Content for  TS 38.300  Word version:  19.0.0

In Air-to-ground (ATG) network, an NG-RAN consisting of ground-based gNBs, which provide cell towers that send signals up to an aircraft's antenna(s) of onboard ATG terminal, with typical vertical altitude of around 10,000m and take-off/landing altitudes down to 3000m. DC is not supported for ATG in this release of the specification.

Support of AI/ML for NG-RAN, as a RAN function, is used to facilitate Artificial Intelligence (AI) and Machine Learning (ML) techniques in NG-RAN. The objective of AI/ML for NG-RAN is to improve network performance and user experience, through analysing the data collected and autonomously processed by the NG-RAN, which can yield further insights, e.g., for Network Energy Saving, Load Balancing, Mobility Optimization, Network Slicing, CCO.

Support of AI/ML for NG-RAN requires inputs from neighbour NG-RAN nodes (e.g., predicted information, feedback information, measurements) and/or UEs (e.g., measurement results). Signalling procedures used for the exchange of information to support AI/ML for NG-RAN, are use case and data type agnostic, which means that the intended usage (e.g., input, output, feedback) of the data exchanged via these procedures is not indicated. AI/ML algorithms and models are out of 3GPP scope. Model-specific performance information, e.g. model performance indicators specified in clause 6 of TS 28.105, is not exchanged over NG-RAN interfaces in TS 38.401. Support of AI/ML for NG-RAN does not apply to ng-eNB. For the deployment of AI/ML for NG-RAN the following scenarios may be supported:

• AI/ML Model Training is located in the OAM and AI/ML Model Inference is located in the NG-RAN node;
• AI/ML Model Training and AI/ML Model Inference are both located in the NG-RAN node.

AI/ML Model Training follows the definition of the "ML model training" as specified in clause 3.1 of TS 28.105. An AI/ML Model needs to be trained, validated and tested before deployment for AI/ML Model Inference. AI/ML Model Inference follows the definition of the "AI/ML inference" as defined in clause 3.1 of TS 28.105.

The following information can be configured to be reported by an NG-RAN node:

• Predicted resource status information, including predicted radio resource status per slice;
• UE performance feedback, including per-slice UE performance;
• Measured UE trajectory;
• Energy Cost (EC);
• Predicted slice available capacity.

The collection and reporting of the above information are configured through the Data Collection Reporting Initiation procedure, while the actual reporting is performed through the Data Collection Reporting procedure. The collection of measured UE trajectory and UE performance feedback is triggered at successful Handover. Cell-based UE trajectory prediction, which can be used, e.g., for the Mobility Optimization use case, is transferred to the target NG-RAN node via the Handover Preparation procedure to provide information for, e.g., subsequent mobility decisions. Cell-based UE trajectory prediction is limited to the first-hop target NG-RAN node. An NG-RAN node may derive predicted CCO issues and the corresponding future coverage states for its affected cells and beams. The NG-RAN node may notify its neighbour NG-RAN nodes, via the NG-RAN Node Configuration Update procedure, about the future coverage state changes together with modification cause and time information. The NG-RAN node may also notify its neighbour NG-RAN nodes that previously notified coverage state changes together with the corresponding modification cause have been cancelled. At any given point in time, for a list of predicted affected cells and beams, there is only one predicted CCO issue generated by a gNB.

OAM configures the following for EC reporting:

• The minimum and maximum energy consumption values corresponding to the minimum and maximum EC index values respectively, based on an implementation-specific mapping rule, which is unified within a defined area; and
• The recommended time interval within which an NG-RAN node selects an implementation-specific time window for averaging of the measurements of the NG-RAN node's consumed energy.

OAM further configures management-based MDT for continuous MDT towards the NG-RAN. OAM provides specific information to the NG-RAN for the NG-RAN to identify continuous MDT.