Real time dataset generation framework for intrusion detection systems in IoT

Abstract

The Internet of Things (IoT) has evolved in the last few years to become one of the hottest topics in the area of computer science research. This drastic increase in IoT applications across different disciplines, such as in health-care and smart industries, comes with a considerable security risk. This is not limited only to attacks on privacy; it can also extend to attacks on network availability and performance. Therefore, an intrusion detection system is essential to act as the first line of defense for the network. IDS systems and algorithms depend heavily on the quality of the dataset provided. Sadly, there has been a lack of work in evaluating and collecting intrusion detection system related datasets that are designed specifically for an IoT ecosystem. Most of the studies published focus on outdated and non-compatible datasets such as the KDD98 dataset. Therefore, in this paper, we aim to investigate the existing datasets and their applications for IoT environments. Then we introduce a real-time data collection framework for building a dataset for intrusion detection system evaluation and testing. The main advantages of the proposed dataset are that it contains features that are explicitly designed for the 6LoWPAN/RPL network, the most widely used protocol in the IoT environment.