Communication control method, base station, and user equipment

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 network node divides a trained AI / ML model into a first divided trained model and a second divided trained model. The communication control method also includes a step in which the network node transmits a first message including the first divided trained model to user equipment. The communication control method furthermore includes a step in which the user equipment acquires position information indicating the position of the user equipment. The communication control method furthermore includes a step in which the user equipment uses the first divided trained model to acquire output result data from the position information. The communication control method furthermore includes a step in which the user equipment transmits a second message including the output result data to the network node. The communication control method furthermore includes a step in which the network node uses the second divided trained model to infer, from the output result data, whether a mobility operation has been executed. The mobility operation includes at least an operation in which a network node transmits a handover request message to another network node.

Description

Communication control method, base station, and user equipment

[0001] The present disclosure relates to a communication control method, a base station, and a user device.

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

[0003] RP-213599, “New SI: Study on Artificial Intelligence (AI) / Machine Learning (ML) for NR Air Interface”RWS-230240, “Motivations on AI / ML-based mobility enhancement”

[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 network node divides a trained AI / ML model into a first divided trained model and a second divided trained model. The communication control method also includes a step in which the network node transmits a first message including the first divided trained model to a user equipment. The communication control method further includes a step in which the user equipment acquires location information indicating a location of the user equipment. The communication control method also includes a step in which the user equipment acquires output result data from the location information using the first divided trained model. The communication control method also includes a step in which the user equipment transmits a second message including the output result data to the network node. The communication control method also includes a step in which the network node infers whether or not a mobility operation is to be performed from the output result data using the second divided trained model. Here, the mobility operation includes at least a step in which the network node transmits a handover request message to another network node.

[0005] A network node according to a second aspect is a network node in a mobile communication system. The network node has a control unit that divides a trained AI / ML model into a first divided trained model and a second divided trained model. The network node also has a transmission unit that transmits a first message including the first divided trained model to a user equipment. The network node also has a reception unit that receives from the user equipment a second message including output result data obtained from location information by the user equipment using the first divided trained model. The control unit infers whether or not a mobility operation is to be performed from the output result data using the second divided trained model. The mobility operation also includes at least the network node transmitting a handover request message to another network node.

[0006] A user device according to a third aspect is a user device in a mobile communication system. The user device has a receiver that divides a trained AI / ML model into a first divided trained model and a second divided trained model, and receives a first message including the first divided trained model from a network node. The user device also has a controller that acquires location information indicating the location of the user device and acquires output result data from the location information using the first divided trained model. The user device also has a transmitter that transmits a second message including the output result data to a network node. The network node infers whether or not a mobility operation is to be performed from the output result data using the second divided trained model. The mobility operation also includes at least a network node transmitting a handover request message to another network node.

[0007] 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 the AI / ML technology according to the first embodiment. FIG. 8 is a diagram showing an example of the layout of functional blocks of the 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 the layout of functional blocks of the AI / ML technology according to the first embodiment. FIG. 11 is a diagram showing an example of the layout of functional blocks of the AI / ML technology according to the first embodiment. FIG. 12 is a diagram showing an example of the layout of functional blocks of the AI / ML technology according to the first embodiment. FIG. 13 is a diagram showing an example of operation according to the first embodiment. FIG. 14 is a diagram showing an example of a setting message according to the first embodiment. Fig. 15 is a diagram showing an example of the configuration of functional blocks of AI / ML technology according to the first embodiment. Fig. 16 is a diagram showing an example of a trained model according to the first embodiment. Fig. 17 is a diagram showing an example of a trained model according to the first embodiment. Fig. 18 is a diagram showing an example of an operation according to the first embodiment. Fig. 19 is a diagram showing an example of a trained model according to the first embodiment. Fig. 20 is a diagram showing an example of a trained model according to the first embodiment.

[0008] The present disclosure aims to efficiently utilize a trained AI / ML model while ensuring the confidentiality of inference result data inferred from the trained AI / ML model. Another aim of the present disclosure is to reduce the load on the trained model.

[0009] [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.

[0010] (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.

[0011] 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.

[0012] 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), 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).

[0013] 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").

[0014] 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.

[0015] The 5GC20 includes an AMF (Access and Mobility Management Function) and a UPF (User Plane Function) 300. The AMF performs various mobility controls for the UE 100. The AMF manages the mobility of the UE 100 by communicating with the UE 100 using NAS (Non-Access Stratum) signaling. The UPF controls data forwarding. The AMF and the UPF 300 are connected to the gNB 200 via an NG interface, which is an interface between a base station and a core network. The AMF and the UPF 300 may be core network devices included in the CN 20.

[0016] 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.

[0017] The receiving unit 110 performs various types of reception under the control of the control unit 130. The receiving unit 110 includes an antenna and a receiver. The receiver converts a radio signal received by the antenna into a baseband signal (received signal) and outputs the baseband signal to the control unit 130.

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] The PDCP layer performs header compression / decompression, encryption / decryption, and the like.

[0033] 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.

[0034] 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).

[0035] 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.

[0036] 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.

[0037] 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).

[0038] (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.

[0039] The functional block configuration example shown in FIG. 6 includes a data collection unit A1, a model learning unit A2, a model inference unit A3, and a data processing unit A4.

[0040] The data collection unit A1 collects input data, specifically, data for learning and data for inference. The data collection unit A1 outputs the data for learning to the model learning unit A2. The data collection unit A1 also outputs the data for inference to the model inference unit A3. The data collection unit A1 may acquire data in the device in which the data collection unit A1 is provided as input data. The data collection unit A1 may acquire data in another device as input data. Data collection refers to a process of collecting data in a network node, a management entity, or a UE 100, for example, to perform AI / ML model learning, data analysis, and inference. Based on the data collected by the data collection unit A1, subsequent AI / ML model learning and AI / ML model inference are performed. Note that an AI / ML model refers to, for example, a data-driven algorithm that applies AI / ML technology to generate a series of outputs based on a series of inputs. Hereinafter, the terms "model" and "AI / ML model" may be used interchangeably.

[0041] The model learning unit A2 performs model learning. Specifically, the model learning unit A2 optimizes parameters of the learning model through machine learning using the learning data, and derives (or generates, or updates) a learned model. The model learning unit A2 outputs the derived learned model to the model inference unit A3. For example, considering y = ax + b, a (slope) and b (intercept) are parameters, and optimizing these corresponds to machine learning. Generally, machine learning is classified into supervised learning, unsupervised learning, and reinforcement learning. Supervised learning is a method that uses correct answer data as learning data. Unsupervised learning is a method that does not use correct answer data as learning data. For example, in unsupervised learning, feature points are memorized from a large amount of learning data, and the correct answer is determined (range estimation). Reinforcement learning is a method of assigning a score to an output result and learning how to maximize the score. Hereinafter, supervised learning will be described, but unsupervised learning or reinforcement learning may be applied as machine learning. This process of learning an AI / ML model (by learning the relationship between input and output) in a data-driven manner and obtaining a trained AI / ML model is referred to as AI / ML model learning, for example. Hereinafter, "AI / ML model learning" may be referred to as "model learning." Furthermore, a trained AI / ML model may be referred to as a "trained model."

[0042] The model inference unit A3 performs model inference. Specifically, the model inference unit A3 infers an output from inference data using a trained model and outputs the inference result data to the data processing unit A4. For example, in the equation y = ax + b, x corresponds to the inference data and y corresponds to the inference result data. Note that "y = ax + b" is a model. A model with optimized slope and intercept, for example, "y = 5x + 3," is a trained model. Various model approaches are available, including linear regression analysis, neural networks, and decision tree analysis. The above "y = ax + b" can be considered a type of linear regression analysis. The model inference unit A3 may provide model performance feedback to the model learning unit A2. This process of generating a series of outputs based on a series of inputs using a trained AI / ML model is referred to as AI / ML model inference. Hereinafter, "AI / ML model inference" may also be referred to as "model inference."

[0043] The data processing unit A4 receives the inference result data and performs processing that utilizes the inference result data.

[0044] FIG. 7 is a diagram illustrating an example of operation in the AI / ML technique according to the first embodiment.

[0045] The transmitting entity TE is, for example, an entity in which machine learning is performed. The transmitting entity TE may perform machine learning to derive a trained model. The transmitting entity TE then generates inference result data as an inference result using the trained model. The transmitting entity TE can transmit the inference result data to the receiving entity RE.

[0046] On the other hand, the receiving entity RE is, for example, an entity in which machine learning is not performed. The receiving entity RE can receive inference result data transmitted from the transmitting entity TE. The receiving entity RE performs various processes using the inference result data. The receiving entity RE may perform machine learning to derive a trained model. In this case, the receiving entity RE transmits the derived trained model to the transmitting entity TE.

[0047] Note that an entity may be, for example, a device, a functional block included in a device, or a hardware block included in a device.

[0048] 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.

[0049] As shown in Fig. 8, in step S1, the transmitting entity TE transmits control data related to AI / ML technology to the receiving entity RE and receives the control data from the receiving entity RE. The control data may be an RRC message, which is signaling of the RRC layer (i.e., Layer 3). The control data may be a MAC Control Element (CE), which is signaling of the MAC layer (i.e., Layer 2). The control data may be Downlink Control Information (DCI), which is signaling of the PHY layer (i.e., Layer 1). The downlink signaling may be UE-specific signaling. The downlink signaling may be broadcast signaling. The control data may be a control message in a control layer (e.g., AI / ML layer) dedicated to artificial intelligence or machine learning.

[0050] (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.

[0051] For example, there are three use cases in which AI / ML technology is applied:

[0052] (1.1) "CSI (Channel State Information) Feedback Enhancement"

[0053] (1.2) "Beam management"

[0054] (1.3) "Positioning Accuracy Enhancement" Below, an example of the placement of functional blocks for each use case will be described.

[0055] (1.1) Example of functional block arrangement in "CSI feedback improvement" "CSI feedback improvement" represents a use case in which, for example, machine learning technology is applied to CSI fed back from UE100 to gNB200. CSI is information about the channel state in the downlink between UE100 and gNB200. The CSI includes at least one of a channel quality indicator (CQI), a precoding matrix indicator (PMI), and a rank indicator (RI). The gNB200 performs, for example, downlink scheduling based on the CSI feedback from UE100.

[0056] 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.

[0057] 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).

[0058] 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.

[0059] 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).

[0060] 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.

[0061] 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.

[0062] 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.

[0063] FIG. 9 is a diagram illustrating an example of operation in "CSI feedback improvement" according to the first embodiment.

[0064] 9, in step S101, 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.

[0065] In step S102, gNB200 may send a switching notification to UE100 to start learning mode.

[0066] In step S103, the UE 100 starts the learning mode.

[0067] In step S104, the gNB 200 transmits the full CSI-RS. The receiver 110 of the UE 100 receives the full CSI-RS, and the CSI generator 131 generates (or estimates) CSI based on the full CSI-RS. In the learning mode, the data collector A1 collects the full CSI-RS and CSI. The model learning unit A2 creates a learned model using the full CSI-RS and the CSI as learning data.

[0068] In step S105, UE100 transmits the generated CSI to gNB200.

[0069] Thereafter, in step S106, when the model learning is completed, the UE 100 transmits a completion notification indicating that the model learning is completed to the gNB 200. The UE 100 may transmit a completion notification when the creation of the learned model is completed.

[0070] In step S107, in response to receiving the completion notification, gNB200 sends a switching notification to UE100 to switch UE100 from learning mode to inference mode.

[0071] In step S108, in response to receiving the switching notification, the UE 100 switches from the learning mode to the inference mode.

[0072] In step S109, the gNB 200 transmits a partial CSI-RS. The receiver 110 of the UE 100 receives the partial CSI-RS. In the inference mode, the data collector A1 collects the partial CSI-RS. The model inference unit A3 inputs the partial CSI-RS as inference data into the trained model, and obtains CSI as the inference result.

[0073] In step S110, the UE 100 feeds back (or transmits) the CSI, which is the inference result, to the gNB 200 as inference result data. In the UE 100, by repeating model learning in the learning mode, a trained model with a predetermined accuracy or higher can be generated. It is expected that the inference result using the trained model generated in this way will also have a predetermined accuracy or higher.

[0074] In addition, in step S111, if UE100 determines that model learning is necessary, it may send a notification indicating that model learning is necessary to gNB200 as control data.

[0075] 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".

[0076] 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:

[0077] (X1) RSRP (Reference Signals Received Power), RSRQ (Reference Signal Received Quality), SINR (Signal-to-interference-plus-noise ratio), or AD converter output waveform (These measurements may be CSI-RS. These measurements may also be other received signals received from the gNB 200.)

[0078] (X2) Bit Error Rate (BER) or Block Error Rate (BLER) (The total number of transmitted bits (or the total number of transmitted blocks) is known, and the BER (or BLER) may be measured based on the CSI-RS.)

[0079] (X3) The movement speed of UE100 (which may be measured by a speed sensor within UE100). The data set to be used for machine learning may be set. For example, the following processing may be performed. That is, UE100 transmits capability information indicating which type of input data it can handle in machine learning to gNB200 as control data. The capability information may represent, for example, any of the data or information shown in (X1) to (X3). The capability information may be information in which learning data and inference data are separately specified. Then, gNB200 transmits data type information to be used as the data set to UE100 as control data. The data type information may represent, for example, any of the data or information shown in (X1) to (X3). Furthermore, the data type information may specify separately data type information to be used as learning data and data type information to be used as inference data.

[0080] (1.2) Example of functional block arrangement in "beam management" Next, an example of functional block arrangement in "beam management" will be described. "Beam management" represents a use case in which, for example, machine learning technology is used to manage which beam is the optimal beam among the beams transmitted from gNB200.

[0081] In "beam management," gNB200 sequentially transmits beams with different directivities. Each beam includes, for example, a reference signal. UE100 measures the reception quality of each beam using the reference signal included in each beam. UE100, for example, determines the beam with the best reception quality as the optimal beam.

[0082] Figure 10 is a diagram showing an example of the arrangement of each functional block in "beam management". In the example of "beam management" shown in Figure 10, a data collection unit A1, a model learning unit A2, and a model inference unit A3 are included in the control unit 130 of UE100. On the other hand, a data processing unit A4 is included in the control unit 230 of gNB200. That is, Figure 10 shows an example in which model learning and model inference are performed in UE100. Figure 10 shows an example in which the transmitting entity TE is UE100 and the receiving entity RE is gNB200.

[0083] As shown in FIG. 10, the UE 100 has an optimal beam determination unit 132. The optimal beam determination unit 132 determines the optimal beam based on, for example, the reception quality of the reference signal included in each beam. As with "CSI feedback," an example will be described in which a CSI-RS is used as the reference signal, but a demodulation reference signal (DMRS) may also be used as the reference signal. The transmitter 120 transmits information representing the determined optimal beam to the gNB 200 as the "optimal beam."

[0084] An example of the operation in "beam management" can be implemented by replacing "CSI feedback" with "optimal beam" in FIG.

[0085] In the learning mode (step S103), the gNB 200 sequentially transmits beams with different directivities to the UE 100 (step S104). Each beam includes a full CSI-RS. In the learning mode, the data collection unit A1 of the UE 100 collects the full CSI-RS and the optimal beam (information representing the optimal beam). The model learning unit A2 creates a learned model using the CSI-RS and the optimal beam (information representing the optimal beam) as learning data. The full CSI-RS is an example of a first reference signal, and the partial CSI-RS is an example of a second reference signal.

[0086] In the inference mode (step S108), the gNB 200 sequentially transmits beams with different directivities. Each beam includes a partial CSI-RS. In the inference mode, the data collection unit A1 collects the partial CSI-RS. The model inference unit A3 inputs the partial CSI-RS as inference data into the trained model, and obtains the optimal beam (information representing the optimal beam) as the inference result. The UE 100 transmits the inference result (optimal beam) to the gNB 200 as inference result data.

[0087] In "beam management," in addition to "CSI-RS" and "optimal beam," at least one of the following data or information may be used as data in the data set.

[0088] (Y1) SSB (Synchronization Signal Block) received from gNB200

[0089] (Y2) RSRP, RSRQ, SINR, or AD converter output waveform (these measurements may be CSI-RS. These measurements may also be other received signals received from gNB200.)

[0090] (Y3) BER or BLER (BER (or BLER) may be measured based on CSI-RS with the total number of transmission bits (or the total number of transmission blocks) known.)

[0091] (Y4) Number of beams or beam pattern

[0092] (Y5) Beam measurement value(s)

[0093] (Y6) The movement speed of UE100 (may be measured by a speed sensor within UE100). UE100 may transmit capability information indicating which type of input data it can handle in machine learning to gNB200 as control data. The capability information may include any of the information or data from (Y1) to (Y6), or may include any of the information or data from (Y1) to (Y6) separately for learning data and inference data. In addition, gNB200 may transmit data type information used as a data set to UE100 as control data. The data type information may include, for example, any of the data or information shown in (Y1) to (Y6), or may include any of the information or data from (Y1) to (Y6) separately for learning data and inference data.

[0094] (1.3) Example of Arrangement of Functional Blocks in "Improvement of Location Accuracy" Next, an example of arrangement of functional blocks in "Improvement of Location Accuracy" will be described. "Improvement of Location Accuracy" represents a use case in which, for example, the accuracy of location information measured by the UE 100 is improved by using machine learning technology.

[0095] 11 is a diagram showing an example of the arrangement of each functional block in "improving location accuracy". In the example of "improving location accuracy" shown in FIG. 11, 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, FIG. 11 shows an example in which model learning and model inference are performed in the UE 100. FIG. 11 shows an example in which the transmitting entity TE is the UE 100 and the receiving entity RE is the gNB 200.

[0096] As shown in FIG. 11 , the UE 100 includes a location information generation unit 133. The UE 100 may include a GNSS (Global Navigation Satellite System) receiver 150. The location information generation unit 133 generates location data for the UE 100 based on a positioning reference signal (PRS) (full PRS or partial PRS) received from the gNB 200. The location information generation unit 133 may receive a GNSS signal (full GNSS signal or partial GNSS signal) received by the GNSS receiver 150, and generate location data for the UE 100 based on the GNSS signal.

[0097] Note that the gNB 200 transmits the full PRS using a predetermined amount of first resources (for example, all antenna ports or a predetermined amount of time-frequency resources), similar to the full CSI-RS. Also, the gNB 200 transmits the partial PRS using second resources (for example, half the antenna ports in an antenna panel, or half the predetermined amount of time-frequency resources) that have a smaller amount of resources than the first resources, similar to the partial CSI-RS.

[0098] The full GNSS signal may also be a GNSS signal that is received continuously over time by the GNSS receiver 150. Furthermore, the partial GNSS signal may also be a GNSS signal that is received intermittently by the GNSS receiver 150. That is, a predetermined amount of first resources may be used for the full GNSS signal, and second resources having a smaller amount than the first resources may be used for the partial GNSS signal.

[0099] An example of the operation for "improving location accuracy" can be implemented by replacing "full CSI-RS" with "full PRS," "partial CSI-RS" with "partial PRS," and "CSI feedback" with "location data" in Figure 9.

[0100] In the learning mode (step S103), the location information generation unit 133 generates location data for the UE 100 based on the full PRS received from the gNB 200. The location information generation unit 133 may receive a full GNSS signal received by the GNSS receiver 150 and generate location data for the UE 100 based on the full GNSS signal. The transmission unit 120 feeds back (or transmits) the location data to the gNB 200. The data collection unit A1 collects the full PRS (or full GNSS signal) and location data. The model learning unit A2 creates a learned model using the full PRS (or full GNSS signal) and location data as learning data.

[0101] In the inference mode (step S108), the data collection unit A1 collects the partial PRS received by the receiving unit 110 (or the partial GNSS signal received by the GNSS receiver 150). The model inference unit A3 inputs the partial PRS (or the partial GNSS signal) as inference data into the trained model, and obtains location data as an inference result. The UE 100 transmits the inference result (location data) to the gNB 200 as inference result data.

[0102] In "improving position accuracy," in addition to "PRS" (or "GNSS signal") and "position data," at least one of the following data or information may be used in the data set:

[0103] (Z1) RSRP, RSRQ, SINR (Signal-to-interference-plus-noise ratio), or AD converter output waveform (these measurements may be PRS. These measurements may also be other received signals received from gNB200.)

[0104] (Z2) LOS (Line of Sight) or NLOS (Non Line of Sight)

[0105] (Z3) Measurement timing, accuracy, likelihood

[0106] (Z4) RF Fingerprint (Cell ID and reception quality in the cell of the cell ID)

[0107] (Z5) Angle of Arrival (AOA), reception level for each antenna, reception phase for each antenna, and observed time difference of arrival (OTDOA) for each antenna

[0108] (Z6) Received information of beacons used in wireless LANs (Local Area Networks) such as Wi-Fi (registered trademark) or short-range wireless communications such as Bluetooth (registered trademark)

[0109] (Z7) The movement speed of UE100 (the movement speed may be measured by the GNSS receiver 150. The movement speed may also be measured by a speed sensor within UE100.) UE100 may transmit capability information indicating which types of input data it can handle in machine learning as control data to gNB200. The capability information may include any of the information or data from (Z1) to (Z7), or may include any of the information or data from (Z1) to (Z7) separately for learning data and inference data. In addition, gNB200 may transmit data type information to be used as a data set as control data to UE100. The data type information may include, for example, any of the data or information shown in (Z1) to (Z7), or may include any of the information or data from (Z1) to (Z7) separately for learning data and inference data.

[0110] (1.4) Other Arrangement Examples Next, other arrangement examples will be described.

[0111] Fig. 12 is a diagram showing another example of the arrangement of "CSI feedback improvement" according to the first embodiment. Fig. 12 shows an example in which a data collection unit A1, a model learning unit A2, a model inference unit A3, and a data processing unit A4 are included in a gN200. That is, Fig. 12 shows an example in which model learning and model inference are performed in the gNB200. Fig. 12 shows an example in which the transmitting entity TE is the gNB200 and the receiving entity RE is the UE100.

[0112] 12 shows an example in which AI / ML technology is introduced into CSI estimation performed by gNB200 based on SRS (Sounding Reference Signal). Therefore, gNB200 has a CSI generation unit 231 that generates CSI based on SRS. The CSI is information indicating the channel state of the uplink between UE100 and gNB200. gNB200 (e.g., data processing unit A4) performs, for example, uplink scheduling based on the CSI generated based on SRS.

[0113] (1.5) Example of Model Transfer In (1.1) to (1.4), examples of the arrangement of each functional block of AI / ML technology have been described. Below, model transfer will be described. The model to be transferred may be a trained model used in model inference. The model may also be an untrained (or training) model used in model training.

[0114] (1.5.1) First operation pattern related to model forwarding Figure 13 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 13, 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 13, 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.

[0115] 13, in step S201, the gNB 200 transmits a capability inquiry message to the UE 100 to request transmission of a message including an information element (IE) indicating execution capability for machine learning processing. The UE 100 receives the capability inquiry message. However, the gNB 200 may transmit the capability inquiry message when executing the machine learning processing (when determining that the processing is to be executed).

[0116] In step S202, the UE 100 transmits to the gNB 200 a message including an information element indicating execution capability for machine learning processing (or, from another perspective, the execution environment for machine learning processing). The gNB 200 receives the message. The message may be an RRC message (for example, a "UE Capability" message or a newly defined message (for example, a "UE AI Capability" message, etc.)). Alternatively, the transmitting entity TE may be the AMF 300, and the message may be a NAS message. Alternatively, if a new layer for performing or controlling machine learning processing (AI / ML processing) is defined, the message may be a message of the new layer.

[0117] The information element indicating the execution capability related to the machine learning process may be an information element indicating the capability of a processor for executing the machine learning process and / or an information element indicating the capability of a memory for executing the machine learning process. Specifically, the information element indicating the processor capability may be an information element indicating the product number (or model number) of the AI ​​processor. Furthermore, specifically, the information element indicating the memory capability may be an information element indicating the memory capacity.

[0118] Alternatively, the information element indicating the execution capability of the machine 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 or an information element indicating the time required to execute the inference process (or the response time).

[0119] Alternatively, the information element indicating the execution capability of the machine learning process may be an information element indicating the execution capability of the learning process (model learning). Specifically, the information element indicating the execution capability of the learning process may be an information element indicating the number of concurrent executions of the learning process or an information element indicating the processing capacity of the learning process.

[0120] In step S203, gNB200 determines the model to be configured (or deployed) in UE100 based on the information elements included in the message received in step S202.

[0121] In step S204, gNB200 transmits a message including the model determined in step S203 to UE100. UE100 receives the message and performs machine learning processing (i.e., model learning processing and / or model inference processing) using the model included in the message. A specific example of step S204 will be described in the following second operation pattern.

[0122] (1.5.2) Second operation pattern related to model transfer FIG. 14 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.

[0123] In the example of FIG. 14, 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.

[0124] 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.

[0125] The individual additional information or common additional information may be a model application that specifies a function to which a model is to be applied (e.g., "CSI feedback," "beam management," "positioning," etc.), or may be a model selection criterion that applies (executes) a corresponding model depending on whether a specified criterion (e.g., moving speed) is met.

[0126] (1.6) Example of Functional Block Configuration Functional blocks for AI for wireless communication have been described using Fig. 6. Currently, the 3GPP is considering the block diagram shown in Fig. 15 regarding functional blocks for AI for wireless communication.

[0127] Fig. 15 is a diagram illustrating an example of the configuration of functional blocks according to the first embodiment. Compared with the functional block diagram illustrated in Fig. 6, the functional block diagram illustrated in Fig. 15 further includes a model management unit A5 and a model recording unit A6.

[0128] The model management unit A5 manages the AI / ML model. For example, the model management unit A5 requests the model training unit A2 to retrain the trained model or requests the model recording unit A6 to transfer the model. As shown in FIG. 15 , an AI / ML model that has been trained through retraining may be referred to as an updated model. Furthermore, for example, the model management unit A5 instructs (or requests) the model inference unit A3 to select a model, (de)activate a model, switch a model, and / or fallback. The model management unit A5 may evaluate the performance of the trained model using the monitoring data acquired from the data collection unit A1 and the monitoring output acquired from the model inference unit A3, and may request retraining or instruct model switching based on the evaluation results.

[0129] The model recording unit A6 functions as a reference point in the functional blocks. Therefore, the model recording unit A6 does not necessarily have to record the trained model or the updated model on a recording medium.

[0130] The arrangement of the functional blocks shown in FIG. 15 in each use case is currently under consideration in 3GPP.

[0131] Hereinafter, an AI / ML model to be trained may be referred to as a "training model," and an AI / ML model that has already been trained may be referred to as a "trained model." Furthermore, data for inference may be referred to as inference data, and data for training may be referred to as training data.

[0132] (Communication Control Method According to First Embodiment) Next, a communication control method according to the first embodiment will be described.

[0133] As explained in the use case of "improving location accuracy," location information may be used in the AI / ML model. However, location information used in the AI / ML model may raise privacy concerns. 3GPP has pointed out that location information may raise privacy concerns for the UE 100 (e.g., Non-Patent Document 2: RWS-230240).

[0134] For example, when a handover event condition (event A1, event A2, etc.) is satisfied, UE 100 transmits a measurement report (Measurement Report) to the serving cell. When the serving cell decides to hand over UE 100, it transmits a handover request (HANDOVER REQUEST) message to the target cell. When the target cell decides to accept the handover, it transmits a handover request acknowledge (HANDOVER REQUEST ACKNOWLEDGE) message including information for accessing the target cell to the serving cell. The serving cell transmits a message (e.g., an RRC Reconfiguration message) including the information to UE 100. This allows UE 100 to perform handover from the serving cell to the target cell.

[0135] However, the measurement report may include location information (LocationInfo) of the UE 100.

[0136] Some users of the UE 100 may not want to use the location information from the viewpoint of privacy. For example, user A may not want to include the location information of the user A's home in the measurement report and transmit it. On the other hand, the network side may want to efficiently perform the handover procedure by using the measurement report including the location information of the UE 100.

[0137] Therefore, it is also possible for gNB200 to transmit a trained model to UE100. That is, the trained model is an AI / ML model in which location information is used as inference data and whether or not a handover request message is transmitted from a serving cell (or serving gNB) to a target cell (or target gNB) is used as inference result data. UE100 inputs its own location information into the trained model, and upon obtaining information indicating whether or not a handover request message is transmitted from the trained model, transmits a message including this information to gNB200. Based on this information, gNB200 can determine whether or not to transmit a handover request message to the target cell. This allows gNB200 to transmit a handover request message to the target cell without UE100 transmitting a measurement report including location information to gNB200.

[0138] However, by transmitting the trained model to UE 100, the network side may notify UE 100 of information that should be kept secret. For example, the occurrence of a ping-pong phenomenon in which handovers to a target cell are repeated multiple times within a certain period of time at a specific location despite sufficient coverage in the source cell may be information that should be kept secret by the operator (or telecommunications carrier). In UE 100, obtaining inference result data indicating that handover request messages are being transmitted at multiple nearby locations may imply the occurrence of a ping-pong phenomenon. There is also a concern that UE 100 may transmit the information to another operator (a base station or access point owned by that operator). In this way, the inference result data obtained from the trained model in UE 100 may be information that should be kept secret.

[0139] Furthermore, when gNB200 transmits a trained model having a certain amount of information or more to UE100, it may become a burden for UE100 to maintain and use the trained model.

[0140] Therefore, the first embodiment aims to enable efficient use of a trained AI / ML model while ensuring the confidentiality of inference result data inferred from the trained AI / ML model. Also, the first embodiment aims to reduce the load on the trained model.

[0141] Therefore, in the first embodiment, first, a base station (e.g., gNB200) divides a trained AI / ML model into a first divided trained model and a second divided trained model. Second, the base station transmits a first message including the first divided trained model to a user device (e.g., UE100). Third, the user device acquires location information indicating the location of the user device. Fourth, the user device acquires output result data from the location information using the first divided trained model. Fifth, the user device transmits a second message including the output result data to the base station. Sixth, the base station uses the second divided trained model to infer whether or not a mobility operation is to be performed from the output result data. Here, the mobility operation includes at least the base station transmitting a handover request message to another base station.

[0142] As described above, in the first embodiment, the trained model is divided, and the trained model is executed in a distributed manner between the UE 100 and the gNB 200. Then, it becomes possible to acquire inference result data using the trained model in the gNB 200. This allows, for example, information indicating whether the gNB 200 will transmit a handover request message to another base station to be inferred in the gNB 200 without the UE 100 having to infer it, thereby making it possible to efficiently use the trained AI / ML model while ensuring the confidentiality of the inference result data.

[0143] Furthermore, in the first embodiment, the entire trained model is not transmitted to the UE 100, and a first divided trained model obtained by dividing the trained model is transmitted to the UE 100. Therefore, in the first embodiment, it is possible to reduce the load on the UE 100 compared to when the entire trained model is transmitted to the UE 100.

[0144] (Example of division of trained model according to first embodiment) Next, an example of division of trained model according to the first embodiment will be described.

[0145] FIG. 16 is a diagram illustrating an example of division of a trained model according to the first embodiment.

[0146] FIG. 16 shows an example of a trained model that does not have an intermediate layer (or hidden layer) and is composed of an input layer and an output layer. Such a trained model may be called a single-layer model (or single-layer perceptron). The trained model shown in FIG. 16 also shows an example of a model that uses location information as inference data and infers inference result data indicating whether or not the gNB 200 will transmit a handover request message to the target gNB (or target cell). The gNB 200 transmitting a handover request message to another gNB is an example of a mobility operation.

[0147] Then, as shown in FIG. 16 , one trained model is executed in a distributed manner between the UE 100 and the gN 200. The gNB 200 may divide the trained model. The divided trained model including at least the input layer may be referred to as a first divided trained model. Furthermore, the divided trained model including at least the output layer may be referred to as a second divided trained model.

[0148] After dividing the learned model, the gNB200 transmits the first divided learned model to the UE100. Specifically, the gNB200 transmits, as the first divided learned model, location information x1, a weighting v11 for y1 (transmitting a handover request message) for the location information x1, and a weighting v12 for y2 (not transmitting a handover request message) for the location information x1. The gNB200 also transmits, as the first divided learned model, location information x2, a weighting v21 for y1 for the location information x2, and a weighting v22 for y2 for the location information x2. In this way, the gNB200 transmits each piece of location information included in the learned model, a weighting for y1 for the location information, and a weighting for y2 for the location information, as the first divided learned model.

[0149] In addition, in order to conceal y1 and y2, gNB200 may transmit a list of weightings. For example, a weighting list "v11 v12" may be transmitted, in which case the first entry represents the weighting for y1 and the last entry represents the weighting for y2. Alternatively, gNB200 may simply transmit a list "y1 v11 y2 v12". In this case, the weighting for y1 represents v11 and the weighting for y2 represents v12. However, the notation that y1 means "send a handover request message" and the notation that y2 means "do not send a handover request message" are not included in the entries (or the first divided learned model transmitted to UE100).

[0150] After receiving the first divided trained model, the UE 100 performs processing corresponding to the input layer using the first divided trained model.

[0151] That is, the UE 100 acquires location information using the GNSS receiver 150 or the PRS. Then, the UE 100 obtains output result data corresponding to the location information using the first divided trained model. For example, the UE 100 may obtain the output result data as follows.

[0152] That is, assume that the location information acquired by UE100 is x1. In this case, based on the model, the UE acquires a weighting v11 from location information x1 to y1 (i.e., sending a handover request message) and a weighting v12 from the acquired location information x1 to y2 (i.e., not sending a handover request message) using the first divided learned model. Then, UE100 transmits weighting v11 and weighting v12 to gNB200. UE100 may transmit a list of weights "v11 v12" to gNB200. In this case, the first entry in the weighting list represents a weighting for y1 (sending a handover request message), and the last entry represents a weighting for y2 (not sending a handover request message). The weighting list may represent the output result (i.e., output result data) of the first divided learned model. In UE100, when output result data corresponding to the location information is obtained using the first divided learned model, the output result data is transmitted to gNB200.

[0153] After receiving the output result data, the gNB 200 performs processing corresponding to the output layer using the second divided trained model. For example, the gNB 200 may perform the following processing.

[0154] That is, gNB200 receives the weighting list as output result data, and as processing in the output layer, grasps the weighting v11 of y1 (sends a handover request message) and the weighting v12 of y2 (does not send a handover request message). gNB200 may compare the two weightings v11 and v12 and perform the larger processing (y1 or y2).

[0155] Note that weighting can also be expressed as probability. For example, in Fig. 16, the probability from location information x1 to y1 is v11, and the probability from location information x1 to y2 is v12.

[0156] Fig. 17 is a diagram illustrating an example of division of a trained model according to the first embodiment. The example illustrated in Fig. 17 also represents an example of a single-layer model.

[0157] The learned model used in Figure 17 is a model that obtains inference result data in which location information is used as inference data and gNB200 transmits handover parameters (according to the location information) to UE100. Setting (or transmitting) the handover parameters used for handover is an example of mobility operation.

[0158] In the case of the trained model shown in FIG. 17 , the trained model is also executed in a distributed manner between UE100 and gNB200. Therefore, gNB200 transmits the first divided trained model to UE100. Specifically, gNB200 transmits the probability w11 of z1 (transmitting handover parameter #1) for location information x1, the probability w12 of z2 (transmitting handover parameter #2) for location information x1, the probability w13 of z3 (transmitting handover parameter #3) for location information x1, and so on. For example, gNB200 may transmit a list of probabilities for location information (e.g., "x1 w11 w12 w13 ...") as the first divided trained model. In this case, the probability for z1, the probability for z2, the probability for z3, and so on may be expressed in the order of "w11 w12 w13 ...". In addition, gNB200 may list z1, z2, ... as the first learned model (for example, "x1 z1 w11 z2 w12 z3 w13 ..."). In this case, the probability from location information x1 to z1 is w11, the probability from location information x1 to z2 is w12, and so on.

[0159] Then, after acquiring the location information, UE100 uses the first divided learned model to obtain probabilities (e.g., probability w11, probability w12, probability w13, ...) for location information (e.g., location information x1). UE100 transmits the probabilities to gNB200 as output result data. UE100 may also list the probabilities ("w11 w12 w13 ...") and transmit them. gNB200 can perform processing on the second divided learned model by understanding that the first probability w11 in the list is the probability for z1, the next probability w12 is the probability for z2, and so on.

[0160] (Example of Operation According to First Embodiment) Next, an example of operation according to the first embodiment will be described.

[0161] FIG. 18 is a diagram illustrating an example of operation according to the first embodiment.

[0162] As shown in FIG. 18, in step S10, UE 100 is in an RRC connected state.

[0163] In step S11, the gNB 200 divides the trained model into a first divided trained model and a second divided trained model. The first divided trained model includes at least the input layer of the trained model. Meanwhile, the second divided trained model includes at least the output layer of the trained model. For example, the control unit 230 of the gNB 200 divides the trained model.

[0164] In step S12, the gNB 200 transmits an RRC message (e.g., a first message) including the first divided learned model, execution interval information, and a threshold value to the UE 100. In this way, the gNB 200 does not transmit all of the learned models to the UE 100, and therefore the load on the UE 100 can be reduced.

[0165] First, in the case of the trained model shown in FIG. 16, the first divided trained model includes each piece of location information and probabilities v11, v12, ... for y1 and y2. Also, in the case of the trained model shown in FIG. 17, the first divided trained model includes each piece of location information and probabilities w11, w12, w13, ... for z1, z2, z3, ... However, from the perspective of ensuring confidentiality, the first divided trained model does not include the contents of the inference result data (for example, y1 representing "sending a handover request message").

[0166] Second, the execution interval information indicates an interval at which the first divided trained model is executed. The UE 100 executes the process for the first divided trained model at the timing of the execution interval in accordance with the execution interval.

[0167] Third, the threshold is a threshold used by UE100 to determine whether or not to transmit output result data to gNB200. UE100 transmits the output result data to gNB200 when the output result data is equal to or greater than the threshold, and does not transmit the output result data to gNB200 when the output result data is less than the threshold.

[0168] For example, the transmitting unit 210 of the gNB 200 transmits the RRC message. The UE 100 receives the RRC message. For example, the receiving unit 110 of the UE 100 receives the RRC message.

[0169] In step S13, UE 100 measures its own location information. For example, control unit 130 of UE 100 may acquire location information using GNSS receiver 150. Alternatively, control unit 130 of UE 100 may receive PRS from gNB 200 and transmit measurement results for PRS (e.g., received signal strength or time differences from multiple cells) to a network (e.g., LMF (Location Management Function)) to acquire location information from the network.

[0170] In step S14, UE100 inputs the location information to the first divided trained model at the timing indicated by the execution interval according to the execution interval to obtain output result data. Specifically, control unit 130 of UE100 obtains probabilities for the location information (e.g., probability v11 of y1 (transmitting a handover request message) for location information x1 and probability v12 of y2 (not transmitting a handover request message) for location information x1) from the first divided trained model as output result data.

[0171] In step S15, it is determined whether or not to transmit the output result data based on the threshold value (step S12). For example, the control unit 130 of the UE 100 determines to transmit the output result data when the output result data (i.e., the probability of obtaining the inference result data) is equal to or greater than the threshold value, and determines not to transmit the output result data when the output result data is less than the threshold value. In the following, it is assumed that the UE 100 determines to transmit the output result data.

[0172] In step S16, UE100 transmits control data (e.g., a second message) including output result data to gNB200. For example, transmitter 120 of UE100 transmits the control data to gNB200. gNB200 receives the control data. For example, receiver 220 of gNB200 receives the control data.

[0173] In step S17, the gNB200 obtains inference result data from the output result data using the second divided trained model. For example, the gNB200 performs the following processing.

[0174] That is, when the learned model is the model shown in Figure 16, gNB200 infers from the output result data whether to send or not send a handover request message. According to the inference result, gNB200 transmits the handover request message to the target cell, or does not transmit the handover request message to the target cell.

[0175] In addition, when the learned model is the model shown in Figure 17, gNB200 infers handover parameters corresponding to the location information from the output result data and transmits the handover parameters. When gNB200 obtains inference result data indicating that a handover request message is to be transmitted in combination with the model shown in Figure 16, it may transmit handover parameters corresponding to the location information to UE100.

[0176] For example, the control unit 230 of the gNB 200 acquires inference result data from the output result data using the second divided learned model. Then, the backhaul communication unit 250 of the gNB 200 may transmit a handover request message to another gNB according to the inference result data, or the transmission unit 210 of the gNB 200 may transmit parameters for handover according to the inference result data.

[0177] The handover parameters may be, for example, at least one of the following:

[0178] First, the handover parameter may be a modified value of a timer used when starting execution of a handover procedure. The timer may be T312, which is started when the UE 100 transmits a measurement report (or when an event condition is satisfied). By modifying the timer, the UE 100 (or the network) can delay or advance the start timing of the handover procedure compared to when the timer is not modified.

[0179] Second, the handover parameter may be an offset value for the signal strength of the neighboring cell. The UE 100 adds the offset value to the signal strength of the neighboring cell in response to obtaining the offset value from the learned model, and transmits the added value in a measurement result (MeasurementResult).

[0180] Third, the handover parameter may be a parameter related to neighbor cell coordination. Specifically, the parameter may be a cell ID of a cell that is not selected as a candidate cell for handover. The UE 100 may obtain measurement results for cells other than the cell ID without selecting the cell ID as a candidate for handover.

[0181] (Another Division Example 1 of Trained Model According to First Embodiment) In the first embodiment, a single-layer model has been described as the trained model, but the trained model used in the first embodiment is not limited to a single-layer model. For example, in the first embodiment, a trained model having a hidden layer (or an intermediate layer) may be used.

[0182] FIG. 19 is a diagram illustrating another example of division of the trained model according to the first embodiment.

[0183] Figure 19 shows an example where the hidden layer exists in gNB200.

[0184] As shown in Figure 19, each node in the hidden layer has a common activation function. The activation function is, for example, a sigmoid function (y = 1 / (1 + e －ｘ )), hyperbolic tangent function (y = (e ｘ -e－ｘ ) / (e ｘ +e －ｘ ) or a ReLU (Rectified Linear Unit) function (y = max(0, x)). For example, the x input to each function is x = x1v11 + x2v21 + x3v31 + ... for the activation function at the top of the hidden layer, and x = x1v12 + x2v22 + x3v32 + ... for the activation function at the next stage. The activation function makes it possible to obtain two outputs, "send a handover request message" and "do not send a handover request message", even when there are many inputs.

[0185] 20 is a diagram illustrating another example of division of the trained model according to the first embodiment. In FIG. 20, an example in which a hidden layer exists in the UE 100 is illustrated.

[0186] As shown in FIG. 20, each node included in the hidden layer includes a common activation function, as in FIG. 19. The activation function may be a known activation function. When a hidden layer exists in the UE 100, the gNB 200 transmits information related to the input layer and information related to the hidden layer of the learned model to the UE 100. For example, the gNB 200 transmits to the UE 100, as information related to the input layer, each piece of location information x1, x2, x3, ..., a probability v11 to the top-stage activation function, and a probability v12 to the next-stage activation function, .... Furthermore, for example, the gNB 200 transmits to the UE 100, as information related to the hidden layer, the activation function included in each node in the hidden layer and the probabilities u11, u12, u21, ... from each node to the output layer. The gNB200 may include information related to the input layer and information related to the hidden layer as a split learned model in an RRC message and transmit it to the UE100 (step S12 of FIG. 18).

[0187] Although FIG. 20 illustrates an example in which the gNB 200 transmits all of the hidden layers to the UE 100, this is not limiting. For example, the gNB 200 may transmit only a portion of the hidden layers to the UE 100. While FIG. 20 illustrates an example in which there is one hidden layer, there may be multiple hidden layers (when multiple hidden layers exist horizontally in FIG. 20). For example, if there are four hidden layers, it is possible to have two layers close to the input layer in the UE 100 and two layers close to the output layer in the gNB 200. In this case, the gNB 200 transmits to the UE 100, as information related to the input layer, each piece of position information x1, x2, x3, ..., the probability v11 for the top-stage activation function, and the probability v12 for the next-stage activation function, ... Further, for example, gNB200 transmits to UE100, as information related to the hidden layer, the activation functions included in each node in the second layer close to the input layer, the probability from each node to each node in the second layer, and the probability from each node in the second layer to each node in the third layer. gNB200 may transmit the information related to the input layer and the information related to the hidden layer as a divided learned model in an RRC message to UE100 (step S12 of FIG. 18). In this way, when multiple layers are included in the hidden layer, gNB200 may transmit a portion of the hidden layer to UE100.

[0188] (Another operation example according to the first embodiment) In the first embodiment, an example (step S11) in which the gNB 200 transmits an RRC message including the learned model and the execution interval information has been described. For example, the gNB 200 may transmit the learned model and the execution interval information to the UE 100 by including them in a U-plane message.

[0189] [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.

[0190] The above-described operational flows are not limited to being implemented independently, but can 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. Furthermore, the order of steps in each flow may be changed as appropriate.

[0191] 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. That is, the UE 100 may be a terminal function unit (a type of communication module) for the base station to control a relay that relays signals. Such a terminal function unit is referred to as an MT. Examples of MTs include, in addition to IAB-MT, NCR (Network Controlled Repeater)-MT and RIS (Reconfigurable Intelligent Surface)-MT.

[0192] 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.

[0193] A program may be provided that causes a computer to execute each process performed by the UE 100 or the gNB 200. The program may be recorded on a computer-readable medium. Using a computer-readable medium, the program can be installed on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transitory recording medium. The non-transitory recording medium is not particularly limited, and may be, for example, a recording medium such as a CD-ROM or a DVD-ROM. Furthermore, circuits that execute each process performed by the UE 100 or the gNB 200 may be integrated, and at least a portion of the UE 100 or the gNB 200 may be configured as a semiconductor integrated circuit (chip set, SoC: System on a chip).

[0194] The functions performed by the apparatus according to the above-described embodiments may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), a central processing unit (CPU), conventional circuits, and / or combinations thereof, programmed to perform the described functions. A processor includes transistors and other circuits and is considered to be circuitry or processing circuitry. A processor may also be a programmed processor that executes a program stored in a memory. In this disclosure, circuitry, units, and means refer to hardware that is programmed to perform or executes the described functions. The hardware may be any hardware disclosed herein or any hardware known to be programmed or capable of performing the described functions. If the hardware is a processor, the circuitry, means, or unit is a combination of the hardware and software used to configure the hardware and / or processor.

[0195] 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.

[0196] 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.

[0197] This application claims priority from Japanese Patent Application No. 2023-140371 (filed August 30, 2023), the entire contents of which are incorporated herein by reference.

[0198] (Supplementary Note) (Supplementary Note 1) A communication control method in a mobile communication system, comprising: a step in which a network node divides a trained AI / ML model into a first divided trained model and a second divided trained model; a step in which the network node transmits a first message including the first divided trained model to a user device; a step in which the user device acquires location information indicating a location of the user device; a step in which the user device acquires output result data from the location information using the first divided trained model; a step in which the user device transmits a second message including the output result data to the network node; and a step in which the network node infers whether or not a mobility operation is to be performed from the output result data using the second divided trained model, wherein the mobility operation includes at least the network node transmitting a handover request message to another network node.

[0199] (Supplementary Note 2) The communication control method according to Supplementary Note 1, wherein the first message includes a threshold value, and the step of transmitting the second message includes a step in which the user device transmits the second message when the output result data is equal to or greater than the threshold value, and does not transmit the second message when the output result data is less than the threshold value.

[0200] (Supplementary Note 3) The communication control method according to Supplementary Note 1 or Supplementary Note 2, wherein the first divided trained model includes at least an input layer of the trained AI / ML model, and the second divided trained model includes at least an output layer of the trained AI / ML model.

[0201] (Supplementary Note 4) The communication control method according to any one of Supplementary Notes 1 to 3, wherein the mobility operation includes setting of parameters used in handover.

[0202] (Supplementary Note 5) The communication control method according to any one of Supplementary Notes 1 to 3, wherein the first message includes execution interval information indicating an execution interval of the first divided trained model, and the step of acquiring the output result data includes a step of the user device acquiring the output result data from the location information using the first divided trained model in accordance with the execution interval.

[0203] (Supplementary Note 6) A network node in a mobile communication system, comprising: a control unit that divides a trained AI / ML model into a first divided trained model and a second divided trained model; a transmission unit that transmits a first message including the first divided trained model to a user device; and a reception unit that receives a second message from the user device that includes output result data obtained by the user device from location information using the first divided trained model, wherein the control unit uses the second divided trained model to infer whether or not a mobility operation is to be performed from the output result data, and the mobility operation includes at least the network node transmitting a handover request message to another network node.

[0204] (Supplementary Note 7) A user device in a mobile communication system, comprising: a receiver unit that receives a first message including a trained AI / ML model divided into a first divided trained model and a second divided trained model from a network node; a controller that acquires location information indicating a location of the user device and acquires output result data from the location information using the first divided trained model; and a transmitter unit that transmits a second message including the output result data to the network node, wherein the network node infers whether or not a mobility operation is to be performed from the output result data using the second divided trained model, and the mobility operation includes at least the network node transmitting a handover request message to another network node.

[0205] 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

Claims

1. A communication control method in a mobile communication system, comprising: a network node dividing a trained AI (Artificial Intelligence) / ML (Machine Learning) model into a first divided trained model and a second divided trained model; the network node transmitting a first message including the first divided trained model to a user device; the user device acquiring location information indicating a location of the user device; the user device acquiring output result data from the location information using the first divided trained model; the user device transmitting a second message including the output result data to the network node; and the network node inferring whether or not a mobility operation is performed from the output result data using the second divided trained model, wherein the mobility operation includes at least the network node transmitting a handover request message to another network node.

2. The communication control method according to claim 1, wherein the first message includes a threshold value, and sending the second message includes the user device sending the second message when the output result data is equal to or greater than the threshold value, and not sending the second message when the output result data is less than the threshold value.

3. A communication control method as described in claim 1, wherein the first split trained model includes at least an input layer of the trained AI / ML model, and the second split trained model includes at least an output layer of the trained AI / ML model.

4. The communication control method according to claim 1, wherein the mobility operation includes setting of parameters used in handover.

5. A communication control method as described in claim 1, wherein the first message includes execution interval information indicating an execution interval of the first divided trained model, and acquiring the output result data includes the user device acquiring the output result data from the location information using the first divided trained model in accordance with the execution interval.

6. A network node in a mobile communications system, comprising: a control unit that divides a trained AI / ML model into a first divided trained model and a second divided trained model; a transmitting unit that transmits a first message including the first divided trained model to a user equipment; and a receiving unit that receives from the user equipment a second message including output result data obtained by the user equipment from location information using the first divided trained model, wherein the control unit uses the second divided trained model to infer whether or not a mobility operation is performed from the output result data, and the mobility operation includes at least the network node transmitting a handover request message to another network node.

7. A user equipment in a mobile communication system, comprising: a receiver unit that receives a first message including the first split trained model from a network node, the first message including a trained AI / ML model being divided into a first split trained model and a second split trained model; a control unit that acquires location information indicating a location of the user equipment and acquires output result data from the location information using the first split trained model; and a transmitter unit that transmits a second message including the output result data to the network node, wherein the network node infers whether or not a mobility operation is to be performed from the output result data using the second split trained model, and the mobility operation includes at least the network node transmitting a handover request message to another network node.

Images