Channel Foundation models and AI-native systems for wireless communications, sensing, and radio access networks.
GREAT Wireless AI is an open research organization developing reusable AI models and systems for wireless communications. The organization was established around integrated sensing and communication (ISAC) research and has expanded toward broader Wireless AI and AI-RAN research.
- Channel foundation models
- AI-native radio access networks
- Integrated sensing and communication
- Channel representation, generation, and adaptation
- Reproducible wireless simulation and evaluation
| Project | Description |
|---|---|
| CSI-CLIP | A contrastive channel foundation model based on CIR-CSI consistency |
| CSI-MAE | A masked-autoencoder channel foundation model for wireless communication and sensing tasks |
| Awesome Channel Foundation Models | A curated collection of research on channel foundation models |
For each project, we aim to release research code, model weights, loading examples, and reproducible data-generation pipelines where licensing permits. Each repository documents its own license, training-data provenance, preprocessing contract, and model limitations.
The GitHub organization retains the GREAT-ISAC handle for continuity with
existing papers, repositories, citations, and community links. Our broader
public-facing name is GREAT Wireless AI.