A Novel Distributed Filter for Sensor Networks Under Deception Attacks

This study proposes a novel distributed filter in which each sensor simultaneously transmits both predicted states and measurements to neighboring sensors, designed to operate effectively even when subjected to deception attacks. By introducing the upper bound of the estimation error covariance matrix, the proposed filter can operate without knowledge of whether the received information has been compromised by attacks. Moreover, eliminating the need to iterate cross-covariance matrices significantly reduces the computational complexity of the filter implementation. The effectiveness of the proposed distributed filter is demonstrated through simulation results.