This paper presents a network-aware automated machine learning (AutoML) framework for detecting distributed denial of service (DDoS) attacks in software-defined sensor networks (SDSN). The proposed framework selects an ideal machine learning algorithm for detecting DDoS attacks in network-constrained environments, using metrics like variable traffic load, heterogeneous traffic rate, and detection time. The paper contributes by investigating the trade-off between the efficiency of ML algorithms and network/traffic state in DDoS detection, and by designing a software architecture with open-source network tools and multiple ML algorithms. The authors demonstrate that their framework ensures traffic packets are delivered within the network under DDoS attacks, albeit with additional delays.
Publication date: 20 Oct 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2310.12914
