1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Complex Fourier Series

  1. Sep 28, 2009 #1
    Having some trouble with this, any help is appreciated

    1. The problem statement, all variables and given/known data
    Give the complex fourier series for [tex]f(x) = 2 - x, -2<x<2[/tex]

    2. Relevant equations
    [tex]f(x) = \sum_{n=-\infty}^\infty C_ne^{\frac{i n \pi x}{l}[/tex]

    [tex]C_n=\frac{1}{2l} \int_{-l}^l f(x)e^{\frac{-i n \pi x}{2}} dx [/tex]

    3. The attempt at a solution
    [tex]f(x) = 2 - x, l = 2[/tex]

    [tex]f(x) = \sum_{n=-\infty}^\infty \frac{1}{4} \int_{-2}^2 ((2-x)e^{\frac{-i n \pi x}{2}}dx) e^{\frac{i n \pi x}{2} [/tex]

    now here, I don't know which steps I should take next. Should I take this integral for general n? Or should I break it up into the case for when n=0, n>0, and n<0?

    I've kind of tried both to no avail :(
  2. jcsd
  3. Sep 29, 2009 #2


    User Avatar
    Homework Helper

    from a quick look here's a ew helpful (hopefully) tips:

    the sum over negative & positive n values in the complex exponent, is equivalent to summing only over positive n's with both sin & cos terms, & this one might be easier to evaluate in terms of sin's & cos's

    also you should find a proof of the integral of sin(2.pi.n.x/L).sin(2.pi.n.x/L) over L = 0 for integers m not equal p and similar for cosines & mixes, this will allow you to write each coefficient in terms of single integral

    now consider your function
    g(x) = 2-x

    first consider the constant part, 2. Any integral of a sinusoid over an integer times period will be zero, real or complex so only the n= 0 (or constant cos part will contribute)

    now -x is an odd function, the cos integrals will be zero, (assuming it is just repeated),so only sin's will contribute...

    so you should be able to reduce the integrals to required to ~x.sin(n.pi.x/2)
  4. Sep 29, 2009 #3
    yeah, i think that was part a), they had us find the fourier series in sin/cos form, now they want us to re-derive it with complex exponentials

    as a side note, i didnt think about splitting up the 2 - x into both 2 and the -x part, that makes sense now and seems quicker for part a :)
  5. Sep 29, 2009 #4


    User Avatar
    Homework Helper

    not sure if you are allowed, but you could always write your complex exponential as:
    [tex] e^{i x } = cos(x) + i sin(x) [/tex]
    then use what you know about sin & cos integrals - evaluating the integrals will essentially be the same then...

    and the negative n ones will just be complex conjugates of the poistives
  6. Sep 29, 2009 #5
    i think they want it purely in the e^ix form :(

    just stuck on whether to worry about positive/negative cases and whether i do or dont how to simplify it to something reasonable
  7. Sep 29, 2009 #6


    User Avatar
    Homework Helper

    hmm.. by separating 2 & -x should be able to integrate *2 easy,for n = 0

    then simple int by parts on the *(-x) for when n is not equal to zero

    knowing what is going to happen to the terms based on what you have already done (sin & cos), i would attempt it for a generic n & -n, then show what happens when you sum the result... so express you results as a sum from 0 to n
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook