Estimating the Distribution Parameter of Non-prolonging Random Dead Time Duration in Recurrent Semi-Synchronous Events Flow Through Maximum Likelihood

The paper explores the recurrent semi-synchronous events flow that is a common mathematical model of information flows of messages that operate in telecommunication and information-computing networks. This model belongs to the category of doubly stochastic events flows meaning that it is a stochastic process with both random arrival times and random intensity. General and special cases of this flow are presented. Functionality of the flow is considered under random non-prolonging dead time that has uniform distribution. Dead time refers to the time of event processing by the recording device. The aim of this work is to estimate a distribution parameter of the dead time duration applying maximum likelihood method. Results of statistical experiments are presented that demonstrate the high quality of the estimation.