Min-max normalization is a data preprocessing technique that rescales raw CSI amplitude values to a fixed range, typically [0, 1], by subtracting the minimum observed value and dividing by the range (max minus min). In WiFi sensing, this step is critical for removing hardware-induced offsets and environmental biases that cause CSI measurements to vary across devices, antenna pairs, or subcarriers, ensuring that subsequent feature extraction or model training operates on comparable scales. Common variants include global normalization applied across all subcarriers simultaneously and per-subcarrier or per-sample normalization, the choice of which can affect how well localized signal variations — such as those exploited in near-field multi-person separation or colormap-based amplitude encoding — are preserved and distinguished by the downstream classifier.
Source Papers
- Channel State Information (CSI) Amplitude Coloring Scheme for Enhancing Accuracy of an Indoor Occupancy Detection System Using Wi-Fi Sensing ↗ — Channel State Information (CSI) Amplitude Coloring Scheme fo
- Human Activity Recognition via Wi-Fi and Inertial Sensors With Machine Learning ↗ — Human Activity Recognition via Wi-Fi and Inertial Sensors Wi
- MUSE-Fi: Contactless MUti-person SEnsing Exploiting Near-field Wi-Fi Channel Variation ↗ — MUSE-Fi: Contactless MUti-person SEnsing Exploiting Near-fie