UL-ECE-DDoS-H-IoT-Datasets2025: MQTT and UDP DDoS Datasets for Healthcare IoT

Overview

UL-ECE-DDoS-H-IoT-Datasets2025 is a collection of two labeled datasets for studying DDoS detection in Healthcare-IoT (H-IoT) networks, released on Mendeley Data. The datasets, UL-ECE-MQTT-DDoS-H-IoT2025 and UL-ECE-UDP-DDoS-H-IoT2025, support research on DDoS detection and mitigation in 5G H-IoT.

Data Generation and Scenario

• H-IoT scenarios with medical sensor nodes (e.g., body temperature, oxygen saturation, heart rate) were simulated using Cooja and ns-3, communicating over MQTT and UDP.
• Both benign and malicious nodes were modeled, with differing data rates to emulate realistic DDoS traffic patterns.
• Raw simulator logs were processed and exported to CSV using Python scripts provided with the dataset.

Datasets and Size

• UL-ECE-MQTT-DDoS-H-IoT2025 contains on the order of twenty thousand processed, labeled records.
• UL-ECE-UDP-DDoS-H-IoT2025 contains nearly one hundred thousand records with a documented set of features.
• Each dataset includes raw traces, processed CSV files, and preprocessing scripts organized into MQTT and UDP folders.

Features and Labels

• Features include timestamp, payload size, message frequency, and node-level connection statistics.
• A binary outcome column marks each sample as normal (0) or attack (1).

Use Cases

• Evaluating DDoS detection and mitigation algorithms in Healthcare-IoT over MQTT and UDP.
• Benchmarking temporal convolutional networks (TCN) and alternative models using realistic H-IoT traffic.

Access and License

The datasets are hosted on Mendeley Data under DOI 10.17632/2bw34ght8b.2. A companion article details their structure and usage; licensing and reuse terms are provided on the Mendeley page.