Domain adaptation is a machine learning technique that transfers knowledge or model parameters learned in one environment (the source domain) to enable accurate inference in a different environment (the target domain), addressing the performance degradation that occurs when CSI-based sensing models are deployed in settings other than where they were trained. It matters critically for Wi-Fi CSI sensing because channel state information is highly sensitive to environmental geometry, furniture layout, and multipath propagation characteristics, meaning a model trained in one room will often fail in another without adaptation. Key variants include supervised domain adaptation, which requires some labeled target-domain data for fine-tuning, unsupervised domain adaptation, which leverages unlabeled target-domain samples through techniques such as adversarial training or distribution alignment, and transfer learning approaches that freeze or partially retrain pretrained feature extractors to reduce the cost of recollecting and relabeling CSI data in each new deployment environment.
Source Papers
- A CSI Dataset for Wireless Human Sensing on 80 MHz Wi-Fi Channels ↗ — A CSI Dataset for Wireless Human Sensing on 80 MHz Wi-Fi Cha
- CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing ↗ — CrossSense: Towards Cross-Site and Large-Scale WiFi Sensing
- Deep Learning-Enhanced Human Sensing with Channel State Information: A Survey ↗ — Deep Learning-Enhanced Human Sensing with Channel State Info
- Investigation of Environment Dependence in Wi-Fi CSI-Based Crowd Counting Systems ↗ — Investigation of Environment Dependence in Wi-Fi CSI-Based C
- SenseFi: A library and benchmark on deep-learning-empowered WiFi human sensing ↗ — SenseFi: A library and benchmark on deep-learning-empowered
- Time matters: Empirical insights into the limits and challenges of temporal generalization in CSI-based Wi-Fi sensing ↗ — Time matters: Empirical insights into the limits and challen
- Towards Environment Independent Device Free Human Activity Recognition ↗ — Towards Environment Independent Device Free Human Activity R
- WiFi as Infrastructure: Valuation Impact of CSI Sensing on Smart Buildings and REIT Portfolios ↗ — WiFi as Infrastructure: Valuation Impact of CSI Sensing on S
- WiGNN: WiFi-Based Cross-Domain Gesture Recognition Inspired by Dynamic Topology Structure ↗ — WiGNN: WiFi-Based Cross-Domain Gesture Recognition Inspired