The return of fiber-deep residential architectures

The return of `fiber-deep` residential architectures

Donald T. Gall and Mitch Shapiro

Favorable prices and improved performance foster renewed interest in this network paradigm.

After more than a decade of hearing about fiber-to-the-home, then fiber-to-the-curb, followed by HFC, MMDS, DBS, Tele-TV, Americast, etc., it`s easy to be skeptical about the notion of aggressive telephone company deployments of video services and "fiber-deep" residential networks.

This having been said, even we skeptics have to admit that progress has continued in the development of reliable and cost-effective fiber-deep broadband-capable telephone company architectures. And while entertainment video has not had staying power as a primary economic driver for these networks, fiber-deep economics have continued to improve as equipment prices have fallen and performance has improved relative to copper loops for new-growth "greenfield" construction and high-density urban rehabilitation projects.

Add to this the emergence of high-speed data as a rapid-growth, highly competitive residential and small business market, and the "fiber-deep" business model begins to look more attractive. If to this carrot we add the stick of expanding "cable-phone" competition from operators like Cox and MediaOne and the prospect of an even broader-based IP-voice-via-HFC initiative led by AT&T/TCI, it`s no surprise that some large telephone companies are beginning to get serious again about fiber-deep architectures that can deliver voice, video, and high-speed data.

Going deep

The most notable examples of this resurgent interest in fiber-deep deployments are the decisions by US West and GTE to deploy a "fiber-to-the-node" version of Next Level Communication`s (NLC) N3 Unified Access System and Residential Gateway. The platform includes a broadband digital terminal linked by fiber to universal services access multiplexers (USAMs) deployed within 4000 ft of customer premises and capable of delivering 26 Mbits/sec to each living unit on twisted pairs using very high-speed digital subscriber line (VDSL) technology.

Less aggressive in terms of fiber deployment than NLC`s fiber-to-the-curb system--which US West has deployed in a Scottsdale, AZ, private development--this fiber-to-the-node (FTTN) platform is designed to deliver broadband network capabilities from fiber-fed USAMs located at feeder distribution interface/serving area interface locations.

US West got the telephone company FTTN ball rolling in April, when it announced plans to offer its VDSL-based TeleChoice service to a 400,000-line service area in Phoenix this year, beginning with a limited trial in Gilbert, AZ. The company also declared that the Phoenix launch would be followed by efforts in other markets in 1999 and beyond. Its multi-year contract with NLC calls for an initial 450,000 lines delivered to residential gateways.

GTE officially joined the FTTN camp late this summer, with the launch of an initial market trial in the Glenwood Acres development in Clearwater, FL (where it also operates an HFC network) and a multi-year purchase agreement with NLC for FTTN deployments in key GTE markets following the trial.

Looking up

Though US West found the HFC overlay and proprietary set-top it deployed in Omaha several years ago to be too expensive, the company apparently believes it can make a full-service FTTN work in markets like Phoenix.

One reason for this optimism is continued improvements in the price and performance of key technologies such as VDSL, digital video, and opto-electronics. And with relatively new plant in high-growth markets like Phoenix, US West is fairly well positioned for cost-effective FTTN deployments in terms of installed fiber plant beyond the central office and "reusable" copper distribution plant. According to a NextLevel official, tests by US West and Bellcore suggest that more than 90% of the installed base of copper distribution cables in the Phoenix and Denver markets will be able to deliver data rates of 23 to 26 Mbits/sec using new low-cost, low-power VDSL chip technology developed by Broadcom. Assuming average data rates of 6 Mbits/sec, this would allow US West to deliver video signals to three televisions per home, with plenty of capacity still available for high-speed data services. Relatively high data rates such as these are considered necessary to deliver a viable video service via copper.

Continued cost reductions in digital video components such as MPEG decoders have also helped make FTTN economics more attractive. For example, NextLevel`s Residential Gateway device will include three MPEG decoders to support up to three television sets per home. Today these decoder chips are likely to cost only about $35. By sharing other components, including power supplies, memory, and processors, the three-TV gateway device is expected to cost substantially less than the combined cost of three individual digital set-tops.

While US West`s FTTN-based TeleChoice product can be viewed as a defensive measure against Cox`s high-speed data and voice initiatives, it is also an attempt to leapfrog the typical cable operator`s service offering by using a more integrated platform. For example, US West plans to offer "on-screen" integration of popular telephone features such as caller ID and voice messaging with its television service. This capability is supported by the NLC platform`s ability to read telephone signals at a multiservice network interface device on the side of a home and direct them to the television set.

The jury is still out in our view as to whether the Phoenix project will mark the first wave of near-term mass deployments by US West, GTE, and perhaps other telephone companies. In fact, it is not yet a given that the Phoenix deployment will move forward as planned.

Among the hurdles still to be overcome is receipt of the necessary franchises from the 24 municipal entities found within the greater Phoenix area. Sources familiar with the franchise situation cite two major issues that may prove difficult to resolve. One is US West`s request to serve only selected areas within these 24 communities. This practice, commonly referred to as "cherry-picking," is frowned upon by many franchise authorities and usually challenged by incumbent operators who generally are required to offer service to the entire community.

Another issue is US West`s request for a franchise provision that would allow it to withdraw from the project if the results prove disappointing. This "escape clause," which also appears unpopular with some Phoenix-area franchise authorities, has led some to suggest that US West may be less committed to TeleChoice than its public statements suggest. The process of franchise negotiation in the months ahead bears watching as an indicator of US West`s commitment to its TeleChoice strategy and whether large-scale deployments of broadband FTTN systems are right around the corner.

While the decision by US West and GTE to deploy NLC`s broadband FTTN platform has dominated the "fiber-deep" headlines in recent months, several RBOCs have continued to quietly ramp up the pace of their FTTC deployments. BellSouth, for example, has been deploying Reltec FTTC systems in virtually all of its new-growth areas with buried plant since the fall of 1995. It currently is serving more than 200,000 homes with this plant and is adding new FTTC homes at a rate of roughly 7000 per month. While most of these are POTS-only lines, BellSouth is equipping some areas with video capability, typically in markets like Atlanta and New Orleans, where the RBOC has deployed a wireless MMDS system. DSC has also been approved by BellSouth as a second FTTC vendor and Sprint has been similarly aggressive in deploying the Reltec FTTC platform in new housing developments.

Up north, in sections of Boston, New York City, and northern New Jersey, Bell Atlantic has also been deploying narrowband FTTC platforms supplied by NLC (Boston and New York) and DSC (New Jersey). The focus here has been mainly on rehabilitation in urban areas with old, high-maintenance plant. Following a first office application in 1997, the NLC deployment is expected to ramp up from 32,000 lines in 1998 to roughly 100,000 lines in 1999. This deployment ramp is likely to continue, given the dramatic improvements in trouble call rates Bell Atlantic has experienced thanks to FTTC (e.g., from more than four monthly calls per 100 lines to less than one). Bell Atlantic`s fiber-deep focus is, however, likely to remain on narrowband rehabilitation in the near-term, with its plate quite full with other investment priorities.

In our next column, we will take a closer look at the design and economics of the various fiber-deep platforms being deployed by telephone companies and developed by their vendors. q

Donald T. Gall has been involved with the cable-TV industry for the last 28 years. He was an integral part of the team at Time Warner that developed the first practical applications of hfc analog fiber and networks. He is currently a consultant with Pangrac & Associates (Port Aransas, TX) and can be reached at [email protected].

Mitch Shapiro is an independent consultant (and cable modem user) specializing in research and analysis focused on competitive broadband markets and technologies. He can be reached at (760) 753-2890 or via e-mail at mitchshapiro@ home.com.