ML-assisted Security for Anomaly Detection in Industrial IoT (IIoT) Applications
Bharath Konatham, Tabassum Simra, Ashutosh Ghimire, Fathi Amsaad, Mohamed I. Ibrahem, Noor Zaman Jhanjhi · 2023
The industrial Internet of Things (IIoT) uses intelligent sensors and actuators, etc., to facilitate the application of IoT in different industries. This paper presents a comparative analysis using several machine learning (ML)-assisted deep learning techniques to develop efficient cybersecurity anomaly detection within IoT applications. To accomplish that, several deep learning techniques are investigated and analyzed, including Convolutional Neural Networks (CNN), Gated Recurrent Units (GRU), their hybrid (CNN+GRU), and Long Short-Term Memory (LSTM). The performance of the developed deep learning models is evaluated in terms of accuracy, precision, recall, F1-score, and False Alarm Rate (FAR). Our experimental results show that the developed hybrid CNN+GRU model outperforms the others, achieving an accuracy of 94.94%, a recall of 92.29%, a precision of 98.49%, an F1-score of 95.24%, and a low false alarm rate of 0.001. However, it is essential to note that the hybrid model requires a longer convergence time, indicating a trade-off between performance and computational efficiency. Notably, individual CNN and GRU models also showcase strong performance as time-efficient alternatives. In conclusion, our adopted comprehensive dataset and rigorous evaluation prove that we have developed practical deep-learning approaches to obtain an accurate measure for an efficient IIoT anomaly detection framework.