Subject: Re: [linux-audio-dev] sound cancelation with anti-sound
From: Robert Jonsson (robert.jonsson_AT_dataductus.se)
Date: Thu Apr 04 2002 - 14:35:52 EEST
Hi,
Haven't read the article yet. However, this is a technique sometimes used in
very loud industries where sound is predictable as to location and content.
There might be issues with this approach with placement of the cancelation
speaker. It has to be placed VERY close the originator of the noise. If you
don't get it right you could very well end up boosting the noise at some angles
rather than the opposite.
At work we have an extremely loud fan in our lunch room. I've been thinking
about this same issue and it's possible application on this fan.
This far I'm thinking it will be too hard to implement in reality. Getting the
noise source cancelled out at for all possible listeners will not work unless
the canceller is mounted, somehow, all over the vent, which is quite large.
Though, I haven't made any calculations, if the fans you want to silence are
sufficiently close together and you mount the speaker in between you might be
able to handle it. I guess you won't know until you try it :-)
Since the noise is fairly constant it should be quite possible to invert it,
possibly even without software.
.02 Euro
/Robert
Maarten de Boer wrote:
> i have been thinking about the following before, and now
> that i read about it on slashdot, and it got me thinking
> a bit more.
>
> the issue is sound cancelation. the article mentioned
> on slashdot,
> http://www.newscientist.com/news/news.jsp?id=ns99992094
> talks about a (very expensive) general approach of
> removing sound with anti-sound.
>
> personally, i am more interested in a very specific
> noise: the noise of the cooling fans in my pc.
>
> i am sure it would be possible to have a background
> application running that uses the soundcard (could be
> a cheap one) to do this. the sound of each fan has
> almost constant pitch, and i suppose it is harmonic.
>
> the application should record the noise, analyse it:
> determine the pitch of each fan and seperate the sounds
> of all fans, invert each signal, and play it back.
> this should work reasonably well as long as the pitch
> doesn't change. by reanalising constantly, any changes
> in the sampled signal have the be analyses and added
> to the canceling inverted signal.
>
> the problem is the sound seperation. The application
> should be low on CPU usage, so I guess time domain
> processing is the only option. but we are talking about
> let's say 3 different signals with rather simple content...
>
> any suggestions / remarks?
>
> maarten
>
>
This archive was generated by hypermail 2b28 : Thu Apr 04 2002 - 14:23:04 EEST