How can I remove white noise from my signals?

Dear Colleagues,

I have a problem. I have velocity signals x(t) recorded with an acoustic
Doppler velocimeter. These signals are contaminated with a white noise
n(t), so x(t) = d(t) + n(t), where d(t) is a true velocity signal.

I can know the noise variance, noise power spectrum Gnn(f) and the true
power spectrum as 
Gdd(f) = Gxx(f) � Gnn(f).

I am looking for a digital filter which removes the white noise from my
signals in the time domain, Is there a filter for that? I wish to obtain
d(t).

I am using MatLab.

Thanks for your attention

Sincerely 

Mart�n.

Mat=EDnArg wrote:
> Dear Colleagues,
>
> I have a problem. I have velocity signals x(t) recorded with an acoustic
> Doppler velocimeter. These signals are contaminated with a white noise
> n(t), so x(t) =3D d(t) + n(t), where d(t) is a true velocity signal.
>
> I can know the noise variance, noise power spectrum Gnn(f) and the true
> power spectrum as
> Gdd(f) =3D Gxx(f) - Gnn(f).
>
> I am looking for a digital filter which removes the white noise from my
> signals in the time domain, Is there a filter for that? I wish to obtain
> d(t).
>
> I am using MatLab.
>
> Thanks for your attention
>
> Sincerely
>
> Mart=EDn.

clues only for homework problems...

what is the bandwidth of your desired signal?

what is the bandwidth of the undesired white noise signal?

Mark

"Mat&#4294967295;nArg" <mromagnoli@ina.gov.ar> wrote in message 
news:__ydnfOux9Jg6_XZnZ2dnUVZ_tidnZ2d@giganews.com...
> Dear Colleagues,
>
> I have a problem. I have velocity signals x(t) recorded with an acoustic
> Doppler velocimeter. These signals are contaminated with a white noise
> n(t), so x(t) = d(t) + n(t), where d(t) is a true velocity signal.
>
> I can know the noise variance, noise power spectrum Gnn(f) and the true
> power spectrum as
> Gdd(f) = Gxx(f) - Gnn(f).
>
> I am looking for a digital filter which removes the white noise from my
> signals in the time domain, Is there a filter for that? I wish to obtain
> d(t).
>
> I am using MatLab.
>
> Thanks for your attention
>
> Sincerely
>
> Mart&#4294967295;n.
>

Hello Martin,

Look into Kalman filtering. It works quite well for this kind of thing.

Clay

Those algorithms for Speech Enhancement may be able to fix your problem.
And if the noise is stationary, the simplest spectral substraction with a
simple VAD will be OK.

>Dear Colleagues,
>
>I have a problem. I have velocity signals x(t) recorded with an acoustic
>Doppler velocimeter. These signals are contaminated with a white noise
>n(t), so x(t) = d(t) + n(t), where d(t) is a true velocity signal.
>
>I can know the noise variance, noise power spectrum Gnn(f) and the true
>power spectrum as 
>Gdd(f) = Gxx(f) � Gnn(f).
>
>I am looking for a digital filter which removes the white noise from my
>signals in the time domain, Is there a filter for that? I wish to obtain
>d(t).
>
>I am using MatLab.
>
>Thanks for your attention
>
>Sincerely 
>
>Mart�n.
>
>
>
>
>