R1-2500546
discussion
AI/ML based Positioning
From Google
Google's prior position on
9.1.2
at
RAN1#119
· AI-synthesized, paraphrased
verify sources →
Advocates for transparent model performance monitoring that does not require UEs to disclose location information to the network, supporting UE-side monitoring (Option A) over network-side monitoring. Pushes for flexible channel measurement options including phase information and configurable sample-based versus path-based reporting. Supports reuse of existing measurement frameworks rather than creating new mechanisms.
Summary
Google presents five proposals for 3GPP RAN1 regarding AI/ML-based positioning, focusing on extending enhanced path-based measurements to the UE side, reporting L1-SINR to handle measurement errors, and simplifying model monitoring. The document argues for explicit provision of TRP geographical coordinates (Info #7) and opposes redundant network-side monitoring options to reduce overhead and privacy risks.
Position
Google proposes extending enhanced path-based measurement to the UE side to reduce reporting overhead, arguing that UE-side path selection is simpler than gNB-side selection. They support reporting L1-SINR alongside path-based measurements to enable the network to filter results or improve accuracy based on potential measurement errors. For model performance monitoring in Case 1, Google supports Option A with a simple 1-bit failure indication based on UE-calculated metrics and NW-configured thresholds, while explicitly opposing Option B due to functional redundancy and privacy concerns regarding UE location disclosure. Regarding assistance information, Google supports Alternative 4, requiring Info #7 (TRP geographical coordinates) to be provided explicitly from the LMF to the UE to enable absolute location identification, arguing that implicit provision via associated ID would unnecessarily increase dimensionality.
Key proposals
- Proposal 1 (Enhanced path-based measurement): Support extending enhanced path-based measurement to UE-side measurement, where the UE performs path selection to reduce reporting overhead.
- Proposal 2 (Enhanced path-based measurement): Support reporting L1-SINR in addition to path-based measurement to allow the network to assess potential measurement error and decide on further processing.
- Proposal 3 (Model monitoring): For Option A based model performance monitoring for Case 1, support a 1-bit indication on whether a performance failure is detected, with the UE calculating the metric against a NW-configured threshold.
- Proposal 4 (Model monitoring): Do not support Option B based model performance monitoring for Case 1, citing redundancy with Option A and potential privacy risks regarding UE location disclosure.
- Proposal 5 (Assistance information): Support Alternative 4 where Info #7 (geographical coordinates of TRPs) is provided explicitly from the LMF to the UE, rather than implicitly via associated ID.