A Peer-to-Peer Cooperative Set-Membership Protocol for Estimation Over WSNs

This paper addresses the peer-to-peer (P2P) set-membership state estimation problem for discrete-time systems over wireless sensor networks (WSNs) under unknown but bounded perturbations. A new communication architecture is introduced that overcomes the limitations of existing methods and leads to a fully P2P estimator design over a WSN. Under the proposed architecture, the state estimation problem is solved, and the design conditions are developed in terms of a new type of cooperative recursive linear matrix inequalities (CRLMIs). In addition to the process and measurement noises, the communication noise is considered to represent the undesired network phenomena such as channel interference, electromagnetic disturbances, or environmental variations. Compared to the previous studies, the derived conditions are fully P2P and utilize neighbors’ information to calculate the optimal error bounds. Finally, the approach is validated through simulation.