Show that \[\int_0^\infty \cos(x^2)\,dx=\int_0^\infty \sin(x^2)\,dx=\sqrt{\frac{\pi}{8}}\]
[ A relatively simple problem compared to the recent ones :) ]
I can write the result as below: \[\Im \left( \int\limits_{0}^{+\infty} e ^{ix ^{2}} dx\right)\] then I use this substitution: \[ix ^{2}=y\] and observing that \[\sqrt{i}=\frac{ 1 }{ \sqrt{2} }(1\pm i)\] I got this:\[-\frac{ 1 }{ 2 \sqrt{i} }\Gamma \left( \frac{ 1 }{ 2 } \right)\] so the final result is: \[\pm \frac{ \sqrt{\pi} }{ 2\sqrt{2} }\]
where:\[\Im \] stands for imaginary part of a complex number
oops...I used: \[ix ^{2}=-y\]
Well done, I had a similar method, but I keep tripping up with the complex computation: \[\begin{align*} \int_{-\infty}^\infty e^{-x^2}\,dx&=\int_{-\infty}^\infty e^{(ix)^2}\,dx\\\\ \sqrt\pi&=\int_{-\infty}^\infty\bigg(\cos (ix)^2+i\sin(ix)^2\bigg)\,dx\\\\ &=\int_{-\infty}^\infty\cos(x^2)\,dx-\underbrace{\mathrm{Im}\left[\int_{-\infty}^\infty e^{-ix^2}\,dx\right]}\end{align*}\] The bracketed part is the part that's particularly bothersome. At any rate, once I figure this out I'd divide by 2 and the result follows.
I get that the bracketed part amounts to \(-\dfrac{\sqrt{\pi}}{\sqrt2}\), so I suppose there's a mistake in my reasoning somewhere along the line...
It's probably due to \[\large\int_\mathbb{R}e^{-ax^2}\,dx=\sqrt{\frac{\pi}{a}}\quad\text{for Re}(a)>0\] and not for complex \(a\).
you are right! @SithsAndGiggles
*not for \(\text{Re}(a)\le0\), I mean
I think so!
very nice question :) @SithsAndGiggles
Thanks! I originally considered approaching the integral via differentiation under the integral sign but I couldn't set up the proper parameterization.
Here's an interesting approach: http://math.stackexchange.com/questions/187729/evaluating-int-0-infty-sin-x2-dx-with-real-methods/190293#190293
I understand
I see
thanks! for your website @SithsAndGiggles
Join our real-time social learning platform and learn together with your friends!