We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
An efficient centralized DDoS attack detection approach for Software Defined Internet of Things.
- Authors
Chauhan, Pinkey; Atulkar, Mithilesh
- Abstract
Both software defined networks and the Internet of Things are new topics that are being heavily employed in the information technology industry and academia. Due to their fame, they have attracted many attacks in the SD-IoT (Software Defined Internet of Things). The attackers' purpose could be to steal or block users' information, as well as to waste network resources on pointless operations, causing the resources to deny legitimate user requests. Distributed denial of service (DDoS) attack is one such attack among those. This paper presents a study on feature extraction for detecting DDoS attack on SD-IoT. Using the Distributed Internet Traffic Generator and hping3 tools, normal and DDoS attack traffic were created, and six features were retrieved from the generated traffic. Some well-known classifiers, such as Random Forest, Light Gradient Boosting Machine (LGBM), Support Vector Machine, and K Nearest Neighbor, are trained and tested with these 6 vector tuples of normal and DDoS attack data to check the efficiency of the extracted features. Since LGBM has been shown to be the most effective classifier across a wide range of metrics, including accuracy, precision, recall, F1, Testing Time, False Alarm Rate, Cohen's Kappa Coefficient, and AUC value, it has been compared with the performance of some other state-of-the-art works. It is found that LGBM with these extracted 6 features is outperforming those works except in one case where the accuracy of proposed work is less than accuracy of that work. So, finally LGBM is deployed in the SDN controller so that it can identify attacks in real-time traffic.
- Subjects
DENIAL of service attacks; INTERNET of things; SOFTWARE-defined networking; INFORMATION technology industry; INTERNET traffic
- Publication
Journal of Supercomputing, 2023, Vol 79, Issue 9, p10386
- ISSN
0920-8542
- Publication type
Article
- DOI
10.1007/s11227-023-05072-y