Fault Detection Filter Design for Networked Systems with Deception Attacks and Communication Delays

This paper treats the problem of H-infinity fault detection for discrete-time networked systems under deception attacks, communication delays, and perturbations. To this end, a mode-dependent fault detection filter (FDF) is presented as a residual generator to detect the occurrence of the fault. As phenomena are occurring randomly via network communication, both the delays and attacks can be appropriately described by mutually independent Markov stochastic process and Bernoulli random variable, respectively. By using the Lyapunov stability theory and linear matrix inequality (LMI) technique, a new sufficient condition for the H-infinity performance analysis of the augmented systems is obtained. Furthermore, and based on the obtained condition, the mode-dependent FDF that ensures the stochastic stability and a prescribed H-infinity performance level of the corresponding augmented system is designed in terms of linear matrix inequalities (LMIs). Finally, a numerical example is presented to show usefulness of the designed mode-dependent FDF.