R1-2601810
discussion
Discussion on downlink transmission scheme(s) for downlink control channels
From Spreadtrum
Spreadtrum's prior position on
10.5.2.1
at
RAN1#124
· AI-synthesized, paraphrased
verify sources →
Proposes inheriting NR PDCCH physical structure (CCE/REG/REG bundle framework, ALs 1/2/4/8/16, time-first REG mapping, hash-based search space) as baseline for 6GR. Prefers one port precoder-cycling between REG bundles for transmit diversity over SFBC, arguing comparable performance with simpler implementation. Proposes studying longer CORESET duration (4/6/12 symbols) to address capacity in hotspot areas and coverage for narrow-bandwidth UEs. Requires a unified framework for sub-slot/slot/slot group-level PDCCH monitoring to replace three separate monitoring capability types from NR, reducing UE and gNB implementation complexity. Proposes studying a harmonized PDCCH monitoring adaptation mechanism, criticizing NR's overlapping functionalities (PDCCH skipping, SSSG switching, DCI format 2_6, PEI, LP-WUS) and suggesting alternatives like two-stage DCI or dynamic candidate indication by LP-WUS.
Summary
This Spreadtrum contribution to the 3GPP 6G study item provides a comprehensive framework for the 6G-R (6GR) downlink control channel (PDCCH) design, largely reusing NR as a baseline. It contains 45 total proposals and 4 observations covering general PDCCH structure, transmission parameters, CORESET design, search space configuration, blind decoding limits, DMRS, transmission schemes, and multi-TRP operation.
Position
Spreadtrum proposes largely inheriting the NR PDCCH structure (CORESET, CCE, REG, REG bundles, hash functions) as the baseline for 6GR, requiring REG to consist of one RB in one OFDM symbol to avoid resource fragmentation. They require supporting only QPSK modulation for 1-stage DCI due to robustness and BD complexity concerns, while proposing further study of higher-order modulations only for 2nd-stage DCI in a 2-stage DCI structure. They propose studying longer CORESET durations beyond NR's 3-symbol limit to address capacity and coverage for IoT devices with narrow bandwidths, and support time-first REG indexing within CORESETs. Spreadtrum requires defining upper limits on monitored PDCCH candidates and non-overlapped CCEs per slot, applied to all device types, and supports only slot-based monitoring capability while rejecting span-based and slot-group-based monitoring to reduce implementation complexity. They propose transparent MU-MIMO transmission via overlapping CORESET resources with different DMRS scrambling IDs, and require studying PDCCH repetition for reliability enhancement as well as transparent or non-transparent time-frequency resource sharing between 5G and 6G CORESETs for MRSS coexistence.
Key proposals
- Proposal 1 (Sec 2.1.1): Support the NR-derived building blocks for 6GR PDCCH including CORESET, CCE, REG, time-first REG indexing, REG bundles, interleaved/non-interleaved CCE-to-REG mapping, hash-function-based search spaces, and blind decoding principles.
- Proposal 2 (Sec 2.1.2): For 6G PDCCH transmission, only support QPSK modulation for 1-stage DCI.
- Proposal 6 (Sec 2.1.3): Study transparent or non-transparent time-frequency regions sharing for 5G and 6G CORESETs.
- Proposal 8 (Sec 2.2.1): Study longer CORESET duration in 6GR beyond NR's 3-symbol maximum.
- Proposal 10 (Sec 2.2.2): For a CORESET overlapping the SBFD DL/UL boundary, it is up to gNB to avoid PDCCH candidate overlap with the boundary in SBFD symbols.
- Proposal 12 (Sec 2.3.1): Consider hash function based random search space structure as starting point.
- Proposal 13 (Sec 2.3.2.1): Define the upper limit of the number of monitored PDCCH candidates / non-overlapped CCEs per slot for a serving cell in 6GR.
- Proposal 16 (Sec 2.3.2.1): Support slot based PDCCH monitoring, rejecting span-based and slot-group-based monitoring capabilities to reduce UE and gNB implementation complexity.
- Proposal 17 (Sec 2.3.2.2): Consider NR PDCCH candidate dropping rules as starting point, keeping BD number for SCell and CSSs in PCell, and mapping USS candidates with lower ID before higher ID, dropping subsequent SS sets when capacity is exceeded.
- Proposal 19 (Sec 2.4): For 6GR PDCCH DMRS design, consider NR's length-31 Gold sequence, single antenna port with 3 DMRS per RB, and precoder granularity equal to REG-bundle or contiguous RBs in the CORESET as starting point.
- Proposal 20 (Sec 2.5.1): For 6GR PDCCH transmit diversity, consider one port precoder-cycling between REG bundles as starting point.
- Proposal 21 (Sec 2.5.2): Support transparent MU-MIMO transmission for 6GR PDCCH.
- Proposal 22 (Sec 2.5.3): Study PDCCH repetition to improve reliability in 6GR.
- Proposal 23 (Sec 2.6): Multi-DCI/single-DCI based scheme can be a starting point for M-TRP transmission.
- Proposal 33 (Sec 2.3.1): A search space is associated with a single CORESET for 6G PDCCH.