A Support Vector Machine (SVM) is a supervised machine learning classifier that finds the optimal hyperplane to separate data points of different classes in a high-dimensional feature space, often employing kernel functions to handle non-linear boundaries. In the context of Wi-Fi/CSI sensing and related sensing tasks, SVMs serve as a classical baseline or lightweight classification backend, receiving hand-crafted or learned features (such as CSI statistics or CNN-extracted representations) to perform tasks like activity recognition, crowd counting regression, or occupancy detection. Key variants include the linear SVM, the radial basis function (RBF) kernel SVM for non-linear problems, and the Support Vector Regression (SVR) formulation used when the output is continuous rather than categorical, all valued for their strong generalization on small labeled datasets where deep learning may overfit.
Source Papers
- A Survey on Fusion-Based Indoor Positioning ↗ — A Survey on Fusion-Based Indoor Positioning
- DASECount: Domain-Agnostic Sample-Efficient Wireless Indoor Crowd Counting via Few-Shot Learning ↗ — DASECount: Domain-Agnostic Sample-Efficient Wireless Indoor
- SenseFi: A library and benchmark on deep-learning-empowered WiFi human sensing ↗ — SenseFi: A library and benchmark on deep-learning-empowered