The Signal Tendency Index (STI) is a statistical feature derived from Channel State Information (CSI) amplitude data that captures the directional trend or pattern of signal variation over time, quantifying how CSI measurements collectively shift in response to the presence and movement of people in an environment. It matters for device-free sensing because it provides a robust, aggregated representation of subtle signal perturbations caused by occupants, enabling more reliable occupancy detection and crowd counting without requiring individuals to carry any device. In practice, STI is computed across multiple CSI subcarrier streams to reduce noise and improve sensitivity, and its formulation may vary slightly between systems — for instance, in WiFree and FreeCount it is used as a key input feature to machine learning models for distinguishing different occupancy levels or crowd counts in indoor spaces.
Dictionary term