July 1998

xDSL Technology & The Internet (Part II)

Copper Applications in Electronic & Communications

By Marc J. Zionts CEO, Westell, Inc.

Presented at Copper Development Association Inc.'s
Annual Spring Meeting
New York City
June 4, 1998

I want to talk about an important new technology called xDSL, how it is relating to the rapid emergence of the Internet and why it is important to the copper industry. But first, a few words about my company.

About Westell, Inc.

Westell, founded in 1980 and headquartered in Aurora, Illinois, employs more than 800 team members and is a leader in telephone company local loop equipment technologies. We are also a leading provider in audio and video conferencing bridging services and in xDSL products. We have particularly strong customer/partner links with Bell Atlantic, British Telecom, GTE and Sprint among the telephone operating companies, and with equipment manufacturers including DSC, Fujitsu Telecom Europe Limited, Lucent and Texas Instruments. In addition to Westell's DSL systems, we also enable DSL on our partners' products, including DSC's LiteSpan 2000 Digital Loop Carrier and Lucent's #5ESS Central Office switch and their SLC5 and SLC2000 Digital Loop Carrier systems.

The Internet

In the telecommunications business the buzz these days is all about the Internet. The growth in everything to do with the Internet is phenomenal in all respects. By the end of 1996 the Internet had already grown rapidly to the point where there were nearly 40 million users worldwide. One year later there were more than 100 million (and that's been six long months ago, so it's higher still today). In terms of domain names (which roughly relates to Internet sites), the number grew by about the same proportion, from 627,000 at the end of 1996 to 1,500,000 a year later. When you combine the increased numbers of users with more places to go, and more content and complexity within those sites - and therefore more bits of data moving from those sites—we have the truly amazing statistic that the traffic on the Internet is doubling every 100 days.

Contrary to what you might have read, people are making lots of money on the Internet. Cisco Systems booked $100 million in sales on the Internet in 1996; in 1997—$3.2 billion! Dell Computers reported daily sales of $6 million several times during the December 1997 holiday season through their web site and subsequently, as of June 1998, is averaging over $5 million per day in sales. Auto-By-Tel is generating 100,000 purchase requests each month for new and used cars, generating monthly sales of $500 million.

Folks, the Internet is here and only one thing stands in the way—bandwidth!

Bandwidth Alternatives

Eckhardt Pfeiffer, Chairman and CEO of Compaq Computer, put it succinctly in a recent Comdex presentation, "We're sucking information through a straw when what we really need is a fire hose." There are basically four bandwidth alternatives: ether, coax, glass and copper.

Wireless Systems, using the air as a medium, include numerous satellite systems, some of them still under development. A dizzying array of acronyms and names include:

LEOs NEOs Teligent
Iridium Teledisic Iconet
Spaceway Cyberstar Skybridge
Celestri Ecco Globalstar
Ellipso Astrolink DirecPC

Some of them will eventually disappear into the ether, while others may evolve into serious competition for wire-based systems. In most cases the problem for maintaining a viable Internet service is the difficulty of establishing the reverse link—the upstream link. Television systems, for instance, are inherently one-way broadcasts and require a separate uplink, usually the wire-based phone line.

Coaxial Cable is the traditional medium over which cable TV channels are delivered to the customer. The coax extends into the house and connects to the TV equipment. Coax has one major advantage and five major disadvantages. On the plus side, the bandwidth capacity of coaxial cable is large, and downstream Internet service can be easily accommodated along with many TV channels.

On the other side of the ledger, only about 70% of US homes are hooked up to cable TV, vs. 95%-plus having telephones. Second, the technology was designed for one-way communication, and only recently have some cable TV providers begun to provision two-way service, which in most cases consists of carrying fiber optic links closer to the customer, combined with better quality coax and electronics to carry two-way signals. The coax loop is shared by many customers in a neighborhood, so that, during busy periods of Internet use, transmission speed (and available bandwidth for a given customer) can shrink dramatically. This sharing results in the possibility of Internet communications being intercepted, so that security is generally inferior to that for point-to-point systems. Finally, the cable TV system has a reputation of being far less reliable than does the telephone system.

Despite its drawbacks, coax is the short-term competition to copper-based systems for delivery of the Internet and major investments are being made today. We welcome the competition, and in fact see it driving the telephone-based Internet business to implement their solutions in a more timely manner.

Fiber Optics delivered over glass has, for all practical purposes, unlimited capacity and bandwidth. Many observers see it as a long-term replacement for copper in the local loop of the telephone system, and indeed it is presently driving ever closer to the customer. But for that last mile the investment in an all-fiber system would be enormous and would take many years, probably decades. The Internet needs bandwidth today.

Copper solutions are available right now, and are being implemented as we speak. The copper twisted-pair local loop, under the control of local telephone companies, is available virtually everywhere; one might say it is ubiquitous.

There is a common misconception that the copper network is the bottleneck to the information superhighway. In fact, analog voice-grade telephony uses less than 1% of the copper transport capability. The rest is essentially going to waste.


So, what is xDSL and what does it do for users?

xDSL is a generic term for different "flavors" of DSL, or digital subscriber line. It is in fact a solution to the "last mile problem." It converts an otherwise analog local loop into a digital line, employing advanced modulation techniques that enhance the loop to provide greater throughput. The upstream and downstream speeds that can be attained depend on several factors: distance to be traveled, modulation techniques used and network design.

If you'll permit me to throw a couple of technical terms your way, xDSL delivers high-bandwidth data, packet video and voice transmission at fiber optic quality over copper, by using the available frequencies above the 4 kilohertz voice band.

Advantages to Users
All this is fine, but what does xDSL mean to users? First and foremost, the connection is "always on," one of the holy grails for serious Internet users. No dialing, no network bottlenecks, no busy signals, no hassles. This means no call set-up, and no user name and password to enter. Just click your browser and you're connected. A standard Ethernet interface is used and the same applications and drivers are used as on a corporate local area network. In fact, one of the major uses for xDSL is to connect users to their corporate networks.

But doesn't this "always on" feature clog up the phone system, particularly the switches at the local exchange? (After all, this is one of the problems with accessing the Internet through dial-up modems.) The answer is no, because this data traffic doesn't pass through the voice switch. It uses data networks, such as those now being deployed using asynchronous transfer mode (ATM) technology. Thus the conversion from traditional, low-speed modems to high-speed xDSL will actually help by relieving phone-switch congestion.

One other advantage of xDSL should be mentioned—security. Rather than being a shared Internet loop into which multiple users tap, as is the case with cable TV-based systems, xDSL provides a point-to-point link between the user and his Internet service provider or other point of contact. Thus, individual virtual channels link users, over their individual copper cables, directly to their host sites, resulting in a virtual dedicated line between them—a secure line.

The Copper Advantage

The vendors of xDSL equipment such as Westell and many others are not going it alone. Significant computer industry support exists, featuring such major players as Intel, Compaq and Microsoft. In fact, these companies and others have formed a consortium to speed the deployment of xDSL, and are rapidly moving toward standardizing a relatively simple version known by various names, including ADSL-Lite, G.Lite or Universal ADSL. The notion of this simplified version is to provide a consumer-installable, plug-n-play solution that is as easy to install and operate as today's standard modems.

Copper-based xDSL systems have two huge cost advantages over cable modems and other new systems that might be deployed. First, the equipment only need be installed for those customers that want the service. There is no general infrastructure investment needed at the customer end (though central office and network investments must be, and are being, made). Thus, the economics are largely based on customers served rather than customers passed. Second, the copper in the local loop is already there, an existing, paid-for asset. For you, the copper industry, the added advantage is that, when xDSL is deployed, incremental investments are more likely to be in copper.


Let's put it in simple terms: xDSL turns copper into gold! Progressive telephone operators around the world recognize this and companies like Bell Atlantic, British Telecom (BT) and GTE are working with Westell to offer xDSL-based services to their customers starting this calendar year. AT&T's pending merger with TCI only adds fuel to the fire in the race for bandwidth and will most likely cause deployment plans for all telephone operators to accelerate.

Also in this Issue:


2007   |   2006   |   2005   |   2004   |   2003   |   2002   |   2001   |   2000   |   1999   |   1998   |   1997