IoT Network Traffic Dataset for User and Device Activity Classification

The dataset provides a comprehensive collection of IoT network traffic captured from multiple consumer IoT devices. This project is designed to support research into IoT activity inference, network traffic classification and cybersecurity/forensic analysis. 

The dataset consists of both raw packet-level data and processed, windowed datasets. The raw dataset contains just over 1.4 million packet headers extracted using tshark, ensuring that key network-level features are preserved. In addition, several processed windowed datasets are available, these variations allow for flexibility for different machine learning approaches. 

Data was recorded in a real home network using 5 IoT devices, an Amazon Alexa, Tapo C200 Camera, Tapo L530B Smart Bulb, Tapo P100 Smart Plug, Tapo T100 Motion sensor paired with the Tapo H100 Smart Hub. 

Traffic capture was performed by running tcpdump on a Raspberry Pi 5 that was port mirroring the DrayTek Vigor902 Access Point ensuring complete and accurate collection of all packets from IoT devices. 

The dataset is particularly useful for: 

Developing models for IoT device activity recognition

Creating dashboards to infer user behaviour from the IoT device activities

Investigating privacy leakage through network metadata 

Building and evaluating intrusion detection systems for IoT devices

All data has been processed to exclude sensitive or personally identifiable information. The dataset is released under the Creative Commons Attribution 4.0 (CC BY 4.0) license.