Event-Triggered Robust State Estimation for Nonlinear Networked Systems with Measurement Delays against DoS Attacks

Area of Science:

• Control Systems Engineering
• Networked Systems Security
• State Estimation Theory

Background:

• Nonlinear networked systems are vulnerable to denial-of-service (DoS) attacks and linearization errors from Extended Kalman Filter (EKF) computations.
• Constant measurement delays complicate traditional state estimation algorithms.
• DoS attacks disrupt measurement transmission, constraining communication rates and impacting estimator performance.

Purpose of the Study:

• To develop an event-triggered robust state estimation algorithm for nonlinear networked systems.
• To address challenges posed by denial-of-service (DoS) attacks and linearization errors.
• To overcome limitations imposed by constant measurement delays.

Main Methods:

• An event-triggered robust state estimation algorithm was derived using sensitivity penalization.
• An explicit packet arrival parameter was incorporated to manage communication constraints.
• A novel state augmentation method was developed to handle measurement delays.

Main Results:

• The proposed algorithm effectively defends against denial-of-service (DoS) attacks.
• Linearization errors inherent in the Extended Kalman Filter (EKF) were mitigated.
• Numerical simulations demonstrated a significant improvement in state estimation accuracy.

Conclusions:

• The developed event-triggered robust state estimator enhances accuracy in nonlinear networked systems.
• The approach successfully counters denial-of-service (DoS) attacks and linearization errors.
• The state augmentation method effectively handles measurement delays, improving overall system performance.