K-Nearest Neighbors (KNN) is a non-parametric supervised classification algorithm that assigns a label to an unknown sample by identifying the K most similar labeled samples in a feature space, typically using distance metrics such as Euclidean distance. In WiFi/CSI and RF-based sensing research, KNN serves as a standard baseline classifier for tasks such as occupancy detection and passenger counting, where extracted signal features are matched against training examples to produce activity or count estimates. While the basic algorithm is straightforward, key variants include distance-weighted KNN, where closer neighbors exert greater influence on the classification outcome, and the choice of K itself represents a critical tuning parameter that directly affects the bias-variance tradeoff in sensing accuracy.
Source Papers
- A Survey on Wireless Device-free Human Sensing: Application Scenarios, Current Solutions, and Open Issues ↗ — A Survey on Wireless Device-free Human Sensing: Application
- Channel State Information from Pure Communication to Sense and Track Human Motion: A Survey ↗ — Channel State Information from Pure Communication to Sense a
- CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing ↗ — CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing
- Human Activity Recognition via Wi-Fi and Inertial Sensors With Machine Learning ↗ — Human Activity Recognition via Wi-Fi and Inertial Sensors Wi
- NeRF2: Neural Radio-Frequency Radiance Fields ↗ — NeRF2: Neural Radio-Frequency Radiance Fields
- WiFi Sensing with Channel State Information ↗ — WiFi Sensing with Channel State Information