IJCAS

In this paper, the distributed H∞ filtering problem is investigated for a class of nonlinear systems over wireless sensor networks (WSNs) with successive packet dropouts (SPDs) and multiplicative noise under an eventtriggered communication mechanism (ETCM). A series of uncorrelated random variables which obey the Bernoulli distribution are used to characterize the induced phenomenon of SPDs occurring in network communication. The ETCM based on the relative error between the estimated signals is introduced to reduce the burden of communication and achieve the purpose of saving network resources. The measurement information on each sensor node can be transmitted to its neighboring nodes only when the certain triggering condition is satisfied. In view of WSNs, a distributed filtering error system model is firstly built to take SPDs together with multiplicative noise into account. Then, sufficient conditions are established to ensure both the prescribed H∞ performance requirement and the filtering error variance constraint over a finite-horizon. Moreover, the design method of filter parameters is given in terms of matrix inequalities. Finally, a numerical example is presented to verify the effectiveness and feasibility of the proposed distributed H∞ filters design scheme.

Keywords: Distributed H∞ filtering, event-triggered communication mechanism, filtering error variance constraint, successive packet dropouts, wireless sensor networks.