On Strong Mixing Conditions for Stationary Gaussian Processes

This paper considers conditions, which guarantee strong mixing of stationary random Gaussian process $\xi (t)$. It is proved, for example, that if the spectral density $f(\lambda )$ of the process $\xi(t)$ is continuous and positive (parameter t is discrete) or $f(\lambda )$ is positive and uniformly continuous, and for large $\lambda $\[ \frac{m}{{\lambda ^k }} \leqq f(\lambda ) \leqq \frac{M}{{\lambda ^{k - 1} }} \] (parameter t is continuous), then strong mixing takes place.