cpsd

Hi folks,

Can anyone help me explain why I get negative cross power spectral density
(cpsd) values? I do my calculations in matlab (cpsd) and I always get positive
values until now.

Knowing the usefulness of it would greatly help me in my analysis.

Regards,

Ibra Start a New Thread

Hi guys!

Thanks for the help and suggestions!

Yes, I used a different data as Jeff said, but the same nfft points.

I found the answer as 'strongly negative cpsd' (which I calculated together with
coherence) means 'high negative coherence'.

Cheers,

Ibrahim

________________________________
From: Syed Meeran
To: Ibrahim muhammed
Cc: m...
Sent: Thu, August 19, 2010 5:54:37 AM
Subject: Re: [matlab] cpsd

Hi,

Can you elaborate more on cpsd function which one do you use.

Do you use the function along with window and nfft points .

Thanks

Syed
On Tue, Aug 17, 2010 at 8:40 PM, Ibrahim muhammed wrote:

>Hi folks,
>
>Can anyone help me explain why I get negative cross power spectral density
>(cpsd) values? I do my calculations in matlab (cpsd) and I always get positive
>values until now.
>
>Knowing the usefulness of it would greatly help me in my analysis.
>
>Regards,
>
>Ibra
> Start a New Thread

Hi,

Can you elaborate more on cpsd function which one do you use.

Do you use the function along with window and nfft points .

Thanks

Syed

On Tue, Aug 17, 2010 at 8:40 PM, Ibrahim muhammed wrote:

> Hi folks,
>
> Can anyone help me explain why I get negative cross power spectral density
> (cpsd) values? I do my calculations in matlab (cpsd) and I always get
> positive values until now.
>
> Knowing the usefulness of it would greatly help me in my analysis.
>
> Regards,
>
> Ibra
>
>
> Start a New Thread

Ibrahim-

> Can anyone help me explain why I get negative cross power spectral density
> (cpsd) values? I do my calculations in matlab (cpsd) and I always get positive
> values until now.

Well if you got positive values before and now you get negative values, then what did you change? Different input
data? New version of MATLAB? A change in your calculations?

Something must be different.

-Jeff Start a New Thread