I want to get the autocorrelation function of the power spectral density of the wind.
This function is defined by: $$s(\omega)=\frac{c_1}{(1+1.5 \times c_2\omega)^{5/3}}$$ $c_1$ and $c_2$ are constants. Any help please.
I specify that the autocorrelation function is done by the integral of : $$\int_{-\infty}^{+\infty} \! s(\omega) \times \exp(I \times \tau \times \omega) \, \mathrm{d}\omega$$
with $-\infty < \tau <+\infty $ and $I^2=−1$
Assumig that $*$ means usual multiplication and $c_2>0$, the following plan should work:
1.) Rescale $c_2\omega =\tilde\omega$
2.) Substitute $\tilde\omega = q-1$
3.) Integrate by parts, choosing $f'(q)=q^{-5/3}, g(q)=e^{i q \tilde{\tau} }$ with $\tilde{\tau}=\tau/c_2$ to lower the degree of the singulartiy at $0$.
4.) Use formula 318 from this link . It can be proven by contour integration methods, using a half circle contour in the (upper/lower) halfplane excluding the branch cut on the positive/negative imaginary axis
Can you get it from here?