Local Outlier Factor (LOF) is an unsupervised anomaly detection algorithm that assigns each data point a score reflecting the degree to which it is an outlier relative to its local neighborhood, computed by comparing the local reachability density of a point to that of its neighbors. In WiFi CSI sensing, LOF is applied to distinguish between normal environmental states and motion-induced signal anomalies, enabling device-free detection of human presence or activity without requiring labeled training data. Its ability to operate locally rather than globally makes it particularly suited to CSI-based sensing environments where signal distributions vary across space and time, and it is sometimes used alongside density-based variants or combined with other statistical features to improve robustness in cluttered indoor settings.
Dictionary term