On fre, 2004-07-16 at 21:42, Tim Blechmann wrote: Is that a single person at a time or several people walking around at random? With a samplerate of say 48k and two microphones you can measure phase differences down to 1/48ms or 330m/48000 = 0.7cm I don't know what kind of soundsource you are facing, but if it is anything like this: http://www.huutajat.org/HyvaMP3.mp3 ... you are bound to fail ;) mvh // Jens M Andreasen

Hi, Friday, July 16, 2004, 9:42:45 PM, you wrote: TB> i'm currently working on an installation, where i have to track persons TB> in a room ... TB> in theory it should be possible to track a person by the time the sound TB> will need to reach the microphone ... i know, these are only ms, but TB> using a tricky algorithm, it might be possible ... TB> has anyone been trying to work on something like that? if so, i'd like TB> to hear about the results ... For all I know, it is quite tricky and would require multiple microphones. A student at the TU Delft, Sammy Amadja, did his Master's thesis on this problem. If you ask at the TU Delft: www.soundcontrol.tudelft.nl, they should be able to send you the thesis as a .pdf or so. I hope this helps. If I can find back the thesis here on my system, I will send it to you, as I *should* have it somewhere... sincerely, Marije --