Description

XRF55 is a large-scale RF activity dataset that bundles WiFi CSI together with mmWave radar and RFID measurements across 55 indoor human activity classes. It targets RF-based human action recognition at a class scale that is significantly larger than the dozen-or-so classes in earlier CSI datasets, and is intended as a more challenging benchmark for RF foundation-model and self-supervised studies.

Modality / size

  • Modality: WiFi CSI + mmWave radar + RFID (multi-modal RF).
  • Subjects / scenarios: multiple participants, indoor environments.
  • Labels: 55 activity classes.

Used by (papers)

  • Cited as a benchmark for large-class RF HAR and for multi-modal RF fusion.

Notes

  • Public dataset accompanying the XRF55 paper.

2 vault papers evaluate on this dataset

Titles and DOIs only — no abstracts, no analyses.

  • A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techniques, Datasets, and Future Research Prospects 2026 DOI ↗
  • SDP: A Unified Protocol and Benchmarking Framework for Reproducible Wireless Sensing 2026 DOI ↗