h2g2 The Hitchhiker's Guide to the Galaxy: Earth Edition

You might have mentioned that a Fast Fourier Transform is used as the primary analysis routine of SETI@Home, and I believe they're also used to make spectrum analysers and fancier visualisation effects on computer music players. But still, I enjoyed the article!

It's used by lots of folks writing audio analysis software, SETI probably being the most well known. I use a lot of the algorithms for the speech recognition software I'm writing.

Computer, send message...

I think "FM Synthesis" must be related to all this, as in Yamaha's classic DX7 synthesizer form the early 80's. Here is one writeup:

QUOTE

Frequency Modulation (FM) is where the output of one oscillator is used to modulate the pitch of another, the oscillators being called Modulator and Carrier respectively. "Modulate the Pitch"... that's the key phrase! The pitch of the Carrier is being changed (modulated) in tandem (in sync/ going up and down at the same time) by the Modulator.
Think of it as one person singing and another person grabbing the throat of the first and shaking him in a rhythmic manner; the singer being the Carrier and the throttler being the Modulator.

END QUOTE

FM synths are notoriously difficult to program, and just setting random values to the parameters typically results in nothing usable. Casio's competing "Phase Distortion" synths, e.g. the CZ-101 from about the same time, is more likely to give a useable sound with random parameter settings. But some of those FM sounds are great, like the famous electric piano sound. My wife, on first hearing the e-piano, said "That's the "Doogie Howser" sound (from the TV show).

I read an article recently about the development of focused sound, almost like an audio laser. Using ultrasonic frequencies, it is possible to direct a more discrete sound "beam" than with audible frequencies, which have a longer wavelength. Of course, under normal circumstances, the human ear can not hear ultrasound. However, ultrasound frequencies can be distorted by atmospheric conditions to produce audible noise. Using complex algorithms, which must be related to Fourier Theory, researchers managed to create a specific audible sound (i.e., a spoken word) from an original ultrasound frequency. The potential applications of this include advertising pitches directed to individual customers in a store, eliminating the need for radio communication headsets, or to confuse enemy soldiers.

Sorry to carry on, and I'll apologize in advance for any mistakes in the above, but thought it was an interesting tangent.

If I recall, they had to stop testing of this because the ultrasound was making people vomit or in some other way making them ill. Ho hum. But I think I saw first mention of it in New Scientist - as a "sound beam" just like you describe.

Martin - "Any similarity between my memory and Real Life is purely co-incidental"

Good article, another application of Fourier Transformation that maybe deserves as mention is interpretation of NMR spectroscopy.

The first thing a chemist does on recieving an NMR spectra is to run a Fourier transform on it - all of a sudden proper peaks representing various H atmos (normally) leap to the fore. The amount of time that wee bit of maths done by computer saves in terms of trying to clarify the signal using other means is immense.

The ultrasound thing has now been developed into a crowd control 'weapon', as far as I know. Not sure if it has been used in anger yet.