The Radial Basis Function (RBF) kernel, also known as the Gaussian kernel, is a popular kernel function used in Support Vector Machine (SVM) classifiers that maps input features into a high-dimensional space by computing similarity based on the squared Euclidean distance between data points, controlled by a bandwidth parameter gamma. In WiFi/CSI sensing research, it is employed to enable non-linear classification of CSI features for tasks such as occupancy detection and passenger counting, where the relationship between raw signal statistics and occupancy states is inherently complex and non-linear. Its importance lies in its flexibility and strong generalization performance across varying environments and configurations, making it a reliable choice compared to linear kernels; the primary variant of concern is the tuning of the gamma parameter and the regularization constant C, which are typically optimized through cross-validation to balance model fit and generalization.
Source Papers
- CSI-based Passenger Counting on Public Transport Vehicles with Multiple Transceivers ↗ — CSI-based Passenger Counting on Public Transport Vehicles wi
- CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing ↗ — CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing
- Device-Free Passive Identity Identification via WiFi Signals ↗ — Device-Free Passive Identity Identification via WiFi Signals
- Device-free occupancy detection and crowd counting in smart buildings with WiFi-enabled IoT ↗ — Device-free occupancy detection and crowd counting in smart
- Time matters: Empirical insights into the limits and challenges of temporal generalization in CSI-based Wi-Fi sensing ↗ — Time matters: Empirical insights into the limits and challen