DDoS-Sandbox

The DDoS-Sandbox dataset consists of flow-level network traffic records generated in a controlled testbed environment. The dataset is created by combining : Our previous dataset (https://www.kaggle.com/datasets/wardac/applicationlayer-ddos-dataset) generated in the frame of INSPIRE-5Gplus project. This dataset contains Benign, Slowloris, and Hulk traffic. Benign and High-rate HTTP DDoS traffic (generated using GoldenEye tool) extracted from CICIDS-2018 dataset (https://www.unb.ca/cic/datasets/ids-2018.html). High and low-rate HTTP DDoS attack traffic generated in OULU's 5G testbed (See paper [1] for more details on this testbed) using GoldenEye(https://github.com/jseidl/GoldenEye.git), Hulken(https://github.com/hellgrenj/hulken.git), and Slowloris(https://github.com/gkbrk/slowloris.git) tools. [1] C. Benzaid, T. Taleb, A. Samiand O. Hireche, ‘A Deep Transfer Learning-Powered EDoS Detection Mechanism for 5G and Beyond Network Slicing’, presented at the 2023 IEEE Global Communications Conference (GLOBECOM 2023), Kuala Lumpur, Malaysia, Feb. 2024. doi: 10.1109/GLOBECOM54140.2023.10436891. (https://zenodo.org/records/10863938)