R1-2409443
discussion
AI/ML 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
Ericsson presents a comprehensive technical case for Rel-19 AI/ML-based positioning, strongly favoring sample-based measurements over legacy path-based reporting due to lower complexity and better generalization across different channel estimators. The document contains 73 proposals and 58 observations, arguing against the inclusion of phase information (CIR) as model input due to signaling overhead and alignment difficulties, while proposing specific parameter ranges for sample-based reporting and defining consistency mechanisms via associated IDs.
Position
Ericsson argues inapplicability of Rel-18 carrier phase positioning for AI/ML inputs, proposing to down-prioritize CIR model inputs due to high signaling overhead and difficulty aligning phase measurements between training and inference. They present a technical case against path-based measurements, demonstrating that sample-based measurements are robust to channel estimation algorithm mismatches and require lower receiver complexity. Ericsson requires the use of total-power PDP inputs summed over all receive antenna ports to balance accuracy and signaling size. They propose using an associated ID to verify consistency of network-side additional conditions between training and inference phases, rather than explicit signaling of all parameters. For model monitoring, they support self-monitoring by the model inference entity as a baseline, utilizing label-free methods or opportunistic LoS links for label-based monitoring.
Key proposals
- Proposal 1 (Sec 2.1): RAN1 assumes Case 1 is supported using UE based DL-TDOA procedures as a foundation.
- Proposal 2 (Sec 2.2): RAN1 assumes Case 3a is supported using UL-TDOA or multi RTT procedure as a foundation.
- Proposal 3 (Sec 2.3): RAN1 assumes Case 3b with sample-based reporting should be supported using a separate procedure from legacy UL RTOA and gNB RxTxTimeDiff reporting.
- Proposal 4 (Sec 3.1): For Rel-19 AI/ML based positioning Case 3b/2b, the measurement report of (PDP, DP) support sample-based measurement.
- Proposal 10 (Sec 3.1.1.5): For the model input types for Case 3b and 3a, consider input based on DP or PDP samples containing sample powers summed over all receive antenna ports, i.e., total-power PDP.
- Proposal 12 (Sec 3.2): Do not support phase information for determining model input, including CIR and single phase value for first path/sample.
- Proposal 16 (Sec 3.2.4): RAN1 to down-prioritize the signaling approach(es) and/or measurement definitions to support CIR model input types for Case 3b and Case 2b.
- Proposal 22 (Sec 4.1.1): Support reporting (optionally) a LOS/NLOS indicator when timing information is reported for Rel-19 AI/ML positioning Case 3a.
- Proposal 32 (Sec 4.1.2): For AI/ML assisted positioning at gNB (Case 3a), the model output is uplink relative time of arrival (TUL-RTOA).
- Proposal 44 (Sec 5.1.2): For training data collection Part B, support only ground truth labels in the format of UE location coordinates (x, y, z).
- Proposal 55 (Sec 6.1): For AI/ML based positioning, support self-monitoring by model inference entity as a baseline.
- Proposal 67 (Sec 7.2.3): For Case 1, to ensure the consistency of NW-side additional condition across training and inference, support associated ID for indicating NW conditions/configurations.
- Proposal 70 (Sec 7.2.4): For measurements reported to the LMF in case 2b/3b, support the LMF to indicate the PRS/SRS measurement bandwidth.