Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: random time-series displacement data from PSD and back again
Replies: 5   Last Post: Jun 24, 2013 9:01 PM

 Messages: [ Previous | Next ]
 Derek Goring Posts: 3,922 Registered: 12/7/04
Re: random time-series displacement data from PSD and back again
Posted: Jun 24, 2013 9:01 PM

On Tuesday, June 25, 2013 10:31:16 AM UTC+12, Jim Schwendeman wrote:
> TideMan,
>
>
>
> Maybe the better question is this:
>
>
>
> When I create my time-series data, how do I know that it's scaled correctly? When I replicated the complex conjugate prior to the ifft, I doubled the number of points, so I guess I'm not entirely sure.
>
>
>
> I believe I am correct, because if I do the sum(abs(y).^2), that is equal to the mean g^2/Hz from the original PSD, which leads me to believe I've satisfied Parseval's theorem.
>
>
>
> Thoughts?

I can't follow your code. You seem to define PSD as ones, then use it as if it had data. I gave up.............

This is Parseval's Law:
sum(PSD)*df must equal var(y)
The area under the spectrum must be the variance of the time signal.
In your case, I think df is 1.

Date Subject Author
6/21/13 Jim Schwendeman
6/21/13 Derek Goring
6/24/13 Jim Schwendeman
6/24/13 Jim Schwendeman
6/24/13 Jim Schwendeman
6/24/13 Derek Goring