Apparatuses, entities, signals, and methods using models with different configurations for wireless communication networks
By using models with different configurations in wireless communication networks and combining the coordinated operation of the device and network sides, the efficiency problem of channel state information feedback quantization is solved, enabling more efficient CSI evaluation and prediction, and improving network robustness and resource utilization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
- Filing Date
- 2024-08-19
- Publication Date
- 2026-05-29
AI Technical Summary
Existing wireless communication networks cannot fully utilize the optimization range in channel state information (CSI) feedback quantization, resulting in insufficient computational cost, transmission resource usage, and network operational robustness.
By employing models with different configurations and through collaborative operations on the device and network sides, the parameters and characteristics of wireless communication networks are evaluated and predicted. By combining neural network methods, selective quantization and configuration adjustments are performed to improve the efficiency of evaluation and prediction.
It effectively balances computational cost, transmission resource usage, and network operation robustness, improving the determination and transmission efficiency of channel state information.
Smart Images
Figure CN122122970A_ABST