Aussie boffin turbo charges copper broadband An Aussie PhD student has developed an algorithm that he says can turbo charge an ordinary copper phone line to deliver downloads at 200Mbit/s. John Papandriopoulos of the Universtity of Melbourne has patented maths he did for his thesis last year on "Dynamic Spectrum Management" (DSM). The algorithm aims to stomp crosstalk …
You can bet it will be patented and protected to the hilt, as the university will want to make a shedload of cash.
Of course a sensible student would hide any good ideas, develop them, then sell them to a large company. Or set their own company up. Sod your name on a paper, I want hard cash.
Idiot!! who needs fibre indeed...I DO! Fibre can definitely give gigabit speeds now, and do it reliably. All this guy has done is give BT another reason not to start rolling out modern cables.
but hey, what do I know, the victorian copper will probably scale to gigabit speeds (falling off to 2kbps 20 yards from the exchange!)
Yards? Meters!? Clearly you all need to re-read the Reg measurement standards and repost accordingly.
And speaking of which, we need a standard for data transfer speeds. Given the nature of Reg readers (see http://www.theregister.com/2007/10/25/barmaid_fined/comments/), I propose the bfpps (Bulgarian funbag pics per second).
Since images dimensions, quality levels, compression ratios, etc. vary significantly from source to soure, someone will have to do some serious long-term research to develop an overall average. I volunteer, provided you can convince my ISP to implement this new DSL technology in my area*.
*Okay, I'll do it anyway...
Wasn't it him who came up with the concept of DSM (as well as inventing the DMT line code)?
He has a DSM project at Stanford - and a 2005 start-up http://www.assia-inc.com/ which was set up to exploit his research...
The Reg had a story a year ago of one of the DSL vendors set to exploit DSM techniques: http://www.theregister.co.uk/2006/10/12/eci_faster_dsl/
Fibre would be great, but I guess it would take forever to become a "universal service"....
Sorry but even assuming you can mitigate the cross talk, and from what I've read that isn't a certainty, the calculations involved don't take into account the reality of our telephone network. Added to the fact that we can get better data rates with fiber today I can't see the benefit of this undoubtedly expensive kit over putting in the bloody cable network we need.
..good idea except that:
A) cable was only being laid by the Pay-TV companies who have since
B) found out that it's cheaper to move everyone to satellite; and
C) Telstra (who is still a mega-corporation owning just about all the copper lines in Oz as well as providing services over those lines) refuses to upgrade to fibre if it has to share. Isn't it nice to be a monopoly?
In other words, unless you already have fibre in your street or are moving to a new estate meant for the extremley rich (like the recent one in Sydney), you won't get to see Fibre in your lifetime.
DMT as used in ADSL and VDSL addresses this very problem in each of its 4.125kHz bands across it spectrum. So I don't quite understand what the difference it. What would be more interesting is how they've managed to increase the speed to 200Mbps and for what loop lengths?
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
