Simultaneous fault detection and consensus for multi-agent systems under false data injection attacks

Meilin Li

TL;DR: This paper addresses the problem of simultaneous fault detection and consensus under false data injection attacks.

Abstract:

In the realm of multiagent systems (MAS), main-
taining system integrity and achieving consensus in the presence
of malicious activities is a critical challenge. This paper addresses
the problem of simultaneous fault detection and consensus under
false data injection (FDI) attacks. FDI attacks, where adversaries
strategically alter the data exchanged among agents, can lead
to severe disruptions in the system’s functionality, potentially
causing erroneous consensus or total system failure. We propose
a robust detection and consensus mechanism that integrates
advanced fault detection techniques with resilient consensus
algorithms specifically designed to withstand FDI attacks. By
leveraging the inherent structure of multiagent systems and
employing a decentralized approach, our method ensures that
agents can not only detect and isolate compromised nodes but
also maintain accurate consensus across the network. Simulation
results validate the effectiveness of the proposed strategy, demon-
strating its capability to mitigate the impact of FDI attacks and
sustain system performance.

Submission Number: 213