hckrnws
Oh, that's very neat, I can imagine that some future front-end to a SDR will have no antenna at all, just a bunch of solid state.
The abstract of the paper:
"""Coupling a Rydberg vapour medium to both microwave and optical fields enables the benefits of all-optical detection, such as minimal disturbance of the measured field and resilience to very strong signals, since no conventional antenna is required. However, peak sensitivity typically relies on adding a microwave local oscillator, which compromises the all-optical nature of the measurement. Here we introduce an alternative, optical-bias detection, that maintains fully optical operation while achieving high sensitivity. To address laser phase noise, which is critical in this approach, we perform a simultaneous measurement of the noise using a nonlinear process and correct it in real time via data processing. This yields a 35 dB improvement in signal-to-noise ratio compared with the basic method. We demonstrate a sensitivity of 176 nV / cm / sqrt(f Hz) , reliable operation up to 3.5 mV/cm at 13.9 GHz, and quadrature-amplitude modulated data transmission, underlining the ability to detect microwave field quadratures while preserving the unique advantages of all-optical detection."""
Emph. mine, at about -36 dBm that's not super sensitive yet though, but that formula suggests that at lower frequencies it should be a lot more sensitive.
The paper is at: https://www.nature.com/articles/s41467-025-63951-9
The abstract is not very clearly written. The 3.5 mV/cm is the maximum detectable field strength before the system starts to saturate. I would be interested how the sensitivity compares to conventional (cryogenic) semiconductor detectors.
Ah, yes, you are right, I misinterpreted the part the maximum was aimed at, thank you for the correction.
And I too am interested in a like-for-like setup with the best of what is currently on offer as the benchmark.
That raised the question, how do you make Rydberg atoms, and the answer is (always!) with lasers: https://en.wikipedia.org/wiki/Rydberg_atom
A company called Infleqtion already has an RF sensing product that uses Rydberg atoms.
Is it really a product already? There are a number of renders but the 'products' link does not work for me.
That page uses the older versions of "SWaP-C2 Optimized[1]" meme, I would guess it might have not been updated in 3-5 years.
1: Stands for "Size, Weight, and Power [and] Cost [and] Cooling Optimized", defense industry equivalent of self awarded gold medal stickers on product packaging, apparently
Not sure if it's being "actively" used in real-world deployments yet, but it does exist.
https://www.prnewswire.com/news-releases/infleqtions-quantum...
Ok, I found a video:
https://www.youtube.com/watch?v=j18K7MZdyY8
That looks like it is definitely real but also extremely fragile and there is no mention of any performance specs. But that's 3 years old now and no product(s) available yet.
We demonstrated simultaneous reception of neighboring channels with strong isolation between them." This enabled the researchers to monitor numerous radio channels at once, instead of tuning into them individually.
Can anyone elaborate on this? How does a single receiver produce multiple concurrent outputs, and how are they isolated in this context?
Because all of the signals are superimposed. So if your receiver isn't selective it will show all of them at once and if you then demodulate selective parts of the spectrum by filtering you can isolate the signals individually.
Think of any antenna: it is just a rod or a coil, it may have a specific frequency that it particularly likes because that is a nice fraction of its wavelength or close to its own resonance frequency, but that doesn't mean it isn't going to receive all the other signals to greater or lesser extent as well. The ratio between that one that it likes and the rest is called selectivity. The lower the selectivity the more evenly you will receive all signals at the same time.
Usually receivers have a tuned front-end to get as much of the signal you want and to repress the rest as much as possible but that is optional, you can have a wideband front end just the same.
If you're just looking for the modern explanation of how an SDR works: an SDR just measures electrical current flow over time. If you graph current over time as detected it usually just looks like a bunch of signals all superimposed upon each other. A signal like "f = asin(x) + bsin(y) + c*sin(z)" if all the transmitters were just transmitting pure carriers with no modulation.
Somewhere in the SDR, electrical current flow becomes a number using analog to digital conversion on the order of 7-12 bits in most SDRs. At that point you apply something like a FFT and then the time domain signal becomes a frequency domain signal, which can separate out different carriers.
Unlike conventional cars that require expensive safety systems such as air bags and seat belts, the mover3000's top speed of one mile per hour makes it intrinsically safe.
I have no clue, but I would guess that they do not have a single atom but rather an entire crystal of them.
I immediately thought of the old crystal radios and a short search brought up this comparison. While old radios used a natural mineral (galena) to detect radio waves, modern Rydberg radio receivers use a synthetic photonic crystal (often made of silicon or glass) to guide and enhance the radio signal for improved performance.
That's not the function of the crystal though, it is not optical but 100% electrical and serves to demodulate the signal, not to receive it.
Thanks for that, I was partially aware of that, just made me think of crystal radios and how magical they are.
They are magical! A radio powered by the transmitter, it's an incredible thing, really. The trick is that the diode is a very low forward drop one, typically Germanium, which has only 0.2V forward drop. But for more magic, you can also use a razor blade or a little bit of lead and then seek for naturally occurring diodes on the surface of the material. These can be even more sensitive than the Germanium ones.
Interesting, I learnt something new today. My only comment was the noise floor of the simple graph was very high-25dBm which (without having a clue how the physics works and skimming the article) sounds about right for something with no proper RF front-end.
Where do you see -25 dBm noise floor? But chance do you mean a 25 dB SNR?
The picture here <https://scx1.b-cdn.net/csz/news/800a/2026/rydberg-atoms-dete...> though I just saw the X-axis refers to audio frequencies, kind of confusing on first glance.
That's dB not dBm, and the article says it's normalized to whatever power each of the two options received at. Nothing about absolute power.
Comment was deleted :(
What modulation technique?
Considering the mention to Family Radio Service, most likely they're demodulating FM.
"The key here is that the radio frequencies used in handheld radios are far from the natural resonances of the atom, so while the atoms can sense the radiation, they don't respond to the frequency modulation on which the audio is encoded,"
Crafted by Rajat
Source Code