Description

EHUCOUNT is a public WiFi CSI dataset for people-counting / occupancy estimation released by the University of the Basque Country (EHU). It is one of the few publicly available CSI counting datasets and is therefore used together with the RTV dataset as a cross-dataset robustness test for Wi-Fi crowd-counting methods.

Modality / size

  • Modality: WiFi CSI on commodity NICs.
  • Subjects / scenarios: indoor environments with varying numbers of occupants.
  • Labels: occupant count per CSI window.

Used by (papers)

  • Used as a public CSI people-counting benchmark in vault papers, paired with RTV and authors' own captures.
  • Statistical separability between classes is reported as larger on EHUCOUNT and RTV than on smaller in-house captures.

Notes

  • Public dataset; obtain from the EHU release.
  • Directly relevant to the thesis crowd-modelling angle.

2 vault papers evaluate on this dataset

Titles and DOIs only — no abstracts, no analyses.

  • Channel State Information (CSI) Amplitude Coloring Scheme for Enhancing Accuracy of an Indoor Occupancy Detection System Using Wi-Fi Sensing 2024 DOI ↗
  • Human Sensing by Using Radio Frequency Signals: A Survey on Occupancy and Activity Detection 2023 DOI ↗