R1-2410173
discussion
Discussion on other aspects for AI/ML for air interface
From Panasonic
Summary
Panasonic proposes 3 key concepts for AI/ML air interface framework: MNO responsibility for associated ID management, consistency scope within MNO boundaries, and dataset association representation through ID-X. The document addresses model identification, data collection, and transfer mechanisms for two-sided AI/ML models.
Position
Panasonic advocates FOR MNO-centric management of associated IDs to preserve vendor proprietary information while enabling multi-cell AI/ML model consistency. They push AGAINST vendor-only management of associated IDs and support removing brackets around network proprietary information disclosure concerns. They favor ID-X representing common associations among datasets rather than individual model or sequence-based identification.
Key proposals
- Proposal 1 (Discussion): The consistency of the associated ID is at least within an MNO
- Proposal 2 (Discussion): MNO should handle and be responsible to associated ID
- Proposal 3 (Discussion): ID-X should represent the common association among datasets to train the model
- Proposal 2.2 (RAN1#118b): The UE-side burden may be reduced if the UE can assume that NW-side additional conditions with the same associated ID are consistent among multiple cells, with information mapping containing proprietary information not disclosed to other vendors