A Generative Adversarial Network (GAN) is a deep learning framework consisting of two neural networks — a generator and a discriminator — trained in opposition, where the generator learns to synthesize realistic data samples and the discriminator learns to distinguish synthetic from real data. In Wi-Fi and CSI-based sensing research, GANs matter primarily because they address critical data scarcity and domain shift challenges: they can augment limited labeled training data, generate synthetic CSI samples across unseen environments or user configurations, and enable domain adaptation by transforming data distributions across different deployment conditions to improve model generalizability. Key variants employed in the field include Conditional GANs (cGANs), which condition generation on class labels or environmental contexts; CycleGANs, used for unpaired domain translation between source and target sensing environments; and Wasserstein GANs (WGANs), which offer more stable training dynamics — collectively making GANs a central tool for improving the reproducibility and cross-domain robustness of Wi-Fi sensing systems.
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
- A Survey on Wireless Device-free Human Sensing: Application Scenarios, Current Solutions, and Open Issues ↗ — A Survey on Wireless Device-free Human Sensing: Application
- A survey on CSI-based Wi-Fi sensing datasets and models with a focus on reproducibility ↗ — A survey on CSI-based Wi-Fi sensing datasets and models with
- Data-driven Crowd Modeling Techniques: A Survey ↗ — Data-driven Crowd Modeling Techniques: A Survey
- Deep Learning-Enhanced Human Sensing with Channel State Information: A Survey ↗ — Deep Learning-Enhanced Human Sensing with Channel State Info
- Device-Free Wireless Sensing for Gesture Recognition Based on Complementary CSI Amplitude and Phase ↗ — Device-Free Wireless Sensing for Gesture Recognition Based o
- NeRF2: Neural Radio-Frequency Radiance Fields ↗ — NeRF2: Neural Radio-Frequency Radiance Fields
- Radio Radiance Field: The New Frontier of Spatial Wireless Channel Representation ↗ — Radio Radiance Field: The New Frontier of Spatial Wireless C
- Recent trends in crowd analysis: A review ↗ — Recent trends in crowd analysis: A review
- SenseFi: A library and benchmark on deep-learning-empowered WiFi human sensing ↗ — SenseFi: A library and benchmark on deep-learning-empowered
- Towards Environment Independent Device Free Human Activity Recognition ↗ — Towards Environment Independent Device Free Human Activity R