Observer-Based Security Control for Interconnected Semi-Markovian Jump Systems With Unknown Transition Probabilities

Area of Science:

• Control Engineering
• Cybersecurity
• Systems Theory

Background:

• Interconnected semi-Markovian jump systems face security challenges due to unknown transition probabilities and network constraints.
• Limited communication bandwidth necessitates efficient data transmission strategies.

Purpose of the Study:

• To develop an observer-based security control strategy for interconnected semi-Markovian jump systems.
• To address unknown transition probabilities and network-induced limitations.
• To enhance resilience against denial-of-service (DoS) and false-data injection (FDI) attacks.

Main Methods:

• An adaptive event-triggered mechanism (AETM) was designed to dynamically adjust triggering thresholds, reducing network load.
• Two Bernoulli distributed variables modeled the impact of DoS and FDI attacks.
• Sufficient criteria for stochastic stability with H∞ attenuation were derived.
• Linear matrix inequalities (LMIs) were employed for simultaneous observer and controller gain matrix computation.

Main Results:

• The proposed AETM effectively reduces network burden compared to conventional event-triggered mechanisms (ETM).
• The developed control strategy ensures stochastic stability and H∞ performance for the augmented systems.
• The method successfully accounts for unknown transition probabilities and cyberattacks.
• Observer and controller gains were computed efficiently using LMIs.

Conclusions:

• The theoretical results demonstrate the effectiveness of the proposed observer-based security control strategy.
• The adaptive event-triggered mechanism offers a practical solution for secure and efficient control of complex networked systems.
• The study provides a robust framework for handling uncertainties and attacks in semi-Markovian jump systems.