The Fourier transform is a mathematical operation that decomposes a time-domain signal into its constituent frequency components, revealing the spectral content of that signal. In WiFi-based human sensing, it is fundamental for transforming raw CSI or RSSI measurements into the frequency domain, where human motion signatures — such as the subtle bandwidth broadening caused by body fidgeting in stationary crowd counting or periodic movement patterns in activity recognition — become more discriminable and accessible to subsequent processing or deep learning pipelines. Key variants employed in this field include the Short-Time Fourier Transform (STFT), which applies the transform over sliding temporal windows to capture time-varying spectral features, and the Fast Fourier Transform (FFT), an efficient computational algorithm that makes real-time frequency analysis of CSI amplitude and phase data practically feasible.
Source Papers
- Efficient machine learning for Wi-Fi CSI-based human activity recognition using fast Monte Carlo based feature extraction ↗ — Efficient machine learning for Wi-Fi CSI-based human activit
- Fast and Robust Stationary Crowd Counting With Commodity WiFi ↗ — Fast and Robust Stationary Crowd Counting With Commodity WiF
- Human Activity Recognition via Wi-Fi and Inertial Sensors With Machine Learning ↗ — Human Activity Recognition via Wi-Fi and Inertial Sensors Wi
- WiFi-Based Human Sensing With Deep Learning: Recent Advances, Challenges, and Opportunities ↗ — WiFi-Based Human Sensing With Deep Learning: Recent Advances