Event-triggered state estimation for complex networks under deception attacks: a partial-nodes-based approach

This paper addresses the issue of state estimation for a kind of complex network (CN) with distributed delays and random interference through output measurements. In the data transmission, the deception attacks are taken into account by resorting to a sequence of Bernoulli random variables with a given probability. Considering the complexity of the network, the fact that only partial output measurements are available in practical environments presents a new challenge. Therefore, the partial-nodes-based (PNB) state estimation problem is proposed. For the sake of data collision avoidance and energy saving, a general event-triggered scheme is adopted in the design of the estimator. A novel estimator is constructed to consider both cyber attacks and resource limitations, filling the gap in previous results on PNB state estimation. By using the Lyapunov method and several stochastic analysis techniques, a few sufficient conditions are derived to guarantee the desired security and convergency performance for the overall estimation error. The estimator gains are obtained by solving a set of matrix inequalities with nonlinear constraints. At last, two examples and simulations are presented to further show the efficiency of the proposed method.