Hacker News Reader

Comment by agoodthrowaway | Hacker News Reader

One thing that I wonder is whether we could ever identify an alien communication signal as an alien signal is likely to be much more advanced. In that scenario, the alien civilization is likely to be able to send their signal at the Shannon limit. Additionally an advanced civilization is likely to be sending signals optically instead of over radio waves. But even if we assume radio waves, Shannon says that the you can send the most information when using a random code. As you approach the Shannon limit the signal becomes more noise-like as the encoding tends towards random. So it’s entirely possible an alien signal is likely to look like pure noise to us and hence we may miss a true alien signal.

View on HN

It's too short for communication. I think it's burst of particles that are released when spaceship goes out of warp.

https://www.universetoday.com/93882/warp-drives-may-come-wit...

16 day regular cycle is because it's regular transfer between two planets. Other FRBs are random because they are just random ships on random research routes.

I am going to frame your answer.

Love your comment. Sudden daydreaming for me.

Or they are heading straight for us but can only stay in warp for 16 days at a time.

Damn, those are some punctual aliens, never early or late.

Hah inevitably any type of FTL or even near light travel seems to inevitably also make an incredibly powerful weapon. This is even better because it doesn't require crashing the vessel into anything to destroy it.

I really like this view on the incompressibility of a uniform distribution.

However, the channel is probably not the same. Just as we moved from AM to FM, I'd expect then that a more advanced civilisation can make further improvements by changing the channel rather than the encoding, and maybe information encoded in the spin of particles or some other funky physics effect.

Ultimately, whilst the encoding will approach a uniform distribution, it seems like bad design for them to have an entire signal indistinguishable from background radiation, which is something they have to overcome as well.

Humans are not “advanced” in this context yet we already operate at the Shannon limit. Our channels are pristine compared to interstellar space. In order to send any information at interstellar distances you either need AGN levels of power or to slow your signal rate many orders of magnitude.

The inverse square law pretty much guarantees we're not going to ever detect random emissions from some alien civilization. The only way to get discernible signals an interstellar distance is with highly directional transmissions. So if we were to ever detect a signal from an alien civilization it would be a deliberate one.

There's properties of coherent signals that would stand out significantly from natural sources. Even if the message looked like noise the properties of the signal would give it away as being artificial.

A directional signal would be a point source so a simple off-axis check of the receiver would tell you if it's a distant or local signal. In order to maximize ERP a signal would also likely be a very narrow bandwidth. If we have a long enough sampling of a signal it'll also show distinct Doppler effects of coming from a rotating planet in orbit around a star. An intelligent civilization would also likely use some universal mathematical properties in the chosen frequency.

A signal, even one densely coded with information, would stand out against the background of space. Even if it appeared as noise due to dense information coding it would still have a narrow bandwidth, likely some non-natural mathematical properties, and come from a source that isn't some type of known magnetically active phenomenon (pulsar etc).

A signal deliberately sent to other stars is more likely to look like the Arecibo signal than WiFi. It would take an infeasible amount of energy, enough to vaporize the transmitter, to send a densely coded wideband signal that would be intelligible a handful of light years out let alone dozens or hundreds.

Depends what you call a signal. Imagine the earth was just a bit more focused on science, united, and thought a bit longer term (on the order of 100k years). Living sustainable, but curious and motivated to explore.

Imagine an earth where literacy, science, and peace were priorities. All countries joined a UN like organization and decided to spend 1% of the planets resources on exploring.

Assuming earth level tech and the next 1000 years of progress (.000007% of the life of the universe) we manage a ship/converted asteroid/similar that can travel for 500 years at 2% of the speed of light with some combination of power solar power lasers orbiting the sun, solar sails, nuclear power (ion, explosion and/or pulse), or antimatter. That gets us to the nearest 10 stars. Hard to say what earth technology will be like in 1000 years, but no new physics are needed. Would that be a huge ship with kitchens, gyms, entertainment etc? Or maybe just some autonomous AI with some genetic samples, some robots, and a 3d printer?

After another 1000 years things have improved further, and slowly spreading across the galaxy at a few % of the speed of light becomes ever more practical. Maybe just enough so that a 2nd solar system in the galaxy an start launching ships. Said 2nd solar system could make it dramatically easier for travel between those two systems since they could use a laser to help decelerate the incoming light sails.

Those huge solar arrays + lasers for pushing a large ship up to a few % of the speed of light make for a hell of a beacon. Even unintentionally it should be visible from quite a distance, even by astronomical standards. A ring of solar panels near a sun could provide considerably more power than is available on earth today, even crazy inefficient things like producing antimatter might become viable. Things like geo-engineering become feasible, some distant civilization might detect things like an implausibly quick change in the atmosphere of mars. Or unusually energetic particles coming out of ships trying to accelerate or decelerate to a few % of the speed of light.

As technology progresses we might even be able to focus the energy coming from the sun more directly to directly push ships to other stars. After all solar power -> laser is horribly inefficient.

Distant observers on the same plane as our solar system might even start to notice the particular pattern caused by the solar/whatever energy collection.

Earths atmosphere might start to exhibit unusual characteristics. Even nuclear propulsion becomes common and we start mining the outer reaches of the solar system that might exhibit some unusual characteristics.

Of course any similar civilization in the galaxy could similarly expanding and eventually the two civilizations would get close enough to notice each other, even without deliberate signals.

> But even if we assume radio waves, Shannon says that the you can send the most information when using a random code. As you approach the Shannon limit the signal becomes more noise-like as the encoding tends towards random.

The waveforms we currently use for high data rate communications are already pretty close to the Shannon limit. I'm willing to bet the aliens will tolerate a 30% reduction in data rate if it means using an identifiable signal.

Does Shannon's theorem really imply that the closer an information stream is to the limit, the more it resembles noise?

There's a simple proof of this; an optimally compressed signal will have all symbols occuring with equal frequency (otherwise you'd be able to compress it more). So yes, an optimally compressed signal is indistinguishable from noise.

However that's the theory; in practice the range of frequencies the signal occurs over and how strong it is at each frequency tells you a lot about whether it's likely to be an information carrier or not.

They send sognals to be heard and discovered. Why would they try hard to disguise if it's meant to be heard by another civilisation?

Let's apply common sense.

0) First, would it be a good idea to announce your world's existence? Robotic explorers who might exist in your neighborhood might decide your world is worth harvesting or colonizing.

1) If yes, then wouldn't sending a high-power announcement signal need to be obviously unnatural, i.e., containing very little data? Short trains of pulses with pauses that count up in binary perhaps?

1.1) "Visible light" seems like an anthropocentric assumption because there's not necessarily a reason other entities couldn't sense microwaves, IR, UV or even possibly fissile radiation. Furthermore, cybernetic "evolution" seems like there would be many possibilities for different senses. Maybe there is something particular efficient about the visible spectrum, such as polarized laser light remaining distinguishable extreme distances?

2) If said entity wanted to send a little more data as well, wouldn't they use a slightly different carrier or modulation but it wouldn't necessarily reach as far? Or, send it, as mentioned, more slowly?

Seems unlikely. Assuming a civilization could send a force to take over earth from many light years away, what could they possibly find of use?

Any element is easier to get elsewhere (without such a big gravity well) and no monkey descendants grumpily lobbing nukes at ya.

Sure, maybe they would harvest the asteroids and small moons while ignoring our complaints over sent over radio waves. No direct impact on earth. Maybe they would zorch a few suspicious looking sats with high energy signatures while they sat in high orbit and observed.

If we particularly amused them, or they thought we were uniquely tasty they could just grab a few of us and figure out how to replicate us.

Maybe our internet and fascination with cat videos would amuse them, some advanced crawlers running on quantum computers (for those pesky prime number based encryption) might cause some serious, but temporary disruption.

Maybe they would install a world leader with orange hair to sabotage any world wide scientific progress to help ensure we don't make significant progress any time soon.

But the sci-fi troupes of stealing humans to eat or vacuuming up the atmosphere or oceans seems exceedingly unlikely.

If anything they might drop a blackhole into the sun to harvest 10% of the mass that falls in and is converted into energy. That way then can use the energy for something more useful. But if that was common we'd be seeing lots of other stars disappear strangely.

All in all seems rather egocentric to think any really advanced civilization would particularly care about us.

Speaking of Shannon, unless we were the intended recipient, wouldn't an advanced civilization protect the transmission, so that we would not be able to eavesdrop and/or interpret it?

For those wondering about random codings, see Fountain Codes: https://en.m.wikipedia.org/wiki/Fountain_code

The military implementation is called Raptor Code (rapid tornado codings): https://en.m.wikipedia.org/wiki/Raptor_code

if they are sufficiently advanced, they would likely understand that other civilizations require primitive bands/encoding?

It could be considered a clue check. If you're unable to recognize and decode the signal, you failed the test.

Back in the days before radio, peple searched for life on the moon by using telescopes to look for signal fires.

Considering we don't have data to sample from, we have no idea what 'likely' is; That is the greatest challenge if "they're out there". We have no idea what to look for, because all we know about is from our own way of doing things.

Incidental transmissions sure but intentionally sent signals would by design avoid going too close to that limit, at least for the preamble that describes how to decode the normal transmissions.

>One thing that I wonder is whether we could ever identify an alien communication signal as an alien signal is likely to be much more advanced.

This has always been my thing.

Scientists: "Well we've been listening for radio signals but haven't heard anything yet!"

What if they discovered radio but decided it was way too easy for others to spy on it, and be detected off planet, so they just used wired connections for everything eventually moving to 99.9999999% efficient fiberoptics and then when they became a multi system empire, after discovering jump-space, they just switched to using fast messenger probes that could pop in and out of a system and carry large amounts of data to be tightly beamed to and from planets via laser.

Or what if they don't use radio because they began exploiting the Glarfnuralka stream for FTL communications via their subspace ansible implants 50 years after they discovered radio waves?

What if there's something,just right there that we've yet to discover that is far more efficient for communications from an energy standpoint? We discovered radio waves 150ish years ago, we detected the first gravity wave less than 5 years ago. What if it takes us another year, another 1,352 years, to discover Glarfnuralkians and develop a way to reliably transmit and receive data streams in that stream?

Or what if Glarfnuralkians are also considered to be primitive and inefficient due to the limited number of effective channels in the stream and the fact that random bits of the data area always lost no matter what and that the nearest intelligent hegemony is instead using Glurrybluron entanglement to transmit in real time between dedicated units that involve some cesium clocks, some variety of variable gravity lock, proper magnetic containment chambers for dual singularities (and we're talking relatively small singularities here, if your containment chambers are advanced enough you can get this thing down to a backpack sized device), an adequate power source (probably some sort of zero point module that draws from the quantum foam between universes), a way to inject electrons into the containment chambers and sufficient computational power from some universe-on-a-chip quantum ASIC.

Or maybe they're just really into face to face communication because of some sort of cultural/religious custom.

There is also the possibility a sentient species could have sensory organs that involve part of the radio spectrum which would likely prevent them from adapting radio communications.

They could be a species that uses chemical communication, not unlike many insects. You could theoretically convey large amounts of data this way with a moderate amount of evolution.

Plus dispersion of the signal would smear the modulation quite a bit, possibly making it even more difficult to detect and decode.

Except beacons of some sort. I could imagine practical benefits to not put a lot of information in there.

Reply on news.ycombinator.com