General Discussion
  >> Fibre Broadband


Register (or login) on our website and you will not see this ad.


Pages in this thread: 1 | 2 | (show all)   Print Thread
Standard User MaryHinge
(member) Tue 08-Nov-11 08:58:37
Print Post

100mps+ FTTC via "vectoring"?


[link to this post]
 
Q2 2012 results slide 12:

FTTC:
– ‘Band Plan’ frequency change agreed, delivering up to 80Mbps in 2012
– other technologies to enable > 100Mbps


The lower pie chart on the right-hand side mentions After ‘Band Plan’, vectoring and public funding

Nice to see Openreach are considering ways of opening up FTTC to even greater speeds. Slide 13 is also interesting.
Standard User DLS
(newbie) Tue 08-Nov-11 21:16:25
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: MaryHinge] [link to this post]
 
I posted a link to a paper on this a few months ago; here it is again.

In brief, vectoring is a computationally-intensive technique that estimates and corrects crosstalk errors. As a consequence, it can enable a line to achieve speeds close to the theoretical maximum for their length. The paper contains some nice illustrations of this.
Standard User WWWombat
(committed) Wed 09-Nov-11 02:20:11
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: MaryHinge] [link to this post]
 
Isn't "vectoring" a seemingly miraculous innovation? Just as many people are starting to get bitten by speed drops, as cable fill increases, along comes an invention to solve it.

And combined with the 17a adoption, the prediction of 100Mbps+ doesn't appear to be limited to those living inside the cabinet either!

At this rate, FTTC isn't looking like such a dumb decision - and could extend the life of the copper network by another decade or two easily.

For those not cooped up in crowded city-centres, the concept of a "mini DSLAM" (from page 13) must be interesting. It shows that BT might start pushing small cabinets further out into the network, opening up those at the end of long lines.


Register (or login) on our website and you will not see this ad.

Standard User WWWombat
(committed) Wed 09-Nov-11 02:24:09
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: DLS] [link to this post]
 
I found this document (also by Ericsson) that shows some of the future promise of vectoring. It isn't so technical, and might be a bit easier to follow for most people.

The only downside to this document is that it is a little older, and the graphs for vectoring don't give such high download figures - but they still show the fact that crosstalk is effectively eliminated.
Standard User lockyatlrg
(fountain of knowledge) Wed 09-Nov-11 08:24:40
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: WWWombat] [link to this post]
 
Can't they enable 30a line profile for speeds of up to 200mbs.?

BT Infinity
ROUTER:-Netgear WNDR37AV
JDSU Stats
Sync 39999D 9995U
Attenuation: 9.6 SNR: 11.4
Line Length 300meters
Standard User DLS
(newbie) Wed 09-Nov-11 13:52:24
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: WWWombat] [link to this post]
 
Nice find, thanks.

To your earlier remark, yes, vectoring - and much of DSL for that matter - does seem miraculous at times (for me, 56K/s into the home as recently as eight years ago, now 40 Mbit/s). It's largely a consequence of being able to put an enormous amount of processing power into small cheap cards and boxes.

Years ago I used to do a lot of what would be termed signal processing, and ideas weren't in short supply but were were generally limited by what we could run in a finite time, sometimes for hours on a supercomputer. Some of those "tricks" seemed pretty miraculous. I haven't done the detailed maths, but it's fair to say that the kind of thing sitting inside a tiny VDSL modem and similar technologies is effectively ~100-1000x the power of one of those big water-cooled monsters from ~30 years ago.
Standard User Ignitionnet
(knowledge is power) Wed 09-Nov-11 14:12:43
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: lockyatlrg] [link to this post]
 
Not approved by Ofcom, goes up to 30MHz while BT only have approval to use up to 17.
Administrator MrSaffron
(staff) Wed 09-Nov-11 15:31:10
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: Ignitionnet] [link to this post]
 
The ANFP has to take account of the cross-talk and possible interference to and from other sources.

Some countries operate differently and deploy the new profiles, dsl standards and then fix things as they go.

BT and others can generally trial services using things that are part of the ANFP, but are expected to do this simply to gather data in the real world to support ANFP inclusion. Lab testing can demonstrate only so much.

Andrew Ferguson, andrew@thinkbroadband.com
www.thinkbroadband.com - formerly known as ADSLguide.org.uk
The author of the above post is a thinkbroadband staff member. It may not constitute an official statement on behalf of thinkbroadband.
Standard User WWWombat
(committed) Wed 09-Nov-11 15:54:53
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: DLS] [link to this post]
 
In reply to a post by DLS:
for me, 56K/s into the home as recently as eight years ago, now 40 Mbit/s

I haven't done the detailed maths, but it's fair to say that the kind of thing sitting inside a tiny VDSL modem and similar technologies is effectively ~100-1000x the power of one of those big water-cooled monsters from ~30 years ago.

Probably more recent than that. As I understand it, all the current DSL technologies are designed around N discreet 56K modems, each working with 4.3125kHz of spectrum - called a "tone".

The 17a profile comes with 4096 such tones - so even with a 50:50 split of up:down, and an allowance that not all of these are runnning at 56K, a modern VDSL2 modem is doing, in parallel, roughly the same job as 1000-2000 old modems, of circa 1995 technology.

And that makes the work being done inside one of the street cabinets even more amazing!
Standard User WWWombat
(committed) Wed 09-Nov-11 16:07:35
Print Post

Re: 100mps+ FTTC via "vectoring"?


[re: MrSaffron] [link to this post]
 
If vectoring is a way to reduce cross-talk, it might act as an enabler for higher frequencies.

However, my understanding is that vectoring happens within a DSLAM, taking account of the neighbouring lines only within the same cabinet.

That probably means that crosstalk & interference will still happen between colocated cabinets owned by different operators, which is exactly the kind of thing that the ANFP acts to prevent.

It could be that the Ofcom desire to keep competition open as much as possible actually leads to a reduction in the (future) technical capability of the copper access network. It'll be interesting to see how this plays out with SubLLU in the future.
Pages in this thread: 1 | 2 | (show all)   Print Thread

Jump to