How to generate a Gaussian time correlated process?
in [DSP]
|
Prev: radiographic images
Next: Whitened Matched Filter
From: Andor on 9 Mar 2010 18:12 On 9 Mrz., 15:22, "msarovar" <mo...(a)grommit.com> wrote: > Hi, > > What is the most efficient way to generate a noise process that is Gaussian > correlated in time from a white noise process? > > I'm having trouble coming up with a simple FIR filter that will do this. > > Any suggestions? > > Thanks, > ~mohan This link is three years old but still works: http://math.mit.edu/~stevenj/preprints/iir-colored-noise-2007-03-13.pdf It describes the use of IIR filters to efficiently (wrt to computer time and memory) generate correlated sequences from white sequences. Regards, Andor
From: msarovar on 9 Mar 2010 21:30 >On 9 Mrz., 15:22, "msarovar" <mo...(a)grommit.com> wrote: >> Hi, >> >> What is the most efficient way to generate a noise process that is Gaussian >> correlated in time from a white noise process? >> >> I'm having trouble coming up with a simple FIR filter that will do this. >> >> Any suggestions? >> >> Thanks, >> ~mohan > >This link is three years old but still works: > >http://math.mit.edu/~stevenj/preprints/iir-colored-noise-2007-03-13.pdf > >It describes the use of IIR filters to efficiently (wrt to computer >time and memory) generate correlated sequences from white sequences. > >Regards, >Andor > Great, thanks guys. I will read more about Chebyshev filters. They sound like my solution. Cheers, ~mohan
From: Andor on 10 Mar 2010 08:44 On 10 Mrz., 05:35, Jerry Avins <j...(a)ieee.org> wrote: > Andor wrote: > > On 9 Mrz., 15:22, "msarovar" <mo...(a)grommit.com> wrote: > >> Hi, > > >> What is the most efficient way to generate a noise process that is Gaussian > >> correlated in time from a white noise process? > > >> I'm having trouble coming up with a simple FIR filter that will do this. > > >> Any suggestions? > > >> Thanks, > >> ~mohan > > > This link is three years old but still works: > > >http://math.mit.edu/~stevenj/preprints/iir-colored-noise-2007-03-13.pdf > > > It describes the use of IIR filters to efficiently (wrt to computer > > time and memory) generate correlated sequences from white sequences. > > Any low-pass filter's output is correlated at high-enough frequencies. Of course. The trick is to get the correlation you want as cheap as possible.
From: HardySpicer on 11 Mar 2010 04:06 On Mar 10, 3:22 am, "msarovar" <mo...(a)grommit.com> wrote: > Hi, > > What is the most efficient way to generate a noise process that is Gaussian > correlated in time from a white noise process? > > I'm having trouble coming up with a simple FIR filter that will do this. > > Any suggestions? > > Thanks, > ~mohan Bit confused here. If you pass Guassian white noise through a LTI filter the output is still Guassian but coloured. Hardy
From: Tim Wescott on 11 Mar 2010 11:16
HardySpicer wrote: > On Mar 10, 3:22 am, "msarovar" <mo...(a)grommit.com> wrote: >> Hi, >> >> What is the most efficient way to generate a noise process that is Gaussian >> correlated in time from a white noise process? >> >> I'm having trouble coming up with a simple FIR filter that will do this. >> >> Any suggestions? >> >> Thanks, >> ~mohan > > Bit confused here. If you pass Guassian white noise through a LTI > filter the output is still Guassian but coloured. He wants noise that is colored in such a way that the autocorrelation is Gaussian. -- Tim Wescott Control system and signal processing consulting www.wescottdesign.com |