Event-triggered resilient filtering with missing measurements

This paper investigates the remote state estimation problems for a class of linear discrete-time systems. An event-triggered scheme that schedules the transmissions between the sensor and the remote estimator is introduced so as to preserve the network resources. The communication process is assumed to suffer from the missing measurement phenomenon described by a Bernoulli distribution. Additionally, the random estimator gain perturbations are considered in the realization of state estimation problems. To deal with such issues, we propose an event-triggered resilient filter algorithm. Note that the analytical expressions of the error covariance of the proposed filter cannot be computed directly. Consequently, we construct its upper bound as an alternative and subsequently design the suboptimal filter gains in order that such a bound is minimized at each step. The effectiveness of the proposed filtering algorithm is illustrated by a numerical simulation.