FEATURE DWDM AND ROADM NETWORKS
involving in-line devices, such as optical amplifiers, should enable a modular approach, so customers can follow a ‘pay-as-you-grow’ model, lighting additional bands as needed. Currently, Coriant is working with its
partners to obtain the first results on wide spectrum transmission modelling. ‘Te next steps will be on the wide-band component design and network management,’ Napoli said. Coriant conjectures that 200Tb/s in metro areas and 150Tb/s in the longer haul could be realistic targets. ‘We are ramping up to say, OK, these are the possible capacities and the possible scenarios given these values and amplifiers in these bands,’ he said. ‘Our numbers might be conservative. In the metro area, to interconnect data centres, the capacity could be larger.’ An example of what’s possible comes
courtesy of Lawrence Livermore National Laboratory (LLNL) in California. Researchers stumbled onto potential E-band amplifier technology when working on neodymium- doped optical-fibre lasers, which lase at 1330nm, 1064nm and 920nm. Tey were designing a microstructured optical fibre to suppress emission at 1064nm and amplify 920nm emission, said Jay Dawson, deputy programme director at LLNL. ‘Tere was spontaneous amplified emission at 1440nm,’ he said. ‘Recognising that was unusual, we took it on.’
Having previously worked in optical
communications before taking on his current, defence-focussed role, Dawson recalled that previous attempts to produce amplifiers at this wavelength had faced big problems. In the 1980s, excited-state absorption in the 1330nm region had increased fibre loss when pump light was applied, severely limiting gain. ‘We got 20dB and 1W in an initial cladding pumped amplifier,’ Dawson underlined. However, this initial result was a high power fibre-laser result, not a telecom compatible amplifier.
Everything to gain Te LLNL team therefore designed a novel neodymium-doped microstructured optical fibre to use in a telecom amplifier, tailored to preferentially enhance E-band optical signal gain. Te fibre’s graded index inclusions direct wavelengths from 850nm-1150nm into surrounding cladding, Dawson explained. While excited-state absorption still prevents amplification at 1330nm, the approach can address the 1390nm to 1460nm range. Te resulting neodymium-doped fibre
amplifier is intentionally designed so that telecoms companies are comfortable with it. It
16 FIBRE SYSTEMS Issue 20 • Summer 2018
@fibresystemsmag |
www.fibre-systems.com
NIF & Photon Science postdoctoral researcher Leily Kiani tests a new optical fibre that could double the bandwidth of fibre-optic cables
could also at least double the proportion of light illuminating the low-loss window in glass fibre if commercialised. ‘I look at this and think it’s got the potential to have the form, fit and function of an EDFA and feel very similar to it in another band,’ Dawson says. ‘Te work we’ve done has shown that we can get neodymium to open up more bandwidth.’ However, LLNL’s defence focus means that it has little capacity to devote to support further NDFA development, and Dawson would welcome interest from other groups. ‘We’d be very happy to collaborate,’ he says, referring interested parties to LLNL’s Innovation and Partnerships Office. While such approaches work alongside
EDFAs, Aston’s Nick Doran is working on a technology that could replace it, known as parametric amplification. ‘We can anticipate very wide gain regions of maybe four times a standard EDFA,’ he said. Parametric amplifiers contain highly
non-linear undoped silica fibres, with very small cores, in order to exploit four-wave mixing, a standard non-linearity in optical fibres that limits transmission capacity. Aston researchers use this to combine a high-power pump laser and the signals that they want to
EDFAs themselves have gone as far as they can. I would think it would be surprising if there isn’t a replacement
amplify, outputting higher power signals. ‘Tere’s no energy lost and we can get 0db noise figures,’ Doran explains. However, he added that the ‘revolution, where they replace EDFAs, is some way down the line’.
Two fibre stages Te revolution will require parametric amplifiers to eliminate their current intrinsic sensitivity to polarisation. With optical communication today exploiting polarised multiplexed signals, a parametric amplifier must amplify all polarisations equally – but until recently, they didn’t. Te Aston team therefore tried separating the pump beam into two polarised components that enter the amplifying fibre loop in opposite directions. ‘We find that when we do high gain that doesn’t work, because of a stimulated Brillouin scattering interaction between the components,’ Doran said. Tey now use two fibre stages, one for the anticlockwise component, and one for the clockwise one. ‘Tat has eliminated polarisation issues,’ Doran said, enabling 15dB amplification from 1532nm-1550nm in a 4x75km system with 1.3dB OSNR penalty. All these new device designs show the desire
to expand what’s currently available. ‘It’s unthinkable that in 10 years’ time the only amplifiers we’ll have will be EDFAs,’ Doran says. But while the communications industry is currently working with what it knows, he thinks EDFAs’ days are numbered. ‘EDFAs themselves have gone as far as they can,’ Doran observed. ‘I would think it would be surprising if there isn’t a replacement.’l
Andy Extance is a science writer based in Exeter, UK
Jason Laurea
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44
