LightGBM (Light Gradient Boosting Machine) is an efficient, tree-based ensemble machine learning algorithm that uses gradient boosting with a leaf-wise growth strategy and histogram-based split finding to achieve fast training speed and low memory usage. In WiFi CSI sensing research, it is applied to tasks such as occupancy estimation and crowd counting by learning discriminative patterns from CSI-derived features, offering competitive accuracy with significantly reduced computational overhead compared to deeper models. Its scalability and robustness to high-dimensional tabular feature inputs make it well-suited for deployment in resource-constrained or real-time indoor sensing scenarios, and it is commonly used in both regression and classification variants depending on whether the sensing task involves estimating continuous quantities like people count or predicting discrete categories like crowd location zones.
Source Papers
- A Framework to Estimate Classroom Occupancy using WiFi Channel State Information ↗ — A Framework to Estimate Classroom Occupancy using WiFi Chann
- Wi-CaL: WiFi Sensing and Machine Learning Based Device-Free Crowd Counting and Localization ↗ — Wi-CaL: WiFi Sensing and Machine Learning Based Device-Free