So ET won't be picking up the Olympic Games in 1936 Berlin?

May 12, 2009 11:18 AM

Do radio waves attenuate and become noise or do they go on forever?

In Carl Sagan's Cosmos and his novel Contact, he writes how radio waves go on forever. I just finished seeing the Discovery Planet special "Life after People", where in the closing segment, the narrator states that new research has shown that radio waves probably attenuate after 4 light years to become nothing more than noise. This occurs because of cosmic dust, radiation, planetary systems, asteroids, ect.

I thought radio waves go on indefinitely in a vacuum, and space is mostly a vacuum because of the distances involved, even with all the cosmic dust and whatnot.

So has another cherished belief of mine hit the dustbin?

posted by MrMulan to science & nature (14 comments total)

Do the two beliefs have to be mutually exclusive?

In other words, you need two points for radio waves to be useful. An origination point and a receiver. As long as there is no interference between the two then the vacuum maintains the signal. But if there is then the signal can attenuate. It really seems that in a perfect condition you could extend the signal.

I'm by far no expert on this, just a laymans guess.
posted by bitdamaged at 11:29 AM on May 12

I believe they get exponentially weaker as you get farther from the source. Something to do with the quadratic equation. Anyway, yeah, aliens would need to have hyper-sensitive receivers to hear anything from us, unless we target at a specific region of the sky.
posted by wastelands at 11:33 AM on May 12

...and I think they eventually get lost in the background noise left over from the big bang.
posted by hippybear at 11:39 AM on May 12

In a way, they are both true. Imagine ripples on a pond (this is a slightly misleading metaphor in that electromagnetic radiation does not have a medium like water, but for our purposes it's fine) being used to send a message. You have a buoy on one side of the pond that you jiggle up and down and there is another one that your friend watches on the other side. Imagine lots of people are doing this - it would become difficult to interpret the resulting pattern of waves and recover the signal you are interested in. One way to make it easier is to filter all but one frequency of jiggling and then encode your information by varying the amplitude (how much you move the buoy up and down). A somewhat more sophisticated, but harder to understand, way to send a signal so it can be seperated from the noise is Frequency Modulation.

Anyway, like waves on a pond radio signals get much harder to detect with distance from the transmitter - they are half as powerful for each doubling of distance. Does the signal ever half no magnitude? Well, here we have to get a bit philosophical. Assuming the amplitudes of radio waves are truly continuous than mathematically speaking you can halve them indefinately and it will get indefinately small. But it never reaches zero. It just gets infinitely close to zero. It gets closer and closer and closer, as far as you want to take it, but it never gets there. See Zeno's Paradox.

Practically speaking, though, there is a sort of universalnoise floor. Imagine our pond is not a still pond, but one that is constantly agitated by all manner of wave creating thing going on in it. Well, at some point the amplitude of the signals we send out are significantly lower than the average amplitude of random ripples on the pond. To try my hand at a saganesque saying: Our voice gets lost in the great.... cosmic... conversation.
posted by phrontist at 11:41 AM on May 12

In a perfect vaccum, a radio wave will indeed carry on forever, or at least until it hits something solid. When it hits something solid it's likely that some of its energy will be absorbed. This energy might be transformed into kinetic energy (heat or movement), or might be re-emitted to continue as it was before. Almost certainly a compromise somewhere between the two, which means that the signal will continue, but weakened slightly.

Another big effect is the weakening of the wave as it spreads out. As the transmission gets further from earth it'll tend to spread out, just like the light from a torch spreads out over a wide area from a tiny bulb. So as the signal travels away from earth, it has the same amount of energy spread over an ever-increasing area. There's a nice explanation of the logic here, and the bottom line is that every time you double the distance from Earth, the radio signal's strength seen from that point is reduced by a square root.

Even after the signal has been partially absorbed by dust and spread thinly over a huge area, it does still exist. The problem is that the universe is full of radio noise emitted by stars and excited gases, left over as echoes from the Big Bang, etc. The point at which we can say the signal is completely attenuated is when we calculate that the signal has been weakened so much that it's indistinguishable from the background noise of the universe.
posted by metaBugs at 11:45 AM on May 12

They attenuate, and they go on forever. Eventually the amplitude declines to the point that it is indistinguishable from background noise, but the signal is still there. (It becomes part of the background noise.)

(It's a quarter as powerful for each doubling of distance.)
posted by Chocolate Pickle at 11:46 AM on May 12 [1 favorite]

Does the signal ever have no magnitude?

It's also worth pointing out that we know today that electromagnetic radiation is quantized, not continous. So at some point, practicalities of information recovery aside, it seems to me as though your wave would truly cease to propagate - but I know very little about QM, so I have no idea whether that interpretation is correct.
posted by phrontist at 11:50 AM on May 12

.... however, signals below the noise floor can be recovered somewhat. Not all signals completely disappear in noise.

The modes we use to encode intelligence on a carrier signal are primarily AM and FM (amplitude and frequency modulation), sometimes phase modulation. There's nothing sacred about these modulation formats. There are others.

Digital signal processing (DSP) techniques exist that spread signal out over a wide band and make it possible to discern the intelligence by mathematics. So in these cases (which go by lovely acronyms like Orthogonal Frequency Division Multiplexing (OFDM or COFDM), signals are routinely recovered below the local noise floor.

Practically speaking, to recover them, you need to know the pseudorandom parts used to spread the signal, so someone on the other side of the universe would see the signal as only noise, unless they knew when/where to look for the pieces.

This is one of those things that I sometimes mention to folks who want to know what math is good for. Every time you use a cell phone, there is a ton of DSP crap going on that does this very trick. Way cool.
posted by FauxScot at 11:51 AM on May 12 [2 favorites]

I believe they get exponentially weaker as you get farther from the source

As phrontist notes, it's an inverse square law, not an exponential decay. (I've begrudingly resigned myself to the colloquial use of "exponentially" for functions that are not mathematically exponential in most cases, but since we're talking actual physics here it seems important to make the distinction.)
posted by DevilsAdvocate at 11:54 AM on May 12

(Argh, yeah, it's a quarter as powerful with each doubling.)
posted by phrontist at 11:58 AM on May 12

There are lots of natural sources of radio waves, which your radio receiver will happily convert into sound. Elelctrical storms on earth and on Jupiter both emit part of their energy as radiofrequency photons. There are enough astrophysical sources that it's profitable to build radio telescopes.

If you build a radio transmitter that sends energy in all directions, that energy is spread over the surface of a sphere centered on the transmitter. As time passes the sphere gets bigger, and so the same energy is more thinly spread out. The area of a sphere goes like its radius squared, so every time the radius of the sphere doubles, the energy density on the surface falls by 75%. Suppose there was a radio station you could just barely get from 100 miles away. From 200 miles away it would be 4 times too weak for you to get with the same receiver; from 300 miles away it would be 9 times too weak for you to get, and so on. A light year is 6×10¹² miles; at a light-year away this same radio station would be too weak for your receiver by a factor 3.6×10²¹. At this point you can start to wonder how many individual radio photons your original signal had, and how big your detector would have to be to interact with just one of them.

You can of course build an antenna that sends all the radio signal in a "beam" going generally the same direction, so that the power density doesn't change so much as the signal travels. But even then you'll get divergence in the beam.

The calculation that broadcast signals fall below the general galactic radio noise after about four light-years is something I only heard about recently. I'm surprised it's so small, but I'm not astonished.
posted by fantabulous timewaster at 12:11 PM on May 12

For example, right now I'm listening to this radio station which sends out 1.6×10³⁰ photons per second. Even if none of them land in my radio, at a light-year away that's one per square meter every thousand seconds. An Arecibo-sized dish would get sixty or eighty photons per second from my radio station. Not much.
posted by fantabulous timewaster at 12:37 PM on May 12 [1 favorite]

One other thing to consider is that the inverse square law only applies if the radio waves are being emitted isotropically (i.e. with no directionality). If you were to target a beam of radio waves at something, the drop-off in signal would be a lot less depending on how tightly you can confine the beam. That means that ET would have a much better chance of detecting a targeted signal, much like we'd have a much better chance of detecting ETs that are targeting us. There isn't a whole lot of hope in detecting just the general communications radio signals from other planets but if they are specifically targeting us (maybe they think we might be here because they saw our planet transit the Sun with their equivalent of the Kepler satellite---if we can be completely anthropocentric for a second :-) it would be much easier to detect them.
posted by kms at 1:27 PM on May 12

Broadcast radio isn't of that much interest from a SETI perspective. Military radars are the more powerful signal. Which is maybe even more depressing than the possibility of remote civilisations picking up transmissions of "I'm a Celebrity, Get Me Out Of Here!"
posted by edd at 4:53 AM on May 14