Hacker News Reader

Comment by raldi | Hacker News Reader

The article doesn't state what distance they were able to transmit over.

During the Cold War, I've heard rumors of one directional communication system with submarines using neutrinos.

I don't know what the utility of the system described in the article though - neutrons will be absorbed pretty quickly, so the communication must have a very short range.

It takes something like 18 light years of lead to attenuate a neutrino beam by half, so they’d need a pretty big receiver :)

It's possible to detect neutrinos with a scintillation chamber, but in general only a tiny fraction of neutrinos passing through would be observed. They really don't interact with regular matter much.

However, it's the transmitter that would be nearly impossible, because blocking neutrinos or affecting them in such a way as to modulate the beam would be extremely difficult.

I think we can probably rule out neutrino based communications during the cold war, especially since VLF did the job just fine.

The "rumor" referred to an array of scintillators dragged behid the submarine, with the ocean water as a detecting medium. In theory, with clear enough water and large enough detector volume, that idea might work. I vaguely recall data rate of less than one bit per second.

That sounds a lot like the ELF radio system they used, which basically was used to order submarines to surface (or near enough) to receive new orders as the bandwidth was <1 baud.

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

The data-rate would be far below 1 bps. The event-rate in SuperK [1] from the Sun was ~0.1/day/kiloton of water.

Signalling to submarines would require a Fermilab or T2K-caliber accelerator with an ability to direct a beam, with a properly-oriented pion-decay hall, generally in the known direction of the submarine and highly-synchronized clocks (or a clever time-structure strategy) to allow coincidence-detection for background-suppression.

For scale, the MINOS detector [2] detected a couple thousand events with a couple of years of beam-time.

[1] https://en.wikipedia.org/wiki/Super-Kamiokande#/media/File:N...

[2] https://arxiv.org/pdf/1103.0340.pdf

Sure you can detect neutrinos that way, especially at higher energies. But you're going to be swamped by natural neutrinos (e.g. atmospherics), unless you have quite the beam. And how do you know where to point the beam?

Modulating neutrinos is easy. You just control the production rate. For example ramping a fission reactor up and down will create pulses of neutrinos (though it's not terribly efficient). For practical neutrino communications, you'd want to use a muon beam to generate your neutrinos. The neutrinos are emitted roughly in the same direction the muons were moving when they decay, so you can make a directional beam, and by controlling the number of muons decaying at any given point in time, either by directly controlling the number of muons or how much time dilation they experience in the beam, you can ramp up and down neutrino production.

For that modulation to be reflected in the detection array, the beam would have to be accurately targeted on the submarine?

How could a nuclear-powered submarine possibly host a neutrino detector? You're sitting meters away from one of the brightest neutrino sources on the planet!

please see my comment below in this thread.

“… using neutrinos”

“… neutrons will be absorbed…”

I made the same confusion for a second. Neutrons are easily absorbed. Neutrinos (little neutrons) are almost impossible to interact with (and therefore absorb)

Yea, I'm curious about this as well and what bit rate they were able to achieve. Wondering if it could be used for high frequency trading in the future since I would think neutrons going straight from source to destination would be faster than any other communication method (no idea if this would be too dangerous or not tho).

I would be very surprised if neutrons travel faster than a microwave beam.

Radio waves already travel at the speed of light, I'm curious what speed-up you're imagining? A shortcut through the earth is the only option I can think of to beat radio's latency.

Yeah, you would significant time on some routes (e.g. NYC to Sydney or whatever), but in practice, the modulation time is probably going to kill you even if the neutrinos make it there faster. In the most extreme case (antipodal communication), the neutrinos take 43 ms to traverse the Earth and radio takes 67 ms to go around the Earth (ok, in practice a bit more since your fastest path would be via LEO satellites, and each hop adds a bit of latency, but it could be made small). But RF signals have MHz-GHz of bandwidth and huge SNR.

To take advantage of being 20 ms early, you'd need to be able to detect neutrinos at greater than 50 Hz from the other side of the world. For context, the planned 40 kton DUNE far detector a mere 1300 km from the world's most powerful neutrino beam at Fermilab, will have a neutrino rate from the beam of a few a day. So yes, highly impractical.

If you managed to crack the detection problem, you’d still likely be so saturated with solar / background neutrinos as to confound any real prospect of a signal.

There are two handles on this: reconstructing the energy/direction and making sure it's consistent with the beam and gating on the beam duty cycle (since at least the beams I'm familiar with produce neutrinos in short bunches, but I'm no expert).

You can also add modulation to the beam to encode more information to help discriminate against backgrounds but that might require even more rate...

You're taking about neutrinos, but the original comment was about neutrons; I think that must add another layer of infeasibility (or was a typo).

indeed, a sibling was talking about neutrinos and somehow I got mixed up where it started.

Radio waves in a vacuum travel at c. They move more slowly in things like copper. Light in fiber cables travel at about 2/3 c so a neutrino beam would be faster and potentially have a shorter distance by going through the earth. Actually designing and implementing a communications system using neutrinos is an exercise left to the reader.

With neutrons, no. They’re slow, isotropic and thermalize easily.

With neutrinos that’d be a different story. They travel at the speed of light (or just below it - depends who you ask) and they aren't readily absorbed by anything. Therefore they can travel through the earth instead of around it.

Problem?

Generation, modulation and detection of neutrinos.

> They’re slow, isotropic

Not necessarily, if produced with an accelerator. (Potentially a nuisance to gamma-ray spectroscopists, generating spurious anisotropic scattering peaks.) Otherwise, yes.

https://arxiv.org/abs/1203.2847

Yep, I was still asleep, good call.

Reply on news.ycombinator.com