SHARP (Sensing with High-resolution and Accurate Radio Propagation) is a Wi-Fi CSI-based human sensing framework and associated dataset designed to leverage wide 80 MHz IEEE 802.11ac channels, providing high-resolution channel state information for wireless human sensing tasks such as activity recognition and localization. Its significance lies in offering the research community a richly detailed, publicly accessible benchmark that captures fine-grained multipath propagation characteristics across a large subcarrier space, enabling more accurate and generalizable machine learning models for CSI-based sensing. Key variants include its use as both a standalone sensing methodology and as a reference dataset against which feature extraction pipelines, such as fast Monte Carlo approaches, are evaluated and validated for human activity recognition performance.
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
- Efficient machine learning for Wi-Fi CSI-based human activity recognition using fast Monte Carlo based feature extraction ↗ — Efficient machine learning for Wi-Fi CSI-based human activit
- Time matters: Empirical insights into the limits and challenges of temporal generalization in CSI-based Wi-Fi sensing ↗ — Time matters: Empirical insights into the limits and challen