Double-Layer Dynamic Switching Observer for Secure Multi-Agent State Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing secure state reconstruction methods for multi-agent systems are computationally expensive and struggle with time-varying attacks, particularly in decentralized systems, and are limited in their ability to identify multiple simultaneous attacks, making them vulnerable to network attacks that can disrupt system operations and personal safety.
Innovation Solution
A distributed secure state reconstruction method using a double-layer dynamic switching observer is introduced, which constructs a specific dynamics model of the multi-agent system, builds a double-layer observer with a residual generator, and dynamically switches communication topology based on residual signal thresholds to identify and mitigate malicious attacks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a centralized sparse attack identification method is used, then accurate attack identification can be achieved, but computational complexity becomes too high for large-scale systems
Solution Approach 1:
The centralized attack identification problem is segmented into distributed local estimation problems. Each agent maintains its own state estimator and only exchanges limited information with neighbors, transforming the high-dimensional centralized optimization into multiple low-dimensional local estimations that can be parallelized.
Solution Approach 2:
The problem is transformed from the measurement space to the state space by constructing a dynamic observer. Instead of directly identifying attacks from measurement residuals in the original space, the system evolves the state estimate dynamically, allowing attack identification to emerge from the state reconstruction process rather than direct measurement analysis.
2Device complexity
If a simple decentralized state estimation method is used, then computational cost is reduced, but the system becomes vulnerable to malicious attacks
Solution Approach 1:
A dynamic observer is introduced as an intermediary between the measurements and the state estimation. This observer filters out the effects of sparse attacks by dynamically adjusting the state reconstruction based on residual analysis, allowing the simple decentralized structure to achieve robustness against attacks without requiring complex centralized coordination.
3Measurement precision
If an observer collects measurement data within a certain duration to construct optimization problem, then secure state estimation can be achieved, but estimation delay increases for time-varying attacks
Solution Approach 1:
The static optimization approach is replaced with a dynamic observer that continuously updates state estimates in real-time. The observer dynamics allow the system to track time-varying attacks by adapting the state reconstruction continuously, eliminating the need for batch data collection and reducing estimation delay while maintaining accuracy.
4Device complexity
If a decentralized observer is used with residual threshold for attack positioning, then distributed computation is achieved, but only single attack identification is possible in neighbor subsystem
Solution Approach 1:
The dynamic observer incorporates continuous feedback from residual signals to adjust the state estimation and attack identification. By monitoring the evolution of residuals over time and using feedback mechanisms, the system can distinguish between multiple simultaneous attacks and identify each one separately, overcoming the limitation of single-attack detection in decentralized structures.
Data Source
AI summary
The present disclosure discloses a distributed secure state reconstruction method based on a double-layer dynamic switching observer. The method includes the following steps: constructing a dynamics model of a sensing channel of a multi-agent system after the sensing channel is attacked according to the multi-agent system; building a double-layer observer for each multi-agent in combination with a specific multi-agent system model, constructing a proper observation communication topology, and designing a corresponding residual generator; analyzing dynamic information generating a residual threshold aiming at an observation model, checking a magnitude between each residual signal and the threshold, dynamically switching the communication topology between the observers according to the compared magnitude, and performing a new data communication interaction; and performing iterative updating to generate new observation data in combination with self observation data and received neighbor observation information, and taking whether the residual signal is greater than a current threshold or not as a standard for determining whether a corresponding communication channel is attacked or not. According to the present disclosure, all transmission channels subjected to malicious attacks can he correctly identified, and the real state of the system can be securely reconstructed.