R1-2509303
discussion
Discussion on DL DFT-s-OFDM for 6GR
From CATT
Summary
This document from CATT and co-signatories summarizes observations and provides two proposals for studying DFT-s-OFDM as an additional DL waveform for 6G. It contains 37 observations summarizing prior meeting contributions on PAPR reduction, link performance, and use cases, followed by 2 proposals regarding evaluation criteria and technical study details.
Position
CATT and co-signatories propose a structured framework for evaluating DL DFT-s-OFDM as a 6G additional waveform using a Net Gain metric defined as Tx power gain minus link loss at target KPI, or network energy saving gain relative to baseline. They propose that RAN1 study specific target channels/signals including CSS PDCCH, unicast/common PDSCH (including Msg4 and SIB1), SSB, and Wake-up signal. They present two explicit options for DFT transform precoding application when multiplexing UEs (before vs. after multiplexing) and two options for DMRS multiplexing with DFT-s-OFDM (excluding DMRS REs from precoding vs. TDM of DMRS with data). They require realistic PA models and consideration of RAN4 metrics for Tx power calculation.
Key proposals
- Proposal 1 (Sec 2): For downlink low-PAPR proposals the primary evaluation criterion is one of: Net Gain [dB] = Tx power gain - link loss relative to the reference @ Target KPI; or Network energy saving gain relative to baseline for BS.
- Proposal 2 (Sec 2): For study of DFT-s-OFDM as potential additional DL waveform, RAN1 should consider target channels/signals including PDCCH (CSS), PDSCH (unicast, common, SIB1), SSB, and Wake-up signal.
- Proposal 2 details (Sec 2): How to apply DFT transform precoding when multiplexing UEs/channels into one symbol - Option 1 (before multiplexing, DFT size equals assigned subcarriers per channel) or Option 2 (after multiplexing, DFT size can be larger than per-channel assigned resources).
- Proposal 2 details (Sec 2): How to multiplex DMRS for PDCCH/PDSCH - Option 1 (DFT precoding applied to REs excluding DMRS-containing REs) or Option 2 (UE expects DMRS always TDMed with data RE at OFDM symbol level).
- Proposal 1 Note (Sec 2): A realistic PA model should be used; when calculating Tx power gain, RAN4 metrics on Tx power should be taken into account.
- Proposal 1 Note (Sec 2): For link loss relative to the reference, fading channel and non-ideal channel estimation including DMRS configuration and equalization is encouraged. Companies to report how to calculate Tx power gain, modulation and coding.
- Observation 1 (Sec 2): RAN1#122bis meeting observed benefit of adopting DFT-s-OFDM for additional DL waveform with target use cases: IoT coverage, NTN coverage, FR2/FR3 coverage, Network Energy Saving.
- Observation 9-17 (Sec 2): PAPR reduction values range from 0.5 dB to 3.2 dB over CP-OFDM depending on number of FDMed DL transmissions and precoding application method (separate vs joint).
- Observation 37 (Sec 2): One company observed that DFT/IDFT module can be implemented by IDFT/DFT with conjugate module, respectively.
- Proposal 2 Note (Sec 2): Which channels will be evaluated is up to company; other options for precoding and DMRS multiplexing are not precluded.