BT Trials New Hollow Core UK Fibre Optic Cable at its Labs
BT has announced that they’ve begun trials of a new type of hollow core optical fibre cable at their BT Labs R&D facility in Adastral Park (Ipswich), which could in theory be more resistant to damage, while also delivering better performance (e.g. latency) for future broadband ISP and mobile connectivity services.
At present most conventional single-mode optical fibre cables work to guide laser light through solid glass cores, which need to be extremely transparent in order to avoid signal loss. Nevertheless, even the clearest of fibres will still have a slight loss over distance (akin to around 0.142 decibel per kilometre) and solid glass fibres struggle to carry very high-power transmissions, especially in short pulses.
Some years ago (2013 example) we began reporting on the development of a new type of optical fibre cable – hollow core fibre – that could put right some of these problems. Essentially, these cables have an air-filled central core (i.e. light travels faster through the air than glass), with an outer ring of glass to help guide the laser beam. Since then the technology has continued to be refined.
Advertisement
According to BT, such cables have the “potential to reduce the latency, or signal delay, caused by the light travelling through glass, by up to 50%.” Naturally, the operator is keen to trial this, and so they’ve teamed up with Lumenisity, a Southampton University spin out company, and Open Radio Access Network (O-RAN) mobile vendor Mavenir.
The trial involves a 10km long hollow core fibre cable provided by Lumenisity (CoreSmart® hollowcore). Working with Mavenir, BT has shown that using such cables can increase the distance between street antennas and the back-end processing in exchanges. Due to the low latencies, use of hollow core in the Radio Access Network (RAN) could also reduce mobile network costs by allowing more 5G antennas to be served from one exchange or cabinet. High-Frequency Trading (HFT) is another area that may benefit.
Professor Andrew Lord, BT’s Head of Optical Network Research, said:
“We’re excited to begin trialling hollow core fibre and to discover the potential opportunities and benefits of deploying this technology in certain scenarios. This new type of fibre cable could play an important role in the future of the world’s communications infrastructure, heralding a step-change in capability and speed, to keep up with the demands for high-speed, low latency communications driven by 5G networks, streaming, and more.”
Just to be absolutely clear, this does NOT mean that BT will in the near future task Openreach with having to replace all of today’s optical fibre cables. Indeed, we aren’t even close to maximising the capability of existing optical fibres in the local access network (e.g. those used for Fibre-to-the-Premises (FTTP) broadband), which may well exist for many decades or possibly longer.
However, we could see such hollow core fibres potentially being deployed as part of future / new core network links, or for providing backhaul style capacity over longer distances etc. The technology is still developing and so BT’s trial is all about testing different use cases and seeing how it performs.
Advertisement
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, BlueSky, Threads.net and Linkedin.
« Merged Broadband and Mobile Giant Virgin Media O2 Launches
CityFibre and Vodafone UK Extend ISP Partnership to Ethernet »
It would be an interesting upsell for those companies in the financial space (i.e. HFT firms) if BT could provide them lower latency routes compared to their competitors.
~15ms difference London to New York.
Even the stock market ain’t that quick.
Sorry 5ms.
Mmm, wonder how you splice these?
Easy, you just stick a hole in the middle to join them together 😉
No they will use tiny little nano bots which travel along the fibre and weld the fibres together from the inside, ensuring a perfect joint.
I’m sure I’ve seen that on TV somewhere 😉
Methods/tools to splice hollow core fibre already exist.
Hollow Core Fibre is not new but it is developing and will become the norm eventually. Lower latency and higher capacities are very possible with this fibre tech.
At the moment we have fibre to the cabinet and copper tails to properties. We are busy replacing the copper with fibre but eventually we will replace the fibre to the cabinets with hollow core fibre.
What we will call it then will be interesting. HCFTTC or HCFTTCFP?
It does not lead to higher capacities and certainly nothing useful for the home user. The lower latencies are so meaninglessly low (fractions of a millisecond) in the distribution network that unless you are a bottom feeding high frequency trader you absolutely don’t care.
It will be used in the long distance transmission network for sure if they can make it work, but fibre being put in today in Openreach’s and other networks to peoples home will never be replaced with hollow core fibre unless it’s because the fibre has failed.
What is the absolute latency difference?
I can’t imagine it will be relevant to consumer ineternet ping times unless the distances are vast.
Speed of light in regular fibre is about 2/3rd of speed of light. The best-case for hollow core is the speed of light.
For a 100km link, round-trip you are talking about a drop from 1ms to 0.67ms at best.
At best ? I think you’re overlooking imperfections of density in solid core fibre.
Indeed. Consumers generally don’t spend enough to make it worthwhile.
@JmJohnson that would be ordinary single mode fibre that can carry 1Tbps across an ocean. You can already buy 25Gbps BiDi SFP28’s today for those networks using point to point, and 25Gbps symmetric GPON is in the process of being standardised, and there is no technical reason why you could not go higher but at the moment there is no market for it to justify the development costs.
All hollow core fibre does is bring lower latency at a higher cost. For the fraction of a milli second you will save in the round trip between the exchange and your house, which is for the vast majority of people in the UK going to be under 40km round trip hollow core fibre is utterly pointless.
Would t it be better if thay just focus on getting full fiber out to everybody 1st same with virgin still waiting for my 1gbps with a pittful 50 upload
No, you must always look to develop something better. Thus technology will become very useful in backhaul and undersea cables. Imagine if they can get latency times from the UK to Australia to something akin to what UK to ROI is now.
No. No it wouldn’t.
The two are not mutually exclusive.
Roll out today but research for tomorrow! 🙂
@Danny given Australia is ~9,500 miles away, you’d struggle to get a round trip time below 100ms without exceeding the speed of light…
You think that telling all the research scientists to stop what they’re doing and start digging up the road instead would be beneficial? It seems quite to very unlikely.
So wait a minute……. the fibre cable that we are all “patiently” waiting for to be installed into our houses, is not THAT future proof then. 15-20 years time we’ll be waiting for our fibre cables to be replaced by hollow fibre cables?
I wonder how fast Elon would have pushed Starlink too by then?
The last two paragraphs answer that – this announcement isn’t about HOME connectivity.
But it’s important to remember that all man-made materials wear out over time, so it’s probably best to think about any new cable on a ‘generational’ time-span of change (i.e. 75 years or longer). Likewise, the bits of kit that go on either end of that cable will, from time to time, need replacement as they reach end-of-life. Nothing stays the same forever.
Apologies, I missed those!
Mark Jackson@ there is already single mode fibre in place that is over 30 years old all over the place (most universities will have a bunch of it). Glass itself has a very low rate of degradation, there are plenty of medieval windows out there to attest to that fact. The only thing that might degrade is the plastic sheathing and well I keep hearing in the news that plastic takes many decades if not hundreds of years to degrade. If it is not snapped there is going to be very little fibre installed today in the distribution network (aka from exchanges to peoples homes) that is replaced while people born today are still alive.
@Jonathan. Yep, that’s more or less what I was getting at :). I think when you buy the fibre it often comes attached to a guarantee of about 25 years (not sure if that’s still the case, perhaps an ISP can clarify), while other studies have suggested a lifespan of 75 years or much longer may be probable (depending on lots of different factors – temperature / environment, stretching etc.)? As you say, most or all people reading this will probably be dead by the time any fibre going into the ground today begins to fail naturally.
In all seriousness there are some major benefits to using HCF. Although speed/latency is one, it requires less power to transmit the light. There is also increased capacity. There are lots of applications for the real world.
In theory for instance you could reduce the amount of active equipment used in today’s FTTP architecture. But you will see more of this installed. It doesn’t actually cost that much more to make and with demand it will get down to costs similar to traditional fibre. It has been around for 20 years!
I think most people will be dead before they even get any fibre..
Dust gets inside fibre, through break or at splice time, how do you get it out? Dust, anybody? No? Dust?
(Little Britain)
You don’t if it can’t be sorted by a simple clean, you splice a new segment of fibre.
I can’t say in my experience this has been a big deal. When fibre breaks happen providers don’t just take the dirty, jagged ends and fuse them.
I do wish that bt and open reach will upgrade the fibre cabinet in Pengegon Camborne To full fiber to the premises and 70mg 28up load is rubbish its troubles all the time tr14 needs fttP and most residents will pay for that service to have the best
There is always one!!!!
Hmm. Lots of good reasons for that to happen but the neighbourhood being likely to have high uptake of the more expensive services doesn’t look like one.
https://www.cornwalllive.com/news/cornwall-news/inside-most-deprived-neighbourhood-cornwall-3500113
Think the town is on the lists, though.
That’s right, just you start work on another one when most of the country doesn’t even have FTTP with the original ones yet…
Why is 38Mbps called super fast? It is rather slow Compared to some other countries.
That hails from 2010 and the Government’s definition of 24Mbps+ being “superfast” (largely to reflect the move beyond ADSL), which was more recently tweaked to 30Mbps+ in order to match definitions from Ofcom and the EU. Obviously, by modern standards, there’s nothing “superfast” about such speeds, but that’s how they still communicate it.
Would be useful if data traveling over hollow fiber moved both faster AND safe from predator’s eavesdropping taps.
HFT users do not encrypt their data so a bit of intrinsically secure fiber that reduces latency is attractive!
It may not just be about latency. Normal single mode fibre is also constrained by attenuation and maximum power. Outside of very low latency applications the best use case may be to extend distances that you can go between regen sites (which also increase latency). I assume it will also use different optics (SFPs) as the standard wavelengths are used to avoid problematic dispersion in single/multi mode cables. I guess that’s the other area of development as specialist SFPs are easy to make, or cheap.
I did some up front research work on this for application in the defence industry back in 2007. One of the key challenges highlighted at that time was being able to connectorise easily and maintain cleanliness (of a hole !), Particularly for high power applications Have things developed since then in this area ?
Loving the complaints here.
Next will be complaints that CERN are using loads of scientists in their particle accelerator that should be fixing our lighting and appliances.
Forget this superconductor malarky, my fridge is sounding dicey.