How to integrate

$$\int\limits_0^\infty e^{-a x^2}\cos(b x) dx$$

where $a>0$

The real problem is this integral

$$\lim\limits_{\alpha\rightarrow 2}\int\limits_0^\infty e^{-a x^\alpha}\cos(b x) dx$$

I tried integration by parts and then the change of variable $z=x^2$ but it does not work.

  • 1
    $\begingroup$ The standard way of doing this is using contour integration and cauchy's integral theorem. $\endgroup$ – user159517 Feb 7 '16 at 18:07

Using Euler's identity, we get:

$$ \int\limits_0^\infty e^{-a x^2}\cos(b x) dx=Re \left( \int\limits_0^\infty e^{-a x^2} e^{ibx} dx \right) $$

$$ \int\limits_0^\infty e^{-a x^2} e^{ibx} dx = \int\limits_0^\infty e^{-a x^2+ibx} dx $$

Let's forget about imaginary unit and take $ib=\beta$ for simplicity:

$$ -ax^2+\beta x=-a (x^2-\frac{\beta}{a}x+\frac{\beta^2}{4a^2})+\frac{\beta^2}{4a}=-a(x-\frac{\beta}{2a})^2+\frac{\beta^2}{4a} $$

$$ \int\limits_0^\infty e^{-a x^2+\beta x} dx=e^{\frac{\beta^2}{4a}} \int\limits_0^\infty e^{-a(x-\frac{\beta}{2a})^2} dx $$

I believe you will not have trouble with the rest.




  • $\begingroup$ One should be careful about the change of variables, since $\beta/(2a)$ is not real. $\endgroup$ – Robert Israel Feb 7 '16 at 18:52
  • $\begingroup$ @RobertIsrael, how can I improve my answer to account for this? I know this gets the correct value of the integral anyway, but I don't want to confuse the OP with my sloppy math $\endgroup$ – Yuriy S Feb 7 '16 at 19:11
  • $\begingroup$ In the complex approach, you estimate the integral from $R$ to $R-\beta/(2a)$. $\endgroup$ – Robert Israel Feb 7 '16 at 19:21
  • $\begingroup$ @Yuriy S : you need the analytic continuation argument : that if it works for $\beta \in [a;b]$ and that the defined function is analytic in $\beta$ and always bounded (so it is an entire function), then it has to work also for any $\beta \in \mathbb{C}$ (another exactly equivalent way that @ vnd did is expanding $e^{i\beta x}= \sum_{k=0}^\infty \frac{(i\beta x)^k}{k!}$ in your integrand) $\endgroup$ – reuns Feb 7 '16 at 21:40

I will consider the integral without the limit process. $$I(a,b)=\int\limits_0^\infty e^{-a x^2}\cos(b x) dx=\sum_{n=0}^{\infty}\frac{(-1)^n.b^{2n}}{(2n)!}\int_0^{\infty}x^{2n}e^{-ax^2}dx$$ by expanding the cosine. The integrals in the above sum are the familiar Gaussian Integrals defined by $$I_m=\int_0^{\infty}x^me^{-ax^2}dx$$ for non negative integral m. One may note that $I_0=\frac{1}{2}\sqrt{\frac{\pi}{a}}$ and that $I_{2n}=(-1)^n\frac{d^n}{da^n}I_0$. It follows that $$I(a,b)=\frac{\sqrt\pi}{2}\sum_{n=0}^{\infty}\frac{b^{2n}}{(2n)!}\frac{d^n}{da^n}a^{-\frac{1}{2}}.$$ It is not difficult to see that $$\frac{d^n}{da^n}a^{-\frac{1}{2}}=\frac{-1^n}{\sqrt a}\frac{1.2...(2n-1)}{2^n}\frac{1}{a^n}.$$ Substituting the above in the expression for $I(a,b)$ one has $$I(a,b)=\sqrt{\frac{\pi}{4a}}\sum_{n=0}^{\infty}\frac{1}{n!}\left(-\frac{b^2}{4a}\right)^n$$ $$\implies I(a,b)=\sqrt{\frac{\pi}{4a}} e^{-\frac{b^2}{4a}} .$$


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.