Distributed Set-Membership Filtering Over Wireless Sensor Networks With Coding-Decoding Mechanism

This paper investigates the distributed set-membership filtering for a class of discrete systems over wireless sensor networks. A new type of coding-decoding method with mixed quantizer is proposed to balance the upper bound of decoding error and the communication burden. Then, a distributed filter is designed based on decoding measurements, and a recursive technique is utilized to ascertain sufficient conditions of the feasibility of the proposed filtering design. A fusion rule is further presented to achieve a more compact effective region by determining the intersection of the local estimation and prediction ellipsoids. A fusion estimate is then derived by addressing two optimization problems, which outperforms all local estimates. Finally, a simulation example is introduced to verify the effectiveness of the developed filtering algorithm.