R1-2409789
discussion
Discussion on AI/ML-based CSI prediction
From Apple
Summary
Apple discusses the necessity of network-side additional conditions for AI-based CSI prediction, arguing that the need for consistency signaling depends on whether the UE model relies on spatial signatures. The document presents three proposals regarding the evaluation of generalization cases, the definition of assisted information identifiers, and UE signaling capabilities during data collection requests.
Position
Apple argues that the requirement for NW-side additional conditions depends on the specific UE-side model design, distinguishing between models that do not depend on spatial signatures (e.g., time domain LSTM, time/frequency domain 2D CNN) and those that do (e.g., 3D CNN). They propose that if generalization case 2 evaluation shows performance degradation, consistency signaling is needed for the UE side model. For models relying on spatial signatures, Apple proposes defining assisted information or an identifier to abstract antenna virtualization and multi-TRP details. Furthermore, they propose enabling the UE to signal whether this assisted information is required during the data collection request procedure.
Key proposals
- Proposal 1 (Discussion on NW side additional condition): Evaluate generalization case 2 to identify NW-side additional conditions; if performance degradation is observed, consistency signaling is needed.
- Proposal 2 (Discussion on NW side additional condition): Define assisted information or an identifier to abstract antenna virtualization and multi-TRP configurations to ensure consistency of training and inference.
- Proposal 3 (Discussion on NW side additional condition): Enable a UE request data collection procedure where the UE can signal whether the assisted information/identifier is required in the request.