Event-Triggered Reliable Dissipative Filtering for Delayed Neural Networks with Quantization

This paper investigates the event-triggered reliable dissipative filtering for delayed neural networks with quantization. First, an event-triggered scheme is introduced to save limited network resources, by which whether or not sampled signals should be transmitted to the quantizer depends on a predefined event-triggered condition. Second, with the event-triggered scheme, a new unified sampled-data filtering error system is established to deal with the issue of dissipative filtering for the neural networks with quantization. Third, by using the Lyapunov-Krasovskii functional method, a sufficient criterion is obtained to ensure asymptotic stability and strict (Q, S, R)-alpha-dissipativity for the filtering error system. Then, based on solutions to a set of linear matrix inequalities, both proper event-triggered parameters and filter parameters can be co-designed. Finally, the effectiveness and the superiority of the proposed method are verified by numerical simulation via two examples.