German researchers demonstrate single laser 26 Tb/s transmission rates

[gary]

BBC News is reporting on an article that has appeared in Nature Photonics
where researchers at the Karlsruhe Institute of Technology in Germany
have demonstrated using an optical Fast Fourier Transform (FFT) encoding/decoding
scheme to transmit data from a single laser at rates up to 26 Tera bits / second
over 50km of optical fiber. A new world record for use with a single laser.

In the BBC article, one of the co-authors of the paper, Wolfgang Freude, says that
previously other researchers had demonstrated transmission speeds of 100 Tb/s using
370 lasers and orthogonal frequency division multiplexing techniques. However, as
Freude points out, 370 lasers is expensive and consumes several kilowatts of power.
The Karlsruhe team claim to now be able to create comparable data rates from a single
laser using exceedingly short pulses.

With applications such as cloud-computing and 4K video streaming set to explode,
the world demand for faster data transmission rates is not about to slow anytime
soon.

BBC article here -
http://www.bbc.co.uk/news/science-environment-13469924

Abstract of article in Nature Photonics here -
http://www.nature.com/nphoton/journa...n.2011.74.html

[renormalised (Carl)]

Only one "small" problem with this, Gary. How powerful was the laser they used to transmit the data?? It's most likely not something that will translate into anything commercially useful for some time.

[gary]

Originally Posted by renormalised

Only one "small" problem with this, Gary. How powerful was the laser they used to transmit the data?? It's most likely not something that will translate into anything commercially useful for some time.

Hi Carl,

I couldn't spot a quote of the laser power used, but I did find a reference to the
commercial laser used by the same investigators in a previous experiment
where they used a single laser to achieve 10.8 Tbits/s down an optical fiber.
That laser was quoted as having an average output power of less than 5mW.

Specifically, I noted one of the research team's commercial partners is a
company called Time-Bandwidth Products that specializes in picosecond and femtosecond lasers.

They claim -

Originally Posted by Time-Bandwidth Products

Time-Bandwidth Products pioneered the development of high pulse repetition rate, fundamentally modelocked lasers based on Er:Yb glass material. The result is the ERGO-XG laser, which has demonstrated superior performance for many years in leading laboratories across the world. Telecomm applications where the ERGO can play a critical role include ultrahigh speed transmission systems (100s of Gb or Tb / second), optical clocking, multiwavelength source, continuum generation, and frequency metrology, to name a few.

More details of their Ergo-XG laser can be fond here -
http://www.time-bandwidth.com/product/view/id/34

It mentions -

Fundamentally modelocked solid state laser exhibits gigahertz repetition rate (9.95328, 12.5, or 25 GHz) for fiber-optic and telecom-related applications. Pulsewidth of <2 ps is tunable within C-band 1540 to 1565 nm, with >5 mW of average output power.

One of the Nature paper's author's, D. Hillerkuss, is available for contact here -
http://www.nature.com/nphoton/foxtro...=D.+Hillerkuss

The BBC quotes one of the co-authors, Wolfgang Freude, as conceding the idea
of the optical FFT is a complex one. However, he goes onto say, that he
is convinced "that it will come into its own as the demand for ever-higher data rates
drives innovation."

He then says -

Originally Posted by Professor Freude

"Think of all the tremendous progress in silicon photonics," he said. "Nobody could have imagined 10 years ago that nowadays it would be so common to integrate relatively complicated optical circuits on to a silicon chip."

So it could be alluding to the fact that in order to viably commercialize the scheme,
one needs to design and fab a suitable photonics device.

I note one of their partners is Micram, a
fab-less ASIC provider in Germany that designs silicon for fiber-optical
networking and that another partner is Agilent, a world-leading photonics
manufacturer.

Thus the scheme may not be that far "out there" and certainly in the communications
research world many things progress pretty rapidly to commercialization compared
to other areas of endeavor, such as, for example, medical research.

[renormalised (Carl)]

Thus the scheme may not be that far "out there" and certainly in the communications
research world many things progress pretty rapidly to commercialization compared
to other areas of endeavor, such as, for example, medical research. http://www.iceinspace.com.au/forum/....s/thumbsup.gif

That's true. You here them say..."It'll be another 5-10 years before procedure 'x' or medicine 'y' will be available to the public", and then that's the last you here of it. 5-10 years passes by and there's still nothing to show for it. Might work too good for the big pharmaceuticals...can't cure everyone and get rid of a disease when it's too profitable to just "manage" it

[Tandum (Robin)]

This is why they only have to lay optic fibre once. The technology on each end of it is what get's better.