I was studying this theorem :

Suppose that $f$ is an integrable function on the circle with $\widehat{f}(n) = 0$ for all n ∈ Z. Then $f(θ_0)$ = 0 whenever $f$ is continuous at the point $θ_{0}$.

from the Princeton Lectures in Analysis, Fourier Series. In the proof the author then says that since the Fourier coefficients of $f$ are zero for all n and $p_k$ is a trigonometric polynomial, we must have $\int_{-\pi}^{\pi} f(\theta)p_k(\theta)\ d\theta = 0 $ for all $k$.

I have not been able to understand why this must be so. I started with the condition:

$\widehat{f}(n) = \frac{1}{L}\int_{a}^{b} f(\theta)e^{\frac{-2\pi inx}{L}} \ d\theta = 0 \ \forall n \in \mathbb{Z}$.

But I couldn't prove that this implies that any integral with trigonometric polynomial is also zero. How can this be seen?


I'm afraid I don't have the book, so I don't know what $p_k(\theta)$ is!

Fortunately, this doesn't matter. By definition, a trigonometric polynomial is a finite sum of the form, $$ p(\theta) = \sum_{n = -N}^N c_n e^{i n \theta}.$$ Given that $ \int_{-\pi}^\pi d\theta \ f(\theta) e^{i n \theta} = 2\pi \hat f(-n) = 0$ for all $n \in \mathbb N$, we have $$ \int_{-\pi}^\pi d\theta \ f (\theta) p(\theta) = \int_{-\pi}^\pi d\theta \sum_{n = -N}^N c_n f(\theta) e^{in \theta} = \sum_{n = -N}^Nc_n \int_{-\pi}^\pi d\theta \ f(\theta) e^{in\theta} =0.$$ (It is legitimate to take the sum out of the integral because the sum is over finitely many values of $n$.)

  • $\begingroup$ I did not understand how you changed the limits of integration from pi to infinity. $\endgroup$ – erdoswiles May 11 '17 at 15:26
  • $\begingroup$ @erdoswiles Sorry, that was a typo! $\endgroup$ – Kenny Wong May 11 '17 at 15:41
  • $\begingroup$ Oh ok! Thanks for the answer. $\endgroup$ – erdoswiles May 11 '17 at 15:43

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.