Distributed event-triggered H∞ consensus filtering in sensor networks

This paper investigates the issue of distributed H∞ consensus filtering in mobile sensor networks where the transmission of each sensor is triggered by an event. For each node, the event triggering mechanism is given by an event-based data processor (EDP), which determines whether to transmit the sampled-data or not. The event-based filter at each node is built on the sampled data of this sensor and those received from its neighbors. The distributed event-triggered H∞ consensus filtering problem is solved by transforming it into the H∞ stability problem of an error system with an interval time-varying delay. The solution to the latter problem is derived by introducing a new event-based bounded real lemma (EBRL) using Lyapunov-Krasovskii approach. A unified co-design algorithm for the filter gains and the threshold parameters in the event conditions are finally given. The effectiveness of the algorithm is illustrated by a numerical example.