R1-2500339
discussion
Specification support for CSI prediction
From vivo
vivo's prior position on
9.1.3
at
RAN1#119
· AI-synthesized, paraphrased
verify sources →
Identifies TXRU virtualization mapping as a critical network-side additional condition, presenting simulation evidence that mismatches cause substantial SGCS loss (up to -44.4%) in high-speed and outdoor scenarios. Advocates for adopting the associated ID solution established in the Beam Management use case to link data collected under specific network-side conditions to unique identifiers. Pushes against ignoring the significant performance degradation caused by TXRU mapping mismatches between training and inference phases.
Summary
This document from vivo analyzes the impact of TXRU virtualization mapping mismatches on AI-based CSI prediction generalization, demonstrating significant performance losses in high-speed and outdoor scenarios. It proposes using associated IDs to ensure training-inference consistency, establishing separate processing unit pools for AI and legacy CSI, and prioritizing Type 3 and Type 1 performance monitoring with SGCS metrics calculated between precoding vectors.
Position
Vivo identifies TXRU virtualization mapping as a critical NW-side additional condition that causes significant generalization performance loss, particularly at 60km/h or with high proportions of outdoor users. They propose adopting the associated ID solution from the Beam Management use case to ensure consistency between training and inference data for UE-side models. For CSI feedback mechanisms, vivo prefers separate processing unit pools for AI-based CSI and legacy CSI to avoid interference and simplify UE design, rather than a joint pool. Regarding performance monitoring, vivo supports using SGCS as the metric, calculated specifically between the precoding vectors of the predicted and real channels to account for codebook quantization effects. They prioritize Type 3 and Type 1 monitoring configurations over Type 2, citing lower overhead and complexity, and propose reusing the dedicated CSI report configuration mechanism defined for AI beam management Case 2.
Key proposals
- Proposal 1 (Sec 2.1): Identify TXRU mapping as a NW-side additional condition significantly impacting generalization performance.
- Proposal 2 (Sec 2.1): Follow the solution based on associated ID as used in the Beam Management (BM) use case to address training/inference consistency for UE-side models.
- Proposal 3 (Sec 3.1): Establish separate processing unit pools for AI-based CSI and legacy CSI to quantify AI-based CSI processing capabilities.
- Proposal 4 (Sec 3.2): Support SGCS (Spectral Grid Correlation Score) as the performance metric for CSI prediction performance monitoring.
- Proposal 5 (Sec 3.2): Calculate the performance monitoring metric (SGCS) between the precoding vectors of the predicted channel and the real channel, rather than raw channel matrices.
- Proposal 6 (Sec 3.2): Prioritize Type 3 and Type 1 performance monitoring over Type 2 due to lower overhead and UE complexity.
- Proposal 7 (Sec 3.2): Configure a dedicated CSI report configuration for monitoring associated with the inference configuration, linking each monitoring RS occasion to an inference occasion following the AI beam management Case 2 mechanism.