Fourier Series: Can even functions be changed to odd?

Click For Summary

Discussion Overview

The discussion revolves around the possibility of modifying even functions to odd functions when creating Fourier series, particularly in the context of solving heat equations. Participants explore the implications of such modifications and their effects on the representation of functions defined in a limited region.

Discussion Character

  • Exploratory
  • Technical explanation
  • Conceptual clarification
  • Debate/contested

Main Points Raised

  • One participant questions whether it is feasible to change an even function, such as f(x) = x sin x, to an odd function by modifying it to f(x) = |x| sin x while retaining the function's characteristics in the interval [0, L].
  • Another participant seeks clarification on what is meant by "retaining the desired information" and asks for specifics regarding the intended outcome of such modifications.
  • A participant explains that the application of these functions is often in solving heat equations, where the function must be odd to correspond with Fourier sine series.
  • One participant concludes that to find a sine series converging to a function f, it is necessary to extend the function to an odd function over the entire interval [-L, L], supporting the idea of using |x| sin x on [-π, π].

Areas of Agreement / Disagreement

Participants express differing views on the implications of modifying functions for Fourier series. While there is some agreement on the need for odd functions in certain applications, the discussion remains unresolved regarding the specific consequences and validity of such modifications.

Contextual Notes

The discussion does not resolve the mathematical implications of changing function parity, nor does it clarify the assumptions underlying the modifications proposed by participants.

thelema418
Messages
131
Reaction score
4
When creating a Fourier series for a function [itex]f(x)[/itex], I consider whether the function is odd or even first. Yet, often these functions are in the positive region [itex][0, L][/itex].

Since [itex]f(x)[/itex] is only defined in this region, can I change the function to get a desired parity? By example, my concern originated with the function [itex]f(x) = x \sin x[/itex]. This is an even function, but I could modify the function as [itex]f(x) = |x| \sin x[/itex] to make it odd while retaining the desired information in [itex][0, L][/itex].

Can this be done? And are there problems with doing this?

Thanks.
 
Physics news on Phys.org
What do you mean with "retaining the desired information"? What exactly is it that you want to do?
 
micromass said:
What do you mean with "retaining the desired information"? What exactly is it that you want to do?

Meaning that [itex]x \sin x[/itex] on [itex][0,\pi][/itex] is the same as [itex]|x| \sin x[/itex] on [itex][0,\pi][/itex].

The application is usually for solving heat equations. So the region is usually x = 0 to x = L in the problems I see. Again, another example: When I work out all the details of a solution by eigenfunction expansion based on homogeneous Dirichlet boundary conditions, I end up with a Fourier sine series equal to some function [itex]f(x)[/itex], where [itex]f(x)[/itex] is the initial condition u(x,0). But to have a Fourier sine series, [itex]f(x)[/itex] must be an odd function, right?
 
thelema418 said:
Meaning that [itex]x \sin x[/itex] on [itex][0,\pi][/itex] is the same as [itex]|x| \sin x[/itex] on [itex][0,\pi][/itex].

The application is usually for solving heat equations. So the region is usually x = 0 to x = L in the problems I see. Again, another example: When I work out all the details of a solution by eigenfunction expansion based on homogeneous Dirichlet boundary conditions, I end up with a Fourier sine series equal to some function [itex]f(x)[/itex], where [itex]f(x)[/itex] is the initial condition u(x,0). But to have a Fourier sine series, [itex]f(x)[/itex] must be an odd function, right?

OK. You're right then. The right way to find a sine series converging to ##f## is to extend it to an odd function on entire ##[-L,L]##. So you'll have to work with ##|x|\sin(x)## on ##[-\pi,\pi]##, like you suggested.
 

Similar threads

  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 33 ·
2
Replies
33
Views
4K
  • · Replies 11 ·
Replies
11
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 8 ·
Replies
8
Views
5K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
3
Views
2K