WiFi CSI (Channel State Information) sensing is a technique that uses the fine-grained physical layer information extracted from WiFi signals — specifically amplitude and phase data across multiple subcarriers — to passively detect, classify, and monitor human activities, presence, and movement in indoor environments without requiring dedicated sensors or wearable devices. It matters to the field because it leverages ubiquitous WiFi infrastructure to enable non-intrusive, privacy-preserving applications such as occupancy detection, people counting, gesture recognition, fall detection, and activity recognition, making it a cost-effective alternative to camera-based or dedicated radar systems. Key variants include static versus dynamic environment sensing, single-link versus multi-link approaches, and learning-based pipelines ranging from traditional machine learning to deep learning architectures such as CNN+LSTM hybrids, with a growing emphasis on generalizability — ensuring models perform reliably across different environments, hardware platforms, and user demographics beyond controlled laboratory conditions.
Source Papers
- A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techniques, Datasets, and Future Research Prospects ↗ — A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techni
- CSI-Based People Counting in WiFi Networks: Leveraging Occupancy Detection ↗ — CSI-Based People Counting in WiFi Networks: Leveraging Occup
- CSI-Chain: A Complete End-to-End Framework for WiFi CSI Sensing ↗ — CSI-Chain: A Complete End-to-End Framework for WiFi CSI Sens
- Device-Free Wireless Sensing for Gesture Recognition Based on Complementary CSI Amplitude and Phase ↗ — Device-Free Wireless Sensing for Gesture Recognition Based o
- Human Activity Recognition via Wi-Fi and Inertial Sensors With Machine Learning ↗ — Human Activity Recognition via Wi-Fi and Inertial Sensors Wi
- WiFi CSI-based device-free sensing: from Fresnel zone model to CSI-ratio model ↗ — WiFi CSI-based device-free sensing: from Fresnel zone model