Blog Post

Gigabit DSL: Yes, it will happen

Stay on Top of Enterprise Technology Trends

Get updates impacting your industry from our GigaOm Research Community
Join the Community!

Copper is the cockroach of the telecom world – it just doesn’t go away. And if telecom technologists have their way, it could soon be carrying data at speed of gigabit per second.
Last week, ECI Telecom and a bunch of other companies announced a new consortium that would work on a technology called the Dynamic Spectrum Management (DSM). The Chief Scientist Office of the Israeli Government has financed the consortium with a grant of about $10 million. DSM is widely viewed as the next evolutionary step after VDSL2. DSM, when commercialized could help provide fiber optic like speeds over copper, the consortium says. DSM addresses one of the biggest issues with the DSL technology – interference also known as crosstalk.

“The main obstacle for the advancement of DSL technology is the interference (“crosstalk”) generated from different DSL lines that share the same telephone cable binder,” said Professor John Cioffi, Professor of Engineering at Stanford University, a pioneer of DSM research, who is also recognized as the inventor of the DMT line code. “DSM is a promising technology for the future evolution of broadband access networks using existing copper infrastructure.”

People should pay attention to what Cioffi says.

He was recently awarded the Marconi Prize (and is the 2006 Marconi Fellow.) He is a soothsayer when it comes to all things DSL. According to DSL Prime, in 1990 he predicted that DSL could deliver between 5-to-10 megabits per second. Then in 2002 he predicted 100 megabits per second over copper. That happened. By 2004 companies like Ikanos and Metalink were showing off chips that could do 100 mega up and down. So now lets take what he is saying very seriously.

“Phone lines are big antennas that radiate into one another,” Cioffi says. “They are their own worst enemies when they are all bundled together. Any kind of [electromagnetic] noise from AM radios, fluorescent lights or your vacuum cleaner can get into these things and cause problems.” (via Stanford Report.)

Back in the 1990s his solution was to transmit data between two modems – say one at home and one at the telco central office – and connect them with each other via 256 different 4 kilobit per second channels. The traffic would flow over the less congested channels, and interference would be overcome.

With DSM, Cioffi is taking copper to the next level. DSM packs more channels and also uses the higher frequency bands that have not been useable because of extreme interference. He is betting that DSM is going to be big, and has decided to start a new company, Adaptive Spectrum and Signal Alignment (ASSIA) Inc. (more details to follow!)

29 Responses to “Gigabit DSL: Yes, it will happen”

  1. failquick

    J.M. Cioffi, B. Lee, M. Mohseni, A. Leshem and L. Youming, “Gigabit DSL,”
    T1E1.4 contribution 2003/487R1, Washington, DC, August 2004.

  2. Why should copper go away? It’s been in use for over two hundred years as a telecommunications medium. Every new generation finds new uses for it. Same thing with ADSL. And our children will find yet another use for copper wire.

  3. I think you are misinformed. There is no mentioning of 1 Gbps speeds.

    Page 42 of Prof. Cioffi’s document on DSM:

    “…200 Mbps symmetric transmission on 500 meters of cable is possible, and 100 Mbps at 800 meters, 50 Mbps at 1.5 km and 25 Mbps in excess of 2 km. As fiber enters the loop plant and reduces loop lengths, DSM-charged DSL will enable 100 Mbps 100BT speeds…”

  4. John Thacker

    The world market is MUCH bigger than the US one and rolling high percentage fibre cover in many places just isnt practical, e.g… Australia

    And it just isn’t practical in much of the US, too, you know. Denmark has a pretty high population density, after all. Canada does the best of any of the G-7 countries, and the US is second, also ahead of the EU-15 average. The top 10 countries, other than South Korea, are all fairly far north, Canada and the Nordic countries. The UK, France, and Germany are a little behind the US. Australia a bit further. New Zealand, Italy, Spain all lag. Ireland lags quite a bit.

  5. Oh, we are SO FAR AWAY from this in Australia. Telstra (the main telco who owns all the last mile) spouts about high uptake and delivering value to the end user. Most Australians (even as close as 10km/6mi from a major CBD) can’t get ADSL and are limited to dialup or satellite, with even the dialup cruelled sometimes due to Telstra’s policy of installing pair-gained lines.
    I am on 1500/256 ADSL, which is the best most Australians who can get ADSL have access to. I am data-limited to 10Gb daytime/10GB offpeak data per month. All this costs me around AU$60/month.
    ADSL2/2+ rollout is limited and is being undertaken by the larger second-tier ISPs as they are sick of waiting for Telstra to install infrastructure in exchanges.
    Live, used, cable exists only in Sydney, Melbourne and Brisbane AFAIK (happy to be corrected).

    Like I said, Australia is SO FAR AWAY from this.

  6. ..and John, like it’s all about what the americans(ATT & Verizon) are doing right? Lets take a quick look at where the USA is in bandwidth/population…oh yeah …12th. Go Denmark this months no. 1 with a bullet. (USA = biggest in volume tho..).

    The world market is MUCH bigger than the US one and rolling high percentage fibre cover in many places just isnt practical, e.g… Australia

  7. ATT (SBC) can’t multiplex HD signals in Project Lightspeed tests at 20 Mb/s on VDSL D-Slams supposed to deliver 50 plus. Can’t wait to see them try to multiplex gig over copper to 10 million customers using DSM…by the time it arrives commercially (if at all), Verizon will have 30 million FiOS pops installed.