Partial-Nodes-Based Scalable H∞-Consensus Filtering With Censored Measurements Over Sensor Networks

This paper deals with the scalable distributed H-infinity-consensus filtering problem for a class of discrete timevarying systems subject to multiplicative noises and censored measurements over sensor networks (SNs). For the underlying SN, it is assumed that only the measurement outputs from partial sensor nodes are available. Also, the phenomenon of censored measurements is taken into account to reflect the limited capability in measuring. A new H-infinity-consensus performance index is put forward to evaluate the disturbance rejection level of the filters against the simultaneous presence of external disturbances, initial conditions, as well as censoring effects. By utilizing the vector dissipativity theory and the recursive matrix inequality technique, sufficient conditions are established under which the prescribed H-infinity-consensus performance index is achieved. The parameters of the desired distributed filters are calculated via solving certain matrix inequalities, where such a calculation is conducted in a local sense so as to preserve the scalability of the filter design. Finally, a numerical simulation example is provided to demonstrate the validity and applicability of the proposed filtering strategy.

Automated summary

This paper addresses the scalable distributed H-infinity-consensus filtering problem for discrete time-varying systems subject to multiplicative noises and censored measurements over sensor networks. It introduces a new H-infinity-consensus performance index and establishes sufficient conditions for achieving the desired performance index using certain techniques.