Densely Connected Neural Networks for Detecting Denial of Service Attacks on Smart Grid Network
Tala Talaei Khoei, Naima Kaabouch
Abstract
Smart grid has several benefits, including efficiency and reliability. However, this network is prone to several cyber-attacks and has limited security. One of the main damaging attacks targeting smart grid is denial of service attacks. These attacks can disrupt the normal process of a network by limiting the network availability. Over the last decades, a number of studies have developed techniques to detect denial of service attacks on smart grid. However, the majority of these techniques suffer from high false alarm and misdetection rates. For this purpose, this paper proposes convolutional neural network models, densely connected neural networks with 121, 169, 201, and 264 layers, for detecting and classifying denial of service attacks on smart grid. The performance is done using eight evaluation metrics. The results show that the proposed model with 264 layers outperforms the other models in terms of the considered metrics.