Estimation of events flow generated with ergodic semi-Markov processes

The paper considers the events flow generated with the ergodic semi-Markov processes which are rather adequate formalisms to describe informational-measuring systems under the Von Neumann computer control. Pearson's regression and correlation criteria for estimation of the proximity events flow to the Poisson flow are investigated. It is shown that the procedure used for such criteria calculation has a high computational complexity. The authors newly obtained the formulae for a parametric criterion and the criterion based on the waiting function properties. The authors demonstrated that the waiting function criterion has the least computational complexity. The waiting function criterion applicability is verified with the Monte-Carlo method.