ACI-IoT-2023: A Robust Dataset for Internet of Things Network Security Analysis

• Published in: MILCOM IEEE Military Communications Conference pp. 1 - 6
• Main Authors: Nack, Emily A., McKenzie, Morgan C., Bastian, Nathaniel D.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 28.10.2024
• Subjects: Computer security; Data collection; Dataset; Internet of Things; Network security; Reviews; Telecommunication traffic; Testing; Topology; Usability; Wireless networks
• ISSN: 2155-7586
• DOI: 10.1109/MILCOM61039.2024.10773916

The dynamic and evolving landscape of cybersecurity demands robust datasets for developing and evaluating Internet of Things (IoT) network security analytical solutions. This paper introduces the ACI-IoT-2023 dataset, designed to address the existing gaps and challenges in current IoT network security datasets. Informed by a comprehensive review of established datasets, ACI-IoT-2023 offers a rich, diverse, and realistic collection of network traffic data from a meticulously emulated home IoT environment. The methodology involves deploying a variety of physical IoT devices, capturing both wired and wireless network traffic, and simulating real-world scenarios through benign and malicious activities. Data collection was conducted over five days, covering reconnaissance, denial of service, brute force, spoofing attacks, and normal network traffic. Advanced tools and scripts were employed for data capture, feature extraction, cleaning, and labeling to ensure the dataset's accuracy and comprehensiveness. Additionally, we generated a complementary dataset with labeled packet capture files, enhancing the dataset's utility for artificial intelligence based IoT network security research. The ACI-IoT-2023 dataset stands as a significant contribution to the field, providing researchers and practitioners with a robust foundation for developing and testing next-generation IoT network security analytical solutions.