MathOverflow is a question and answer site for professional mathematicians. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top


It may be a stupid question, i'm trying to find a closed form for an integral similar to a Fourier transform on $S^{n}$ but i'm stuck... Let $\alpha>0$, the integral i can't solve is


where $p,q\in S^{n}$, $\left< \cdot,\cdot \right>$ is the euclidean scalar product on $\mathbb{R}^{n+1}$ and $d\mu_{S^{n}}$ is the measure induced by euclidean measure on $\mathbb{R}^{n+1}$. so the question is: is there a closed form for $I(p,\alpha)$ only in terms of $p$ and $\alpha$?

Thank you in advance!

share|cite|improve this question
up vote 3 down vote accepted

The integral obviously doesn't depend on $p$, since it must be rotationally invariant, and, using polar coordinates centered on $p$, one obtains $$ I(p,\alpha) = \mathrm{vol}(S^{n-1})\int_{-1}^1 e^{i\alpha t}(1{-}t^2)^{(n-2)/2}dt =\frac{2\ \pi^{n/2}}{\Gamma(n/2)}\int_{-1}^1 \cos(\alpha t)(1{-}t^2)^{(n-2)/2}dt. $$ This latter integral can be evaluated by standard techniques. For odd values of $n$, it is expressed in Bessel functions, while, for even values of $n$ it is expressed in terms of elementary functions. Thus, for $n=2$, one has $$ \int_{-1}^1 \cos(\alpha t)(1{-}t^2)^{(2-2)/2}dt = \frac{2\sin\alpha}{\alpha}, $$ while, for $n=3$, one has $$ \int_{-1}^1 \cos(\alpha t)(1{-}t^2)^{(3-2)/2}dt = \frac{\pi\ J_1(\alpha)}{\alpha}, $$ where $J_1$ denotes the Bessel function of the first kind, and so on.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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