Sonet equipment, long the darling of the telco industry, is now being embraced by cable operators for the transport of new, high-speed data services, as well as for a support platform for their forays into voice services. But while the Sonet (synchronous optical network) platform has long been optimized for voice and data, the question remains — will Sonet soon see its day in the sun as a video transport platform for the cable industry?

The answers to that question are as varied as the respondents, and many cable operators still seem to be gathering enough data to answer the many, unresolved questions they have about the technology's suitability for video transport.

Cox Communications will soon have Sonet transmission gear "sitting in every single hub site" it owns for the transport of its ATM data traffic and its competitive access and residential telephony traffic, according to Mark Davis, director of engineering, telephony technology; however, the company still has one major concern about the technology's suitability for video transport: cost.

"We are still struggling with the cost of Sonet for pure video transport, and there is a good reason for that," says Davis. "Sonet is a baseband video transport, when you use the codecs that are available today. The fact that you have to remodulate that signal at every single hub site back to an AM signal, that is where the costs are. If you just looked at it on a transport-only basis, then the time has come. But when you factor in the additional signal processing, the modulation and the scrambling, things that you must do at every hub site with baseband video, it is still more expensive than doing a proprietary solution, or even analog 1550 fiber optics."

In fact, Cox recently announced that it has chosen ADC Telecommunications to supply digital video transport equipment (gear that is not Sonet-based) for a deployment in some of its large clusters in the U.S. The MSO will deploy the company's DV6000 equipment as part of a contract award estimated to be worth more than $10 million over the course of three years.

In its Phoenix system, though, Cox does own a five-hub, dual OC-48 backbone that is dedicated to video transport, courtesy of Times Mirror, which had installed the equipment before the system was purchased by Cox. And, the MSO has been taking a closer look at Sonet for video, says Davis, in light of benefits such as redundancy and high reliability, as well as flexibility.

"It also gives you other nice things, such as the ability to cross-connect channels on the fly, and you can do sub-tending rings off of fiber rings, which can help in some of your hub configurations."

But Davis also remains optimistic about progress in pushing the cost of Sonet farther down, noting that the technology is "a lot closer than it has ever been, price-wise" to making good sense for video.

Shaking hands with the outside world

InterMedia Partners Inc.'s Chief Technical Officer, Ken Wright, agrees that the platform is looking better and better for video, but is still uncertain whether the train has actually arrived at the station. Right now, the MSO is wrestling with the question of which technology to use for a digital interconnect it plans in Tennessee, which will link up about a quarter of a million subscribers in Nashville and six surrounding counties. InterMedia is in the process of rebuilding and interconnecting the systems, which have stood alone for years. To that end, the operator has put an RFP (request for proposal) out on the street for using a digital solution to interconnect the systems (replies were due just after press time). Until the replies are in and analyzed, the operator is in a wait-and-see mode about whether to use Sonet for the interconnect, or some other digital solution.

Why would InterMedia want to use Sonet for the Tennessee interconnect? "The reason is to have interconnectivity to the outside world," explains Wright. "As long as all we are using it for is transporting our own traffic internally, it's not that big of an issue whether it's proprietary or Sonet-based. If we ever wanted to connect to an adjacent operator, as long as we had the same pick on a proprietary vendor, it would be OK. (But) once we want to connect to a long distance carrier or a local telephone company that we want to hand off voice and data traffic to . . . then it certainly makes life a lot easier if we are using a standards-based approach."

On the cost side, Wright says that vendors are telling him that Sonet is now competitive in price with other digital solutions. "There is a little bit behind that statement. I think that they are saying (the technologies) are competitive on an apples-to-apples approach. We are hearing that once you factor in the number of fibers and the route miles, they are cost-competitive on an end-to-end basis, rather than just comparing (Sonet) equipment to (non-Sonet) equipment."

Other than cost, another issue that requires close examination is the ability of Sonet-based approaches to transport different types of signals, says Wright. Specifically, how well, and by what means, will the platform be able to transport scrambled video or BTSC-encoded signals?

"At least one vendor says it will have (a solution to the scrambling issue) in six months," adds Wright.

Regardless of the answers it receives from its RFP, by the end of this year, InterMedia will have its digital interconnect in place in Tennessee.

Vendors respond to system trends

Equipment suppliers such as NextLevel Systems (formerly known as General Instrument) say that part of the rationale behind their entry into the Sonet video market is a response to architectural trends within the cable industry. In one example, it's becoming more and more common for cable systems to cluster their systems, and consolidate headends.

"Quite frankly, that's why we got into this in the first place," says Don Raskin, principal engineer, Transmission Networks Systems Business Unit, Broadband Networks Group, NextLevel Systems. "We found an increasing number of our customers were coming to us with a need to regionalize. And a digital transport scheme was really the only way to accomplish it. Then, as we got further into it, it became clear that you could do a lot of other things, including integration of your services."

In other words, as cable operators move from being solely video providers into the realm of communications companies offering not just video but voice and data services as well, they might gain some efficiency of operations by using one network to transport multiple services, efficiency which would be realized, for example, in the need for one network management system instead of two.

But what about cable engineers' concerns about the cost of Sonet for video?

"The number-one aspect that makes Sonet appropriate now, whereas it wasn't in the past, is that it is less expensive than it used to be," says Raskin. To put that in perspective, executives at NextLevel say that the cost of Sonet-based video delivery vs. a non-compressed, proprietary digital scheme used to be two times, and even three times greater, while with recent developments, Sonet is "essentially at cost parity with other digital transport methods," according to the executives.

NextLevel offers a platform which carries compressed digital video signals from the headend out onto the cable system. Sonet, because it was born in the telco industry, utilizes standard, digital interfaces; video is compressed so that it is compatible with those standard interfaces. The drawbacks of Sonet for video transport, says Raskin, relate to those very interfaces.

"There is a tradeoff between very high fiber efficiency and the amount of equipment that an operator would need at the remote sites, the hubs," he explains. "With a proprietary digital transport system, you can carry an IF output from the headend and then just upconvert that onto the cable system. In the system that we have been discussing, you need to have baseband transport, and if the signal needs to be scrambled, you'd have to put a scrambler in the remote site."

Manufacturers are making strides in developing interfaces that are friendly to video. "Basically, we now know how to transport all of the signals, and by the end of this year, we'll have interfaces to carry everything that a cable operator needs to carry," Raskin emphasizes.

But having to put in additional equipment at remote hubs is not all bad, either, notes Raskin. There are some advantages to having the equipment located at remote sites, such as the ability to do local ad insertion and change program lineups locally, too, he says.

Another plus for a Sonet-based video system, says Raskin, is that operators can establish backup headends in a relatively straightforward manner. "If you had a real disaster and a headend went out, or if you just lost one bank of channels, it's very natural to have a backup at some other location that's on the same Sonet ring," he notes.

But even more important, Sonet is a standards-based transport system. "You can go to multiple vendors, and their equipment talks to each other. And you can even marry some data equipment that you have from one Sonet vendor with video transport from another Sonet vendor," adds Raskin.

Scientific-Atlanta has also entered the Sonet-video arena, with the announcement of its Prisma Digital Transport platform. In contrast to NextLevel, the company will offer a system that utilizes uncompressed digital transport "in a Sonet-compliant format," according to information released by S-A.

"In addition to the traditional Sonet low-speed interfaces, we have a number of low-speed plug-ins that are optimized to interface video — just like the proprietary systems," says Tim Wilk, director of strategic planning, Terrestrial Network Systems, Scientific-Atlanta Inc. "The big difference is, we are mapping (those interfaces) directly into Sonet. We've taken the concept of video encoders and decoders from the linear digital systems, and have merged them with the key aspects of Sonet transport technology."

For a variety of reasons, S-A is very bullish on Sonet's prospects for the delivery of multiple services, including video. Key among those is the competitive edge it can provide to cable operators concerned about system reliability.

"Sonet is introducing a high degree of intelligence into the interconnect network," says Wilk. "You can really provide superior performance monitoring and a self-healing capability, if there is a fiber cut, or an electronic failure."

"Sonet is also able to direct traffic in a cost-effective way," says Wilk. "Sonet can do drop/add/pass very cost-effectively. It doesn't have to demultiplex its whole digital stream to try to find one channel."

Not surprisingly, there are varying cost estimates for Sonet vs. other digital solutions coming out of the vendor community. For example, executives with ADC Broadband Communications say that because a Sonet-based, compressed video system requires more equipment than a non-Sonet, uncompressed digital video system, a user would most likely incur an additional cost of 30 to 40 percent.

"Sonet is really designed for a lot of people talking to each other," says Wes Simpson, director of ADC Broadband's Digital Video Business Unit. "A true video network is designed to take mass quantities of video and give them to a site economically."

Universal uncompressed digital video networks, say ADC executives, enable the cable system to carry scrambled signals and vertical blanking interval information (VBI) intact, which translates into a cost savings because the processing equipment is centralized, instead of distributed at each hub site. In addition, because the video is uncompressed, its quality will be higher, they say.

But the company is by no means anti-Sonet. Bob Harris, marketing manager, Digital Video Business Unit, notes that the company has developed a variety of standard voice and data interfaces to hook its system up to a Sonet platform. And Simpson continues, "For voice and data, Sonet makes a lot of sense, because those are almost always two-way services."

But the very attributes that make Sonet so well-suited for voice and data delivery make it unsuited for video delivery, says Harris. "A true Sonet network, with its full duplex bidirectional capabilities, is actually wasting a great deal of bandwidth on the return path coming back, because most of your video is going down uncompressed or slightly compressed off of hub sites. That's an economic (factor)."

When to make the move?

The cable industry is well-known for making the most out of its available resources, whether that means personnel or equipment. To that end, operators have continued to push the envelope of what analog technology can do, rather than make the move to digital fiber at all.

"Some operators, if they need to go a greater distance, will go to 1550 fiber," notes Mediacom Senior VP and COO Joe Van Loan, who adds that, "for the most part, we get where we need to be with 1310 nm analog fiber."

The Sonet debate will undoubtedly continue for some time, as operators and manufacturers evaluate the synergies between various digital fiber platforms and cable operators' forays into new, advanced service offerings. But what was once seen only as "that telco technology" is looking brighter for video.

1. "Cost Analysis of Self-Healing Video Transport Network Capabilities," By John Thoma, ADC Broadband Communications.
2. "Digital Transport," a white paper published by Scientific-Atlanta, ©1997.
3. "Regional Networks for Broadband Cable Television Operations," By Donald Raskin and Curtiss Smith, NextLevel Systems — Broadband Networks Group. A paper submitted to NFOEC '97.
4. "DV6000 Key Applications," a presentation produced by ADC Broadband Communications.

Canadians bullish on Sonet

As early adopters of fiber optic technology, Canadian cable operators are now looking to Sonet-based fiber networks as a critical component to their future. That's because the industry's plans for delivering digitally compressed video, Internet access and telephony all rely on the technology for carriage.

This is why "the big five or eight cable companies in the country are all pursuing some measure of fiber construction," says Brian Nelson, vice president of Fundy Telecom, a division of cable MSO Fundy Communications. "Certainly Videotron, Shaw and Rogers are pursuing fiber builds very aggressively."

Of course, the actual amount of fiber installed varies by company. Fundy Communications has put in about 1,875 linear miles of fiber across Nova Scotia and New Brunswick and recently completed a high density OC-48 Sonet ring supplied by Alcatel. The company also uses Lucent OC-48 and OC-12 fiber installations both to cover its territory, and to tie into conventional telcos such as AT&T Canada and Sprint.

The fiber has two purposes, says Nelson. First, it allows the company to compete with the established local telco in providing business telephony. That's a market worth about U.S.$150 million, he estimates. Second, the network is also used to supply digitally compressed video from various TV studios or third parties "and feed it to all our headend distribution."

Meanwhile, Rogers Communications Inc., the parent company of Rogers Cablesystems, has essentially wired together all of its systems with fiber. What this means is that "with the exception of one or two smaller systems, everything we have is accessible from either our Toronto or Vancouver main headend," says Roger Keay, RCI's vice president of technology and strategic planning.

So well-fibered is its network, in fact, that the company was able to survive the failure of a Canadian satellite a few years ago. While other cable companies were running blank screens, Rogers was still offering live video.

In Quebec, cable MSO Videotron is busy upgrading all of its networks with fiber. "We're bringing them up to 750 MHz," says Director of Public Affairs Sylvia Moran.

Currently, the company relies on 1,500 miles of fiber to deliver cable TV, two-way Internet access and business telephone services. "We put in basically what the client wants," says Moran. "If you want T-1, we just bring fiber directly into the building. Then we connect it to our extensive fiber optic network, which already covers Montreal, Quebec City and other urban areas in the province."

However, Videotron has other plans for its fiber — namely, local telephony. A current PCS trial with provider Microcell Telecommunications has Videotron feeding Microcell's PCS signals down its network to pole-mounted transmitters. These then provide the final link to telephone subscribers via wireless. The stakes here are high: if practical, this PCS strategy could give the company an economical way to break into local residential telephony.

Finally, Cogeco Cable Inc. is also moving ahead with hybrid fiber/coax architecture, installing Sonet fiber rings in southern Ontario. "The purpose is to be able to provide competitive data services, and eventually, telephone services to local business," says Denis Belanger, Cogeco's vice president of engineering and development. "We also want to prepare to go digital on our systems, in order to provide more programming choices," as well as compete with DBS and others in providing DTV in the future.