ABSTRACT
In this chapter, the event-triggered state estimation problem is investigated for discrete-time neural networks. For the purpose of energy saving, the event-triggered mechanism is adopted, and the measurement outputs are only transmitted to the estimator when a certain triggered condition is met. Firstly, we develop a new event-triggered estimation technique for the delayed neural networks with stochastic parameters and incomplete measurements. The incomplete information under consideration includes randomly occurring sensor saturations and quantizations. A Lyapunov functional is constructed to obtain sufficient conditions under which the estimation error dynamics is exponentially ultimately bounded in the mean square. The other research focus of this chapter is to design the event-triggered H∞ state estimators for a class of discrete-time stochastic genetic regulatory networks (that can be reviewed as a special case of neural networks) such that, in the presence of Markovian jumping parameters and time-varying delays, the estimation error dynamics is stochastically stable with a prescribed H∞ performance level. Finally, some numerical examples are presented to illustrate the effectiveness of the propose event-triggered state estimation scheme.
