JP-WO2025033515-A5 - Communication control method and user device, mobile communication system, program, and chipset

Dates

Publication Date

20260511

Application Date

20240808

Description

This disclosure relates to a communication control method and user equipment, a mobile communication system, a program, and a chipset. Figure 13 is a diagram showing another configuration example of the "CSI feedback improvement" according to the first embodiment. In Figure 13, the data collection unit A1, the model learning unit A2, the model inference unit A3, and the data processing unit A4 are included in the gNB 200. That is, Figure 14 is an example in which model learning and model inference are performed in the gNB 200. In Figure 14, the transmitting entity TE is the gNB 200 and the receiving entity RE is the UE 100. In step S24, the transmitting entity TE may transmit location information to the receiving entity RE. The gNB200 may transmit a control message containing location information to the UE100. The transmitting entity TE, i.e., the LMF400, may transmit an LPP message containing location information to the UE100. The transmitting entity TE may use the location information itself. In step S32, the LMF 400 transmits the trained model to the UE100. The UE100 receives the trained model. In step S33, the LMF 400 transmits the learning record data to the UE 100. The LMF 400 may also transmit information to the UE 100 instructing the UE to retrain the learning model if it determines that the input data was not used for model training of the learning model (hereinafter sometimes referred to as "model retraining instruction information"). The LMF 400 may transmit the learning record data and the model retraining instruction information in a single message. The UE 100 receives at least the learning record data. In step S34, the LMF 400 sends information representing an instruction to confirm the learning record data (hereinafter sometimes referred to as "learning record data confirmation instruction information") to the UE 100. The LMF 400 may also send the learning record data, the model retraining instruction information, and the learning record data confirmation instruction information in a single message. The UE 100 receives the learning record data confirmation instruction information. In step S463, in response to receiving the fallback request information, the LMF400 transmits information to the UE100 instructing it to perform a fallback (hereinafter sometimes referred to as "fallback instruction information"), and also transmits information to the UE100 instructing it to perform model training while the fallback is in progress (hereinafter sometimes referred to as "training start instruction information"). The reason for instructing the UE100 to start model training during the fallback is that the UE100 is located in a place where it has never performed training before (NO in step S35), and by having it perform model training in that location, the aim is to resume use with a newly trained model (described later). The fallback instruction information may also include instructions for deactivating the trained model and for starting the use of legacy operation. In step S867, the LMF400 may send a learning start instruction to the UE100, instructing it to perform model learning while fallback is running.