DL-SkLSTM approach for cyber security threats detection in 5G enabled IIoT

DL-SkLSTM approach for cyber security threats detection in 5G enabled IIoT

연구 분야: Verification

학회: International Journal of Information Technology

The advancement of 5G technology has enabled the IIoT (Industrial Internet of Things) to integrate artificial intelligence, cloud computing, and edge computing in real-time, leading to an improvement in industrial procedures in terms of efficiency. Despite the benefits of 5G technology for the IIoT, it also introduces new security risks and complexity to the control systems used in these ecosystems. Recent cyber-attacks are increasingly targeting vulnerable IoT devices, highlighting the need for enhanced security and privacy measures. To address this issue, this study proposes a 5G-based system that utilizes the DL-SkLSTM (Deep Learning- Stacked Long Short-Term Memory) based architecture to detect and classify the cyber-attack on a publicly accessible IIoT dataset, namely the Edge-IIoTset. SkLSTM is used to differentiate various cyberattacks. Finally, conducting a comprehensive analysis and comparison, we have identified that the proposed system outperforms several state-of-the-art DL and machine learning techniques.