Probabilistic and statistical analysis of chaotic information flows in telecommunication networks by using the method of moving separation of mixtures

The behavior of chaotic information flows in complex telecommunication networks is studied. A stochastic model of telecommunication network is proposed which has the form of a superposition of some simple structures. As a result, this model generates mixtures of gamma- distributions for the execution times of claims. The problem of separation of mixtures of exponential and gamma-distributions is solved by the EM-algorithm. An attempt is made of the interpretation of the obtained results.