Device Movement Detection Using <scp>WiFi</scp> Channel State Information

Abstract For IoT device management and operation, it is useful to automatically detect when a device is physically moved. In this letter, we propose a simple method to detect WiFi device movements by applying unsupervised anomaly detection algorithms to WiFi Channel State Information collected at WiFi access points. Our proposed method combines multiple anomaly detection algorithms to detect various types of movements. The evaluation results show that the proposed method can detect device movement with 89% accuracy, which is higher than when using RSSI or a single anomaly detection algorithm. © 2025 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC.