1. Not finding help here? Sign up for a free 30min 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!

Fourier Analysis

  1. Nov 1, 2008 #1
    1. The problem statement, all variables and given/known data
    I need to find the Fourier coefficients for a function f(t)=1 projected onto trigonometric polynomials of infinite order


    2. Relevant equations
    Equation finding the first coefficient, the constant term:
    [​IMG]


    3. The attempt at a solution
    So I feel quite stupid because this should be a very simple integral. The integral of a monotonic function over an interval that is symmetric about the origin SHOULD be equal to zero.

    It seems my book has just integrated from 0 to pi for first Fourier coefficient. I'm here pulling my hair out trying to figure out why/how this is correct because the two integrals certainly are not equivalent. Can someone help me out here?
     
  2. jcsd
  3. Nov 1, 2008 #2

    gabbagabbahey

    User Avatar
    Homework Helper
    Gold Member

    Shouldn't that equation be:

    [tex]a_0=\frac{1}{\pi} \int_{-\pi}^{\pi} f(t)dt[/tex]

    which gives [itex]a_0=0[/itex]? ...Or were you trying to say that the answer key gives [itex]a_0=\frac{1}{\sqrt{2}}[/itex]?...Are you sure that you are looking for the first term [itex]a_0[/itex] and not the first non-zero term?
     
  4. Nov 2, 2008 #3
    See, that's what I think it should be (and yes I'm looking for the first term a(sub 0)). But the answer key gives the answer I posted above. I'm beginning to think its wrong. Although, going by what we think the answer is, the transform of the function f(t)=1 would just be 0, which doesn't make sense.
     
  5. Nov 2, 2008 #4
    I can post the entire answer key answer if you would like to see it
     
  6. Nov 2, 2008 #5

    gabbagabbahey

    User Avatar
    Homework Helper
    Gold Member

    Are you looking for the Fourier Transform of f(t), or the Fourier Series representation? Assuming you are looking for the latter; just because the first term in the series is zero, doesn't mean the entire series is zero.
     
  7. Nov 2, 2008 #6
    I'm looking for the transform, not the series.
     
  8. Nov 2, 2008 #7

    gabbagabbahey

    User Avatar
    Homework Helper
    Gold Member

    Well the fourier transform is [tex]\hat{f}(\omega)= \frac{1}{\sqrt{2 \pi}} \int_{-\infty}^{\infty} f(t) e^{-i\omega t} dt[/tex] which is not an infinite series with coefficients: this is why I assumed you were looking for the fourier series ...your original question makes no sense to me if you are looking for the transform and not the series!

    I'm pretty sure your looking for the series, not the transform....are you sure f(t)=1? If I were a betting man, I'd bet dollars to dimes that the question actually has f(t)=1(t), where 1(t) is the unit (Heaviside) step function, namely:

    [tex]1(t)= \left\{ \begin{array}{rl} 0, & t<0 \\ 1, & t \geq 0[/tex]

    In which case, the equation

    [tex]a_0=\frac{1}{\sqrt{2} \pi} \int_{-\pi}^{\pi} f(t)dt[/tex]

    should give you the correct series coefficient.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: Fourier Analysis
Loading...