Dear Matthew,
Am enclosing some matlab I've written for
extracting the relevant filter matrices.
It has the following steps:
1. We generate 1/f noise from a nonlinear dynamical system (just
so as to have some test data)
2. We look at the power spectral density
3. We model it with a 1/f curve
4. We inverse FFT to get the autcorrelation
5. We compute the corresponding filter matrix, K
6. As a 'sanity check' we compute the autocorrelation
directly using the matlab xcorr function
Note the disclaimer in the code.
All the best,
Will.
Matthew Brett wrote:
> Dear All,
>
> I was interested in generating fitted 1/f K matrices from real data a la
> Zarahn at al 1997; has anyone got any code or suggestions? I would be
> very grateful if so...
>
> Thanks a lot,
>
> Matthew
--
William D. Penny
Wellcome Department of Cognitive Neurology
University College London
12 Queen Square
London WC1N 3BG
Tel: 020 7833 7478
FAX: 020 7813 1420
Email: [log in to unmask]
URL: http://www.fil.ion.ucl.ac.uk/~wpenny/
|