Device-free human sensing refers to the detection, localization, and recognition of human activities or presence using ambient wireless signals — most commonly WiFi channel state information (CSI) — without requiring the monitored individuals to carry or interact with any dedicated device. This approach is significant because it enables passive, unobtrusive monitoring for applications such as indoor localization, gesture recognition, fall detection, and activity classification, making it practical in real-world settings where wearable or handheld devices are unavailable or undesirable. Key variants in the field include model-driven approaches grounded in physical propagation models such as the Fresnel zone model and the CSI-ratio model, as well as data-driven deep learning approaches that leverage architectures such as CNNs, RNNs, and ResNets to learn discriminative features directly from raw CSI measurements.
Source Papers
- SenseFi: A library and benchmark on deep-learning-empowered WiFi human sensing ↗ — SenseFi: A library and benchmark on deep-learning-empowered
- 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
- Towards Environment Independent Device Free Human Activity Recognition ↗ — Towards Environment Independent Device Free Human Activity R
- WiFi CSI-based device-free sensing: from Fresnel zone model to CSI-ratio model ↗ — WiFi CSI-based device-free sensing: from Fresnel zone model
- WiGNN: WiFi-Based Cross-Domain Gesture Recognition Inspired by Dynamic Topology Structure ↗ — WiGNN: WiFi-Based Cross-Domain Gesture Recognition Inspired