R1-2601818 discussion

Discussion on HARQ related aspects

From Spreadtrum
Status: not treated
WI: FS_6G_Radio
Agenda: 10.5.4.3
Release: Rel-20
Source: 3gpp.org ↗
Spreadtrum's prior position on 10.5.4.3 at RAN1#124 · AI-synthesized, paraphrased
verify sources →
Proposes studying a unified and integrated scheduling framework that natively supports all combinations of dynamic scheduling schemes (single-slot, repetition, multi-PDSCH/PUSCH, multi-cell, and multi-cell multi-PDSCH/PUSCH). Proposes reducing the number of HARQ-ACK codebook types from the current five in NR, specifically addressing pain points such as Type-1's large feedback bits and complex generation, Type-2's last DCI issue and inter-cell coordination dependency, and the complexity of enhanced Type-2 for two PDSCH groups. Proposes adopting a single basic PDSCH/PUSCH processing time per device type as the starting point, eliminating the dual-capability structure of NR, while also proposing differentiated processing times for diverse device types and further study on relaxation for energy efficiency and large bandwidth configurations. Proposes studying L1-based dynamic indication for faster ARQ retransmission triggered by HARQ process status, requiring definition of the previous TB associated with a HARQ process ID. Proposes cross-carrier HARQ retransmission within the Multi-Carrier Single Cell (MCSC) architecture, where an initial transmission on one carrier can be retransmitted on another carrier within the same cell.

Summary

This Spreadtrum contribution discusses 6G HARQ-related aspects, presenting 22 observations and 15 proposals across feedback timing, codebook design, processing timelines, ARQ integration, and HARQ process management. The document advocates for simplifying NR's complex HARQ framework by reducing codebook types, adopting a single reference time for feedback, and introducing cross-carrier retransmission for Gothia cells.

Position

Spreadtrum proposes simplifying 6G HARQ timing by introducing a single reference time and one point for HARQ-ACK on PUCCH and UCI mapping on PUSCH, arguing against NR's complex symbol-level overlap checking with variable PUCCH resource sets and PRI-based resource determination that changes within the same slot. They recommend retaining DCI-indicated slot offset K1 for DL HARQ feedback timing while advocating for a single basic PDSCH/PUSCH processing time per device type, with additional relaxation for energy efficiency and large bandwidth configurations (e.g., 200 MHz or 400 MHz). For codebook design, they propose continuing Type-2 HARQ-ACK codebook with enhancements to address the last DCI issue and inter-cell DAI coordination dependency, while reducing the total number of codebook types from five. They additionally propose studying cross-carrier HARQ retransmission for the Gothia cell multi-carrier architecture and DCI-based UL faster ARQ retransmission indications triggered by HARQ process failure status.

Key proposals

Your notes

Private to your account