Sequential Covariance Intersection-Based Distributed Nonlinear State Estimation Under Denial of Service

This paper investigates the nonlinear distributed secure state estimation problem in peer-to-peer sensor networks under denial of service (DoS) attacks. During data transmission, DoS attackers block the exchange of information by jamming the wireless communication channels, which can degrade the filtering performance. To ensure estimation consistency in attack-prone environments, a consensus on information approach is utilized to fuse the available local information. Moreover, considering the cross-correlations between different nodes, a sequential covariance intersection-based distributed cubature Kalman filtering (SCIDCKF) algorithm is proposed in a nonlinear setting. Afterward, the boundedness of estimation error covariances in the proposed fusion framework is rigorously analyzed through theoretical derivation. Finally, the effectiveness of the theoretical results is validated by a simulation example.