Distributed Fault Estimation for Nonlinear Systems With Sensor Saturation and Deception Attacks Using Stochastic Communication ProtocolsOACSTPCDEI

This paper is aimed at the distributed fault estima-tion issue associated with the potential loss of actuator efficiency for a type of discrete-time nonlinear systems with sensor satura-tion.For the distributed estimation structure under consideration,an estimation center is not necessary,and the estimator derives its information from itself and neighboring nodes,which fuses the state vector and the measurement vector.In an effort to cut down data conflicts in communication networks,the stochastic commu-nication protocol(SCP)is employed so that the output signals from sensors can be selected.Additionally,a recursive security estimator scheme is created since attackers randomly inject mali-cious signals into the selected data.On this basis,sufficient condi-tions for a fault estimator with less conservatism are presented which ensure an upper bound of the estimation error covariance and the mean-square exponential boundedness of the estimating error.Finally,a numerical example is used to show the reliability and effectiveness of the considered distributed estimation algo-rithm.