It bears a similarity to MIMO, in that MIMO initially promised "infinite" capacity. MIMO did give an improvement in capacity, but it wasn't infinite, the limit being related to the volume occupied by the antenna array (see papers by Leif Hanlen). One has to think that this technology will turn out to have a similar limit on fuller analysis. In fact one has to wonder whether the limit will turn out to be exactly the same, and whether it turns out to be a form of space-time coding? After all, one could presumably emulate the "slotted parabolic dish" antenna mentioned by using a suitably coded antenna array?
I'm not sure why anyone writing a science article would ever talk about anything being "potentially infinite". You could claim that the ability to broadcast on different frequencies gives you room for a potentially infinite number of broadcasts, arguing that frequencies are real numbers and there are an infinite number of them between any two points on the radio dial. Unfortunately, you can't simply think in terms of frequencies available, because you need a certain amount of bandwidth around each one to carry information. The more closely you pack the broadcast frequencies, the narrower the slice of bandwidth you have for each, so the more frequencies you transmit on simultaneously, the less information you can transmit on each one.
I suspect that this discovery adds another dimension, as if you took the frequency line and expanded it to a plane, but that each potential transmission "point" on the "airwave plane" would need a finite area around it in which to pack information. This new dimension may allow us to use a lot more capacity that we were previously wasting (maybe), but I'm pretty skeptical that the usable area will be "potentially infinite" except in the limited sense that the usable frequency line was already "potentially infinite".
Even so, any large increase in the effective wireless bandwidth would be cause for celebration.
I, for one, think this is neat stuff. (Practicality might well be a different matter.)
Poynting himself mentioned the angular momentum present -- From the arxiv paper:
Poynting, J. H. The wave motion of a revolving shaft, and a suggestion as to the angular momentum in a beam of circularly polarised light. Proc. Roy. Soc. London A 82, 560–567 (1909).
The presence of "suggestion" in the title is also interesting. Wish I had time to dig through this history.
Me, too thinks that this is just (a subset of) MIMO in disguise. The advantage of the antenna array you mention is, that it can precode/decode with any channel matrix, while these antennas implement one fixed matrix.
I get the same feeling. MIMO basically exploits that there can be multiple propagation "paths" over the same frequency and that those can each carry information up to the Shannon capacity. Paths can be separated by geometry (reflections etc) or polarity. This Orbital Angular Momentum is probably just another way to create more "paths" that can be exploited with MIMO.
In complex radio environments (like urban cellular networks) you get so many reflections that the limitation to MIMO is usually not too few separate paths but rather too many, with too much correlation between them. I'm not sure this would help very much there.
But in simple radio environments, like point-to-point links you often are limited by the number of separate paths (you can have two based on Spin Angular Momentum aka polarity and there are lots of products out there that do that).
In short; I think these guys are barking up the wrong tree in trying to adapt this for cellular, they should stay in point-to-point. They will be very welcome in that space if they can go from 2x2 to 4x4 MIMO across a point-to-point link.
This doesn't have to do with different propagation paths. The Orbital Angular Momentum (OAM) is an inherent wave property, like Wavelength or Frequency. Just like you can listen to radio on different frequencies (eg 660AM, 1010AM), you can think about this technology as "listening" on different OAM channels.
The twist in the wavefront can be thought of as a newly available subset of channels.
This has been a a field of research that has been primarily been developed in Optics, in fact, my PhD research included creating ultrafast (femtosecond), supercontinuum (white light) vortices that are capable of transmitting information over 2^L channels where L is the amount of twist the light has.
You're right that multipath is very different from SAM and OAM from a theoretical point of view. Not so much from a practical point of view though.
In current point-to-point systems you often use SAM to give you two "channels" across the link, most often referred to as horizontal and vertical polarity, but you then hook that up to the same MIMO technology that you would use for multipath.
I would expect OAM to be exploited in a similar maner for point-to-point.
The fact that SAM is very difficult to exploit for capacity in cellular applications makes me think it will be even more difficult to exploit OAM. IF you ever see it implemented my money would be on using MIMO to exploit multipath + SAM + OAM, where multipath would dominate in complex radio environments and SAM + OAM in simple more point-to-point like environments.
Circular polarization of light (or waves in general) corresponds to a spin angular momentum that arises as the polarization of the light is "spinning".
This "twist" is the wavefronts phase rotating or being staggered as it travels forward. Think of the wavefront of the twisted wave as looking like a piece of spiral pasta.
It is this spiraling that corresponds to orbital angular momentum.
Under what circumstance can you change the OAM in order to exert a torque? (or maybe vice-versa: What is an example of torque exerting a change in OAM on the wave?)
You can prepare waves in such a way that forces them into this spiraled waveform state. This can be done in a cavity (in the case of lasers) as there are solutions of the wave equation that give rise to OAM; or this can be done by using diffractive optics, like a hologram, that somewhat force the wave into this state. It is this case that you can think of a torque being exerted onto the waves.
Waves too, can exert torque on small particles (micron sized polystyrene spheres for instance). Light with spin or orbital angular momentum can be used to make these small object rotate!
From that article, it's impossible to understand how this differs from ordinary cirularly polarized radio waves created routinely with helical antennas.
I assume the key difference is that the electric and magnetic field are not 90 degrees out of phase, but some other amount?
It's hard to get a real message through, so many outrageous claims and layers of "popularization" and jargon.
http://www.nature.com/news/2011/110222/full/news.2011.114.ht...
It bears a similarity to MIMO, in that MIMO initially promised "infinite" capacity. MIMO did give an improvement in capacity, but it wasn't infinite, the limit being related to the volume occupied by the antenna array (see papers by Leif Hanlen). One has to think that this technology will turn out to have a similar limit on fuller analysis. In fact one has to wonder whether the limit will turn out to be exactly the same, and whether it turns out to be a form of space-time coding? After all, one could presumably emulate the "slotted parabolic dish" antenna mentioned by using a suitably coded antenna array?