4
$\begingroup$

Is there a way to prove this relation?:

$$\sum_{\text{n}=0}^\infty\frac{1}{\text{s}^2+\left(1+2\text{n}\right)^2\omega^2}=\frac{\pi\tanh\left(\frac{\pi\text{s}}{2\omega}\right)}{4\text{s}\omega}$$

And find the conditions for which this equality hold?

$\endgroup$
  • 2
    $\begingroup$ hint1: mittag leffler theorem. hint2: product representation of $\sin$. hint3: contour integration $\endgroup$ – tired Nov 20 '16 at 16:30
  • 1
    $\begingroup$ then consult the internet $\endgroup$ – tired Nov 20 '16 at 16:36
  • 1
    $\begingroup$ i give you a starting point: math.stackexchange.com/questions/141470/… $\endgroup$ – tired Nov 20 '16 at 16:39
  • 1
    $\begingroup$ Besides the very great hint given by @tired I would suggest you to read and study: " Complex Variables, theory and applications " by Murray S. Spiegel, in which you will learn about infinite series, residues, complex variables, contour integration et cetera. After that, I bet you'll be able to solve that. $\endgroup$ – Von Neumann Nov 20 '16 at 16:41
  • 1
    $\begingroup$ @SimpleArt The problem is that it sound strange to me that he has to evaluate that with Real Analysis. Sure it has to be possible, but.. quite sadistic, isn't it? :D $\endgroup$ – Von Neumann Nov 20 '16 at 17:06
3
$\begingroup$

A real-analytic derivation through Fourier series.


You may remove a useless parameter by setting $s=\omega t$ then look for a closed form for $$ S(t)=\sum_{n\geq 0}\frac{1}{t^2+(2n+1)^2}\tag{1} $$ Integration by parts gives a useful lemma: $$ \forall a,b>0,\qquad\int_{0}^{+\infty}\frac{\sin(a x)}{a} e^{-bx}\,dx = \frac{1}{a^2+b^2}\tag{2} $$ and $$ \sum_{n\geq 0}\frac{\sin((2n+1)x)}{2n+1}=W(x)\tag{3} $$ is a $2\pi$-periodic rectangle wave that equals $\frac{\pi}{4}$ over $(0,\pi)$ and $-\frac{\pi}{4}$ over $(\pi,2\pi)$.
By exploiting $(2)$ and $(3)$, $$ S(t) = \int_{0}^{+\infty}W(x) e^{-tx}\,dx =\frac{\pi}{4}\sum_{n\geq 0}(-1)^n\int_{n\pi}^{(n+1)\pi}e^{-tx}\,dx\tag{4}$$ and by computing the last integrals and the resulting geometric series, $$ \sum_{n\geq 0}\frac{1}{t^2+(2n+1)^2}=\color{red}{\frac{\pi}{4t}\,\tanh\left(\frac{\pi t}{2}\right)}\tag{5}$$ follows. The same can be achieved through the Poisson summation formula.

$\endgroup$
1
$\begingroup$

Using the identity $$\sum_{n\in\mathbb{Z}}f\left(n\right)=-\sum\left\{ \textrm{Residues of }\pi\cot\left(\pi z\right)f\left(z\right)\textrm{ at }f'\textrm{s poles}\right\} $$ which follows from the residue theorem we have $$\sum_{n\geq0}\frac{1}{s^{2}+\left(2n+1\right)^{2}\omega^{2}}=\frac{1}{2}\sum_{n\in\mathbb{Z}}\frac{1}{s^{2}+\left(2n+1\right)^{2}\omega^{2}}$$ and since we have poles at $z=-\frac{1}{2}\pm\frac{i\sqrt{s^{2}}}{2\sqrt{\omega^{2}}}$ we can conclude that $$\sum_{n\geq0}\frac{1}{s^{2}+\left(2n+1\right)^{2}\omega^{2}}=\color{red}{\frac{\pi\tanh\left(\frac{\pi s}{2\omega}\right)}{4s\omega}}$$ as wanted.

$\endgroup$
0
$\begingroup$

$\newcommand{\bbx}[1]{\,\bbox[8px,border:1px groove navy]{\displaystyle{#1}}\,} \newcommand{\braces}[1]{\left\lbrace\,{#1}\,\right\rbrace} \newcommand{\bracks}[1]{\left\lbrack\,{#1}\,\right\rbrack} \newcommand{\dd}{\mathrm{d}} \newcommand{\ds}[1]{\displaystyle{#1}} \newcommand{\expo}[1]{\,\mathrm{e}^{#1}\,} \newcommand{\ic}{\mathrm{i}} \newcommand{\mc}[1]{\mathcal{#1}} \newcommand{\mrm}[1]{\mathrm{#1}} \newcommand{\pars}[1]{\left(\,{#1}\,\right)} \newcommand{\partiald}[3][]{\frac{\partial^{#1} #2}{\partial #3^{#1}}} \newcommand{\root}[2][]{\,\sqrt[#1]{\,{#2}\,}\,} \newcommand{\totald}[3][]{\frac{\mathrm{d}^{#1} #2}{\mathrm{d} #3^{#1}}} \newcommand{\verts}[1]{\left\vert\,{#1}\,\right\vert}$ \begin{align} &\sum_{\mrm{n} = 0}^{\infty} {1 \over \mrm{s}^{2} + \pars{1 + 2\mrm{n}}^{2}\omega^{2}} = {1 \over 4\omega^{2}}\sum_{\mrm{n} = 0}^{\infty} {1 \over \pars{\mrm{n} + 1/2}^{2} + \mrm{s}^{2}/\pars{4\omega^{2}}} \\[5mm] = &\ {1 \over 4\omega^{2}}\sum_{\mrm{n} = 0}^{\infty}{1 \over \bracks{\mrm{n} + 1/2 + \mrm{s}\,\ic\,/\pars{2\omega}} \bracks{\mrm{n} + 1/2 - \mrm{s}\,\ic\,/\pars{2\omega}}} \\[5mm] = &\ {1 \over 4\omega^{2}}\, {\Psi\pars{1/2 + \mrm{s}\,\ic\,/\bracks{2\omega}} - \Psi\pars{1/2 - \mrm{s}\,\ic\,/\bracks{2\omega}}\over \mrm{s}\,\ic\,/\omega} \\[5mm] = &\ -\,{\ic \over 4\,\mrm{s}\omega}\, \bracks{\pi\cot\pars{\pi\bracks{{1 \over 2} - {\mrm{s} \over 2\omega}\,\ic}}} = -\,{\pi\ic \over 4\,\mrm{s}\omega}\, \tan\pars{{\pi\,\mrm{s} \over 2\omega}\,\ic} = -\,{\pi\ic \over 4\,\mrm{s}\omega}\, \bracks{\ic\tanh\pars{\pi\,\mrm{s} \over 2\omega}} \\[5mm] = &\ \bbox[15px,#ffe,border:1px dotted navy]{\ds{{\pi \over 4\,\mrm{s}\omega}\, \tanh\pars{\pi\,\mrm{s} \over 2\omega}}} \end{align}

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.