SETI may have a hard time finding anything at all, even if the universe is teeming with life. This article, which I picked up over at PhysOrg, cites new research which says that once you get good at compressing your data, radio transmissions become indistinguishable from those naturally emanating from stars. SETI looks for inefficient transmissions like the ones we currently create, but it’s likely that any civilizations out there who have had radio for more than the paltry 8 decades we have will be communicating in very efficient, compressed code.
Bummer. We’re probably staring at hundreds of really, really cool feeds from 20,000-year-old disc jockeys when we look at the night sky, but our gear doesn’t recognize the signal. (And we don’t have their public key, either.)
13 thoughts on “Sorry, SETI!”
We’re not that far from being at that level ourselves, actually…we just don’t do it that way because of massive infrastructure inertia. Imagine the HDTV cutover, with two orders of magnitude higher cost. (Aside: I still haven’t heard the question to which HDTV is the answer, unless you’re an equipment manufacturer.) Even now, however, we routinely generate communications signals that are indistinguishable from noise unless you’re listening in just the right way.
Actually really good compression will be as close to perfectly random as you can get, over a narrow range of points which will probably be a multiple of 2.
So if we see a narrow band of pulses, say 256, and each one has the same level of entropy (that being that every value occurs with the same probability of every other value) we probably found a data stream. It’s a very exact thing, almost too exact to happen often by accident.
Keep in mind that good encryption will have the same characteristics.
Of course SETI is looking for a single some other species sent out into the universe, not the radio chatter. Considering our control sample (that is us) did this once, for a very short time and in a specific direction the odds of SETI acutely finding said signal is pretty low.
Actually compression may not make as big of a difference as you (or the guy from PhysOrg) may think.
After all, at this time the only thing SETI is looking for is a carrier wave (I.e. an increase in signal strength over a given area of the RF spectrum).
They are not (yet) interrested in a message, just where to listen for one.
That’s fine as far as it goes – but with spread spectrum transmission, or just some forms of wideband modulation, there’s no good carrier wave to find.
Given the right technique, you can find cell phone traffic without knowing where to look first. A detector with sufficient bandwidth and sufficient dwell time – or a quasi-peak detector – will do wonders. I’m pretty sure they know about that stuff at SETI.
Jay’s right. SETI is SOL because with optimized radio comms as described in the article, you don’t HAVE a carrier.
Hey, maybe the glass is half-full. Wouldn’t this mean any signals we recognise are intended to be decypherable by backward Earthmen? So maybe we get “Hi, how are you” instead of “Foobax and the Monch trio are rocketing up the charts with ‘Baby I Wanna Hold Your Pseudopod’.”
There’s a lot of variables involved. We only know what our chances of finding a signal are if we don’t look.
(Currently working towards my 26,000th hour running the SETI@Home screen saver.)
I like how you put that 🙂
Signal to Noise
In Signal to Noise by Eric S. Nylund he promoted this as the plot device. They whole novel revolved around the protagonist picking information from out of what appears to be noise using his own filtering technology the message teachs him how to build a communicator ….
It’s the first place I heard of a one electron quantum computer and some other nifty things.
It’s great and I highly recommend it.
I won’t spoil the ending but it’s hot!
Said article seems to be cheerfully unencumbered by any hard data. Lacking that, it’s difficult to envision any rf data transmission method that would NOT be detectable as an indicator of intelligence if looked at the right way.
For example, somebody mentioned cell phones in this dicussion earlier on, using the fabulous buzzword “spread spectrum”, which is incidentally NOT spread spectrum per se, but a somewhat lesser version of ‘frequency agile’ techniques used by the military.
Now, said cell phones operate BECAUSE they operate around common center frequencies within a fixed bandwidth. Individual packets are sent at different frequencies depending on what is available at the time, but they all cluster around that center frequency. Using the right arpeture size and dwell time, one can easily view a cell service’s activity in a plain old spectrum analyzer. Nothing fancy is required, and I’m pretty sure that the stuff SETI employs qualifies are “pretty darned fancy”.
It doesn’t matter if we can tell what it says – if we can discern that there is a cluster of activity over time around a center frequency, we will have established evidence that there is possibly intelligence behind it. That’s all that SETI’s mandate requires.
Granted, eetee might not have cell phones, the FCC to regulate the spectrum, or anything else we might recognize, but we’re not playing with certainties here – we’re looking at statistical probabilities, and those indicate that there’s a good chance that we will find multiple instances of something reckognizable given a sufficient search for it. Drake equation, et cetera, et cetera.
Something else said article neglects to take into account is that in order to GET to those cell phones or something slicker, a civilization will most likely progress through amplitude, frequency, and phase modulation techniques same as we did.
That’s actually an assumption that goes back to before the times of SETI. I recall reading something Asimov wrote about this when SETI was being considered, and he reckoned that we had around a 100 year window AT BEST to catch a civilization’s emissions before they went totally to narrowband satellite or optical, neither which is condusive to the goals of SETI. So in a way this is extremely old news (the book I read this in was published in the 70’s and was from collected essays from before that time).
The essence of this article is “it ain’t gonna be easy”, to which the jade SETI veteran might reply, “what took you so long?”
Now, if more data were available, I’d perhaps be less cynical about the worth of the article, but on the weight of its merits alone, I’m disinclined to bow to its logic.
I didn’t say “cell phone”. You did. I’m well aware of the details of cellphone CDMA emission, as well as true spread spectrum. True spread spectrum emissions are indistinguishable from noise if the spreading method and some idea of the sequence parameters aren’t known. The only reason I haven’t looked at it on a spectrum analyzer is that mine doesn’t go above 1 GHz.
The only way to detect such activity is to monitor the noise floor and look for an increase in its level. This, in turn, only works if you have a “before” figure to compare to.
To me, the best way to think of it is that each civilization, once it reaches the stage of radio, puts out a spherical shell of detectable emissions n light-years thick, where n is the length of time it takes them to go from first transmission to widespread adoption of non-radiating or undetectable transmission methods. We haven’t established our own value of n yet. Even so, however, the implication is that we will only be able to detect emissions from outside our own solar system that happen to have civilizations pass through their n years however far in the past corresponds to the distance from their home to ours. This is a chancy thing, at best. I don’t share the article’s degree of pessimism, but it is clearly going to be pure luck if we do find something that way.
A major cornerstone of SETI is being in the right place at the right time. The assumption is that we have a small window. The asumption is that the galaxy is big – REALLY BIG – and that even with a pessimistic estimate, the odds are good that there’s something to see if one is looking.
Incidentally, (talking shop), the ability to detect a spread spectrum emission depends on the amount of dithering applied. The assumption is that the entity generating it doesn’t care if it can be detected so much. Otherwise, they dither the stuffing out of it and it’s hard to detect 3 meters away, much less a few light years’ distance.
(Please note that as I write this, I haven’t yet read the article. I’m just making this up as I go, so feel free to ignore it as a totally uninformed opinion.)
I noticed none of you mentioned UWB, which as I understand it is sort of a super spread-spectrum. Because of the way it works it inherently looks like background noise unless you have the key to receive it. And that’s not counting any compression or encryption you might perform on the data.
It has limited distance, though, so that particular technique is unlikely to the basis for an entire civilization’s telecommunications. Still, if we can do that then there may be (almost certainly are) even better techniques that we haven’t discovered yet.
But I think there’s still hope for SETI, for a couple reasons. First, biological beings tend to follow certain patterns, and while I would not assume that our characteristics are definitive, I’m inclined to think that we’re fairly typical. An organism will naturally try to get the most advantage from the least effort. Anything else is inefficient and will tend to be overcome by something more efficient in the long run. But that includes the fact that established organisms will tend to be conservative and stick with what works. We have old-fashioned broadcast radio playing side-by-side with the latest emerging WiFi and cell phone standards. Why? Because it works and it’s cheap, and we have a lot of resources invested in the infrastructure. Why throw it away while it’s still useful? Certainly it’s on the wane, but it will be a while before it’s dead. And even then, how long before we lose cheap walkie-talkies and FRS radios?
I could go on about that (I am a pedant) but I’ll cut it short.
My point is that new technologies will supplant the old, but never entirely, and it will take a while. And the hobbyists will doubtless take over what the commercial broadcasters abandon.
Yes, that may reduce the amount of open traffic to the point where it’s indistinguishable from noise. It might not. I don’t know. And I don’t know how long inertia will keep us using the old-fashioned stuff.
But the other thing is that no matter how much you compress or encrypt, intelligent communication (and that includes Aunt Marge gossiping on the phone, I’m not judging the content 🙂 consists of patterns. If you completely remove the patterns, then it’s totally random, and there are no data, there’s no intelligence. Discerning patterns in highly compressed data might be difficult, but as we go on we get better at it.
I don’t know. Maybe our hypothetical aliens will start coming up with better shielding and highly directional transmissions for everything, or all broadcast will be short-haul and all long-haul stuff will go through fiber (the atmosphere blocking the short-haul stuff) and the gosh darn planet just won’t radiate.
I’d also hope there’s more than one other planet with intelligent life and we might find them in different stages of development. We might miss the really advanced one but find one a bit more primitive.
I don’t know. But I do know that harry_whodunnit was right when he said, “We only know what our chances of finding a signal are if we don’t look.”
Comments are closed.