2

We wish to calculate the Fourier Transform of $\text{Sech}(x)$. I have tried to do this using the same technique as similar questions. We attempt to directly calculate the integral using integration by parts twice:

$\mathcal{F}[\text{Sech}(x)](\xi) = \int^{\infty}_{-\infty} \text{Sech}(x) e^{-2\pi i x \xi} \text{d}x $

$= \int^{\infty}_{-\infty} \text{Sech}(x) \text{Cos}(2\pi x \xi) \text{d}x - i\int^{\infty}_{-\infty} \text{Sech}(x) \text{Sin}(2\pi x \xi) \text{d}x $

We consider first the integral $ \int^{\infty}_{-\infty} \text{Sech}(x) \text{Cos}(2\pi x \xi) \text{d}x $

$ = [-\text{Sech}(x) \frac{\text{Sin}(2 \pi x \xi)}{2 \pi \xi} ]^{x=\infty}_{x=-\infty} - \int^{\infty}_{-\infty} \text{Sech}(x) \text{Tanh(x)} \frac{\text{Sin}(2 \pi x \xi)}{2 \pi \xi} \text{d}x $

The term on the left is $0$, so we are left with just $ - \frac{1}{2\pi \xi} \int^{\infty}_{-\infty} \text{Sech}(x) \text{Tanh(x)} \text{Sin}(2 \pi x \xi) \text{d}x $

$ = -\frac{1}{2 \pi \xi} ( [\text{Sech}(x) \text{Tanh(x)} \frac{\text{Cos}(2 \pi x \xi)}{2 \pi \xi}]^{x=\infty}_{x=-\infty} - \int^{\infty}_{-\infty} \{\text{Sech}^3 (x) - \text{Sech}(x)\text{Tanh}^2(x) \} \frac{\text{Cos}(2 \pi x \xi)}{2 \pi \xi} \text{d}x ) $

Again, the term on the left is $0$, so we are left with:

$ +(\frac{1}{2 \pi \xi})^2 \int^{\infty}_{-\infty} \{\text{Sech}^3 (x) - \text{Sech}(x)\text{Tanh}^2(x) \}\text{Cos}(2 \pi x \xi) \text{d}x $

$ = (\frac{1}{2 \pi \xi})^2 \int^{\infty}_{-\infty} \text{Sech}(x)\text{Cos}(2 \pi x \xi) \{\text{Sech}^2 (x) - \text{Tanh}^2(x) \} \text{d}x $

So we have performed integration by parts twice and are left with the equation:

$ \int^{\infty}_{-\infty} \text{Sech}(x) \text{Cos}(2\pi x \xi) \text{d}x = (\frac{1}{2 \pi \xi})^2 \int^{\infty}_{-\infty} \text{Sech}(x)\text{Cos}(2 \pi x \xi) \{\text{Sech}^2 (x) - \text{Tanh}^2(x) \} \text{d}x $

Unfortunately, this does not seem usable for this situation, and it does not seem like a useful result will appear with any further uses of integration by parts. I am left to assume we need to use a totally different method. If anyone has any suggestions, I would be very grateful. Thank you.

gbnhgbnhg
  • 457
  • Maybe this can help? https://math.stackexchange.com/questions/799616/fourier-transform-of-1-cosh – Alessio K Jan 26 '20 at 09:08
  • 1
    @Extended Thank you very much for sharing, but I find the answers provided in this question a bit unclear and short of explanation. My complex analysis isn't very good, so some of the steps being taken by the answerers don't make sense to me... – gbnhgbnhg Jan 26 '20 at 09:53
  • You can get displayed equations by enclosing them in double instead of single dollar signs. You can get properly sized paired delimiters (braces, brackets, parentheses) that adapt to the size of their content by preceding them with \left and \right. – joriki Jan 26 '20 at 10:11

1 Answers1

2

$$\mathcal{F}[sech(x)](\xi)=2\int_{-\infty}^{\infty}\frac{e^{2\pi ix\xi}}{e^{2x}+1}e^xdx\overbrace{=}^{e^{2x}=z}\int_{0}^{\infty}\frac{z^{\pi i\xi+\frac{1}{2}-1}}{z+1}dz$$ $$\mathcal{F}[sech(x)](\xi)=\mathfrak{B}\left(\pi i\xi+\frac{1}{2},1-\pi i\xi-\frac{1}{2}\right)=\pi csc\left(\pi\left(\pi i\xi+\frac{1}{2}\right)\right)=\pi sech(\pi^2 \xi)$$

Teruo
  • 1,860