# Inverse fourier transform of characteristic function of a discrete random variable

So we are given a measure $$\mathbb{P}$$ such that $$\mathbb{P}(X\in\mathbb{Z})=1$$. And we are asked to prove $$\mathbb{P}(X=n)=\frac1{2\pi}\int_0^{2\pi}\exp(-itn)\varphi_X(t)\ dt$$

where $$\varphi_X(t)$$ is a characteristic function of X.

I can believe the statement is true as I can spot inverse fourier trasform of $$\varphi_X(t)$$ which in fact is a fourier transform of $$X$$. What I did is I simply substituded $$\varphi_X(t)$$ from the definition and try to play with it but I get:

$$RHS=\frac1{2\pi}\int_0^{2\pi}\exp(-itn)\sum_{k\in\mathbb{Z}}\exp(ikt)\mathbb{P}(t=k)\ dt$$

I tried some algebraic manipulations it took me nowhere closer to the result and what's more I am not even sure if I can swap the order of the limits.

I would like to get some hint on that one and explanation why I can swap the order of the limits we are taking (if this is the right way)

Hint: First, $$\varphi_X(t) = \sum_{k \in \mathbb{Z}} e^{ik t} \mathbb{P}(X = k)$$, so you should correct those typos. Then, what is $$\frac{1}{2\pi}\int_0^{2\pi} e^{it(k-n)}\,dt$$ equal to?
Hint pt 2: To justify changing the order, show that $$\int_0^{2\pi} \sum_{k \in \mathbb{Z}} \left| e^{i t(k - n)} \mathbb{P}(X = k) \right| < \infty$$ and then use Fubini's theorem.