So I have a question about power spectral densities and correlations of noise. I know that there is the Wiener–Khinchin theorem that can connect the two with a Fourier transform, and that a white noise process (flat power spectral density) has no correlations between the points; the correlation function is a delta function. But what if we now take a white noise process with a cutoff? Say it is white from 0 to 200 MHz, with some amplitude A, but after that it is 0

You can think of the cutoff as a fermi dirac type of cutoff if you will; it just goes from A to 0 very rapidly after 200 MHz. What kind of correlations does this noise have? It is no longer a delta function, but what is it