Real-Time Man in the Middle Attack Detection and Mitigation in Cyber-Physical Systems Using Neural Networks and Remote Monitoring

This article presents a novel method for detecting and mitigating Man-in-the-Middle (MIM) attacks in cyber-physical systems (CPS) using neural networks. MIM attacks are particularly dangerous because attackers can intercept and manipulate communications, gaining control over secure data. Existing detection tools often require high computational resources and are costly. Our approach focuses on a DC-DC converter, a crucial CPS component, where the wireless communication controller is an Arduino Nano IoT, and real-time monitoring is enabled via a LabVIEW GUI on a remote PC. We developed a neural network-based detection system to identify abnormal behavior indicative of MIM attacks. The results demonstrate its effectiveness in maintaining system integrity and reliability. Additionally, we explore a remote monitoring method as an alternative to the LabVIEW GUI, enhancing flexibility and scalability. The control algorithm is developed in MATLAB 2022a, with real-time hardware implementation validating the proposed method.