Hi, Thank you for the suggestions, I will try it via the fft. Regarding the noise it is hard to assess how it looks like. To be more specific about what I want to do: I acquired fMRI data and trial averaged all the data, resulting in a single response profile for a specific event. This profile could be assumed to be the convolution of a neuronal response with a hemodynamic response function represented by a gamma kernel. However, the underlying neuronal activity is not just the response to a distinct event, but would be more complex, thus my idea to obtain the response by a deconvolution. With the extensive trial averaging I am hoping to cope with the severe noise, so that the averaged time course might represent more closely the result of the convolution. However, I might be wrong, and another problem would be, that I do not know anything about the baseline and thus have to determine one based on some assumptions. Therefore I am not entirely sure whether my approach makes any sense or is just waste of time. But if I could obtain reasonable results this would just be great.
Thanks again for your suggestions, wolf Moshe Olshansky wrote: > Hi Wolf, > > Without noise you could use FFT, i.e. FFT of a convolution is the product of > the individual FFTs and so you get the FFT of your input signal and using > inverse FFT you get the signal itself. > When there is noise you must experiment. You may want to filter the response > before doing FFT. Whay do you know about the noise? > > Regards, > > Moshe. > > > --- On Mon, 25/8/08, wolf zinke <[EMAIL PROTECTED]> wrote: > > >> From: wolf zinke <[EMAIL PROTECTED]> >> Subject: [R] deconvolution: Using the output and a IRF to get the input >> To: r-help@r-project.org >> Received: Monday, 25 August, 2008, 8:22 AM >> Hi, >> >> Maybe someone could give me some pointers for my problem. >> So far I have >> not found a good solution, maybe it is just ill posed? >> >> I have a signal that is the result of an input signal >> convolved with a >> given impulse response function (IRF) plus noise. I want to >> use the this >> signal and the IRF to determine the underlying input >> signal. In my >> naivety I thought this just might be a deconvolution >> problem. But here I >> found only routines that use the input signal and the >> output signal to >> get the IRF. Is it possible to derive the input signal when >> output and >> IRF are given? If so, how could I do this with R? >> >> Thanks a lot for any hints, >> wolf >> >> ______________________________________________ >> R-help@r-project.org mailing list >> https://stat.ethz.ch/mailman/listinfo/r-help >> PLEASE do read the posting guide >> http://www.R-project.org/posting-guide.html >> and provide commented, minimal, self-contained, >> reproducible code. >> > > [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
