Optimal Discrete Estimation of Reference Points of a Discrete-Continuous Markov Process against the Background of Correlated Markov Noise

In this paper, we solve the problem of synthesizing optimal and quasi-optimal algorithms for estimating the reference points of continuous components of a vector discrete-continuous Markov random process with allowance for the known statistical characteristics of additive Markovian correlated noise using the methods of the Markov theory of estimation of random processes. The method of difference measurements was used while synthesizing the algorithms. A block diagram of a quasi-optimal digital filter is given. Using a simple example, simulation shows the performance of a quasi-optimal algorithm in statistically uncertain situations.