Security Sliding Mode Control of Interval Type-2 Fuzzy Systems Subject to Cyber Attacks: The Stochastic Communication Protocol Case

• Published in: IEEE transactions on fuzzy systems Vol. 29; no. 2; pp. 240 - 251
• Main Authors: Zhang, Zhina, Niu, Yugang, Cao, Zhiru, Song, Jun
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.02.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Communication; Control systems; Control theory; Cyber attack; Cyberattack; Cybersecurity; Fuzzy control; Fuzzy systems; interval type-2 fuzzy system; Mathematical model; Protocols; Scheduling; Sensors; Sliding mode control; stochastic communication protocol; Stochastic processes; Symmetric matrices
• ISSN: 1063-6706; 1941-0034
• DOI: 10.1109/TFUZZ.2020.2972785

This article addresses the security control problem of a class of interval type-2 fuzzy systems via the sliding mode control strategy. A stochastic communication scheduling protocol is utilized to govern the transmission from the sensors to the controller, by which only one sensor node has the chance to transmit its value at every instant. Meanwhile, cyber attacks from malicious adversaries might be launched in vulnerable communication channels. To quantitatively analyze the effect of the stochastic communication protocol and cyber attacks, their mathematical model is first constructed based on a compensation scheme. Since the scheduling signal may be unavailable once cyber attacks are activated, a desirable sliding mode control law is synthesized with token-independent control gains, whose membership functions are mismatched with those of the fuzzy system. To deal with these mismatched membership functions, the relations between the membership functions of the system and the control law are reconstructed. Consequently, the favorable property of perfectly matched memberships could be employed. Sufficient conditions are derived so that the resultant closed-loop interval type-2 fuzzy system is stochastically stable and, at the same time, the state trajectories can be forced into a small domain around the prescribed sliding surface. The proposed control design approach is verified by two examples.