R1-2410718
discussion
Summary #5 of specification support for positioning accuracy enhancement
From Ericsson
Ericsson's prior position on
9.1.2
at
RAN1#118bis
· AI-synthesized, paraphrased
verify sources →
Advocates for sample-based measurements over legacy path-based measurements for superior performance and lower complexity, while strongly opposing phase information inclusion in model inputs due to deployment costs and minimal accuracy gains.
Summary
This RAN1 document from Ericsson presents 95 proposals across 6 major technical areas for AI/ML-based positioning enhancement in NR, covering model input definitions, model output specifications, training data collection, inference procedures, and performance monitoring frameworks.
Position
Ericsson advocates FOR sample-based measurements as the primary approach for AI/ML positioning input (supporting majority view over compromise approaches), FOR reusing existing legacy signaling frameworks and IEs where possible to minimize specification impact, and FOR label-free monitoring methods with UE-side metric calculation. They push AGAINST supporting CIR/phase information for model input due to implementation complexity, AGAINST mandatory reporting of both sample-based and path-based measurements (preferring sample-based only), and AGAINST overly complex new signaling when legacy mechanisms can be enhanced.
Key proposals
- Proposal 2.1.1.2-1 (Sec 1.1): For sample-based measurement, the starting time of Nt consecutive samples is the timing of the first detected sample within a search window, with the search window determined by configured offset relative to reference time, reusing existing 'Search Window Information' IE in 38.455
- Proposal 2.1.3.2-1B (Sec 1.3): Support Alternative (a) sample-based measurement report only for Rel-19 AI/ML based positioning (majority view over supporting both sample-based and path-based)
- Proposal 3.1.2-1A (Sec 2.1): For AI/ML assisted positioning Case 3a, LOS/NLOS indicator reuses meaning from TS 38.455 providing likelihood of line-of-sight propagation path, with existing IE 'LoS/NLoS Information' format supporting soft/hard indicators
- Proposal 3.1.5-1 (Sec 2.2): For AI/ML assisted positioning Case 3a measurement reports, LMF shall be able to distinguish whether timing information is obtained by legacy method or Rel-19 AI/ML
- Proposal 4.1.3-2A (Sec 3.1.1): For AI/ML based positioning Case 1, time stamp includes both NR-TimeStamp and UTCTime from TS 37.355, with UTCTime as optional field present at least when SFN wraps around
- Proposal 4.1.4-2 (Sec 3.1.1): For AI/ML based positioning, when label is UE location, reuse existing IE using geographic shapes from TS 23.032 where uncertainty and confidence serve as quality indicator
- Proposal 5.1.7-1B (Sec 4): For AI/ML based positioning Case 1, all assistance information from legacy UE-based DL-TDOA except info #7 can be provided from LMF to UE, with four alternatives for handling geographical coordinates of TRPs
- Proposal 6.1.2-1 (Sec 6.1): Support label-free monitoring methods where model inference entity performs self-monitoring without external ground truth information, with UE/gNB sending monitoring decision to LMF when model becomes inappropriate
- Proposal 6.2.5-3 (Sec 6.2): For AI/ML positioning Case 1 label-based monitoring, support at least Option A-1 and A-2 with legacy framework/measurements/signaling reuse, monitoring metric definition up to UE implementation
- Conclusion 2.4.2-1 (Sec 1.6): No consensus in RAN1 to support CIR (Channel Impulse Response) for determining model input in Rel-19 AI/ML based positioning
- Conclusion 2.4.2-2 (Sec 1.6): No consensus to support using one phase value for first path or first sample only as model input for AI/ML based positioning
- Agreement (Sec 4): For sample-based measurement definition, starting time of Nt consecutive samples determined as first detected path rounded down with timing granularity T for gNB/TRP measurement
- Agreement (Sec 6.1): For AI/ML positioning Case 1 model performance monitoring, target UE side performs monitoring metric calculation and may signal monitoring outcome to LMF
- Proposal 2.1.3.5-1A (Sec 1.3): For Case 3b, additionally support enhanced measurement with Nt' values selected from Nt consecutive channel response values, with candidate sets Nt'≤24, Nt={32,64,128}, timing granularity T=2^k×Tc
- Proposal 2.1.1.2-2 (Sec 1.1): Sample-based measurement candidate values include Nt≥{16,32,64,128}, Nt'≥{9,16,32} where Nt'≤Nt, and timing reporting granularity factor k≥{0...5}