Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. 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

I am trying to solve a problem part of which includes the following integral ($j=\sqrt{-1}$):

$$\int_{k_1}^{k_2} k e^{-jk\sigma} J_n(\rho k) \, \mathrm{d}k$$

The $e^{-jk\sigma}$ term is making my usual approach to the problem ineffective. However, unless I am mistaken, this same term makes the above a forward Fourier transform where $\sigma$ would be the frequency term and $k$ the time/space term:

$$\rho^{-1} \int_{k_1}^{k_2} \rho k e^{-jk\sigma} J_n(\rho k) \, \mathrm{d}k = \rho^{-1} \mathcal{F}_\sigma \left( \rho k J_n(\rho k) \right) $$

I've been looking for this transform pair (i.e., $\mathcal{F}_x(xJ_n(x))$ but have not been able to find anything. Is there a analytic solution to this?

share|cite|improve this question

I don't know if this helps but first try solving

$I(\sigma):=\int e^{-jk\sigma} J_n(\rho k)dk$

it is probably easier. Then use the fact that your integral above is simply $j\frac{dI}{d\sigma}$.

share|cite|improve this answer
It does and it doesn't in that I can get a solution for an integral with infinite limits but not for one with finite limits, unless I include convolution with a sinc to account for it. – AnonSubmitter85 Nov 13 '12 at 4:05
you can extend the finite limits to infinite limits if you multiply the integrand by two opposite heaviside functions (the pulse function) – Dar Far Nov 13 '12 at 12:04
Could you please rephrase this? I don't see how multiplying by the pulse function accomplishes this. Would the sifting property not be the result of multiplying the integrand by a pulse function? – AnonSubmitter85 Nov 13 '12 at 22:36
well, the pulse function ${\bf 1}_{[k_1,k_2]}$ is defined over all $\mathbb{R}$ and has compact support in the interval you are interested in $[k_1,k_2]$. So, you can use this $\int_{-\infty}^{\infty}{\bf 1}_{[a,b]}f(x)dx=\int_{a}^{b}f(x)dx$ to extend your integral to the whole $\mathbb{R}$ and then use fourier transforms. Of course, when you take the derivative of the pulse function, it must be in a distributional sense (some dirac deltas will appear). – Dar Far Nov 16 '12 at 12:18

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.