Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Unified Attack Detection Strategy for Multi-Agent Systems over Transient and Steady Stages

Jinming Gao, Yijing Wang, Wentao Zhang, Rui Zhao, Yang Shi, Zhiqiang Zuo

TL;DR

This work tackles the challenge of detecting cyber-physical attacks in multi-agent systems across multiple layers and stages. It advances three complementary detectors: a watermarking-based scheme with KL divergence for the communication layer, an envelope-based detector for the agent layer, and a trusted two-hop framework for hybrid attacks, ensuring effectiveness in both transient and steady states. The approach yields theoretical detection conditions and demonstrates practical viability through platooning simulations, showing timely attack identification with modest graph and transmission requirements. Collectively, the methods offer a unified, scalable path toward resilient MAS operation in dynamic environments.

Abstract

This paper proposes a unified detection strategy against three kinds of attacks for multi-agent systems (MASs) which is applicable to both transient and steady stages. For attacks on the communication layer, a watermarking-based detection scheme with KullbackLeibler (KL) divergence is designed. Different from traditional communication schemes, each agent transmits a message set containing two state values with different types of watermarking. It is found that the detection performance is determined by the relevant parameters of the watermarking signal. Unlike the existing detection manoeuvres, such a scheme is capable of transient and steady stages. For attacks on the agent layer, a convergence rate related detection approach is put forward. It is shown that the resilience of the considered system is characterized by the coefficient and offset of the envelope. For hybrid attacks, based on the above detection mechanisms, a general framework resorting to trusted agents is presented, which requires weaker graph conditions and less information transmission. Finally, an example associated with the platooning of connected vehicles is given to support the theoretical results.

A Unified Attack Detection Strategy for Multi-Agent Systems over Transient and Steady Stages

TL;DR

This work tackles the challenge of detecting cyber-physical attacks in multi-agent systems across multiple layers and stages. It advances three complementary detectors: a watermarking-based scheme with KL divergence for the communication layer, an envelope-based detector for the agent layer, and a trusted two-hop framework for hybrid attacks, ensuring effectiveness in both transient and steady states. The approach yields theoretical detection conditions and demonstrates practical viability through platooning simulations, showing timely attack identification with modest graph and transmission requirements. Collectively, the methods offer a unified, scalable path toward resilient MAS operation in dynamic environments.

Abstract

This paper proposes a unified detection strategy against three kinds of attacks for multi-agent systems (MASs) which is applicable to both transient and steady stages. For attacks on the communication layer, a watermarking-based detection scheme with KullbackLeibler (KL) divergence is designed. Different from traditional communication schemes, each agent transmits a message set containing two state values with different types of watermarking. It is found that the detection performance is determined by the relevant parameters of the watermarking signal. Unlike the existing detection manoeuvres, such a scheme is capable of transient and steady stages. For attacks on the agent layer, a convergence rate related detection approach is put forward. It is shown that the resilience of the considered system is characterized by the coefficient and offset of the envelope. For hybrid attacks, based on the above detection mechanisms, a general framework resorting to trusted agents is presented, which requires weaker graph conditions and less information transmission. Finally, an example associated with the platooning of connected vehicles is given to support the theoretical results.
Paper Structure (10 sections, 7 theorems, 45 equations, 10 figures, 2 tables, 3 algorithms)

This paper contains 10 sections, 7 theorems, 45 equations, 10 figures, 2 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. Notations and Graph Theory
  4. System Description
  5. Main Results
  6. Detection Strategy for Attacks on the Communication Layer
  7. Detection Strategy for Attacks on the Agent layer
  8. Detection Strategy for Hybrid Attacks
  9. Platooning Simulations
  10. Conclusion

Key Result

Proposition 1

For the attacked system $(GOeq1)$, $\mathbb{E}(\widetilde{y}^{a}_{ijl}(k)) \in \left[ \epsilon _1,\epsilon _2 \right]$ if and only if

Figures (10)

  • Figure 1: KL divergences of detection strategy in mustafa2020resilient.
  • Figure 2: Block diagram of Algorithm\ref{['alg1']} for attacks on the communication layer.
  • Figure 3: Block diagram of Algorithm\ref{['alg2']} for attacks on the agent layer.
  • Figure 4: Block diagram of Algorithm\ref{['alg3']} for hybrid attacks.
  • Figure 5: Information flow topology for platoons.
  • ...and 5 more figures

Theorems & Definitions (25)

  • Remark 1
  • Proposition 1
  • proof
  • Definition 1
  • Proposition 2
  • proof
  • Theorem 1
  • proof
  • Remark 2
  • Remark 3
  • ...and 15 more