I'll take this opportunity to follow up on something. Rokhlin developed a fast hankel transform using a mixture of the fft and fast multipole, see
http://www.cs.yale.edu/publications/techreports/tr1045.pdf This would be a very useful tool to have, but no one seems to have distributed code for it. I've been looking for a collaborator/student who'd be interested in developing it. It would make a nice addition to the GSL. -gideon On Jun 6, 2010, at 11:35 PM, Gideon Simpson wrote: > What you're trying to do is perform a Hankel transform of order 0 at wave > number 1. See http://en.wikipedia.org/wiki/Hankel_transform. I think you're > best bet, off the shelf, is to use the DHT, the discrete hankel transform, > which is part of the GSL library. If your function F(x) decays sufficiently > rapidly, the DHT should be satisfactory. > > -gideon > > On Jun 2, 2010, at 6:33 AM, [email protected] wrote: > >> Dear Sir, >> >> I am a beginer with gsl and I am trying to do an integration of the form: >> >> \int_0^\infty [ x J0(x) F(x) ]. >> >> J0(x) being oscillatory makes the integrtal +ve and -ve within its >> consecutive zero's. Form of F(x) is such that the overall integrand is a >> decaying function of x. >> >> How to handle this type of integration using gsl. >> >> I tried using "gsl_integration_qag", but its not giving the correct results. >> >> with best regards, >> Prithwish >> >> _______________________________________________ >> Help-gsl mailing list >> [email protected] >> http://lists.gnu.org/mailman/listinfo/help-gsl > > > _______________________________________________ > Help-gsl mailing list > [email protected] > http://lists.gnu.org/mailman/listinfo/help-gsl _______________________________________________ Help-gsl mailing list [email protected] http://lists.gnu.org/mailman/listinfo/help-gsl
