Hello experties, Any body have an idea how to calculate the inverse impulse response in the time domaine of an impuse response h using MATLAB ? I'll be happy if you help me. Thank you in advance ----- Stef
Inverse impulse response
Started by ●March 31, 2006
Reply by ●March 31, 20062006-03-31
stef wrote:> Hello experties, > > Any body have an idea how to calculate the inverse impulse response in the > time domaine of an impuse response h using MATLAB ? > I'll be happy if you help me. > Thank you in advanceYou can't do it in general. Dividing by very small values sees to that. There's material on deconvolution in Smith at http://dspguide.com (Chapter 17 p.300) that will give you perspective about what is possible. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●March 31, 20062006-03-31
stef wrote:> Hello experties, > > Any body have an idea how to calculate the inverse impulse response in the > time domaine of an impuse response h using MATLAB ? > I'll be happy if you help me. > Thank you in advanceYou probably want something like this: http://www.ramsete.com/aurora/inverse.htm
Reply by ●March 31, 20062006-03-31
>stef wrote: > >> Hello experties, >> >> Any body have an idea how to calculate the inverse impulse response inthe>> time domaine of an impuse response h using MATLAB ? >> I'll be happy if you help me. >> Thank you in advance > >You probably want something like this: > >http://www.ramsete.com/aurora/inverse.htm > >Hello, I used Levinson-Durbin algorithm. I used the MATLAB source code in the following adress: http://www.dsprelated.com/compdsppost.php?id=55071 The problem is: The approximation of the inverse impulse response is a noise. Have any one used this algorithm? Do you have an other idea how to solve this problem, or other suggessions? I'll be happy for any answer. Thank you ---------- Stef
Reply by ●March 31, 20062006-03-31
>>stef wrote: >> >>> Hello experties, >>> >>> Any body have an idea how to calculate the inverse impulse responsein>the >>> time domaine of an impuse response h using MATLAB ? >>> I'll be happy if you help me. >>> Thank you in advance >> >>You probably want something like this: >> >>http://www.ramsete.com/aurora/inverse.htm >> >> >Hello, >I used Levinson-Durbin algorithm. I used the MATLAB source code in the >following adress: http://www.dsprelated.com/compdsppost.php?id=55071SORRY: I mean this adress: http://www.musicdsp.org/showone.php?id=188>The problem is: The approximation of the inverse impulse response is a >noise. >Have any one used this algorithm? >Do you have an other idea how to solve this problem, or other >suggessions? >I'll be happy for any answer. >Thank you >---------- >Stef > >
Reply by ●March 31, 20062006-03-31
"stef" <kabbaj@gmx.de> wrote in message news:r9ydnRHw1YvoG7DZnZ2dnUVZ_vqdnZ2d@giganews.com...> Hello experties, > > Any body have an idea how to calculate the inverse impulse response in the > time domaine of an impuse response h using MATLAB ? > I'll be happy if you help me. > Thank you in advance > ----- > StefIf it is acoustic then the chances are that it will be non-minimum phase and won't have an inverse.There are a number of papers on this however. Tam
Reply by ●April 1, 20062006-04-01
stef wrote: ...> I used Levinson-Durbin algorithm. I used the MATLAB source code in the > following adress: http://www.dsprelated.com/compdsppost.php?id=55071 > The problem is: The approximation of the inverse impulse response is a > noise. > Have any one used this algorithm? > Do you have an other idea how to solve this problem, or other > suggessions?Sounds like a case for ... SuperBob! :-)
Reply by ●April 1, 20062006-04-01
> >stef wrote: > >... >> I used Levinson-Durbin algorithm. I used the MATLAB source code in the >> following adress: http://www.dsprelated.com/compdsppost.php?id=55071 >> The problem is: The approximation of the inverse impulse response is a >> noise. >> Have any one used this algorithm? >> Do you have an other idea how to solve this problem, or other >> suggessions? > >Sounds like a case for ...What do you mean SuperBob...? If you don't understand my question, please ask me or pay attention what the real DSP experts write. -------- stef> >SuperBob! > >:-) > >
Reply by ●April 1, 20062006-04-01
Hi, Tam , Can you say a little bit more about acoustic and nonminmum phase ? If possible, what papers talked about this inverse problem ? Thanks in advance, Xinhui
Reply by ●April 1, 20062006-04-01
>Hi, Tam , > > Can you say a little bit more about acoustic and nonminmum phase ? If >possible, what papers talked about this inverse problem ? > > Thanks in advance, > >Xinhui > >Hello, Xinhui, I found two papers in internet talking about this problem: -Acoustic characterisation of virtual musical instruments using MLS technique on ancient violins. From Angelo Farina, Andreas Langhoff, Lamberto Tonchin. -Software implementation of an LMS analyzer with tools for convolution,auralization, and inverse filtering. From Angelo Frarina and Fabrizion Righini. I've not yet solved the problem! If you have an idea how to solve it, please write a message. Thank you in advence. ----- Stef
