Event-based Filtering with Individual Triggering Thresholds in Wireless Sensor Network: Distributed Detectability Analysis

In this paper, we investigate the distributed filtering problems for a class of nonlinear continuous-time stochastic systems over wireless sensor networks. From a resource-efficient perspective, a multi-channel event-based mechanism is proposed to trigger the signal transmission that meets certain predefined conditions with hope to reduce the communication rate. A novel individual triggering threshold is put forward for each state component in order to reflect the fact that the system states might have different triggering rates according to the engineering practice. Our results demonstrate that the distributed filtering system under the proposed event-based communication can achieve the mean-square stability when the strongly connected network is distributedly detectable. Finally, the efficiency of the proposed filtering strategy is verified by a numerical simulation.