Some facts about the choice of the weighting matrices in Larimore type of subspace algorithms (Q1614294)

The effect of some weighting matrices on the asymptotic variance of the estimate of linear discrete-time state space systems estimated using subspace methods is investigated. The analysis refers to the Larimore type of subspace procedures and deals with white or without observed inputs. The main result expresses the asymptotic variance of the system matrix estimates in canonical form as function of some of the user choices, clarifying the question on how to choose them optimally. It is shown that, in comparison to the canonical correlation analysis, the weighing scheme leads to optimal accuracy. The new expression for the asymptotic variance is given. It is shown that it can be implemented for the computation of the asymptotic variance more efficiently than previously published.