Communication control method

Abstract

A communication control method according to one aspect of the present invention is a communication control method for a mobile communication system. The communication control method includes a step in which a node transmits a transmission request, which requests transmission of an AI / ML model, to another node. In addition, the communication control method includes a step in which the node receives the AI / ML model from the other node. Here, the transmission request includes an application condition. The node also receives mapping information that satisfies the application condition. The mapping information indicates a correspondence relationship between the AI / ML model and collection data for use in the AI / ML model.

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) of mobile communication systems.

[0003] 3GPP TR 38.843 V18.0.0 (2023-12)

[0004] A communication control method according to a first aspect is a communication control method in a mobile communication system. The communication control method includes a step in which a node transmits a transmission request requesting transmission of an AI / ML model to another node. The communication control method also includes a step in which the node receives the AI / ML model from the other node. Here, the transmission request includes an application condition. The node also receives mapping information that satisfies the application condition. The mapping information indicates a correspondence between the AI / ML model and collected data used in the AI / ML model.

[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 AI / ML technology according to the first embodiment. FIG. 7 is a diagram showing an example of operation in AI / ML technology according to the first embodiment. FIG. 8 is a diagram showing an example of the arrangement of functional blocks of AI / ML technology according to the first embodiment. FIG. 9 is a diagram showing an example of operation according to the first embodiment. FIG. 10 is a diagram showing an example of operation according to the first embodiment. FIG. 11 is a diagram showing an example of a configuration message according to the first embodiment. FIG. 12 is a diagram showing a first example of operation according to the first embodiment. FIG. 13 is a diagram showing an example of mapping information according to the first embodiment. FIG. 14 is a diagram showing a second example of operation according to the first embodiment. FIG. 15 is a diagram showing a third example of operation according to the first embodiment. FIG. 16 is a diagram showing a fourth example of operation according to the first embodiment. Fig. 17 is a diagram illustrating a fifth operation example according to the first embodiment. Fig. 18 is a diagram illustrating a seventh operation example according to the first embodiment. Fig. 19 is a diagram illustrating an eighth operation example according to the first embodiment. Fig. 20 is a diagram illustrating an operation example according to the second embodiment.

[0006] The present disclosure aims to properly identify AI / ML models.

[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. Furthermore, devices included in the core network CN may be referred to as core network devices.

[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 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 UPF 300 may be core network devices included in the CN 20. The core network device and the gNB 200 may be collectively referred to as a network device.

[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. Alternatively, the gNB 200 may be an example of a network node.

[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 (Data Collection) A1, a model training unit (Model Training) A2, a model inference unit (Inference) A3, a management unit (Management) A5, and a model storage unit (Model Storage) A6.

[0038] The functional block configuration example shown in FIG. 6 represents a functional framework of a general AI / ML technology. Therefore, depending on a hypothetical use case, some of the functional block configuration example (e.g., model recording unit A6, etc.) may not be included in the functional block configuration example. The functional block configuration example shown in FIG. 6 may also be distributed between the UE 100 and a network-side device. Alternatively, some functions of the functional block configuration example (e.g., model learning unit A2 or model inference unit A3, etc.) may be located in both the UE 100 and the network-side device.

[0039] The data collection unit A1 provides input data to the model learning unit A2, the model inference unit A3, and the management unit A5. The input data includes training data for the model learning unit A2, inference data for the model inference unit A3, and monitoring data for the management unit A5.

[0040] The training data is data required as input when the AI / ML model is learning. The inference data is data required as input when the AI / ML model is inferring. The monitoring data is data required as input when the AI / ML model is managing.

[0041] In addition, data collection may refer to the process of collecting data at a network node, a management entity, or a UE 100, for example, to train an AI / ML model, manage an AI / ML model, and perform inference on an AI / ML model.

[0042] The model learning unit A2 performs AI / ML model training, AI / ML model validation, and AI / ML model testing.

[0043] AI / ML model learning is the process of learning an AI / ML model from input / output relationships to obtain a trained AI / ML model to be used for inference. For example, considering y = ax + b, AI / ML model learning may be the process of optimizing a (slope) and b (intercept) by providing input (x) and output (y) (i.e., providing learning data).

[0044] Generally, machine learning includes supervised learning, unsupervised learning, and reinforcement learning. Supervised learning is a method that uses correct answer data as training data. Unsupervised learning is a method that does not use correct answer data as training data. For example, unsupervised learning memorizes feature points from a large amount of training data and determines the correct answer (estimates the range). Reinforcement learning is a method that assigns a score to the output result and learns how to maximize the score. Although supervised learning will be described below, either unsupervised learning or reinforcement learning may be applied as machine learning.

[0045] The model learning unit A2 outputs a trained AI / ML model (Trained Model) obtained by AI / ML model learning to the model recording unit A6, and also outputs an updated AI / ML model (Updated Model) obtained by relearning the trained AI / ML model to the model recording unit A6.

[0046] In the following, AI / ML model learning may be referred to as "model learning" or "learning."

[0047] The model inference unit A3 performs AI / ML model inference. Specifically, the model inference unit A3 applies the inference data provided by the data collection unit A1 to the trained AI / ML model (or updated AI / ML model) to obtain inference output data. For example, in the equation y = ax + b, x corresponds to the inference data and y corresponds to the inference output data. Note that "y = ax + b" is an AI / ML model. A model with optimized slope and intercept, for example, "y = 5x + 3," is a trained AI / ML model. There are various model approaches, including linear regression analysis, neural networks, and decision tree analysis. The above "y = ax + b" can also be considered a type of linear regression analysis.

[0048] The model inference unit A3 outputs inference output data to the management unit A5. The model inference unit A3 also receives management instructions from the management unit A5. For example, the management instructions include selection of an AI / ML model, activation (deactivation) of an AI / ML model, switching of an AI / ML model, and fallback (performing inference without using an AI / ML model). The model inference unit A3 performs model inference in accordance with the management instructions.

[0049] Note that AI / ML model inference is, for example, a process of obtaining a set of outputs from a set of inputs using a trained AI / ML model (or an updated AI / ML model). Alternatively, model inference may be a process of obtaining inference output data from inference data using a trained AI / ML model (or an updated AI / ML model). Hereinafter, AI / ML model inference may be referred to as "model inference" or "inference."

[0050] In the following, an AI / ML model that is currently being trained (or updated) may be referred to as a training AI / ML model (or an updating AI / ML model). In the following, when there is no need to distinguish between a training (or updating) AI / ML model and a trained (or updated) AI / ML model, they may be simply referred to as an "AI / ML model."

[0051] The management unit A5 supervises operations on the AI / ML model (selection, activation, deactivation, switching, fallback, etc.). The management unit A5 also supervises monitoring of the AI / ML model. The management unit A5 can also perform operations to ensure appropriate inference operations based on monitoring data and inference output data. To this end, the management unit A5 outputs a model transfer and / or model delivery request (Model Transfer / Delivery Request) to the model recording unit A6, and causes the trained (or updated) AI / ML model recorded in the model recording unit A6 to be output to the model inference unit A3. The management unit A5 also outputs management instructions to the model inference unit A3 and supervises operations on the AI / ML model. Furthermore, the management unit A5 can output performance feedback and a re-learning request to the model learning unit A2, causing the model learning unit A2 to re-learn the AI / ML model (i.e., update the learned AI / ML model).

[0052] FIG. 7 is a diagram illustrating an example of operation in the AI / ML technique according to the first embodiment.

[0053] In Fig. 7, the transmitting entity TE is an entity capable of performing model inference and transmitting inference output data to the receiving entity RE. Meanwhile, the receiving entity RE is an entity capable of receiving inference output data from the transmitting entity TE. Model training may be performed in the transmitting entity TE. The model training may also be performed in the receiving entity RE. If the model training is performed in the receiving entity RE, the trained AI / ML model may be transmitted from the receiving entity RE to the transmitting entity TE.

[0054] The entity may be, for example, a device, a functional block included in the device, or a hardware block included in the device.

[0055] 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.

[0056] 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., an AI / ML layer) dedicated to artificial intelligence or machine learning. Alternatively, the control data may be a NAS message in the NAS layer. The control data may include a performance feedback request and / or a re-learning request transmitted from the management unit A5 to the model learning unit A2. Alternatively, the control data may include a model transfer request and / or a model delivery request sent from the management unit A5 to the model recording unit A6, or a management instruction sent from the management unit A5 to the model inference unit A3.

[0057] (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.

[0058] For example, there are three use cases in which AI / ML technology is applied:

[0059] (X1.1) "CSI (Channel State Information) Feedback Enhancement"

[0060] (X1.2) "Beam management"

[0061] (X1.3) “Positioning accuracy enhancement”

[0062] (X1.1) Example of functional block arrangement in "CSI feedback improvement" "CSI feedback improvement" represents a use case in which AI / ML technology is applied to CSI fed back from UE100 to gNB200, for example. 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.

[0063] 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.

[0064] 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).

[0065] 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.

[0066] 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).

[0067] 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.

[0068] 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.

[0069] 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.

[0070] FIG. 9 is a diagram illustrating an example of operation in "CSI feedback improvement" according to the first embodiment.

[0071] 9, in step S10, the gNB 200 may notify or set the CSI-RS transmission pattern (puncture pattern) in the inference mode to the UE 100 as control data. 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.

[0072] In step S11, gNB200 may send a switching notification to UE100 to start learning mode.

[0073] In step S12, the UE 100 starts a learning mode.

[0074] In step S13, 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 uses the full CSI-RS and the CSI as learning data to create a learned AI / ML model.

[0075] In step S14, UE100 transmits the generated CSI to gNB200.

[0076] Thereafter, in step S15, 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.

[0077] In step S16, in response to receiving the completion notification, gNB200 sends a switching notification to UE100 to switch UE100 from learning mode to inference mode.

[0078] In step S17, in response to receiving the switching notification, the UE 100 switches from the learning mode to the inference mode.

[0079] In step S18, 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.

[0080] In step S19, 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.

[0081] In addition, in step S20, if UE100 determines that model learning is necessary, it may send a notification indicating that model learning is necessary to gNB200 as control data.

[0082] In the example shown in Fig. 9, 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".

[0083] 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:

[0084] (Y1) 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 or other received signals received from gNB200).

[0085] (Y2) 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.)

[0086] (Y3) 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 (Y1) to (Y3). 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 (Y1) to (Y3). 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.

[0087] An example of the arrangement of functional blocks in (X1.1) "CSI feedback" has been described above. The above-mentioned arrangement example is just one example, and in 3GPP, the arrangement example of functional blocks is still in the process of being studied. Similarly, (X1.2) "Beam management" and (X1.3) "Position accuracy improvement" are also still in the process of being studied.

[0088] (X1.4) Example of Model Transfer Next, we will explain the transfer of an AI / ML model (Model Transfer). Note that the terms "transfer of an AI / ML model" and "delivery of an AI / ML model" have the same meaning.

[0089] (X1.4.1) First operation pattern related to model forwarding Figure 10 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 10, the receiving entity RE will be described as mainly being the UE 100, but the receiving entity RE may be the gNB 200 or the AMF 300. Also, in the example shown in Figure 10, the transmitting entity TE will be described as being the gNB 200, but the transmitting entity TE may be the UE 100 or the AMF 300.

[0090] 10, in step S25, the gNB 200 transmits a capability inquiry message to the UE 100 to request transmission of a message including an information element (IE) indicating the execution capability for the learning process. The UE 100 receives the capability inquiry message. However, the gNB 200 may transmit the capability inquiry message when it executes the learning process (when it determines that it will execute the learning process).

[0091] In step S26, the UE 100 transmits to the gNB 200 a message including an information element indicating execution capabilities for the learning process (in another respect, the execution environment for the learning process). 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 is defined for performing or controlling the learning process (AI / ML process), the message may be a message of the new layer.

[0092] The information element indicating the execution capability related to the learning process may be an information element indicating the capability of a processor for executing the learning process and / or an information element indicating the capability of a memory for executing the 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.

[0093] Alternatively, the information element indicating the execution capability of the learning process may be an information element indicating the execution capability of the inference process (model inference). Specifically, the information element indicating the execution capability of the inference process 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 process.

[0094] Alternatively, the information element indicating the execution capability related to the 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.

[0095] In step S27, gNB200 determines the model to be configured (or deployed) in UE100 based on the information elements contained in the message received in step S26.

[0096] In step S28, gNB200 transmits a message including the model determined in step S27 to UE100. UE100 receives the message and performs a learning process (i.e., a model learning process and / or a model inference process) using the model included in the message. A specific example of step S28 will be described in the following second operation pattern.

[0097] (X1.4.2) Second Operation Pattern Related to Model Transfer FIG. 11 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.

[0098] In the example of FIG. 11, 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.

[0099] 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.

[0100] 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.

[0101] (Model Identification According to First Embodiment) Next, model identification according to the first embodiment will be described.

[0102] Currently, 3GPP is discussing model identification of AI / ML models. For example, in the framework shown in FIG. 6 , model identification may be performed during model training. Specifically, model identification may be performed in UE 100 (or a network device) when it is determined to which AI / ML model training data corresponds. Alternatively, model identification may be performed during model inference. Specifically, model identification may be performed in UE 100 (or a network device) when it is determined to which AI / ML model inference data corresponds. Alternatively, model identification may be performed during management. Specifically, model identification may be performed in UE 100 (or a network device) when it is determined to which AI / ML model training data corresponds.

[0103] In 3GPP, model identification is classified into two types, Type A and Type B. Type A is, for example, a type in which model identification is performed without using signaling between the UE 100 and the network device. On the other hand, Type B is, for example, a type in which model identification is performed via signaling.

[0104] In addition, 3GPP classifies Type B model identification into the following three options.

[0105] (Option 1) Model identification accompanying data collection settings and / or instructions

[0106] (Option 2) Model identification accompanying dataset transmission

[0107] (Option 3) Model Identification in Model Transfer from Network to UE 100 In the first embodiment, the description will be focused on (Option 1). Note that there are also the following options 4 and 5, but they have not yet been agreed upon in 3GPP.

[0108] (Option 4) Model identification via reference model

[0109] (Option 5) Model Identification via Model Monitoring

[0110] In the case of (Option 1), for example, when data is collected in the UE 100 or the network device, it is assumed that model identification is performed when determining which AI / ML model the collected data corresponds to. In the UE 100 or the network device, model identification makes it possible to input the collected data to a specific AI / ML model as learning data or inference data, for example.

[0111] Therefore, the first embodiment aims to appropriately identify the AI / ML model.

[0112] Therefore, in the first embodiment, first, a node (e.g., UE 100) transmits a transmission request requesting transmission of an AI / ML model to another node (e.g., a network device). Second, the node receives the AI / ML model from the other node. Here, the transmission request includes application conditions. Furthermore, the node receives mapping information that satisfies the application conditions. The mapping information indicates a correspondence relationship between the AI / ML model and collected data used in the AI / ML model.

[0113] As described above, in the first embodiment, for example, the UE 100 transmits a model transmission request including application conditions to a network device and receives mapping information that satisfies the application conditions from the network device. The UE 100 can easily identify (or specify) an AI / ML model that corresponds to collected data by using the mapping information. At this time, the UE 100 receives mapping information that satisfies the application conditions, and therefore, for example, the UE 100 does not receive mapping information related to an AI / ML model that is not expected to be used depending on the location where the UE 100 is located. Therefore, the UE 100 can appropriately identify an AI / ML model that corresponds to collected data based on the mapping information.

[0114] In the following, a device or entity included in the core network (5GC20) in the mobile communication system 1 may be referred to as a "core network device." Also, in the following, in the mobile communication system 1, the gNB200 and the core network device may be referred to as a "network device." The network device may be an LPP (LTE Positioning Protocol) server. The network device may be an OAM (Operations, Administration and Maintenance) server. Furthermore, the network device and the UE100 may be referred to as a "node." The node may be the UE100 or a core network device.

[0115] In the following, model learning and model inference may be referred to as "learning" and "inference", respectively. Furthermore, when there is no need to distinguish between a training AI / ML model and a trained AI / ML model, they may be simply referred to as an "AI / ML model".

[0116] An example of operation according to the first embodiment will be described below in the following order.

[0117] (1) UE-sided model

[0118] (1-1) First Operation Example: UE-Identified and NW-Initiated

[0119] (1-2) Second Operation Example: UE-Identified and UE-Initiated

[0120] (1-3) Third Operation Example: NW-Identified and NW-Initiated

[0121] (1-4) Fourth Operation Example: NW-Identified and UE-Initiated

[0122] (1-5) Fifth Operation Example: NW-Identified and Logged MDT (Minimalization of Drive Tests) Model Collection

[0123] (2) Network-side model (NW-sided model): Sixth operation example

[0124] (3) Two-sided model: NW identification and UE identification: Seventh operation example

[0125] (3-1) Eighth Operation Example: NW Identification: Eighth Operation Example The "UE side model" is, for example, a model in which inference is performed on the UE 100 side. On the other hand, the "network side model" is, for example, a model in which inference is performed on the network device side.

[0126] Furthermore, "UE identification" indicates that model identification is performed on the UE 100 side, for example. On the other hand, "NW identification" indicates that model identification is performed on the network device side, for example.

[0127] Furthermore, "NW start" means, for example, that processing is started from a network device. On the other hand, "UE start" means, for example, that processing is started from the UE 100. However, in the first embodiment, for model identification (Option 1) accompanying data collection, the case where data collection is started from a network device is described as "NW start", and the case where data collection is started from the UE 100 is described as "UE start".

[0128] (1-1) First Operation Example: UE-Identified (UE-Identified) and NW-Initiated (NW-Initiated) First, a first operation example will be described. That is, an operation example in which inference is performed on the UE 100 side (UE-side model), model identification is performed on the UE 100 side (UE identification), and processing is initiated from the UE 100 side (UE-initiated) will be described.

[0129] FIG. 12 is a diagram illustrating a first operation example according to the first embodiment. Note that FIG. 12 illustrates an OTT (Over The Top) server. The OTT server is an external server located outside the mobile communication system 1. A network device may be used instead of the OTT server. In the first operation example, the OTT server is described as holding an AI / ML model (e.g., a trained AI / ML model). The same applies to the OTT server in the following operation examples.

[0130] As shown in Fig. 12, in step S30, the transmitter 120 of the UE 100 transmits a model transmission request to the OTT server. The model transmission request is an example of a transmission request requesting transmission of an AI / ML model. In the first embodiment, the transmitter 120 transmits the model transmission request including application conditions. The application conditions represent, for example, conditions when the collected data is used in the AI / ML model. Specifically, they are as follows:

[0131] First, the application condition may include a geographical condition. The geographical condition may be expressed by latitude and altitude. Alternatively, the geographical condition may be expressed by latitude, altitude, and altitude. Alternatively, the geographical condition may be expressed by a PCI (Physical Cell ID), a TAI (Tracking Are Identifier), an RA (Registration Area), or a PLMN (Public Land Mobile Network). For example, if "PCI=A" is an application condition, it indicates that data collected at PCI=A is applied to the AI / ML model.

[0132] Second, the application conditions may include a time condition. The time condition may be expressed by time information based on date and time. The time condition may be expressed by a time period. For example, when the time is included in the application conditions, it indicates that data collected at that time is applied to the AI / ML model.

[0133] Third, the application condition may include the movement speed of the UE 100. For example, when the application condition is "movement speed = 30 km / h", it indicates that collected data collected at a movement speed = 30 km / h is applied to the AI / ML model.

[0134] Fourth, the application conditions may include at least one of the following pieces of information:

[0135] (B1) Use cases (e.g., CSI feedback improvement)

[0136] (B2) Performance (or KPI) of the AI / ML model (e.g., inference accuracy (%) or NMSE (Normalized Mean Squared Error))

[0137] (B3) AI / ML model development history or learning history

[0138] (B4) Signaling (e.g., reference signal (RS)) or technical conditions (e.g., codebooks applied to the AI / ML model) applied to the AI / ML model

[0139] (B5) AI / ML model size

[0140] (B6) Computing power required to run AI / ML models

[0141] (B7) AI / ML model format (open format, proprietary model, or standard library type)

[0142] (B8) Reference AI / ML model The UE 100 requests mapping information that satisfies the application conditions by transmitting a model transmission request including the application conditions. The UE 100 may request transmission of an AI / ML model that satisfies the application conditions by transmitting a model transmission request including the application conditions. The OTT server receives the model transmission request including the application conditions.

[0143] In step S31, in response to receiving the model transmission request, the OTT server transmits mapping information together with the AI / ML model to the UE 100. The AI / ML model to be transmitted is a model that satisfies the application conditions. The mapping information to be transmitted is mapping information that satisfies the application conditions.

[0144] The mapping information indicates the correspondence between the AI / ML model and the collected data used in the AI / ML model. Fig. 13 shows an example of the mapping information according to the first embodiment.

[0145] First, the mapping information uses a model ID as an AI / ML model corresponding to the collected data. In the example shown in Fig. 13, the mapping information indicates a one-to-one relationship between the collected data and the model ID. The mapping information only needs to indicate the correspondence between the collected data and the model ID, and multiple model IDs may correspond to one collected data. In the mapping information, multiple collected data may correspond to one model ID.

[0146] Second, the mapping information may include an application condition. The application condition may be an application condition transmitted by the UE 100.

[0147] Third, the mapping information may include a priority for each AI / ML model. The priority indicates the priority of which AI / ML model should be applied when multiple model IDs correspond to one collection data according to the mapping information. For example, in the example of FIG. 13, when the collection data is "L1-RSRP*4", two models correspond: the AI / ML model "#BM1" and the AI / ML model "#BM2". In this case, since the priority of "#BM1" is higher than that of "#BM#2", the AI / ML model "#BM1" is applied. The application (or selection) of the AI / ML model based on the priority is performed by the UE 100.

[0148] The receiver 110 of the UE 100 receives the AI / ML model and the mapping information.

[0149] 12, in step S32, the transmitter of the network device (for example, the transmitter 210 of the gNB 200) transmits a mapping transmission request including application conditions to the OTT server. In step S32, since the network device has not acquired the mapping information acquired by the UE 100, at this stage, the network device performs processing to acquire the mapping information.

[0150] The mapping transmission request is a request for acquiring mapping information that satisfies an application condition. The application condition may be the same as the application condition transmitted by the UE 100. Specifically, the application condition may be at least one of the following:

[0151] (C1) Geographical and / or temporal conditions

[0152] (C2) Any of the above (B1) to (B4). Note that the application conditions transmitted by the network device have a certain relationship with the application conditions (step S30) transmitted by the UE 100, even if the content itself is different. For example, if the network device is a gNB 200, the UE 100 exists under the gNB 200. Therefore, when the application conditions are geographical conditions, it is assumed that the application conditions transmitted by the UE 100 and the application conditions transmitted by the network device have a certain relationship, such as the former being included in the latter. Even when the application conditions are temporal conditions, it is also assumed that the two application conditions are included within a certain range. Therefore, it is possible to assume that the mapping information (step S31) transmitted by the OTT server to the UE 100 and the mapping information (step S33) transmitted to the network device are not completely unrelated, but that a certain relationship exists. Therefore, the mapping information acquired by the UE 100 and the mapping information acquired by the network device have a certain (or common) relationship, and it is possible to identify the same model ID from the collected data in both the UE 100 and the network device. In the following, it is assumed that there is such a relationship.

[0153] The OTT server receives a mapping transmission request including the application conditions (step S32).

[0154] In step S33, in response to receiving the mapping transmission request, the OTT server transmits mapping information that satisfies the application conditions to the network device. The OTT server may transmit to the network device the same mapping information as the mapping information transmitted to the UE 100. A receiver of the network device (for example, the receiver 220 of the gNB 200) receives the mapping information.

[0155] In step S34, the control unit of the network device (e.g., the control unit 230 of the gNB 200) decides to use the AI / ML model. Note that in step S34, the control unit of the network device does not specifically identify the AI / ML model, but may instead decide the application for which the AI / ML model is to be used.

[0156] In step S35, the transmitting unit of the network device transmits a data collection instruction to the UE 100. The transmitting unit of the network device may identify the collection data corresponding to the use determined in step S34 based on the mapping information received in step S33. The data collection instruction may include information indicating what kind of information the data to be collected is. For example, this information may be "L1-RSRP" or "power information". The data collection instruction may include a plurality of pieces of this information. For example, the data collection instruction includes "L1-RSRP" and "power information". The receiving unit 110 of the UE 100 receives the data collection instruction.

[0157] In step S36, the control unit 130 of the UE 100 collects collection data in response to receiving the data collection instruction. The control unit 130 determines whether the collection data is used in an AI / ML model based on the mapping information. The control unit 130 may identify which AI / ML model the collection data corresponds to by using the mapping information. That is, the control unit 130 performs model identification using the mapping information. The model identification may be determining whether the collection data is used in an AI / ML model. The model identification may be determining which AI / ML model the collection data is used in. When the collection data corresponds to multiple AI / ML models, the control unit 130 may identify one of the AI / ML models using a priority order. The control unit 130 may identify the AI / ML model in which the collection data is used by using a model ID.

[0158] In step S37, the transmitting unit 120 of the UE 100 transmits the identified model ID to the network device. If the UE 100 does not hold an AI / ML model corresponding to the data collection instruction (step S35), the transmitting unit 120 of the UE 100 may transmit a message to that effect to the network device. The receiving unit of the network device receives the model ID.

[0159] In step S38, in response to receiving the model ID, the control unit of the network device identifies the AI / ML model corresponding to the model ID and transmits a start-of-use instruction for the AI / ML model to the UE 100. The start-of-use instruction may include the model ID. The receiving unit of the UE 100 receives the start-of-use instruction.

[0160] In step S39, in response to receiving the instruction to start use, control unit 130 of UE 100 starts using the AI / ML model.

[0161] In a first operation example, the UE 100 may operate as a node, and the network device and the OTT server may operate as other nodes. In this case, model identification is performed in the UE 100, which is a node.

[0162] (1-2) Second Operation Example: UE-Identified and UE-Initiated Next, a second operation example will be described. The second operation example is an operation example in which inference is performed on the UE 100 side (UE-side model), model identification is performed on the UE 100 side (UE identification), and processing is initiated from the UE 100 side (UE-initiated).

[0163] FIG. 14 is a diagram illustrating a second operation example according to the first embodiment.

[0164] In FIG. 14, steps S40 and S41 are the same as steps S30 and S31, respectively, of the first operation example.

[0165] In step S42, the control unit 130 of the UE 100 determines to use the AI / ML model.

[0166] In step S43, the control unit 130 of the UE 100 collects collection data. Since the second operation example is an operation example in the case of UE-initiated, the control unit 130 of the UE 100 starts the data collection process at its own discretion. The control unit 130 identifies which AI / ML model the collected collection data corresponds to using mapping information. That is, the control unit 130 performs model identification using the mapping information. The control unit 130 identifies the model ID of the AI / ML model. The subsequent processes (steps S44 to S46), including step S43, are the same as those in the first operation example.

[0167] In the second operation example, the UE 100 may also operate as a node, and the network device and the OTT server may also operate as other nodes. Model identification is performed in the UE 100, which is a node.

[0168] (1-3) Third Operation Example: NW Identified (NW-Identified) and NW Initiated (NW-Initiated) Next, a third operation example according to the first embodiment will be described. The third operation example is an operation example in which inference is performed on the UE 100 side (UE side model), model identification is performed on the network device side (NW identification), and processing is initiated from the network device side (NW-Initiated).

[0169] FIG. 15 is a diagram illustrating a third operation example.

[0170] Step S50 is the same as step S30 in the first operation example.

[0171] In step S51, the OTT server transmits the AI / ML model to the UE 100. In this case, the model identification is performed on the network device side, so the OTT server does not transmit mapping information. The OTT server transmits the model ID of the AI / ML model to be transmitted.

[0172] Steps S52 and S53 are the same as steps S32 and S33 in the first operation example. The network device has the mapping information, and the UE 100 does not have the mapping information.

[0173] In step S54, the control unit of the network device decides to use the AI / ML model.

[0174] In step S55, the control unit of the network device identifies collected data. The control unit may identify the target collected data from the collected data included in the mapping information.

[0175] In step S56, the transmitting unit of the network device transmits a data collection instruction to the UE 100. The data collection instruction may include information indicating the collection data specified by the network device. The receiving unit 110 of the UE 100 receives the data collection instruction.

[0176] In step S57, the control unit 130 of the UE 100 collects data in response to receiving the data collection instruction, and the transmission unit 120 of the UE 100 transmits the collected data to the network device. If the UE 100 does not hold an AI / ML model corresponding to the data collection instruction, the transmission unit 120 of the UE 100 may transmit a message to that effect to the network device. The reception unit of the network device receives the collected data.

[0177] In step S58, the control unit of the network device identifies which AI / ML model the collected data corresponds to, that is, performs model identification.

[0178] In step S59, the control unit of the network device transmits a start-of-use instruction for the identified AI / ML model to the UE 100. The start-of-use instruction includes the model ID of the AI / ML model specified (or identified) by the model identification. The receiving unit 110 of the UE 100 receives the start-of-use instruction.

[0179] In step S60, in response to receiving the instruction to start use, control unit 130 of UE 100 starts using the specified AI / ML model.

[0180] In the third operation example, the network device may operate as a node, and the OTT server may operate as another node. In this case, model identification (step S58) is performed in the network device that is the node.

[0181] (1-4) Fourth Operation Example: NW-Identified and UE-Initiated Next, a fourth operation example will be described. That is, an operation example in which inference is performed on the UE 100 side (UE-side model), model identification is performed on the network device side (NW identification), and processing is initiated from the UE 100 side (UE-initiated) will be described.

[0182] FIG. 16 is a diagram illustrating a fourth operation example according to the first embodiment.

[0183] Steps S70 and S71 shown in Fig. 16 are the same as steps S50 and S51 (Fig. 15) of the third operation example, respectively. Steps S72 and S73 shown in Fig. 16 are the same as steps S52 and S53 (Fig. 15) of the third operation example, respectively. The network device has mapping information, and the UE 100 does not have mapping information.

[0184] In step S74, the control unit 130 of the UE 100 determines to use the AI / ML model.

[0185] In step S75, the control unit 130 of the UE 100 performs data collection. Since the fourth operation example is a UE-initiated model, the data collection process is initiated by the UE 100.

[0186] In step S76, the transmitting unit 120 of the UE 100 transmits the collected data to the network device, and the receiving unit of the network device receives the collected data.

[0187] In step S77, the control unit of the network device identifies to which AI / ML the collected data collected by the UE 100 corresponds, i.e., performs model identification. The control unit of the network device identifies the AI / ML model using the mapping information received in step S73.

[0188] In step S78, the transmitting unit of the network device transmits a usage start instruction for the identified AI / ML model to the UE 100. The subsequent operations are the same as those in the third operation example.

[0189] In the fourth operation example, the network device may operate as a node, and the OTT server may operate as another node. In this case, model identification (step S77) is performed in the network device that is the node.

[0190] (1-5) Fifth Operation Example: NW-Identified and Logged MDT Next, a fifth operation example will be described. In the fifth operation example, inference is performed on the UE side (UE-side model), model identification is performed on the network device side (NW identification), and model collection is performed in the logged MDT.

[0191] FIG. 17 is a diagram illustrating a fifth operation example according to the first embodiment.

[0192] In step S90, UE 100 is in an RRC connected state with gNB 200. However, it is sufficient that UE 100 is in the RRC connected state at a timing before the timing of step S95.

[0193] Steps S91 to S94 are the same as steps S30 to S33 in the first operation example, respectively. The UE 100 and the network device acquire mapping information.

[0194] In step S95, the network device configures the logged MDT for the UE 100. When the network device is a gNB 200, the transmitter 210 of the gNB 200 may configure the logged MDT by transmitting a LoggedMeasurementConfiguration message (RRC message). In configuring the logged MDT, collection of data used in the AI / ML model may be instructed. The instruction may include information regarding the data to be collected.

[0195] In step S96, the UE 100 transitions to an RRC idle state.

[0196] In step S97, the control unit 130 of the UE 100 acquires a log in accordance with the log MDT setting. The acquired log may include log data used in the AI / ML model and log data not used in the AI / ML model. Therefore, the control unit 130 determines whether the acquired log data (which is also collected data) is used in the AI / ML model using the mapping information received in step S92. Furthermore, if the acquired log data is used in the AI / ML model, the control unit 130 may determine which AI / ML model the log data is used in. The control unit 130 generates additional information indicating whether the log data is used in the AI / ML model. The control unit 130 may associate the additional information with the log data. The additional information may be represented by 1-bit information. Alternatively, the additional information may include a model ID indicating which AI / ML model the log data is used in. The control unit 130 may store the log data and the additional information in a memory within the UE 100.

[0197] In step S98, UE100 transitions to an RRC connected state with gNB200.

[0198] In step S99, the transmitter 120 of the UE 100 transmits the log data and the additional information to the network device, and the receiver of the network device receives the log data and the additional information.

[0199] In step S100, the control unit of the network device records the additional information in memory. The control unit of the network device may record the additional information in memory in association with the log data.

[0200] In step S101, the control unit of the network device decides to use the AI / ML model. The control unit of the network device may use additional information to confirm that collected data used in the AI / ML model has been acquired in UE 100, thereby deciding to use the AI / ML model. The control unit of the network device may use the additional information to determine which AI / ML model is supported. Model identification may be performed through the process of deciding to use the AI / ML model. Steps S102 and onward are the same as steps S78 and S79 (FIG. 16) of the fourth operation example.

[0201] In the fifth operation example, the network device may operate as a node, and the OTT server may operate as another node (or the UE 100 may operate as another node). In this case, model identification (step S101) is performed in the network device that is a node.

[0202] (1-5-1) Other Examples of the Fifth Operation Example In the fifth operation example, the logged MDT has been described as an example, but this is not limited to this. For example, immediate MDT (Immediate MDT) is also applicable. In this case, the UE 100 maintains the RRC connected state with the gNB 200, and the process shown in FIG. 17 is performed. In addition, instead of the logged MDT, any of the processes shown below may be performed.

[0203] (1-5-1-1) Layer 3 Measurement (for example, Measurement Report): The log transmission in step S99 can be the transmission of a measurement report.

[0204] (1-5-1-2) Layer 1 Measurement (e.g., CSI Report): The log transmission in step S99 can be a CSI report.

[0205] (1-5-1-3) UE Assistance Information (UAI) or UE Capability Information: In this case, these RRC messages are transmitted instead of the log transmission in step S99.

[0206] (1-5-1-4) Early measurement: When the UE 100 performs measurement in the RRC idle state and enters the RRC connected state, the UE 100 transmits the measurement result instead of transmitting the log in step S99.

[0207] (1-5-1-5) LTE Position Protocol (LPP): Instead of transmitting the log in step S99, location information is transmitted. In this case, the network device becomes an LPP server.

[0208] (2) Sixth Operation Example Next, a sixth operation example will be described. The sixth operation example is an example in which inference is performed on the network device side (network side model).

[0209] In the case of the network-side model, if the network device can acquire mapping information from the OTT server (for example, steps S72 and S73 in FIG. 16 ), the network device can perform model identification, and thereafter can determine the use of the AI / ML model by itself. Although it is assumed that data collection is performed in the UE 100, if the network device side instructs which data to collect, there is no particular need for the UE 100 to perform model identification.

[0210] (3) Seventh Operation Example: Two-Side Model, NW Identification, and UE Identification Next, a seventh operation example will be described. The seventh operation example is an example in which inference is performed on both the UE 100 side and the network device side (two-side model), and model identification is performed on both the network side and the UE side (NW identification and UE identification).

[0211] Fig. 18 is a diagram illustrating a seventh operation example according to the first embodiment. Fig. 18 illustrates an example of "CSI feedback improvement" as a use case in which the AI / ML model is used.

[0212] In the seventh operation example, a pair ID is used to link the AI / ML model used on the UE 100 side with the AI / ML model used on the network device side. The model ID is global identification information, so it is identification information that is longer than a certain length. On the other hand, the pair ID is identification information for linking two AI / ML models, so it is identification information that is shorter than the model ID.

[0213] Steps S110 and S111 shown in FIG. 18 are the same as steps S30 and S31 (FIG. 12) of the first operation example. However, in the mapping information, a pair ID may be used instead of a model ID. Steps S112 and S113 are also the same as steps S32 and S33 of the first operation example, but a pair ID may be used instead of a model ID as the mapping information. The UE 100 and the network device may acquire mapping information including a pair ID. In the following description, a case where a pair ID is used will be described as an example.

[0214] In step S114, the control unit of the network device determines the use of an AI / ML model. The AI / ML model may be determined to be used not only for the AI / ML model of the network device itself, but also for the AI / ML model present on the UE 100 side.

[0215] In step S115, the transmitting unit of the network device transmits a data collection instruction to the UE 100. The control unit of the network device may identify the collection data to be used in the AI / ML model determined to be used in step S114 by using the mapping information. The transmitting unit of the network device may transmit a data collection instruction for the identified collection data. The receiving unit 110 of the UE 100 receives the data collection instruction.

[0216] In step S116, the control unit 130 of the UE 100 collects the data instructed by the data collection instruction and performs model identification using the mapping information. At this time, the control unit 130 may identify the AI / ML model that uses the collected data by using a pair ID. By identifying the pair ID, the control unit 130 may determine that the AI / ML model that uses the collected data is usable in the UE 100.

[0217] In step S117, the transmitter 120 of the UE 100 transmits model usability information including the pair ID to the network device. The model usability information is information indicating that the AI / ML model indicated by the pair ID can be used. The receiver of the network device receives the model usability information including the pair ID.

[0218] In step S118, the control unit of the network device identifies the AI / ML model indicated by the pair ID. Model identification is also performed in the network device.

[0219] In step S119, the transmitting unit of the network device transmits a start-of-use instruction for the AI / ML model indicated by the pair ID to the UE 100. The start-of-use instruction may include the pair ID of the pair to be started. The receiving unit 110 of the UE 100 receives the start-of-use instruction.

[0220] In step S120, in response to receiving the instruction to start use, control unit 130 of UE 100 starts using the AI / ML model.

[0221] In step S121, the transmitting unit 120 of the UE 100 transmits compressed CSI data to the network device. The compressed CSI data is data obtained by compressing the CSI inferred by the UE 100 using the AI / ML model. The receiving unit of the network device receives the compressed CSI data.

[0222] In step S122, the control unit of the network device starts using the AI / ML model. Since the example shown in Fig. 18 is for the use case of "CSI feedback enhancement," the control unit of the network device uses the AI / ML model for the compressed CSI data to obtain expanded (or decompressed) CSI.

[0223] In the seventh operation example, the UE 100 and the network device may operate as nodes, and the OTT server may operate as another node. In this case, model identification (steps S116 and S118) is performed in the UE 100 and the network device, which are nodes.

[0224] (3-1) Eighth Operation Example Next, an eighth operation will be described. In the eighth operation example, as in the seventh operation example, inference is performed in the UE 100 and the network device (two-sided model), but unlike the seventh operation example, model identification is not performed on the UE 100 side, and model identification (NW identification) is performed on the network device side. As in the seventh operation example, the use case is an example of "CSI feedback improvement."

[0225] FIG. 19 is a diagram illustrating an eighth operation example according to the first embodiment.

[0226] Steps S130 to S133 in Fig. 19 are the same as steps S110 to S113 (Fig. 18) in the seventh operation example. Pair IDs are also used in Fig. 9. Mapping information is acquired by both the UE 100 and the network device.

[0227] In step S134, the network device configures the CSI report for the UE 100. The CSI report configuration may include information regarding data to be collected. When the network device is the gNB 200, the configuration may be performed using an RRC message including a CSI report configuration (CSI-ReportConfiguration) as an information element. The transmitter of the network device transmits a message regarding the CSI report configuration, and the receiver 110 of the UE 100 receives the message.

[0228] In step S135, the control unit 135 of the UE 100 collects data to be reported as a CSI in accordance with the CSI report setting. At this time, the control unit 135 may use the mapping information to determine whether the collected data is used in an AI / ML model, as in the fifth operation example, and generate additional information as a result of the determination. Alternatively, the control unit 135 may use the mapping information to include in the additional information a pair ID of the AI / ML model in which the collected data is used.

[0229] In step S136, the transmitting unit 120 of the UE 100 transmits the CSI report (collected data) and the additional information to the network device.

[0230] In step S137, the control unit of the network device records the additional information in memory.

[0231] In step S138, the control unit of the network device decides to use the AI / ML model.

[0232] In step S139, the control unit of the network device specifies (or identifies) the pair ID of the AI / ML model that has been determined to be used. The control unit of the network device may specify the pair ID by using the mapping information received in step S133.

[0233] In step S140, the transmitting unit of the network device transmits a start-of-use instruction for the identified AI / ML model to the UE 100. The start-of-use instruction may include a pair ID of the target AI / ML model. The receiving unit 110 of the UE 100 receives the start-of-use instruction.

[0234] In step S141, the control unit 130 of the UE 100 starts using the AI / ML model that is the target of the start of use instruction. The subsequent processing is the same as that in the seventh operation example (steps S121 and S122).

[0235] (Another Operation Example 1 According to the First Embodiment) In the first embodiment, an example has been described in which requests, settings, data, information, and the like are transmitted between the UE 100 and the network device. The transmission may be transmitted, for example, using control data ( FIG. 7 ). Alternatively, the transmission may be transmitted using a message conforming to the protocol of a layer newly established for AI / ML (for example, an AI / ML layer). Transmission between the network device and the OTT server may also be performed using control data (or a new message for AI / ML).

[0236] (Another Operation Example 2 According to First Embodiment) In the first embodiment, an example has been described in which a model ID is used to identify an AI / ML model. A function ID may be used to identify the AI / ML model. Alternatively, a local ID that can be recognized between the UE 100 and the network device may be used.

[0237] Second Embodiment Next, a second embodiment will be described.

[0238] In the first embodiment, it is assumed that UE-initiated and NW-initiated are determined in advance. In the second embodiment, UE-initiated and NW-initiated are determined by negotiation between nodes.

[0239] Specifically, first, a user equipment (e.g., UE 100) receives a framework information transmission request from a network device. Second, in response to receiving the frame network information transmission request, the user equipment transmits framework information to the network device. Here, the framework information includes information about the framework when the AI / ML model is executed.

[0240] This allows the network device to grasp, for example, what kind of framework the UE 100 has, and to determine UE initiation or NW initiation based on the framework. Therefore, the network device can appropriately determine whether to perform UE initiation or NW initiation. Furthermore, the UE initiation or NW initiation determined in this manner can also be used to perform the operations described in the first embodiment. Therefore, in the second embodiment, as in the first embodiment, it is also possible to appropriately identify the AI / ML model.

[0241] Here, examples of the framework used in the second embodiment are as follows.

[0242] (D1) UE-side model, network-side model, or two-side model

[0243] (D2) Physical entity or address information corresponding to each entity in the functional framework shown in FIG. 6

[0244] (D3) Whether the AI / ML model is an open format or a proprietary format

[0245] (D4) Supported model identification method (one of Option 1 to Option 5)

[0246] (D5) UE identification or NW identification, UE initiation or NW initiation ("UE initiation or NW initiation" may be used when it is determined in advance which one will be initiated). (D6) Supported LCM method (e.g., model ID-based LCM or function-based LCM).

[0247] (D7) Supported model transfer method (for example, a case number indicating the model transfer method specified in "Table 4.3-1" of 3GPP TR38.843 V18.0.0) The framework information indicates information about the framework when the AI / ML model is executed. The framework information may include at least any of the above (D1) to (D7). The framework information may also include information about frameworks other than the above (D1) to (D7).

[0248] (Example of Operation According to Second Embodiment) Next, an example of operation according to the second embodiment will be described.

[0249] Fig. 20 is a diagram illustrating an example of operation according to the second embodiment. In Fig. 20, for example, the operation is performed under the following assumptions.

[0250] That is, there is a UE-side OTT server connected to the UE 100 and a NW-device-side OTT server connected to the network device. The UE-side OTT server may have framework information of the UE 100. Furthermore, the NW-device-side OTT server may have framework information of the network device. However, the network device cannot necessarily connect to the UE-side OTT server, and the UE 100 cannot necessarily connect to the network-device-side OTT server. Therefore, it is assumed that neither the network device nor the UE 100 knows what framework the other has.

[0251] 20, the network device transmits a model transmission request to the network device-side OTT server (step S150), and the network device-side OTT server transmits the AI / ML model to the network device (step S151). The receiving unit of the network device receives the AI / ML model.

[0252] The UE 100 also transmits a model transmission request to the UE-side OTT server (step S152), and the UE-side OTT server transmits the AI / ML model to the UE 100. The receiving unit 110 of the UE 100 receives the AI / ML model. At this stage, both the UE 100 and the network device have AI / ML models, but are not yet aware of the framework under which the other party's AI / ML model is executed.

[0253] The order of steps S152 and S153 and steps S150 and S151 may be reversed.

[0254] In step S154, the UE 100 makes an initial connection to the network device. If the network device is the gNB 200, the control unit of the UE 100 performs a random access procedure with the gNB 200, and further performs a setup procedure to establish an RRC connection.

[0255] In step S155, the transmitting unit of the network device transmits a framework information transmission request to the UE 100. The receiving unit 110 of the UE 100 receives the framework information transmission request.

[0256] In step S156, the transmitter 120 of the UE 100 transmits the framework information to the network device in response to receiving the framework information transmission request. The control unit 130 of the UE 100 may determine the framework based on the AI / ML model received in step S153. For example, depending on the AI / ML model, it is possible that some models can only be executed as a UE-side model, and some models can only be executed as a network-side model. The receiver of the network device receives the framework information.

[0257] In step S157, the control unit of the network device determines the use of the framework based on the framework information. For example, the control unit of the network device may determine here whether the AI / ML model should be UE-initiated or NW-initiated. The control unit of the network device may determine whether the AI / ML model should be UE-initiated or NW-initiated based on the framework information received in step S156.

[0258] In step S158, the transmitting unit of the network device transmits a framework use instruction to the UE 100. The use instruction may include information indicating the determined framework. The use instruction may include information indicating whether UE initiation or NW initiation should be performed. The receiving unit 110 of the UE 100 receives the use instruction. As a result, the UE 100 acquires information regarding UE initiation or NW initiation determined by the network device. Thereafter, the UE 100 and the network device may perform the operation according to the first embodiment in accordance with UE initiation or NW initiation.

[0259] (Another Operation Example According to the Second Embodiment) In the second embodiment, control data may be used for transmission between the UE 100 and the network device. For the transmission, a new message according to a protocol newly defined for AI / ML may be used.

[0260] [Other Embodiments] In the first embodiment described above, supervised learning has been mainly described, but the present invention is not limited to this. For example, unsupervised learning or reinforcement learning may be applied to the first embodiment.

[0261] 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.

[0262] 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.

[0263] That is, the UE 100 may be a terminal function unit (a type of communication module) for a base station to control a repeater that relays signals. Such a terminal function unit is referred to as an MT. Examples of the MT include, in addition to the IAB-MT, an NCR (Network Controlled Repeater)-MT and a RIS (Reconfigurable Intelligent Surface)-MT.

[0264] 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.

[0265] A program may be provided that causes a computer to execute each process performed by the UE 100, the gNB 200, or the network device 400. 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 and / or a DVD-ROM. Furthermore, circuits that execute each process performed by the UE 100, the gNB 200, or the network device 400 may be integrated, and at least a portion of the UE 100, the gNB 200, or the network device 400 may be configured as a semiconductor integrated circuit (chip set, SoC: System on a chip).

[0266] 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.

[0267] The functions performed by the UE 100 or base station 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), a CPU (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to perform the described functions. A processor includes transistors and / or 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 a memory.

[0268] In this specification, a circuitry, unit, or means is hardware that is programmed to realize or performs the described functions, which may be any hardware disclosed herein or any hardware known to be programmed to realize or perform the described functions.

[0269] If the hardware is a processor considered to be a type of circuitry, the circuitry, means, or unit is a combination of hardware and software used to configure the hardware and / or processor.

[0270] Although one embodiment has 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 within the scope of the gist. Furthermore, the embodiments, operation examples, and processes can be appropriately combined within the scope of not being inconsistent.

[0271] This application claims priority from Japanese Patent Application No. 2024-061007 (filed April 4, 2024), the entire contents of which are incorporated herein by reference.

[0272] (Addendum) The above can be summarized as follows.

[0273] (Supplementary Note 1) A communication control method in a mobile communication system, comprising: a step in which a node transmits a transmission request requesting transmission of an AI / ML model to another node; and a step in which the node receives the AI / ML model from the other node, wherein the transmission request includes an application condition, and the receiving step includes a step in which the node receives mapping information that satisfies the application condition, and the mapping information indicates a correspondence between the AI / ML model and collected data used in the AI / ML model.

[0274] (Supplementary Note 2) The communication control method according to Supplementary Note 1, wherein the application conditions indicate conditions under which the collected data is used in the AI / ML model.

[0275] (Supplementary Note 3) The communication control method according to Supplementary Note 1 or Supplementary Note 2, wherein the receiving step includes the node receiving the AI / ML model that satisfies the application condition.

[0276] (Supplementary Note 4) The communication control method according to any one of Supplementary Notes 1 to 3, wherein the mapping information includes the application condition.

[0277] (Supplementary Note 5) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the mapping information includes a priority for each of the AI / ML models.

[0278] (Supplementary Note 6) The communication control method according to any one of Supplementary Notes 1 to 5, further comprising the step of the node performing model identification using the mapping information.

[0279] (Supplementary Note 7) The communication control method according to any one of Supplementary Notes 1 to 6, further comprising: a step of the node collecting the collected data; and wherein the step of performing model identification includes the node performing model identification on the collected data by using the mapping information.

[0280] (Supplementary Note 8) The communication control method according to any one of Supplementary Note 1 to Supplementary Note 7, wherein the step of performing model identification includes a step in which, when the collected data corresponds to a plurality of the AI / ML models, the node performs the model identification using the priority order.

[0281] (Supplementary Note 9) A communication control method, wherein the node is a user device and the other node is a network device, further comprising: a step in which the user device receives a framework information transmission request from the network device; and a step in which the user device transmits framework information to the network device in response to receiving the frame network information transmission request, wherein the framework information includes information regarding a framework when the AI / ML model is executed.

[0282] 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 250: Backhaul communication unit 400: Network device

Claims

1. A communication control method in a mobile communication system, comprising: a node transmitting a transmission request to another node requesting the transmission of an AI (Artificial Intelligence) / ML (Machine Learning) model; and the node receiving the AI / ML model from the other node, wherein the transmission request includes application conditions; and the receiving includes the node receiving mapping information that satisfies the application conditions, and the mapping information indicates a correspondence between the AI / ML model and collected data used in the AI / ML model.

2. The communication control method according to claim 1, wherein the application conditions indicate conditions under which the collected data is used in the AI / ML model.

3. The communication control method according to claim 1, wherein said receiving includes said node receiving said AI / ML model that satisfies said application condition.

4. The communication control method according to claim 1, wherein the mapping information includes the application conditions.

5. The communication control method according to claim 1, wherein the mapping information includes a priority for each of the AI / ML models.

6. The communication control method according to claim 1, further comprising the step of: said node performing model identification using said mapping information.

7. The communication control method according to claim 5, further comprising the node collecting the collected data, and performing the model identification includes the node performing model identification on the collected data using the mapping information.

8. The communication control method according to claim 5 or claim 7, wherein performing the model identification includes the node performing the model identification using the priority order when the collected data corresponds to a plurality of the AI / ML models.

9. A communication control method, wherein the node is a user device and the other node is a network device, further comprising: the user device receiving a framework information transmission request from the network device; and the user device transmitting framework information to the network device in response to receiving the frame network information transmission request, wherein the framework information includes information regarding the framework when an AI / ML model is executed.

Images