WiFi CSI Dataset: Channel State Information Captures from 4 Subjects and 3 Receivers

A structured dataset of WiFi Channel State Information (CSI) captures recorded as text logs. The data includes repeated trials for 4 class labels and 4 subjects, recorded simultaneously by 3 receiver devices. The dataset was authored by ilyakolosov and last updated on 2026-05 01:34:15.

Use Cases

• Human activity or gesture recognition based on WiFi CSI signal patterns.
• Developing device-free sensing models based on multi-receiver signal data.
• Benchmarking signal processing algorithms for Channel State Information.
• Studying the variability of wireless signals across different subjects.

Strengths

• Data is structured with repeated trials for 4 distinct class labels.
• Includes data from 4 different subjects and 3 separate receiver devices.
• Dataset is described as 'Hugging Face–ready' for ease of use.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file size are unknown, which may limit suitability assessment.

Provenance

Source

huggingface

Collection Method

Recorded WiFi CSI captures from receiver devices.

Time Range

null

Freshness

Last updated 2026-05-01 01:34:15; freshness should be verified.

Geography

null