Data-Driven Channel Prediction for 5G and Beyond Networks
A new research paper proposes a data-driven approach for channel estimation in 5G and beyond wireless networks, aiming to reduce computational and communication complexity compared to traditional pilot-based methods. The study, published on arXiv, focuses on the 7GHz frequency band and uses ray tracing for data generation and machine learning to predict channel coefficients based on transmitter and user locations. The work can be deployed as a digital twin in future networks.
Key facts
- arXiv paper 2605.01777 proposes data-driven channel estimation for 5G and beyond.
- Traditional methods use periodic pilots, increasing complexity.
- The approach uses ray tracing for data generation.
- Machine learning models predict channel coefficients from transmitter and user locations.
- Focuses on 7GHz frequency band.
- Can be deployed as a digital twin in wireless networks.
- Aims to improve user experience by reducing complexity.
- Published on arXiv.
Entities
Institutions
- arXiv