Algebraic approach to stochastic optimal controller design—multivariable case

The paper generalizes, to the multivariable case, the results obtained by the authors in a previous paper concerning the design of an optimal digital controller. The stochastic plant is described by its transfer function matrix and the objective is the minimization of a linear combination of quadratic measures of system inputs and outputs on an infinite time range. The design procedure uses polynomial algebraic techniques and requires only the solution of a system of two polynomial matrix equations. The results are extended to the case when the matrix obtained from spectral factorization is not of full rank.