R1-2407795
discussion
Discussion and reply LS on applicable functionality reporting for beam management UE-sided model
From ZTE
Summary
ZTE provides a comprehensive response to RAN2's liaison statement on AI/ML beam management functionality reporting, addressing 10 specific questions with 14 detailed proposals covering UE capability signaling, network-side conditions, and activation procedures.
Position
ZTE advocates FOR flexible and optional network-side signaling to minimize overhead while maintaining functionality, supporting sub-use-case level granularity for AI/ML beam management, and reusing existing CSI framework mechanisms. They push AGAINST mandatory associated ID signaling and unnecessary inference configuration in Step 3, arguing for implementation flexibility and backward compatibility with future AI/ML extensions.
Key proposals
- Proposal 1 (Sec 2): For AI/ML beam management, the feature group is defined at least per sub use case level (i.e., BM-Case1, BM-Case2) due to the intrinsic difference of the two sub use cases, and the component values within each feature group are signaled to the NW by UE capability indication in Step 2
- Proposal 2 (Sec 2): To ensure better compatibility with future extensions of AI/ML technologies, the condition combinations supported by the UE are aligned between NW and UE outside the UE capability indication
- Proposal 3 (Sec 2): Clarify that the granularity of functionality, expected to be at least at the sub-use case level or even finer, is not related to the UE capability discussion in Step 2
- Proposal 4 (Sec 3): Associated ID can be seen an implicit representation of the NW-side additional conditions for UE-sided models in beam management. Discussion about the content of NW-side additional conditions is not necessary
- Proposal 5 (Sec 3): A single functionality might be associated with one or multiple associated IDs, and conversely, different functionalities might be associated with the same associated ID
- Proposal 6 (Sec 3): In Step 3, the provision of associated ID is optional, which is an implementation behavior of the NW side
- Proposal 7 (Sec 3): The UE can train and recommend applicable functionalities without the signaling of associated ID, while the final decision on functionality applicability is made by the network
- Proposal 8 (Sec 3): To avoid unnecessary signaling consumption, the provision of configuration (e.g. inference configuration) other than NW-side additional condition in Step 3 for UE to determine applicable functionalities is not necessary
- Proposal 9 (Sec 3): The content of applicable functionality reporting in Step 4 includes condition combinations with model availability and applicable associated ID
- Proposal 10 (Sec 3): The content of inference configuration in Step 5 comprises configurations of the associated CSI report
- Proposal 11 (Sec 4): If inference configuration is provided in Step 3, it does not activate the functionality immediately upon receiving Step 3
- Proposal 12 (Sec 4): If inference configuration is not provided in Step 3, configuration in Step 5 does not necessarily activate the functionality immediately upon receiving Step 5, depending on the temporal characteristics of the associated CSI report
- Proposal 13 (Sec 4): If more than one functionality are configured in Step 3 or Step 5, multiple/all applicable functionalities can be activated if the simultaneous processing of these functionalities doesn't exceed the UE capabilities
- Proposal 14 (Sec 4): L1/L2 signaling for functionality activation/deactivation is needed, which can reuse the activation/deactivation signaling for CSI report