What is the Correct Fourier Series for e^x?

AI Thread Summary
The discussion revolves around the calculation of the Fourier series for the function e^x over the interval [-π, π]. The user presents their derived coefficients a_0, a_n, and b_n, leading to a Fourier series expression that differs from their professor's by a sign in the summation term. Other participants suggest that the discrepancy likely arises from errors in the integration steps, particularly regarding the sign of the integral involving e^x. Additionally, there is a query about the integration of e^x and its relation to the hyperbolic sine function. Clarification on these calculations and concepts is sought to resolve the differences in the Fourier series results.
twoflower
Messages
363
Reaction score
0
Hi all,

I've been having little problems getting Fourier series of e^x.

I have given

<br /> f(x) = e^{x}, x \in [-\pi, \pi)<br />

Then

<br /> a_0 = \frac{1}{\pi}\int_{-\pi}^{\pi} e^{x}\ dx = \frac{2\sinh \pi}{\pi}<br />

<br /> a_{n} = \frac{1}{\pi}\int_{-\pi}^{\pi} e^{x}\cos (nx)\ dx = <br /> <br /> \frac{1}{\pi}\left\{\left[ \frac{1}{n}\ e^{x} \sin (nx)\right]_{-\pi}^{\pi} - <br /> <br /> \frac{1}{n}\int_{-\pi}^{\pi} e^{x}\sin (nx)\ dx\right\} = <br /> <br /> \frac{(-1)^{n+1}\left(e^{-\pi} - e^{\pi}\right)}{\pi(1+n^2)}<br />

<br /> b_{n} = \frac{1}{\pi}\int_{-\pi}^{\pi}e^{x}\sin (nx)\ dx = \frac{1}{\pi}\left\{ <br /> <br /> \left[ -\frac{1}{n}\ e^{x}\cos (nx)\right]_{-\pi}^{\pi} + <br /> <br /> \frac{1}{n}\int_{-\pi}^{\pi} e^{x}\cos(nx)\ dx\right\} = <br /> <br /> \frac{n(-1)^{n}\left(e^{-\pi} - e^{\pi}\right)}{\pi(1+n^2)}<br />

So the Fourier series looks like this:

<br /> S(f,x) = \frac{\sinh \pi}{\pi}\ +\ \frac{2\sinh \pi}{\pi}\sum_{n=1}^{\infty} <br /> <br /> \frac{(-1)^{n+1}}{1+n^2}\left(\cos(nx) - n\sin(nx))<br />

Anyway, our professor gave us another right (I hope so) result:

<br /> S(f,x) = \frac{\sinh \pi}{\pi}\ +\ \frac{2\sinh \pi}{\pi}\sum_{n=1}^{\infty} <br /> <br /> \frac{(-1)^{n}}{1+n^2}\left(\cos(nx) - n\sin(nx))<br />

Obviously my series is just of opposite sign than it should be, but I can't find
the mistake, could you help me please?
 
Last edited:
Physics news on Phys.org
twoflower,

Am I missing something? The two solutions look the same to me!
 
jdavel, twoflower's solution has (-1)^(n + 1) inside the sum, while the professor's solution has (-1)^n.
 
Last edited:
Muzza said:
jdavel, twoflower's solution has (-1)^(n + 1) inside the sum, while the professor's solution has (-1)^n.

Muzza,

Doh!
 
You must have made a mistake in the integrals. It's hard to tell because you've skipped 3 or 4 steps in your post, but it looks like at least your bottom integral is wrong (wrong sign) for the case where n is odd (the only case I felt like doing in my head).
 
as far as i can tell, when you have taken (e^-pi - e^pi) out from the summation you have left it as sinh(pi) whereas it should be -sinh(pi), this should then match with the professors solution.
I have a similar question, using this series i must show that \sum1/(1 + n^2) = 1/2(picoth(pi) -1), however when i calculate it to be 1/2(picoth(pi) + pi - 1), can anyone help?
 
Why is integration of e^x=2sinh (pie)/pie ?
i am not able to understand this concept can some body enlighten this for me !
 
Sinh(x)= 1/2 [e^x - e^(-x)]
 
Back
Top