R1-2601813 discussion

Discussion on beam management for downlink and uplink for 6GR

From Spreadtrum
Status: not treated
WI: FS_6G_Radio
Agenda: 10.5.2.4
Release: Rel-20
Source: 3gpp.org ↗
Spreadtrum's prior position on 10.5.2.4 at RAN1#124 · AI-synthesized, paraphrased
verify sources →
Proposes hybrid beamforming as the fundamental beamforming architecture for 6G ultra large antenna arrays, supporting digital, analog, and hybrid architectures. Proposes studying a unified CSI framework that integrates beam management across network-controlled and UE-initiated cases, single TRP and multi-TRP, and intra-cell and inter-cell scenarios, including LTM. Requires studying beam measurement schemes with low latency and overhead, and considers designing a unified event-triggered reporting procedure covering BFR, UE-initiated beam management, and event-triggered TRP/cell switching. Supports carrying forward NR BM-Case1 (spatial domain DL Tx beam prediction) and BM-Case2 (temporal DL Tx beam prediction) into 6G, and extending AI/ML beam prediction to inter-cell/multi-TRP and cross-frequency scenarios in 6G Day1. Considers NR's unified TCI framework as the baseline for 6G TCI framework and proposes studying collaborative beam measurement and reporting based on UE aggregation to reduce overall overhead and power consumption.

Summary

Spreadtrum presents 2 observations and 11 proposals on 6G beam management, arguing NR beam management effectively improves performance but suffers from excessive spec complexity across releases. They propose a unified 6G framework integrating NW-initiated and UE-initiated beam management, enhanced with AI/ML for inter-cell/multi-TRP predictions, while using NR procedures like P1/P2/P3 and unified TCI as a baseline.

Position

Spreadtrum proposes adopting hybrid beamforming as the fundamental 6G architecture and using NR beam measurement procedures (P1/P2/P3, U1/U2/U3) and the NR unified TCI framework as the baseline for 6G design. They propose studying a unified beam management framework that integrates NW-initiated and UE-initiated procedures, single-TRP and multi-TRP, and intra-cell and inter-cell scenarios to avoid the fragmented spec effort seen in NR. For beam reporting, they propose designing a unified and flexible format accommodating inter-cell, multi-TRP, and UL beam selection, jointly considering AI and non-AI content. They propose extending AI/ML beam prediction (BM-Case1 and BM-Case2) to inter-cell/multi-TRP scenarios and cross-frequency prediction, while also studying collaborative beam measurement via UE aggregation to reduce overhead.

Key proposals

Your notes

Private to your account