wcelliott
9th November 2007 - 02:41 PM
QUOTE
So it can be used for all these things, but according to you, it can't be used to detect the difference between a drum beat and C4 exploding?? hahahahahahaha!! lol
You should speak with Wcelliot too, according to him, he was working on a sonogram program which would be able to identify small aircraft by the engine/prop noise.
http://forum.physorg.com/index.php?showtop...ndpost&p=275207QUOTE ("Wcelliot")
I was PROGRAMMING sonogram software for IDENTIFYING small aircraft by their engine/prop-noises back in 1985.
How the hell does a sonogram detect the difference between engine/prop noises, yet according to you Trippy, it can't detect the difference between a wire snapping, a drum beat and C4 going off??
That must mean they share the same frequency, amplitude etc etc.....lol
Well, Stundie, nobody can make you look as stupid as you can, yourself.
Trippy's statement is basically correct, impulse events all share the same basic characteristics visible in sonograms, and that is a simultaneous vertical stripe of energy across all frequencies from the zeroth-order all the way up. As I've explained to AlK
ad nauseum, explosives detonate really quickly, a lot faster than any audio recorder can fully characterize. C-4 detonates at over 26,000feet/second, so a four-inch block (enough to transect a steel beam) would start-to-finish, detonate in 10 MICROSECONDS.
Maybe you can do the arithmetic that AlK hasn't been able to do, yet, and tell me what the sampling interval associated with a 40kHz sampling rate? (Hint: 1/sampling rate = sampling interval) So how many samples do you expect to have for a 10 MICROSECOND event?
Hint: ___________|_______________
That's what an explosive detonation looks like in the time domain, absent any reflections of the explosion off the local environment. Reflections will be attenuated by several factors, and will always be delayed, so that same ____|____ will bounce around off walls, floors, ceilings, etc., and each reflection will be delayed and smaller, and they'll all add up to that "characteristic N-shape".
The vertical spike is the detonation, the downward "\" slope is just the reflections off the environment, which means that it's ARTIFACT, not an intrinsic CHARACTERISTIC of the detonation itself.
All impulse-driven events have that _______|________ shape in time, and any impulse that reflects around inside a building will sound like an impulse reflecting around inside a building.
As I've explained to AlK a dozen times.
BTW, Prop aircraft have specific numbers of prop-beats/rotation of the engine, and engines in aircraft are either piston or jets, and each sound different. You can count the number of piston-engine exhausts per prop rotation, which will be characteristic for each type of aircraft manufactured, and jet engines have a characteristic frequency associated with the engine rpms and compressor blades/"exhaust buckets", and some planes have jet engines driving props, so there's that characteristic, and different planes have different numbers of engines. All of those sounds reflect off their airframes, which is aspect-dependent. That's how aircraft can be identified (mostly, some planes' characteristics overlap).
And the software that Trippy provided links to were pretty similar, basically, to the software that I'd written way back in 1985 when you were probably still in diapers (when I was a Senior Staff Scientist for Science Applications International Corporation). That was about the same time I designed a digital signal processor to do acoustic beamforming. From scratch, with adder-chips. Under a million-dollar contract based on a ten-page white-paper I wrote.