R1-2600056
discussion
Discussion on beam management for downlink and uplink for 6GR air interface
From FUTUREWEI
Summary
Futurewei provides 3 observations and 7 proposals on 6G MIMO framework design and beam management, advocating for evolution from 5G NR with specific enhancements for upper midband, fast beam acquisition via mixed antenna architecture, AI/ML integration from day-1, and cross-frequency beam prediction using sub-6GHz channel information.
Position
Futurewei proposes adopting the 5G NR MIMO/RS/CSI framework as the starting point for 6G development, arguing no fundamental paradigm shift is needed while supporting substantial evolution for upper midband (UMB) around 7 GHz and 15 GHz. They propose a mixed antenna architecture combining low-resolution 1-bit all-digital receiver arrays for one-shot beam acquisition with analog/hybrid arrays for data transmission, supported by feasibility results showing near-ideal beamforming gain using 1-bit receivers. They require QCL/TCI enhancements to enable UL signals (SRS) as QCL source RS for DL TCI states, extending the Rel-17/18 unified TCI framework. They propose AI/ML integration as an integral part of the 6G MIMO framework from day-1, identifying cross-frequency beam prediction (FR1-to-FR2) as a candidate use case requiring study of data collection configuration, inference-related resource configuration, and LCM operations including fallback and model switching.
Key proposals
- Proposal 1 (Sec 2.2): Adopt 5G NR MIMO/RS/CSI framework as starting point for 6G and study enhancements for UMB support, general SE/UPT performance, and fast beam acquisition for FR2
- Proposal 2 (Sec 2.3): RAN1 to discuss high-level MIMO framework principles enabling AI/ML-based functionality integration from day-1 of 6G
- Proposal 3 (Sec 2.3): Identify a few AI/ML use cases to facilitate study and assessment of AI/ML-ready MIMO framework design
- Proposal 4 (Sec 3.1): Support mixed antenna architecture where a low-resolution digital array receiver is included for fast beam acquisition
- Proposal 5 (Sec 3.2): Consider the unified TCI framework developed in 5G NR as a starting point for 6GR QCL/TCI framework
- Proposal 6 (Sec 3.2): Support QCL/TCI enhancements to enable the use of UL signal (e.g., SRS) as QCL source RS of DL TCI states
- Proposal 7 (Sec 3.3.4): Consider AI/ML-based cross-frequency beam management predicting optimal mmWave beams using sub-6GHz channel information as a candidate approach for BM enhancement
- Observation 1 (Sec 2.3): AI/ML should be considered as part of the MIMO framework from day-1 since AI/ML-assisted operation may impact fundamental air-interface tradeoffs (overhead vs. performance and robustness)
- Observation 2 (Sec 3.1): One-shot beam acquisition based on 1-bit all-digital receiver is feasible
- Observation 3 (Sec 3.3.3): Using FR1 channel measurements to predict optimal FR2 beams is feasible in co-located scenario; angular-delay domain inputs achieve better Top-K performance than raw channel matrix inputs