Remote audio surveillance - possible by 'Laser Dopplering' of air molecules?

Sorry for the clumsy title; there’s only so much space available in the title box.

What I am trying to ask is whether remote audio surveillance (of, say, a room in some distant building) is possible by examining the motion and vibration of the air molecules in the room.

Phrased differently (and I hope more clearly), let’s assume that the air molecules in a room are moving around in a random manner. Then, if people start talking in the room, the sound waves made by their speech should jostle the room’s air molecules in a way that imparts to them various non-random motions and vibrations. Presumably, these non-random motions will be a function of the sound waves created by the occupants’ conversation and, thus, in some sense ‘encode’ the conversation.

Could you not aim a laser (?laser doppler) into the ‘empty space’ of the room to detect and analyze the motion of the room’s air molecules, and thereby detect and decipher what is being said there?

You’re probably aware that lasers can be aimed at any smooth, reflecting surface in a room to capture the sound waves of what’s being spoken therein (the classic example is bouncing a laser off one of the room’s windows). But that’s old stuff. What I’m interested to know is whether you even need a smooth, reflecting surface anymore. Can you just bounce your laser off the room’s air molecules?

I doubt I’m being as clear as I would have wanted to be, but am pretty sure you’ll still get what I’m talking about.

Can it be done? Is it being done?


It can be (and has been) done by pointing the laser at a window, and then looking at the modulated return beam.

I don’t see how you could do it with just air, though. Maybe using dust in the air.
But, you would need to time-gate the return signal to filter out the near and far reflections.

In principle, you could use Rayleigh scattering off of the air molecules and look at the variations in the intensity of the light due to increased or decreased density. The problem is that for most sounds this fluctuation is incredibly small; for a 120 dB sound it would be about one part in 5000, and for every 20 dB below that the density fluctuation would decrease by a factor of 10. So a normal conversation at about 60 dB would correspond to intensity fluctuations in the scattered light of about one part in five million.

Doppler shift is a no-go. The individual air molecules in a room are each moving at a pretty good clip (hundreds of meters per second) in basically random directions. By contrast, the effects of the 120-dB sound wave cause the molecules to add an extra few centimeters per second onto their speed. If you were to look at the change of frequency due to the motions of the molecules, all you would see is the Doppler broadening due to their random motions; I doubt that you could see the overall shift in the frequency due to the sound wave on top of that.

That was extremely informative MikeS and I apologize for taking so long to acknowledge your post.

Now, back to the drawing board . . .

Just checking the wiki entry on Laser Microphones…and, apparently, someone actually has built a prototype laser microphone that uses a stream of smoke as a “diaphragm.”

The device shown and demonstrated in the video, however, is intended as a prototype for a studio mic, not a surveillance device.