1. for a deterministic function, correlation function of the function and a time-delayed replica
2. for a stationary random function, mathematical expectation of the product of the function and a time-delayed replica:

   ${\displaystyle C(t)=E[f(\tau )f(t+\tau )]}$

Note 1 to entry: The autocorrelation function of a deterministic function or a stationary random function is the inverse Fourier transform of its power spectral density.

Note 2 to entry: When a stationary random function can be considered as ergodic, its autocorrelation function can be calculated from a particular sample:

${\displaystyle C(t)=\underset{T\to \infty }{\text{lim}}\frac{1}{2T}{\int }_{-T}^{+T}f(\tau )f(t+\tau )\mathrm{d}\tau }$