Contour Integration

Calculus Level 3

log ( 1 + x 2 ) 1 + x 2 d x = A π log A \large \int_{-\infty}^{\infty} \frac{\log(1+x^{2})}{1+x^{2}} \, dx=A\pi\log{A} If the equation above is true for some positive integer A A , find A A .


The answer is 2.

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

2 solutions

Put x = tan y x=\tan y , the integral turns , I = π 2 π 2 ln ( sec 2 y ) d y = 4 0 π 2 ln ( cos y ) d y \large \displaystyle I = \int_{-\frac{\pi}{2}}^{\frac{\pi}{2}} \ln (\sec^2 y) dy = -4 \int_{0}^{\frac{\pi}{2}} \ln (\cos y) dy

Easy result , 0 π 2 ln ( cos x ) d x = π 2 ln 2 \displaystyle \int_{0}^{\frac{\pi}{2}} \ln (\cos x) dx = -{\frac{\pi}{2}}\ln 2

So , I = 2 ln 2 \displaystyle I = \color{#3D99F6}{2}\ln \color{#3D99F6}{2} .

@Aditya Sharma , I think it should be 4 0 π 2 ln ( cos y ) d y -4 \int_{0}^{\frac{\pi}{2}} \ln (\color{#D61F06}{\cos y}) dy .

Chew-Seong Cheong - 4 years, 11 months ago

Log in to reply

Oh Thanks, Though both are same.

Aditya Narayan Sharma - 4 years, 11 months ago

Log in to reply

You are right. I didn't see that.

Chew-Seong Cheong - 4 years, 11 months ago
James Yang
Sep 22, 2016

\newcommand{\LIM}[2]{\lim\limits_{#1\rightarrow #2}} \newcommand{\PLOG}[1]{\text{log}\PAREN{#1}} \newcommand{\diff}{\mathop{}\!\mathrm{d}} \newcommand{\PAREN}[1]{\left(#1\right)} \text{This is for anyone who's interested in the contour integration method.}\\ \\ \text{The branch cut is \$\forall z = ib, -ib, b>1\$.}\\ \text{Since there are no singularities in each closed contour, and the integrands are analytic,}\\ \oint_{C_1} \frac{\PLOG{1-iz}}{1+z^2}\diff z + \oint_{C_2} \frac{\PLOG{1+iz}}{1+z^2}\diff z = 0 \\ \text{For the first integral, by changing the direction of integration in the positive direction,}\\ \oint_{C_1} \frac{\PLOG{1-iz}}{1+z^2}\diff z = \int_{-R}^R + \int_{|z| = R_+} + \int_{|z| = R_-} - \int_{|z|=h}\\ \text{where \$R\_+\$ and \$R\_-\$ represent the semicircle contour with radius \$R\$ on the positive real side and negative real side, }\\ \text{respectively, and \$h\$ is any small arbitrary real number.}\\ \text{Notice that the vertical contours converge to the same contour but in opposite directions,}\\ \text{so the sum of integral on both contours converge to zero.}\\ \text{As \$|z| = R\to \infty\$, the integrand approaches 0 since the order of \$\PLOG{z}\$ is less than \$z\$ and \$z/\PAREN{1+z^2}\$ approaches 0. Hence,}\\ \LIM{R}{\infty} \PAREN{\int_{-R}^R + \int_{|z| = R_+} + \int_{|z| = R_-} - \int_{|z|=h}} = \LIM{R}{\infty} \PAREN{\int_{-R}^R} - \int_{|z|=h}\\ \text{Let \$\epsilon\$ be the distance between the two vertical contours}\\ \text{As \$\epsilon \to 0\$, the contour where \$|z|=h\$ converges to a complete circle. Since the integrand is singular at the center \$i\$, the integral}\\ \oint_{|z|=h}\frac{\PLOG{1-iz}}{1+z^2}\diff z \to 2\pi i Res\PAREN{\frac{\PLOG{1-iz}}{1+z^2}, i} = 2\pi i \cdot \frac{\log(2)}{2i} = \pi \log(2)\\ \text{Therefore,}\\ \oint_{C_1} \frac{\PLOG{1-iz}}{1+z^2}\diff z = \LIM{R}{\infty} \PAREN{\int_{-R}^R \frac{\PLOG{1-ix}}{1+x^2}\diff x} - \pi \log(2)\\ \text{Similarly,}\\ \oint_{C_2} \frac{\PLOG{1+iz}}{1+z^2}\diff z = \LIM{R}{\infty} \PAREN{\int_{-R}^R \frac{\PLOG{1+ix}}{1+x^2}\diff x} - \pi \log(2)\\ \text{Substituting to the first equation,}\\ \LIM{R}{\infty} \PAREN{\int_{-R}^R \frac{\PLOG{1-ix}}{1+x^2}\diff x} - \pi \log(2) + \LIM{R}{\infty} \PAREN{\int_{-R}^R \frac{\PLOG{1+ix}}{1+x^2}\diff x} - \pi \log(2) = 0\\ \LIM{R}{\infty} \PAREN{\int_{-R}^R \frac{\PLOG{1-ix} +\PLOG{1+ix}}{1+x^2}\diff x} = 2\pi \log(2)\\ \int_{-\infty}^\infty \frac{\log(1+x^2)}{1+x^2}\diff x = 2\pi \log(2)\\

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...