Anomaly detection in WiFi/CSI sensing refers to the task of identifying unusual or unexpected patterns in wireless channel measurements that deviate significantly from normal behavior, such as detecting the presence of unauthorized individuals, abnormal crowd dynamics, or atypical motion events in a monitored environment. It matters to the field because it enables passive, infrastructure-light security and safety monitoring without requiring cameras or wearable sensors, making it well-suited for privacy-sensitive or resource-constrained deployments. Key variants include occupancy-based anomaly detection, where deviations from expected crowd counts or densities trigger alerts, and activity-based anomaly detection, where irregular motion signatures in CSI data indicate events such as falls, intrusions, or sudden crowd surges.
Source Papers
- A survey of recent advances in CNN-based single image crowd counting and density estimation ↗ — A survey of recent advances in CNN-based single image crowd
- Fast and Robust Stationary Crowd Counting With Commodity WiFi ↗ — Fast and Robust Stationary Crowd Counting With Commodity WiF
- Recent trends in crowd analysis: A review ↗ — Recent trends in crowd analysis: A review