Out of convolution operation and inverse filtering operation,which operation is better regarding the stability point of view(minimum phase system), if i introduce noise into system. Thanks Riz
Selection between convolution and inverse filtering operation
Started by ●September 19, 2007
Reply by ●September 19, 20072007-09-19
On Sep 19, 9:49 am, "riz" <rizwan....@gmail.com> wrote:> Out of convolution operation and inverse filtering operation,which > operation is better regarding the stability point of view(minimum phase > system), if i introduce noise into system. > > Thanks > RizDo you mean "deconvolution" instead? From a practical point of view, inverse filtering is not usually advisable when noise is present, as the filter will amplify any noise present at frequencies that were attenuated by the original filter. In that case, a deconvolution algorithm might be more appropriate. Jason
Reply by ●September 19, 20072007-09-19
>On Sep 19, 9:49 am, "riz" <rizwan....@gmail.com> wrote: >> Out of convolution operation and inverse filtering operation,which >> operation is better regarding the stability point of view(minimumphase>> system), if i introduce noise into system. >> >> Thanks >> Riz > >Do you mean "deconvolution" instead? From a practical point of view, >inverse filtering is not usually advisable when noise is present, as >the filter will amplify any noise present at frequencies that were >attenuated by the original filter. In that case, a deconvolution >algorithm might be more appropriate. > >Jason > >No,actually i can get my desired goal by two methods which are independent.one method involves inverse filtering,other method convolution.
Reply by ●September 19, 20072007-09-19
On Sep 19, 4:16 pm, "riz" <rizwan....@gmail.com> wrote:> >On Sep 19, 9:49 am, "riz" <rizwan....@gmail.com> wrote: > >> Out of convolution operation and inverse filtering operation,which > >> operation is better regarding the stability point of view(minimum > phase > >> system), if i introduce noise into system. > > >Do you mean "deconvolution" instead? From a practical point of view, > >inverse filtering is not usually advisable when noise is present, as > >the filter will amplify any noise present at frequencies that were > >attenuated by the original filter. In that case, a deconvolution > >algorithm might be more appropriate. > > No,actually i can get my desired goal by two methods which are > independent.one method involves inverse filtering,other method > convolution.Convolution and filtering are the same thing, so what is the distinction you are making? Are you distinguishing between FFT-based filtering vs. convolution-based filtering? Or inverse-filtering vs. Wiener-filtering (deconvolution)? -- Oli
Reply by ●September 19, 20072007-09-19
>On Sep 19, 4:16 pm, "riz" <rizwan....@gmail.com> wrote: >> >On Sep 19, 9:49 am, "riz" <rizwan....@gmail.com> wrote: >> >> Out of convolution operation and inverse filtering operation,which >> >> operation is better regarding the stability point of view(minimum >> phase >> >> system), if i introduce noise into system. >> >> >Do you mean "deconvolution" instead? From a practical point of view, >> >inverse filtering is not usually advisable when noise is present, as >> >the filter will amplify any noise present at frequencies that were >> >attenuated by the original filter. In that case, a deconvolution >> >algorithm might be more appropriate. >> >> No,actually i can get my desired goal by two methods which are >> independent.one method involves inverse filtering,other method >> convolution. > >Convolution and filtering are the same thing, so what is the >distinction you are making? Are you distinguishing between FFT-based >filtering vs. convolution-based filtering? Or inverse-filtering vs. >Wiener-filtering (deconvolution)? > >-- >Oli > >actually,i have two different methods to achieve same objective.One method uses convolution/filtering/fft-based filtering.other uses inverse filtering.
Reply by ●September 20, 20072007-09-20
On Sep 19, 10:50 pm, "riz" <rizwan....@gmail.com> wrote:> >On Sep 19, 4:16 pm, "riz" <rizwan....@gmail.com> wrote: > >> >On Sep 19, 9:49 am, "riz" <rizwan....@gmail.com> wrote: > >> >> Out of convolution operation and inverse filtering operation,which > >> >> operation is better regarding the stability point of view(minimum > >> phase > >> >> system), if i introduce noise into system. > > >> >Do you mean "deconvolution" instead? From a practical point of view, > >> >inverse filtering is not usually advisable when noise is present, as > >> >the filter will amplify any noise present at frequencies that were > >> >attenuated by the original filter. In that case, a deconvolution > >> >algorithm might be more appropriate. > > >> No,actually i can get my desired goal by two methods which are > >> independent.one method involves inverse filtering,other method > >> convolution. > > >Convolution and filtering are the same thing, so what is the > >distinction you are making? Are you distinguishing between FFT-based > >filtering vs. convolution-based filtering? Or inverse-filtering vs. > >Wiener-filtering (deconvolution)? > > actually,i have two different methods to achieve same objective.One > method uses convolution/filtering/fft-based filtering.other uses inverse > filtering.That still doesn't make sense. "Inverse filtering" is still filtering, so to carry out "inverse filtering", one must use either convolution or an FFT-based approach. So I'm afraid I still don't see the distinction you're making here. -- Oli
