Multilayer Perceptron-Based Defense Mechanisms for Securing Industrial IoT in Industry 4.0 Environments

Abstract

In the Industry 4.0 era, the Industrial Internet of Things (IIoT) has transformed manufacturing by facilitating seamless connectivity and real-time data exchange between physical devices and systems. This transformation has bolstered efficiency, productivity, and decision-making in industrial settings. However, the increased connectivity also brings heightened cybersecurity risks. Securing the IIoT environment is critical to safeguard critical infrastructure, data, and operations against cyber threats. As IIoT adoption expands across sectors, ensuring system security and resilience becomes imperative to maintain operational continuity and preserve trust. This paper proposes a deep learning-based approach, leveraging the CIDDS, BOT-IoT, and Edge_IIoTset datasets, to fortify IIoT and manufacturing systems against cyber threats. Multilayer Perceptron (MLP) is identified as the top-performing model, achieving an accuracy of 99.26%, precision of 98.74%, and recall of 98.86% on the CIDDS dataset. Similar superior performance was observed on the BOT-IoT (99.52%, 99.52%, and 99.99%) and Edge_IIoTset (99.93%, 99.93%, and 99.99%) datasets, making MLP a robust solution for anomaly detection in industrial IoT environments.