Protocol-Based Distributed Security Fusion Estimation for Networked Systems With Unknown Bounded Noise Under Quantization

Abstract

In this article, a novel distributed fusion estimation approach is proposed for time-varying systems with unknown bounded noises subject to bandwidth-constrained networks and denial-of-service (DoS) attacks, where the round-robin scheduling protocol and the quantization scheme are, respectively, employed to ease the burden of networks. It is assumed that the network connecting sensors and local state estimators is vulnerable to DoS attacks. For resisting the impacts of DoS attacks, a compensation strategy is adopted. A new method has been developed to devise the local state estimators. The innovation signals from local estimators will be quantized first before entering the network. Gains of the local state estimators and the fusion weighting matrices are acquired by solving a linear matrix inequality. Finally, the effectiveness of the proposed methods is verified by a target tracking system.