Observer-Based Adaptive Neural Sliding Mode Control of Fuzzy Systems With Sojourn-Probability-Based Multimode Attacks

Area of Science:

• Control Systems Engineering
• Cybersecurity
• Nonlinear Dynamics

Background:

• Observer-based sliding mode control (SMC) is crucial for fuzzy nonlinear systems.
• Multimode denial of service (DoS) and deception attacks pose significant threats to system stability.
• Existing attack models often fail to capture the dynamic and stochastic nature of complex cyberattacks.

Purpose of the Study:

• To develop a novel multimode DoS attack model with time-varying sojourn probabilities.
• To model deception attacks using unbounded nonlinear functions and adaptive neural networks (NNs).
• To design a robust observer-based SMC strategy for fuzzy nonlinear systems under cyberattacks.

Main Methods:

• Introduced a novel multimode DoS attack model incorporating time-varying sojourn probabilities.
• Modeled deception attacks as unbounded nonlinear functions approximated by adaptive NNs.
• Designed a fuzzy sliding surface and an observer-based SMC law ensuring mean square estimation upper bound.

Main Results:

• The proposed DoS model accurately represents stochastic attack behaviors, overcoming Markov-based model limitations.
• Adaptive NNs effectively mitigated the impact of deception attacks on system stability.
• The developed observer-based SMC law guaranteed the mean square estimation upper bound for fuzzy nonlinear systems.

Conclusions:

• The proposed control strategy effectively ensures the stability of fuzzy nonlinear systems against multimode DoS and deception attacks.
• The novel attack modeling and adaptive NN approach offer a significant advancement in robust control under cyber threats.
• Validation through a tunnel diode circuit model confirms the strategy's superiority and effectiveness.