R1-2410194
discussion
Discussions on CSI prediction
From LG Electronics
Summary
This LG Electronics contribution discusses AI/ML-based CSI prediction for NR air interface, focusing on consistency between training and inference, and presents simulation results showing marginal performance loss for gNB antenna tilt angle variations. The document contains 4 proposals and 1 observation addressing specification requirements, performance monitoring priorities, and CSI processing criteria.
Position
LG Electronics advocates FOR simplifying specification requirements by concluding that no additional spec support is needed for gNB antenna tilt angle consistency, while pushing FOR prioritizing Type 1 and Type 3 performance monitoring over Type 2 due to lower reporting overhead and higher accuracy. They are AGAINST Type 2 performance monitoring due to its larger payload requirements and quantization loss issues.
Key proposals
- Proposal 1 (Consistency section): For CSI prediction using UE-sided model, conclude that specification support is not needed to ensure consistency between training and inference with respect to gNB antenna tilt angles
- Proposal 2 (Inference section): It may be necessary to discuss whether the legacy CSI processing criteria and timeline (e.g., the required number and/or occupation time of CPUs, the values of Z/Z', and total number active/simultaneous CSI-RS resource/ports) are sufficient
- Proposal 3 (Performance monitoring section): Prefer to prioritize type 1 and 3 performance monitoring for CSI prediction using UE-sided model
- Proposal 4 (Performance monitoring section): It may be necessary to discuss how to handle cases where multiple predicted CSIs exist in the work item phase related to performance monitoring issues