Security Interval Type-2 Fuzzy Sliding Mode Control under Multistrategy Injection Attack: Design, Analysis, and Optimization

Yekai Yang,Yugang Niu,James Lam

DOI: https://doi.org/10.1109/tfuzz.2023.3239930

IF: 12.253

2023-01-01

IEEE Transactions on Fuzzy Systems

Abstract:This article considers the security sliding mode control problem for interval type-2 (IT2) fuzzy systems, where the sensor-controller channel may be subject to the false data injection (FDI) attack with a multistrategy attack model. The different attack patterns at each instant are described via a random categorical distribution. It is noted that the injected false data may result in the system state to exceed the modeling domain. In order to handle the above phenomenon, a reconstruction method of the controller's membership functions (MFs) is proposed via analyzing the variation range of the attacked state, and then, a feasible IT2 fuzzy sliding mode security controller is designed. The explicit relation between the raw state signal and attacked state is established, based on which stability conditions dependent on the regional boundaries of MFs are obtained for the input-to-state stability in probability of the closed-loop system. Meanwhile, a framework of security performance optimization is established via searching a desirable sliding matrix, which may be solved via the particle swarm optimization algorithm. Furthermore, the proposed security control scheme is extended to imperfectly matched MFs case, where the controller does not demand the same fuzzy rules as the controlled fuzzy system. Finally, the simulation verification is carried out via numerical and practical examples.