Attack Isolation via an Improvement Cubature Kalman Filter Algorithm for BD2/SINS Integrated Navigation Systems

Motivated by the necessity to provide secure navigation information in this context, especially against potential data injection attacks, the paper proposes an improvement cubature Kalman filter (CKF) algorithm by introducing attack detection and isolation ideas combined with residual chi-square test. The CKF is used to estimate state through the measurements. Then the residual chi-square test is incorporated through the proposed CKF framework. By calculating the value of residual chi-square test and comparing with the detection threshold, we can detect the attacks. Finally, an isolation strategy is proposed to avoid attack measurements being injected into the system. To be concrete, the BeiDou-2 (BD2)/ Strap-down Inertial Navigation System (SINS) integrated navigation system is simulated to evaluate the performance of the proposed algorithm, which contains two types of measurement attacks ranging from a constant bias attack to a time varying bias attack.