Communication method

Abstract

A communication method according to one aspect of the present invention is used in a mobile communication system. The communication method includes a step in which, regarding an AI / ML model capable of estimating a measurement value for a first beam included in a first set from a measurement value for a second beam included in a second set, a user device detects that the AI / ML model is unavailable. In addition, the communication method includes a step in which, in response to satisfying a prescribed condition, the user device transmits, to a network node, a second measurement value for the second beam received after the AI / ML model becomes unavailable, or a first measurement value for the first beam estimated using the AI / ML model immediately before the AI / ML model becomes unavailable.

Description

Communication method

[0001] The present disclosure relates to a communication method used in a mobile communication system.

[0002] In recent years, studies have been conducted in 3GPP (Third Generation Partnership Project) (registered trademark, the same applies hereinafter), a standardization project for mobile communication systems, to apply artificial intelligence (AI) technology, particularly machine learning (ML) technology, to wireless communication (air interface) of mobile communication systems.

[0003] 3GPP TR 38.843 V18.0.0 (2023-12) 3GPP TS 38.000 V18.4.0 (2024-12) 3GPP TS 28.105 V19.1.0 (2024-12)R2-2500997, “LCM for UE-side Beam Management”, 3GPP TSG-RAN WG2 Meeting #129, Athens, Greece, 17 - 21 February 2025

[0004] The communication method according to the first aspect is a communication method in a mobile communication system. The communication method comprises the step of: detecting, by a user equipment, that an AI (Artificial Intelligence) / ML (Machine Learning) model capable of inferring measurement values for first beams included in a first set from measurement values for second beams included in a second set cannot be used. The communication method further comprises the step of: transmitting, by the user equipment to a network node, either a second measurement value for a second beam received after the AI / ML model becomes unavailable, or a first measurement value for a first beam inferred using the AI / ML model immediately before the AI / ML model becomes unavailable, in accordance with satisfaction of a predetermined condition.

[0005] The second aspect of the communication method is a communication method in a mobile communication system. The communication method includes the step of a user device receiving configuration information from a network node. The communication method also includes the step of the user device detecting that it is not possible to use an AI / ML model capable of inferring measurement values ​​for a first beam included in a first set from measurement values ​​for a second beam included in a second set. Furthermore, the communication method includes the step of the user device transmitting to the network node, in accordance with the configuration information, at least one of the second measurement values ​​for the second beam received after it became impossible to use the AI / ML model, and the first measurement values ​​for the first beam inferred using the AI / ML model immediately before it became impossible to use the AI / ML model.

[0006] Figure 1 is a diagram showing an example configuration of a mobile communication system according to the first embodiment. Figure 2 is a diagram showing an example configuration of a UE (User Equipment) according to the first embodiment. Figure 3 is a diagram showing an example configuration of a network node (base station) according to the first embodiment. Figure 4 is a diagram showing an example configuration of a protocol stack according to the first embodiment. Figure 5 is a diagram showing an example configuration of a protocol stack according to the first embodiment. Figure 6 is a diagram showing an example configuration of a functional block of AI / ML technology according to the first embodiment. Figure 7(A) is a diagram showing an example configuration of a functional block of a mobile communication system according to the first embodiment, and Figure 7(B) is a diagram showing an example configuration of a functional block of a UE according to the first embodiment. Figure 8 is a diagram showing an example configuration of a functional block of a mobile communication system according to the first embodiment. Figures 9(A) and 9(B) are diagrams showing an example configuration of a functional block of a mobile communication system according to the first embodiment. Figures 10(A) and 10(B) are diagrams showing an example configuration of a functional block of a mobile communication system according to the first embodiment. Figure 11 is a diagram for explaining the problems according to the first embodiment. Figure 12 is a diagram for explaining the problems according to the first embodiment. Figure 13 is a diagram showing an example of operation according to the first embodiment. Figures 14(A) and 14(B) are diagrams showing examples of beams according to the first embodiment. Figure 15 is a diagram showing an example of operation according to the first embodiment. Figure 16 is a diagram showing an example of operation according to the first embodiment. Figure 17 is a diagram showing an example of operation according to the first embodiment.

[0007] This disclosure aims to provide a communication method that enables a user device to transmit useful information to a network node.

[0008] The mobile communication system according to the first embodiment will be described with reference to the drawings. In the drawings, identical or similar parts are denoted by the same or similar reference numerals.

[0009] [First Embodiment] The configuration of the mobile communication system according to the first embodiment will now be described. Figure 1 is a diagram showing an example of the configuration of the mobile communication system 1 according to the first embodiment. The mobile communication system 1 conforms to the 5th Generation System (5GS) of the 3GPP standard. In the following description, 5GS will be used as an example, but the mobile communication system may also have an LTE (Long Term Evolution) system applied to it at least partially. The mobile communication system may also have a 6th Generation (6G) system or later applied to it at least partially.

[0010] The mobile communication system 1 comprises a network (NW) 10 and a user device (UE) 100. The UE 100 is a mobile communication device that performs wireless communication with the NW 10. The UE 100 may be any device used by a user, such as a mobile phone terminal (including a smartphone) and / or a tablet terminal, a notebook PC (Personal Computer), a communication module (including a communication card or chipset), a sensor or a device installed on a sensor, a vehicle or a device installed on a vehicle (Vehicle UE), or an aircraft or a device installed on an aircraft (Aerial UE).

[0011] NW10 includes a radio access network (RAN) 20 and a core network (CN) 30. When the mobile communication system is a fifth-generation system (5GS), RAN20 is referred to as NG-RAN (Next Generation Radio Access Network) and CN30 is referred to as 5GC (5G Core Network).

[0012] RAN20 includes multiple network nodes 200 (network nodes 200a to 200c in the example in Figure 1). The network nodes 200 are interconnected via inter-network node interfaces. In RAN20, network nodes 200 are sometimes referred to as base stations. When a network node 200 is a base station, it consists of a CU (Central Unit) and a DU (Distribution Unit) (i.e., functionally divided), and the two units may be connected by a front-haul interface. When the mobile communication system 1 is 5GS, the network nodes 200 are referred to as gNBs, the inter-network node interfaces as Xn interfaces, and the front-haul interfaces as F1 interfaces.

[0013] Furthermore, if at least a part of the mobile communication system 1 is an LTE system, the network node 200 may be an eNB (evolved Node B) which is an LTE base station. Also, if the mobile communication system 1 is a sixth-generation system or later, the network node 200 has the function of a base station and may be a device equivalent to a gNB or eNB.

[0014] Each network node 200 manages one or more cells. Each network node 200 performs wireless communication with the UE 100 that has established a connection with its own cell. Each network node 200 has functions such as wireless resource management (RRM), routing of user data (also simply referred to as "data"), and measurement and control functions for mobility control and scheduling. The term "cell" is used to indicate the smallest unit of a wireless communication area. The term "cell" is also used to indicate a function or resource that performs wireless communication with the UE 100. One cell belongs to one carrier frequency. One cell may be associated with one downlink component carrier and one uplink component carrier. The bandwidth corresponding to one cell (system bandwidth) may be divided into multiple bandwidth parts (BWP: Bandwidth Part). In the following explanation, the gNB may be used as an example of a network node 200.

[0015] CN30 includes a CN (Core Network) device 300. The CN device 300 may include a C-plane device corresponding to the control plane (C-plane) and a U-plane device corresponding to the user plane (U-plane). The C-plane device performs various mobility controls and paging for the UE100. The C-plane device communicates with the UE100 using NAS (Non-Access Stratum) signaling. The U-plane device controls data transfer. When the mobile communication system is 5GS, the C-plane device is called AMF (Access and Mobility Management Function), the U-plane device is called UPF (User Plane Function), and the interface between the network node 200 and the CN device 300 is called the NG interface.

[0016] In the following, the network node 200 and the CN device 300 may be referred to as the network device. The network device may also be the network node 200. The network device may also be the CN device 300.

[0017] Figure 2 shows an example configuration of UE100 (user device) according to the first embodiment. UE100 comprises a receiving unit 110, a transmitting unit 120, and a control unit 130. The receiving unit 110 and the transmitting unit 120 constitute a communication unit 140 that performs wireless communication with the network node 200. UE100 is an example of a communication device.

[0018] The receiving unit 110 performs various types of reception under the control of the control unit 130. The receiving unit 110 includes an antenna and a receiver. The receiver converts the radio signal received by the antenna into a baseband signal (received signal) and outputs it to the control unit 130.

[0019] The transmitting unit 120 performs various types of transmissions under the control of the control unit 130. The transmitting unit 120 includes an antenna and a transmitter. The transmitter converts the baseband signal (transmission signal) output by the control unit 130 into a wireless signal and transmits it from the antenna.

[0020] The control unit 130 performs various control and processing operations in the UE 100. Such processing includes processing in each layer described later. The control unit 130 includes at least one processor and at least one memory. The memory stores programs executed by the processor and information used in processing by the processor. The processor may include a baseband processor and a CPU (Central Processing Unit). The baseband processor performs modulation, demodulation, encoding, and decoding of baseband signals. The CPU executes programs stored in memory and performs various processing operations. Note that processing or operations performed in the UE 100 may also be performed in the control unit 130.

[0021] Figure 3 shows an example configuration of a network node 200 according to the first embodiment. The network node 200 includes a transmitting unit 210, a receiving unit 220, a control unit 230, and an NW communication unit 240. The transmitting unit 210 and the receiving unit 220 constitute a communication unit 250 that performs wireless communication with the UE 100. The NW communication unit 240 constitutes a backhaul communication unit that communicates with the CN 30. The network node 200 is an example of a communication device.

[0022] The transmitting unit 210 performs various types of transmissions under the control of the control unit 230. The transmitting unit 210 includes an antenna and a transmitter. The transmitter converts the baseband signal (transmission signal) output by the control unit 230 into a wireless signal and transmits it from the antenna.

[0023] The receiving unit 220 performs various types of reception under the control of the control unit 230. The receiving unit 220 includes an antenna and a receiver. The receiver converts the radio signal received by the antenna into a baseband signal (received signal) and outputs it to the control unit 230.

[0024] The control unit 230 performs various control and processing operations on the network node 200. Such processing includes processing at each layer described later. The control unit 230 includes at least one processor and at least one memory. The memory stores programs executed by the processor and information used for processing by the processor. The processor may include a baseband processor and a CPU. The baseband processor performs modulation, demodulation, encoding, and decoding of baseband signals. The CPU executes programs stored in memory and performs various processing operations. Processing or operations performed at the network node 200 may also be performed by the control unit 230.

[0025] The NW communication unit 240 is connected to an adjacent network node via the Xn interface, which is an inter-network node interface. The NW communication unit 240 is connected to the CN device 300 via the NG interface, which is an inter-network node-core network interface. The network node 200 may consist of a central unit (CU) and a distributed unit (DU) (i.e., functionally divided), and the two units may be connected by the F1 interface, which is a front-haul interface.

[0026] Figure 4 shows an example of the protocol stack configuration for a user plane wireless interface that handles data.

[0027] The user plane radio interface protocol comprises a physical (PHY) layer, a media access control (MAC) layer, a radio link control (RLC) layer, a packet data convergence protocol (PDCP) layer, and a service data adaptation protocol (SDAP) layer.

[0028] The PHY layer performs encoding / decoding, modulation / demodulation, antenna mapping / demapping, and resource mapping / demapping. Data and control information are transmitted between the PHY layer of UE100 and the PHY layer of network node 200 via a physical channel. The PHY layer of UE100 receives downlink control information (DCI (Downlink Control Information)) transmitted from network node 200 over the physical downlink control channel (PDCCH (Physical Downlink Control Channel)). Specifically, UE100 performs blind decoding of the PDCCH using a Radio Network Temporary Identifier (RNTI (Radio Network Temporary Identifier)) and acquires the successfully decoded DCI as the DCI addressed to its own UE. The DCI transmitted from network node 200 has a CRC (Cyclic Redundancy Check) parity bit added, which has been scrambled by RNTI.

[0029] In NR, UE100 can use a bandwidth narrower than the system bandwidth (i.e., the cell bandwidth). Network node 200 configures UE100 with a bandwidth portion (BWP) consisting of consecutive PRBs (Physical Resource Blocks). UE100 sends and receives data and control signals in the active BWP. For example, up to four BWPs may be configured for UE100. Each BWP may have a different subcarrier spacing. The frequencies of these BWPs may overlap. If multiple BWPs are configured for UE100, network node 200 can specify which BWP to apply by controlling the downlink. This allows network node 200 to dynamically adjust the UE bandwidth according to the amount of data traffic on UE100, thereby reducing UE power consumption.

[0030] The network node 200 can configure, for example, up to three control resource sets (CORESETs) for each of up to four BWPs on a serving cell. A CORESET is a radio resource for control information that the UE 100 should receive. The UE 100 may have up to twelve or more CORESETs configured on a serving cell. Each CORESET may have an index from 0 to 11 or more. A CORESET may consist of six resource blocks (PRBs) and one, two, or three consecutive OFDM (Orthogonal Frequency Division Multiplexing) symbols in the time domain.

[0031] The MAC layer performs data priority control, retransmission processing using Hybrid ARQ (Hybrid Automatic Repeat reQuest), and random access procedures. Data and control information are transmitted between the MAC layer of UE100 and the MAC layer of network node 200 via a transport channel. The MAC layer of network node 200 includes a scheduler. The scheduler determines the transport format for the up and down links (transport block size, modulation and coding scheme (MCS)) and the resource blocks to be allocated to UE100.

[0032] The RLC layer transmits data to the receiving RLC layer using the functions of the MAC layer and PHY layer. Data and control information are transmitted between the RLC layer of UE100 and the RLC layer of network node 200 via a logical channel.

[0033] The PDCP layer performs header compression / decompression, encryption / decryption, etc.

[0034] The SDAP layer maps IP flows, which are the units under which the core network performs QoS (Quality of Service) control, to wireless bearers, which are the units under which the access layer (AS: Access Stratum) performs QoS control. Note that if the RAN is connected to the EPC (Evolved Packet Core), an SDAP is not required.

[0035] Figure 5 shows the configuration of the protocol stack of the wireless interface of the control plane that handles signaling (control signals).

[0036] The protocol stack of the control plane's radio interface includes a Radio Resource Control (RRC) layer and a Non-Access Stratum (NAS) layer, instead of the SDAP layer shown in Figure 4.

[0037] RRC signaling for various settings is transmitted between the RRC layer of UE100 and the RRC layer of network node 200. The RRC layer controls the logical channel, transport channel, and physical channel in response to the establishment, re-establishment, and release of the wireless bearer. If there is a connection (RRC connection) between the RRC of UE100 and the RRC of network node 200, UE100 is in the RRC connected state. If there is no connection (RRC connection) between the RRC of UE100 and the RRC of network node 200, UE100 is in the RRC idle state. If the connection between the RRC of UE100 and the RRC of network node 200 is suspended, UE100 is in the RRC inactive state.

[0038] The NAS, located above the RRC layer, handles session management and mobility management, among other things. NAS signaling is transmitted between the UE100's NAS and the AMF's NAS. In addition to the wireless interface protocol, the UE100 also has an application layer, etc. Furthermore, the layer below the NAS is called the AS (Access Stratum).

[0039] (AI / ML Technology) Next, the AI / ML (Artificial Intelligence / Machine Learning) technology according to the embodiment will be described. Figure 6 is a diagram showing an example of the configuration of the functional block of the AI / ML technology in the mobile communication system 1 according to the first embodiment.

[0040] The example of the functional block configuration shown in Figure 6 includes a Data Collection unit A1, a Model Training unit A2, a Model Inference unit A3, a Management unit A5, and a Model Storage unit A6.

[0041] The example functional block configuration shown in FIG. 6 represents a general functional framework of common AI / ML techniques. Therefore, depending on a hypothetical use case, some parts of the example functional block configuration (e.g., model recording unit A6, etc.) may not be included in the example functional block configuration. Further, the example functional block configuration shown in FIG. 6 may be distributed and disposed between the UE 100 and a network device. Alternatively, in the example functional block configuration, some functions (e.g., model learning unit A2 or model inference unit A3, etc.) may be disposed in both the UE 100 and the network device.

[0042] Data collection unit A1 provides input data to model learning unit A2, model inference unit A3, and management unit A5. The input data includes training data for the model learning unit A2, inference data for the model inference unit A3, and monitoring data for the management unit A5.

[0043] The training data is data required as input when an AI / ML model performs learning. The inference data is data required as input when an AI / ML model performs inference. Furthermore, the monitoring data is data required as input when managing an AI / ML model.

[0044] Note that Data collection may be, for example, a process of collecting data at a network node, a management entity, or the UE 100 for performing AI / ML model learning, AI / ML model management, and AI / ML model inference.

[0045] Model learning unit A2 performs AI / ML model training, AI / ML model validation, and AI / ML model testing.

[0046] AI / ML model learning refers to the process of training an AI / ML model based on input-output relationships to obtain a trained AI / ML model that is used for inference. For example, when considering y=ax+b, providing the input (x) and output (y) (that is, providing training data) and then optimizing a (the slope) and b (the intercept) may constitute AI / ML model learning.

[0047] Generally, machine learning includes supervised learning, unsupervised learning, and reinforcement learning. Supervised learning is a method that uses ground-truth data in training data. Unsupervised learning is a method that does not use ground-truth data in training data. For example, in unsupervised learning, feature points are learned from a large amount of training data, and correct determination (range estimation) is performed. Reinforcement learning is a method that assigns a score to an output result and learns a method to maximize the score. Although supervised learning is described below, unsupervised learning may alternatively be applied as machine learning. Reinforcement learning may alternatively be applied as the machine learning.

[0048] The model learning unit A2 outputs the trained AI / ML model (Trained Model) obtained by AI / ML model learning to the model storage unit A6, and the model learning unit A2 also outputs the updated AI / ML model (Updated Model) obtained by retraining the trained AI / ML model to the model storage unit A6.

[0049] Note that hereinafter, AI / ML model learning may be referred to as "model learning" or "learning".

[0050] The model inference unit A3 performs AI / ML model inference. Specifically, the model inference unit A3 applies the inference data provided by the data collection unit A1 to a trained AI / ML model (or an updated AI / ML model) to obtain inference output data. For example, considering y = ax + b, x is the inference data and y is the inference output data. Note that "y = ax + b" is an AI / ML model. A model with optimized slope and intercept, for example "y = 5x + 3", is a trained AI / ML model. Here, there are various modeling methods (approaches), including linear regression analysis, neural networks, and decision tree analysis. The above "y = ax + b" can also be considered a type of linear regression analysis.

[0051] The model inference unit A3 outputs inference output data to the management unit A5. The model inference unit A3 also receives management instructions from the management unit A5. For example, management instructions include selecting an AI / ML model, activating (deactivating) an AI / ML model, switching between AI / ML models, and fallback (performing inference without using an AI / ML model). The model inference unit A3 performs model inference according to the management instructions.

[0052] AI / ML model inference refers to the process of obtaining a set of outputs from a set of inputs using, for example, a trained AI / ML model (or an updated AI / ML model). Alternatively, model inference may be the process of obtaining inference output data from inference data using a trained AI / ML model (or an updated AI / ML model). Hereafter, AI / ML model inference may be referred to as "model inference" or "inference."

[0053] In the following text, an AI / ML model that is being trained (or updated) may be referred to as a "Training AI / ML model" (or "Updating AI / ML Model"). Also, in the following text, if there is no distinction between an AI / ML model being trained (or updated) and an AI / ML model that has been trained (or updated), it may simply be referred to as an "AI / ML model."

[0054] The management unit A5 supervises operations on the AI / ML model (selection, activation, deactivation, switching, fallback, etc.). The management unit A5 also supervises monitoring of the AI / ML model. Based on monitoring data and inference output data, the management unit A5 can also perform actions to ensure appropriate inference operation. Therefore, the management unit A5 outputs a Model Transfer / Delivery Request to the Model Recording Unit A6, and causes the trained (or updated) AI / ML model recorded in the Model Recording Unit A6 to output to the Model Inference Unit A3. Furthermore, the management unit A5 outputs management instructions to the Model Inference Unit A3, supervising operations on the AI / ML model. Furthermore, the management unit A5 can output performance feedback and retraining requests to the model learning unit A2, which can then retrain the AI / ML model (i.e., update the trained AI / ML model).

[0055] (Use Cases) Next, we will explain use cases in which AI / ML technology is applied. For example, there are three use cases in which AI / ML technology is applied:

[0056] (X1.1) "CSI (Channel State Information) Feedback Improvement"

[0057] (X1.2) "Beam management"

[0058] (X1.3) “Positioning accuracy enhancement”

[0059] (X1.1) "Improved CSI Feedback" "Improved CSI Feedback" describes a use case where AI / ML technology is applied to the CSI that is fed back from UE100 to network node 200. CSI is information about the channel status in the downlink between UE100 and network node 200. CSI includes at least one of the following: Channel Quality Indicator (CQI), Precoding Matrix Indicator (PMI), and Rank Indicator (RI). Network node 200 performs, for example, downlink scheduling based on the CSI feedback from UE100.

[0060] In the "CSI Feedback Improvement" use case, there are two sub-use cases: CSI compression in the frequency domain and CSI prediction in the time domain.

[0061] (X1.1.1) Sub-use case: CSI compression In CSI compression, the CSI inferred using the trained AI / ML model in UE100 is compressed in UE100. The compressed CSI is transmitted from UE100 to network node 200.

[0062] Figure 7(A) is a diagram showing an example of the configuration of a functional block in the mobile communication system 1 when CSI compression is used. As shown in Figure 7(A), the UE 100 has a CSI generation unit 101, and the network node 200 has a CSI reconstruction unit 201.

[0063] The CSI generation unit 101 includes a CSI generation inference unit 1010 and a quantization unit 1011. The CSI generation inference unit 1010 infers CSI (inference output data) from the input using a trained AI / ML model. The input to the CSI generation inference unit 1010 may be, for example, a partial (or punctured) CSI. The partial CSI may be a CSI measured using a CSI reference signal (CSI-RS (Channel State Information-Reference Signal)) (or demodulation reference signal (DMRS)) transmitted using a certain amount of resources or less. Alternatively, the input to the CSI generation inference unit 1010 may be a CSI reference signal. The output of the CSI generation inference unit 1010 is a CSI with a larger number of CSIs than the partial CSI input, if a partial CSI is input. Also, the output of the CSI generation inference unit 1010 is a CSI if a CSI reference signal is input. Hereinafter, the output of the CSI generation inference unit 1010 will be referred to as the full CSI. The quantization unit 1011 quantizes the full CSI. Compression is performed through quantization. The quantization unit 1011 transmits the quantized full CSI to the network node 200 as CSI feedback.

[0064] The CSI reconstruction unit 201 includes an inverse quantization unit 2010 and a CSI reconstruction inference unit 2011. The inverse quantization unit 2010 receives CSI feedback, inverse quantizes the quantized full CSI, and outputs the full CSI. The CSI reconstruction inference unit 2011 uses a trained AI / ML model to infer the reconstructed full CSI from the output of the inverse quantization unit 2010 (full CSI). The reconstructed full CSI is output from the CSI reconstruction unit 201.

[0065] Furthermore, the quantization unit 1011 may be merged with the CSI generation inference unit 1010, and the inverse quantization unit 2010 may also be merged with the CSI reconstruction inference unit 2011. In addition, a preprocessing unit may be provided before the CSI generation inference unit 1010. A postprocessing unit may be provided after the CSI reconstruction inference unit 2011.

[0066] As shown in Figure 7(A), the sub-use case of CSI compression is based on a two-sided model in which inference is performed using a trained AI / ML model on both the UE100 side and the network node 200 side.

[0067] (X.1.1.2) Sub-use case: CSI prediction In CSI prediction, a trained AI / ML model is used to infer (predict) future CSIs from the history of past CSIs.

[0068] Figure 7(B) shows an example of the configuration of a functional block in UE100 when CSI prediction is used. CSI prediction is based on the UE-side model in UE100, where inference is performed using a pre-trained AI / ML model.

[0069] As shown in Figure 7(B), UE100 has a CSI prediction model 102. The CSI prediction model 102 has a trained AI / ML model that takes past CSI history as input and infers future CSI (predicted CSI). The CSI prediction model 102 may also be a CSI prediction inference unit. UE100 (CSI prediction model 102) transmits the predicted CSI as CSI feedback to the network node 200.

[0070] Furthermore, the CSI prediction model 102 may have a pre-processing unit before it. The CSI prediction model 102 may also have a post-processing unit after it.

[0071] (X1.2) "Beam Management" In beam management use cases, there are two sub-use cases: spatial-domain downlink beam prediction, which performs beam prediction in the spatial direction, and temporal downlink beam prediction, which performs beam prediction in the temporal direction. Spatial-domain downlink beam prediction is called "BM Case 1" (BM-Case 1), and temporal downlink beam prediction is called "BM Case 2" (BM-Case 2).

[0072] Figure 8 is a diagram showing an example of the configuration of a functional block in the mobile communication system 1 when BM case 1 is used. In the case of BM case 1, a UE side model in which inference is performed at UE 100 may be applied. A NW side model in which inference is performed on the network side may also be applied. Therefore, as shown in Figure 8, the AI / ML model 103 that performs inference may reside at UE 100. The AI / ML model 103 that performs inference may reside at NW 10 (a network node 200 or CN device 300 included in NW 10).

[0073] In BM Case 1, the input to the AI / ML model 103 is the measured value for each beam included in beamset B. On the other hand, the output from the AI / ML model 103 (inference output data) is the probability that each (downstream) beam included in (predicted) beamset A will be the top beam. Beamset A and beamset B may be different. Alternatively, beamset B may be a subset of beamset A. The measured value of beamset B, which is the input to the AI / ML model 103, may be expressed as RSRP (Reference Signal Received Power).

[0074] Figure 8 also shows an example configuration of the mobile communication system 1 in the case of BM case 2. In the case of BM case 2, the UE side model may also be applied. In the case of BM case 2, the NW side model may also be applied. In the case of BM case 2, the AI / ML model 103 is located in either UE 100 or NW 10.

[0075] In BM Case 2, the input to the AI / ML model 103 is the history of measurements for each beam included in beamset B. The previously measured measurements for each beam are input to the AI / ML model 103. On the other hand, the output from the AI / ML model 103 (inference output data) is the probability that each (downstream) beam included in (predicted) beamset A will be the top beam, similar to BM Case 1. Beamset A and beamset B may be different. Beamset B may be a subset of beamset A. Also, in BM Case 2, beamset A and beamset B may be the same.

[0076] In both BM Case 1 and BM Case 2, the UE side model allows UE100 to report prediction results to NW10. Also, in both BM Case 1 and BM Case 2, the NW side model can predict the top beam based on the measured values ​​for each beam included in beamset B reported by UE100.

[0077] (X1.3) "Positioning Accuracy Enhancement" In the use case for improving positioning accuracy, there are two sub-use cases: Direct AI / ML positioning, which directly infers the position of UE100 using a trained AI / ML model, and AI / ML assisted positioning, which infers intermediate position measurements. In the latter, AI / ML assisted positioning, the position of UE100 is measured or inferred in the LMF (Location Management Function) using intermediate position measurements. Intermediate position measurements can serve as assisting information for measuring or inferring the position of UE100 in the LMF. The LMF may have a pre-trained AI / ML model, and uses this pre-trained AI / ML model to infer the position of UE100.

[0078] (X1.3.1) Sub-use case: Direct AI / ML positioning Figure 9(A) is a diagram showing an example of the configuration of a functional block in the mobile communication system 1 when direct AI / ML positioning is used. In the case of direct AI / ML positioning, the UE side model and the network side model are applied. Therefore, the AI / ML model 104 used for inference may reside in the UE 100. The AI / ML model 104 used for inference may reside in the NW 10.

[0079] In the case of direct AI / ML positioning, the input to the AI / ML model 104 is measured values ​​at each measurement point (TRP: Transmission and / or Reception Point). These measured values ​​may include, for example, channel impulse response (CIR), power delay profile (PDP), or fingerprint. For instance, both CIR and PDP represent the delay time for a signal at a specific frequency, but CIR represents the instantaneous delay time, while PDP represents the statistical delay time. On the other hand, the output from the AI / ML model 104 (inference output data) is the position information of the UE 100. This position information may also be represented by a fingerprint. The fingerprint, for example, represents the measurement information for the cells of the UE 100.

[0080] (X1.3.2) Sub-use case: AI / ML assisted positioning Figures 9(B) to 10(B) are diagrams showing examples of the configuration of functional blocks in the mobile communication system 1 when AI / ML assisted positioning is used. In the case of AI / ML assisted positioning, the UE side model and the network side model are also applied. Therefore, the AI / ML model 105 used for inference may reside in the UE 100. The AI / ML model 105 used for inference may reside in the NW 10. In the case of AI / ML assisted positioning, there are cases where one AI / ML model 105 is used for multiple inputs (Figure 9(B)), where the same AI / ML model is used for each of the multiple inputs (Figure 10(A)), and where different AI / ML models are used for each of the multiple inputs (Figure 10(B)).

[0081] In either case, the input to the AI / ML model 105 is the channel measurement value at each measurement point (TRP). The channel measurement value may be CIR, PDP, or fingerprint, similar to direct AI / ML positioning. On the other hand, the output from the AI / ML model 105 (inference output data) is, in either case, an intermediate position measurement value used for position measurement. The intermediate position measurement value may be LOS or NLOS identification, measurement timing and / or measurement angle, or likelihood of measurement.

[0082] (LCM) Currently, 3GPP is discussing the Life Cycle Management (LCM) of the AI / ML model.

[0083] The LCM of an AI / ML model may specifically include at least one of the following operations: • (L1) Data collection; • (L2) Model training; • (L3) Identification; • (L4) Model delivery / transfer; • (L5) Model inference operation; • (L6) Selection, activation, deactivation, switching, and fallback operations; • (L7) Monitoring; • (L8) Model update (L9) UE capability.

[0084] On the network side, for example, by controlling each operation of the LCM, it becomes possible to properly manage everything from the generation of AI / ML models to their deletion (or disposal). Fallback refers to switching from an AI / ML model to a model that does not use an AI / ML model. A model that does not use an AI / ML model is sometimes called a "legacy model."

[0085] 3GPP specifies two types of LCM: function-based LCM and model ID-based LCM.

[0086] A function-based LCM may be an operation performed on a function of an AI / ML model. Specifically, a function-based LCM may perform one of the following operations on a function of an AI / ML model: activation, deactivation, switching, or fallback. The network side can, for example, use 3GPP signaling (RRC messages, MAC CE, or DCI) to instruct the LCM operation on a function of an AI / ML model. In a function-based LCM, UE100 may have one AI / ML model for one function. UE100 may have multiple AI / ML models for a single function.

[0087] On the other hand, a model ID-based LCM may be an operation performed on individual AI / ML models using the model ID. The model ID is identification information used to distinguish one AI / ML model from other AI / ML models. Specifically, a model ID-based LCM may use the model ID to perform one of the following operations on each AI / ML model: activation, deactivation, switching, or selection.

[0088] (Problems of the First Embodiment) Next, the problems of the first embodiment will be described.

[0089] As mentioned above, there are two use cases for beam management: "BM-Case 1," which performs beam prediction in the spatial direction, and "BM-Case 2," which performs beam prediction in the temporal direction.

[0090] In "BM Case 1," the prediction of beams belonging to Set A (or set A) is made based on the measurement results of beams belonging to Set B (or set B). Set A and Set B may be different sets. Set B may be a subset of Set A. For example, in "BM Case 1," the prediction of beams transmitting CSI-RS may be made from the measurement results of beams transmitting SSB (Synchronization Signal Block). Alternatively, in "BM Case 1," the prediction of a certain number of beams transmitting SSB may be made from the measurement results of a certain number of beams transmitting SSB.

[0091] On the other hand, in "BM Case 2," predictions for beams belonging to Set A are made based on past measurement results of beams belonging to Set B. Similar to "BM Case 1," Set A and Set B may be different sets. Set B may be a subset of Set A. For example, CSI-RS may be predicted from past measurement results of SSB, or the number of beams above a certain number that transmit SSB may be predicted from past measurement results for a certain number of beams below a certain number that transmit SSB.

[0092] In the first embodiment, the following description will focus on "BM Case 1". In particular, in the first embodiment, the UE side model in which inference is performed by the AI / ML model in UE 100 will be described.

[0093] Here, we will explain the content proposed by 3GPP regarding "BM Case 1" in the UE side model.

[0094] Figure 11 is a diagram illustrating the problem that forms the basis of the proposal. The problem will be explained using Figure 11. In the proposal, SSB is used as example set B and CSI-RS as example set A. Therefore, the examples shown in Figure 11 will also be explained using SSB as example set B and CSI-RS as example set A.

[0095] As shown in Figure 11, in step S10, the network node 200 sends an RRC reconfiguration message containing configuration information to the UE 100. The configuration information is for beam prediction using "BM case 1" to be performed in the UE 100.

[0096] In step S11, UE100 reports to network node 200 whether the configuration information is applicable using the RRC Reconfiguration Complete message. Figure 11 shows an example of the report indicating that the information is applicable. The applicable information may include identification information (ID) of the configuration information that has become applicable.

[0097] In step S12, UE100 sets up an AI / ML model capable of beam prediction using "BM Case 1" based on the configuration information. Then, UE100 starts up the AI / ML model. Using the AI / ML model, UE100 infers the measurement results of the CSI-RS transmission beam and reports the inference results as a beam report (CSI beam report) to the network node 200 (step S13). The CSI-RS transmission beam is a beam belonging to set A of "BM Case 1".

[0098] Here, UE100 assumes that the configuration information (step S10) is not applicable (step S14). In this case, UE100 cannot use the AI / ML model. Then, UE100 reports to network node 200 that the configuration information is not applicable (step S15). UE100 may also send information indicating that the configuration information is not applicable using a UAI (UE Assistance Information) message.

[0099] In this case, since UE100 cannot use the AI / ML model, it becomes unclear what should be reported to network node 200.

[0100] To resolve the ambiguity in the operation of UE100, it was proposed that UE100 report the measured values ​​from set B (specifically, the measurement results for SSB) rather than the beam prediction from set A (the beam report from CSI-RS) (Non-Patent Literature 4).

[0101] Figure 12 is a diagram illustrating an example of the proposed content. Figure 12 also serves to explain the problems according to the first embodiment. Figure 12 shows an example of operation after the AI / ML model is activated in UE100 (step S12 in Figure 11).

[0102] In Figure 12, steps S20 to S23 show an example of operation using the AI / ML model, while steps S24 onwards show the specific proposed content.

[0103] As shown in Figure 12, in step S20, the network node 200 transmits SSB using the beam belonging to set B (hereinafter referred to as the "SSB transmission beam"). Beam sweeping is used for SSB transmission. Beam sweeping is a technique that switches the transmission direction of the beam at predetermined time intervals. By beam sweeping, the network node 200 can transmit SSB in time-division multiplexing to different beam directions. The SSB transmission beam is the beam belonging to set B.

[0104] In step S21, the UE100 receives each SSB transmitted in a different beam direction and measures the RSRP (Reference Signal Received Power) of each beam (or each SSB).

[0105] In step S22, UE100 uses an AI / ML model to infer the top k beam for CSI-RS (hereinafter, the beam used for CSI-RS transmission will be referred to as the "CSI-RS transmission beam") from the RSRP of each SSB transmission beam (step S22). For example, since the AI / ML model can infer the probability that each CSI-RS transmission beam will be the top beam, the top k beam may be inferred based on this probability. Note that the CSI-RS transmission beam will be the beam belonging to set A.

[0106] Then, in step S23, UE100 reports the inference result (i.e., the beam prediction result for set A) to network node 200 as a CSI beam report.

[0107] Subsequently, in step S24, the setting information set in step S10 becomes inapplicable.

[0108] In step S25, the network node 200 transmits SSB using the SSB transmission beam.

[0109] In step S26, UE100 measures the RSRP of each SSB. Then, in step S27, UE100 transmits the measurement results to the network node 200.

[0110] However, with the UE100, the configuration information is not applicable, so the AI / ML model cannot be used, and therefore the CSI-RS transmission beam cannot be predicted.

[0111] Therefore, UE100 does not report the prediction result for the CSI-RS transmission beam, but rather reports the measured value of set B, i.e., the RSRP of SSB (step S26).

[0112] Thus, in UE100, if the setting information is not applicable, the 3GPP proposal is to stop the beam prediction for set A and report the measured values ​​for set B.

[0113] The following issues exist regarding the above proposal. Specifically, at network node 200, although the goal is to obtain predicted values ​​for the CSI-RS transmission beam, receiving SSB measurement values ​​may be completely useless. While there should be a certain relationship between the predicted values ​​for the CSI-RS transmission beam and the SSB measurement values, the two results are fundamentally different, and at network node 200, the SSB measurement values ​​may not be meaningful in some cases.

[0114] Therefore, the objective of the first embodiment is to enable UE100 to transmit useful information to network node 200.

[0115] (Communication control method according to the first embodiment) Therefore, in the first embodiment, when predetermined conditions are met, UE100 transmits either the measured value of SSB (set B) or the predicted value of CSI-RS (set A) inferred immediately before the AI / ML model could no longer be used.

[0116] Specifically, firstly, the user device (e.g., UE100) detects that it is not possible to use an AI / ML model capable of inferring a measurement value (e.g., a predicted value) for the first beam (e.g., a CSI-RS transmission beam) included in the first set (e.g., set A) from a measurement value for the second beam (e.g., an SSB transmission beam) included in the second set (e.g., set B). Secondly, depending on whether predetermined conditions are met, the user device transmits to a network node (e.g., network node 200) at least one of the second measurement value for the second beam (e.g., an SSB measurement value) received after the AI / ML model became unusable, and the first measurement value for the first beam (e.g., a CSI-RS predicted value) inferred using the AI / ML model immediately before the AI / ML model became unusable.

[0117] Thus, in the UE 100 according to the first embodiment, when it is not possible to use the AI / ML model, the measurement results of set B are not transmitted uniformly as in the proposed version, but are transmitted only when predetermined conditions are met. Therefore, in the first embodiment, the likelihood of the information being useful to the network node 200 can be increased to a certain level compared to the proposed version. For example, if there is a predetermined condition that there is a certain relationship between the prediction result of set A and the measurement result of set B, and the UE 100 transmits the measurement result of set B that satisfies this predetermined condition, the network node 200 can use this relationship to obtain the prediction result of set A from the measurement result of set B. Therefore, the network node 200 can use the measurement result of set B transmitted from the UE 100 as useful information.

[0118] Furthermore, in the UE 100 according to the first embodiment, if predetermined conditions are met, it is also possible to transmit the prediction results of set A that were inferred immediately before the AI / ML model could no longer be used. The network node 200 can acquire a CSI-RS transmission beam from the prediction results, and can then use that beam to transmit CSI-RS to the UE 100. Therefore, the network node 200 can utilize the prediction results of set A transmitted from the UE 100 as useful information.

[0119] In the first embodiment, SSB will be used as an example for set B, and CSI-RS as an example for set A. Specifically, the beam for SSB transmission (an example of a second beam) will be described as a beam belonging to set B, and the beam for CSI-RS transmission (an example of a first beam) will be described as a beam belonging to set A.

[0120] Furthermore, in the first embodiment, the measured values ​​for the beam and the measured values ​​of the signal received using the beam may not be distinguished. For example, the measured values ​​for the SSB transmission beam and the measured values ​​of the SSB received using the beam may not be distinguished. Also, for example, the measured values ​​(or predicted values) for the CSI-RS transmission beam and the measured values ​​(or predicted values) of the CSI-RS received using the beam may not be distinguished.

[0121] (Overall Operation Example According to the First Embodiment) First, an overall operation example according to the first embodiment will be described, and then a detailed operation example according to the first embodiment will be described.

[0122] Figure 13 is a diagram illustrating an example of the overall operation. In Figure 13, steps S20 to S26 are the same as in Figure 12. That is, UE100 reports the inference results using the AI / ML model (steps S20 to S23), and then detects that the setting information cannot be applied (step S24). Then, UE100 measures the RSRP of each SSB transmission beam (steps S25 and S26).

[0123] In step S30, the control unit 130 of UE100 determines whether a predetermined condition is met. If the predetermined condition is met (YES in step S30), the process proceeds to step S31. On the other hand, if the predetermined condition is not met, the process proceeds to step S32.

[0124] (1) Example of predetermined conditions 1 The predetermined conditions may be based on the measured value of SSB (an example of a second measured value) and the measured value (or predicted value) of CSI-RS (an example of a second measured value).

[0125] Here, the SSB measurements include, for example, measurements for SSB received after the configuration information becomes inapplicable and the AI / ML model can no longer be used (step S26). The SSB measurements may also include RSRP for each (SSB transmission) beam. Specifically, the SSB measurements may include the beam ID for each beam and the RSRP for that beam ID.

[0126] Furthermore, the CSI-RS predictions include, for example, the CSI-RS predictions inferred by the AI / ML model immediately before the AI / ML model becomes unusable (step S22). The CSI-RS predictions may also include the predicted RSRP for each beam (for CSI-RS transmission) and the probability that the beam will be the top beam. Specifically, the CSI-RS predictions may include the beam ID for each beam, the predicted RSRP for that beam ID, and the probability that it will be the top beam. The CSI-RS predictions may also include the top k beam IDs, the predicted RSRP for each beam ID, and the probability that it will be the top beam.

[0127] Specific examples of the specified conditions are as follows:

[0128] Firstly, the predetermined condition may be that the difference between the measured value of SSB (step S26) and the predicted value of CSI-RS (step S22) is less than or equal to the difference threshold.

[0129] If the difference between the measured SSB value and the predicted CSI-RS value is below the difference threshold, the difference between the two RSRPs will be below a certain level, since both the measured and predicted values ​​include RSRP. Furthermore, the measured SSB value was measured after the AI / ML model became unusable (step S24) in Figure 13 (step S26), and the predicted CSI-RS value was predicted just before the AI / ML model became unusable (step S22). In other words, if the difference between the two RSRPs is below a certain level, it can be assumed that UE100 has hardly moved and is remaining in the same location, since the difference between the two RSRPs is below a certain level before and after the AI / ML model becomes unusable.

[0130] Figure 14(A) shows an example of an SSB transmission beam, and Figure 14(B) shows an example of a CSI-RS transmission beam. As shown in Figures 14(A) and 14(B), if the difference between the two is below a certain level, it can be assumed that UE100 has not moved before and after the AI / ML model becomes unusable.

[0131] Therefore, even if network node 200 transmits CSI-RS using the beam corresponding to the beam used for SSB transmission, considering that UE100 has not moved before and after the AI / ML model becomes unusable, it can be assumed that this beam may be the best beam for UE100.

[0132] Furthermore, the CSI-RS transmission beam obtained from the CSI-RS prediction values ​​is the beam predicted just before the AI / ML model became unusable. Considering that UE100 does not move before or after the AI / ML model becomes unusable, it can be assumed that this beam may be the best beam for UE100.

[0133] Based on the above, the transmitting unit 120 of UE100 transmits either the measured value of SSB (step S26 in Figure 13) or the predicted value of CSI-RS (step S22) (step S31) depending on whether the predetermined conditions are met (YES in step S30). As a result, the network node 200 can use the measured value or the predicted value as useful information.

[0134] Secondly, the predetermined conditions may also be that the difference between the measured value of SSB (step S26) and the predicted value of CSI-RS (step S22) is greater than the difference threshold, and the correlation between the measured value of SSB (step S26) and the predicted value of CSI-RS (step S22) is greater than the correlation threshold.

[0135] Even if the two RSRP values—one included in the SSB measurement and the other in the CSI-RS prediction—are greater than the difference threshold, if there is some correlation between the two beams, the network node 200 can use that correlation to obtain the CSI-RS transmission beam from the SSB transmission beam.

[0136] For example, in Figures 14(A) and 14(B), if the difference between the two RSRPs is greater than a certain amount, it can be assumed that there is a certain or greater probability that UE100 has moved between the time the predicted value of CSI-RS was obtained (step S22) and the time the measured value of SSB was obtained (step S26). However, if there is a correlation between the SSB transmission beam and the CSI-RS transmission beam, the network node 200 can obtain the CSI-RS transmission beam from the SSB transmission beam. For example, the network node 200 can select the beam with the largest SSB RSRP based on the measured value of SSB, and then use the correlation to select the beam for CSI-RS transmission from that beam. By transmitting CSI-RS using the beam obtained using the correlation, the network node 200 may receive CSI-RS with that beam as the best beam.

[0137] Alternatively, network node 200 can directly use the predicted CSI-RS values ​​to select a beam for CSI-RS transmission and transmit CSI-RS using that beam. In this case, since the inference is performed just before the AI / ML model can no longer be used, UE100 can expect a certain level of probability that this will be the best beam.

[0138] Based on the above, the transmitting unit 120 of UE100 transmits either the measured value of SSB (step S26 in Figure 13) or the predicted value of CSI-RS (step S22) depending on whether the predetermined conditions are met (YES in step S30). As a result, the network node 200 can use the measured value or the predicted value as useful information.

[0139] On the other hand, a case where the predetermined conditions are not met (NO in step S30) may be when the difference between the measured value of SSB and the predicted value of CSI-RS is greater than the difference threshold, and the correlation between the measured value of SSB and the predicted value of CSI-RS is less than or equal to the correlation threshold.

[0140] If the difference between the two beams is greater than the difference threshold, and the correlation between the two beams is below a certain level, the network node 200 does not know which beam would be the best beam for UE100 to use for CSI-RS transmission. Therefore, in this case, UE100 does not transmit the SSB measurement value (step S26).

[0141] In this case, the transmitting unit 120 of UE 100 may transmit the CSI-RS prediction value that it inferred immediately before (step S22) to the network node 200 (step S32) if the predetermined conditions are not met. This is because the network node 200 can use this as useful information, as it is the result of inference made immediately before the AI / ML model could no longer be used, and can therefore be used as a beam for CSI-RS transmission.

[0142] Alternatively, the transmitting unit 120 of UE 100 may send a fallback request to the network node 200 if the predetermined conditions are not met (step S32). This is because, if the difference between the two beams is greater than the difference threshold and the correlation between the two beams is below a certain level, it can be assumed that selecting the best beam from all beams for CSI-RS transmission will be more accurate without performing inference using an AI / ML model. The fallback request may also be sent using an RRC message, MAC CE, UCI (Uplink Control Information), or a new AI / ML layer message.

[0143] Alternatively, the transmitting unit 120 of UE100 does not have to report anything if the predetermined conditions are not met.

[0144] (2) Example of predetermined conditions 2 The predetermined conditions may be based on the measurement values ​​for the SSB transmission beam received after the AI / ML model can no longer be used (an example of a second measurement value) (step S26) and the measurement values ​​for the SSB transmission beam received immediately before the AI / ML model can no longer be used (an example of a third measurement value) (step S21). In other words, the predetermined conditions may be based on the measurement values ​​from step S26 and step S21.

[0145] Specifically, the predetermined condition may be that the difference between the two measured values ​​is less than or equal to the difference threshold.

[0146] If the difference between the two measured values ​​is less than or equal to the difference threshold, considering that the measured values ​​include the RSRP of each beam, the SSB transmission beam with the largest RSRP will remain almost unchanged before and after the AI / ML model becomes unusable. In other words, it can be assumed that UE100 has not moved from its location before and after the AI / ML model becomes unusable. In this case as well, similar to the example above, even if network node 200 transmits CSI-RS using the beam corresponding to the beam used for SSB transmission, it can be assumed that there is a certain or greater probability that UE100 will receive CSI-RS with that beam as the best beam. Furthermore, even if network node 200 transmits CSI-RS using the beam obtained from the predicted CSI-RS value, it can be assumed that UE100 has not moved, so there is a certain or greater probability that it will receive CSI-RS with that beam as the best beam.

[0147] Based on the above, the transmission unit 120 of UE 100 transmits either the measured value of SSB or the predicted value of CSI-RS depending on whether the predetermined conditions are met (step S31). As a result, the network node 200 can use the measured value or the predicted value as useful information.

[0148] On the other hand, the case where the predetermined conditions are not met (NO in step S30) is when the difference between the two measured values ​​is greater than the difference threshold.

[0149] As shown in Figure 14(A), if the difference between the measurement value obtained in step S21 and the measurement value obtained in step S26 is greater than a certain value, it can be assumed that UE100 moved (at high speed) before and after the AI / ML model became unusable. In such a case, it is meaningless for network node 200 to obtain the SSB measurement value (step S21).

[0150] In this case, the transmitter 120 of UE100 may transmit the predicted CSI-RS value (step S22) inferred immediately before the AI / ML model became unusable. Since this predicted value is the result of inference immediately before the AI / ML model became unusable, it can be assumed that the CSI-RS transmission beam obtained from this predicted value has a certain level of reliability.

[0151] Alternatively, the transmission unit 120 of UE 100 may send a fallback request to the network node 200 if the predetermined conditions are not met (step S32). If the two measured values ​​are larger than a certain level, it can be assumed that UE 100 is moving, and therefore the predicted value itself, which was predicted just before the AI / ML model became unusable, may become meaningless. For this reason, it is also possible to prevent AI / ML model inference from being performed on UE 100.

[0152] Alternatively, the transmitting unit 120 of UE100 does not have to report anything if the predetermined conditions are not met.

[0153] (3) Example of a predetermined condition 3 The predetermined condition is a condition based on the transmission timing of the SSB. Specifically, the predetermined condition is when the transmission timing (or reception timing) of the downlink signal does not coincide with the transmission timing of the SSB. With respect to SSB, multiple SSBs called SSB bursts are transmitted periodically. The periodic period can be selected from 5 [ms], 10 [ms], 20 [ms], 40 [ms], 80 [ms], and 160 [ms], but 20 [ms] may be set as the default period. The periodic period is set by the network node 200.

[0154] Thus, since the SSB transmission timing is predetermined, UE100 can determine which beam in the received beam corresponds to SSB. If UE100 is unable to perform inference, and has confirmed that the downlink signal is SSB, there is no point in transmitting the SSB measurement result. This is because network node 200 wants to receive the predicted value of CSI-RS, which is the inference result. In other words, if the downlink signal reception timing coincides with the SSB transmission timing, UE100 does not report the downlink signal reception result. On the other hand, if the downlink signal reception timing is not the SSB transmission timing, i.e., if the predetermined conditions are met, UE100 reports the measured value of the downlink signal (or the predicted value of CSI-RS).

[0155] In summary, if the transmission timing of the downlink signal is the same as the transmission timing of SSB, the predetermined conditions are not met (NO in step S30), and one of the following is performed: a predicted value from CSI-RS, a fallback request, or no report is made (step S32).

[0156] On the other hand, in UE100, if the transmission timing of the downlink signal is other than the transmission timing of SSB, it is assumed that a predetermined condition is met (YES in step S30), and the measured value of the downlink signal (or the predicted value of CSI-RS) is reported (step S31).

[0157] (Detailed Operation Example According to the First Embodiment) Next, a detailed operation example according to the first embodiment will be described.

[0158] Figures 15 to 17 illustrate detailed operation examples. While some operations in these detailed examples overlap with those in the overall operation example shown in Figure 13, these operations will be omitted as appropriate in the explanation. As with the overall operation example, Figures 15 to 17 use the UE side model, "BM Case 1," as an example for explanation.

[0159] Note that the actions shown in Figure 15 represent examples of operations when the AI / ML model is performing inference normally, while Figures 16 onwards represent examples of operations after the AI / ML model has become unusable.

[0160] As shown in Figure 15, in step S40, the transmitting unit 210 of the network node 200 transmits an RRC reconfiguration message containing configuration information. The configuration information, similar to step S10 in Figure 11, indicates the configuration information for beam prediction using "BM case 1" to be performed in the UE 100. Hereinafter, this will be referred to as configuration information and prediction configuration information. The prediction configuration information may be included, for example, in the CSI report configuration (CSI-ReportConfig). Multiple prediction configuration information entries may be included in list format. In this case, the prediction configuration information may be included in an information element (csi-ReportConfigToAddModList) that includes multiple CSI report configurations. The receiving unit 110 of the UE 100 receives the RRC reconfiguration message containing the prediction configuration information.

[0161] In step S41, the transmission unit 210 of the network node 200 may send configuration information to the UE 100 indicating the operation to be performed when inference cannot be performed in the AI / ML model. Hereinafter, this configuration information will be referred to as the unavailable configuration information. In the example shown in Figure 15, the unavailable configuration information is shown to be sent by an RRC reset message different from the prediction configuration information, but both configuration information may be sent in a single RRC reset message. In that case, flag information for distinguishing between the unavailable configuration information and the prediction configuration information may be included in the RRC reset message.

[0162] Firstly, the unavailable setting information may include a difference threshold and / or a correlation threshold. If the unavailable setting information includes a difference threshold, it may also include information indicating that the difference between the measured value of SSB (Set B) and the predicted value of CSI-RS (Set A) should be reported. Furthermore, if the unavailable setting information includes a correlation threshold, it may also include information indicating that the correlation between the measured value of SSB (Set B) and the predicted value of CSI-RS (Set A) should be reported.

[0163] Secondly, the disabled setting information may include specification information that specifies the processing to be performed in UE100.

[0164] The specified information may include information indicating that the measured values ​​for set B will be reported. Specifically, the specified information may include information indicating that the measured values ​​for SSB will be reported. In this case, the specified information may include information indicating that all SSB transmission beams will be reported. Alternatively, the specified information may include information indicating that the top k SSB transmission beams will be reported. Alternatively, the specified information may include information indicating that the best beam among the SSB transmission beams will be reported. Note that the SSB to be reported may be an SSB received after the AI / ML model became unusable. The SSB to be reported may also be an SSB received immediately before the AI / ML model became unusable. The specified information may include information indicating which SSB it is.

[0165] Alternatively, the specified information may include information indicating that the predicted values ​​for set A, inferred immediately before the AI / ML model became unavailable, should be reported. Specifically, the specified information may include information indicating that the predicted values ​​for CSI-RS should be reported. In this case as well, the specified information may include information indicating that all CSI-RS transmission beams, the top k CSI-RS transmission beams, and the top beam among the CSI-RS transmission beams should be reported.

[0166] Alternatively, the specified information may include information indicating that nothing will be reported.

[0167] Alternatively, the specified information may include information indicating that a fallback request may be sent. In this case, the specified information may include information indicating that it should be reported that the AI / ML model could not perform inference. Alternatively, the specified information may include information indicating that it should be reported why the AI / ML model could not perform inference.

[0168] Alternatively, the specified information may include information indicating that it is left to the implementation dependency of UE100.

[0169] Thirdly, the unavailable setting information may include information indicating the type of measured and predicted values. The information indicating the type may represent the reception quality of Set B or the prediction quality of Set A. The information indicating the type may be, for example, any of the following: (A1) RSRP; (A2) CIR (Carrier to Interference Ratio), CINR (Carrier to Interference and Noise Ratio), or SINR (Signal to Interference and Noise Power Ratio); (A3) AoA (Angle of Arrival); (A4) RSSI (Received Signal Strength Indicator); (A5) MCS (Modulation and Coding Scheme); (A6) Frequencies used for the SSB transmission beam and the CSI-RS transmission beam; (A7) Current position of UE100 or moving speed of UE100; (A8) Beam reception time.

[0170] Fourth, the unavailable setting information may be transmitted using RRC messages, MAC CE, or DCI. The unavailable setting information may also be transmitted using messages of the newly established AI / ML layer for AI / ML.

[0171] The receiving unit 110 of the UE100 receives an RRC reset message that includes information about the unavailable setting.

[0172] In step S42, the transmission unit 120 of UE 100 sends an RRC reset completion message to the network node 200, which includes information indicating whether or not the setting information is applicable. The information indicating whether or not the setting information is applicable may indicate whether or not it is applicable to the predicted setting information. The information indicating whether or not the setting information is applicable may also indicate whether or not it is applicable to the unavailable setting information. The information indicating whether or not the setting information is applicable may include identification information (ID) of the setting information that has become applicable. If there are multiple sets of setting information that have become applicable, the applicable information may include multiple identification information. Figure 11 shows an example in which applicable information indicating that the predicted setting information and the unavailable setting information have become applicable is included. The reception unit 220 of the network node 200 receives the RRC reset completion message. When the control unit 230 of the network node 200 receives the RRC reset completion message, it may activate the applicable setting information based on the applicable information. Activation may be performed using MAC-CE. Activation may also be performed using DCI. Upon activation, the AI / ML model may be configured in UE100.

[0173] In step S43, the control unit 130 of UE 100 sets up an AI / ML model based on the prediction setting information and the unavailable setting information, and starts the AI / ML model. UE 100 may also start the AI / ML model in response to receiving an activation instruction for the AI / ML model transmitted from the network node 200. This activation instruction may be made using MAC CE or DCI.

[0174] In step S44, the transmitting unit 210 of the network node 200 transmits an SSB (SetB) using an SSB transmission beam. The transmitting unit 210 transmits multiple SSB transmission beams using beam sweeping. The receiving unit 110 of the UE 100 receives each SSB transmitted using each SSB transmission beam.

[0175] In step S45, the control unit 130 of UE100 measures the RSRP of each SSB.

[0176] In step S46, the control unit 130 of the UE 100 uses an AI / ML model to infer predicted CSI-RS values ​​from the SSB RSRP measured in step S45. The control unit 130 may also infer predicted values ​​for the top k CSI-RS transmission beams.

[0177] In step S47, the transmitting unit 120 of UE 100 transmits the predicted value inferred in step S46 as an inference result to the network node 200. The inference result may be transmitted using an RRC message (e.g., a UAI message), MAC CE, or UCI. The inference result may also be transmitted using a newly defined AI / ML layer message. The receiving unit 220 of the network node 200 receives the inference result.

[0178] In step S48, the transmitting unit 210 of the network node 200 transmits CSI-RS using the top k CSI-RS transmission beams based on the inference results. The transmitting unit 210 transmits the top k CSI-RS transmission beams using beam sweeping. The receiving unit 110 of the UE 100 receives the k CSI-RS.

[0179] In step S49, the transmitting unit 120 of UE 100 selects the best CSI-RS transmitting beam from the top k CSI-RS transmitting beams and transmits information about the best beam to the network node 200. The control unit 130 of UE 100 may select the best beam based on the RSRP of the received CSI-RS transmitting beam (or k CSI-RS). The information about the best beam may include the beam ID of the best beam and the RSRP of that beam. The information about the best beam may be transmitted using an RRC message, MAC CE, UCI, or a new AI / ML layer message. The receiving unit 220 of the network node 200 receives the information about the best beam.

[0180] In step S50, the transmitting unit 210 of the network node 200 selects the best beam based on the information regarding the best beam and transmits the CSI-RS using the best beam. Subsequently, the UE 100 receives the CSI-RS and transmits a measurement report regarding the CSI-RS to the network node 200. The network node 200 then performs scheduling based on the measurement report. User data is transmitted and received between the UE 100 and the network node 200 based on the scheduling result (steps S51 and S52).

[0181] In step S53 (Figure 16), the control unit 130 of UE100 detects that the AI / ML model cannot be used. The control unit 130 may also detect that the prediction setting information (step S40 in Figure 15) cannot be applied.

[0182] Firstly, the control unit 130 may detect that the AI / ML model cannot be used if the memory capacity inside the UE100 falls below a certain level or if the CPU resources fall below a certain level.

[0183] Secondly, the control unit 130 may detect that the AI / ML model cannot be used if it detects that the usage conditions for the AI / ML model are not met. The usage conditions may include resource settings such as the number of beams. The usage conditions may also include the position of the UE 100. The usage conditions may be included in the prediction setting information.

[0184] Thirdly, the control unit 130 may detect that the AI / ML model cannot be used if it detects that the inference accuracy of the AI / ML model is below a certain level. Specifically, if the difference between the output of the AI / ML model (e.g., the predicted RSRP of CSI-RS) and the past measurement results (actual values) of CSI-RS is greater than or equal to the inference accuracy threshold, the control unit 130 may detect that the AI / ML model cannot be used because the inference accuracy is below a certain level. The inference accuracy threshold may be included in the prediction setting information.

[0185] In step S54, the transmitting unit 210 of the network node 200 transmits an SSB (SetB) using an SSB transmission beam. The transmitting unit 210 transmits multiple SSB transmission beams by beam sweeping. The receiving unit 110 of the UE 100 receives each SSB transmitted using each SSB transmission beam.

[0186] In step S55, the control unit 130 of UE100 measures the RSRP of each SSB.

[0187] In step S56, the control unit 130 of UE100 confirms the disabled setting information (step S41 in Figure 15).

[0188] In step S57, the control unit 130 of UE100 determines the content of the report.

[0189] Firstly, if the unusable setting information includes a differential threshold and / or a correlation threshold, the control unit 130 determines the content of the report depending on whether a predetermined condition (step S30 in Figure 13) is met.

[0190] As explained in the overall operation example (Figure 13), if the predetermined conditions are met, the SSB measurement value or the CSI-RS predicted value may be the reported content (step S31). Therefore, if the predetermined conditions are met, the control unit 130 decides to report either the SSB measurement value (hereinafter referred to as "Option A") or the CSI-RS predicted value (hereinafter referred to as "Option B"). On the other hand, if the predetermined conditions are not met, the control unit 130 decides to report nothing (hereinafter referred to as "Option C"), request a fallback (hereinafter referred to as "Option D"), or report the CSI-RS predicted value ("Option B").

[0191] Secondly, if the unusable setting information includes specified information, the control unit 130 determines the report content according to the specified information.

[0192] In other words, if the specified information specifies that the measured value of set B be reported, specifically the measured value of SSB ("Option A"), the control unit 130 should decide to report Option A. Also, if the specified information specifies that the predicted value of set A be reported, specifically the predicted value of CSI-RS ("Option B"), the control unit 130 should decide to report Option B. Furthermore, if the specified information includes information indicating that nothing should be reported ("Option C"), the control unit 130 should decide to report Option C (i.e., nothing should be reported). Furthermore, if the specified information includes information indicating that a fallback request ("Option D") may be sent, the control unit 130 may decide to report Option D. Each Option can be summarized as follows. • (B1) Option A: Measured values ​​for Set B (RSRP and beam ID, etc.); • (B2) Option B: Predicted values ​​for Set A (RSRP and beam ID, etc.) predicted just before the AI / ML model became unavailable; • (B3) Option C: Do nothing; • (B4) Option D: Request fallback.

[0193] In steps S58 to S61, the transmitter 120 of the UE 100 transmits to the network node 200 according to the determined report content. In the example shown in Figure 16, in step S58, the transmitter 120 of the UE 100 transmits the measured value of SSB (Set B). In step S59, the transmitter 120 of the UE 100 transmits the predicted value of CSI-RS (Set A) that was predicted just before the AI / ML model became unusable. In step S60, the transmitter 120 of the UE 100 does not report anything. In step S61, the transmitter 120 of the UE 100 transmits a fallback request.

[0194] Thus, the UE100 transmits at least the measured value of SSB (step S55) or the predicted value of CSI-RS (step S46) to the network node 200 according to the specified information included in the unavailable setting information.

[0195] Furthermore, UE100 may transmit the thresholds used to determine the report content (difference threshold and / or correlation threshold) and the determination result when the report content was judged, along with the report content, to the network node 200.

[0196] Steps S58, S59, and S61 may be transmitted using an RRC message, MAC CE, UCI, or a new AI / ML layer message.

[0197] The receiving unit 220 of the network node 200 receives the report content transmitted from UE 100 (steps S58, S59, and S61).

[0198] As shown in Figure 17, in the case of Option A and Option B, steps S62 and S63 are performed.

[0199] That is, in step S62, the control unit 230 of the network node 200 determines the beam for CSI-RS (Set A) transmission. When the control unit 230 receives an SSB measurement value, it may determine the beam corresponding to the beam ID included in the measurement value as the beam for CSI-RS transmission. Alternatively, when the control unit 230 receives an SSB measurement value, it may use correlation to determine the beam corresponding to the beam ID included in the measurement value as the beam for CSI-RS transmission. Alternatively, when the control unit 230 receives a predicted CSI-RS value, it may determine the beam having the beam ID included in the predicted value as the beam for CSI-RS transmission. The control unit 230 may select the top k beams. The control unit 230 may select the best beam.

[0200] In step S63, the transmitting unit 210 of the network node 200 transmits the CSI-RS using the beam determined in step S62. The receiving unit 110 of the UE 100 receives each CSI-RS.

[0201] In step S66, the transmitting unit 120 of UE100 selects the best beam from the CSI-RS received in step S63, acquires the measurement result, and transmits the information about the best beam and the measurement result to the network node 200. Subsequently, steps S50 and onward shown in Figure 15 are performed.

[0202] On the other hand, in the case of Option C and Option D, steps S64 and S65 are performed.

[0203] In other words, in step S65, the control unit 230 of the network node 200 detects that there is no report from UE 100, or detects a fallback request. The control unit 230 may also detect that there is no report if it does not receive anything from UE 100 even after a predetermined time has elapsed since transmitting the SSB (step S54).

[0204] In step S65, the transmitting unit 210 of the network node 200 transmits CSI-RS by beam sweeping using all beams for CSI-RS transmission.

[0205] Then, UE100 selects the best beam from the received CSI-RS transmission beams and transmits it to network node 200 along with the measurement results (step S66).

[0206] (Another example of operation according to the first embodiment) In the first embodiment, SSB was used as example set B and CSI-RS as example set A, but the embodiment is not limited to these.

[0207] For example, SSB may be used as set B, and SSB may also be used as set A. This corresponds to the case in "BM Case 1" where set B is a subset of set A. Therefore, the number of beams used for the SSB in set B will be less than that used for the SSB in set A. In this case, it can be implemented by replacing CSI-RS, which was described as being included in set A in the first embodiment, with SSB.

[0208] Alternatively, SSB may be used as set B, and "data" may be used as set A. In this case, the implementation can be carried out by replacing CSI-RS, which was described as being included in set A in the first embodiment, with data.

[0209] Furthermore, CSI-RS may be used as set B, and data as set A. In this case, the SSB described as being included in set A in the first embodiment can be read as CSI-RS, and the CSI-RS described as being included in set B can be read as data.

[0210] Furthermore, data may be used as both Set B and Set A. In this case, Set B is a subset of Set A. Therefore, the number of beams used for the data in Set B will be less than that used for the data in Set A. In this case, it can be implemented by replacing the SSB, which was described as being included in Set A in the first embodiment, with data, and the CSI-RS, which was described as being included in Set B, with data.

[0211] (Another Operation Example 2 According to the First Embodiment) In the first embodiment, examples were described in which data and information transmitted and received between the UE 100 and the network node 200 are transmitted using RRC messages, MAC CE, or UCI (or DCI). However, at least a portion of such data and information may be transmitted using new messages of the AI / ML layer newly defined for the AI / ML model (e.g., AI / ML layer messages).

[0212] [Other Embodiments] The above-described operation flows are not limited to being performed separately and independently; two or more operation flows can be combined and performed. For example, some steps of one operation flow may be added to another operation flow, or some steps of one operation flow may be replaced with some steps of another operation flow. It is not necessary to execute all steps in each flow; only some steps may be executed. Also, the order of steps in each flow may be changed as appropriate.

[0213] In the embodiments and examples described above, an example in which the base station is an NR base station (gNB) was described, but the base station may also be an LTE base station (eNB) or a 6G base station. Furthermore, the base station may also be a relay node such as an IAB (Integrated Access and Backhaul) node. The base station may also be a DU of an IAB node. Furthermore, UE100 may be an MT (Mobile Termination) of an IAB node. That is, UE100 may be a terminal function unit (a type of communication module) for the base station to control a relay device that performs signal relay. Such a terminal function unit is referred to as an MT. Examples of multi-transmission architectures (MTs) include IAB-MT, NCR (Network Controlled Repeater)-MT, and RIS (Reconfigurable Intelligent Surface)-MT.

[0214] Furthermore, the term "network node" primarily refers to a base station, but may also refer to a core network device or a part of a base station (CU, DU, or RU). Additionally, a network node may consist of a combination of at least a part of the core network device and at least a part of a base station.

[0215] A program may be provided that causes a computer to execute each process performed by the UE 100 or the network node 200. The program may be recorded on a computer-readable medium. Using a computer-readable medium, it is possible to install the program on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transient recording medium. The non-transient recording medium is not particularly limited, but may be a recording medium such as a CD-ROM and / or DVD-ROM. Alternatively, the circuits that execute each process performed by the UE 100 or the network node 200 may be integrated, and at least a part of the UE 100 or the network node 200 may be configured as a semiconductor integrated circuit (chipset, SoC (System on a chip)).

[0216] The functions realized by the above-described communication device (UE100 or network node 200, etc.) may be implemented in a circuit or processing circuit, including a general-purpose processor, application-specific processor, integrated circuit, ASICs (Application Specific Integrated Circuits), CPU (a Central Processing Unit), conventional circuitry, and / or a combination thereof, programmed to realize the described functions. The processor includes transistors and / or other circuits and is considered a circuit or processing circuit. The processor may also be a programmed processor that executes a program stored in memory. In this specification, circuit, unit, and means are hardware programmed to perform or execute the functions described herein. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to perform or execute the functions described herein. If such hardware is a processor that is considered to be of the type of circuit, such circuit, means, or unit is a combination of hardware and software used to constitute such hardware and / or processor.

[0217] The phrases “based on” and “depending on / in response to” as used in this disclosure do not mean “based solely on” or “in response solely” unless otherwise specified. “Based on” means both “based solely on” and “at least partially on.” Similarly, “depending” means both “at least partially on” and “in at least partially on.” The terms “include,” “comprise,” and variations thereof do not mean that they include only the listed items, but may include only the listed items or may include additional items in addition to the listed items. Furthermore, the term “or” as used in this disclosure is not intended to mean exclusive OR. Additionally, any reference to elements using designations such as “first,” “second,” etc., as used in this disclosure does not limit the quantity or order of those elements in general. These designations may be used herein as a convenient way to distinguish between two or more elements. Therefore, references to the first and second elements do not imply that only two elements may be adopted therein, or that the first element must precede the second element in any way. In this disclosure, where articles are added by translation, such as a, an, and the in English, these articles shall be plural unless it is clearly indicated from the context that they are not.

[0218] Although the embodiments have been described in detail above with reference to the drawings, the specific configuration is not limited to those described above, and various design changes can be made without departing from the gist of the invention.

[0219] This application claims priority to Japanese Patent Application No. 2025-052191 (filed on March 26, 2025), the entirety of which is incorporated into the specification of this application.

[0220] (Note) The above embodiments can be summarized as shown in the note, but the note does not limit the embodiments.

[0221] (Note 1) A communication method in a mobile communication system, comprising: a step of detecting that a user device cannot use an AI / ML model capable of inferring a measurement value for a first beam included in a first set from a measurement value for a second beam included in a second set; and a step of transmitting to a network node, depending on predetermined conditions, either a second measurement value for the second beam received after the AI / ML model became unusable, or a first measurement value for the first beam inferred using the AI / ML model immediately before the AI / ML model became unusable.

[0222] (Note 2) The communication method described in Note 1, wherein the predetermined conditions are conditions based on the second measurement value and the first measurement value.

[0223] (Note 3) The communication method according to Note 1 or Note 2, wherein the predetermined condition is that the difference between the second measurement value and the first measurement value is less than or equal to the difference threshold.

[0224] (Note 4) The communication method according to any one of Notes 1 to 3, wherein the predetermined conditions are that the difference between the second measurement and the first measurement is greater than the difference threshold, and the correlation between the second measurement and the first measurement is greater than the correlation threshold.

[0225] (Note 5) The communication method according to any one of Notes 1 to 4, further comprising the step of transmitting the first measurement value to the network node, transmitting a fallback request for the AI / ML model to the network node, or not reporting anything, depending on whether the user device does not satisfy the predetermined conditions.

[0226] (Note 6) The communication method according to any one of Notes 1 to 5, wherein the predetermined conditions are based on the second measurement value and the third measurement value for the second beam received immediately before the AI / ML model could no longer be used.

[0227] (Note 7) The communication method according to any one of Notes 1 to 6, wherein the predetermined condition is that the difference between the second measurement value and the third measurement value is less than or equal to the difference threshold.

[0228] (Note 8) The communication method according to any one of Notes 1 to 7, further comprising the steps of transmitting the first measurement value to the network node, transmitting a fallback request for the AI / ML model to the network node, or not reporting anything, depending on whether the user device does not satisfy the predetermined conditions.

[0229] (Note 9) The communication method described in any of Notes 1 to 8, wherein the predetermined conditions are conditions based on the transmission timing of the SSB transmitted using the second beam.

[0230] (Note 10) The communication method according to any one of Notes 1 to 9, further comprising the step of the user device receiving setting information including differential thresholds and / or correlation thresholds.

[0231] (Note 11) The communication method according to any one of Notes 1 to 10, wherein the second beam is a beam used to transmit SSB and the first beam is a beam used to transmit CSI-RS.

[0232] (Note 12) A communication method in a mobile communication system, comprising: the steps of: a user device receiving configuration information from a network node; the user device detecting that it is not possible to use an AI / ML model capable of inferring a measurement value for a first beam included in a first set from a measurement value for a second beam included in a second set; and the user device transmitting to the network node, in accordance with the configuration information, at least one of a second measurement value for the second beam received after it became impossible to use the AI / ML model, and a first measurement value for the first beam inferred using the AI / ML model immediately before it became impossible to use the AI / ML model.

[0233] 1: Mobile communication system 20: RAN 30: CN 100: UE 101: CSI generation unit 102: CSI prediction model 110: Receiving unit 120: Transmitting unit 130: Control unit 140: Communication unit 200: Network node 201: CSI reconstruction unit 210: Transmitting unit 220: Receiving unit 230: Control unit 240: NW communication unit 250: Communication unit 300: CN device 1010: Inference unit for CSI generation 1011: Quantization unit 2010: Inverse quantization unit 2011: Inference unit for CSI reconstruction A1: Data acquisition unit A2: Model learning unit A3: Model inference unit A5: Management unit A6: Model recording unit

Claims

A communication method in a mobile communication system, The user device detects that it cannot use an AI (Artificial Intelligence) / ML (Machine Learning) model that is capable of inferring measurement values ​​for the first beam in the first set from measurement values ​​for the second beam in the second set, The user device transmits to a network node, depending on whether predetermined conditions are met, either a second measurement value for the second beam received after it becomes impossible to use the AI / ML model, or a first measurement value for the first beam inferred using the AI / ML model immediately before it becomes impossible to use the AI / ML model. Communication method. 　 The predetermined conditions are conditions based on the second measurement value and the first measurement value. The communication method according to claim 1. 　 The predetermined condition is that the difference between the second measurement value and the first measurement value is less than or equal to the difference threshold. The communication method according to claim 2. 　 The predetermined conditions are that the difference between the second measurement and the first measurement is greater than the difference threshold, and the correlation between the second measurement and the first measurement is greater than the correlation threshold. The communication method according to claim 2. 　 The user device further includes, depending on whether it does not meet the predetermined conditions, transmitting the first measurement value to the network node, transmitting a fallback request for the AI / ML model to the network node, or not reporting anything. The communication method according to claim 4. 　 The predetermined conditions are based on the second measurement and the third measurement of the second beam received immediately before the AI / ML model could no longer be used. The communication method according to claim 1. 　 The predetermined condition is that the difference between the second measurement value and the third measurement value is less than or equal to the difference threshold. The communication method according to claim 6. 　 The user device further includes, depending on whether it does not meet the predetermined conditions, transmitting the first measurement value to the network node, transmitting a fallback request for the AI / ML model to the network node, or not reporting anything. The communication method according to claim 7. 　 The predetermined conditions are conditions based on the transmission timing of the SSB transmitted using the second beam. The communication method according to claim 1. 　 The user device further includes receiving configuration information including differential thresholds and / or correlation thresholds. The communication method according to claim 1. 　 The second beam is used to transmit SSB, and the first beam is used to transmit CSI-RS. The communication method according to claim 1. 　 A communication method in a mobile communication system, The user device receives configuration information from the network node, The user device detects that it cannot use an AI / ML model that is capable of inferring measurement values ​​for the first beam included in the first set from measurement values ​​for the second beam included in the second set, The user device transmits to the network node, in accordance with the setting information, at least one of the second measurement value for the second beam received after the AI / ML model can no longer be used, and the first measurement value for the first beam inferred using the AI / ML model immediately before the AI / ML model can no longer be used. Communication method.

Images