Accuracy of discrete Markov approximation in the problems of estimation of random processes characteristics

Many computational algorithms related to Markov processes contain the covariance matrix of measurements. Approximation of covariance matrix of measurements of observed random process by the covariance matrix of Markov process is of great interest. Because it gives an opportunity to develop computationally efficient algorithms for analysis of Markov processes (parametric identification, filtering, interpolation and others). This paper presents the algorithm of approximation of the covariance matrix of the observed process by the covariance matrix of a multiply connected (m-connected) Markov process. It has been shown that for many problems such approximation provides necessary accuracy.