City College of New York researchers have manipulated the polarization of a laser beam to create shapes that could boost data transmission rates:
Using special devices called "q-plates," the researchers manipulated a laser beam's polarization into novel shapes some of which Milione referred to as "radial" and "azimuthal." "While light's polarization (linear and circular) is used for many modern technologies, such as, 3D television, its shape is often left untouched," he said.
The researchers showed that each shape could carry an additional data stream. While the researchers used only four shapes, in principal, the number that can be used is unlimited. "The amount of data that can be transmitted on a single laser beam can be scaled to terabits or even petabits," said Alfano. "This technology is potentially compatible with building to building communication in NYC or even between Google data centers."
The research is published in Optics Letters [abstract].
(Score: 3, Informative) by VLM on Monday April 27 2015, @09:58PM
http://en.wikipedia.org/wiki/Polarization_mode_dispersion [wikipedia.org]
For every action there's an equal and opposite reaction, sorta.
Just like different wavelengths travel at different speeds in fiber spreading pulses, so do different polarizations.
It will none the less add some capacity of course.
(Score: 4, Interesting) by c0lo on Monday April 27 2015, @11:50PM
The problem that I'd be afraid is mixing polarizations due to imperfections, which would result in mixing communication channels
https://www.youtube.com/watch?v=aoFiw2jMy-0 https://soylentnews.org/~MichaelDavidCrawford
(Score: 3, Informative) by coolgopher on Tuesday April 28 2015, @04:00AM
Yeah, you don't want to cross the streams. That would be bad.
(Score: 3, Funny) by TK-421 on Tuesday April 28 2015, @01:57PM
Tell him about the Twinkie.
(Score: 3, Interesting) by bd on Tuesday April 28 2015, @07:48AM
I'm afraid it will not add capacity for fiber optics communications, as this method only works in free-space data transmission, due to the usage of higher-order transverse modes (you would have to use multi-mode fibers, and those are suspect to an effect called mode dispersion, completely messing up this signal).
Oh, and it apparently only works over a distance of 1 m, as even the smallest optical table would provide for far more than one meter of propagation length, provided one cleverly uses a bunch of mirrors. I have only read the abstract as I am not at work, but I can only assume they tried and it didn't work.
(Score: 2) by bob_super on Wednesday April 29 2015, @03:53PM
On top of this, the fiber's capacity usually (99.99..%) isn't the problem in current systems.
Demuxing a full set of DWDM wavelengths at 25Gb/s each, still takes so much power that adding twice or four times the capacity in the box is a recipe for meltdown. If you have a long-reach link and can't dig to add more fibers, this kind of trick will help in a few years. But for now, most people are much better off adding another fiber and back-end.