An Input-Output Approach to Structured Stochastic Uncertainty

TL;DR

This paper develops an input-output framework for analyzing linear systems with stochastic uncertainties, providing conditions for stability and performance that are applicable even with correlated uncertainties and without relying on state space models.

Contribution

It introduces a purely input-output approach to characterize mean-square stability in stochastic systems, extending applicability to distributed systems and correlated uncertainties.

Findings

Necessary and sufficient spectral radius conditions for stability.

LMI equivalents for systems with state space realizations.

Applicability to correlated uncertainties and distributed systems.

Abstract

We consider linear time invariant systems with exogenous stochastic disturbances, and in feedback with structured stochastic uncertainties. This setting encompasses linear systems with both additive and multiplicative noise. Our concern is to characterize second-order properties such as mean-square stability and performance. A purely input-output treatment of these systems is given without recourse to state space models, and thus the results are applicable to certain classes of distributed systems. We derive necessary and sufficient conditions for mean-square stability in terms of the spectral radius of a linear matrix operator whose dimension is that of the number of uncertainties, rather than the dimension of any underlying state space models. Our condition is applicable to the case of correlated uncertainties, and reproduces earlier results for uncorrelated uncertainties. For cases where state space realizations are given, Linear Matrix Inequality (LMI) equivalents of the input-output conditions are given.