Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
New quantum receiver the first to detect entire radio frequency spectrum (phys.org)
184 points by dnetesn on Feb 6, 2021 | hide | past | favorite | 74 comments


Another StarTrek tech entering the reality realm.

When Voyager stumbled upon a industrial-age civilization Kim immediately notified Janeway that he's detecting "Old style" FM broadcasts. I recall thinking "why would a 24th century subspace-communicator equipped star ship even monitor those?"

I got my answer, quantum receivers, they monitor everything.


Why would a 24th century all-purpose info-tool not actively monitor and attempt to characterize all signals received?

I mean look at all the sensors we can pack into a phone nowadays: crazy zooms, infrared, a wide range of radios. And that's in just 10 years, imagine how much improvement can be made in another 400 years. Honestly, if I were regularly beaming down to visit planets at wildly varying and unknown levels of technology I'd want to know as much as my reality-refractor could tell me.


I mean look at all the sensors we can pack into a phone nowadays

Mobile phones used to include a FM receiver but more recent ones have stopped including it because it wasn't deemed useful enough.


No, it mostly stopped due to carriers not wanting FM due to it being competing with traffic bills - this despite pressure from civil society; think emergency responders, not wanting to drop the functionality for obvious reasons.


That doesn't explain its absence from most phones from manufacturers that do not have any chance on getting sold directly by carriers. You will find occasional features on those phones - such as multiple SIM slots - that are rare on phones sold directly by carriers, but FM radio has become rare even on those.


Multiple sim slots are extraordinarily common outside the united states.


The catalog of one of the most popular retail networks in my country shows that 78% of the smartphones they sell have two sim slots


On phones sold directly from carriers? Where? Not doubting that it's possible, but I've not seen that anywhere I've looked.

I know it is common outside of what carriers sell - every phone I've owned for the last decade have had multiple SIM slots. But none of the phones offered by my carrier does, at least last I checked.

In any case, the point is that even on phones by manufacturers that have no reason to care what the carriers think - and none of the phones I've had in recent years have been carried by any carriers as far as I can tell -, FM radio is no longer a common feature, and so pressure from carriers is not a good explanation.


I don't know about current models, but in the past a trace was cut and a resistor placed in order to _disable_ FM. Multiple SIM support is also simply sabotaged due to carrier pressure and a price to pay per registered IMEI.


From what I understand the lack of use (in the US) is largely due to carriers being reluctant to enable direct FM receivers on the phones on their network. Can’t charge data when the customer can get free FM.

Discontinuation may also be due to the requirement of using headphones as the receiver antenna.


Buy a Nokia, Xiaomi, Motorola, or even the korean version of Samsung devices.

All of them have FM, some even DVB-T and DVB-H support.

It's only the "premium" class in the US that doesn't include these feature (or dual sim, or micro sd slots, or 3.5mm jack, etc)


As I understand it the FM radio feature requires physical headphones to act as an antenna. Maybe enough people are on bluetooth headphones (Airpods etc.) that it's no longer cost-effective to include the physical headphone circuitry which is required for the FM receiver to work.


Maybe it's just me, but this reads to me is as a big load of cleverly worded (misleading) statements that say little, yet make all kind of vague suggestions.

I don't know any details about this particular tech, but I think the article says this can detect radio waves over a relatively large spectrum (0Hz..20GHz) simultaneously. It doesn't mention anything about focusing on any particular frequency, or even if it can say anything about the frequency of a signal. From the article itself, it sounds more like this tech can just detect if ANY signal within this range is present, not particularly what kind of signal. Maybe the tech can "hone in" on a particular signal, but I don't remember reading anything about that.

All the rest, about all the marvelous new possibilities this tech will provide (once developed further) sounds more like somebody from marketing went all out on it. Just as so many things with "quantum" in their name, it sounds mostly like a nothing-burger blown completely out of proportion and context.


It's not an antenna like I thought at first. It's a tuner with 4 MHz instantaneous bandwidth and a very wide tuning range, https://arxiv.org/abs/2009.14383 - the actual (laser intensity) measurements are done within the RF electrical field above a coplanar waveguide track on a PCB.

It's a cool RF tuner. But it still need a real, full size, antenna. It's also not clear to me if they can get the phase informations out of the laser intensity measurements. I suppose it's all a matter of how fast they can sample the light.


I think the point is that they are improving the technique. In absolute terms the tuning range and bandwidth aren’t particularly notable compared with current technology. For example Lime Microsystems next generation RFIC under development is targeting a range of 100 GHz and 2 GHz of bandwidth.

https://limemicro.com/technology/


So basically, 3x higher than an hackrf (1MHz-~6GHz)? Not bad, but also not something revolutionary.


Hackrf only captures 20 MHz of spectrum. But is tunable to the range you described. You would need 1000 hackrf to capture the entire range at once.


Bandwidth is 4MHz, according to the article.



This is a bit cynical, but a great rant and it makes some good points. I’m curious what’s your opinion on Bitcoin?


> I’m curious what’s your opinion on Bitcoin

a Ponzi scheme by any other name .. which primarily works because it exploits the uncertainties which many people feel about yet another system we are all invested in (whether we like it or not), sporting an extremely unfair distribution and being fundamentally broken too (in fact, at this point it pretty much functions like a Ponzi scheme too): The Dollar dominated world economy.

It is a Tulip mania in the making, which is going to make smart/early investors very rich (as long as they act quickly) when an sudden event (unexpected for most people) will trigger a serious crash of the world economy (which at this point appears almost inevitable). I'm pretty sure that plenty of serious investors are cynically gambling on exactly just that.

Sadly, an increasing proportion of primarily the US economy (and plenty of foreign ones, somehow dependent/reliant on it) revolves around stocked-up capital (potential power, but not used for anything useful, just speculation and political influence) on one hand and increasingly desperate fraud and deception, trying to access this piled up capital. The daily news is distracting most people away from this reality, but I think it is getting worse, rapidly. Something will crack, eventually.

Bitcoin isn't the real problem. It's mostly a symptom of a far bigger problem (will only be "solved" with a crash).

Similarly, plenty of technical/science news these days appears to be no more than deceptive marketing campaigns. But that is again, just like bitcoin, more like a symptom of an even bigger problem than the actual problem itself.


What sort of exotic signal processing solution would be required to support 20GHz of bandwidth at any useful sampling rate and bit depth? How would you actually collect a meaningful amount of information from every station in range simultaneously?

I feel like this is something that may be possible in physics, but is not well supported in practical applicability. The amount of signal processing hardware required to support this in the field should be a deal breaker with modern hardware, unless there are some really strong constraints regarding the bands & modulation schemes actually being monitored simultaneously.

Assume 40GHz sample rate for bare minimum no aliasing over the 20GHz spectrum, 8 bits per sample. You will wind up with ~320Gbps of data to deal with.



Ask Keysight

https://www.keysight.com/en/pdx-2935683-pn-UXR0802A/infiniiu...

That said most signal processing scopes at high frequencies are analogue, so you are dealing with Giga Samples / Second not bps. But yes to to make sense of the actual data it would require some serious networking and storage but that might not be necessary for many applications.


Once you have performed the sampling, or even before that technically, information theory provides the connection between the analogue and the digital.

> That said most signal processing scopes at high frequencies are analogue

I think the state of the art is DC to a few (10s?) GHz with direct sampling, impressive stuff.


By direct sampling do you mean non-interleaved? The UXR oscilloscope he linked has a set of 256 Gsample/sec digitizers which can do the full DC to 110GHz on all 4 channels.


Clearly I'm not up to speed then - although I had in mind one I might one day be able to buy myself as opposed to be the absolute tail end of T&M equipment worth selling.


Yeah that scope alone costs over $1M, the probes for it probably cost more than what most people would or could pay for their house.


The main application would seem to be no longer needing to scan the spectrum for activity... you'd continuously be scanning (effectively) all of it simultaneously. A least one useful bit information you'd be obtaining from this is simply where in the spectrum activity is occurring.


A quantum signal processor of course!

It is however 5-10 years away.


FPGAs, basically.


My company can create some of the highest density cold Rydberg (Cs) atom setups in the world. We use our tech for focused ion beams, but I'm interested in this application. Our equipment isn't 'miniature' but everything is in a single 19" rack.

If anyone who is an expert on the RF side wants to drop ideas below or contact me at adam@zerok.com to brainstorm please feel free.


Cool tech. I would reach out to technical folks at Keysight to see if there is a potential collaboration.


This article got me interested to learn more. There's a related phenomenon called the Rydberg Blockade[1], wherein a group of Rydberg atoms can be tuned to multiple quantum states and because of their large radii, form a coverage area over space in which those quantum states are excluded for other particles (by Pauli exclusion?). There's work to create large arrays, 1D and 2D of tens to hundreds of elements, of Rydberg atoms with packed configurations that can be continuously replenished atom-by-atom. This bit is being used for optical coupling in quantum computing[2].

But this got me wondering... does this mean that such a 2D array of Rydberg atoms, continually replenished, could effectively operate as an energy capturing force field? I'm imagining a surface that for a given maintenance energy is minimally dense while maximally extensive in space, and medium duration (the Rydberg excitations last longer the more they're pumped, e.g. micro and milliseconds). Also because of the dipole coupling and high-frequency tunability (~100s of atoms in packed arrays < 1s again see 2), it could be quickly restored from perturbation.

If so, this seems to present a candidate for energy capture from photonic emissions in e.g. fusion reactions, but crucially, one that could be maintained without large amounts of mass or that mass ablating away. That might make a handy energy shield for a spaceship fusion engine (I believe that's a key problem for small fusion engine designs), esp if you could recycle the captured energy :)

[1] https://blogs.unimelb.edu.au/atomopt/rydberg-atoms/ [2] Quantum Simulators and Processors Based on Rydberg Atom Arrays https://www.youtube.com/watch?v=7MXIVZLaCII


A bigger problem in getting power from fusion is dealing with the resulting neutrons. You can't redirect them out the exhaust, so it's energy you've lost (unless you harness the thermal power of the reactor rather than straight up using it for thrust); they bombard the reactor and can cause regular matter to become radioactive...


I think that's true for the easier to ignite reactions like D-T, but there are others that are aneutronic, eg Proton+Boron. It's much harder to initiate, but the fuel is cheap and the energy output higher.

"Aneutronic fusion loses much of its energy as light... Since X-rays can go through far greater material thickness than electrons, many hundreds or thousands of layers are needed to absorb the X-rays." https://en.m.wikipedia.org/wiki/Aneutronic_fusion

Now the Army research in the OP doesn't go up to XRay frequencies, but I haven't found a hard upper limit to the coupling effect yet.


What if the fusion container was an x-ray laser? Could it induce fusion, where emitted xrays are in phase with the laser? I can imagine that an extreme high energy, directable, phase aligned, giant laser power source could be pretty useful.


Yeah, the NIF experiment is laser driven fusion, but I don't think that even being in phase would help capture the light. Not sure. It's the Ryberg coupling with light (without the usual requirement of just lots of mass/shielding) that's kind of unique here


Pretty big win for RF. Peak sensitivity -145 dBm/Hz. Little bit of a stretch to say entire RF spectrum, but 0 -> 20 GHz is still really cool.


Apparently the same lab has done related work into the THz range:

"In March 2020, the laboratory announced that its scientists analysed the Rydberg sensor's sensitivity to oscillating electric fields over an enormous range of frequencies—from 0 to 10^12 Hertz"

https://en.m.wikipedia.org/wiki/Rydberg_atom


Right. If it went into the terahertz range, it would help open up that area. 5G cell phones go higher than 20GHz than that.


Even so, this is pretty impressive, not all that long ago this device would be SF and people would doubt it would ever exist.


Not that long ago, the radios in a cell phone would be SF.


Indeed!

The antenna in mobile phones perform well across a wide range of frequencies due to being fractals!

https://en.wikipedia.org/wiki/Fractal_antenna


No, it does so because of solid state tuners, and RF switches.


The article I linked claims:

A fractal antenna's response differs markedly from traditional antenna designs, in that it is capable of operating with good-to-excellent performance at many different frequencies simultaneously.

Is this inaccurate?

The associated electronics are obviously necessary too.


Fractal antennas are mostly snake oil. You’ll always see then driven against a large counterpoise (ground plane), as the counterpoise becomes the radiator. You can make use of Chu’s limit without resorting to fractals.

https://en.m.wikipedia.org/wiki/Chu–Harrington_limit


FYI @madengr, almost all of your recent comments - going back a few months, are in [dead] state. Not sure if you've been shadow-banned or what, but you might want to contact the mods or something.


That opening paragraph is hellishly depressing.


In that particular case tax money were used on peer-reviewed research. Rather than emitting carbon driving a tank or flying yet another airplane. Not that bad.


Once upon a time the people attracting funding for what became the internet were using language just like that. Almost all tech has good and bad applications.


A lot of consumer technology came out of research towards military application.


This still requires an antenna tuned to the appropriate wavelength, and for matching you cannot have all antennas connected at the same time, they must be switched, so this would be unable to receive all frequencies at the same time.

The paper also indicates this is a heterodyne technique, so you have to tune to a center frequency like a radio, and the receptivity looks very narrow.


Miniaturization and ruggedization for practical applications will require a lot of work. Not only are the atoms contained in a heated tube, but, unlike atomic clocks, sophisticated tunable laser technology must also be employed. So only very high value missions, like large satellites or ground installations are practicable.


This is how every tech breakthrough starts. Remember when lasers required high vacuum tubes and high voltage power supplies with ridiculous specs, serious glass making skills and pretty much daily cleaning of all the optics? And now they are just about everywhere and cost pennies.


How much different would the laser tech need to be than what is used today in laser gyros?

Based on how these things tend to advance I would probably give it a decade before this can fit into an ELINT pod.


Preprint of the paper described in the article: https://arxiv.org/abs/2009.14383


Does anyone have insight into why this is published in the open literature, if it is such a militarily-exploitable advance?


i dont think we have any enemies questioning whether or not we listen to radio frequencies?


Yup. The cool thing is that it's an all-in-one system. I'm not sure there's a huge strategic advantage in knowing whether someone has one mega-radio or a bunch of separate modules glued together.

Plus, I linked to the preprint above. It's ten pages long--you're not going to be able to build your own based on it; it's more just documentation that the thing exists.


Think of this disclosure as the "free for the public" version of something way cooler they aren't telling everyone about.


what if they know/calculate what is implied into development/utilization so they know no one but them (or the same class opponent) could replicate it


Very interesting. I was watching skinwalker ranch scifi doc program, and they found a signal with the same oscillations from 25MHz to the "GHz". They had a rocket scientist, Travis Taylor, (PhDs in Optical Engineering and Aerospace System Engineering, MS/E in in Physics and Astronomy) state that it was impossible to have such a huge range in frequencies from HAM band (all the way down to RC car bands) to microwave bands. I would assume if someone need to do this, they would need a huge array of transmitters or something like this to transmit such a broad spectrum.


It seems like lower radio frequencies require really big antennas. Does this somehow work around that enabling very compact low frequency antennas?

Low frequencies are very useful because they penetrate structures more easily.


Magnetic antennas are quite small already.


Is this already possible by scaling vertically with SDR and processing power? By the time the quantum receiver is commodity, I believe you would be able to do the same with old technology for a fraction of the price. But I guess there are benefits that I am not able to see, and the progression of new tech...


These are no longer the days when you had to spin your tuning dial up and down the band listening for 'CQ'. Be great if this stuff could distinguish the intelligence from the lightning and truck ignitions and propaganda stations.


Waiting for my next USRP with a Rydberg frontend.


This could potentially be game changing for SETI.


I only took mechanics and electromagnetism in university physics so anything based on quantum mechanics is like magic to me.


Could this be useful for astronomers?


Fantastic development.


So what's "quantum" about this?


The NSA has just entered the room...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: