A Fog-Based Attack Detection Model Using Deep Learning for the Internet of Medical Things

Internet of Medical Things (IoMT) applications have advanced and become more widespread in recent years. This has driven the need to secure IoMT networks. As a second line of defense, effective security techniques, including deep learning approaches to intrusion detection systems (IDS) have been applied to IoMT to detect network attacks. However, most existing solutions are either cloud-based or are challenging to implement on IoMT devices. This delays attack detection. Furthermore, these detections are centralized and therefore incompatible with the IoMT environment. In addition, fog computing has emerged recently as a new field that complements cloud computing due to its improved location awareness, mobility, scalability, heterogeneity, low latency, and geographical distribution. This work proposes a deep learning-based IDS for early attack detection in IoMT fog. This research is carried out using deep learning approaches based on Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN) and Gated Recurrent Unit (GRU). The IoT-Healthcare Security dataset is used to evaluate the proposed model. The CNN model achieves 99.62% accuracy, reduced detection time and low memory consumption. Compared to existing approaches, the proposed technique showed the highest accuracy.