Application of Synthetic Data Generation Methods to the Detection of Network Attacks on Internet of Things Devices
T. D. Ovasapyan, V. D. Danilov, D. A. Moskvin · Automatic Control and Computer Sciences · 2021
Abstract— The application of synthetic data generation methods to augment the dataset and subsequently improve the accuracy of detecting network attacks on Internet of Things (IoT) devices using machine learning methods is considered. Augmentation and generative adversarial network methods are used to augment the initial dataset. The IOT Network Intrusion Dataset, which includes network traffic from different attacks (DoS, MiTM, and scan) as well as attacks by one of the most common botnets in the Internet of Things, Mirai, is used as the initial dataset.