Independent Component Analysis (ICA) is a statistical signal processing technique that decomposes a multivariate signal into maximally statistically independent components, operating under the assumption that the observed signals are linear mixtures of unknown independent sources. In WiFi/CSI sensing, ICA is applied as a preprocessing or feature extraction method to separate useful human-induced signal variations — such as those caused by respiration, gestures, or movement — from noise, interference, and hardware-induced artifacts that contaminate raw CSI measurements. Its importance lies in improving signal quality and isolating distinct source contributions without requiring labeled data or prior knowledge of the mixing process, making it particularly valuable for activity recognition, vital sign monitoring, and other fine-grained sensing tasks where overlapping signal components would otherwise obscure meaningful patterns.
Source Papers
- A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techniques, Datasets, and Future Research Prospects ↗ — A Survey on Wi-Fi Sensing Generalizability: Taxonomy, Techni
- WiFi Sensing with Channel State Information ↗ — WiFi Sensing with Channel State Information
- WiFi-Based Human Sensing With Deep Learning: Recent Advances, Challenges, and Opportunities ↗ — WiFi-Based Human Sensing With Deep Learning: Recent Advances