Distributed Optimal and Self-Tuning Filters Based on Compressed Data for Networked Stochastic Uncertain Systems with...

Area of Science:

• Control Systems Engineering
• Network Security
• Signal Processing

Background:

• Networked systems are vulnerable to malicious data attacks during sensor data exchange.
• Stochastic deception attacks pose a significant threat to the integrity of distributed estimation.

Purpose of the Study:

• To investigate distributed security estimation for networked stochastic uncertain systems facing stochastic deception attacks.
• To develop robust filtering algorithms that maintain estimation accuracy under attack conditions.

Main Methods:

• A weighted measurement fusion algorithm based on the least-squares method for known attack parameters.
• A correlation function method to identify unknown attack rates and noise variances.
• Development of a distributed optimal filter and a distributed self-tuning filter.

Main Results:

• The proposed distributed optimal filter achieves equivalent estimation accuracy with reduced computational cost compared to methods using uncompressed data.
• The distributed self-tuning filter effectively handles unknown attack parameters by identifying them first.
• Convergence analysis confirms the stability and effectiveness of the developed algorithms.

Conclusions:

• The study presents effective distributed filtering algorithms for networked systems under stochastic deception attacks.
• The developed methods provide a practical solution for enhancing security and reliability in sensor networks.
• Simulation results validate the superior performance and efficiency of the proposed approaches.