R1-2600196
discussion
Other aspects for CSI acquisition
From OPPO
Summary
OPPO presents 7 proposals and 2 observations covering CSI acquisition and RS design for 6G, including no-PMI CSI reporting, fusion of DL CSI feedback and SRS measurements, periodic TRS support, and simplified PTRS design.
Position
OPPO requires supporting CSI report without PMI for reciprocity-based transmission as a 6G day 1 feature, using the NR non-PMI report procedure as a starting point. They propose studying fusion of implicit downlink CSI feedback and SRS measurement at the NW-side, presenting evaluation results showing Option 1 (implicit CSI fusion) achieves SGCS of 0.915/0.844/0.794/0.752 for layers 1-4 respectively, outperforming both SRS-only and explicit CSI fusion. For TRS design, OPPO supports periodic TRS as day 1 with CSI-RS as the starting point, proposes further study on aperiodic TRS necessity, and recommends considering additional usages including XDCP and CJT calibration measurement during design. For PTRS, OPPO argues for a simplified 6G architecture, specifically targeting time/frequency density determination mechanisms, waveform-specific designs, and PTRS-DMRS association rules to reduce specification and implementation complexity.
Key proposals
- Proposal 1 (Sec 2.1): Support CSI report without PMI for reciprocity based transmission in 6GR day 1.
- Proposal 1 (Sec 2.2): Support to study fusion of downlink CSI feedback and SRS measurement in 6GR.
- Observation 1 (Sec 2.2): Fusion of downlink CSI and SRS measurement outperforms SRS only and AI/ML based CSI compression in 5G NR for each layer.
- Observation 2 (Sec 2.2): Fusion of implicit CSI feedback and SRS measurement outperforms fusion of explicit CSI feedback and SRS measurement.
- Proposal 1 (Sec 3.1): Support periodic TRS in 6G day 1.
- Proposal 2 (Sec 3.1): Further study whether to support aperiodic TRS.
- Proposal 3 (Sec 3.1): TRS as specific CSI-RS as the starting point.
- Proposal 4 (Sec 3.1): Consider possible usages during design of TRS, e.g. XDCP and CJT calibration measurement.
- Proposal 1 (Sec 3.2): 6G PTRS should consider simplified design, e.g. for time/frequency density determination, for different waveforms and for PTRS-DMRS association.
- Proposal 4 (Sec 4): Support periodic TRS in 6G day 1.
- Proposal 7 (Sec 4): 6G PTRS should consider simplified design, e.g. for time/frequency density determination, for different waveforms and for PTRS-DMRS association.