K-nearest neighbors (KNN) is a non-parametric classification and regression method that assigns a label or value to a query sample based on the majority vote or averaged output of its K most similar samples in a feature space, typically measured by a distance metric such as Euclidean distance. In WiFi CSI sensing and related domains, KNN serves as a lightweight yet effective baseline for tasks such as indoor crowd counting and traffic state estimation, where it can map extracted signal or state features to discrete or continuous target values without requiring extensive model training. Key variants include weighted KNN, where closer neighbors contribute more to the prediction, and adaptive KNN approaches that dynamically adjust K based on local data density, which are particularly relevant in few-shot or domain-agnostic settings where labeled samples are scarce and generalization across environments is critical.
Source Papers
- DASECount: Domain-Agnostic Sample-Efficient Wireless Indoor Crowd Counting via Few-Shot Learning ↗ — DASECount: Domain-Agnostic Sample-Efficient Wireless Indoor
- Guiding Wi-Fi Sensor Placement for Enhanced CSI-Based Sensing in Stationary Crowd Counting ↗ — Guiding Wi-Fi Sensor Placement for Enhanced CSI-Based Sensin
- Physics-Informed Deep Learning for Traffic State Estimation: A Survey and the Outlook ↗ — Physics-Informed Deep Learning for Traffic State Estimation:
- Recent trends in crowd analysis: A review ↗ — Recent trends in crowd analysis: A review
- SenseFi: A library and benchmark on deep-learning-empowered WiFi human sensing ↗ — SenseFi: A library and benchmark on deep-learning-empowered
- WiFi-Based Human Sensing With Deep Learning: Recent Advances, Challenges, and Opportunities ↗ — WiFi-Based Human Sensing With Deep Learning: Recent Advances