Forgot your password?
typodupeerror
Networking The Internet Upgrades IT News

Welsh Scientists Radically Increase Fiber Broadband Speeds With COTS Parts 72

Posted by timothy
from the believe-it-when-it-appears-in-your-home dept.
Mark.JUK writes "Scientists working under an EU funded (3 Million Euros) project out of Bangor University in Wales (United Kingdom) have developed a commercially-exploitable way of boosting broadband speeds over end-user fibre optic lines by using Optical Orthogonal Frequency Division Multiplexing (OOFDM) technology, which splits a laser down to multiple different optical frequencies (each of which can be used to carry data), and low-cost off-the-shelf components. The scientists claim that their solution has the ability to 'increase broadband transmission by up to two thousand times the current speed and capacity' (most UK Fibre-to-the-Home or similar services currently offer less than 100 Megabits per second) and it can do this alongside a 'significant reduction in electrical power consumption.'"
This discussion has been archived. No new comments can be posted.

Welsh Scientists Radically Increase Fiber Broadband Speeds With COTS Parts

Comments Filter:
  • by Anonymous Coward on Tuesday November 06, 2012 @10:53AM (#41894165)

    ... BT are bloody useless!

    • by L4t3r4lu5 (1216702) on Tuesday November 06, 2012 @10:58AM (#41894253)
      It's not incompetence in this instance; It is actually malice. BT would much rather hold on to this tech for the next 15 years, squeezing an extra few pounds per month out of you for the next tier of service, right up until you're paying more for your internet connection than you are for your mortgage.

      Consider; The identical fibre with this new tech is all of a sudden 2000x times less efficient than it could be. Do you think you'll be charged 1/2000 of the current rate if it's implemented and you elect not to use it?

      (I realise there is more to this, like switching overhead, backbone speed, contention etc).
      • by jeffmeden (135043)

        Would you really be that interested in a 1000x increase in last mile speed, even if it meant that NONE of your actual applications (except maybe bittorrent) were going to go any faster at all? I would rather see them invest in the backbone than trying to out-do a 100mbit last mile which is probably pretty freaking hard to saturate as it is.

        Just sayin'

        • by Anonymous Coward
          Ah, but no-one in Britain has a 100Mbit/s connection. I'm on 1Mbit/s, and there are many on worse connections than that here. We definitely need a drastic improvement in the last mile, and it needs to be properly future-proof otherwise it would be a mostly wasted initiative. 1Gbit/s and above, fibre to the premises.
        • by evilviper (135110)

          I would be immensely interested in high-speed last mile because:

          Two customers of the same ISP could transfer files at ridiculous speeds. Bittorrent and other P2P services would automatically take advantage of this, upon seeing one peer with insane bandwidth. The technically inclined would make good use of this, storing backups off-site once ample last-mile bandwidth is there.

          Edge-network caching services like Akami would now mean many popular websites will be super-fast, not just slightly lower latency...

      • by Kjella (173770)

        Well, maybe. But I saw a documentary on when McDonald's started super sizing meals (no, not "Supersize me") and when you first had the store, staff, equipment, procurement, cleaning etc. delivering extra fries actually cost them very little. I imagine it's quite the same for an ISP, to take my own as an example for 22% more in cost I get 140% more bandwidth compared to the tier below mine. So if delivering super fast broadband is dirt cheap they'll want to push me to another crazy fast tier for money I didn

  • by shoppa (464619) on Tuesday November 06, 2012 @10:56AM (#41894217)
    I don't know if things are better in the UK, but here in the US the bottleneck for fiber-to-end-user is rarely the link from CO to end-user. The bottleneck is aggregate traffic capacity from CO to the backbones, an amount that has to be shared among all users. Giving individual end users more capacity to the CO sounds like it would make the current bottleneck even more apparent.
    • by Anonymous Coward on Tuesday November 06, 2012 @11:03AM (#41894289)

      What is this 'fiber' you speak of? Sounds like an interesting tech.

    • TFA is pretty useless and doesn't indicate what sorts of fiber this works on, or why it is different from other OOFDM-related work; but is there any reason to suspect that a technology that improves fiber transmit rates wouldn't help the CO backbone link speed as well?

      Given the, um, vigorous state of competition in the broadband market, it isn't clear that that will matter much; but if they have some new secret sauce that makes transmissions over fiber faster it would, naively, seem to be something that co

      • by jeffmeden (135043)

        TFA is pretty useless and doesn't indicate what sorts of fiber this works on, or why it is different from other OOFDM-related work; but is there any reason to suspect that a technology that improves fiber transmit rates wouldn't help the CO backbone link speed as well?

        Given the, um, vigorous state of competition in the broadband market, it isn't clear that that will matter much; but if they have some new secret sauce that makes transmissions over fiber faster it would, naively, seem to be something that could be added to any part of the network carried over fiber.

        Maybe it won't because tossing in a router that is capable of processing 1000x more packets is NOT going to happen with "COTS" parts? Fiber is only as fast as the hardware on either end. These are little strands of glass barely wide enough to feel, if doubling/tripling/1000x'ing bandwidth were as simple as tossing a few more in the trench don't you think they would have done that already? Gracefully processing the light at either end is the hard part.

        • by Bengie (1121981)
          They shifted the bottle-neck. A positive multi-magnitude change in any bottle-neck is warmly welcomed.
    • by BeanThere (28381)

      I suspect that if the technology is really 'commercially viable', it could also make it cheaper to upgrade backbone links too.

      • by AlecC (512609)

        It would certainly help if the tale I was spun about my slow broadband was true. According to the droid I spoke to. there was no point in installing more equipment into my local exchange because they were bandwidth limited on the link from that relatively local exchange to the main backbone, and it would mean laying new fibre, which would take a long time and be very expensive. If they can suddenly speed up that link by 10x, let alone 2000x, then the cluster of villages served by that exchange will, in an i

    • by Anonymous Coward

      Well, it means they can charge you the same for using less fiber and power.

      So what they'll do is all new laid fiber to the endpoint will use this technology, inorder to save money, charge you the same, but they'll leave their 100kbps backbone, so that they can claim it's the pirates clogging the pipes.

    • by Degats (1506137)

      I don't know if things are better in the UK, but here in the US the bottleneck for fiber-to-end-user is rarely the link from CO to end-user.

      Currently, for most places in the UK, the bottleneck for fibre-to-end-user is the copper cable between the end-user's house and the cabinet down the street where the fibre terminates.

  • by Anonymous Coward

    First Post!

  • by Bearhouse (1034238) on Tuesday November 06, 2012 @10:58AM (#41894257)

    Not sure ISPs and others would be keen in upgrading their infrastructure to make the theoretical speed really available to home users, sadly...

    • They might consider if it it actually does prove to have power cost reductions.
    • by BeanThere (28381)

      Not sure ISPs and others would be keen in upgrading their infrastructure to make the theoretical speed really available to home users

      I wonder how this [broadbandbreakfast.com] and this [techcrunch.com] happened then?

      Though USA needs to do away with regional monopolies/cartels.

      • Thanks for the input, get your point but was really just saying what others have subsequently done better - the theoretical speeds of 'broadband' are already often far in excess of the speed that you can actually download at.

        • by zlives (2009072)

          the issue is backbone connectivity/BW limit. if this works for backbone connections as well, then end users can finally atleast get what they are actually paying for already.

  • by vlm (69642) on Tuesday November 06, 2012 @11:20AM (#41894531)

    which splits a laser down to multiple different optical frequencies

    No no no thats just WDM for DWDM. Imagine a piece of glass fiber with prisms on each end and separate red, green, blue, etc lasers and detectors. They (can) operate completely independently. You can do the same thing with RF and NTSC signals... its call old fashioned analog cable TV.

    OOFDM is like hyper close packed DWDM and usually made out of different tech. Some games are played to eliminate ISI and crosstalk, assuming the gear is working properly, perfectly linear, etc. Maybe a cruddy analogy would be kinda like two voice signals in one DSB carrier, or another cruddy analogy is its plain ole DSL FDM except coordinated so the FDM slices don't/can't interfere with each other and the leading O means its optical.

    For RF this is "old" stuff like from the 90s. For optical this is pretty impressive and new. Same concept just a couple orders of magnitude higher frequency.

    The wikipedia article is not so bad

    http://en.wikipedia.org/wiki/Orthogonal_frequency-division_multiplexing [wikipedia.org]

    low-cost off-the-shelf components

    HA HA yeah maybe thats in the grant proposal as a goal, or its low cost compared to installing another length of fiber... Its not gonna be low cost as in I could do it in my basement using parts from an old laser printer, or you'll be buying a fiber "ethernet switch" using it for $9.95. It is probably going to be lower-cost compared to any previous design, which IS cool.

    • by Shatrat (855151)

      I suspect they are actually doing some kind of fiddling in the electrical domain and calling it FDM, while the laser is still either On Off Keying or maybe Phase Shift Keying. Since DP-QPSK transmitters and receivers still cost about as much as a luxury car they're hardly COTS.

  • by Hognoxious (631665) on Tuesday November 06, 2012 @11:47AM (#41894929) Homepage Journal

    Unfortunately Llanfairpwllgwyngyllgogerychwyrndrobwllllantysiliogogogoch exceeds the maximum packet size and causes the router to c*@
    ! n o
    c a r r i e r

    • by ccanucs (2529272)
      I can say that out loud though! :-) W.
  • by Anonymous Coward

    The speed of light is the speed of light.
    Would love to know how they made it faster :)

    Yes the effect is improved throughput - ie transfer rates or download/upload speeds, but the packet speed isn't improved at all.

    When we can introduce a photon into one end of a piece of fiber and have it instantaneously come out the other end, we'll have *speed* improvements.
    Until then, we're only increasing capacity.

    That is all. EOL

    • by Bengie (1121981)
      There are two ways to measure speed: 1) Bandwidth 2) Latency

      Because latency is fixed by light, then there is only one type of speed. I don't see the confusion
    • by tehcyder (746570)
      Right, so by your logic, all internet connections run at exactly the same speed, since light travels at a constant speed in the same medium.. When someone says that a 1Gb/s connection is "a thousand times faster" than a 1Mb/s connection, they're simply wrong.

      Makes you wonder what the point of so-called broadband is, doesn't it? We might as well have stuck with our 14.4 kbps modems.
  • by gstoddart (321705) on Tuesday November 06, 2012 @11:59AM (#41895109) Homepage

    But since this was done by a Welshman, nobody will be able to decipher the packets.

    I kid, I kid.

    • by ccanucs (2529272)
      Syr, yr hyn a ddywedwch yn debygol iawn wir!
      • by gstoddart (321705)

        LOL, took some googling, but I think I get "Sir, you're being a douche for sure".

        I can't imagine why the Welsh have a reputation for being indecipherable. :-P

        I met some Welsh guys on vacation once -- the younger guys I could follow, but the older guys might as well have been speaking Klingon their accents were so thick. Nice guys though.

  • First of all, here in the Netherlands the roll-out of fiber to the home seems to be moving at a snail's pace. My impression is that our local telco giant, KPN, who work together with Reggefiber to install fiber optic cabling, is only interested in doing this for new neighborhoods. I once asked what it might take to change their minds and was told that, if I was to survey my neighborhood (around 1,000 homes) and gather signatures from at least 40% who would be interested in such a connection, then they woul

  • throw something up on ehow at least, eh? I've got a private fiber to try this out on.

    poor little me doesn't have a copy of Nature Photonics from May 2011.

  • I thought Wales wasn't going to get the Internet until 2020: http://www.flickr.com/photos/samsmith/3619752897/ [flickr.com] (Text version here if that is easier for you: http://www.morningstarr.co.uk/forum/underworld/25573-internet-reach-wales-2020-a.html [morningstarr.co.uk])
    • by pedros (2555030)
      I'm suprised that the great Barry Chuckle would resort to such stereotypical rubbish.
  • There is a university in Wales.

What this country needs is a dime that will buy a good five-cent bagel.

Working...