Calculation of derivatives fisher information matrix in problem of active identification stochastic linear systems with input signal parameterization
Vladimir M. Chubich, Елена Филиппова · 2016
For stochastic linear continuous-discrete models with unknown parameters included to the equations of states and measurements, to the initial conditions and the covariance matrices of dynamics and measurement noises recurrent analytical relations of derivatives Fisher information matrix in terms of the input signal parameterization are obtained. The corresponding computational algorithm is developed.