Guess whos bringing fiber to the curb?

Guess who`s bringing fiber to the curb?

At the risk of offending the squeamish, here?s a hint: It could be a transmission technology based on copper. Read on?if you dare!

Kurt Krueger Ericsson Network Systems

The long-heralded fiber revolution, which had been all but halted at the central office by the immense challenges of the local loop, may now be poised for a breakthrough. While data is a growing part of overall fiber traffic, today`s optic highways still carry mostly long-haul time-division multiplexed voice traffic. Now changes in edge-of-network technologies and the growing subscriber demand for high-bandwidth services are hastening the long-awaited full deployment of the fiber network.

Operators the world over anticipate an increase in revenue as consumer demand grows for advanced multimedia services. These broadband services--including video-on-demand, home shopping, interactive conferencing, and Internet access--are tailor-made drivers for the fiber revolution. With virtually every small office and increasing numbers of residential customers now using computers, the Internet and telecommuting are fueling rapid growth in demand for high-speed communications at the curb.

But while these new services promise vast new revenue potential, two significant obstacles stand in the way of an interactive broadband future.

First, current networks simply cannot deliver the necessary two-way bandwidth needed on this new interactive data highway. Traffic jams will occur when small/home office users access this pathway to retrieve information from Web sites, FTP servers, and corporate file servers. Multimedia access will also create a heavy traffic flow, as consumers tune into video-on-demand and other broadband services.

The second big hurdle is cost. Access networks traditionally represent the majority of a telecommunications operator`s investment and network operating costs. Most telephone companies have tens if not hundreds of millions of dollars invested in copper-based local access resources. These local loops are designed and built to deliver long-term stability, and to date no single application has emerged that will justify the considerable investment required to revamp the access network.

While a number of technologies and loop infrastructures provide limited broadband services to the access network, the industry has looked hard for a truly affordable way to bring speed and bandwidth to the curb. Now, with the advent of asymmetric digital subscriber line (ADSL) the solution may be at hand.

The ADSL solution

ADSL is an emerging modem technology that converts existing twisted-pair copper telephone lines into highly economical access paths for multimedia and high-speed data communications. An ADSL circuit connects devices on each end of a twisted-pair telephone line, creating distinct information channels--a high-speed downstream channel and a medium-speed upstream channel over a selected "plain old telephone service" (POTS) or Integrated Services Digital Network (ISDN) line.

ADSL can transmit up to 8 Mbits/sec to a subscriber and up to 1 Mbit/sec upstream depending on line length and loop conditions. At these rates, ADSL holds the promise of expanding existing access capacity by a factor of 50 or more using existing copper cabling.

Most ADSL systems consist of a digital subscriber line access multiplexer (DSLAM), which includes an access multiplexer, ADSL line cards, POTS filters, a network concentrator, and customer-premises equipment such as a stand-alone or PC network interface card modem and POTS filter. The DSLAM acts to aggregate a large number of ADSL subscribers into one or several uplink ports to a frame relay or Asynchronous Transfer Mode (ATM) backbone network.

ADSL systems typically also include management elements to control the DSLAM and customer-premises equipment. Remote splitters are used to combine and separate the incoming data and narrowband (POTS and ISDN) streams and transmit both over a single line. Many ADSL transmission systems are now based on a standards-compliant ADSL technology using discrete multitone line code for optimum performance and protection against external noise.

This ADSL broadband solution delivers high-speed access at a competitive cost, while making maximum use of the existing backbone network. It is increasingly accepted as the most promising speed/capacity solution for high-speed Internet access, remote local area network access, and multimedia applications within the family of digital subscriber line (xDSL) technologies. This family of alternatives is fairly broad. For example, high bit-rate DSL (HDSL) is a modem and twisted-pair technology that delivers T1 or E1 speeds using two and three pairs, respectively. IDSL employs ISDN transmission technology to deliver data at 128 kbits/sec to an IDSL "modem bank" linked to a router. Rate-adaptive ADSL (RADSL) employs modem-generated line tests at startup to adapt the operating speed to the maximum line capacity. Symmetric DSL (SDSL) delivers HDSL and POTS over a single telephone line. Very high data-rate DSL technology (VDSL) uses a modem over twisted-pair access to deliver data rates in excess of 51 Mbits/sec.

ADSL applications and benefits

ADSL is generally seen as opening the access doorway for two growing applications: high-speed data communications and interactive multimedia. High-speed data may include Internet access, telecommuting, remote LAN access, and certain kinds of specialized network connections. Interactive multimedia will encompass a growing array of increasingly popular applications such as video-on-demand, video games and catalogs, user-selected TV programming, and information retrieval.

ADSL provides the access bandwidth needed to deliver a wealth of broadband services over existing public networks, which now approach 750 million lines globally. Because it brings affordable high-speed access directly to the customer premises, ADSL extends the benefits of the fiber network to millions of corporate, residential, and telecommuting subscribers. That means providers can bring enhanced services to consumers immediately, thus opening these lucrative new markets to both telephone companies and application suppliers.

ADSL is an "always on" technology that allows providers to offer services at nearly 300 times faster than 24.4-kbit/sec modems, more than 100 times faster than 56-kbit/sec modems and fully 70 times faster than 128-kbit/sec ISDN. Of course, the actual data rates available from ADSL depend on local line length and loop conditions. But with those impressive speeds and economies, ADSL opens the doorway to multimedia and broadcast-quality applications, including collaborative computing, videoconferencing, distance learning, and video-on-demand.

Because ADSL data travels on a private, nonshared line, where transmission speeds are not affected by other users, the technology creates a highly secure and private communications channel. ADSL also allows line-hungry providers to use one line for new data services while maintaining telephone service on the same line.

The industry is rapidly converging to set ADSL standards that support interoperability and mass-marketing programs. Major service providers have tested ADSL, and major equipment manufacturers now offer mature, cost-efficient second- and third-generation ADSL systems. Finally, ADSL networks integrate easily with ATM traffic, giving telephone companies a smooth upgrade path to the future.

There are a number of issues affecting deployment of ADSL. In many cases, subscribers live or work too far away from a provider`s central office to receive telephony, ISDN, or DSL service. In these instances, local-exchange carriers are required to provide a digital loop carrier (DLC) as an element of its access network. DLC systems support more than 25% of POTS customers in North America and are viewed as a key driver for wide-scale deployment of ADSL services. Because interoperable, multiservice devices integrated into existing equipment platforms are seen as a major requirement for the acceptance of ADSL, DLCs are expected to be a major competitive context for the introduction of ADSL.

Having established discrete multitone as the ADSL standard, industry analysts now say that perhaps the biggest ADSL-related challenge for service providers will not be technology or equipment costs but deployment and operations, administration, maintenance, and provisioning challenges in the complex local-access environment.

ADSL and the future

Ironically, a copper-based technology will dramatically increase the volume of traffic carried on the fiber-optic network. As new interactive multimedia services are introduced, large numbers of high-megabit subscribers will be attracted to the networks. Business customers, especially, are expected to drive the push for accelerated fiber-optic deployments. In fact, ADSL will bring thousands of multi-megabit subscribers to the edge of the network, providing the impetus and revenues needed to fuel a fiber-optic expansion.

While existing networks were engineered to handle voice calls with an average length of about three minutes, data and multimedia ADSL traffic can last for hours. Thus, this "always on" service will also dramatically increase the duration of traffic on the networks. These increases in traffic volume and duration will place huge new demands on the fiber-optic network connecting the central offices. Industry observers are debating the implications of these developments, but several key changes seem likely to accompany the advent of ADSL.

In the short term, ADSL may well drive the deployment of wavelength-division multiplexing toward the edge of the network. Owners and operators will be reluctant to trench up the ground and bury more fiber; moving from OC-3 (155 Mbits/sec) to OC-12 (622 Mbits/sec) and on to OC-192 (10 Gbits/sec) is prohibitively costly. Wavelength-division multiplexing suggests itself as a practical and cost-efficient interim fix.

As a "last mile" alternative, ADSL may naturally impact the trial and deployment of "fiber-to-the-door." Considering the traffic and network maintenance challenges faced by pushing fiber this far, most observers say it will take upwards of a decade before fiber reaches subscribers in significant numbers. By bringing broadband services to consumers now, ADSL may provide the economic incentive needed to drag fiber those last few feet to the subscriber`s doorway. ADSL may also ensure the near-term viability of Synchronous Optical Network (SONET), in that there are now some efforts to deploy ATM and Internet protocol (IP) over SONET.

Finally, and perhaps most important, ADSL will drive the integration of a wide variety of access network technologies. Near-the-curb networks are currently distinguished by the number and complexity of deployed technologies, with the basic voice networks overlaid with time-division multiplexing (TDM) for data traffic and additional overlays for ATM and frame relay traffic. ADSL may actually be the driver that will force a beneficial integration of access technologies. As ADSL spurs the rollout of broadband technologies, there will be an at-the-edge move to integrate frame relay, ATM, TDM, and possibly IP traffic.

Copper driving fiber

Fiber optics still holds virtually unlimited potential as a way to bring an impressive new menu of advanced broadband services to the marketplace. But surprisingly, it may be a copper-based technology that feeds multimedia into the optical path and finally brings fiber all the way to the curb. As competition, liberalization, and subscriber demand drive the deployment of enhanced services, copper-based ADSL may be the springboard that finally delivers the full benefits of the fiber-optic network. u

Kurt Krueger is deputy director, product planning and management for the xDSL Multi-Service Access Group of Ericsson Network Systems (Richardson, TX).