WiFi sensing datasets are curated collections of Channel State Information (CSI) or other wireless signal measurements, captured under controlled or real-world conditions, that record how radio frequency signals are affected by human presence, movement, or activity in an environment. These datasets are foundational to the field because they enable the development, training, and benchmarking of machine learning models for tasks such as gesture recognition, fall detection, localization, and activity classification, while also serving as a critical resource for evaluating how well sensing systems generalize across different environments, hardware configurations, and user populations. Key variants include datasets distinguished by channel bandwidth (e.g., 20 MHz versus 80 MHz), sensing scenario (single-room versus multi-room), activity type, number of subjects, antenna configurations, and collection environment, with broader and more diverse datasets being particularly valued for advancing generalizability research.
Source Papers
- A CSI Dataset for Wireless Human Sensing on 80 MHz Wi-Fi Channels ↗ — A CSI Dataset for Wireless Human Sensing on 80 MHz Wi-Fi Cha
- A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techniques, Datasets, and Future Research Prospects ↗ — A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techni
- OPERAnet, a multimodal activity recognition dataset acquired from radio frequency and vision-based sensors ↗ — OPERAnet, a multimodal activity recognition dataset acquired