Openreach Expand Large Scale Trial of SRA on FTTC Broadband Lines
Openreach (BT) has today announced that they will conduct a “large scale” UK trial of Seamless Rate Adaption (SRA) technology on Fibre-to-the-Cabinet (FTTC / VDSL2) based superfast broadband ISP lines (40-80Mbps), which could improve service stability by varying the speed more effectively.
The technical way of explaining SRA is to say that it can be used to “reconfigure the total data rate by modifying the framing parameters and the bits and fine gains parameters.” If the noise condition (interference) improves then SRA can be used to gradually increase the data rate again, potentially taking it all the way back up to what you had before a nasty bit of sudden noise impacted the line.
SRA is technically nothing new to Openreach’s hybrid fibre network and it’s a standard part of the latest G.fast standard. Prior to that the operator has also previously tested it alongside the problematic VDSL2 Vectoring technology (2014 to 2015) and they recently tried to incorporate it as part of their Long Reach VDSL tests (here), which were shelved.
Earlier this year the operator decided that it might be better to try implementing SRA on its own and so they setup a small Proof of Concept (PoC) trial with 200 lines via Huawei based street cabinets (here). The service was applied gradually (line-by-line) and Dynamic Line Management (DLM) was also disabled. This ran for several months and after that Openreach began reviewing its impact, which brings us back to today.
This briefing is to inform all CPs that following the successful proof of concept earlier this year we will be trialling Seamless Rate Adaption (SRA) on a larger volume of lines
We’ve published a briefing for CPs relating to SRA, which will vary the speed of a line as noise margin changes reducing the propensity to retrain, whereas today on NGA1, if the noise margin drops beyond a certain threshold, the line will retrain. Openreach will be running a large-scale trial to prove the benefits of this change.
Sadly the briefing doesn’t say anything more than that, but we understand from our sources that the larger trial is likely to target around 100,000 lines (at least that was the original plan) and will probably run until around the start of Spring 2019.
As usual a big question mark sits over whether or not the larger trial will include any ECI cabinets, which have a history of being quite problematic (unlike Huawei’s otherwise dependable kit).
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on X (Twitter), Mastodon, Facebook and Linkedin.
« JT Picks ZTE to Deploy 5G Mobile on the British Channel Islands
5G Mobile Trial in Japan Delivers Broadband Speeds of 27Gbps »
Latest UK ISP News
- FTTP (5515)
- BT (3514)
- Politics (2537)
- Openreach (2297)
- Business (2262)
- Building Digital UK (2244)
- FTTC (2043)
- Mobile Broadband (1973)
- Statistics (1788)
- 4G (1664)
- Virgin Media (1619)
- Ofcom Regulation (1461)
- Fibre Optic (1395)
- Wireless Internet (1389)
- FTTH (1381)
My gran would have called this ‘getting blood out of a stone’, but the superfarce rocks on with our incumbent determined to keep the legacy phone network going as long as it possibly can. Thrift I can appreciate, but as long as they describe it as such and not call it fibre broadband then that’s fine, but it still isn’t doing much for our digital future if all we are gonna get is tarted up phone line broadband from obsolete cabinets.
There are over 7 million users of FTTC across about 95% of all premises in the UK. Improving the reliability of their services while simultaneously deploying FTTP is a worthwhile enterprise.
This activity isn’t taking manpower or resources from FTTP rollout. It will still take years to cover the majority of the UK with FTTP and things like this in the interim are important and useful.
I guess you’d be happier if there were still 50%+ of the UK unable to get anything better than ADSL? Had Openreach done zero FTTC they’d likely still be less than half way through the country with FTTP.
My Gran would have said, Great at blowing their own fibre. Even better at blowing their own trumpet.
When are you going to sort out Northern Ireland the Internet is crap here
The rollout has already started in northern Ireland
The main problem I see with FTTC was it was oversold to the public and politicians alike. People were promised upgrades that never materialised due to its limited range. According to the stats 20% of my area still does have access to ‘superfast’ speeds and many can’t even get ADSL2+
So although the idea of these changes are worthy, it will seem to many like simply rearranging the deckchairs.
yet F*** g.fast – nice
if you think 40-80Mbps (4-7 MB/s) is “superfast broadband” you need to see a doctor
want to remind anyone of the speeds in countries like Sweden ? 1000 Mbps +
I have lived and experienced the first boom of fiberoptic Sweden. It was very exciting and big money was spent to setup the infrastructure early 2000. It is the lack of investment in infrastructure in this country which is problematic.
When we write “superfast broadband” it’s done with reference to the country’s own definition, which is 24Mbps+ via the UK Government or 30Mbps+ via Ofcom/EU and recent BDUK contracts. We just follow those definitions to remain consistent.
It would be nice if they actually sorted out the ECI issue, G.INP, as a priority, before trying out other ‘features’ that could potentially improve VDSL2/FTTC. If SRA does end up being rolled out to ECI, which I doubt, I wonder if that will also experience technical issues and be reverted? 😛
Just get on with FTTP already.
Exactly. The figures shown by by as covered by fast broadband or even fttc is a barefaced lie.
Myself and many many people in the heart of London don’t get anymore than a guaranteed 4.5mb/s. Yes 4.5. BT tell you your ISP is lying and that the exchange is enabled and handed over, your ISP tells you the exchange is enabled but the ports don’t work. Then you investigate further and find out BT wants you to do a community funding. Paah
@Bernie
If you’re suggesting that whole exchange areas that are shown as having fibre available, whether FTTC or FTTP, in fact have none then please list them here. Or are you talking about the exchange being fibre enabled and your home or office not being covered, which is a very different thing?
As to 4.5Mbps in central London, the laws of physics apply there too, as does basic economics. Why should it be different just because you live or work in the centre of a large city?
Anything OR can do to improve FTTC service while it exists should be welcomed.
However I am still surprised how many homes I visit on either ADSL or VDSL that have not sought to maximise their speed. My father in-law recently had Sky Q installed and up to this point never needed broadband. The Sky installer has simply inserted a filter into their extension telephone socket without any regard to the historic telephone wiring in the house and its routing from outside. No testing and no attempt to place the Router near the external line. It worked so job done.
Whether asking a friend or professional there is sizeable percentage of DSL users that could improve their speeds.
“Anything OR can do to improve FTTC service while it exists should be welcomed.”
Given the fact they tested SRA on ADSL for years on and off and this is not the first time they have also fiddled with it for FTTC im pretty confident in saying ill eat my hat if it is even rolled out nationally.
Must of been a slow PR day… Next months new will be them trying G.INP on ECI kit for about the 6th time.
Millions of users will be on FTTC for the foreseeable. Whether it is upgrading firmware, changing settings or “copper rearrangement” it is good that OR are still looking at ways of improvement. This was not PR. This is a technical note to ISPs out of courtesy. The way I read it is they are constantly working with both ECI and Huawei, they then test changes in the labs and then need to trial on real pairs to see what the net affects are to a cabinet population. It may actually solve a problem on some lines but have an adverse effect on others.
I prefer this rather than OR simply excepting limitations of the telephony network or previous purchasing decisions such as ECI.
Talk about what goes around comes around. The same conversations and PR I’m sure were about with ADSL way back when, but ADSL rushed to market and SRA became an untested afterthought, and so was never used outside of a few select ISPs with their own kit. When I tried it on my line with an LLU ISP it didn’t achieve anything really, disconnect rates were no better, perhaps worse.
Also it was more appropriate with ADSL given the much longer line lengths and so varying noise picked up by those longer wires, but VDSL is running over shorter distances and if it is disconnecting for a customer every few hours because of swings in noise, surely the source of the noise should be tracked down and not papered over with SRA.
The main problem with introducing these extra bits of the spec so late in the day rather than insisting they are supported from day one is that no modem manufacturer up until then has had to worry about bullet proof support on the UK cabinets for SRA, so SRA will be troublesome because it is untested and been ignored, with many modems needing firmware updates that have long been dropped and forgotten about by the manufacturer. We’ve seen exactly this with G.INP and Vectoring and ECI kit unable to work with it.
These sorts of trials come down to PR only really. Why doesn’t BT lift the sync rate and go for VDSL MAX, i.e. let lines sync up to the maximum they can achieve without being capped at 80/20. A lot of people could see 100/30, but of course that’s not going to happen as it makes G.Fast a lot less attractive for those same people getting the speed boost.
It was issued as a technical briefing to customers – it’s not a press release for the media or public that’s published on the website, so hardly PR.
Openreach have given up on ECI. There will be uogrades/improvements to ECI cabinets. If anything, they will start replacing them with Huawei cabinets while they expand FTTP coverage.
My profile on fttc is stuck, wish they could reset it!
People have been asking for a reset for years.
Whilst its accepted that bthere’s lack of investment in this country, there would appear to be no lack of increased and exorbitant charges, even for the most modest improvements of capability.
According to that Dutch company (Featured in a video linked to one thread on this site), the roll-out cost of G.Fast systems to infrastructure owners over a ten-year period averages out to about 17 Euro per household.
Whereas, I understand that some people who are already receiving the only the first stage roll-out of G.Fast (There are four) from BT are already paying £17.50 EXTRA PER MONTH over current VDSL broadband charges.
I presume that the mid-morning aerobics session in the BT Chairman’s Office goes something like this:-
https://www.youtube.com/watch?v=kBBhQRWbAtA
Correzione: That G.Fast roll-out cost was 17 Euros, per household, PER YEAR.
Don Corleone says that there is no truth in the rumour that there are benzodiazpines in the OFCOM office coffee supply.
So the extra bandwidth required to ensure customers can burst end to end at 300Mb rather than 76Mb is free?
No 1. Will the average customer notice the difference ? The real pay-off is at the provider end, giving servers more free-time and producing a much better return on capital employed. Envitably, that’s going to mean more pop-up annoyance adverts on web-sites for crap that the embedded AI marketing tools have decided that you want.
No. 2 Even Talk Talk are reported as charging £15 a month for the new service. That means that in every 2.5 year implemnetation period the ISP’s will be making 29 months pure profit – that’s 2900% profit putting the returns on Pay Day loans in the shadows. Its even worse when you consider that the audience for G. Fast will be bigger than for Pay Day loans.
No.3 There’s something really sic and wrong with a corporate sector that requires that level of profit, because it isn’t required to payback any investment the ISPs have made and its ultimate effect is to curtails economic growth in other areas. Its typical of unmediated capitalism action, which gives us “Roller-coaster” / “Crash and Burn economics – good for the very few shareholders, bad for the many.
No.4. Give it another 10 years and we will have another “Authority”, probably a robotic one (Successor to the Maybot ?) telling us that the crash in the telecomms industry couldn’t have been anticipated and that they’ll be lessons to be learned. When, every decade in the UK, haven’t you heard that phrase being used ?
No. 5 And who would have thought that OFCOM would be permamnently asleep at the wheel ?
Perhaps they could wake-up and stipulate that a portion of this excess profit be used to substantially accelerate the installation of broadband cabling to all corners of the UK?
Answering.
#1 – No-one is going to be forced to take G.fast. Not an issue.
#2 – TalkTalk don’t build the G.fast infrastructure. The cost of the initial build is irrelevant to their retail pricing. Their charging more reflects their increased backhaul costs and rental costs from Openreach which makes this entire point irrelevant.
#3 – I’ve no idea what you’re talking about. Broadband provision is very competitive in the UK. A number of operators wish they could charge more.
#4 – Doesn’t really have much to answer.
#5 – Ofcom have a lot to answer for – this isn’t one of those things.
You seem to have watched that one presentation but not actually understand the issues. Openreach don’t get the equipment for free, it costs money to purchase and install it so the 17 Euros per year makes no sense. The actual cost to install the things was I believe around the 100GBP per premises passed mark.
You also ignore that Openreach have reigned back on G.fast and they and everyone else is now pursuing full fibre with far more enthusiasm. No-one considers G.fast anything other than a stopgap and I strongly suspect Openreach regret pursuing the technology as they have.
Other than those an excellent rant.
Correzione: Point 2; That should be 27.5 months pure profit. So that’s only 2750% rate of return on capital employed.
Addendum:
No:6 Whilst all this is going on, the UK broadband system will still be reliant for operation on 70-100 year old copper cables, largely unrenewed/refreshed/remanufactured from their original date of manufacture. How much longer is the insulation on these cables going to remain functional. You can see the disaster waiting in the wings, when the whole of the UK is totally dependent on this ancient system and it starts to go phut. Just like the “Banking crisis”, the taxpayer will pick-up the bill.
The excess profits being made under the proposed charging schema should be devoted to renewal of the system and replacement of copper with fibre optic. Instead all this money is going to shareholders and perks for the corporaty.
@NR – the UK network is based on fibre and is not reliant on copper in that way. When copper is in place for local ends it will be replaced in time. No disaster waiting, unless you know more.
@TheFacts: The UK access networks are mainly based on copper, not fibre, unlike the backbones. And there is no prospect of a large-scale copper switch-off with fibre replacement in the foreseeable future. The big looser will be BT.
It will be interesting to see if the Landline fraternity can maintain this stance when 5G really takes off – though I notice that the Dutch Boy’s presentation made reference to incorporating wireless as a “Relief” path for excess data usage in landlines. Isn’t that just re-enforcing the prospective landline strategic vulnerability i.e. throwing good money after bad !
Surely with less callout costs wouldn’t it be better to sell off the copper replacing with fibre and have routers or household fibre terminal blocks with phone connectors or routers with phone connectors as the new bt fttp hub.