IEEE Transactions on Systems, Man, and Cybernetics | 2019
Event-Triggered H∞ Filtering for Networked Switched Systems With Packet Disorders
Abstract
This paper deals with event-triggered H∞ filtering problem for switched continuous-time systems with quantization. A multiple periodic sampling-based event-triggering strategy is adopted to select those ``necessary sampled signals to be transmitted. As a result, the amount of communication and the frequency of signal updates can be reduced significantly. Regarding the nonideal communication network, we consider network-induced delays and packet disorders, which are common during packet transmission through networks. In order to cope with packet disorders, an active packet loss approach is introduced. As such, a new time-delay system model is established, by which the filtering error system is modeled as a switched system with an interval time-varying delay. Then, by employing piecewise Lyapunov functional method and average dwell time technique, some criteria are derived to ensure that the filtering error system can achieve a prescribed level of H∞ performance. Moreover, the relationship between switching instants and the data updating instants in filter is discussed in detail, and the filter gains and event-triggering parameters can be jointly designed in terms of solutions to some linear matrix inequalities. Finally, numerical simulation is provided to verify the effectiveness of the proposed method.