What is the kernel $K(t)$ of the following Laplace transform equation: $$\int_{0}^{+\infty}e^{-(x+y)t} K(t) dt= \sum_{n=0}^{\infty}\varphi_{n}^{\alpha}(x)\varphi_{n}^{\alpha}(y),$$ where $\varphi_{n}^{\alpha}(x)=(\frac{n!}{\Gamma(\alpha+n+1)})^{1/2} \frac{e^{x}}{n!} \frac{d^{n}}{x^{n}}(e^{-x}x^{n})$ is the normalized Laguerre polynomial. If the result is not known, where can I find this type of equation?

  • $\begingroup$ I presume your expression for $\varphi^\alpha_n(x)$ is missing a factor $x^{-\alpha}$ before the derivative, and a factor $x^\alpha$ after the derivative, and that the denominator $x^n$ should be $dx^n$. $\endgroup$ Apr 30, 2021 at 13:03
  • $\begingroup$ Yes you are right sorry $\endgroup$ Apr 30, 2021 at 15:00

1 Answer 1


Let me check this for $\alpha=0$, when $\varphi_n^\alpha(x)=L_n(x)$ is the Laguerre polynomial. The completeness relation is (see eq. 1.10 of Li and Wong - Integral and series representations of the Dirac delta function) $$\sum_{n=0}^\infty L_n(x)L_n(y)=e^{(x+y)/2}\delta(x-y).$$ This depends on both the sum and difference of $x$ and $y$, so the representation given in the OP, which depends only on $x+y$, does not exist.

  • $\begingroup$ Thank you very much $\endgroup$ Apr 30, 2021 at 15:07
  • $\begingroup$ Is it possible to put $K(t)=\delta(x-y) e^{(x+y)/2}e^{(x+y)t}e^{-t}$? $\endgroup$ Apr 30, 2021 at 15:29
  • 2
    $\begingroup$ sure, but it's kind of pointless to allow $K(t)$ to depend on $x$ and $y$ as well, then why introduce this integral? $\endgroup$ Apr 30, 2021 at 15:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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