Estimation of random parameters by the method of prior probability maximum (Q1914339)

A method for random parameters estimation is considered. An approach is applied to solve the problem of constructing an a posteriori density function which is used to form an estimation criterion. The basic computational problems arise from the complexity of the system models and they are more complicated in the case of nonlinear system identification. The author proposes a simpler way to obtain the most probable estimates in non-Gaussian cases. The paper contains results which generalize the least-squares method and the discrete Kalman-Bucy filter, to the case of correlated perturbations and interferences. The paper will be of interest for the specialists and engineers, working on nonlinear system state estimation and identifiction.