Why is this function not ##L^1(\mathbb{R} \times \mathbb{R})##?

Click For Summary
SUMMARY

The discussion centers on the inability to swap integrals in the expression for the Fourier inverse theorem due to the function \( f(u)e^{i\omega(t-u)} \) not being in \( L^1(\mathbb{R} \times \mathbb{R}) \). Participants clarify that the complex exponential is not always bounded, and properties of \( f(u) \) must ensure rapid decay for integrability. The discussion emphasizes that neither \( \sin \) nor \( \cos \) is integrable over \( \mathbb{R} \), which contributes to the divergence of the inner integral when attempting to swap the order of integration.

PREREQUISITES
  • Understanding of Fourier transforms and inverse Fourier transforms
  • Knowledge of \( L^1 \) spaces and integrability conditions
  • Familiarity with properties of complex exponentials
  • Basic concepts of boundedness and convergence in integrals
NEXT STEPS
  • Study the properties of \( L^1(\mathbb{R}) \) and conditions for functions to belong to this space
  • Learn about the implications of the Riemann-Lebesgue lemma on Fourier transforms
  • Explore examples of functions that are not integrable over \( \mathbb{R} \) and their implications in Fourier analysis
  • Investigate the conditions under which integral swapping is valid in multiple integrals
USEFUL FOR

Mathematicians, physicists, and students studying Fourier analysis, particularly those interested in the properties of integrable functions and the behavior of Fourier transforms.

laurabon
Messages
17
Reaction score
0
Hi everyone in the following expression
##f(t)=\frac{1}{2 \pi} \int\left(\int f(u) e^{-i \omega u} d u\right) e^{i \omega t} d \omega ##

the book says I can't swap integrals bacause the function

##f(u) e^{i \omega(t-u)}## is not ## L^1(\mathbb{R} \times \mathbb{R})##

why ? complex exponential is not always bounded?
 
Physics news on Phys.org
What properties does ##f(u)## have?
 
FactChecker said:
What properties does ##f(u)## have?
it is Fuorier inverse theorem with ##f## in ##L^1(\mathbb{R})##
 
laurabon said:
Hi everyone in the following expression
##f(t)=\frac{1}{2 \pi} \int\left(\int f(u) e^{-i \omega u} d u\right) e^{i \omega t} d \omega ##

the book says I can't swap integrals bacause the function

##f(u) e^{i \omega(t-u)}## is not ## L^1(\mathbb{R} \times \mathbb{R})##

why ? complex exponential is not always bounded?
Note that ##\sin## and ##\cos## are not integrable over ##\mathbb R##.
 
  • Like
Likes   Reactions: FactChecker
PeroK said:
Note that ##\sin## and ##\cos## are not integrable over ##\mathbb R##.
I came up with this . Is correct? ##
\begin{aligned}
& \left|f(u) e^{i \omega(t-u)}\right|=|f(u)| \\
& \iint|f(u)|=\int_{-\infty}^{\infty} c = \infty
\end{aligned}##

how should I use the fact that sin and cosine are not integrable ?I mean to verify that is in L1 I always have to use the fact that ##|e^{i \omega(t-u)}|=1## I think
 
laurabon said:
I came up with this . Is correct? ##
\begin{aligned}
& \left|f(u) e^{i \omega(t-u)}\right|=|f(u)| \\
& \iint|f(u)|=\int_{-\infty}^{\infty} c = \infty
\end{aligned}##

how should I use the fact that sin and cosine are not integrable ?I mean to verify that is in L1 I always have to use the fact that ##|e^{i \omega(t-u)}|=1## I think
I don't think that works. First, let's take ##t = 0##, where the integral reduces to:
$$f(0) = \int_{-\infty}^{+\infty} \bigg ( \int_{-\infty}^{+\infty}f(u)e^{-iwu}\ du \bigg ) \ dw$$If we try to swap the integrals and look at:$$\int_{-\infty}^{+\infty} f(u) \bigg ( \int_{-\infty}^{+\infty}e^{-iwu}\ dw \bigg ) \ du$$We can see immediately that the inner integral does not converge and does not represent a well-defined function of ##u##. Because neither ##\cos(wu)## nor ##\sin(wu)## is integrable with respect to ##w## on all of ##\mathbb R##.
 
  • Like
Likes   Reactions: FactChecker
laurabon said:
Hi everyone in the following expression
##f(t)=\frac{1}{2 \pi} \int\left(\int f(u) e^{-i \omega u} d u\right) e^{i \omega t} d \omega ##

the book says I can't swap integrals bacause the function

##f(u) e^{i \omega(t-u)}## is not ## L^1(\mathbb{R} \times \mathbb{R})##

why ? complex exponential is not always bounded?
Notice f(x)==1 is also bounded. To be able to be in ##L^1(X)## , where ##X## itself is unbounded, your f has to decrease fast-enough.
 

Similar threads

  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 0 ·
Replies
0
Views
2K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K