Data-Driven Distributed Mitigation Strategies and Analysis of Mutating Epidemic Processes

Abstract

In this paper we study a discrete-time SIS (susceptible-infected-susceptible) model, where the infection and healing parameters and the underlying network may change over time. We provide conditions for the model to be well-defined and study its stability. For systems with homogeneous infection rates over symmetric graphs, we provide a sufficient condition for global exponential stability (GES) of the healthy state, that is, where the virus is eradicated. For systems with heterogeneous virus spread over directed graphs, provided that the variation is not too fast, a sufficient condition for GES of the healthy state is established. Appealing to the first stability result, we present two data-driven mitigation strategies that set the healing parameters in a centralized and a distributed manner, respectively, in order to drive the system to the healthy state.