A hybrid blockchain and machine learning approach for intrusion detection system in Industrial Internet of Things
Song Wu, Xiangyuan Zhu, Sheng Ren, Wenxue Tan, Yibo Peng
Abstract
The Industrial Internet of Things (IIoT) is a key component of Industry 4.0, which enables manufacturing to be automated and data collected in real-time. Edge IoT devices are subject to cybersecurity threats and unauthorised access. Decentralisation and resource limitations of IIoT often prevent traditional security mechanisms from addressing these threats. Intrusion detection systems (IDSs), which are used to detect intrusions in IIoT environments, are presented in this paper as hybrid machine learning-blockchain approaches. Blockchain technology ensures data integrity, secures communication, and prevents unauthorised modifications through the proposed system. To reduce false positives and improve threat detection accuracy, XGBoost is able to reduce the number of false positives. Using the BOT-IoT dataset, the model is demonstrated to be superior to conventional intrusion detection systems. This approach ensures enhanced security and trustworthiness of IIoT networks by offering a scalable, efficient, and secure solution.