Solution to one class of navigation information processing problems by using adaptive Bayesian filters

The discrete stationary linear filtering problem with unknown covariance matrices of generating (system) and measurement noise is considered. An adaptive algorithm for calculating the optimal, in the root-mean square sense, estimate and an algorithm for calculating the Cramer-Rao Lower Bound (CRLB) are proposed to analyze the efficiency of suboptimal algorithms that do not require significant computational resources. The possibility of using them for these purposes is illustrated by an example.