R1-2600116
discussion
Discussion on downlink transmission scheme(s) for downlink shared channels for 6GR
From Spreadtrum
Summary
This Spreadtrum contribution provides 15 proposals and 5 observations on 6G downlink transmission schemes, codeword-to-layer mapping, DMRS design, and MRSS aspects. The company advocates for unified transmission schemes to minimize complexity, specifies DMRS overhead reduction using AI/ML receivers, and proposes baseline NR-reusing design principles with conservative enhancements only for significant performance gains.
Position
Spreadtrum proposes unified 6GR downlink transmission schemes to minimize UE and NW implementation complexity, explicitly opposing multiple schemes for the same scenario and scheme enhancements yielding less than 10% performance gain. They retain NR baselines for maximum layers (8), maximum codewords (2), DMRS port count, and PN sequence for CP-OFDM, while requiring a single DMRS type for both DL and UL. For AI/ML, they present simulation results showing AI-based channel estimation achieving similar BLER with reduced DMRS overhead (2-symbol DMRS matching 3-symbol non-AI performance in time domain, 1/3 RE reduction in frequency domain) and propose prioritizing sparse orthogonal DMRS (sub-case A) over superimposed pilots (SIP), which they consider insufficiently evaluated for multi-user interference and standardization impact. For NR-6GR MRSS, they propose semi-static FDM/TDM and rate matching mechanisms while opposing SDM between NR and 6GR due to negative performance impacts.
Key proposals
- Proposal 1 (Sec 2.1): Strive for unified transmission scheme(s) to minimize downlink transmission schemes and configurations to reduce implementation complexity and the cost for both UE and NW sides; avoid multiple schemes for the same scenario; avoid unnecessary scheme enhancement to pursue small incremental performance gain, e.g. less than 10%.
- Proposal 2 (Sec 2.1): Taking NR as the starting point, study the following downlink transmission scheme aspects in 6GR: closed-loop and open-loop, sTRP and mTRP (including NCJT and CJT), SU/MU-MIMO, transmit diversity, and others.
- Proposal 3 (Sec 2.2): For PDSCH scheduling, the maximum number of layers is 8 and the maximum number of codewords is 2.
- Proposal 4 (Sec 2.2): For rank=1, schedule 1 codeword; for rank 2-4, study whether to support 1 or 2 scheduled codewords; for rank 5-8, schedule 2 codewords.
- Proposal 5 (Sec 2.2): Consider the layer mapping rule of first half layers mapped to CW0 and remaining layers mapped to CW1 as baseline for 2 scheduled CWs.
- Proposal 6 (Sec 2.2): Consider a single MCS for one codeword as baseline.
- Proposal 7 (Sec 2.3.1): Strive to define a single DMRS type for both DL and UL.
- Proposal 8 (Sec 2.3.1): Regarding the maximum number of DMRS ports, consider the maximum number of DMRS ports supported by 5G NR as baseline.
- Proposal 10 (Sec 2.3.1): For DMRS port multiplexing, FDM, TDM and CDM can be considered.
- Proposal 11 (Sec 2.3.2): Support DMRS overhead reduction sub-case A for 6GR Day 1 AI use case, considering UE-side model for DL DMRS overhead reduction in time, frequency, or joint domains.
- Proposal 12 (Sec 2.3.2): DMRS pattern discussion in 6GR Day1 should consider lower overhead for AI/ML based receiver.
- Proposal 13 (Sec 2.3.2): SIP needs to have sufficient evaluation results before selected as the priority use case, with at least interference from multiple users and standardization impact (e.g., DMRS design) investigated.
- Proposal 14 (Sec 2.4): High-level aspects to consider for NR-6GR MRSS include UE/NW implementation complexity, resource allocation coordination between NR-6GR, radio resource utilization, operating bands at least existing FR1, aligned numerology and waveform as NR in shared band, unified design between MRSS and non-MRSS scenario, and no impact to legacy NR UE behavior.
- Proposal 15 (Sec 2.4): Consider semi-static FDM/TDM between NR and 6GR and rate matching of 6GR PDSCH around NR signals/channels as potential MRSS mechanisms.