R1-2600150
discussion
Beam management for downlink and uplink
From Huawei
Summary
This Huawei contribution proposes a simplified and unified 6G beam management framework, consisting of 4 observations and 13 proposals covering beam reporting, beam indication, beam failure recovery, enhancement solutions including long-term channel information and AI/ML, and evaluation methodology.
Position
Huawei argues that the 5G beam management framework suffers from redundancy due to separately introduced features across releases, and proposes a simplified and unified 6G BM framework that merges MIMO and mobility beam reporting into a single CSI framework, eliminating the separate LTM-CSI-ReportConfig. They propose studying long-term channel information-based beam management to reduce reliance on exhaustive beam sweeping for both initial access and connected mode, enabling faster beam/link recovery without blind search. They require studying enhanced IMR configuration that decouples CMR-IMR association and explicit interference beam reporting to enable beam-based scheduling. They propose studying enhanced beam reporting with UE-reported beam combination coefficients (similar to Type II PMI) for dynamic beam generation to address mismatch between pre-defined DFT beams and UE channel paths. They propose studying two one-sided AI/ML models with enhanced beam report contents including confidence/reward/score fields, with continuous online finetuning via near-real time LCM mechanisms.
Key proposals
- Proposal 1 (Sec Overview): 6G beam management should study a simplified and unified beam management framework that accommodates existing 5G BM features, and further enhancements on BM performance and efficiency.
- Proposal 5 (Sec Simplified and unified beam reporting framework): A unified beam reporting framework should be supported for MIMO and mobility.
- Proposal 6 (Sec TCI-state framework): The unified TCI framework can be considered as a starting point of 6G.
- Proposal 7 (Sec QCL framework): The QCL framework, including QCL type and QCL chain, should be refined based on real commercial use cases.
- Proposal 8 (Sec Combination of BFR and UEIBR): Consider to merge BFR procedure into UE initiated beam reporting in 6G.
- Proposal 9 (Sec Long-term channel information-based beam management): 6G should study long-term channel information-based beam management, covering both initial beam acquisition and connected mode beam tracking, to reduce overhead, latency and energy consumption.
- Proposal 10 (Sec Interference beam reporting): Study enhanced IMR configuration and interference reporting to provide explicit interference information among beams for beam-based scheduling.
- Proposal 11 (Sec Beam reporting for dynamic beam generation): Study enhanced beam reporting to enable dynamic beam generation.
- Proposal 12 (Sec AI based beam management): Study AI/ML for beam management and extension in 6G, at least including enhanced beam report contents for more efficient UE/BS interaction.
- Proposal 13 (Sec Evaluation Method and KPI): Consider the evaluation of BM performance using link-level simulations in the current stage, and the use of system-level simulations in further evaluations.