Real-Time Detection of Injection Attacks in Industrial Multi-agent Systems

This paper proposes an industrial network data injection detection method using predictive modeling based on the Markov state transition matrix. Focusing on industrial networks within the Industrial Internet, we employ a Markov state transition matrix to establish state prediction models. By using this method promptly identify potential false data injection and data tampering. This method ensures real-time identification of data injection attacks with low computational overhead, thereby enhancing the security and robustness of industrial network systems. Experimental results demonstrate that the proposed approach achieves high detection accuracy and real-time performance in various data injection attack scenarios.