R1-2601791
discussion
Discussion on aspects of downlink-based CSI acquisition for 6GR air interface
From FUTUREWEI
FUTUREWEI's prior position on
10.5.3.1
at
RAN1#124
· AI-synthesized, paraphrased
verify sources →
Proposes adopting the 5G NR MIMO/RS/CSI framework as the baseline for 6G without fundamental paradigm shifts, focusing evolutionary enhancements on upper midband (UMB) support. Proposes studying hybrid antenna architectures for UMB base stations with large antenna element counts (e.g., >512 elements, specifically 2048-element combinations with 32-256 TXRUs at 7/15 GHz) and presents technical case for simultaneous multi-beam codebooks enabling fast full CSI acquisition in N/K_s orthogonal beams per OFDM symbol. For CSI-RS overhead reduction, proposes structured port-to-RE mapping exploiting beam-domain channel sparsity and a joint channel estimation algorithm based on IDFT, contrasting with AI/ML-based approaches requiring both UE-sided and NW-sided model evaluation. Proposes CSI framework simplification through streamlined CSI-ReportConfig/CSI-ResourceConfig settings and requires proper conventional non-AI/ML baselines be identified for all AI/ML evaluations. Proposes adopting 5G NR two-sided AI/ML CSI compression and further studying JSCC and downloadable codebook approaches while deferring joint source-channel-modulation coding due to higher ecosystem impact.
Summary
Futurewei presents 12 proposals and 5 observations on 6G downlink-based CSI acquisition, covering CSI-RS overhead reduction via structured port-to-RE mapping and AI/ML-based prediction, AI/ML-based CSI compression and feedback with JSCC and downloadable codebooks, CSI acquisition for hybrid antenna architectures supporting fast full CSI via simultaneous multi-beam, unified codebook design with network-indicated CSI-RS precoding matrix and reporting basis, and advanced interference measurement and reporting including interference statistics and prospective probing.
Position
Futurewei proposes studying structured mapping between CSI-RS ports and REs in spatial, frequency, and time domains to facilitate CSI-RS overhead reduction, demonstrating that a beam-domain-first transformation approach achieves approximately 1.5–3 dB better performance at 0 dB SNR compared to frequency-interpolation-first methods for 256 ports at 1/8 RE/RB/port density. They propose considering AI/ML-based approaches for CSI prediction from sparse CSI-RS, presenting results showing AI/ML consistently outperforms linear interpolation baselines in low SNR regimes, with gains more significant when sparsity is introduced in the spatial domain. For hybrid antenna architectures at UMB, they propose studying simultaneous multi-beam codebook designs enabling fast full CSI acquisition, arguing that for 2048 elements with 128 RF chains, full CSI port sweeping can be completed in 16 OFDM symbols versus 2048 symbols using a DFT codebook. They propose studying a unified codebook design where the network indicates the CSI-RS precoding matrix and CSI reporting basis/dictionary matrix to the UE, enabling full CSI acquisition for hybrid architectures, higher-resolution reporting, and incorporating existing 5G NR codebooks as special cases. For interference, they propose enhancements for reporting interference statistics (standard deviation/variance, maximum, minimum, percentile values) and prospective DL interference probing and reporting based on NZP CSI-RS based IMR measurements.
Key proposals
- Proposal 1 (Sec 2): For 6GR DL-based CSI acquisition, support enhancements for CSI-RS overhead reduction, enhancements for CSI report overhead reduction, and enhancements for acquiring interference information.
- Proposal 2 (Sec 2): In 6GR, CSI framework simplification such as streamlining the CSI report/resource configurations/settings may be considered.
- Proposal 3 (Sec 2): In 6GR, when evaluating AI/ML-based approach(es) for downlink-based CSI acquisition enhancements, proper conventional non-AI/ML baseline and common assumptions applicable for both AI/ML and non-AI/ML methods should be identified for each use case/enhancement option.
- Proposal 4 (Sec 3.1): Investigate CSI-RS designs with structured mapping between CSI-RS port and RE in spatial, frequency and time domain to facilitate overhead reduction.
- Proposal 5 (Sec 3.2): In 6GR, when studying enhancements to reduce CSI-RS overhead, consider AI/ML-based approach(es) for downlink-based CSI acquisition as one of the enhancement approaches.
- Proposal 6 (Sec 4.1): Consider adopting 5GNR AI/ML-based CSI compression via two-sided model(s) use case in 6GR study with further evaluations/verifications if needed.
- Proposal 7 (Sec 4.1.5): For AI/ML-based CSI compression and feedback enhancements, consider including CSI Compression with JSCC and CSI compression with downloadable codebook(s) in the 6G study.
- Proposal 8 (Sec 5.1): For UMB base stations, hybrid antenna architecture with a large number of antenna elements and moderate number of TXRUs supporting hybrid beamforming should be a key focus area for antenna panels with > 512 elements, carriers with ≥ 100 MHz bandwidth, or frequencies at the high upper midband.
- Proposal 9 (Sec 5.2): For hybrid BS antennas, study the issue of how to acquire full CSI faster for efficient MU MIMO support.
- Proposal 10 (Sec 5.3): Study extensions of 5G NR CSI-RS design and Type II codebook framework for fast full CSI acquisition by hybrid BS antennas.
- Proposal 11 (Sec 6): For 6GR DL-based CSI acquisition, study indication of CSI-RS precoding matrix and indication of CSI reporting basis/dictionary matrix to UE.
- Proposal 12 (Sec 7): For 6GR DL-based CSI acquisition, support advanced interference measurement and reporting: enhancements for reporting interference statistics and enhancements for prospective DL interference probing and reporting.