Content for TS 38.300 Word version: 18.3.0

1… 4… 4.7… 5… 5.3… 5.4… 6… 6.2… 6.6… 7… 8… 9… 9.2.2… 9.2.2.5… 9.2.3… 9.2.3.2… 9.2.3.3… 9.2.4… 9.2.6… 9.3… 10… 11… 15… 15.5… 16… 16.2… 16.3… 16.4… 16.8… 16.9… 16.10… 16.12… 16.12.5… 16.12.6… 16.12.6.3 16.12.7 16.13… 16.14… 16.15… 16.18… 16.19… 16.21… 16.21.3… 17… 18… 19 20… 21… A… B… C… G…

16.19 Support for Air to Ground Networks 16.19.1 Overview 16.19.2 Timing and Synchronization 16.19.2.1 Scheduling and Timing 16.19.2.2 Timing Advance 16.19.3 Mobility 16.19.3.1 Mobility in RRC_IDLE and RRC_INACTIVE 16.19.3.2 Mobility in RRC_CONNECTED 16.19.3.2.1 Handover 16.19.3.2.2 Conditional Handover 16.20 Support of AI/ML for NG-RAN 16.20.1 General 16.20.2 Principles 16.20.3 Data Collection and Reporting
...

16.19 Support for Air to Ground Networks |R18| p. 221

16.19.1 Overview p. 221

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.

CA and DC are not supported for ATG in this release of the specification.

16.19.2 Timing and Synchronization p. 221

16.19.2.1 Scheduling and Timing p. 221

To accommodate the propagation delay in ATG, the timing relationship is enhanced by one cell-specific offset K_offset:

K_offset is a configured scheduling offset that needs to be larger or equal to the max RTT.

The scheduling offset K_offset is used to allow the UE sufficient processing time between a downlink reception and an uplink transmission, see TS 38.213.

16.19.2.2 Timing Advance p. 221

For the serving cell, the network broadcasts coarse gNB location information. The UE shall have valid GNSS position as well as gNB location before connecting to an ATG cell. To achieve synchronisation, before and during connection to an ATG cell, the UE shall compute the RTT between UE and the gNB based on the GNSS position and the gNB location parameters, and autonomously pre-compensate the TTA for the RTT between the UE and the gNB.

In connected mode, the UE shall be able to continuously update the Timing Advance.

The UE may be configured to report Timing Advance during Random Access procedures or in connected mode. For an RRC_CONNECTED UE, event-triggered reporting of the Timing Advance is supported.

16.19.3 Mobility p. 221

16.19.3.1 Mobility in RRC_IDLE and RRC_INACTIVE p. 221

The same principles as described in clause 9.2.1 apply to mobility in RRC_IDLE for ATG and the same principles as described in clause 9.2.2 apply to mobility in RRC_INACTIVE for ATG unless hereunder specified. The gNB reference location of the serving cell and neighbour cell(s) provided in SIB22 can be used for the UE to perform the idle/inactive measurement and mobility.

The UE can determine whether a network is for ATG connectivity implicitly by the existence of cellBarredATG in SIB1.

16.19.3.2 Mobility in RRC_CONNECTED p. 221

16.19.3.2.1 Handover p. 221

The same principle as described in clause 9.2.3.2 applies to ATG unless hereunder specified.

16.19.3.2.2 Conditional Handover p. 221

The same principle as described in clause 9.2.3.4 applies to ATG unless hereunder specified.

ATG supports the following additional trigger conditions upon which UE may execute CHO to a candidate cell, as defined in TS 38.331:

The RRM measurement-based event A4;
A location-based trigger condition.

A location-based trigger condition is always configured together with one of the measurement-based trigger conditions (CHO events A3/A4/A5) as defined in TS 38.331.

16.20 Support of AI/ML for NG-RAN |R18| p. 222

16.20.1 General p. 222

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.

16.20.2 Principles p. 222

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, and the details of model performance feedback are also out of 3GPP scope.

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.

AI/ML Model Inference follows the definition of the "AI/ML inference" as defined in clause 3.1 of TS 28.105.

16.20.3 Data Collection and Reporting p. 222

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

Predicted resource status information;
UE performance feedback;
Measured UE trajectory;
Energy Cost (EC).

The collection and reporting 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.