A Local Approach to Distributed $H_{\infty }$-Consensus State Estimation Over Sensor Networks Under Hybrid Attacks: Dynamic Event-Triggered Scheme

Abstract

This paper deals with the distributed $H_{\infty }$-consensus state estimation problem for a class of discrete time-varying systems with integral measurements over sensor networks. The addressed target plant has its output modeled as integral measurement so as to account for the interval time taken for sample collection. The signal transmissions between an individual sensor node and its neighboring nodes are first scheduled by a dynamic event-triggered scheme (ETS) for the purpose of energy saving, and such transmissions are further prone to hybrid cyber-attacks (comprising denial-of-service and deception attacks). A distributed estimator is constructed for each node by using the available information from itself and its neighboring nodes such that the estimation error dynamics achieves the prescribed $H_{\infty }$-consensus performance in mean square sense. A local performance analysis method is developed to establish sufficient conditions that ensure the existence of the desired distributed estimators, and the corresponding estimator gains are then obtained by solving certain recursive matrix inequalities. The effectiveness of the proposed distributed estimation algorithm is illustrated through extensive simulation studies, where comparative experiments are conducted on time-triggered scheme, static ETS and dynamic ETS.