Boffins biff over ‘twisted radio’ Six months after an Italian/Swedish group set the comms world alight with a wireless technology of theoretically infinite capacity, debate over his work is becoming increasingly bitter. To recap the original story: today’s wireless technologies use a handful of venerable modulation schemes to carry information (amplitude, …
The second dissenting paper can be easily answered: run the test multiplexing three spins rather then two. Either it won't work (and these guys are right) or it will work, and there is some new things to explore in physics.
TBH: I hope they can and do get it to work with three. It will likely lead to exciting advancements in quantum mechanics!
As well as infinite sensitivity and infinite resolution and zero noise. It sounds very unlikely. They say that the 'spin' can be varied infinitely; what they mean is that it can be varied continuously. Well, the amplitude and frequency of a radio signal can be varied continuously but nobody would suggest that this implies an infinite number of detectable states that can be used to encode data.
Things like 56k modems used multiple states of amplitude and phase to put more data down a telephone line, even though the actual baud rate was limited by telephone bandwidth.
So there is a precedent for varying something continuously to have no definite limit to the amount of data you can transmit: the actual limit is set by noise. If there was no noise on telephone lines, you could use ten times - or a thousand times - as many amplitude levels, and transmit more data.
The 56k limit was imposed by the digital-analogue conversion at the exchange (8 kHz sampling at 8 bits, with one bit being 'robbed' for other purposes leaving 56 kbit/sec). You can indeed transfer much more data down the analogue bit of the line, but that'll all be lost when it's converted to digital by the exchange - indeed, that's exactly what ADSL and VDSL do, adding much better equipment at the exchange to recover multiple megabits a second of cleverly-modulated data the PSTN exchange can't handle.
frank ly: "the amplitude and frequency of a radio signal can be varied continuously..." As can the phase, which is a more appropriate comparison to what they're trying to do with OAM. But yes, you're correct that this is no more "infinite", in practice, than phase modulation is.
John Savard: Yes, many protocols have used modulations in different domains and/or digital modulations with more than two values. QPSK, for example, uses four phase values for two bits per cycle. That's not in dispute nor the relevant point behind the OAM claims.
The issue of bandwith (or 'quantity' of information) of electromagnetic radiation is not as complex as the authors of the 'twisted state' stories present it. Let me brake it down for you:
There are four properties of electromagnetic radiation: (i) the amplitude (strength), (ii) the frequency, (iii) the polarization, and (iv) the localization.
The limit of information content as function of (i) is trivial: in the ultimate limit one might use single photons to transmit a '1' and the absence thereof to transmit a '0'. In practice, there is noise to consider.
Point (ii) is a bit tricky to classify: i you want to transmit information with a well-defined frequency (leave the neighboring frequency open for another sender), then you need a long pulse (the wave frequency of a short pulse is not well defined). So communication becomes slow. If you use shorter pulses, you can communicate fast, but need more frequency space for it. In the end, Heisenbergs uncertainty limit tells you how much information you can get through a limited frequency range. Turns out here is a hard limit, it does not matter whether you use short pulses with broad frequency or long pulses with narrow frequency.
Point (iii) is trivial: there are two polarization states for the photon. You could call them horizontal and vertical or right handed and left-handed, but whatever you call them you can create your polarization state of choice by the sum of horizontal and vertical polarized beams. So by using polarization you can double the information content.
Point (iv) is the hardest to grasp, hence the big discussion about 'twisted' states. You can interfere multiple beams, which will lead to interference and a spatial localization of your photons. Indeed, a simple laser beam propagating through space can be properly described by the sum of many spherical waves which only interfere constructively in the forward propagating direction and destroy each other in all other directions. The 'twisted' states are nothing else but a slightly more complex superposition of waves. If you localize the intensity onto multiple detectors, the you can multiplex the information and transmit more thereof. But there is no magic to it, you could create an interference of 20 beams onto 20 detectors to multiplex your data 20-fold, or you could just send 20 beams separately. In the end it comes down to the technical practicality of the transmission and reception setup, but there is no magic increase in information density with those twisted pulses.
Hope this helps.
Can I suggest reading up even a little before spouting this kind of mock-intellectual mumbo jumbo. Modulate the amplitude and you introduce harmonics. Modulate the frequency and you introduce harmonics. Both increase the signal's bandwidth without any need to resort to quantum theory. Indeed there is ultimately little that can't be explained using traditional wave theory, Shannon and the Fourier transform. Introducing things like quantum theory when they are complete irrelevances is like those idiots who try (and fail) to apply GR to phenomena that can be satisfactorily explained with Newtonian mechanics.
Ultimately it is this one research team that are making outlandish claims and they don't have the facts to back up their case. I was dubious on first hearing about this even without reading their reports in full so frankly I not surprised now. When your claims fly in the face of received wisdom it doesn't necessarily make you wrong, but the onus is very much on you to prove your assertions than for others to disprove them.
Actually, what Shultz said is not "mock-intellectual mumbo jumbo", is was in fact reasonably well founded.
For myself, I am unsurprised that you can get extra bandwidth by adding a spatial encoding as well, although its practical utility might turn out to be rather limited. But then they published in (the open access) New. J. Physics,
so their idea of utility is not the same as that of an engineer.
And "like those idiots who try (and fail) to apply GR to phenomena that can be satisfactorily explained with Newtonian mechanics" ... actually this is a reasonable thing to do. I have on my bookshelf a copy of Schutz "A first course in general relativity", and one interesting exercise in there is to show how GR reduces to a Newtonian-like theory in the appropriate limit. It might be a bit pointless from an engineering perspective, but not for physics.
Similarly, minor typo's aside, the post was a good breakdown of the four information-carrying aspects of an EM wave. In particular, the introduction of the uncertainty principle as an explanation for Shannon's theorem was interesting - you know that it has no other basis in any physical law (afaik) - it ties in broadly with entropy, but apart from these two, and possibly the uncertainty principle, there is no physics that links the concept of information with anything else. Information does however seem to be "a conserved quantity" of some sort, and there should ultimately be some new physics coming from it.
In particular, the introduction of the uncertainty principle as an explanation for Shannon's theorem was interesting - you know that it has no other basis in any physical law
That has been well explained. As Rebecca hinted at the instant you begin modulating a carrier to carry data you introduce harmonics - those are not stray artifacts of the hardware but fundamental to the manipulation you are undertaking.
Consider your starting position of a pure sinusoidal carrier wave at say 1MHz. 1MHz is the only frequency present in the signal because it is a pure sine wave. Then you start modulating it - say you employing amplitude modulation to superimpose another 5kHz sine wave. The resulting signal composes four frequencies - at 5kHz and at 0.995, 1.000 and 1.005 MHz. If instead of a sine wave you superimpose a square wave - i.e. turn the 1MHz signal on and off - you get at infinite number of harmonics going up at 1.005, 1.015, 1.025 ... MHz and downwards at 0.995, 0.985, 0.975 ... MHz.
The explanation for this is not physical, it is simply what happens when waves are added together. Joseph Fourier showed this 200 years ago well before Shannon. The proof is pure maths and therefore if anything more convincing than any physical explanation since countering it is essentially arguing along the lines that 2 + 2 ≠ 4.
What Shannon showed was that if the bandwidth of the resulting signal is constrained - either deliberately to control spectrum usage or parasitically by limitations of the transmission system - the data that can be reconstructed from the resulting distorted waveform is fundamentally limited by that bandwidth. If we take that 1MHz carrier turning on and off at 5kHz and pass it through a band pass filter that passes over the range 0.995-1.005MHz we don't see the true on-off nature of the signal at all, because we have stripped out the harmonics carrying that information. Instead we see a 1MHz carrier modulated with a sinusoidally varying 5kHz signal. If the filter is narrowed further to 0.999-1.001 MHz we see nothing but a continuous carrier wave even during the periods the source is turned off. In other words, the entire data content is lost.
None of this requires any "new" physics - it is 200 year old pure mathematics.
The apostrophe in "typo's" is justified, though "typos" is equally easy to argue for.
Since the apostrophe in English indicates the elision of letters and/or spaces, and "typo" is a contraction of "typographical error", then "typo's" is simply a contraction of "typographical errors". That's perfectly in keeping with English orthography and punctuation.
Conversely, "typo" as a shorter version of "typographical error" is in common usage as a word on its own,[1] and so "typos" is a reasonable plural. (Some might prefer to form the plural as "typoes", in parallel to common spellings such as "heroes", but nothing requires the additional "e". Personally, I think it looks awful with the "e", and while Google shows some results on the web, Google n-gram search fails to find any instances of it in Google Books.)
[1] If anyone's curious, the Google Books n-gram search shows "typos" passing "typographical errors" around 1995. Though the word "typos" has other meanings, I think they're now archaic or obscure, so that suggests "typos" in this sense is becoming not only common but dominant. Also, Google's scans frequently misrecognized "types" as "typos" - odd, since the former is considerably more common. (Retrain that model, Google.) The earliest use I found of "typos" in this sense, in a quick skim through some of the results, was 1968.
> That is like arguing that quantum physics violates Newtonion mechanics.
Except of course Newton never got to deal with ferrets.
Radio frequencies change with the weather.
During periods of sustained anticyclones terrestrial TV became impossible and that was after decades of hi-tech development.
"That is like arguing that quantum physics violates Newtonion [sic] mechanics."
Why is it?
Remember that Newtonian mechanics had experimental evidence against it *before* QM, SR and GR were invented. People didn't just pull those theories out of their ass.
There is no pre-existing evidence against the 2nd law; in fact there are strong reasons to suspect such evidence will never arrive (Google "perpetual motion" if you've really forgotten all this stuff).
QM in particular does not violate 2TM:
"The time development operator in quantum theory is unitary, because the Hamiltonian is hermitian. Consequently the transition probability matrix ... implies the Second Law of Thermodynamics."
Hence, anything derived from QM which violates 2TM is wrong.
There's a definite pecking order to physical laws, and while the second law of thermodynamics is often portrayed as an absolute inviolate in reality it is pretty near the bottom of that pecking order.
Essentially it reduces to a statement not about the Universe but about probability: there are far more disordered states than ordered ones so in there is a free choice as what the final state may be a disordered one is far more likely than a highly ordered one. ALMOST always the case is not the same thing as ALWAYS the case and there are plenty of documented "violations".
Consider a simple example by analogy: you have a deck of cards with a loose ordering to them - say it alternates between red and black cards throughout the deck. If you shuffle them in all probability you will destroy that loose ordering. There is no physical reason that it may not arrange into a MORE ordered state - say they sort into suits and by value within those suits. It's highly improbable but not impossible, and it is possible to rig the odds.
Suppose we carry one shuffling the cards, checking each time for that specific order. If it isn't there we shuffle again. If it is, we stop. After a suitably long amount of time it is no longer highly improbable that the cards are so shuffled - it is a virtual inevitability.
This isn't a purely abstract argument. Consider all the elementary particles required to form a complex, highly ordered object - YOU, for example. On the face of it there is no way in hell those particles could assemble themselves to form you. However, nature works in stages - protons and neutrons form into atoms and are then stable. If they don't they try again in another attosecond or two. Once formed the atoms form into molecules, or try again if they don't, and so on ultimately up to cells then organs and finally the completed individual. Each particular step is improbable but given enough roles of the dice such outcomes become an inevitability.
Your card-shuffling example is rubbish, since you're pumping energy into the system by shuffling the cards.
The same applies to the "ordered assemblage of particles" argument, and all such: the surface of the Earth is under a net influx of energy, and that has to be dissipated, and increasing complexity is one aspect of that dissipation.
There's no violation of the second law in either of those cases.
If someone points out to you that your pet theory of the universe is in disagreement with Maxwell's equations — then so much the worse for Maxwell's equations. If it is found to be contradicted by observation — well, these experimentalists do bungle things sometimes. But if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for it but to collapse in deepest humiliation.
Sir Arthur Stanley Eddington, The Nature of the Physical World (1915), chapter 4
Most likely is that using more "spin" states will result in inter user interference. This will put a limit on the rate of information that can be reliably sent using the technology. There are ways of mitigating the effects such as those used currently in MIMO systems, though they're not perfect.
Let's for a moment assume that you can encode information in Spin states. That doesn't mean "infinite" any more than than normal phase, fdm or amplitude is (COFDM uses all 3 by multiple carriers each modulated QAM). You can also polarisation, But only paths with no reflections are any use and more than two states isn't useful.
With any encoding system there is a "distance" between symbols. The more states you have the less distance there is i.e. QAM16 vs QAM2048 so signal is less robust. On Cable 256 QAM is feasible and on fibre even 2048, but on RF it's QPSK (very long range), 16, 32 or at most 64.
If this is a real mode it's unlikely that symbol distance ( immunity to noise) will allow more than two spin states to be useful in an open path. Which may at least double capacity on microwave links. Unlikely to be applicable to portable devices.
Of course the experiment they did might not be using spin at all but only two polarisations.
The author's original paper states: "This novel radio technique allows the implementation of, in principle, an infinite number of channels in a given, fixed bandwidth, even without using polarization, multiport
or dense coding techniques."
Note the "in principle", this is a pretty strong qualifier, essentially meaning (in this context), "ignoring all practical constraints, and only considering (this) simple theoretical model".
The big constraint here is alignment - the antennae need to be aligned, because you wont be able to get good enough OAM information from the edge of the beam, especially if its centre is off the receiving antennae. So mobile might be ok, but only if lined up carefully .... not if you are wandering down the street! This is more of a prepared fixed-point to fixed-point scheme.
No - there are two circular polarization states; but this is not the same thing as the (orbital) angular momentum of a light beam. You can have a "plane wave" in LH or RH circular polarization states without any spatial structure, but for OAM you need it. E.g. try looking up "optical vortex" on wikipedia.
Greater bandwidth per square kilometer is available on mobile networks by using more and smaller cells and lower power and range. If a cell transmitter is on every lamppost and domestic WiFi, and the signal level required based on weather and day/night conditions is negotiated by transmitters and receivers, frequencies are reusable at shorter distances between transmitters sharing the same frequency. Perhaps we could call this 'Cell Division Multiplexing'.
Spin-1 particles have 3 spin states, only two of which are observable along the axis (right-handed or left-handed, or horizontal or vertical, depending on how you build your detector). While you can vary the mixture of photons in your signal, you don't get extra multiplexing out of that because within each channel you'll still get the same old interference. Doubling your bandwidth is nice, but I don't see how you can get more than that without violating quantum theory.
Yes, polarization and oam are distinct, but did the experiment actually create individual photons with angular momentum, or just a tubular beam who's path of greatest intensity circled the tube's Center as a function of time. I suspect the later unless the antenna was actually some sort of magical black hole beam gun. Rather as Schultz says, this is just a particularly funky shape of mimo antenna.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
