R1-2601843
discussion
Views on HARQ related aspects in 6GR
From Nokia
Nokia's prior position on
10.5.4.3
at
RAN1#124
· AI-synthesized, paraphrased
verify sources →
Proposes that 6GR shall retain NR's asynchronous adaptive HARQ operation for UL-SCH and DL-SCH. Presents a technical case against NR's tight coordination between UL and DL schedulers in HARQ-ACK reporting, observing that DAI dependencies in Type-2 HARQ-ACK codebooks and timing indication in DL assignments increase scheduling complexity, and proposes studying mechanisms that decouple DL assignment from HARQ-ACK feedback triggering to enable parallel CA processing. Argues that a 6G flexible CA framework with decoupled UL and DL carriers requires HARQ-ACK reporting with decoupled scheduling, targeting a design that allows decoupling of DL scheduling information between cell groups. Proposes studying enhanced feedback techniques including multi-value NACK to differentiate PDSCH decoding failure from missed DL assignment, and indication of detected PDCCH count, to limit the impact of missed DL assignments on adaptive DL HARQ operation.
Summary
Nokia's contribution discusses HARQ operation for 6GR, presenting 9 proposals and 9 observations focused on reducing UL/DL scheduling coordination, mitigating missed DL assignments, defining HARQ-ACK payload sizes, studying L1/L2 feedback signaling, analyzing NACK-to-ACK error impact, exploring finer granularity feedback, and triggering faster ARQ for UL.
Position
Nokia proposes studying HARQ enhancements that decouple UL and DL scheduling to reduce tight coordination requirements present in NR, specifically by removing the link between HARQ-ACK feedback transmission timing and DL assignment timing and decoupling DL scheduling information across cells or cell groups. They propose studying PDCCH detection feedback techniques, including multi-value NACK and explicit counts of detected PDCCHs or failed PDSCH decodings, to mitigate the severe impact of missed DL assignments on fully adaptive HARQ re-transmissions using reserved MCS or non-self-decodable RVs. They propose a HARQ-ACK payload range from 1 to 1536 bits as a working assumption, based on NR values for CBG sizes, TBs per PDSCH, and realistic assumptions on reported HARQ processes and carriers. They present an observation from system-level simulations that NACK-to-ACK errors at 0.1% probability only marginally degrade simulated application layer throughput except at the very low end of the distribution, supporting NR's design target. Nokia proposes studying finer granularity HARQ-ACK feedback and discussing DCI-based triggering of a faster ARQ procedure for UL to avoid RLC status report latency.
Key proposals
- Proposal 1 (Sec 2): Studies on 6GR HARQ feedback should aim for a HARQ feedback design that maintains a simple structure also with more complex configurations and facilitates parallel processing of different CA serving cells without requiring low-latency inter-carrier dependencies of schedulers and DCI.
- Proposal 2 (Sec 2.1): Study HARQ enhancements to reduce the coordination between UL and DL schedulers for single cell and multiple cells, including decoupling the DL assignment from the triggering of the HARQ-ACK feedback report in the UL and transmission of HARQ-ACK feedback on PUSCH or PUCCH.
- Proposal 3 (Sec 2.1): Study Type-3 and Type-2 HARQ-ACK codebooks and enhancements necessary to support HARQ-ACK reporting with decoupled scheduling.
- Proposal 4 (Sec 2.1): Target for a 6G HARQ-ACK feedback design that allows decoupling of DL scheduling information between cells or cell groups.
- Proposal 5 (Sec 2.2): Study feedback techniques with reasonable overhead for PDCCH detection as part of the 6GR PDSCH HARQ feedback framework to limit the impact of missed DL assignments on the adaptive DL HARQ operation, including NACK with multiple values and information on the number of detected PDCCH(s) or failed PDSCH decoding(s).
- Proposal 6 (Sec 2.3): HARQ-ACK payload range from 1 to 1536 bits is used as working assumption in the 6GR design.
- Proposal 7 (Sec 2.4): Study L1 and L2 HARQ-ACK feedback to understand which advantages the L2 signalling vs. the L1 signalling may have.
- Proposal 8 (Sec 2.6): Study finer granularity HARQ-ACK feedback in addition to TB level granularity taking into account achievable gains in throughput, overhead reduction and UE energy efficiency.
- Proposal 9 (Sec 2.7): RAN1 to discuss the possibility for triggering fast ARQ procedure for UL with DCI signalling.