Distributed State Estimation for Mixed Delays System Over Sensor Networks With Multichannel Random Attacks and Markov...

Area of Science:

• Control Systems Engineering
• Network Security
• Stochastic Systems

Background:

• Distributed state estimation is crucial for networked systems.
• Mixed delays and unknown attacks pose significant challenges to system stability and performance.
• Existing attack models often fail to capture the complexity of real-world network intrusions.

Purpose of the Study:

• To develop a novel multichannel random attack model for distributed systems.
• To design a robust distributed state estimator capable of handling dynamic network topologies and complex attacks.
• To ensure the asymptotic mean-square stability of the estimation error system under H∞ disturbance rejection.

Main Methods:

• A new multichannel random attack model considering simultaneous packet modification across multiple channels.
• Dynamic network topology modeled using a Markov chain.
• Lyapunov theory and stochastic analysis for stability proofs.
• Linearization method for solving estimator parameter matrices.

Main Results:

• The proposed distributed state estimator is proven to be asymptotically mean-square stable.
• The system achieves a guaranteed H∞ antidisturbance index, demonstrating robustness against attacks.
• Simulation examples validate the effectiveness of the designed algorithm.

Conclusions:

• The novel multichannel random attack model accurately represents complex network threats.
• The developed distributed state estimation algorithm effectively handles mixed delays, dynamic topologies, and sophisticated attacks.
• The approach provides a robust solution for secure and reliable state estimation in networked systems.