How can I Numerically integrate a function multiplied with a dirac delta function?
6 views (last 30 days)
Show older comments
for instance, integrate f=x.^2.*dirac(x), type function from -1 to 1.
Thanks
0 Comments
Answers (1)
Roger Stafford
on 18 Mar 2017
Edited: Roger Stafford
on 18 Mar 2017
The integral of f(x) = x.^2.*dirac(x) w.r. x from -1 to +1 is zero. That is because the dirac(x) function is considered 0 at all points x except at x = 0, and at x = 0 its value is so large that the integral across the single point x = 0 is just 1. However in the case of your f(x) that one is multiplied by x^2 so that makes the integral zero. If you had a function g(x) = cos(x)*dirac(x), its integral from -1 to +1 would be just cos(0)*1 = 1.
Obviously this cannot be justified in the ordinary numerical sense with finite numbers and only has its basis in what mathematicians call “measure theory”. In particular, you cannot obtain the correct answer using numerical integration.
2 Comments
Walter Roberson
on 18 Mar 2017
And if you must integrate it numerically then you are going to have to use one of the approximations to the dirac delta; https://en.wikipedia.org/wiki/Dirac_delta_function#Approximations_to_the_identity
Roger Stafford
on 3 Jan 2018
Edited: Walter Roberson
on 3 Jan 2018
@Mostafa A. Jouybari: In answer to your question concerning this "Answers" entry, you can find the following sentence, "A rigorous approach to regarding the Dirac delta function as a mathematical object in its own right requires measure theory or the theory of distributions" at the Wikipedia website
See Also
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!