|
|
|
|
Australian breakthrough could deliver 1000 times the bandwidthSydney: Presentation at 11.45 Tuesday, 25 October 2005, Hilton Sydney, level 4, room 3.A small ‘smart’ scratch in a piece of glass is set to replace the electronic components in global fibre optics networks, letting us reach data speeds up to one thousand times greater than currently possible. Researchers at CUDOS, the Sydney-based Centre for Ultrahigh Bandwidth Devices for Optical Systems, announced their breakthrough today at an international optics conference - LEOS - in Sydney. They also announced a suite of other developments towards their ultimate goal – light-powered switching transistors and light-powered computers. Light in fibre-optic cables carries data at speeds and quantities far greater than the old copper phone lines – that’s why fibre-optics are the backbone of Australian communications and of the global internet. But every so often the light signals need to be cleaned up, amplified and retransmitted. Currently, we do that with electronic circuits which are much slower than light. These amplifiers are becoming a bottleneck on the internet. “Our discovery will clear this bottleneck” says Ben Eggleton, Director of CUDOS. “To get more bandwidth, we need new technology that’s faster than electrons – and light’s the fastest. We just need to build the components that can handle it. And now we’ve got one of the first problems solved,” he says. CUDOS have developed a small plate of glass with a carefully engineered scratch in it that can regenerate a new clean signal from an old noisy one. The glass is just placed in the light flow. The “smart” scratch first separates out the signal from the noise as the light is guided along it. The light then passes through a series of finely etched channels that recognise which bits are the separated signal and which bits are noise, and allow only the signal pulses past. Any noise is trapped by the channels. CUDOS have tested this device at speeds over ten times higher than the fastest cables in use in Australia. “We hope to see the technology implemented over the next decade by the major data carriers as new ultrahigh bandwidth communications systems are rolled out,” says Ben. With this looming traffic jam cleared, the next challenge for CUDOS is to build an optical switch - the light equivalent of the transistor switch. The invention of the transistor in 1947 triggered the electronic age – an Intel Pentium 4 computer chip can have up to one hundred million transistors. During the LEOS conference, CUDOS team members from the University of Sydney; Australia National University; Macquarie University; Swinburne University and the University of Technology, Sydney, will also be announcing some of the steps they’ve made towards a light switching ‘transistor’ and a photonic chip – the building block of ultra-fast light powered computers. LEOS is Lasers and Electro-Optics Society of the IEEE ((the worldwide Institute for Electrical and Electronic Engineers) Background information on CUDOS and conference abstracts are available online at www.scienceinpublic.com. For further information and interviews contact Ben Eggleton on 0413 385 715, Jacob O'Shaughnessy, Media Officer, University of Sydney, Tel: +61 (2) 9351 4312, mobile: 0421 617 861, jacob@media.usyd.edu.au or Niall Byrne on 0417 131 977, niall@scienceinpublic.com
Background Who’s in CUDOS? (and what’s in a name) CUDOS = the Centre for Ultrahigh Bandwidth Devices for Optical Systems. CUDOS is a research consortium between five Australian Universities: The University of Sydney, Macquarie University, University of Technology Sydney, Australian National University and Swinburne University of Technology. Funding comes from the Australian Research Council under the Centres of Excellence program, the five universities and from the NSW State Government. The Research Director is Professor Ben Eggleton, with Professor Yuri Kivshar as Deputy Director. In 2004 the Prime Minister awarded Ben Eggleton the Malcolm McIntosh Prize for Physical Scientist of the Year (link to media releases and background below).
Current issues with optical communications Most of our telecommunications backbones now run on fibre optics. Fibre optics carrying photons of light give us much greater speed and carrying capacity than electron-based transmission. The signal processing though is still done with electronics components, that clean and regenerate signals, reamplify and retransmit them, buffer data, and monitor the network for problems and faults. The electronics components are made with transistors, which we won’t be able to make much smaller or faster. But our demands for high speed transmission of data are still going up. So these signal processing components are becoming bottlenecks on the network. If we could go all-optical it would solve our current speed limitations. Current network speeds can reach up to 40 gigabits per second. CUDOS expect to develop components that will work at 160 gigabits per second and beyond. The physics of photonic interactions compared to electronic interactions means that light starts reaching its carrying capacity at a data speed about 10,000 times greater than electrons can manage. Some optical components have been built and demonstrated previously but they don’t offer the same advantages of size and manufacturability that a silicon integrated circuit chip does. CUDOS have built an effective signal regenerator that compares in size to a silicon chip, and are continuing to research other components to replace electronic network devices. They plan over five years to develop these concepts and miniaturise them into an integrated photonic chip that can replace the electronic silicon chips being used in our networks.
A signal regenerator needs to have some way of distinguishing the signal from the noise and then separating out the noise so that only the signal is passed on. The idea of a Mamyshev regenerator (what CUDOS have built) is that it uses a non-linear material to distinguish signal and noise. Non-linear because the stronger (more powerful) that a light pulse passing through is, the more its frequency is changed by the material. So as long as your signal is still louder than the noise, its frequency will be changed more. Put a band-pass filter on the end of this material to allow only your highly-changed frequencies through, and you get a clean signal with none of the noise. CUDOS found that chalcogenide glass – glass doped with sulphur, selenium and tellurium – worked as a non-linear separator. Signal pulses passing through an optical waveguide in the glass separated to a different, broader frequency range than the noise did. That gave them the first half of the regenerator. To create the band-pass filter, they etched a frequency-selective diffraction grating into the waveguide that would filter out any of the noise. The results showed a clean, accurate signal with the noise removed.
Being held 23 - 27 October 2005
Hilton Sydney, Sydney, Australia LEOS is the IEEE Lasers and Electro-Optics Society (IEEE is the Institute for Electrical and Electronic Engineers, worldwide) The papers CUDOS are presenting at LEOS are:
That paper’s conclusions: CUDOS has demonstrated experimentally the first integrated pulse regenerator using a non-linear waveguide followed by linear filtering through a double Bragg grating band-pass filter. They achieve good agreement between their theory and experimental results. Session details: Integrated All-Optical Chalcogenide Waveguide Pulse Regenerator: Experiment and Modeling V. G. Ta'eed, M. Shokooh-Saremi, CUDOS, University of Sydney, Sydney, NSW, Australia, L. Fu, D. Moss, M. Rochette, I. C. M. Littler, University of Sydney, Sydney, NSW, Australia, B. J. Eggleton, CUDOS, University of Sydney, NSW, Australia, Y. Ruan and B. Luther-Davies, Australian National University, Canberra, ACT, Australia We present an integrated, all-optical, chalcogenide waveguide pulse regenerator based on linear filtering of self phase modulated pulses. We demonstrate a nonlinear transfer function with 1.5 ps optical pulses and find good agreement with theory.
Further reading: CUDOS What CUDOS does (from annual report
2004 menu) Ben Eggleston wins the 2004 Malcolm
McIntosh Prize for Physical Scientist of the Year Lighting up Australia’s future, by
Steven Keeping
TuF1 08.30 - 09.00
(Invited) |
These are now archive pages: for 2009 and beyond visit www.scienceinpublic.com/blog General
enquiries: please contact the people and organisations mentioned in our media
releases
|