IoT-DH Dataset

Published: 7 August 2026| Version 3 | DOI: 10.17632/8dns3xbckv.3
Contributors:
syaifuddin saif,
,

Description

The IoT-DH (IoT DDoS Honeypot) Dataset is a comprehensive and original collection of data designed to enhance the understanding and detection of Distributed Denial of Service (DDoS) attacks targeting Internet of Things (IoT) devices. The collected data comprises information obtained through an IoT honeypot—a system intentionally designed to attract cyberattacks and unveil the strategies, methods, and procedures employed by malicious actors to compromise and overwhelm devices with traffic. To ensure high authenticity and avoid replication of existing data, this dataset was originally generated from public network traffic. The main goal is to assist in monitoring attack traffic, identifying potential threats, and analyzing assailant techniques. The raw network traffic (PCAP) was accurately parsed using the Argus Framework and converted into a raw CSV format for convenient data access. Furthermore, to guarantee precision, the dataset's labeling process was highly automated using a Variational Autoencoder (VAE)-based anomaly detection method.

Files

Institutions

Categories

Computer Science, Engineering, Computer Vision, Machine Learning, Internet of Things, Informatics, Deep Learning

Licence