Comtrend, a company that designs and builds DSL routers and modems for IPTV. "The capex of putting cat 5 cable inside the home is high compared with the cost of giving the customer a couple of adapters and leveraging home wiring."
Verizon, after studying all the options, chose coax, while SureWest selected copper. AT&T’s approach is to mimic the cable companies by selecting a technology called home phoneline network alliance (HPNA).
"We can adapt anything that runs analog cable TV reliably," Rackley says. "In almost all cases, modifications have to be made to ensure the integrity of the connectors, junction points in the network and splitter devices. We have to make sure there isn’t a point of failure that would inhibit the services we deliver."
That may be so, but cable operators can claim a lead when it comes to testing and maintaining home networks–simply because they’ve been doing it for so much longer. For telcos, the learning curve has slowed down some video deployments.
Embarq, a spin-off from U.S. operator Sprint, is trying to address the issue. "During our video trials we are developing a strategy for dealing with the issue of home wiring," says Steve Carter, vice president of network planning at Embarq. "You need the right balance between service quality and cost, because otherwise you can spend an awful lot on wiring somebody’s home. That’s our focus: looking at service quality, and assessing the most cost-effective solutions."
Such concerns might rule out the solution favored by Eric Presworsky, vice president at home-networking vendor Zhone. "The best way for a telco to succeed is run point-to-point Ethernet to every different port in the house," he says.
Telcos are not wanton spenders, no matter how much they crave new subscribers, and the Ethernet path likely will entail high levels of investment.
What might give the telcos some advantage is the poor track record of cable companies, whose past is pockmarked with examples of arrogant behavior, rate hikes and poor customer service.
Unfortunately for telcos, cable has got its video act together. Only a superior telco offering will wrest away users.
|
