Proving the Fourier Series for a Real, Odd Function

Click For Summary

Discussion Overview

The discussion revolves around the proof of the Fourier series representation for a real, odd function, specifically focusing on the coefficients \( b_n \) in the series expansion. Participants explore the mathematical derivation and the necessary integrals involved in establishing the relationship between the function and its Fourier series representation.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • Robert presents a question regarding the proof of the Fourier series for a real, odd function and seeks assistance in understanding the derivation of the coefficients \( b_n \).
  • Warren explains that the coefficients \( b_n \) can be derived from the integral of the function multiplied by \( \sin(n \omega t) \), using the periodicity condition of the function.
  • Robert requests clarification on a specific step in Warren's explanation, particularly how the summation over \( n \) simplifies to a single term during integration.
  • A later reply introduces trigonometric identities to evaluate integrals involving products of sine functions, highlighting that the integral evaluates to zero for \( n \neq m \) and provides the result for \( n = m \).
  • The discussion includes a detailed breakdown of the integration process and the justification for dropping the summation to focus on the non-zero term, leading to the expression for \( b_n \).

Areas of Agreement / Disagreement

Participants engage in a technical exploration of the proof, with some agreement on the steps involved in deriving \( b_n \), but there is no consensus on the clarity of the explanation provided, as Robert seeks further detail on specific steps.

Contextual Notes

The discussion includes assumptions regarding the periodicity of the function and the validity of the trigonometric identities used in the integration process. There are unresolved aspects related to the transition from summation to integration, which some participants find unclear.

Who May Find This Useful

This discussion may be useful for students and educators in mathematics and physics, particularly those interested in Fourier analysis and the properties of odd functions in relation to Fourier series.

robgb
Messages
17
Reaction score
0
The Fourier Series!

Hi guys, I'm having a bit a trouble helping my daughter with this question on the Fourier series approximation:

The Fourier series for a real, odd function, f(t) can be written as:
f(t) = [SUM to infinity, n=1, of]: b[subscipt n] sin(nwt)
where f(t=T)=f(t) and w=(2[pie])/T

Prove that b[subscipt n] = 2/T [integral between T/2 & -T/2 of]: f(t)sin(nwt)dt

Sorry about not knowing how to do all the symbols etc, but if you write it out I'm pretty sure it makes sence.

Anyway, if anyone could give me a hand with how to go about proving this, it would be greatly appreciated.

Many thanks, Robert.
 
Physics news on Phys.org
Hey Robert,

You have a function of t represented by the series

[tex]f(t) = \sum_{n=1}^\infty b_n \sin{n \omega t}[/tex]

And you'd like to know what the [tex]\inline{b_n}[/tex] are. You're also given the condition [tex]\inline{f(t + T) = f(t)}[/tex], which just indicates the function's period is [tex]\inline{T}[/tex].

You're not even required to figure out what the [tex]\inline{b_n}[/tex] are; you're just asked to demonstrate that they are, in fact

[tex]b_n = \int_{-T/2}^{T/2} f(t) \sin{n \omega t}\ dt[/tex].

To do this, first note the following fact:

[tex]\begin{equation*}<br /> \begin{split}<br /> \int_{-T/2}^{T/2} \sin^2{\left( \frac{2 \pi n t}{T} \right) }\ dt &= \frac{T}{2}<br /> \end{split}<br /> \end{equation*}[/tex]

for all n, and use it to perform the following steps:

[tex]\begin{equation*}<br /> \begin{split}<br /> \int_{-T/2}^{T/2} f(t) \sin{n \omega t}\ dt &= \int_{-T/2}^{T/2} \left[ \sum_{n=1}^\infty b_n \sin{n \omega t} \right] \sin{n \omega t}\ dt\\<br /> &= \sum_{n=1}^\infty \int_{-T/2}^{T/2} b_n \sin^2{n \omega t}\ dt\\<br /> &= \frac{T}{2} b_n<br /> \end{split}<br /> \end{equation*}[/tex]

Does this make sense?

- Warren
 
Thanks mate, yep i essentially understand, just 1 question though:

How did you get from the second line to the third line in the following:

[tex]\begin{equation*}<br /> \begin{split}<br /> \int_{-T/2}^{T/2} f(t) \sin{n \omega t}\ dt &= \int_{-T/2}^{T/2} \left[ \sum_{n=1}^\infty b_n \sin{n \omega t} \right] \sin{n \omega t}\ dt\\<br /> &= \sum_{n=1}^\infty \int_{-T/2}^{T/2} b_n \sin^2{n \omega t}\ dt\\<br /> &= \frac{T}{2} b_n<br /> \end{split}<br /> \end{equation*}[/tex]

Could you work me though that bit in a little more detail mate?

Thanks a lot,
Robert.
 
Originally posted by robgb
Thanks mate, yep i essentially understand, just 1 question though:

How did you get from the second line to the third line in the following:

Could you work me though that bit in a little more detail mate?

Thanks a lot,
Robert.
i think chroot pulled a fast one on you. there is a bit more to the trick that you have to realize to make the infinite summation turn into a single term, with one integration.

to see this, first let me remind you of some trigonometric identities:

[tex] \cos(a\pm b)=\cos a\cos b \mp\sin a\sin b[/tex]
so
[tex] \begin{align}<br /> &\frac{1}{2}(\cos(a-b)-\cos(a+b))\\<br /> &=-\frac{1}{2}\cos a\cos b+\frac{1}{2}\sin a\sin b\\<br /> & -\frac{1}{2}\cos a\cos b+\frac{1}{2}\sin a\sin b\\<br /> &= \sin a\sin b<br /> \end{align}[/tex]

we need to use this identity to evaluate some integrals:

[tex] \begin{align}<br /> &\int_{-T/2}^{T/2}\sin(\frac{2\pi mt}{T})\sin(\frac{2\pi nt}{T})\ dt=\\<br /> &\frac{1}{2}\int_{-T/2}^{T/2}\cos\frac{2\pi(n-m)t}{T}\ dt-\frac{1}{2}\int_{-T/2}^{T/2}\cos\frac{2\pi(n+m)t}{T}\ dt=\\<br /> &\left\frac{T}{4\pi(n-m)}\sin\frac{2\pi(n-m)t}{T}\right]^{T/2}_{-T/2}\\<br /> &-\left\frac{T}{4\pi(n+m)}\sin\frac{2\pi(n+m)t}{T}\right]^{T/2}_{-T/2}\\<br /> &=0<br /> \end{align}[/tex]
note that this integration is not valid when n=m, since it involves a division by zero. in the n=m case, i have:
[tex] \begin{align}<br /> \int_{-T/2}^{T/2}\sin^2\frac{2\pi nt}{T}\! dt\\<br /> =\frac{1}{2}\int_{-T/2}^{T/2}dt-\frac{1}{2}\int_{-T/2}^{T/2}\cos\frac{4\pi nt}{T}dt\\<br /> =T/2<br /> \end{align}[/tex]

so, now, armed with these integrals, we can do the problem.

i am going to rename the dummy index n, in the infinite summation to m. you will see why.
[tex] f(t) = \sum_m^\infty b_m \sin\frac{2\pi mt}{T}[/tex]

then


[tex]\begin{equation*}<br /> \begin{split}<br /> \int_{-T/2}^{T/2} f(t) \sin{n \omega t}\ dt &= \int_{-T/2}^{T/2} \left[ \sum_m^\infty b_m \sin\frac{2\pi mt}{T}\right] \sin\frac{2\pi nt}{T}dt\\<br /> &= \sum_m^\infty b_m\int_{-T/2}^{T/2} \sin\frac{2\pi nt}{T}\sin\frac{2\pi mt}{T}\ dt<br /> \end{split}<br /> \end{equation*}[/tex]

this summation contains an infinite number of terms, one for every value of m, but only when m=n is the term nonzero, so i will drop the summation, and change m to n, so i am only considering one integral, and the value of that integral is T/2. this is the only nonzero term of the infinite summation.

[tex] \int_{-T/2}^{T/2} f(t) \sin{n \omega t}\ dt =\frac{T}{2}b_n[/tex]
or
[tex] b_n=\frac{2}{T}\int_{-T/2}^{T/2} f(t) \sin{n \omega t}\ dt[/tex]

and that s it. any questions?
 
Last edited:

Similar threads

  • · Replies 3 ·
Replies
3
Views
2K
Replies
3
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 33 ·
2
Replies
33
Views
4K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 4 ·
Replies
4
Views
4K