I've spent a lot of times trying to show that $$ \mathcal{F}[e^{-x^2/2} G(x,t)] = e^{-k^2/2} G(k, -it) $$ with $G(x,t)$ being the generating function of Hermite polynomial, $$ G(x,t) = e^{2tx - t^2} $$
My attempt
We want to show that \begin{align*} \mathcal{F}[e^{-x^2/2} G(x,t)] &= e^{-k^2/2} G(k, -it) \\ &= e^{-k^2/2} e^{-2itk+t^2} \\ &= e^{-\frac{k^2}{2} - 2itk + t^2} \end{align*}
Computing the Fourier transform: \begin{align*} \mathcal{F}[e^{-x^2/2} G(x,t)] &= \mathcal{F}[e^{-x^2/2} e^{2tx - t^2}] \\ &= \frac{1}{\sqrt{2\pi}} \int_{-\infty}^{+\infty} e^{-x^2/2} e^{2tx - t^2} e^{-ikx} dx \end{align*}
We know that \begin{align*} \sqrt{2\pi} &= \sqrt{\frac{\pi}{1/2}} \\ &= \int_{-\infty}^{+\infty} e^{-\frac{1}{2}x^2} dx \end{align*}
I've been trying to factor the exponential isolating something like $-\frac{1}{2}x^2$. In fact I should get something like $e^{-\frac{1}{2}x^2} e^{-\frac{k^2}{2} + 2itk + t^2}$ in the integrant so I can get rid of the $\frac{1}{\sqrt{2\pi}}$ and get the desired result but I can't do it.
see related computations https://math.stackexchange.com/questions/2114701/computing-the-fourier-transform-of-h-kxe-x2-2-where-h-k-is-the-hermit
– prolea Aug 27 '20 at 10:36