Invisible Shield: Real-Time Industrial Threats Detection for IoT-based Systems

The Industrial Internet of Things' (IIoT) explosive growth has significantly changed industrial environments by linking smart devices via Supervisory Control and Data Acquisition (SCADA) systems. This integration also introduces significant cyber security vulnerabilities even though it brings about flexibility, resource efficiency, and operational agility. Current IDS using traditional machine learning fail to classify cyber attacks precisely because of complexity in data, limited availability, and mislabeling issues. To overcome these challenges, this study presents a scalable and effective ensemble detection framework that utilizes Pyramidal Recurrent Units (PRUs) and Decision Tree (DT) models. This framework aims to detect and counter cyber attacks across large IIoT networks, especially in SCADA-based environments. The propose of developing this project is to ensure scalable and efficient Deep Learning and Decision Tree based ensemble cyber attack detection framework to resolve trustworthiness issues in the SCADA based IIoT networks. Our proposed detection method can be applied to various IIoT domains. It is easy to implement and deploy, improving efficiency and accuracy while addressing the limitations of earlier approaches.This framework enhances the security of SCADA-based IIoT systems, making industrial networks more reliable, trustworthy, and resilient. The results indicates that the system performs well across different network settings, showcasing its adaptability and robustness in detecting IIoT-based SCADA systems.