Communication control method, network node, and user device

Abstract

One aspect relates to a communication control method in a mobile communication system. The communication control method comprises a step of user equipment transmitting, to a base station, at least one of: a use condition indicating a condition for using each of a plurality of AI / ML models; and an execution condition indicating a condition for executing an operation with respect to each of the plurality of AI / ML models.

Description

Communication Control Method

[0001] The present disclosure relates to a communication control method.

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

[0003] 3GPP contribution: RP-213599, “New SI: Study on Artificial Intelligence (AI) / Machine Learning (ML) for NR Air Interface”

[0004] A communication control method according to one aspect is a communication control method in a mobile communication system, the communication control method including a step of transmitting, by a user equipment, to a network node, at least one of use conditions indicating conditions for using each of a plurality of AI / ML models and execution conditions indicating conditions for executing an operation for each of the plurality of AI / ML models.

[0005] FIG. 1 is a diagram showing an example of the configuration of a mobile communication system according to the first embodiment. FIG. 2 is a diagram showing an example of the configuration of a UE (user equipment) according to the first embodiment. FIG. 3 is a diagram showing an example of the configuration of a gNB (base station) according to the first embodiment. FIG. 4 is a diagram showing an example of the configuration of a protocol stack according to the first embodiment. FIG. 5 is a diagram showing an example of the configuration of a protocol stack according to the first embodiment. FIG. 6 is a diagram showing an example of the configuration of functional blocks of the AI / ML technology according to the first embodiment. FIG. 7 is a diagram showing an example of operation in the AI / ML technology according to the first embodiment. FIG. 8 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. FIG. 9 is a diagram showing an example of reducing CSI-RS according to the first embodiment. FIG. 10 is a diagram showing an example of reducing CSI-RS according to the first embodiment. FIG. 11 is a diagram showing an example of operation according to the first embodiment. FIG. 12 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. FIG. 13 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. FIG. 14 is a diagram illustrating an example of the layout of functional blocks of AI / ML technology according to the first embodiment. FIG. 15 is a diagram illustrating an example of operation according to the first embodiment. FIG. 16 is a diagram illustrating an example of the layout of functional blocks of AI / ML technology according to the first embodiment. FIG. 17 is a diagram illustrating an example of operation according to the first embodiment. FIG. 18 is a diagram illustrating an example of operation according to the first embodiment. FIG. 19 is a diagram illustrating an example of a setting message according to the first embodiment. FIG. 20 is a diagram illustrating an example of usage conditions according to the first embodiment. FIG. 21 is a diagram illustrating an example of operation according to the first embodiment. FIG. 22(A) is a diagram illustrating an example of an AI / ML model associated with a priority according to the first embodiment, and FIG. 22(B) is a diagram illustrating an example of information related to Activate according to the first embodiment. FIGS. 23(A) and 23(B) are diagrams illustrating examples of execution conditions according to the second embodiment. FIG. 24 is a diagram illustrating an example of operation according to the second embodiment. FIG. 25 is a diagram illustrating an example of priority according to the second embodiment.

[0006] The present disclosure aims to enable a user device to appropriately perform model inference using an AI / ML model.

[0007] [First embodiment] A mobile communication system according to a first embodiment will be described with reference to the drawings. In the description of the drawings, the same or similar parts are denoted by the same or similar reference numerals.

[0008] (Configuration of mobile communication system) The configuration of a mobile communication system according to the first embodiment will be described. FIG. 1 is a diagram showing an example of the configuration of a 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. Although 5GS will be described below as an example, the mobile communication system may also be at least partially applied to an LTE (Long Term Evolution) system. The mobile communication system may also be at least partially applied to a sixth generation (6G) system or later system.

[0009] The mobile communication system 1 includes a user equipment (UE) 100, a 5G radio access network (NG-RAN: Next Generation Radio Access Network) 10, and a 5G core network (5GC: 5G Core Network) 20. Hereinafter, the NG-RAN 10 may be simply referred to as the RAN 10. Furthermore, the 5GC 20 may be simply referred to as the core network (CN) 20.

[0010] The UE 100 is a mobile wireless communication device. The UE 100 may be any device that is used by a user. For example, the UE 100 may be a mobile phone terminal (including a smartphone) and / or a tablet terminal, a notebook PC, a communication module (including a communication card or a chipset), a sensor or a device provided in a sensor, a vehicle or a device provided in a vehicle (Vehicle UE), or an aircraft or a device provided in an aircraft (Aerial UE).

[0011] The NG-RAN 10 includes a base station (called a "gNB" in a 5G system) 200. The gNBs 200 are connected to each other via an Xn interface, which is an interface between base stations. The gNB 200 manages one or more cells. The gNB 200 performs wireless communication with a UE 100 that has established a connection with its own cell. The gNB 200 has a radio resource management (RRM) function, a routing function for user data (hereinafter simply referred to as "data"), a measurement control function for mobility control and scheduling, and the like. 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 for wireless communication with the UE 100. One cell belongs to one carrier frequency (hereinafter simply referred to as "frequency").

[0012] In addition, gNBs can also be connected to the Evolved Packet Core (EPC), which is the core network of LTE. LTE base stations can also be connected to 5GC. LTE base stations and gNBs can also be connected via an inter-base station interface.

[0013] The 5GC20 includes an AMF (Access and Mobility Management Function) and a UPF (User Plane Function) 300. The AMF performs various mobility controls for the UE 100. The AMF manages the mobility of the UE 100 by communicating with the UE 100 using NAS (Non-Access Stratum) signaling. The UPF controls data forwarding. The AMF and the UPF 300 are connected to the gNB 200 via an NG interface, which is an interface between a base station and a core network. The AMF and the UPF 300 may be core network devices included in the CN 20.

[0014] 2 is a diagram showing an example of the configuration of a UE 100 (user equipment) according to the first embodiment. The UE 100 includes 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 that performs wireless communication with the gNB 200. The UE 100 is an example of a communication device.

[0015] The receiving unit 110 performs various reception operations under the control of the control unit 130. The receiving unit 110 includes an antenna and a receiver. The receiver converts a radio signal received by the antenna into a baseband signal (received signal) and outputs the baseband signal to the control unit 130.

[0016] The transmitting unit 120 performs various transmissions under the control of the control unit 130. The transmitting unit 120 includes an antenna and a transmitter. The transmitter converts a baseband signal (transmission signal) output by the control unit 130 into a radio signal and transmits it from the antenna.

[0017] The control unit 130 performs various controls and processes in the UE 100. Such processes include processes of each layer described below. 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 the processes by the processor. The processor may include a baseband processor and a CPU (Central Processing Unit). The baseband processor performs modulation / demodulation, encoding / decoding, etc. of baseband signals. The CPU executes programs stored in the memory to perform various processes. Note that the processes or operations performed in the UE 100 may be performed in the control unit 130.

[0018] 3 is a diagram showing an example of the configuration of a gNB 200 (base station) according to the first embodiment. The gNB 200 includes a transmitter 210, a receiver 220, a controller 230, and a backhaul communication unit 250. The transmitter 210 and the receiver 220 constitute a communication unit that performs wireless communication with the UE 100. The backhaul communication unit 250 constitutes a network communication unit that communicates with the CN 20. The gNB 200 is another example of a communication device.

[0019] The transmitting unit 210 performs various transmissions under the control of the control unit 230. The transmitting unit 210 includes an antenna and a transmitter. The transmitter converts a baseband signal (transmission signal) output by the control unit 230 into a radio signal and transmits it from the antenna.

[0020] 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 a radio signal received by the antenna into a baseband signal (received signal) and outputs the baseband signal to the control unit 230.

[0021] The control unit 230 performs various controls and processes in the gNB 200. Such processes include processes for each layer described below. 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 in the processes by the processor. The processor may include a baseband processor and a CPU. The baseband processor performs modulation / demodulation, encoding / decoding, etc. of baseband signals. The CPU executes programs stored in the memory to perform various processes. Note that the processes or operations performed in the gNB 200 may be performed by the control unit 230.

[0022] The backhaul communication unit 250 is connected to adjacent base stations via an Xn interface, which is an interface between base stations. The backhaul communication unit 250 is connected to the AMF / UPF 300 via an NG interface, which is an interface between a base station and a core network. The gNB 200 may be configured (i.e., functionally divided) with a central unit (CU) and a distributed unit (DU), and the two units may be connected via an F1 interface, which is a fronthaul interface.

[0023] FIG. 4 is a diagram showing an example of the configuration of a protocol stack of a radio interface of a user plane that handles data.

[0024] The user plane air interface protocol includes a physical (PHY) layer, a medium 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.

[0025] 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 gNB200 via a physical channel. The PHY layer of UE100 receives downlink control information (DCI) transmitted from gNB200 on a physical downlink control channel (PDCCH). Specifically, UE100 performs blind decoding of the PDCCH using a radio network temporary identifier (RNTI) and acquires successfully decoded DCI as DCI addressed to the UE. The DCI transmitted from gNB200 has a CRC (Cyclic Redundancy Code) parity bit scrambled by the RNTI added.

[0026] In NR, UE100 can use a bandwidth narrower than the system bandwidth (i.e., the cell bandwidth). The gNB200 configures the UE100 with a bandwidth portion (BWP) consisting of contiguous PRBs (Physical Resource Blocks). The UE100 transmits and receives data and control signals in the active BWP. For example, up to four BWPs may be configurable for the UE100. Each BWP may have a different subcarrier spacing. The BWPs may overlap in frequency. When multiple BWPs are configured for the UE100, the gNB200 can specify which BWP to apply by controlling the downlink. This allows the gNB200 to dynamically adjust the UE bandwidth according to the amount of data traffic of the UE100, etc., thereby reducing UE power consumption.

[0027] The gNB 200 can configure, for example, up to three control resource sets (CORESETs) for each of up to four BWPs on the serving cell. The CORESET is a radio resource for control information to be received by the UE 100. Up to 12 or more CORESETs may be configured on the serving cell for the UE 100. Each CORESET may have an index of 0 to 11 or more. The CORESET may consist of six resource blocks (PRBs) and one, two, or three consecutive Orthogonal Frequency Division Multiplex (OFDM) symbols in the time domain.

[0028] The MAC layer performs data priority control, retransmission processing using Hybrid Automatic Repeat reQuest (HARQ), random access procedures, etc. Data and control information are transmitted between the MAC layer of the UE 100 and the MAC layer of the gNB 200 via a transport channel. The MAC layer of the gNB 200 includes a scheduler. The scheduler determines the uplink and downlink transport format (transport block size, modulation and coding scheme (MCS)) and the resource blocks to be allocated to the UE 100.

[0029] 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 the UE 100 and the RLC layer of the gNB 200 via a logical channel.

[0030] The PDCP layer performs header compression / decompression, encryption / decryption, and the like.

[0031] The SDAP layer maps IP flows, which are units for Quality of Service (QoS) control by the core network, to radio bearers, which are units for QoS control by the access stratum (AS). Note that if the RAN is connected to the EPC, SDAP may not be required.

[0032] FIG. 5 is a diagram showing the configuration of a protocol stack of a radio interface of a control plane that handles signaling (control signals).

[0033] The protocol stack of the radio interface of the control plane includes a radio resource control (RRC) layer and a non-access stratum (NAS) instead of the SDAP layer shown in FIG.

[0034] RRC signaling for various settings is transmitted between the RRC layer of UE100 and the RRC layer of gNB200. The RRC layer controls logical channels, transport channels, and physical channels according to the establishment, re-establishment, and release of radio bearers. When there is a connection (RRC connection) between the RRC of UE100 and the RRC of gNB200, UE100 is in an RRC connected state. When there is no connection (RRC connection) between the RRC of UE100 and the RRC of gNB200, UE100 is in an RRC idle state. When the connection between the RRC of UE100 and the RRC of gNB200 is suspended, UE100 is in an RRC inactive state.

[0035] The NAS, which is located above the RRC layer, performs session management, mobility management, etc. NAS signaling is transmitted between the NAS of the UE 100 and the NAS of the AMF 300. Note that the UE 100 has an application layer and the like in addition to the radio interface protocol. Also, the layer below the NAS is called an Access Stratum (AS).

[0036] (AI / ML Technology) Next, the AI / ML technology according to the embodiment will be described. Fig. 6 is a diagram showing an example of the configuration of functional blocks of the AI / ML technology in the mobile communication system 1 according to the first embodiment.

[0037] The functional block configuration example shown in FIG. 6 includes a data collection unit A1, a model learning unit A2, a model inference unit A3, and a data processing unit A4.

[0038] The data collection unit A1 collects input data, specifically, learning data and inference data. The data collection unit A1 outputs the learning data to the model learning unit A2. The data collection unit A1 also outputs the inference data to the model inference unit A3. The data collection unit A1 may acquire data in the device on which the data collection unit A1 is provided as input data. The data collection unit A1 may also acquire data in another device as input data.

[0039] The model learning unit A2 performs model learning. Specifically, the model learning unit A2 optimizes parameters of the learning model through machine learning using the learning data, and derives (or generates, or updates) a learned model. The model learning unit A2 outputs the derived learned model to the model inference unit A3. For example, considering y = ax + b, a (slope) and b (intercept) are parameters, and optimizing these corresponds to machine learning. Generally, machine learning is classified into supervised learning, unsupervised learning, and reinforcement learning. Supervised learning is a method that uses correct answer data as learning data. Unsupervised learning is a method that does not use correct answer data as learning data. For example, in unsupervised learning, feature points are memorized from a large amount of learning data, and the correct answer is determined (range estimation). Reinforcement learning is a method of assigning a score to an output result and learning how to maximize the score. Although supervised learning will be described below, unsupervised learning or reinforcement learning may also be applied as machine learning.

[0040] The model inference unit A3 performs model inference. Specifically, the model inference unit A3 infers an output from inference data using a trained model and outputs the inference result data to the data processing unit A4. For example, in the case of y = ax + b, x corresponds to the inference data and y corresponds to the inference result data. Note that "y = ax + b" is a model. A model in which the slope and intercept are optimized, for example, "y = 5x + 3", is a trained model. Here, there are various model approaches, such as linear regression analysis, neural networks, and decision tree analysis. The above "y = ax + b" can be considered a type of linear regression analysis. The model inference unit A3 may provide model performance feedback to the model learning unit A2.

[0041] The data processing unit A4 receives the inference result data and performs processing that utilizes the inference result data.

[0042] FIG. 7 is a diagram illustrating an example of operation in the AI / ML technique according to the first embodiment.

[0043] The transmitting entity TE is, for example, an entity in which machine learning is performed. The transmitting entity TE performs machine learning to derive a trained model. The transmitting entity TE then generates inference result data as an inference result using the trained model. The transmitting entity TE transmits the inference result data to the receiving entity RE.

[0044] On the other hand, the receiving entity RE is, for example, an entity in which machine learning is not performed. The transmitting entity TE performs various processes using the inference result data received from the transmitting entity TE.

[0045] The entity may be, for example, a device, a functional block included in the device, or a hardware block included in the device.

[0046] For example, the transmitting entity TE may be the UE 100, and the receiving entity RE may be the gNB 200 or a core network device. Alternatively, the transmitting entity TE may be the gNB 200 or a core network device, and the receiving entity RE may be the UE 100.

[0047] As shown in Fig. 7, in step S1, the transmitting entity TE transmits control data related to AI / ML technology to the receiving entity RE and receives the control data from the receiving entity RE. The control data may be an RRC message, which is signaling of the RRC layer (i.e., Layer 3). The control data may be a MAC Control Element (CE), which is signaling of the MAC layer (i.e., Layer 2). The control data may be Downlink Control Information (DCI), which is signaling of the PHY layer (i.e., Layer 1). The downlink signaling may be UE-specific signaling. The downlink signaling may be broadcast signaling. The control data may be a control message in a control layer (e.g., AI / ML layer) specialized for artificial intelligence or machine learning.

[0048] (Layout Examples and Use Cases) Next, a description will be given of how the functional blocks shown in Fig. 6 are arranged in the mobile communication system 1. Below, layout examples of the functional blocks will be described along with specific use cases.

[0049] For example, there are three use cases in which AI / ML technology is applied:

[0050] (1.1) "CSI (Channel State Information) Feedback Enhancement"

[0051] (1.2) "Beam management"

[0052] (1.3) "Positioning Accuracy Enhancement" Below, an example of the placement of functional blocks for each use case will be described.

[0053] (1.1) Example of functional block arrangement in "CSI feedback improvement" "CSI feedback improvement" represents a use case in which, for example, machine learning technology is applied to CSI fed back from UE100 to gNB200. CSI is information about the channel state in the downlink between UE100 and gNB200. The CSI includes at least one of a channel quality indicator (CQI), a precoding matrix indicator (PMI), and a rank indicator (RI). The gNB200 performs, for example, downlink scheduling based on the CSI feedback from UE100.

[0054] 8 is a diagram showing an example of the arrangement of each functional block in "CSI feedback improvement". In the example of "CSI feedback improvement" shown in FIG. 8, a data collection unit A1, a model learning unit A2, and a model inference unit A3 are included in the control unit 130 of the UE 100. On the other hand, a data processing unit A4 is included in the control unit 230 of the gNB 200. That is, model learning and model inference are performed in the UE 100. FIG. 8 shows an example in which the transmitting entity TE is the UE 100 and the receiving entity RE is the gNB 200.

[0055] In "CSI feedback improvement", the gNB 200 transmits a reference signal for the UE 100 to estimate the downlink channel state. As the reference signal, a CSI reference signal (CSI-RS) will be described as an example below, but the reference signal may be a demodulation reference signal (DMRS).

[0056] First, in model learning, UE100 (receiving unit 110) receives a first reference signal from gNB200 using a first resource. Then, UE100 (model learning unit A2) derives a learned model for inferring CSI from the reference signal using learning data including the first reference signal and CSI. Such a first reference signal is sometimes referred to as a full CSI-RS.

[0057] For example, the CSI generation unit 131 performs channel estimation using the received signal (CSI-RS) received by the receiving unit 110 to generate CSI. The transmitting unit 120 transmits the generated CSI to the gNB 200. The model learning unit A2 performs model learning using a set of the received signal (CSI-RS) and the CSI as learning data, and derives a learned model for inferring the CSI from the received signal (CSI-RS).

[0058] Second, in model inference, the receiver 110 receives a second reference signal from the gNB 200 using a second resource that is less than the first resource. Then, the model inference unit A3 uses the trained model to infer CSI as inference result data using the second reference signal as inference data. Hereinafter, such a second reference signal may be referred to as a partial CSI-RS or a punctured CSI-RS.

[0059] For example, the model inference unit A3 inputs the partial CSI-RS received by the receiving unit 110 as inference data into the trained model, and infers CSI from the CSI-RS. The transmitting unit 120 transmits the inferred CSI to the gNB 200.

[0060] This enables UE 100 to feed back (or transmit) accurate (complete) CSI to gNB 200 from the small amount of CSI-RS (partial CSI-RS) received from gNB 200. For example, gNB 200 can reduce (puncture) CSI-RS when intended to reduce overhead. In addition, UE 100 can respond to situations where the radio conditions deteriorate and some CSI-RS cannot be received normally.

[0061] 9 and 10 are diagrams illustrating an example of reducing CSI-RS according to the first embodiment.

[0062] FIG. 9 shows an example of reducing CSI-RS by reducing the number of antenna ports that transmit CSI-RS. For example, the gNB 200 performs the following process. That is, when the UE 100 is in a mode in which model learning is performed (hereinafter, sometimes referred to as the "learning mode"), the gNB 200 transmits CSI-RS from all antenna ports of the antenna panel. On the other hand, when the UE 100 is in a mode in which model inference is performed (hereinafter, sometimes referred to as the "inference mode"), the gNB 200 reduces the number of antenna ports that transmit CSI-RS and transmits CSI-RS from half the antenna ports of the antenna panel. This reduces overhead, improves antenna port utilization efficiency, and achieves a reduction in power consumption. Note that the antenna port is an example of a resource.

[0063] On the other hand, FIG. 10 shows an example in which the radio resources used for transmitting the CSI-RS, specifically, the gNB 200, reduces the time-frequency resources. For example, the gNB 200 performs the following process. That is, when the UE 100 is in learning mode, the gNB 200 transmits the CSI-RS using predetermined time-frequency resources. On the other hand, when the UE 100 is in inference mode, the gNB 200 transmits the CSI-RS using time-frequency resources that are less than the predetermined time-frequency resources. This reduces overhead, improves the utilization efficiency of radio resources, and reduces power consumption.

[0064] As shown in Figures 9 and 10, gNB200 transmits full CSI-RS using a predetermined amount of first resources and transmits partial CSI-RS using second resources that have a smaller amount of resources than the first resources.

[0065] FIG. 11 is a diagram illustrating an example of operation in "CSI feedback improvement" according to the first embodiment.

[0066] 11, in step S101, the gNB 200 may notify or set the transmission pattern (puncture pattern) of the CSI-RS in the inference mode as control data to the UE 100. For example, the gNB 200 transmits to the UE 100 the antenna port and / or time-frequency resource that transmits or does not transmit the CSI-RS in the inference mode.

[0067] In step S102, gNB200 may send a switching notification to UE100 to start learning mode.

[0068] In step S103, the UE 100 starts the learning mode.

[0069] In step S104, the gNB 200 transmits the full CSI-RS. The receiver 110 of the UE 100 receives the full CSI-RS, and the CSI generator 131 generates (or estimates) CSI based on the full CSI-RS. In the learning mode, the data collector A1 collects the full CSI-RS and CSI. The model learning unit A2 creates a learned model using the full CSI-RS and the CSI as learning data.

[0070] In step S105, UE100 transmits the generated CSI to gNB200.

[0071] Thereafter, in step S106, when the model learning is completed, the UE 100 transmits a completion notification indicating that the model learning is completed to the gNB 200. The UE 100 may transmit a completion notification when the creation of the learned model is completed.

[0072] In step S107, in response to receiving the completion notification, gNB200 sends a switching notification to UE100 to switch UE100 from learning mode to inference mode.

[0073] In step S108, in response to receiving the switching notification, the UE 100 switches from the learning mode to the inference mode.

[0074] In step S109, the gNB 200 transmits a partial CSI-RS. The receiver 110 of the UE 100 receives the partial CSI-RS. In the inference mode, the data collector A1 collects the partial CSI-RS. The model inference unit A3 inputs the partial CSI-RS as inference data into the trained model, and obtains CSI as the inference result.

[0075] In step S110, the UE 100 feeds back (or transmits) the CSI, which is the inference result, to the gNB 200 as inference result data. In the UE 100, by repeating model learning in the learning mode, a trained model with a predetermined accuracy or higher can be generated. It is expected that the inference result using the trained model generated in this way will also have a predetermined accuracy or higher.

[0076] In addition, in step S111, if UE100 determines that model learning is necessary, it may send a notification indicating that model learning is necessary to gNB200 as control data.

[0077] In the example shown in Fig. 11, an example has been described in which the training data is "(full) CSI-RS" and "CSI", and the inference data is "(partial) CSI-RS". Hereinafter, the training data and / or the inference data may be referred to as a "dataset".

[0078] In "improving CSI feedback," in addition to "CSI-RS" and "CSI," at least one of the following data or information may be used as a data set:

[0079] (X1) RSRP (Reference Signals Received Power), RSRQ (Reference Signal Received Quality), SINR (Signal-to-interference-plus-noise ratio), or AD converter output waveform (These measurements may be CSI-RS. These measurements may also be other received signals received from the gNB 200.)

[0080] (X2) Bit Error Rate (BER) or Block Error Rate (BLER) (The total number of transmitted bits (or the total number of transmitted blocks) is known, and the BER (or BLER) may be measured based on the CSI-RS.)

[0081] (X3) The movement speed of UE100 (which may be measured by a speed sensor within UE100). The data set to be used for machine learning may be set. For example, the following processing may be performed. That is, UE100 transmits capability information indicating which type of input data it can handle in machine learning to gNB200 as control data. The capability information may represent, for example, any of the data or information shown in (X1) to (X3). The capability information may be information in which learning data and inference data are separately specified. Then, gNB200 transmits data type information to be used as the data set to UE100 as control data. The data type information may represent, for example, any of the data or information shown in (X1) to (X3). Furthermore, the data type information may specify separately data type information to be used as learning data and data type information to be used as inference data.

[0082] (1.2) Example of functional block arrangement in "beam management" Next, an example of functional block arrangement in "beam management" will be described. "Beam management" represents a use case in which, for example, machine learning technology is used to manage which beam is the optimal beam among the beams transmitted from gNB200.

[0083] In "beam management," gNB200 sequentially transmits beams with different directivities. Each beam includes, for example, a reference signal. UE100 measures the reception quality of each beam using the reference signal included in each beam. UE100, for example, determines the beam with the best reception quality as the optimal beam.

[0084] Figure 12 is a diagram showing an example of the arrangement of each functional block in "beam management". In the example of "beam management" shown in Figure 12, a data collection unit A1, a model learning unit A2, and a model inference unit A3 are included in the control unit 130 of UE100. On the other hand, a data processing unit A4 is included in the control unit 230 of gNB200. That is, Figure 12 shows an example in which model learning and model inference are performed in UE100. Figure 12 shows an example in which the transmitting entity TE is UE100 and the receiving entity RE is gNB200.

[0085] As shown in FIG. 12, the UE 100 has an optimal beam determination unit 132. The optimal beam determination unit 132 determines the optimal beam based on, for example, the reception quality for the reference signal included in each beam. As with "CSI feedback," an example will be described in which a CSI-RS is used as the reference signal, but a demodulation reference signal (DMRS) may also be used as the reference signal. The transmitter 120 transmits information representing the determined optimal beam to the gNB 200 as the "optimal beam."

[0086] An example of the operation in "beam management" can be implemented by replacing "CSI feedback" with "optimal beam" in Figure 11.

[0087] In the learning mode (step S103), the gNB 200 sequentially transmits beams with different directivities to the UE 100 (step S104). Each beam includes a full CSI-RS. In the learning mode, the data collection unit A1 of the UE 100 collects the full CSI-RS and the optimal beam (information representing the optimal beam). The model learning unit A2 creates a learned model using the CSI-RS and the optimal beam (information representing the optimal beam) as learning data. The full CSI-RS is an example of a first reference signal, and the partial CSI-RS is an example of a second reference signal.

[0088] In the inference mode (step S108), the gNB 200 sequentially transmits beams with different directivities. Each beam includes a partial CSI-RS. In the inference mode, the data collection unit A1 collects the partial CSI-RS. The model inference unit A3 inputs the partial CSI-RS as inference data into the trained model, and obtains the optimal beam (information representing the optimal beam) as the inference result. The UE 100 transmits the inference result (optimal beam) to the gNB 200 as inference result data.

[0089] In "beam management," in addition to "CSI-RS" and "optimal beam," at least one of the following data or information may be used as data in the data set.

[0090] (Y1) SSB (Synchronization Signal Block) received from gNB200

[0091] (Y2) RSRP, RSRQ, SINR, or AD converter output waveform (these measurements may be CSI-RS. These measurements may also be other received signals received from gNB200)

[0092] (Y3) BER or BLER (BER (or BLER) may be measured based on CSI-RS with the total number of transmission bits (or the total number of transmission blocks) known)

[0093] (Y4) Number of beams or beam pattern

[0094] (Y5) Beam measurement value(s)

[0095] (Y6) UE100's movement speed (may be measured by a speed sensor within UE100) UE100 may transmit capability information indicating which type of input data it can handle in machine learning to gNB200 as control data. The capability information may include any of the information or data from (Y1) to (Y6). The capability information may include any of the information or data from (Y1) to (Y6), separating learning data and inference data. In addition, gNB200 may transmit data type information used as a data set to UE100 as control data. The data type information may include, for example, any of the data or information shown in (Y1) to (Y6). The data type information may include, for example, any of the information or data from (Y1) to (Y6), separating learning data and inference data.

[0096] (1.3) Example of Arrangement of Functional Blocks in "Improvement of Location Accuracy" Next, an example of arrangement of functional blocks in "Improvement of Location Accuracy" will be described. "Improvement of Location Accuracy" represents a use case in which, for example, the accuracy of location information measured by the UE 100 is improved by using machine learning technology.

[0097] Figure 13 is a diagram showing an example of the arrangement of each functional block in "improving location accuracy". In the example of "improving location accuracy" shown in Figure 9, the data collection unit A1, model learning unit A2, and model inference unit A3 are included in the control unit 130 of the UE 100. On the other hand, the data processing unit A4 is included in the control unit 230 of the gNB 200. That is, Figure 13 shows an example in which model learning and model inference are performed in the UE 100. Figure 13 shows an example in which the transmitting entity TE is the UE 100 and the receiving entity RE is the gNB 200.

[0098] As shown in FIG. 13 , the UE 100 includes a location information generation unit 133. The UE 100 may include a GNSS (Global Navigation Satellite System) receiver 150. The location information generation unit 133 generates location data for the UE 100 based on a positioning reference signal (PRS) (full PRS or partial PRS) received from the gNB 200. The location information generation unit 133 may receive a GNSS signal (full GNSS signal or partial GNSS signal) received by the GNSS receiver 150, and generate location data for the UE 100 based on the GNSS signal.

[0099] Note that the gNB 200 transmits the full PRS using a predetermined amount of first resources (for example, all antenna ports as shown in FIG. 9, or a predetermined amount of time-frequency resources as shown in FIG. 10), similar to the full CSI-RS. Also, the gNB 200 transmits the partial PRS using second resources (for example, half the antenna ports in the antenna panel as shown in FIG. 9, or half the predetermined amount of time-frequency resources as shown in FIG. 10) that have a smaller amount of resources than the first resources, similar to the partial CSI-RS.

[0100] The full GNSS signal may also be a GNSS signal that is received continuously over time by the GNSS receiver 150. Furthermore, the partial GNSS signal may also be a GNSS signal that is received intermittently by the GNSS receiver 150. That is, a predetermined amount of first resources may be used for the full GNSS signal, and second resources having a smaller amount than the first resources may be used for the partial GNSS signal.

[0101] An example of the operation for "improving location accuracy" can be implemented by replacing "full CSI-RS" with "full PRS," "partial CSI-RS" with "partial PRS," and "CSI feedback" with "location data" in Figure 11.

[0102] In the learning mode (step S103), the location information generation unit 133 generates location data for the UE 100 based on the full PRS received from the gNB 200. The location information generation unit 133 may receive a full GNSS signal received by the GNSS receiver 150 and generate location data for the UE 100 based on the full GNSS signal. The transmission unit 120 feeds back (or transmits) the location data to the gNB 200. The data collection unit A1 collects the full PRS (or full GNSS signal) and location data. The model learning unit A2 creates a learned model using the full PRS (or full GNSS signal) and location data as learning data.

[0103] In the inference mode (step S108), the data collection unit A1 collects the partial PRS received by the receiving unit 110 (or the partial GNSS signal received by the GNSS receiver 150). The model inference unit A3 inputs the partial PRS (or the partial GNSS signal) as inference data into the trained model, and obtains location data as an inference result. The UE 100 transmits the inference result (location data) to the gNB 200 as inference result data.

[0104] In "improving position accuracy," in addition to "PRS" (or "GNSS signal") and "position data," at least one of the following data or information may be used in the data set:

[0105] (Z1) RSRP, RSRQ, SINR (Signal-to-interference-plus-noise ratio), or AD converter output waveform (these measurements may be PRS. These measurements may also be other received signals received from gNB200.)

[0106] (Z2) LOS (Line of Sight) or NLOS (Non Line of Sight)

[0107] (Z3) Measurement timing, accuracy, likelihood

[0108] (Z4) RF Fingerprint (Cell ID and reception quality in the cell of the cell ID)

[0109] (Z5) Angle of Arrival (AOA), reception level for each antenna, reception phase for each antenna, and observed time difference of arrival (OTDOA) for each antenna

[0110] (Z6) Received information of beacons used in wireless LANs (Local Area Networks) such as Wi-Fi (registered trademark) or short-range wireless communications such as Bluetooth (registered trademark)

[0111] (Z7) Moving speed of UE 100 (The moving speed may be measured by GNSS receiver 150. The moving speed may be measured by a speed sensor in UE 100.)

[0112] The UE 100 may transmit capability information indicating which type of input data the UE 100 can handle in machine learning to the gNB 200 as control data. The capability information may include any of the information or data from (Z1) to (Z7). The capability information may include any of the information or data from (Z1) to (Z7), separating the learning data and the inference data. The gNB 200 may also transmit data type information used as a data set to the UE 100 as control data. The data type information may include, for example, any of the data or information shown in (Z1) to (Z7). The data type information may include, for example, any of the information or data from (Z1) to (Z7), separating the learning data and the inference data.

[0113] (1.4) Other Arrangement Examples Next, other arrangement examples will be described.

[0114] Figure 14 is a diagram showing another example of the arrangement of "CSI feedback improvement" according to the first embodiment. Figure 14 shows an example in which a data collection unit A1, a model learning unit A2, a model inference unit A3, and a data processing unit A4 are included in a gN200. That is, Figure 14 shows an example in which model learning and model inference are performed in the gNB200. Figure 14 shows an example in which the transmitting entity TE is the gNB200 and the receiving entity RE is the UE100.

[0115] 14 shows an example in which AI / ML technology is introduced into CSI estimation performed by gNB200 based on SRS (Sounding Reference Signal). Therefore, gNB200 has a CSI generation unit 231 that generates CSI based on SRS. The CSI is information indicating the channel state of the uplink between UE100 and gNB200. gNB200 (e.g., data processing unit A4) performs, for example, uplink scheduling based on the CSI generated based on SRS.

[0116] FIG. 15 is a diagram illustrating an example of operation in another arrangement example according to the first embodiment.

[0117] 15 , in step S201, the gNB 200 performs SRS transmission configuration for the UE 100. The SRS transmission configuration may include type information of the reference signal transmitted by the UE 100.

[0118] In step S202, gNB200 starts learning mode.

[0119] In step S203, UE100 transmits the full SRS to gNB200 according to the SRS transmission setting (step S201). The receiver 220 of gNB200 receives the full SRS. In the learning mode, the CSI generator 231 generates (or estimates) CSI based on the full SRS. The data collector A1 collects the full SRS and CSI. The model learning unit A2 creates a learned model using the full SRS and CSI as learning data.

[0120] In step S204, the gNB 200 identifies an SRS transmission pattern (puncture pattern) to be input to the learned model as inference data, and sets the identified SRS transmission pattern to the UE 100. The gNB 200 may transmit an SRS transmission configuration including the identified SRS transmission pattern to the gNB 200.

[0121] In step S205, gNB200 switches from learning mode to inference mode. gNB200 starts model inference using the trained model.

[0122] In step S206, UE100 transmits a partial SRS in accordance with the SRS transmission setting (step S204). When gNB200 inputs the SRS as inference data into the trained model to obtain a channel estimation result, it uses the channel estimation result to perform uplink scheduling of UE100 (for example, control of uplink transmission weight, etc.). Note that if the inference accuracy using the trained model deteriorates, gNB200 may reconfigure UE100 to transmit a full SRS.

[0123] (1.5) Example of Arrangement When Federated Learning is Performed Next, an example of the arrangement of each functional block when federated learning is performed will be described. Federated learning is, for example, a machine learning technique in which machine learning is performed in a distributed state without aggregating data (or data sets). In federated learning, each entity does not need to transmit data, so the security of each entity can be ensured. Furthermore, federated learning is said to be able to obtain learning results with the same accuracy as conventional centralized machine learning.

[0124] Figure 16 is a diagram showing an example of a configuration when federated learning according to the first embodiment is performed. The example shown in Figure 16 shows an example where location estimation of UE100 is performed using federated learning. Figure 16 shows an example where UE100 has a data collection unit A1, a model learning unit A2, and a model inference unit A3. That is, it shows an example where model learning and model inference are performed in UE100. Figure 16 shows an example where UE100 is the transmitting entity TE, and gNB200 and / or location server 400 is the receiving entity RE.

[0125] The associative learning shown in FIG. 16 is performed, for example, in the following procedure.

[0126] First, the location server 400 transmits to the UE 100 a model that is the basis for model learning.

[0127] Second, UE 100 (model learning unit A2) performs model learning using data present in UE 100. The data present in UE 100 is, for example, the PRS received from gNB 200 and / or output data (GNSS signal) of GNSS receiver 150. The data present in UE 100 may include location data generated by location information generation unit 133 based on the reception result of the PRS and / or the output data of GNSS receiver 150.

[0128] Third, UE 100 applies the learned model, which is the learning result, in model inference unit A3, and transmits variable parameters (hereinafter, sometimes referred to as "learned parameters") included in the learned model to location server 400. In the above example, the optimized a (slope) and b (intercept) correspond to the learned parameters.

[0129] Fourth, location server 400 (associated learning unit A5) collects learned parameters from multiple UEs 100 and integrates them. Location server 400 may transmit the learned model obtained by the integration to UE 100. Location server 400 can estimate the location of UE 100 based on the learned model and measurement reports from UE 100.

[0130] FIG. 17 is a diagram illustrating an example of operation in the federated learning according to the first embodiment.

[0131] 17, in step S301, the gNB 200 may notify the UE 100 of a model that serves as a basis for learning. The location server 400 may notify the model via the gNB 200.

[0132] In step S302, the gNB 200 instructs the UE 100 to learn the model. The gNB 200 may set the reporting timing (trigger condition) of the learned parameters. The reporting timing may be periodic. The reporting timing may be triggered by the learning proficiency satisfying the condition (i.e., an event trigger).

[0133] In step S303, the UE 100 starts a learning mode. The UE 100 performs model learning using the full PRS (or full GNSS signal) and the position data generated by the position information generating unit 133 as learning data.

[0134] In step S304, when the reporting timing condition is met, UE100 transmits the learned parameters at that time to the network (gNB200 or location server 400).

[0135] In step S305, the location server 400 integrates the learned parameters reported from the plurality of UEs 100.

[0136] (1.6) Model transfer example

[0137] In (1.1) to (1.5), examples of the arrangement of each functional block of AI / ML technology have been described. Below, an example of model transfer will be described. The model to be transferred may be a trained model used in model inference. The model may also be an untrained (or training) model used in model training.

[0138] (1.6.1) First operation pattern related to model forwarding Figure 18 is a diagram showing an example of an operation of the first operation pattern related to model forwarding according to the first embodiment. In the example shown in Figure 18, the description will be given assuming that the receiving entity RE is mainly the UE 100, but the receiving entity RE may be the gNB 200 or the AMF 300. Also, in the example shown in Figure 18, the description will be given assuming that the transmitting entity TE is the gNB 200, but the transmitting entity TE may be the UE 100 or the AMF 300.

[0139] 18, in step S401, the gNB 200 transmits a capability inquiry message to the UE 100 to request transmission of a message including an information element (IE) indicating execution capability for machine learning processing. The UE 100 receives the capability inquiry message. However, the gNB 200 may transmit the capability inquiry message when executing the machine learning processing (when determining that the processing is to be executed).

[0140] In step S402, the UE 100 transmits to the gNB 200 a message including an information element indicating execution capability for machine learning processing (or, from another perspective, the execution environment for machine learning processing). The gNB 200 receives the message. The message may be an RRC message (for example, a "UE Capability" message or a newly defined message (for example, a "UE AI Capability" message, etc.)). Alternatively, the transmitting entity TE may be the AMF 300, and the message may be a NAS message. Alternatively, if a new layer for performing or controlling machine learning processing (AI / ML processing) is defined, the message may be a message of the new layer.

[0141] The information element indicating the execution capability related to the machine learning process may be an information element indicating the capability of a processor for executing the machine learning process and / or an information element indicating the capability of a memory for executing the machine learning process. Specifically, the information element indicating the processor capability may be an information element indicating the product number (or model number) of the AI ​​processor. Specifically, the information element indicating the memory capability may be information indicating the memory capacity.

[0142] Alternatively, the information element indicating the execution capability of machine learning processing may be an information element indicating the execution capability of inference processing (model inference). Specifically, the information element indicating the execution capability of inference processing may be an information element indicating whether a deep neural network model is supported. The information element may also be an information element indicating the time (or response time) required to execute the inference processing.

[0143] Alternatively, the information element indicating the execution capability related to the machine learning process may be an information element indicating the execution capability of the learning process (model learning). Specifically, the information element indicating the execution capability of the learning process may be an information element indicating the number of concurrent executions of the learning process. The information element may be an information element indicating the processing capacity of the learning process.

[0144] In step S403, gNB200 determines the model to be configured (or deployed) in UE100 based on the information elements contained in the message received in step S402.

[0145] In step S404, gNB200 transmits a message including the model determined in step S403 to UE100. UE100 receives the message and performs machine learning processing (i.e., model learning processing and / or model inference processing) using the model included in the message. A specific example of step S404 will be described in the following second operation pattern.

[0146] (1.6.2) Second operation pattern related to model transfer FIG. 19 is a diagram showing an example of a configuration message including a model and additional information according to the first embodiment. The configuration message may be an RRC message transmitted from the gNB 200 to the UE 100 (for example, an "RRC Reconfiguration" message, or a newly defined message (for example, an "AI Deployment" message or an "AI Reconfiguration" message, etc.). Alternatively, the configuration message may be a NAS message transmitted from the AMF 300 to the UE 100. Alternatively, when a new layer for performing or controlling machine learning processing (AI / ML processing) is defined, the message may be a message of the new layer.

[0147] In the example of FIG. 19, the setting message includes three models (Model #1 to #3). Each model is included as a container in the setting message. However, the setting message may include only one model. The setting message further includes, as additional information, three individual additional information (Info #1 to #3) provided individually corresponding to each of the three models (Model #1 to #3), and common additional information (Meta-Info) commonly associated with the three models (Model #1 to #3). Each of the individual additional information (Info #1 to #3) includes information unique to the corresponding model. The common additional information (Meta-Info) includes information common to all models in the setting message.

[0148] The individual additional information may be a model index indicating an index (index number) assigned to each model, or may be a model execution condition indicating the performance (e.g., processing delay) required to apply (execute) the model.

[0149] The individual additional information or the common additional information may be a model usage that specifies a function to which a model is to be applied (e.g., "CSI feedback," "beam management," "positioning," etc.). The individual additional information or the common additional information may be a model selection criterion that applies (executes) a corresponding model depending on whether a specified criterion (e.g., a moving speed) is satisfied.

[0150] (2) Communication Control Method According to First Embodiment Next, a communication control method according to the first embodiment will be described.

[0151] In 3GPP, it is agreed that UE 100 may have multiple AI / ML models. However, when UE 100 has multiple AI / ML models, there is no clear guideline as to which model may be used.

[0152] When UE 100 has a predetermined AI / ML model in a predetermined use case (e.g., CSI feedback), the predetermined AI / ML model may be the best model in a certain area but may not be the best model in another area. Also, when UE 100 has a vendor-specific AI / ML model (i.e., a vendor-specific model (proprietary model)) in a predetermined use case (e.g., CSI feedback), the predetermined AI / ML model may be the best model in a RAN area provided by the vendor but may not be the best model in another RAN area provided by another vendor.

[0153] In this way, when UE 100 has a plurality of AI / ML models for the same use case, which model to use (as the best model) may differ depending on the situation or condition in which the model is used.

[0154] Therefore, in the first embodiment, an object is to enable the UE 100 to appropriately perform model inference using an AI / ML model.

[0155] Therefore, in the first embodiment, a user device (e.g., UE100) transmits to a base station (e.g., gNB200) at least one of usage conditions indicating the conditions for using each of the multiple learned models and execution conditions indicating the conditions for performing operations on each of the multiple learned models.

[0156] In the first embodiment, an example will be described in which the UE 100 transmits usage conditions to the gNB 200. For example, the gNB 200 can receive usage conditions from other UEs, and can select the optimal trained model for the UE 100 by taking into consideration not only the usage conditions from the UE 100 but also the usage conditions from other UEs. The UE 100 can perform model inference appropriately using an AI / ML model by performing model inference using the trained model. Note that the execution conditions will be described in the second embodiment.

[0157] (Terminology) Here, the term "AI / ML model" will be explained. An "AI / ML model" is, for example, a data-driven algorithm that applies AI / ML technology to generate a series of outputs based on a series of inputs. The process of learning an "AI / ML model" may be "AI / ML model learning." "AI / ML model learning" is performed, for example, in the model learning unit A2. Furthermore, the process of performing inference using the learned "AI / ML model" is "AI / ML model inference." "AI / ML model inference" is performed, for example, in the model inference unit A3.

[0158] Hereinafter, an "AI / ML model" may be simply referred to as a "model." Also, below, a model that does not use AI / ML technology may be referred to as a "non-AI / ML model" or a "legacy model." An example of a non-AI / ML model is a probability function. When an input value is input to a probability function, an output value can be obtained without using an AI / ML model. Alternatively, in the example of a CSI feedback model, a conventional method of obtaining CSI from CSI-RS measurements using a codebook may be a non-AI / ML model.

[0159] (Conditions of Use According to First Embodiment) Next, a description will be given of conditions of use according to the first embodiment. Items used as conditions of use include, for example, the following.

[0160] In the above items, for example, "region or area" represents the region or area in which the AI / ML model is used as a condition of use for the AI / ML model. For example, if the UE 100 (and / or the gNB 200) performs model learning in the "region or area" and creates a trained model, the "region or area" can become a condition of use.

[0161] Also, for example, if UE 100 creates a trained model using training data acquired using a "frequency or frequency range" (for example, CSI-RS received using frequency f), the "frequency or frequency range" can become a usage condition.

[0162] Furthermore, for example, if UE100 creates a learned model at a "time or time range," the "time or time range" can become a usage condition.

[0163] In this way, each item shown in Table 1 can be expressed as a use condition of the AI / ML model. The use condition may be expressed as a single item. The use condition may also be expressed as a combination of items. The use condition may include all of the items shown in Table 1. Furthermore, the use condition not only indicates the conditions for using the AI / ML model, but also indicates the situation in which the AI / ML model is used.

[0164] FIG. 20 is a diagram illustrating an example of the use conditions according to the first embodiment.

[0165] Fig. 20 shows an example in which UE 100 has multiple AI / ML models for the same use case (e.g., CSI feedback). As shown in Fig. 20, the usage conditions are shown for each AI / ML model. In the example of Fig. 20, the usage conditions for the AI / ML model with model ID = "1" are "usage conditions #1", and the usage conditions for the AI / ML model with model ID = "2" are "usage conditions #2".

[0166] By expressing the usage conditions for each AI / ML model, when UE100 transmits the usage conditions to gNB200, gNB200 can easily determine which model is optimal for each AI / ML model based on the usage conditions.

[0167] (Example of Operation According to First Embodiment) Next, an example of operation according to the first embodiment will be described.

[0168] FIG. 21 is a diagram illustrating an example of operation according to the first embodiment.

[0169] As shown in FIG. 21, in step S501, UE100 transmits usage conditions to gNB200 indicating the conditions for use for each of multiple AI / ML models.

[0170] First, the UE 100 may transmit the usage conditions to the gNB 200 using an RRC message such as a UE Capability message. Alternatively, the UE 100 may transmit the usage conditions to the gNB 200 using a new RRC message related to the AI / ML model (e.g., a Model Registration Request message). Alternatively, a new layer related to the AI / ML model (e.g., an AI / ML layer) may be defined, and the UE 100 may transmit the usage conditions to the gNB 200 using a new message in that layer.

[0171] Secondly, the UE 100 may transmit a model ID that identifies each AI / ML model to the gNB 200 along with the usage conditions. Each AI / ML model is identified by the model ID. The usage conditions may be transmitted in list form for each AI / ML model, as shown in FIG. 20. Furthermore, if the model indicated by the model ID is a non-AI / ML model, information indicating that it is a non-AI / ML model (or a legacy model) may be included in the usage conditions.

[0172] Third, the UE 100 may filter the use conditions. That is, the UE 100 may transmit to the gNB 200 an AI / ML model that matches the use conditions filtered from a plurality of use conditions. For example, if the filtered use condition is "cell ID = xx" ("region or area"), the UE 100 transmits to the gNB 200 an AI / ML model that matches "cell ID = xx". The use conditions to be filtered in the UE 100 may be specified (or set) in advance by the gNB 200 to the UE 100. The use conditions may be transmitted from the gNB 200 to the UE 100 as control data.

[0173] Fourth, the UE 100 may associate a priority with each AI / ML model. FIG. 22(A) is a diagram illustrating an example of an AI / ML model associated with a priority according to the first embodiment. The priority may represent the priority when using the AI / ML model in the UE 100. The priority may be designated (or set) in advance by the gNB 200 to the UE 100 by control data. The UE 100 may determine the priority based on the circumstances of the UE 100 when model learning is performed using the AI / ML model. Examples of the circumstances of the UE 100 include the power consumption, execution time, and learning amount of the UE 100. The UE 100 may also determine the priority based on the circumstances of the AI / ML model. Examples of the circumstances of the AI / ML model include the inference accuracy of the AI / ML model (better than a predetermined accuracy) and the learning time (longer than a predetermined learning time). The UE 100 may determine the priority based on the circumstances of the UE and the circumstances of the AI / ML model. The UE 100 may also transmit the reason for the priority along with the priority to the gNB 200, linking it to each AI / ML model. For example, "Priority: 7, Reason: Fastest execution time." The example shown in FIG. 22(A) shows an example in which the UE 100 transmits the priority to the gNB 200 separately from the usage conditions, but the UE 100 may transmit the priority to the gNB together with the usage conditions. The example shown in FIG. 22(A) shows an example in which "7" is the highest priority and "0" is the lowest priority, but the order may be reversed.

[0174] Fifth, UE100 may transmit to gNB200, together with the usage conditions, retention information indicating whether UE100 holds the AI / ML model. The retention information may also be transmitted to gNB200 in a form linked to each IA / ML model. In this case, UE100 may transmit to gNB200, together with the retention information, either the reason why UE100 holds the AI / ML model or the reason why UE100 does not hold the AI / ML model. UE100 may delete the AI / ML model that it initially held from memory for some reason, such as a memory shortage, and therefore may also transmit the reason to gNB200.

[0175] Sixth, the UE 100 may associate information about Activate with each AI / ML model and transmit it to the gNB 200. FIG. 22(B) is a diagram showing an example of information about Activate according to the first embodiment. Information about Activate includes information indicating that Activate is possible, information indicating that Activate is not possible, and information indicating that Activation is in progress. Activation is possible, for example, indicates that the AI / ML model is in an executable state. On the other hand, activation is not possible, for example, indicates that the AI / ML model is not in an executable state, or that the AI / ML model is not executable. Activation is in progress indicates that the model is currently being executed. The UE 100 may transmit, together with information related to Activate, a reason related to Activate in association with each AI / ML model. For example, "Activate not possible, reason: insufficient memory or model incompatibility (software cannot be executed on the UE 100)."

[0176] Seventh, UE100 may transmit to gNB200 that it will use multiple models. For example, UE100 may use both the company A model and the company B model and notify the one with the higher likelihood (or accuracy). Alternatively, UE100 may use both the company A model and the company B model and notify the result (Mixing) obtained by inputting the result into the company C model.

[0177] The UE 100 may transmit the use conditions in a format in which the use conditions are linked to each model ID ( FIG. 20 ). The UE 100 may transmit the use conditions in a format in which the model ID is linked to each use condition (for example, “model ID #1” for “use condition #1”, “model ID #2, model ID #3” for “use condition #2”, etc.).

[0178] Returning to FIG. 21, in step S502, gNB200 selects (or determines) an AI / ML model to be used by UE100 from among multiple AI / ML models based on the usage conditions received from UE100. The AI / ML model selected by gNB200 may be referred to as a "selected AI / ML model" below. gNB200 may select multiple selected AI / ML models.

[0179] In addition, when handover is performed in UE100, gNB200 may transmit the usage conditions received from UE100 to the target gNB (or target base station) to which UE100 is handed over. For example, (source) gNB200 may transmit the usage conditions to the target gNB using an Xn message such as a handover request (HANDOVER REQUEST) message. In this case, (source) gNB200 may transmit to the target gNB, together with the usage conditions, information received from UE100 (model ID, filtered conditions, priority (and reason for priority), retention information (and reason for retention (or non-retention)), information regarding Activate, and / or using multiple models).

[0180] In step S503, the gNB200 may transmit information regarding the selected AI / ML model to the UE100 as the usage model for the UE100. That is, the gNB200 may specify the selected AI / ML model selected based on the usage conditions to the UE100. The information regarding the selected AI / ML model may include multiple AI / ML models. That is, the gNB200 may specify multiple selected AI / ML models. The information regarding the selected AI / ML model may be transmitted from the gNB200 to the UE100 as control data. The information regarding the selected AI / ML model may be a list in which information indicating that the model is the selected AI / ML model is added to the list of usage conditions shown in FIG. 20. In this list, the "priority" (FIG. 22(A)) may be overwritten to indicate that the model is the selected AI / ML model (for example, the AI / ML model with the highest priority). Alternatively, the "information regarding Activate" (FIG. 22(B)) may be overwritten to indicate that the selected AI / ML model is an AI / ML model (for example, an AI / ML model indicating "activate possible"). When the gNB200 determines that the selected AI / ML model is not held in the UE100, the gNB200 may transmit to the UE100 the selected AI / ML model that the UE100 does not hold (hereinafter, may be referred to as an "unheld AI / ML model"). In this case, the gNB200 may transmit the unheld AI / ML model to the UE100 by executing a model transfer procedure.

[0181] In step S504, the UE 100 transitions to the inference mode. The UE 100 may transition to the inference mode before step S503.

[0182] In step S505, gNB200 transmits partial (or punctured) CSI-RS.

[0183] In step S506, UE100 performs model inference using the selected AI / ML model specified in step S503. If UE100 receives an unheld AI / ML model from gNB200, UE100 performs model inference using the unheld AI / ML model. Then, UE100 transmits the inferred CSI status report to gNB200.

[0184] (Another example according to the first embodiment) In the first embodiment, an example has been described in which the use conditions are transmitted from UE100 to gNB200, but the destination of the use conditions is not limited to gNB200. For example, the use conditions may be transmitted from UE100 to AMF300. In this case, the use conditions are transmitted from UE100 to AMF300 using a NAS message. AMF300 determines the selected AI / ML model based on the use conditions, and transmits information about the selected AI / ML model to UE100. Information about the selected AI / ML model is also transmitted using a NAS message. Settings such as priority (FIG. 22(A)) may also be configured from AMF300 to UE100 using a NAS message.

[0185] The destination of the usage conditions may be an OTT server that provides various content services such as a video distribution service. In this case, the usage conditions, information on the selected AI / ML model, priority, and the like are transmitted using a message based on a predetermined protocol between the UE 100 and the OTT server.

[0186] In this way, the destination of the usage conditions may be a network device including AMF300 and an OTT server, and the usage conditions, information regarding the selected AI / ML model, and priority, etc. are transmitted using messages according to a predetermined protocol between UE100 and the network device.

[0187] (Another Example 2 According to First Embodiment) In the first embodiment, the use case has been described using CSI feedback as an example, but the use case is not limited to this. For example, other use cases such as beam management or position accuracy improvement may be used.

[0188] Second Embodiment Next, a second embodiment will be described, focusing on the differences from the first embodiment.

[0189] The second embodiment is an embodiment regarding execution conditions. The execution conditions indicate, for example, conditions for executing an operation for each of a plurality of trained models. The "operation" is any one of activation for an AI / ML model, deactivation for an AI / ML model, switching of an AI / ML model, and fallback for an AI / ML model. The execution conditions indicate, for example, conditions for executing each of these operations for each AI / ML model.

[0190] Activation of an AI / ML model means, for example, that the AI / ML model is activated and ready to run. Activation of an AI / ML model may also mean an execution state in which the AI / ML model is running. Deactivation of an AI / ML model means, for example, that the AI / ML model is not ready to run, or that the AI / ML model is not ready to run.

[0191] Furthermore, switching the AI / ML model refers to, for example, switching from a first AI / ML model to a second AI / ML model that is an AI / ML model different from the first AI / ML model, and fallback of the AI / ML model refers to, for example, switching from an AI / ML model to a non-AI / ML model (or a legacy model).

[0192] 23A and 23B are diagrams showing examples of execution conditions according to the second embodiment.

[0193] 23(A) shows an example of an execution condition of activation (or deactivation). As shown in FIG. 23(A), the execution condition of activation (or deactivation) for the AI / ML model with model ID = "1" is "TAI = xx", that is, when the UE 100 enters the tracking area with TAI = xx, activation (or deactivation) for the AI / ML model with model ID = "1" is executed.

[0194] 23B shows an example of a switching (or fallback) execution condition. For example, when FIG. 23B shows the switching execution condition, the execution condition for switching the AI / ML model with model ID = "1" is "likelihood < x". That is, when the likelihood of the AI / ML model with model ID = "1" is lower than "x", the UE 100 switches from the AI / ML model with model ID = "1" to another AI / ML model. Also, when FIG. 23B shows the fallback execution condition, the execution condition for the fallback of the AI / ML model with model ID = "1" is "likelihood < x". That is, when the likelihood of the AI / ML model with model ID = "1" is lower than "x", the UE 100 performs fallback from the AI / ML model with model ID = "1" to a non-AI / ML model.

[0195] The likelihood is, for example, an index representing likelihood. The higher the likelihood, the higher the likelihood of the AI / ML model, and the lower the likelihood, the lower the likelihood of the AI / ML model. The likelihood may represent the likelihood of the AI / ML model when compared with a non-AI / ML model.

[0196] In this way, by expressing the execution conditions for each AI / ML model, when UE100 executes each operation, it is possible to report to gNB200 for each AI / ML model. In gNB200, it is also possible to easily grasp the operating status of each AI / ML model in UE100 for each AI / ML model. In addition, in gNB200, it is also possible to specify the AI / ML model to be used by UE100, or to transmit an AI / ML model that UE100 does not hold (non-held AI / ML model) to UE100. Therefore, UE100 can appropriately perform model inference using the AI / ML model.

[0197] (Example of Operation According to Second Embodiment) Next, an example of operation according to the second embodiment will be described.

[0198] FIG. 24 is a diagram illustrating an example of operation according to the second embodiment.

[0199] As shown in FIG. 24, in step S601, UE100 transmits execution conditions to gNB200.

[0200] First, the UE 100 may transmit the execution condition to the gNB 200 using an RRC message such as a UE capability message, as in the first embodiment. The UE 100 may transmit the execution condition using a new RRC message (e.g., a model request message) related to the AI / ML model. Alternatively, a new layer (e.g., an AI / ML layer) for the AI / ML model may be defined, and the UE 100 may transmit the execution condition using a new message in that layer.

[0201] Secondly, the UE 100 may transmit a model ID that identifies each AI / ML model to the gNB 200 along with the execution conditions. The execution conditions may be transmitted in list form for each AI / ML model, for example, as shown in FIG. 23(A) or 23(B). In this case, identification information indicating which of the four operations is for each AI / ML model may also be added to the list. For example, in FIG. 23(A), information indicating "activation" or "deactivation" may be added to the list, and in FIG. 23(B), information indicating "switching" or "fallback" may be added to the list.

[0202] Third, the execution condition may be expressed by the use condition according to the first embodiment. For example, in the example of FIG. 23A, the execution condition is expressed as a condition using the "region or area" item of the use condition. The execution condition may include a likelihood, as shown in FIG. 23B. The likelihood may represent the accuracy of the AI / ML model relative to the non-AI / ML model.

[0203] Fourth, a priority order may be associated with the execution condition. FIG. 25 is a diagram illustrating an example of a priority order according to the second embodiment. The priority order indicates, for example, the order in which the AI / ML models are executed when there are multiple AI / ML models that satisfy the execution condition. The example of FIG. 25 indicates that when the execution condition is "likelihood < x" (when the likelihood is less than x), each AI / ML model is executed according to the priority order shown in FIG. 25. As in the first embodiment, the priority order may be specified (or set) in advance from the gNB 200 to the UE 100 using, for example, control data. Furthermore, as in the first embodiment, the priority order may be determined based on the circumstances of the UE 100 when the UE 100 performs model learning using the AI / ML model. Alternatively, the UE 100 may determine the priority order based on the circumstances of the AI / ML model. Each of the circumstances may be the same as in the first embodiment. The UE 100 may determine the priority order based on the circumstances of the UE and the circumstances of the AI / ML model. Furthermore, the UE 100 may also transmit the reason for the priority along with the priority to the gNB 200, linking it to each AI / ML model. For example, "Priority: 7, Reason: Fastest execution time". Note that the example shown in Figure 25 represents an example in which the UE 100 transmits the priority to the gNB 200 separately from the execution conditions, but the UE 100 may also transmit the priority to the gNB together with the execution conditions. Also, the example shown in Figure 25 represents an example in which "7" is the highest priority and "0" is the lowest priority, but the order may be reversed.

[0204] Fifth, UE100 may transmit the holding information to gNB200 together with the execution conditions, as in the first embodiment. Together with the holding information, UE100 may transmit to gNB200 either the reason why the AI ​​ / ML model is held or the reason why the AI ​​ / ML model is not held.

[0205] Sixth, the gNB 200 may specify the AI / ML model to be used by the UE 100 to the UE 100 based on the execution conditions. The specification may be performed by using a list overwritten with the list received as the execution conditions (step S601). For example, if the execution conditions include a priority, the AI / ML model to be used by the UE 100 may be specified by the overwritten priority (e.g., the AI / ML model with the highest priority), as in the first embodiment. The specification may be transmitted from the gNB 200 to the UE 100 using control data as information regarding the selected AI / ML model, as in the first embodiment.

[0206] The UE 100 may transmit the execution conditions in a format in which the execution conditions are linked to each model ID (e.g., FIG. 23A). The UE 100 may transmit the execution conditions in a format in which the model IDs linked to each execution condition (e.g., "model ID #1" for "execution condition #1", "model ID #2, model ID #3" for "execution condition #2").

[0207] Returning to FIG. 24 , in step S602, when UE100 executes an operation for the AI / ML model, it transmits execution information to gNB200. The execution information is, for example, information indicating that UE100 has executed an operation for the AI / ML model. The execution information may indicate that activation has been executed, deactivation has been executed, switching has been executed, or fallback has been executed for each AI / ML model. The gNB200 may instruct UE100 to execute the operation. The execution of the operation may also be transmitted from gNB200 to UE100 using control data. The execution of the operation may include the model ID of the target AI / ML model. The UE100 may execute the operation for the AI / ML model in accordance with the execution instruction for the operation.

[0208] In addition, when handover is performed in UE100, gNB200 may transmit the execution conditions (step S601) and / or execution information (step S602) received from UE100 to the target gNB (or target base station) to which UE100 is handed over. For example, (source) gNB200 may transmit the execution conditions and / or execution information to the target gNB using an Xn message such as a handover request (HANDOVER REQUEST) message. In this case, (source) gNB200 may transmit to the target gNB200 the information received from UE100 (model ID, priority (and reason for priority), retention information (and reason for retention (or non-retention)), and / or information on the AI ​​ / ML model that gNB200 has specified for UE100) together with the execution conditions and / or execution information.

[0209] First, UE100 may transmit execution information to gNB200 when the operation is performed.

[0210] Secondly, when the UE 100 executes the operation, it may store (log) the execution information in memory and then transmit it collectively to the gNB 200. In this case, the UE 100 may transmit the stored execution information to the gNB 200 in accordance with a request from the gNB 200. The UE 100 may periodically transmit the stored execution information to the gNB 200. The gNB 200 may instruct the UE 100 to store the execution information and then transmit it to the gNB 200. The request and instruction from the gNB 200 may be transmitted to the UE 100 using control data. In this way, by the UE 100 storing the execution information and transmitting it collectively later, communication efficiency can be improved compared to when the UE 100 transmits the execution information each time it executes the operation.

[0211] In addition, gNB200 may decide to transmit the non-retained AI / ML model to UE100 based on the execution information. For example, if gNB200 determines that "non-activation" continues for a specific AI / ML model, it may determine that the non-retained AI / ML model should be used as a substitute for the specific AI / ML model. In this case, gNB200 may transmit the non-retained AI / ML model to UE100 by executing a model transmission procedure, as in the first embodiment.

[0212] In step S603, the UE 100 transitions to the inference mode. The transition to the inference mode may occur before the transmission of the execution information (step S602).

[0213] In step S604, gNB200 transmits partial CSI-RS to UE100, and UE100 receives the CSI-RS.

[0214] In step S605, the UE 100 infers the CSI status report using, for example, an AI / ML model that satisfies "activation" among the execution conditions. If a priority is indicated, the UE 100 may infer the CSI status report using the AI / ML model with the highest priority. Alternatively, the UE 100 may infer the CSI status report using an AI / ML model specified by the gNB 200. The UE 100 may infer the CSI status report using a non-retained AI / ML model received from the gNB 200.

[0215] (Another example 1 according to the second embodiment) In the second embodiment, the destination of the execution condition has been described as gNB200, but this is not limited to this. For example, the destination of the execution condition may be AMF300. The destination may be an OTT server. The destination may be another network device. When the destination of the execution condition is AMF300, the execution condition, execution information, priority, etc. may be transmitted using a NAS message. Furthermore, when the destination of the execution condition is a network device including an OTT server, the execution condition, execution information, priority, etc. may be transmitted using a message according to a predetermined protocol between the UE100 and the network device.

[0216] (Another Example 2 According to Second Embodiment) In the second embodiment, the use case has been described using CSI feedback as an example, but the use case is not limited to this. For example, a use case such as beam management or position accuracy improvement may be used.

[0217] [Other Embodiments] In the first and second embodiments described above, supervised learning has been mainly described, but the present invention is not limited to this. For example, the first and second embodiments may be applied to unsupervised learning or reinforcement learning.

[0218] The above-described operational flows are not limited to being implemented independently, but can also be implemented by combining two or more operational flows. For example, some steps of one operational flow may be added to another operational flow, or some steps of one operational flow may be replaced with some steps of another operational flow. In each flow, it is not necessary to execute all steps, and only some steps may be executed.

[0219] In the above-described embodiments and examples, an example in which the base station is an NR base station (gNB) has been described, but the base station may be an LTE base station (eNB) or a 6G base station. 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 the IAB node. The UE 100 may also be an MT (Mobile Termination) of the IAB node.

[0220] 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). A network node may also be configured by a combination of at least a part of a core network device and at least a part of a base station.

[0221] Furthermore, a program (e.g., an information processing program) that causes a computer to execute each process or function according to the above-described embodiments may be provided. Alternatively, a program (e.g., a mobile communication program) that causes the mobile communication system 1 to execute each process or function according to the above-described embodiments may be provided. The program may be recorded on a computer-readable medium. Using a computer-readable medium, the program can be installed on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transitory recording medium. The non-transitory recording medium is not particularly limited, and may be, for example, a recording medium such as a CD-ROM or DVD-ROM. Such a recording medium may be memory included in the UE 100 and the gNB 200. Furthermore, circuits that execute each process performed by the UE 100 or the gNB 200 may be integrated, and at least a portion of the UE 100 or the gNB 200 may be configured as a semiconductor integrated circuit (chip set, SoC: System on a chip).

[0222] The functions performed by the UE 100 or the gNB 200 (network node) may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (Central Processing Units), conventional circuits, and / or combinations thereof, programmed to perform the described functions. A processor includes transistors and other circuits and is considered to be circuitry or processing circuitry. A processor may also be a programmed processor that executes a program stored in memory. In this specification, circuitry, unit, or means refers to hardware that is programmed to perform the described functions or hardware that executes them. The hardware may be any hardware disclosed herein or any hardware known to be programmed or capable of performing the described functions. If the hardware is a processor, the circuitry, means, or unit is a combination of hardware and software used to configure the hardware and / or processor.

[0223] As used in this disclosure, the terms "based on" and "depending on / in response to" do not mean "based only on" or "depending only on," unless expressly stated otherwise. The term "based on" means both "based only on" and "based at least in part on." Similarly, the term "depending on" means both "depending only on" and "depending at least in part on." The terms "include," "comprise," and variations thereof do not mean including only the listed items, but may mean including only the listed items or may include additional items in addition to the listed items. Additionally, the term "or," as used in this disclosure, is not intended to mean an exclusive or. Furthermore, any reference to elements using designations such as "first," "second," etc., as used in this disclosure does not generally limit the quantity or order of those elements. These designations may be used herein as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed therein or that the first element must precede the second element in some way. In this disclosure, where articles are added by translation, such as a, an, and the in English, these articles shall include the plural unless the context clearly indicates otherwise.

[0224] Although the embodiments have been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design changes can be made without departing from the spirit of the invention. Furthermore, it is also possible to combine the embodiments, operation examples, or processes within a consistent range.

[0225] This application claims priority from Japanese Patent Application No. 2023-021077 (filed February 14, 2023), the entire contents of which are incorporated herein by reference.

[0226] (Supplementary Note) (Supplementary Note 1) A communication control method in a mobile communication system, comprising a step in which a user device transmits to a network node at least one of use conditions indicating conditions for using each of a plurality of AI / ML models and execution conditions indicating conditions for executing an operation for each of the plurality of AI / ML models.

[0227] (Supplementary Note 2) The communication control method according to Supplementary Note 1, wherein the transmitting step includes a step in which the user equipment transmits to the network node an AI / ML model that matches a usage condition filtered from among the plurality of usage conditions.

[0228] (Supplementary Note 3) The communication control method according to Supplementary Note 1 or Supplementary Note 2, wherein the transmitting step includes a step in which the network node sets the use condition for filtering in the user equipment.

[0229] (Supplementary Note 4) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 3, wherein the transmitting step includes a step in which the user equipment transmits the usage conditions and a priority for each of the plurality of AI / ML models to the network node.

[0230] (Supplementary Note 5) The communication control method according to any one of Supplementary Notes 1 to 4, wherein the transmitting step includes a step in which the user equipment transmits the priority and a reason for the priority to the network node.

[0231] (Supplementary Note 6) The communication control method according to any one of Supplementary Notes 1 to 5, wherein the transmitting step includes a step in which the user device transmits, to the network node, the usage conditions and retention information indicating whether the user device retains the AI / ML model for each of the plurality of AI / ML models.

[0232] (Supplementary Note 7) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 6, wherein the transmitting step includes a step in which the user equipment transmits, to the network node, the retained information and either the reason why the user equipment holds the AI / ML model or the reason why the user equipment does not hold the AI / ML model.

[0233] (Supplementary Note 8) The communication control method according to any one of Supplementary Notes 1 to 7, further comprising: a step in which the network node specifies a selected AI / ML model selected from the plurality of AI / ML models to the user device based on the usage conditions; and a step in which the user device performs model inference using the selected AI / ML model.

[0234] (Supplementary Note 9) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 8, further comprising the steps of: the network node transmitting, to the user equipment, an unheld AI / ML model that is not held in the user equipment based on the usage conditions; and the user equipment performing model inference using the unheld AI / ML model.

[0235] (Supplementary Note 10) The communication control method according to any one of Supplementary Notes 1 to 9, further comprising the step of: the network node transmitting the conditions of use to a target network node to which the user equipment is to be handed over.

[0236] (Supplementary Note 11) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 10, wherein the operation is one of activation of the AI / ML model, deactivation of the AI / ML model, switching of the AI / ML model, and fallback of the AI / ML model.

[0237] (Supplementary Note 12) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 11, wherein the execution condition includes a likelihood or accuracy of the AI / ML model relative to a non-AI / ML model.

[0238] (Supplementary Note 13) The communication control method according to any one of Supplementary Notes 1 to 12, wherein the execution condition is expressed by the use condition.

[0239] (Supplementary Note 14) The communication control method according to any one of Supplementary Notes 1 to 13, wherein the transmitting step includes a step in which the user equipment transmits, to the network node, the execution condition and a priority for the AI / ML model that satisfies the execution condition.

[0240] (Supplementary Note 15) The communication control method according to any one of Supplementary Notes 1 to 14, further comprising the step of: when the user equipment has executed the operation on the AI / ML model, transmitting execution information indicating that the operation has been executed to the network node.

[0241] (Supplementary Note 16) The communication control method according to any one of Supplementary Notes 1 to 15, further comprising the step of the network node instructing the user equipment to transmit the execution information.

[0242] (Supplementary Note 17) The communication control method according to any one of Supplementary Notes 1 to 16, further comprising the step of: the user device transmitting the execution information to the network node after recording the execution information.

[0243] (Supplementary Note 18) The communication control method according to any one of Supplementary Notes 1 to 17, further comprising the step of instructing the user equipment to transmit the execution information to the network node after the network node has recorded the execution information.

[0244] 1: Mobile communication system 20: 5GC (CN) 100: UE 110: Receiving unit 120: Transmitting unit 130: Control unit 200: gNB 210: Transmitting unit 220: Receiving unit 230: Control unit A1: Data collection unit A2: Model learning unit A3: Model inference unit A4: Data processing unit TE: Transmitting entity RE: Receiving entity

Claims

1. A communication control method in a mobile communication system, comprising: The user equipment transmits to the network node at least one of a use condition indicating a condition for using each of a plurality of AI (Artificial Intelligence) / ML (Machine Learning) models and an execution condition indicating a condition for executing an operation for each of the plurality of AI / ML models. Communication control method.

2. The transmitting step includes transmitting, to the network node, an AI / ML model that matches a filtered use condition from among the plurality of use conditions, by the user equipment. The communication control method according to claim 1.

3. The transmitting step includes the network node setting the filtering conditions to the user equipment. The communication control method according to claim 2.

4. The transmitting step includes the user equipment transmitting the conditions of use and a priority for each of the plurality of AI / ML models to the network node. The communication control method according to claim 1.

5. The transmitting step includes the user equipment transmitting the priority and a reason for the priority to the network node.

5. The communication control method according to claim 4.

6. The transmitting includes the user equipment transmitting, to the network node, the use conditions and retention information indicating whether the user equipment retains the AI / ML model for each of the plurality of AI / ML models. The communication control method according to claim 1.

7. The transmitting step includes the user equipment transmitting, to the network node, either a reason why the user equipment holds the AI / ML model or a reason why the user equipment does not hold the AI / ML model, and the holding information.

7. The communication control method according to claim 6.

8. The network node assigns a selected AI / ML model selected from the plurality of AI / ML models to the user equipment based on the usage conditions; and performing model inference using the selected AI / ML model. The communication control method according to claim 1.

9. The network node transmits to the user equipment an unheld AI / ML model that is not held in the user equipment based on the usage conditions; The user device further comprises performing model inference using the unretained AI / ML model. The communication control method according to claim 1.

10. The network node may further transmit the conditions of use to a target network node to which the user equipment is to be handed over. The communication control method according to claim 1.

11. The operation is one of activating the AI / ML model, deactivating the AI / ML model, switching the AI / ML model, and falling back the AI / ML model. The communication control method according to claim 1.

12. The execution conditions include the likelihood or accuracy of the AI / ML model relative to a non-AI / ML model. The communication control method according to claim 11.

13. The execution conditions are expressed by the use conditions. The communication control method according to claim 11.

14. The transmitting step includes the user equipment transmitting the execution condition and a priority for the AI / ML model that satisfies the execution condition to the network node. The communication control method according to claim 11.

15. When the user equipment executes the operation on the AI / ML model, the user equipment transmits execution information indicating that the operation has been executed to the network node. The communication control method according to claim 11.

16. The network node further instructs the user equipment to transmit the execution information. The communication control method according to claim 15.

17. The method further includes the user equipment transmitting the performance information to the network node after recording the performance information. The communication control method according to claim 14.

18. The network node may further include, after recording the performance information, instructing the user equipment to transmit the performance information to the network node.

18. The communication control method according to claim 17.

19. A network node in a mobile communication system, comprising: The system includes a receiving unit that receives from a user device at least one of a use condition indicating a condition for using each of a plurality of AI (Artificial Intelligence) / ML (Machine Learning) models and an execution condition indicating a condition for executing an operation for each of the plurality of AI / ML models. Network node.

20. A user equipment in a mobile communication system, comprising: The system includes a transmitter that transmits to a network node at least one of a use condition indicating a condition for using each of a plurality of AI (Artificial Intelligence) / ML (Machine Learning) models and an execution condition indicating a condition for executing an operation for each of the plurality of AI / ML models. User equipment.

Images