R1-2410204
discussion
Discussion on AIML beam management
From TCL
Summary
TCL proposes comprehensive enhancements to 3GPP NR beam management by integrating AI/ML techniques to simplify conventional P1/P2/P3 beam pairing processes and improve beam failure detection/recovery procedures. The document contains 10 proposals and 4 observations covering beam prediction, reference signal optimization, TCI framework enhancements, and reporting mechanisms for AI/ML-enabled beam management.
Position
TCL advocates for a comprehensive AI/ML integration into NR beam management that goes beyond basic beam prediction to include unified BFD/BFR frameworks, enhanced TCI signaling, and sophisticated reporting mechanisms. They push for significant protocol changes including new TCI state IDs, extended QCL types, and flexible quantization schemes, positioning themselves as supporters of deep AI/ML integration rather than minimal-impact approaches. TCL specifically promotes unified frameworks over separate legacy and AI/ML procedures.
Key proposals
- Proposal 1 (Sec 2.1): Conventional P1/P2/P3 approach to beam pairing/refinement can be simplified by AI/ML beam prediction to two phases, with potential for single-phase operation if AI/ML model can predict both DL-Tx and Rx beam pairs in P1
- Proposal 2 (Sec 2.2): Study to use AI/ML based beam prediction to maintain the candidate beam list for BFR (Beam Failure Recovery)
- Proposal 3 (Sec 2.2): Study to integrate the RS configuration and measurement configuration of BFD and BFR under a unified AI/ML based framework
- Proposal 4 (Sec 2.2): Study to evolve the BFD to BF event prediction while using temporal beam prediction to find the candidate beams for BFR
- Proposal 5 (Sec 2.3): Measurement patterns for AI/ML beam management should be designed for spatial domain beam prediction, and should consider both the fixed regular and random patterns
- Proposal 6 (Sec 2.4): Enhancement on the TCI framework should include additional TCI state ID dedicated for AI/ML BP, new QCL types for Set A and Set B beam association, and timing information for BM-Case2 predicted beams
- Proposal 7 (Sec 2.5): Enhancement on beam management report should indicate model type and/or bitmap/report type to indicate selected report quantities
- Proposal 8 (Sec 2.5): The report for AI/ML BP may include L1-RSRP and/or post processed RSRP
- Proposal 9 (Sec 2.5): RAN1 should consider enhancement on AI/ML beam management report including quantization of report quantities starting with RSRP quantization and two-stage report mechanism using both PUCCH and PUSCH
- Proposal 10 (Sec 2.5): For BM-Case2, overhead reduction approaches including splitting reports into multiple groups and indicating predicted beam selection via reference beam plus bitmap within neighborhood