Conceptual Problem with Convolution Theorem

  • #1
Hi - I'm trying to work out the following convolution problem:

I have the following integral:
[tex]
\int^{\infty}_{-\infty}p(x)U(x)e^{-i \omega x}dx
[/tex]

Where p(x) is any real function which is always positive and U(x) is the step function

Obviously this can easily be solved using the convolution theorem because I have
[tex]
\mathcal{F}[p(x)U(x)] = P(\omega)*U(\omega)
[/tex]

The problem I having is with the very similar integral but the exponential is now positive:
[tex]
\int^{\infty}_{-infty}p(x)U(x)e^{+i \omega x}dx
[/tex]

I don't know how to deal with this integral - even though I suspect I can use the convolution theorem on it.

I've tried to derive the convolution theorem for both exponentials but I get stuck at the stage:

[tex]
\int^{\infty}_{-\infty} \int^{\infty}_{-\infty}P(\omega')U(\omega-\omega')d\omega'e^{-i\omega x}d\omega = p(x)u(x)
[/tex]

And:

[tex]
\int^{\infty}_{-\infty} \int^{\infty}_{-\infty}P(\omega')U(\omega-\omega')d\omega'e^{+i\omega x}d\omega = p(x)u(x)
[/tex]

My problem is this:
If I define the [tex] \int^{\infty}_{-\infty} f(x) e^{-i\omega x}[/tex] integral as the Fourier Transform - then I can write the second equation as:
[tex]
\mathcal{F}^{-1}[P(\omega)*U(\omega)] =p(x)u(x)
[/tex]
And thus applying the inverse Fourier operator to both sides I get:
[tex]
[P(\omega)*U(\omega)] =\mathcal{F}[p(x)u(x)]
[/tex]

But If I set up this convention for my Fourier Transform how do I deal with the first equation:
[tex]
\int^{\infty}_{-\infty} \int^{\infty}_{-\infty}P(\omega')U(\omega-\omega')d\omega'e^{-i\omega x}d\omega = p(x)u(x)
[/tex]

This isn't a Fourier Transform operation anymore - its slightly different. Is there anything I can do from here to show what:
[tex]
\int^{\infty}_{-\infty} p(x)u(x)e^{i\omega x} dx
[/tex]

is in terms of convolutions ?
 

Answers and Replies

  • #2
AlephZero
Science Advisor
Homework Helper
7,002
294
Think about real and imaginary parts of the integral separately.

p(x) and U(x) are real functions.

[itex]e^{i\omega x} = \cos \omega x + i \sin \omega x[/itex]

[itex]e^{-i\omega x} = \cos \omega x - i \sin \omega x[/itex]

That's all you need to answer the question.
 
Last edited:
  • #3
Thanks AlpehZero - I guess it always helps to go back to the fundamental definitions ...
 

Related Threads on Conceptual Problem with Convolution Theorem

  • Last Post
Replies
4
Views
4K
Replies
1
Views
1K
Replies
4
Views
8K
Replies
1
Views
1K
  • Last Post
Replies
2
Views
2K
  • Last Post
Replies
8
Views
9K
  • Last Post
Replies
4
Views
2K
  • Last Post
Replies
1
Views
16K
  • Last Post
Replies
1
Views
2K
Replies
3
Views
2K
Top