Networked Nonlinear Fusion Estimation Under DoS Attacks

This paper is concerned with the distributed fusion estimation problem for nonlinear systems under denial of service (DoS) attacks, where it is not required to know the statistical property of DoS attacks and the bounds of system noises in advance. A compensation model is constructed to defend against DoS attacks in order to reduce performance loss. Following the idea of bounded recursive optimization, two convex optimization problems, which are established in terms of linear matrix inequalities, are designed to obtain the local optimal estimator gains and the optimal weighting matrices by minimizing the upper bounds of estimation errors at each time. Finally, an intelligent vehicle localization system is used to show the validity of the results.