Home » ISP News »

BT Confirm UK Superfast Broadband FTTC Vectoring Trial for Late July

Posted Wednesday, June 26th, 2013 (2:09 am) by Mark Jackson (Score 4,409)
bt openreach engineer working

BTOpenreach has confirmed to ISPreview.co.uk that the first trial of its new VDSL Vectoring (ITU-T G.993.5) technology in the United Kingdom, which could help ISPs to deliver better FTTC broadband speeds over longer copper lines, is now set to begin at the end of July or early August 2013.

Vectoring technology effectively works to remove interference (crosstalk) from the “last mile” run of copper wire(s) between your local street cabinet and home or business. This is not unlike the noise cancellation method used by some modern headphones.

In theory Vectoring could help to push the top speed of Fibre-to-the-Cabinet technology up to 200Mbps (currently 80Mbps) but BT has previously said that it sees this more as a “speed enabler” than a booster. In other words it’s intended to improve existing service speeds, especially for those on longer lines, rather than the top advertised rate.

Having said that there’s no shortage of people whom expect that Vectoring and other technologies will ultimately help BT to push its top headline FTTC speeds beyond 100Mbps (Megabits per second), although we’ve previously been told that there are no plans to do this in 2013.

Vectoring Trial Coverage

As first reported in April (here), the tightly-controlled trial will take place in Barnet (London) and Braintree (Essex) in England. It will be rolled out in phases and thus not all of the related street cabinets will go live at launch. The most recent information we have is that cabs 12, 39 and 74 in Braintree and 26, 41 and 42 in Barnet will be included.

ISPreview.co.uk also queried whether the trial would be compatible with Openreach’s existing Huawei or ECI (VDSL) modems, which can also reflect the use of different local street cabinets by the same companies.

Openreach confirmed that the majority of end customers would be able to use their existing modem (we suspect that some of these might also need a firmware update but that’s not confirmed). However a “small number” could need a replacement modem and BT has said that it would work with ISPs to organise this.

It’s our understanding that BT’s trial will add Vectoring to all FTTC lines that connect to the related street cabinets (i.e. the best way for BT to get enough meaningful data) and they will also all be upgraded to the operators top 80Mbps (20Mbps uploads) service. Street cabinets in Braintree should be the first to benefit and those in Barnet will follow a few weeks later.

The initial trial will run until at least September and affected ISPs are now likely to begin contacting the associated customers. At this stage it’s not known whether BT will officially choose to roll-out Vectoring but the smart money says they will.. eventually.

The main drawback is that the upgrade does come at a slight cost and it’s up to BT’s trial to reveal whether or not the benefits make this worthwhile. It’s far too early to say whether or not this is a cost that would ever be lumped onto consumers but we suspect that BT will treat it more as a direct investment (like Virgin Media’s double speed upgrade).

Delicious
Add to Diigo
Add to Slashdot
Tags: , ,
Leave a Comment
35 Responses
  1. cyclope

    From reading info regarding vectoring the BT modems in use currently only support upto 100mbps on the lan side, so they couldn’t be used for higher speeds in reality as for vectoring the Huawei modems are reported to be compatible with vectoring where as the ECI modems aren’t,Plus there are quite possibly a lot of ECI modems that are in use connected to Huawei dslam’s

    I feel that vectoring should be offered free to all those customers affected by crosstalk,IMO BT has got some neck,to charge for vectoring, as it’s down to their outdated infrastructure (the copper.in some cases alluminimum local loop)that wasn’t even designed for adsl upto 2mbps to run on it let alone 80mbps

    They should be brought to book on this monopoly or not

    • The cost is most likely to be consumed by BT as part of its upgrade work but all the article says is that you can never be sure until the final details are known.

    • MikeW

      Vectoring only brings significant benefit if every line in a cable is vectored. Apply this to every cable from the cabinet, and it is easy to see that every line in the cabinet must be vectored.

      In practical terms, then, it is hard to see this being a price increase for end-users subscribers to buy themselves onto a vectored line.

      There are four gains that spring from this:
      a) It probably(*) allows BT to advertise a higher headline speed, of the order of 100-110-120Mbps. This helps them (and BDUK counties) to meet the EU digital agenda 2020 targets.

      BT can’t exactly charge individual end-users for the benefit of advertising higher speeds, and meeting targets.

      b) It probably(*) allows BT to extend the reach of SFBB speeds by another 300 metres or so, bringing more of those with longer lines (>1km from cabinet) into SFBB coverage. This aids the 2020 targets, and will build on the coverage figures that we’re seeing for BDUK bids in the 2015/2016 timeframe.

      BT can’t exactly charge existing individual end-users (already within range) for the benefit of covering and appealing to more subscribers on a cabinet. However, they will appeal to a whole next segment of the market.

      c) Some people will be able to actually achieve the new higher headline speed – probably(*) more than can achieve the current 80Mbps top-end. This allows ISPs to define new packages to take advantage of that extra headline speed – and Openreach to take a cut.

      This aspect becomes an optional way for existing subscribers to be charged more.

      d) The DSL speed checker will probably(*) give speed uplifts for everyone already connected to an FTTC cabinet – acting as a boost to take-up for those who felt the original uplift to be not worthwhile.

      In the end, this ought to be a win-win for BT to finance, increasing SFBB coverage, increasing take-up of SFBB services, and allowing them to market triple-play services to more and more subscribers.

      The real question is whether BT choose to vector all their cabinets, or whether they apply the same kind of “commercial criteria” to determining which cabinets are worth upgrading.

    • Roberto

      Vectoring can be enabled or disabled per port on the line card, you do not automatically enable or disable the whole card in the cabinet. In fact doing that could end up being a bad thing for a few reasons far too long to go in to here.

      Vectoring speed BOOST wise (at least from what we currently have) will likely only help those with short cable runs. If BT were to market it as a SPEED UPGRADE. Those with longer cable runs it will likely also improve speed but to nothing like the 200Mbps the article mentions. For those a considerable distance that currently only get currently get say 20-30Mbs range it will likely boost them nearer to the current UPTO rates of 40-80Mbs, which others nearer the cabinet already get.

      This is also probably why BT see and will likely advertise it more as a speed “enabler” rather than a speed “upgrade”.

      They could in theory market it as a new product with a higher top speed (say 120-150Mbs) but that would create a situation like we have already IE those near a cabinet getting full whack those further away getting significantly less.

      BTs idea is a good one, provide a service where the speed is consistent no matter where you are distance wise from the cabinet. It will stop complaints we have had for years of “haves” and “have nots” with regards to speed.

    • MikeW

      BT need vectoring to offer a headline speed boost – at least to a figure of 100Mbps.

      The reason? The counties are spending BDUK money as a stepping stone to the EU 2020 targets of 100Mbps to 50%, and universal 30Mbps. The only point in spending money on FTTC as part of that is if it can help reach both of those targets.

      For FTTC to help with the first target (50% at 100Mbps+), it *certainly* needs vectoring to deliver an increase in headline speed.

      For FTTC to help with the second target (universal 30Mbps), it *probably* needs to increase the range that 30Mbps can be attained. Other solutions (such as FTTP and FTTdP) will probably be still too costly to help sufficiently.

      As for whether it is bad to enable vectoring on all ports on a card – I disagree. The perceived wisdom throughout the written material on vectoring (research and trial results) is that any unvectored lines can be catastrophic to the overall improvement; this material seems convincing. If you know different, you probably *should* explain why.

    • Roberto

      I am not aware of any expenditure being used towards any EU guideline of 100Mb. Care to link to that information?

      NO BDUK money is being used to provide 100Mbps to anyone. Again your source on that??

      As to meeting the 30Mbps for all vectoring is not needed for that, in fact for the very noisiest of lines it will not help as vectoring can only cancel out a certain amount of noise. It either works or does not.

    • MikeW

      Oh it’s you again.

      For a county that specifically indicates they are using BDUK funds as a stepping stone to the EU goals, look at the Coventry/Solihull/Warwickshire project. The June newsletter covers it in laymans terms, while their local broadband plan and public consultation documents detail their intent clearly. You need google to find those documents, as they aren’t directly linked on the website any more.

      As for BDUK funds providing 100Mbps… Well, obviously there is nothing that uses FTTC that provides that, yet. However, FTTP is a known component of the BDUK projects, and certainly will provide 100Mbps and more. There are exchanges listed in the Openreach where/when as getting FTTP within BDUK projects – for example in North Yorkshire alone: Ampleforth, North Grimston, West Heslerton, West Lutton, Airton, Brandsby, Burythorpe, Constable Burton, Coverdale, Cracoe, Horton in Ribblesdale, Lastingham, and Sawley. There are plenty more in Lancashire and Wales.

      And whatever your opinion about how much noise vectoring is capable of removing, the fact is that vectoring does increase the range at which 30Mbps can be achieved. Anything that extends the range of SFBB speeds from a cabinet is automatically going to bring higher coverage figures – at least where there are people who live longer distances away from their cabinet.

      As various BDUK county contracts have been signed, the indications are that between 2% and 6% of their premises will be connected to fibre cabinets, but won’t get SFBB speeds. The extended range that vectoring brings will bring some of those premises within the SFBB threshold.

    • Roberto

      So the short version is NO the money concerned is not being used for any 100Mb product.

    • MikeW

      No, the short answer is YES.

      Try reading what I wrote. You might find it educational.

    • Roberto

      The money in your link is for a 24+Mb service can you provide a link to where money is being spent on a 100Mb service or not?

    • MikeW

      Simple. Any exchange being upgraded as part of BDUK that results in deployment of FTTP. Plain native, not on-demand, just ordinary FTTP. Guaranteed availability of 100Mbps to the fortunate few.

      You don’t need any link other than the one to Openreach’s standard where-and-when website to see the plans.

      Of course, you have to do a bit more research and thinking (more things that aren’t your strong points) to work out which exchanges are only there because of BDUK plans. That’s pretty easy in the case of North Yorkshire, where the commercial plans ran out at 18 exchanges, and the BDUK plans take more than 100 exchanges. Even a buffoon could work it out. Can you?

    • Roberto

      So no then you can not provide a link to where funding is for a 100Mb service. That is what i thought.

    • MikeW

      I see you couldn’t reach the heights of being a buffoon then.

      Thanks for confirming though.

    • Robert

      I see you still can not provide a link to where funding is for a 100Mb service.

      As to the name calling. If you had any intelligence to question I would have questioned it already.

    • ascii

      If bt charge end user for this then it’s a scam and needs putting to oftel . Bt charging customers to fix it’s own problems is a scam end of.

  2. X66yh

    The outdated infrastructure is a result of virtually nil investment in it while the telephone system was in public ownership. Additionally rather than use copper in the 1970’s the GPO as they then were used aluminium ‘cos it was cheap at the time….again because HMG prefered to grab and then waste the profits from telephones on their own pet schemes.

    BT should be congratulated for finding ever more innovative ways of making legacy copper/aluminium installed when the concept of broadband was not even invented work in today’s environment.

    Perhaps all the moaners might consider when working on their houses etc that they install infrastructure suitable for the technological advances of the next 40 years……..

    • Anoyed Tax Payer

      If BT were being innovative, why then did they not include Vectoring compatibility in the first place like other countries did?

      I’m all for FTTC, but it should be treated as a stepping stone technology rather than the future, therefore investments into it should be light and not heavy, i.e Best to use what money is available to get as many people on FTTC as possible (Even if it’s a slow speed to start with) and slowly convert the whole network to 100% fibre.

    • MikeW

      It was impossible to include vectoring in live equipment in 2010.

      The vectoring process was only standardised in 2010, and the manufacturers are still running interworking plugfests to verify compatibility (the first of these, between chipsets, was only 8 months ago).

      I agree that FTTC should be seen as stepping-stone technology, but vectoring is a very worthwhile addition. FTTC (with and without vectoring) is still an “up to” technology, but vectoring makes it so much more stable and predictable.

      The range improvements are probably the best aspect, even if the focus is one the headline speed aspect.

  3. Kyle

    Having had to move back to ADSL2+ from VDSL, I would support maybe a pound or two extra per month to enable this to work.

    I’m not sure what the quality/length of lines is like in the trial areas, but being out in the sticks (as are a lot of properties in the village), I hope that the trials are making full use of the problems that exist out there and attacking those cabinets with long/poor lines.

    It will be very interesting to see how this works, but I’m glad this is not being touted as a speed increase when there are those who already benefit from a fibre-enabled cabinet and still can’t make full use of the VDSL speeds available.

  4. MikeW

    A quick mention on the cost…

    At worst, the cost of upgrading the cabinets is restricted to the plug-in shelf of linecards. This turns out to be a small number of £thousand per cabinet, not tens of thousands.

    IIRC, the recent talk-talk document that attempted to justify their complaint to Ofcom had a working life estimate of these linecards as 8 years. If true, the upgrades may be seen as an end-of-life replacement!

    And of course, with 10-20% takeup rates, not many of the cabinets will be fully populated yet.

    • Darren

      BT like to avoid causing downtime so I’m just thinking how they would go about swapping out linecards.

      If downtime is unavoidable then it needs doing sooner rather than later to minimise disruption.

      If downtime is avoidable providing there is spare linecard space then again sooner the better.

      Given the low uptake I don’t think they will be waiting for cards to go EOL, at least I hope. The speed enabler angle suggest to me they want it rolled out sooner rather than later to aid takeup.

    • MikeW

      BT are required to minimise and avoid downtime in voice calls – particularly for emergency service coverage – but are less worried about minimising downtime for broadband. One of my lines (non-fibre) was down recently for 2 hours, while BT did some work in the middle of the night.

      With that in mind, the cutover procedure will have to put in place a means of keeping the voice connection to the exchange, while the linecards themselves are removed and replaced – probably with a change in the steel cage that holds them too (the backplane inside the cage almost certainly needs changing to allow system-wide vectoring to happen).

      My guess would be that they would either insert some temporary connections in the Krone strips that terminate the tie cables, or they’d have some dummy connectors that the plugs (when taken off the real linecards) can be put onto, re-making the physical connection back to the exchange.

      Here’s the inside of a cabinet:
      http://beusergroup.co.uk/technotes/images/e/ee/PICT8304.JPG

      The part that likely needs replacing is the hardware with the flat white fronts. This currently has 2 linecards in place with 2 (hidden) plugs attaching the lines. There is space for 2 more line cards, and the plugs for those are in a pink plastic bag top-centre of the cabinet.

    • Darren

      The phoneline doesn’t route via the FTTC cab, the tie cable from the FTTC cab is jumpered onto the phone line in the copper PCP. So they wouldn’t need to do anything to keep the POTS online.

      They may not be required to minimise downtime on broadband but they will still want to wherever possible.

    • MikeW

      The phone line does route via the FTTC cab.

      One set of tie cables take the (voice only) lines back to the exchange, jumpered appropriately in the PCP. These lines incorporate an additional filter (similar to the ones we have on the NTE in the home) that prevent signals from an exchange-based ADSL modems from interfering with the VDSL modems in the cabinet.

      The other set of tie cables take the line towards the house, jumpered in the PCP too.

      Take a look at this PDF: http://www.bcs.org/upload/pdf/sfisher-090311.pdf
      Page 13.

      Or this one: http://www.niccstandards.org.uk/meetings/003%20-%20George%20Williamson.pdf
      Page 9.

      Or this one: http://www.openreach.co.uk/orpg/home/products/super-fastfibreaccess/fibretothecabinet/fttc/downloads/GEA_FTTC_4.pdf
      This diagram shows the separate jumpers in the PCP, but has lost the tie cables between PCP and DSLAM entirely (in either direction). Shame…

    • MikeW

      Oh – I agree that they want to minimise broadband downtime.

      It is just that they find a couple of hours to be acceptable for data, while for voice the acceptable level is measured in seconds.

      The voice network is massively over-engineered to provide this, while the data network is distinctly “best-effort”.

    • Darren

      Thanks for those references – Happy to be proven wrong.

      In that case it’s definitely a sooner the better situation upgrade wise. Hopefully thought was given to swapping cards in the design so it’s not too much of a PITA.

    • MikeW

      @Darren

      Adding to the previous stuff, I found another way that vectoring will have an effect on downstream speeds… “PHY layer retransmission”.

      Obviously vectoring removes the noise caused by FEXT induced by other lines, but doesn’t remove noise from other sources (including exchange-based ADSL). However, it seems likely that the remaining noise is going to be of less significance.

      That means there is less need to rely on FEC/interleaving as an error-correction technique – which regains 20%ish of capacity to the lines that get afflicted by this choice from DLM.

      But there is an alternative option, known as “PHYR” (PHY layer retransmission), from G.998.4. Using this means that today’s CRC errors will be retransmitted, rather than protected using FEC.

      I haven’t checked enough, but I suspect that PHYR works well with light levels of errors. If the error rate got bad enough, I suspect that the heavier FEC/interleaving settings would still need to be turned on by DLM.

      Again, this is not a gain in speed over the theoretical best single line, but it seems that vectoring makes it more likely that this “light touch” can be used instead.

    • MikeW

      Oops. That last post was aimed at @DTMark, in a later section of comments.

  5. DTMark

    Assuming that the length and quality of a line dictates the maximum attainable rate – for example, a 1km line which might have an attainable downstream rate of somewhere between say 12Mbps and 35Mbps depending on what metal it is and what condition it is in:

    Assume one line of this length is provisioned for FTTC and it manages somewhere in the middle – say 25Mbps.

    Crosstalk then drags that down to say 15Mbps.

    How does vectoring manage to make that line attain *more than* the original 25Mbps defined by the laws of physics down that cable?

    • MikeW

      Vectoring can’t get more speed than could be originally attained for a single line, if it is the only change. Vectoring alone can only remove the effects of crosstalk caused by other lines.

      This means it gives results that are more consistent with distance (or attenuation), so allows the predictions to be higher and more accurate.

      But other changes can be applied on top of vectoring that can make further gains possible.

      One change is in the upstream direction, where vectoring allows the power back-off mechanism to be reduced/removed. This allows upstream to run faster, especially on shorter lines.

      For downstream, the reduction in interference probably means that the use of spectrum and the transmit power can be changed: Higher power for the longer lines would allow higher speeds, but would ordinarily cause more crosstalk, so would not be a solution. With vectoring in place, it can at least be considered and trialled.

    • DTMark

      That was my interpretation.

      So it isn’t correct to say that vectoring helps ISPs deliver “better” FTTC speeds.

      Rather, it might help ISPs deliver speeds that are less bad than they would have been e.g. for someone who got 70Meg when they signed up only to see that drop to 50Meg (have seen quite a few posts like this, those would be very short lines, so crosstalk certainly isn’t a “long line” issue), they might see the 70Meg they were actually estimated to get in the first place.

      Was all of this taken into consideration when proposing that FTTC could potentially deliver “superfast broadband” to 90% of people?

    • FibreFred

      Crosstalk has always been known about. In your example 50Meg would still be classed as superfast?

    • MikeW

      Exactly.

      But remember, only the first line on each cabinet gets to experience that “no crosstalk” environment, and then only for a matter of days perhaps. Everyone else afterwards only ever sees a connection with some amount of crosstalk included (the same thing happens on/between ADSL lines from the exchange, but fewer people know or talk about it).

      But note that the estimates already take crosstalk into account. I saw one set of “FTTx working group” slides (dated a couple of years ago) that said they were aiming at including crosstalk effects for the next 18-24 months.

      So someone who got 70 when they signed up, and saw it drop to 50, ought to have had an estimate of 50 or lower. My estimate is 53, but I’m actually on 80. The original attainable figure from the modem was over 90, but it is currently hovering around 75.

      So the answer to your final question is yes – the effect of crosstalk was known at the outset.

  6. Trevor Harris

    I think I would prefer the money to be spent on FTTP. I am getting 22Mb/s and so I would benifit from vectoring but with FTTP I would get up to 1Gb/s. BT could then recoup some money by selling all that expensive copper laying in the ground. The system should be much more reliable so reducing maintenance costs. The Government and BT need to be more ambitious.

  7. Excellent post. I will be experiencing many of these
    issues as well..

IMPORTANT: Javascript must be enabled to post (most browsers do this automatically). On mobile devices you may need to load the page in 'Desktop' mode to comment.


Comments RSS Feed

* Your comment might NOT appear immediately (the site cache re-syncs periodically) *
* Comments that break site rules, SPAM, TROLL or post via fake IP/anon proxy servers may be blocked *
Promotion
Cheapest Superfast ISPs
  • Sky Broadband £0.00 (*20.00)
    * Speed: 38Mbps - Unlimited
    * Gift: First 6 Month Discount
  • PlusNet £3.99 (*14.99)
    * Speed: 38Mbps - Unlimited
    * Gift: None
  • BT £5.00 (*15.00)
    * Speed: 38Mbps - 20GB
    * Gift: £50 Sainsburys Voucher
  • TalkTalk £6.75 (*13.50)
    * Speed: 38Mbps - Unlimited
    * Gift: £120 Love2shop Voucher
  • Virgin Media £10.00 (*15.50)
    * Speed: 50Mbps - Unlimited (FUP)
    * Gift: None
Poll
* Javascript must be ON to vote *
The Top 20 Category Tags
New Forum Topics
Promotion

Copyright © 1999 to Present - ISPreview.co.uk - All Rights Reserved (Terms, Privacy and Cookie Policy, Links (.), Website Rules)