Solving Gaussian Integral with Complex offset

Fingolfin_Noldo
Messages
2
Reaction score
0

Homework Statement


Given f(x) = e^{-ax^2/2} with a > 0 then show that \^{f} = \int_{-\infty}^{\infty} e^{-i \xi x - ax^2/2} \, \mathrm{d}x = \surd\frac{2}{a} = e^{-\xi^2/2a} by completing the square in the exponent, using Cauchy's theorem to shift the path of integration from the real axis (Im x = 0) to the horizontal line I am x = -\xi/a and finally use the relation \int_{-\infty}^{\infty} e^{- ax^2/2} \, \mathrm{d}x = \surd\pi


Homework Equations



Show \^{f} = \int_{-\infty}^{\infty} e^{-i \xi x - ax^2/2} \, \mathrm{d}x = \surd\frac{2}{a} e^{-\xi^2/2a}

The Attempt at a Solution


I completed the squares and set x = x' + i \xi/a. But now I have an integral from -\infty + i\xi/a to \infty + i\xi/a. Is there something I am missing here? Should I construct some contour? Constructing a semi-circle won't work since the arc does not converge. Thanks
 
Physics news on Phys.org
If you want an explicit contour your could draw a rectangle connecting the line with imaginary part i\xi/a to the real axis. As the ends of the rectangle go to infinity their contribution vanishes. And there are no poles enclosed. So the integral over your line is the same as the integral over the real axis.
 
There are two things I don't understand about this problem. First, when finding the nth root of a number, there should in theory be n solutions. However, the formula produces n+1 roots. Here is how. The first root is simply ##\left(r\right)^{\left(\frac{1}{n}\right)}##. Then you multiply this first root by n additional expressions given by the formula, as you go through k=0,1,...n-1. So you end up with n+1 roots, which cannot be correct. Let me illustrate what I mean. For this...
Back
Top