Compensation Strategy-Based Gaussian Approximation Fusion Filtering for Multi-Sensor Nonlinear Systems Subject to Markov DoS Attacks

The research concentrates on the Gaussian approximation fusion filtering (GAFF) issue for multi-sensor nonlinear systems under Markov denial of service (DoS) attacks. This paper aims to propose a fusion filtering algorithm for the addressed systems to cope with DoS attacks within the Gaussian approximation filtering framework, where the attacks follow the Markov property. In particular, a prediction compensation strategy is adopted to reduce the impact caused by the attacks as much as possible. In light of the available measurements, the Gaussian approximation local filtering unified framework is given, where the Gaussian weighted integrals are addressed by means of the cubature Kalman method. By using the sequential inverse covariance intersection fusion criterion, the GAFF algorithm is obtained to improve estimation accuracy. Finally, a simulation experiment with comparison tests is introduced to validate the effectiveness of the proposed GAFF method.