pth moment noise-to-state stability of stochastic differential equations with persistent noise

TL;DR

This paper investigates the stability of stochastic differential equations with persistent noise, establishing conditions for noise-to-state stability using Lyapunov functions and providing a comprehensive theoretical framework.

Contribution

It introduces new criteria for noise-to-state stability in systems with persistent noise, including a characterization of Lyapunov functions and equivalence relations for stability analysis.

Findings

Established conditions for noise-to-state stability in probability and pth moment.

Developed a framework for comparing Lyapunov functions via inequalities and equivalence classes.

Provided examples demonstrating the application of the theoretical results.

Abstract

This paper studies the stability properties of stochastic differential equations subject to persistent noise (including the case of additive noise), which is noise that is present even at the equilibria of the underlying differential equation and does not decay with time. The class of systems we consider exhibit disturbance attenuation outside a closed, not necessarily bounded, set. We identify conditions, based on the existence of Lyapunov functions, to establish the noise-to-state stability in probability and in pth moment of the system with respect to a closed set. As part of our analysis, we study the concept of two functions being proper with respect to each other formalized via a pair of inequalities with comparison functions. We show that such inequalities define several equivalence relations for increasingly strong refinements on the comparison functions. We also provide a complete characterization of the properties that a pair of functions must satisfy to belong to the same equivalence class. This characterization allows us to provide checkable conditions to determine whether a function satisfies the requirements to be a strong NSS-Lyapunov function in probability or a pth moment NSS-Lyapunov function. Several examples illustrate our results.