I'm trying to verify this \begin{equation} \int{d^3\vec{r}\ } \langle\rho\rangle_T\; = e, \end{equation} where $\langle\rho\rangle_T$ is the temporal average of $\rho(x,t) = e\delta(x-a\sin{\omega t})$ and $e$, $a$ and $\omega$ are constants.
If $\rho(x,t) = e\delta(x-a\sin{\omega t})$, the problem is reduced to \begin{equation} \int_{-\infty}^{\infty}{dx\ } \langle\rho\rangle_T\; = \int_{-a}^{a}{dx\ } \langle\rho\rangle_T, \end{equation} but I don't know how to solve the integral that appeared: \begin{equation} \langle\rho\rangle_T\; = \frac{e}{T} \int_{0}^{T}{dt\ } \delta(x-a\sin{\omega t}). \end{equation} I have tried to use an elementary $u$-sustitution: $u = a\sin{\omega t}$, then
$$du dt\ a\omega\cos{\omega t} \quad \leftrightarrow \quad \frac{1}{a\omega} \left[1-\left(\frac{u}{a}\right)^2\right]^{-1/2}\ du = dt. $$ Or \begin{equation} \langle\rho\rangle_T\; = \frac{e}{a\omega T} \int_{0}^{T}{du\ } \left[1-\left(\frac{u}{a}\right)^2\right]^{-1/2}\delta(x-u). \end{equation}
Is there a property than I can use? Is the $u$-sustitution valid?