US Telco Support > AT&T > AT&T U-verse > Is AT&T investigating use of Vectored VDSL or cuPON?

Is AT&T investigating use of Vectored VDSL or cuPON?

There is an engineer named John Cioffi, he's a Professor of Engineering at Stanford who has been doing cutting edge research on all things DSL for many years. He was awarded the Marconi prize for his contributions to ADSL design.

He's been working for around 6 years now for what is thought to be a potential successor to VDSL2 ... it's a technology called DSM (Dynamic Spectrum Management). The basic idea is that instead of the DSL head end independently and individually managing each customer's line by itself (like a VRAD or DSLAM currently does), the new version would intelligently manage all 50 pairs in an single cable as one unit. The DLSAM/VRAD could then optimize the entire cable bundle together and eliminate the cross-talk that currently prevents VDSL2 from getting higher speeds.

He's estimating that DSM could bring speeds of 1000 Mbps at 1000' from the DLSAM/VRAD over 4 pairs, thus the technology is informally being referred to as "gigabit DSL".

Article on the Subject.

Technical Paper on the MIMO techniques necessary for DSM.

SomeJoe7777,

Thank you so much for the reply. You basically confirmed my impression of the general concepts involved. Also, interesting note on the Marconi prize. I missed that I guess. Impressive guy, indeed.

It's good to know there are alternatives to more than get us by until it makes sense to roll fiber all the way to the home. It sounds like everything else, it will be a gradual transition over time, again, with these type of solutions more than getting us by until then.

Does anyone have any ideas about at&t's general mindset on the successor to VDSL2, especially what time-frames might look like. I don't imagine they'll NEED to start aggressively doing more than VDSL2 until maybe 3-5 years out from now. For two reasons... until around then, there won't be enough reasons for MANY people to need more than a 24/3 Mbps connection badly enough to want to pay the price for it. And related, even when that equation starts to turn, they can compete more aggressively on price to stay competitive and keep cable's speed advantage at bay. But certainly by 5 years out, they will be forced to change up their game.

As another stop-gap, they could start offering more speed profiles and at least provision customers close enough up to 75 Mbps or so, even on the VDSL2 they have today. So they have a lot of options, but again, eventually they'll have to have another plan, and by eventually, I'm thinking no more than 5 years or so with DOCSIS 3 breathing down their throat. Granted, DSL / VDSL technologies almost always offer a better quality connection (compared to DOCSIS, at least the way most cable companies oversubscibe it excessively), but the difference is not enough for the majority of everyday / average people to realize or care much about.

Anyway, so again, anyone have any information about what the big T is thinking? I'd be most curious.

Thanks,
Alan

Five years? More like two years.

Comcast is rolling out DOCSIS 3 faster than any other provider and is destined to be the premier provider in the US if they continue to gobble up cable companies left and right. Comcast has cash to burn and they will continue to speed up the deployment of DOCSIS 3 and buy up the smaller cable companies. Time Warner is starting to show signs of DOCSIS 3 deployments and I'm sure that Cox isn't too far behind.

The market is growing fast, very fast and if AT&T doesn't do something about the changes that are going to be happening within the next two years they will be left in the dust.

Granted, all of this sounds very interesting and the idea of getting more out of the copper sounds like a good idea but honestly, if AT&T is thinking about doing all of this they are going to have to bite the bullet and replace all the aging copper in their plant. There is no reason why there should be neighborhoods that have copper lines that are older than twenty years.
--
Tom

trparky,

I'm not in total disagreement with you, but keep in mind, competition isn't only in speed. From my experience in two different locations, DSL technologies have always been more reliable/stable than DOCSIS, at least the way Comcast was deploying it in those areas (Walnut Creek & Pleasant Hill, CA).

Then there's the pricing aspect. They can just price appropriately until they can get higher-speed alternatives in place. Cable doesn't seem at all too eager to get into a pricing war. They seem perfectly happy being the "Elite" (speed-wise at least) broadband provider, and pricing accordingly.

And lastly, the issue of demand. I think it will take AT LEAST 3 years before there are enough applications (not in the sense of programs, but uses) that demand an aggregate speed over 24/3 Mbps (especially on the downstream side... upstream MAYBE). Sure, there will be a certain percentage of customers that have real uses for speeds over that (especially in the upstream), but the average person won't until at least three years, and likely up to 5.

Also, from some of those original links I put in, the long term inflation of Internet access speeds has historically been 20% per year. Currently the average speed in the US is around 6 Mbps or less. This would also indicate they have plenty of breathing room to map out there next phase.

And to your point about fiber, also from one of those links in my original post, it is quoted that FTTH deployments are about 3 to 4 times more expensive than FTTN deploys are.

Additionally, DSM level 3 will also bring with it very sophisticated means of line issue diagnosis and maintenance. It will auto-correct some issues, and for most others, give very exact readings as to where/what the problem is. So DSM level 3 solutions should vastly reduce Operation and Maintenance costs of present ADSL and VDSL systems.

Given all this, the case to throw in the towel and go FTTH is not such a clear cut decision. Especially when at&t needs to be plowing money into their golden goose, which is wireless.

So continued use of copper seems quite viable for perhaps as long as 20 years. And in the meantime, they can be sensibly expanding to FTTH as opportunity presents (greenfields, where they have to dig a trench anyway, etc.).

To me, I don't care HOW I get the Internet speeds I want/need, as long as I get it. I'm technology agnostic as long as the end result is the same. I do find the engineering feats that are allowing all this unintended use of copper in both coaxial and twisted pair quite fascinating and neat though.

Anyone else have any info or views?

Thanks again SomeJoe and trparky for your posts.

reply to trparky

Wireless? Last time I checked, people spend most of their down time at home so where would wireless come into the picture.

If you ask me, AT&T is fine when it comes to their wireless speeds. And this is coming from a person who own an iPhone. Granted there are areas where coverage is a bit spotty but fix those and they should be fine. To me, the AT&T wireless network works fine. As for speeds, I don't really care about how fast my wireless speeds are. If I need to download something huge, I have my home Internet connection.
--
Tom

reply to nishiko7

reply to trparky

reply to trparky

reply to nishiko7
This news is exciting to me.

I keep hearing people say that we have gotten every last drop out of the copper. I always think.... we said that when 6mbps dsl was out. (Or atleast I hear it.) I really hope that ATT can find a use for something better than VDSL 2 on its copper. VDSL2 is nice, but no one said that Copper transmission improvements are over.

I'm under the impression that the Uverse network lends it self to upgrades more easily than the old DSL network. They went from VDSL to VDSL2 easily. And the VRADS seem like it wouldn't be too hard to upgrade the equipment inside of them.

Comcast is pushing for 100mbps, and IM glad they are. Also with FIOS in some areas pushing for the same speed. No way should ATT just sit on VDSL for 2 years. IMO they should already be testing their next solutions. Because it will take things that long to trickle down into real world use.

I know ATT is working on a fiber network, but for some reason I don't see that happening anytime soon. Please prove me wrong.

For those like me, we cant get 24/3, and there are many of use on uverse. Uverse seems to already be showing signs of age, err issues. Although that's how things like Uverse are. I really look at it as a bridge to hold us over until fiber is here.

reply to djrobx

reply to doublea
doublea,

Agree mostly completely with you.

Also, you should be seeing some hope with pair bonding (which I know some here will think is laughable as it's been talked about for years)... it is coming, especially for people who can't sync at at least the 25 Meg profile.

And I agree at&t better be seriously looking at these now, as it does indeed take time to actually implement once the decision is made.

Anyway, good dialogue from everyone!

reply to nishiko7
Remember there are those who HAVE $$$ to spend on speed-whatever it costs at any cost.
Most of us have a budget to live with and frankly unless you are a power GAMER, or a business, or use P2P I question just how much speed you really have to have.
If ATT wanted to just gain customers, the rural areas are begging for any kind of dsl. There arent dslams in most rural areas.
If the quest was how to economically serve everyone using copper with dsl 6, 8 or 10 miles out from the Central Office instead of "making speed the goal" we would be going somewhere.
Actually, I am afraid ATT may be thinking of doing a Verizon and selling off rural areas to a Fairpoint type company.
That would be a tragedy.

sbcretired, great points.

I too hope at&t doesn't go down Verizon's route regarding rural customers. I think a recent legal ruling against the use of Reverse Morris Trusts should help dissuade them, but no guarantee. There's always new tax loopholes to exploit, especially for moneyed interests (that have the resources to get fancy).

I agree (even though I won't benefit... I'm in the burbs here)... the rural situation should be improved before we focus on anything else. I'm not talking about EVERY rural customer, but where it makes some sense. If there's one person out 5 miles from any other house/business with broadband, then obviously this might be a case where satellite or another solution will have to make due. Most people do have some choice about where they live too.

But in many cases, rural expansion would make sense for at least ADSL service, especially if amortized over a longer period of time, say 10 years. These companies could probably at least break even on 75% of currently unserved rurals. Again, the other 25%, it may not make economic sense to do, but that's a different debate (subsidies, etc... too political for these threads).

The more people who have affordable access to at least a basic level of broadband (at least 1.5 Mbps / 256 or something along those lines), the better this country will fare economically (and other ways) relative to other countries. It just seems so obvious to me.