Sensor fusion is the process of combining data or estimates from multiple heterogeneous sources — such as WiFi, IMU, camera, Bluetooth, and barometer — into a unified representation that is more accurate, robust, and complete than any single source alone. In the context of indoor positioning and CSI-based sensing, it matters because no individual modality reliably handles all environments and conditions, and fusion systematically mitigates the weaknesses of each source through complementary information. Key variants include data-level fusion (combining raw signals before processing), feature-level fusion (merging extracted representations), and decision-level fusion (integrating independent estimates), with algorithmic approaches spanning Kalman filtering, particle filtering, and deep learning-based fusion architectures.
Source Papers
- A Survey on Fusion-Based Indoor Positioning ↗ — A Survey on Fusion-Based Indoor Positioning
- Guiding Wi-Fi Sensor Placement for Enhanced CSI-Based Sensing in Stationary Crowd Counting ↗ — Guiding Wi-Fi Sensor Placement for Enhanced CSI-Based Sensin
- RSSI-Assisted CSI-Based Passenger Counting with Multiple Wi-Fi Receivers ↗ — RSSI-Assisted CSI-Based Passenger Counting with Multiple Wi-