Communication control method, network node, and user device

Abstract

Provided is a communication control method for a mobile communication system. The communication control method includes a step in which a base station sets, for a user device, a prescribed range for wireless quality. The prescribed range for wireless quality represents a wireless quality range in which a user device is permitted to determine the timing for executing conditional handover using an AI / ML model.

Description

Technical Field

[0001] The present disclosure relates to a communication control method , network nodes and user equipment thereof.

Background Art

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

[0003] For example, in Non-Patent Document ① below, AI-based mobility management is discussed. That is, in Non-Patent Document ①, due to reasons such as an increase in the density of network deployment due to higher frequencies, there is a possibility that mobility characteristics may deteriorate, and as a result of simulating AI inferences for mobility impairments including handover command loss and handover failure (HOF), it is discussed that high accuracy has been achieved for mobility impairments. Further, in Non-Patent Document ② below, it is discussed that regarding mobility management, it is necessary to consider the generalization of models from the perspective of high-speed movement.

Prior Art Documents

Non-Patent Documents

[0004]

Non-Patent Document 1

Non-Patent Document 2

[0005] One embodiment of the communication control method is a communication control method in a mobile communication system. The communication control method includes the step of a network node setting a predetermined range of radio quality to a user device. Here, the predetermined range of radio quality represents the range of radio quality in which the user device is permitted to determine the timing of conditional handover execution using an AI / ML model.

[0006] Furthermore, one embodiment of the communication control method is a communication control method in a mobile communication system. The communication control method includes the step of a user device sending log information to a network node, which includes the result of a handover and execution trigger information indicating either that a handover was performed using an AI / ML model or that a handover was performed using network settings without using an AI / ML model. [Brief explanation of the drawing]

[0007] [Figure 1] Figure 1 shows an example configuration of a mobile communication system according to the first embodiment. [Figure 2] Figure 2 shows an example configuration of a UE (User Equipment) according to the first embodiment. [Figure 3] Figure 3 shows an example configuration of a gNB (base station) according to the first embodiment. [Figure 4] Figure 4 shows an example of the configuration of a protocol stack according to the first embodiment. [Figure 5] Figure 5 shows an example of the protocol stack configuration according to the first embodiment. [Figure 6] Figure 6 shows an example of the configuration of a functional block of the AI / ML technology according to the first embodiment. [Figure 7]Figure 7 is a diagram illustrating an example of operation in the AI / ML technology according to the first embodiment. [Figure 8] Figure 8 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. [Figure 9] Figure 9 is a diagram illustrating an example of reducing CSI-RS according to the first embodiment. [Figure 10] Figure 10 is a diagram illustrating an example of reducing CSI-RS according to the first embodiment. [Figure 11] Figure 11 is a diagram illustrating an example of operation according to the first embodiment. [Figure 12] Figure 12 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. [Figure 13] Figure 13 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. [Figure 14] Figure 14 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. [Figure 15] Figure 15 is a diagram illustrating an example of operation according to the first embodiment. [Figure 16] Figure 16 is a diagram showing an example of the arrangement of functional blocks of the AI / ML technology according to the first embodiment. [Figure 17] Figure 17 is a diagram illustrating an example of operation according to the first embodiment. [Figure 18] Figure 18 is a diagram illustrating an example of operation according to the first embodiment. [Figure 19] Figure 19 is a diagram showing an example of a setting message according to the first embodiment. [Figure 20] Figure 20 is a diagram illustrating an example of operation according to the first embodiment. [Figure 21] Figure 21 is a diagram illustrating an example of operation according to the second embodiment. [Modes for carrying out the invention]

[0008] This disclosure aims to enable user devices to appropriately perform wireless communication using AI / ML models.

[0009] [First Embodiment] Referring to the drawings, a mobile communication system according to the first embodiment will be described. 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 the mobile communication system according to the first embodiment will be described. FIG. 1 is a diagram showing a configuration example of a mobile communication system 1 according to the first embodiment. The mobile communication system 1 complies with the 5th generation system (5GS) of the 3GPP standard. Hereinafter, the 5GS will be described as an example, but in the mobile communication system, an LTE (Long Term Evolution) system may be at least partially applied. In the mobile communication system, a system after the 6th generation (6G) system may be at least partially applied.

[0011] The mobile communication system 1 includes a user equipment (UE) 100, a 5G radio access network (NG-RAN) 10, and a 5G core network (5GC) 20. Hereinafter, the NG-RAN 10 may be simply referred to as the RAN 10. Also, the 5GC 20 may be simply referred to as the core network (CN) 20.

[0012] The UE 100 is a movable wireless communication device. The UE 100 may be any device as long as it is a device used by a user. For example, the UE 100 is a mobile phone terminal (including a smartphone) and / or a tablet terminal, a notebook PC, a communication module (including a communication card or a chipset), a sensor or a device provided in the sensor, a vehicle or a device provided in the vehicle (Vehicle UE), an aircraft or a device provided in the aircraft (Aerial UE).

[0013] NG-RAN10 includes base stations (referred to as "gNBs" in 5G systems) 200. The gNBs 200 are interconnected via the Xn interface, which is an inter-base station interface. Each gNB 200 manages one or more cells. The gNB 200 performs wireless communication with UEs 100 that have established a connection with its own cell. The gNB 200 has radio resource management (RRM) functions, user data routing functions (hereinafter simply referred to as "data"), measurement and control functions for mobility control and scheduling, etc. "Cell" is used as a term to indicate the smallest unit of a wireless communication area. "Cell" is also used as a term to indicate a function or resource that performs wireless communication with the UE 100. One cell belongs to one carrier frequency (hereinafter simply referred to as "frequency").

[0014] Furthermore, gNBs can also connect to the EPC (Evolved Packet Core), which is the core network of LTE. LTE base stations can also connect to 5GCs. LTE base stations and gNBs can also be connected via an inter-base station interface.

[0015] 5GC20 includes an AMF (Access and Mobility Management Function) and a UPF (User Plane Function) 300. The AMF performs various mobility controls for the UE100. The AMF manages the mobility of the UE100 by communicating with it using NAS (Non-Access Stratum) signaling. The UPF controls data transfer. The AMF and UPF300 are connected to the gNB200 via the NG interface, which is the base station-core network interface. The AMF and UPF300 may also be core network devices included in CN20.

[0016] Figure 2 shows an example configuration of UE100 (user device) according to the first embodiment. UE100 comprises a receiving unit 110, a transmitting unit 120, and a control unit 130. The receiving unit 110 and the transmitting unit 120 constitute a communication unit that performs wireless communication with gNB200. UE100 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 the radio signal received by the antenna into a baseband signal (received signal) and outputs it to the control unit 130.

[0018] The transmitting unit 120 performs various types of transmissions under the control of the control unit 130. The transmitting unit 120 includes an antenna and a transmitter. The transmitter converts the baseband signal (transmission signal) output by the control unit 130 into a wireless signal and transmits it from the antenna.

[0019] The control unit 130 performs various control and processing operations in the UE 100. Such processing includes processing in each layer described later. The control unit 130 includes at least one processor and at least one memory. The memory stores programs executed by the processor and information used for processing by the processor. The processor may include a baseband processor and a CPU (Central Processing Unit). The baseband processor performs modulation, demodulation, encoding, and decoding of baseband signals. The CPU executes programs stored in memory and performs various processing operations. Note that processing or operations performed in the UE 100 may also be performed in the control unit 130.

[0020] Figure 3 shows an example configuration of the gNB200 (base station) according to the first embodiment. The gNB200 comprises a transmitter 210, a receiver 220, a control unit 230, and a backhaul communication unit 250. The transmitter 210 and receiver 220 constitute a communication unit that performs wireless communication with the UE100. The backhaul communication unit 250 constitutes a network communication unit that communicates with the CN20. The gNB200 is another example of a communication device.

[0021] The transmitting unit 210 performs various types of transmissions under the control of the control unit 230. The transmitting unit 210 includes an antenna and a transmitter. The transmitter converts the baseband signal (transmission signal) output by the control unit 230 into a 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 the radio signal received by the antenna into a baseband signal (received signal) and outputs it to the control unit 230.

[0023] The control unit 230 performs various control and processing operations in the gNB200. Such processing includes processing in each layer described later. The control unit 230 includes at least one processor and at least one memory. The memory stores programs executed by the processor and information used for processing by the processor. The processor may include a baseband processor and a CPU. The baseband processor performs modulation, demodulation, encoding, and decoding of baseband signals. The CPU executes programs stored in memory and performs various processing operations. Note that processing or operations performed in the gNB200 may also be performed by the control unit 230.

[0024] The backhaul communication unit 250 is connected to an adjacent base station via the Xn interface, which is an inter-base station interface. The backhaul communication unit 250 is connected to the AMF / UPF300 via the NG interface, which is an inter-base station-core network interface. The gNB200 may consist of a central unit (CU) and distributed units (DU) (i.e., functionally separated), and the two units may be connected by the F1 interface, which is a fronthaul interface.

[0025] Figure 4 shows an example of the protocol stack configuration for a user-plane wireless interface that handles data.

[0026] The user plane radio interface protocol comprises a physical (PHY) layer, a media access control (MAC) layer, a radio link control (RLC) layer, a packet data convergence protocol (PDCP) layer, and a service data adaptation protocol (SDAP) layer.

[0027] The PHY layer performs encoding / decoding, modulation / demodulation, antenna mapping / demapping, and resource mapping / demapping. Data and control information are transmitted between the UE100's PHY layer and the gNB200's PHY layer via a physical channel. The UE100's PHY layer receives downlink control information (DCI) transmitted from the gNB200 over the physical downlink control channel (PDCCH). Specifically, the UE100 performs blind decoding of the PDCCH using a Radio Network Temporary Identifier (RNTI) and acquires the successfully decoded DCI as the DCI addressed to its own UE. The DCI transmitted from the gNB200 has a CRC (Cyclic Redundancy Code) parity bit added, which is scrambled by the RNTI.

[0028] In NR, the 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 consecutive PRBs (Physical Resource Blocks). The UE100 sends and receives data and control signals in the active BWP. The UE100 may have up to four BWPs configured, for example. Each BWP may have a different subcarrier spacing. The frequencies of these BWPs may overlap. If 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 on the UE100, thereby reducing UE power consumption.

[0029] The gNB200 can, for example, configure up to three control resource sets (CORESETs) for each of up to four BWPs on a serving cell. A CORESET is a radio resource for control information that the UE100 should receive. The UE100 may have up to 12 or more CORESETs configured on a serving cell. Each CORESET may have an index from 0 to 11 or more. A CORESET may consist of six resource blocks (PRBs) and one, two, or three consecutive OFDM (Orthogonal Frequency Division Multiplex) symbols in the time domain.

[0030] The MAC layer performs data priority control, retransmission processing using Hybrid Automatic Repeat request (HARQ), and random access procedures. Data and control information are transmitted between the MAC layer of the UE100 and the MAC layer of the gNB200 via the transport channel. The MAC layer of the gNB200 includes a scheduler. The scheduler determines the transport format for the up and down links (transport block size, modulation and coding scheme (MCS)) and the resource blocks to be allocated to the UE100.

[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 UE100's RLC layer and the gNB200's RLC layer via a logical channel.

[0032] The PDCP layer performs header compression / decompression, encryption / decryption, etc.

[0033] The SDAP layer maps IP flows, which are the units under which the core network performs QoS (Quality of Service) control, to wireless bearers, which are the units under which the access layer (AS: Access Stratum) performs QoS control. Note that if the RAN is connected to the EPC, the SDAP layer may not be necessary.

[0034] Figure 5 shows the configuration of the protocol stack of the wireless interface of the control plane that handles signaling (control signals).

[0035] The protocol stack of the control plane's wireless interface includes a Radio Resource Control (RRC) layer and a Non-Access Stratum (NAS) layer, instead of the SDAP layer shown in Figure 4.

[0036] RRC signaling for various settings is transmitted between the RRC layer of the UE100 and the RRC layer of the gNB200. The RRC layer controls the logical channel, transport channel, and physical channel in response to the establishment, re-establishment, and release of the radio bearer. If there is a connection (RRC connection) between the RRC of the UE100 and the RRC of the gNB200, the UE100 is in the RRC connected state. If there is no connection (RRC connection) between the RRC of the UE100 and the RRC of the gNB200, the UE100 is in the RRC idle state. If the connection between the RRC of the UE100 and the RRC of the gNB200 is suspended, the UE100 is in the RRC inactive state.

[0037] The NAS, located above the RRC layer, handles session management and mobility management, among other things. NAS signaling is transmitted between the UE100's NAS and the AMF300's NAS. The UE100 also has application layers in addition to the wireless interface protocol. Layers below the NAS are called AS (Access Stratum).

[0038] (AI / ML technology) Next, the AI / ML technology according to the embodiment will be described. Figure 6 is a diagram showing an example of the configuration of the functional block of the AI / ML technology in the mobile communication system 1 according to the first embodiment.

[0039] The example of the functional block configuration shown in Figure 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 training data and inference data. The data collection unit A1 outputs the training data to the model training unit A2. The data collection unit A1 also outputs the inference data to the model inference unit A3. The data collection unit A1 may acquire data from its own device as input data. The data collection unit A1 may acquire data from another device as input data.

[0041] Model learning unit A2 performs model learning. Specifically, model learning unit A2 optimizes the parameters of the learning model using machine learning with training data and derives (or generates or updates) a trained model. Model learning unit A2 outputs the derived trained model to model inference unit A3. For example, y = ax + b In this context, a (slope) and b (intercept) are parameters, and optimizing these parameters constitutes machine learning. Generally, machine learning includes supervised learning, unsupervised learning, and reinforcement learning. Supervised learning is a method that uses correct data as training data. Unsupervised learning is a method that does not use correct data as training data. For example, in unsupervised learning, feature points are memorized from a large amount of training data, and correct answers (range estimation) are determined. Reinforcement learning is a method that assigns a score to the output result and learns how to maximize the score. Supervised learning will be explained below, but unsupervised learning or reinforcement learning may also be applied as machine learning.

[0042] Model inference unit A3 performs model inference. Specifically, model inference unit A3 uses a trained model to infer an output from the inference data and outputs the inference result data to data processing unit A4. For example, y = ax + b In this context, x represents the inference data, and y represents the inference result data. Note that "y = ax + b" is the model. A model with optimized slope and intercept, for example "y = 5x + 3", is a trained model. Here, there are various modeling methods (approaches), 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.

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

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

[0045] The sending entity TE is, for example, an entity on which machine learning is performed. The sending entity TE performs machine learning to derive a trained model. Then, the sending entity TE uses the trained model to generate inference result data as the inference result. The sending entity TE sends this inference result data to the receiving entity RE.

[0046] On the other hand, the receiving entity RE is, for example, an entity that does not undergo machine learning. The sending entity TE performs various processes using the inference result data received from the sending entity TE.

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

[0048] For example, the transmitting entity TE may be UE100, and the receiving entity RE may be gNB200 or a core network device. Alternatively, the transmitting entity TE may be gNB200 or a core network device, and the receiving entity RE may be UE100.

[0049] As shown in Figure 7, in step S1, the transmitting entity TE transmits control data related to AI / ML technology to the receiving entity RE, or receives said control data from the receiving entity RE. The control data may be an RRC message, which is signaling at the RRC layer (i.e., Layer 3). The control data may be a MAC CE (Control Element), which is signaling at the MAC layer (i.e., Layer 2). The control data may be Downlink Control Information (DCI), which is signaling at 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 specialized for artificial intelligence or machine learning (e.g., the AI / ML layer).

[0050] (Deployment examples and use cases) Next, we will explain how each functional block shown in Figure 6 is arranged in the mobile communication system 1. Below, we will describe examples of the arrangement of each functional block in accordance with specific use cases.

[0051] Three use cases where AI / ML technology can be applied include the following:

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

[0053] (1.2) "Beam management"

[0054] (1.3) “Positioning accuracy enhancement” The following describes examples of how to arrange functional blocks for each use case.

[0055] (1.1) Example of functional block placement in "CSI Feedback Improvement" "CSI Feedback Improvement" describes a use case where machine learning techniques are applied to the CSI feedback from UE100 to gNB200. CSI is information about the channel state in the downlink between UE100 and gNB200. CSI includes at least one of the following: Channel Quality Indicator (CQI), Precoding Matrix Indicator (PMI), and Rank Indicator (RI). Based on the CSI feedback from UE100, gNB200 performs, for example, downlink scheduling.

[0056] Figure 8 shows an example of the arrangement of each functional block in "CSI Feedback Improvement". In the example of "CSI Feedback Improvement" shown in Figure 8, the data acquisition unit A1, the model learning unit A2, and the model inference unit A3 are included in the control unit 130 of UE100. On the other hand, the data processing unit A4 is included in the control unit 230 of gNB200. In other words, model learning and model inference are performed in UE100. Figure 8 shows an example where the transmitting entity TE is UE100 and the receiving entity RE is gNB200.

[0057] In "CSI Feedback Improvement," the gNB200 transmits a reference signal for the UE100 to estimate the channel state of the downlink. In the following explanation, the CSI reference signal (CSI-RS) is used as an example of the reference signal, but the reference signal may also be a demodulation reference signal (DMRS).

[0058] Firstly, during model training, UE100 (receiver 110) receives a first reference signal from gNB200 using a first resource. Then, UE100 (model training unit A2) uses training data, including the first reference signal and CSI, to derive a trained model for inferring CSI from the reference signal. Such a first reference signal is sometimes referred to as full CSI-RS.

[0059] For example, the CSI generation unit 131 performs channel estimation using the received signal (CSI-RS) received by the receiver unit 110 and generates a CSI. The transmission unit 120 transmits the generated CSI to the gNB 200. The model learning unit A2 uses the set of received signal (CSI-RS) and CSI as training data to perform model learning and derives a trained model for inferring CSI from the received signal (CSI-RS).

[0060] Secondly, in model inference, the receiver 110 receives a second reference signal from the gNB200 using a second resource which is less resource than the first resource. Then, the model inference unit A3 uses the trained model to infer the CSI using the second reference signal as inference data and the inference result data. Hereafter, 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 receiver unit 110 as inference data into the trained model and infers the CSI from the CSI-RS. The transmission unit 120 transmits the inferred CSI to the gNB200.

[0062] This allows the UE100 to feed back (or transmit) the accurate (complete) CSI to the gNB200 from the small amount of CSI-RS (partial CSI-RS) received from the gNB200. For example, the gNB200 can intentionally reduce (puncture) the CSI-RS to reduce overhead. It also allows the UE100 to cope with situations where wireless conditions are poor and some CSI-RS cannot be received properly.

[0063] Figures 9 and 10 illustrate an example of reducing CSI-RS according to the first embodiment.

[0064] Figure 9 illustrates an example of reducing CSI-RS transmission by reducing the number of antenna ports that transmit CSI-RS. For example, the gNB200 performs the following process: When the UE100 is in model learning mode (hereinafter sometimes referred to as "learning mode"), the gNB200 transmits CSI-RS from all antenna ports on the antenna panel. On the other hand, when the UE100 is in model inference mode (hereinafter sometimes referred to as "inference mode"), the gNB200 reduces the number of antenna ports that transmit CSI-RS, transmitting CSI-RS from half of the antenna ports on the antenna panel. This reduces overhead, improves the utilization efficiency of antenna ports, and reduces power consumption. Note that antenna ports are just one example of resources.

[0065] On the other hand, Figure 10 shows an example where the gNB200 reduces the radio resources used for CSI-RS transmission, specifically the time-frequency resources. For example, the gNB200 performs the following processing: When the UE100 is in learning mode, the gNB200 transmits CSI-RS using a predetermined time-frequency resource. Conversely, when the UE100 is in inference mode, the gNB200 transmits CSI-RS using less time-frequency resources than the predetermined time-frequency resource. This reduces overhead, improves the efficiency of radio resource utilization, and reduces power consumption.

[0066] As shown in Figures 9 and 10, the gNB200 transmits a full CSI-RS using a predetermined amount of first resources, and transmits a partial CSI-RS using second resources, which have a smaller resource amount than the first resources.

[0067] Figure 11 is a diagram illustrating an example of operation in "CSI Feedback Improvement" according to the first embodiment.

[0068] As shown in Figure 11, in step S101, the gNB200 may notify or set the CSI-RS transmission pattern (puncture pattern) for inference mode to the UE100 as control data. For example, the gNB200 sends to the UE100 the antenna ports and / or time-frequency resources that transmit or do not transmit CSI-RS in inference mode.

[0069] In step S102, the gNB200 may send a switch notification to the UE100 to initiate learning mode.

[0070] In step S103, UE100 enters learning mode.

[0071] In step S104, the gNB200 transmits the full CSI-RS. The receiver 110 of the UE100 receives the full CSI-RS, and the CSI generation unit 131 generates (or estimates) the CSI based on the full CSI-RS. In learning mode, the data acquisition unit A1 collects the full CSI-RS and the CSI. The model learning unit A2 uses the full CSI-RS and the CSI as training data to create a trained model.

[0072] In step S105, UE100 sends the generated CSI to gNB200.

[0073] Subsequently, in step S106, when model training is complete, UE100 sends a completion notification to gNB200 indicating that model training is complete. UE100 may also send a completion notification when the creation of the trained model is complete.

[0074] In step S107, upon receiving the completion notification, the gNB200 sends a switch notification to the UE100 to switch from learning mode to inference mode.

[0075] In step S108, UE100 switches from learning mode to inference mode in response to receiving a switch notification.

[0076] In step S109, the gNB200 transmits a partial CSI-RS. The receiver 110 of the UE100 receives the partial CSI-RS. In inference mode, the data acquisition 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 CSI as the inference result.

[0077] In step S110, UE100 feeds back (or sends) the CSI, which is the inference result, to gNB200 as inference result data. In learning mode, UE100 can generate a trained model with a predetermined accuracy or higher by repeatedly training the model. It is expected that the inference result using such a trained model will also have a predetermined accuracy or higher.

[0078] In step S111, if UE100 determines that model training is necessary, it may send a notification to gNB200 as control data indicating that model training is necessary.

[0079] In the example shown in Figure 11, the training data consists of "(Full) CSI-RS" and "CSI," while the inference data is "(Partial) CSI-RS." Hereafter, the training data and / or inference data may be referred to as the "dataset."

[0080] In "CSI Feedback Improvement," in addition to "CSI-RS" and "CSI," at least one of the following data or information may be used as the dataset:

[0081] (X1) RSRP (Reference Signals Received Power), RSRQ (Reference Signal Received Quality), SINR (Signal-to-interference-plus-noise ratio), or output waveform of the AD converter (These measurements may also be CSI-RS. These measurements may also be other received signals from the gNB200.)

[0082] (X2) Bit Error Rate (BER) or Block Error Rate (BLER) (BER (or BLER) may be measured based on CSI-RS, assuming the total number of transmitted bits (or total number of transmitted blocks) is known.)

[0083] (X3) The moving speed of the UE100 (may be measured by a speed sensor inside the UE100). The dataset to be used for machine learning may be specified. For example, the following process may be performed: The UE100 sends capability information to the gNB200 as control data, indicating which types of input data it can handle in machine learning. The capability information may represent any of the data or information shown in (X1) to (X3), for example. The capability information may specify training data and inference data separately. The gNB200 then sends data type information to be used as the dataset to the UE100 as control data. The data type information may represent any of the data or information shown in (X1) to (X3), for example. The data type information may also specify data type information used for training data and data type information used for inference data separately.

[0084] (1.2) Example of functional block arrangement in "Beam Management" Next, we will explain an example of the arrangement of functional blocks in "Beam Management." "Beam Management" represents a use case where, for example, machine learning techniques are used to manage which beam is the optimal beam among the beams transmitted from the gNB200.

[0085] In "beam management," the gNB200 sequentially transmits beams with different directivity. Each beam includes, for example, a reference signal. The UE100 uses the reference signal included in each beam to measure the reception quality of each beam. The UE100 then determines, for example, the beam with the best reception quality as the optimal beam.

[0086] Figure 12 shows an example of the arrangement of each functional block in "beam management". In the example of "beam management" shown in Figure 12, the data acquisition unit A1, the model learning unit A2, and the model inference unit A3 are included in the control unit 130 of UE100. On the other hand, the data processing unit A4 is included in the control unit 230 of gNB200. In other words, Figure 12 shows an example in which model learning and model inference are performed in UE100. In Figure 12, the transmitting entity TE is UE100 and the receiving entity RE is gNB200.

[0087] As shown in Figure 12, the UE100 has an optimal beam determination unit 132. The optimal beam determination unit 132 determines the optimal beam based, for example, on the reception quality of the reference signal included in each beam. The reference signal is described in an example where CSI-RS is used, similar to "CSI feedback," but a demodulated reference signal (DMRS) may also be used as the reference signal. The transmitter 120 transmits information representing the determined optimal beam as "optimal beam" to the gNB200.

[0088] An example of operation in "Beam Management" can be implemented in Figure 11 by replacing "CSI Feedback" with "Optimal Beam".

[0089] In learning mode (step S103), the gNB200 sequentially transmits beams with different directivity to the UE100 (step S104). Each beam includes full CSI-RS. In learning mode, the data acquisition unit A1 of the UE100 collects full CSI-RS and information representing the optimal beam. The model learning unit A2 uses the CSI-RS and information representing the optimal beam as training data to create a trained model. Full CSI-RS is an example of a first reference signal, and partial CSI-RS is an example of a second reference signal.

[0090] In inference mode (step S108), the gNB200 sequentially transmits beams with different directivity. Each beam contains partial CSI-RS. In inference mode, the data acquisition 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 (or information representing it) as the inference result. The UE100 transmits the inference result (optimal beam) as inference result data to the gNB200.

[0091] 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 for the dataset.

[0092] (Y1) SSB (Synchronization Signal Block) received from gNB200

[0093] (Y2) RSRP, RSRQ, SINR, or AD converter output waveform (These measurements may also be CSI-RS. These measurements may also be other received signals from the gNB200.)

[0094] (Y3)BER or BLER (BER (or BLER) may be measured based on CSI-RS, assuming the total number of transmitted bits (or total number of transmitted blocks) is known.)

[0095] (Y4) Number of beams, or beam pattern

[0096] (Y5) Beam measurements (including multiple measurements)

[0097] (Y6) The moving speed of UE100 (may be measured by a speed sensor inside UE100) UE100 may send capability information to gNB200 as control data indicating which types of input data it can handle in machine learning. The capability information may include any of the information or data from (Y1) to (Y6), or it may include any of the information or data from (Y1) to (Y6) separately for training data and inference data. In addition, gNB200 may send data type information to UE100 as control data. The data type information may include, for example, any of the data or information shown from (Y1) to (Y6), or it may include any of the information or data from (Y1) to (Y6) separately for training data and inference data.

[0098] (1.3) Example of functional block placement in "Improved positional accuracy" Next, we will explain an example of the placement of functional blocks in "Improving Positional Accuracy." "Improving Positional Accuracy" represents a use case in which, for example, the accuracy of positional information measured by UE100 is improved using machine learning technology.

[0099] Figure 13 shows an example of the arrangement of each functional block in "improving positional accuracy". In the example of "improving positional accuracy" shown in Figure 9, the data acquisition unit A1, the model learning unit A2, and the model inference unit A3 are included in the control unit 130 of UE100. On the other hand, the data processing unit A4 is included in the control unit 230 of gNB200. In other words, Figure 13 shows an example in which model learning and model inference are performed in UE100. In Figure 13, the transmitting entity TE is UE100 and the receiving entity RE is gNB200.

[0100] As shown in Figure 13, the UE100 includes a position information generation unit 133. The UE100 may also include a GNSS (Global Navigation Satellite System) receiver 150. The position information generation unit 133 generates position data for the UE100 based on a positioning reference signal (PRS) (full PRS or partial PRS) received from the gNB200. The position information generation unit 133 may also receive a GNSS signal (full GNSS signal or partial GNSS signal) received by the GNSS receiver 150 and generate position data for the UE100 based on the GNSS signal.

[0101] Furthermore, the gNB200 transmits full PRS using a predetermined amount of first resources (for example, all antenna ports as shown in Figure 9, or a predetermined amount of time-frequency resources as shown in Figure 10), similar to full CSI-RS. Also, the gNB200 transmits partial PRS using a second resource with a smaller amount of resources than the first resource (for example, half of the antenna ports on the antenna panel as shown in Figure 9, or half of a predetermined amount of time-frequency resources as shown in Figure 10), similar to partial CSI-RS.

[0102] Furthermore, the full GNSS signal may be a GNSS signal that the GNSS receiver 150 receives continuously over time. In addition, the partial GNSS signal may be a GNSS signal that the GNSS receiver 150 receives intermittently. That is, a predetermined amount of first resources may be used for the full GNSS signal, and a second resource with a smaller amount of resources than the first resource may be used for the partial GNSS signal.

[0103] An example of operation for "improving positional accuracy" can be implemented in Figure 11 by replacing "full CSI-RS" with "full PRS," "partial CSI-RS" with "partial PRS," and "CSI feedback" with "positional data."

[0104] In learning mode (step S103), the position information generation unit 133 generates position data for UE100 based on the full PRS received from gNB200. The position information generation unit 133 may also receive the full GNSS signal received by the GNSS receiver 150 and generate position data for UE100 based on the full GNSS signal. The transmission unit 120 feeds back (or transmits) the position data to gNB200. The data acquisition unit A1 collects the full PRS (or full GNSS signal) and position data. The model learning unit A2 creates a trained model using the full PRS (or full GNSS signal) and position data as training data.

[0105] In inference mode (step S108), the data acquisition unit A1 collects the partial PRS (or partial GNSS signal received by the GNSS receiver 150) received by the receiver 110. The model inference unit A3 inputs the partial PRS (or partial GNSS signal) as inference data into the trained model and obtains position data as an inference result. The UE100 transmits the inference result (position data) as inference result data to the gNB200.

[0106] In "improving positional accuracy," in addition to "PRS" (or "GNSS signal") and "positional data," at least one of the following types of data or information may also be used as data for the dataset.

[0107] (Z1) RSRP, RSRQ, SINR (Signal-to-interference-plus-noise ratio), or output waveform of the AD converter (These measurements may also be PRS. These measurements may also be other received signals received from the gNB200.)

[0108] (Z2) LOS (Line of Sight) or NLOS (Non-Line of Sight)

[0109] (Z3) Measurement timing, accuracy, likelihood

[0110] (Z4) RF fingerprint (cell ID and reception quality in the cell with that cell ID)

[0111] (Z5) Angle of Arrival (AOA) of the received signal, received level for each antenna, received phase for each antenna, and Observed Time Difference of Arrival (OTDOA) for each antenna.

[0112] (Z6) Receipt information of beacons used in wireless LANs (Local Area Networks) such as Wi-Fi (registered trademark) or short-range wireless communication such as Bluetooth (registered trademark)

[0113] (Z7) The moving speed of UE100 (This moving speed may be measured by the GNSS receiver 150. This moving speed may also be measured by a speed sensor inside UE100.) UE100 may send capability information to gNB200 as control data indicating which types of input data it can handle in machine learning. The capability information may include any of the information or data from (Z1) to (Z7). The capability information may include any of the information or data from (Z1) to (Z7) separately for training data and inference data. In addition, gNB200 may send data type information used as a dataset to UE100 as control data. The data type information may include, for example, any of the data or information shown from (Z1) to (Z7). The data type information may include any of the information or data from (Z1) to (Z7) separately for training data and inference data.

[0114] (1.4) Other arrangement examples Next, we will explain other arrangement examples.

[0115] Figure 14 is a diagram showing another configuration example of "CSI Feedback Improvement" according to the first embodiment. In Figure 14, the data acquisition unit A1, the model learning unit A2, the model inference unit A3, and the data processing unit A4 are all included in gN200. In other words, Figure 14 is an example in which model learning and model inference are performed in gNB200. In Figure 14, the transmitting entity TE is gNB200 and the receiving entity RE is UE100.

[0116] Figure 14 shows an example in which AI / ML technology is introduced into the 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. This CSI is information indicating the channel state of the uplink between UE100 and gNB200. gNB200 (for example, data processing unit A4) performs, for example, uplink scheduling based on the CSI generated based on SRS.

[0117] Figure 15 is a diagram illustrating an example of operation in another arrangement according to the first embodiment.

[0118] As shown in Figure 15, in step S201, gNB200 configures the UE100 for SRS transmission. The SRS transmission configuration may include information about the type of reference signal that the UE100 will transmit.

[0119] In step S202, the gNB200 enters learning mode.

[0120] In step S203, UE100 transmits the full SRS to gNB200 according to the SRS transmission settings (step S201). The receiving unit 220 of gNB200 receives the full SRS. In learning mode, the CSI generation unit 231 generates (or estimates) CSI based on the full SRS. The data acquisition unit A1 collects the full SRS and CSI. The model learning unit A2 uses the full SRS and CSI as training data to create a trained model.

[0121] In step S204, gNB200 identifies the SRS transmission pattern (puncture pattern) to be input as inference data to the trained model, and sets the identified SRS transmission pattern in UE100. gNB200 may also send the SRS transmission setting, including the identified SRS transmission pattern, to gNB200.

[0122] In step S205, the gNB200 switches from training mode to inference mode. The gNB200 then starts model inference using the trained model.

[0123] In step S206, UE100 transmits a partial SRS according to the SRS transmission settings (step S204). gNB200 inputs this SRS as inference data into the trained model to obtain channel estimation results, and then uses these channel estimation results to perform uplink scheduling for UE100 (for example, control of uplink transmission weights). Note that gNB200 may reconfigure UE100 to transmit a full SRS if the inference accuracy of the trained model deteriorates.

[0124] (1.5) Example of configuration when associative learning is performed Next, we will explain an example of the arrangement of each functional block when federated learning is performed. Federated learning is a machine learning method in which machine learning is performed in a distributed state without aggregating data (or dataset). In federated learning, each entity does not need to send data, so the security of each entity can be ensured. Furthermore, federated learning is said to be able to obtain learning results with the same accuracy as conventional centralized machine learning.

[0125] Figure 16 is a diagram showing an example of the configuration when federated learning according to the first embodiment is performed. The example shown in Figure 16 represents an example in which the position estimation of UE100 is performed using federated learning. In Figure 16, UE100 is an example in which UE100 has a data acquisition unit A1, a model learning unit A2, and a model inference unit A3. That is, it represents an example in which model learning and model inference are performed in UE100. In Figure 16, UE100 is an example in which the transmitting entity TE is and gNB200 and / or the location server 400 are an example in which the receiving entity RE is.

[0126] Associative learning, as shown in Figure 16, can be performed, for example, using the following procedure.

[0127] Firstly, the location server 400 sends the model that will serve as the basis for model training to the UE100.

[0128] Secondly, UE100 (model learning unit A2) performs model learning using data present in UE100. The data present in UE100 is, for example, the output data (GNSS signal) of the PRS and / or GNSS receiver 150 received from gNB200. The data present in UE100 may also include location data generated by the location information generation unit 133 based on the PRS reception result and / or the output data of the GNSS receiver 150.

[0129] Thirdly, UE100 applies the trained model, which is the result of the learning process, to the model inference unit A3, and also sends the variable parameters included in the trained model (hereinafter sometimes referred to as "trained parameters") to the position server 400. In the example described above, the optimized a (slope) and b (intercept) correspond to the trained parameters.

[0130] Fourth, the position server 400 (federated learning unit A5) collects trained parameters from multiple UE100s and integrates them. The position server 400 may also transmit the trained model obtained through integration to the UE100s. Based on the trained model and the measurement reports from the UE100s, the position server 400 can estimate the position of the UE100s.

[0131] Figure 17 is a diagram illustrating an example of operation in associative learning according to the first embodiment.

[0132] As shown in Figure 17, in step S301, the gNB200 may notify the UE100 of the base model to be learned. Alternatively, the location server 400 may notify the gNB200 of the model.

[0133] In step S302, gNB200 instructs UE100 to train the model. gNB200 may also set a timing (trigger condition) for reporting trained parameters. The reporting timing may be periodic. The reporting timing may also be triggered (i.e., an event trigger) when the training proficiency meets a condition.

[0134] In step S303, UE100 starts learning mode. UE100 uses the full PRS (or full GNSS signal) and the position data generated by the position information generation unit 133 as training data to perform model learning.

[0135] In step S304, when the reporting timing conditions are met, UE100 sends the learned parameters at that time to the network (gNB200 or location server 400).

[0136] In step S305, the location server 400 integrates the learned parameters reported from multiple UE100s.

[0137] (1.6) Model Transfer Example

[0138] Sections (1.1) through (1.5) described examples of the arrangement of each functional block of AI / ML technology. Below, we will describe an example of model transfer. The model to be transferred may be a trained model used in model inference. The model may also be an untrained (or in-training) model used in model training.

[0139] (1.6.1) First operating pattern for model transfer Figure 18 is a diagram illustrating an example of the operation of a first operation pattern relating to model transfer according to the first embodiment. In the example shown in Figure 18, the receiving entity RE is mainly assumed to be UE100, but the receiving entity RE may be gNB200 or AMF300. Also, in the example shown in Figure 18, the transmitting entity TE is assumed to be gNB200, but the transmitting entity TE may be UE100 or AMF300.

[0140] As shown in Figure 18, in step S401, gNB200 sends a capability query message to UE100 requesting the sending of a message containing information elements (IE) indicating the ability to perform machine learning processing. UE100 receives the capability query message. However, gNB200 may also send the capability query message if it decides to perform machine learning processing.

[0141] In step S402, UE100 sends a message to gNB200 containing informational elements indicating the execution capability for machine learning processing (or, from another perspective, the execution environment for machine learning processing). gNB200 receives the message. The message may be an RRC message (e.g., a "UE Capability" message, or a newly defined message (e.g., a "UE AI Capability" message)). Alternatively, if the sending entity TE is AMF300, the message may be a NAS message. Alternatively, if a new layer for executing or controlling machine learning processing (AI / ML processing) is defined, the message may be a message for that new layer.

[0142] Information elements indicating the execution capability for machine learning processing may include information elements indicating the processor's capability for executing machine learning processing and / or information elements indicating the memory's capability for executing machine learning processing. Specifically, information elements indicating processor capability may include information representing the model number (or part number) of the AI ​​processor. Similarly, information elements indicating memory capability may include information indicating the memory capacity.

[0143] Alternatively, an information element indicating the ability to perform machine learning processing may be an information element indicating the ability to perform inference processing (model inference). Specifically, an information element indicating the ability to perform inference processing may be an information element indicating whether or not deep neural network models are supported. This information element may also be an information element indicating the time (or response time) required to perform the inference processing.

[0144] Alternatively, an information element indicating the execution capability of machine learning processing may be an information element indicating the execution capability of the learning process (model learning). Specifically, an information element indicating the execution capability of the learning process may be an information element indicating the number of simultaneous executions of the learning process. This information element may also be an information element indicating the processing capacity of the learning process.

[0145] In step S403, gNB200 determines the model to configure (or deploy) to UE100 based on the information elements contained in the message received in step S402.

[0146] In step S404, gNB200 sends a message to UE100 containing the model determined in step S403. UE100 receives the message and uses the model contained in the message to perform machine learning processing (i.e., model training and / or model inference). A specific example of step S404 is explained in the following second operation pattern.

[0147] (1.6.2) Second operating pattern for model transfer Figure 19 is a diagram showing an example of a configuration message including a model and additional information according to the first embodiment. The configuration message may be an RRC message sent from gNB200 to UE100 (e.g., an "RRC Reconfiguration" message, or a newly defined message (e.g., an "AI Deployment" message or an "AI Reconfiguration" message)). Alternatively, the configuration message may be a NAS message sent from AMF300 to UE100. Alternatively, if a new layer for executing or controlling machine learning processing (AI / ML processing) is defined, the message may be a message for that new layer.

[0148] In the example in Figure 19, the configuration message includes three models (Model #1 to #3). Each model is included as a container for the configuration message. However, the configuration message may include only one model. The configuration message further includes, as additional information, three individual additional pieces of information (Info #1 to #3) corresponding to each of the three models (Model #1 to #3), and common additional pieces of information (Meta-Info) that are associated with all three models (Model #1 to #3). Each of the individual additional pieces of information (Info #1 to #3) contains information specific to the corresponding model. The common additional pieces of information (Meta-Info) contain information common to all models within the configuration message.

[0149] The individual supplemental information may also be a model index representing the index (index number) assigned to each model. This individual supplemental information may also be model execution conditions indicating the performance (e.g., processing delay) required to apply (execute) the model.

[0150] Individual or common additional information may be model applications that specify the function to which the model applies (e.g., "CSI feedback," "beam management," "positioning," etc.). Such individual or common additional information may also be model selection criteria that apply (execute) the corresponding model when a specified criterion (e.g., travel speed) is met.

[0151] (1.7) Communication control method according to the first embodiment Next, a communication control method according to the first embodiment will be described.

[0152] As mentioned above, Non-Patent Document 1 discusses the achievement of high accuracy in AI inference for mobility failures, including handover command loss and handover failure, as a result of simulations of AI inference for mobility failures. However, Non-Patent Document 1 does not discuss how the AI / ML model was specifically deployed to obtain these simulation results. Similarly, Non-Patent Document 2 does not discuss the specific deployment of the AI / ML model.

[0153] Now, let's consider handover as a form of mobility management. If the AI / ML model exists on the UE100 side, the handover decision is made on the gNB200 during a normal handover. Therefore, the involvement of the AI / ML model is significantly smaller compared to when the AI / ML model exists on the gNB200 side.

[0154] On the other hand, let's consider conditional handover as a type of handover. In this case, since the trigger conditions for the conditional handover are determined on the UE100 side, if an AI / ML model exists on the UE100 side, the likelihood of that AI / ML model being involved is higher compared to when the AI / ML model exists on the gNB200 side.

[0155] However, the trigger conditions for conditional handover are clearly defined. Therefore, even if an AI / ML model exists on the UE100 side, it is expected that the UE100 will have little room to make its own decisions.

[0156] On the other hand, if the UE100 is to make all the decisions regarding conditional handover, it may not always be appropriate from a network control perspective.

[0157] Therefore, the objective of the first embodiment is to enable the UE100 (its AI / ML model) to appropriately determine the timing of conditional handover while retaining network control. Furthermore, the objective of the first embodiment is to enable the UE100 to appropriately perform wireless communication using the AI / ML model by enabling the UE100 to appropriately determine the timing of conditional handover.

[0158] Therefore, in the first embodiment, the gNB200 sets the range of wireless quality over which it may use an AI / ML model to determine the timing of a conditional handover to the UE100.

[0159] Specifically, the base station (e.g., gNB200) sets a predetermined range of radio quality for the user equipment (e.g., UE100). Here, the predetermined range of radio quality represents the range of radio quality for which the user equipment is permitted to determine the timing of conditional handover using an AI / ML model.

[0160] Thus, the UE100 can determine the timing of conditional handover execution using an AI / ML model within a predetermined range of wireless quality, enabling the UE100 to appropriately determine the timing of conditional handover execution. Furthermore, since the predetermined range of wireless communication is determined by the gNB200 and transmitted to the UE100, the UE100 can appropriately determine the timing of conditional handover while maintaining network control. Therefore, the UE100 can appropriately execute wireless communication using an AI / ML model.

[0161] (1.8) Conditional Handover (CHO) Here, a conditional handover according to the first embodiment will be described. A conditional handover is a handover performed by the UE100 when one or more handover execution conditions are met. The UE100 starts evaluating the handover execution conditions when it receives a ConditionalReconfiguration from the gNB200. The handover execution conditions include one or two trigger conditions. The conditional configuration includes candidate cells and trigger conditions. If at least one candidate cell satisfies the handover execution conditions, the UE100 detaches from the source gNB and starts connecting to the selected candidate cell. The conditional configuration is notified to the UE100 from the gNB200 by individual signaling (e.g., an RRCReconfiguration message).

[0162] Thus, unlike a normal handover (which may be referred to as "legacy handover" below) in which the UE100 reports wireless status measurements to the gNB200 and the gNB200 decides on a handover to an adjacent cell based on that report, conditional handover can autonomously perform a handover to a candidate cell that satisfies the trigger conditions.

[0163] (1.9) Example of operation according to the first embodiment Next, an example of operation according to the first embodiment will be described.

[0164] The AI / ML model used in the operational example according to the first embodiment resides on the UE100 ("UE-side one-sided model"). The input to the AI / ML model is information about the wireless environment. Specifically, this may be measurement information for a serving cell or candidate cell. Alternatively, the input may be movement information indicated by the speed or direction of the UE100. Or, the input may be the position information of the UE100. On the other hand, the output to the AI / ML model is the execution timing of the conditional handover (i.e., the timing of access to the target cell).

[0165] In other words, the AI / ML model used in the operation example of the first embodiment is a model that takes information about the wireless environment as input and outputs the timing for executing a conditional handover.

[0166] In the following, the terms "AI / ML model" and "inference model" may not be distinguished. "Inference model" may also refer to a "trained model." Furthermore, the "input" of an inference model may be referred to as "inference data," and the "output" of an inference model may be referred to as the "inference result."

[0167] Figure 20 is a diagram illustrating an example of operation according to the first embodiment.

[0168] As shown in Figure 20, in step S501, UE100 may notify gNB200 that it has the capability to determine (or infer) the timing of the conditional handover. An RRC message (e.g., a UECapability message) may be used for this notification.

[0169] In step S502, the gNB200 configures a conditional handover for the UE100. Specifically, the gNB200 may configure the conditional handover using the ConditionalReconfiguration described above.

[0170] Firstly, the setting for conditional handover includes a range of radio quality that allows the UE100 to make conditional handover decisions. The range of radio quality that allows the UE100 to make conditional handover decisions may be referred to below as the "predetermined range of radio quality." The predetermined range of radio quality may represent the range of radio quality that the UE100 is permitted to use an AI / ML model to determine the timing of conditional handover. For example, the predetermined range of radio quality may be set to a serving cell RSRP of "-80dBm to -90dBm." If the RSRP for the serving cell is within this range, the UE100 can use an AI / ML model to infer (or determine) the timing of conditional handover. The radio quality in question is one of RSRP (Reference Signal Received Power), RSRQ (Reference Signal Received Quality), or SINR (Signal to Interference plus Noise Ratio).

[0171] Specifically, the first condition, which includes a predetermined range of wireless quality, is included in the setting of conditional handover.

[0172] The first condition, for example, is as follows:

[0173] (X1) If event A2 (Serving becomes worse than threshold) is included in the conditional handover settings, this indicates that if the wireless quality of the serving cell (wireless quality within the specified range) deteriorates below the event A2 threshold, the UE100 is allowed to determine the timing of the conditional handover using an AI / ML model.

[0174] (X2) If event A3 (Neighbor becomes offset better than PCel) is included in the conditional handover setting, it indicates that if the wireless quality of the adjacent cell (wireless quality within the specified range) is better than the wireless quality of the serving cell (wireless quality within the specified range) with an offset added, the UE100 is allowed to determine the timing of the conditional handover using an AI / ML model.

[0175] (X3) If event A4 (Neighbor becomes better than threshold) is included in the conditional handover setting, it indicates that if the wireless quality of the adjacent cell (wireless quality within the specified range) is better than the event A3 threshold, the UE100 is allowed to determine the timing of the conditional handover using an AI / ML model.

[0176] (X4) If event A5 (PCell becomes worse than threshold 1 and neighbor becomes better than threshold 2) is included in the conditional handover setting, it indicates that the UE100 is allowed to determine the timing of conditional handover execution using an AI / ML model if the serving cell's wireless quality is worse than the first threshold of event A5 and the neighboring cell's wireless quality is better than the second threshold of event A5.

[0177] The predetermined range of wireless communication may be the range that allows conditional handover by UE100 model inference. This range may be the first condition. Specifically, this range may be, for example, a range indicated by an upper limit and a lower limit. This range may be a range indicated by only an upper limit. This range may be a range indicated by only a lower limit. This range may be expressed as an offset to the second condition, which will be described below.

[0178] Secondly, the conditional handover setting may include radio quality conditions that trigger a conditional handover regardless of the UE100's determination of the timing of the conditional handover. For example, if, for some reason, the AI / ML model fails to determine the timing of the conditional handover, the radio quality condition may be used to force the UE100 to perform a conditional handover when a specific radio quality is reached. The radio quality condition may be used, for example, as a remedy when the UE100 is unable to determine the timing. The radio quality condition may be represented by a radio quality threshold that causes the UE100 to perform a conditional handover without using the AI / ML model. For example, the radio quality condition may be used to ensure that the UE100 always performs a conditional handover when the serving cell's RSRP becomes "-90dBm". The radio quality condition may also be a radio quality threshold that causes the UE100 to perform a conditional handover without using the AI / ML model.

[0179] The wireless quality condition is included in the second condition. The second condition may be the same as the trigger conditions for an existing conditional handover (e.g., events A2 through A5). The wireless quality condition is, for example, a threshold included in the trigger conditions.

[0180] In this way, the gNB200 sets a first condition for conditional handover, which includes a predetermined range of wireless quality, and a second condition for conditional handover, which includes a threshold for wireless quality (i.e., a wireless quality condition), in the UE100.

[0181] In step S503, the UE100 measures the wireless quality of the serving cell and / or adjacent cells.

[0182] In step S504, if the wireless quality satisfies the first condition, the UE100 uses an AI / ML model to infer (or determine) when to perform a conditional handover.

[0183] Then, in step S505, if the AI / ML model infers an appropriate execution timing, the UE100 performs a conditional handover at that execution timing and begins accessing the target cell.

[0184] On the other hand, in step S506, if the AI / ML model is unable to properly infer the execution timing, and the wireless quality satisfies the second condition, the UE100 performs a conditional handover without using the AI / ML model. The UE100 may stop model inference by the AI / ML model.

[0185] (1.10) Other examples according to the first embodiment In the first embodiment, we described allowing the AI / ML model to execute within a range of wireless quality, but we are not limited to this. For example, the range may be permitted not only by wireless quality but also by distance or time, and the AI / ML model may be allowed to execute conditional handover within the set range. For example, the trigger conditions included in the first condition may include the following two:

[0186] (X5) Conditional Event D1 (CondEvent D1): If Conditional Event D1 is included in the conditional handover setting, it indicates that if the distance between UE100 and the first reference position is greater than the first threshold of Event D1, and the distance between UE100 and the second reference position of the conditional reset candidate is shorter than the second threshold of Event D1, then UE100 is allowed to determine the timing of the conditional handover using the AI / ML model.

[0187] (X6) Conditional Event T1 (CondEvent) T1): If conditional event T1 is included in the conditional handover setting, this indicates that if the time measured by UE100 is longer than the event T1 threshold and shorter than (event T1 threshold - predetermined threshold (Threshold) + duration), then UE100 is allowed to determine the timing of the conditional handover using the AI / ML model.

[0188] For example, the distance range for which execution by the AI / ML model is permitted is set based on the distance in conditional event D1. Also, for example, the execution timing of the AI / ML model is permitted based on the measurement time in conditional event T1. time A range is set. If conditional event D1 or conditional event T1 is used in the second condition, a range may be set for each threshold in which a conditional handover is forced to be executed.

[0189] [Second Embodiment] Next, a second embodiment will be described. The second embodiment will be described focusing on the differences from the first embodiment.

[0190] As described in the first embodiment, when the UE100 makes a decision regarding conditional handover, the network may need to collect information on whether the handover failure due to the conditional handover was caused by the UE100's decision or by a network instruction. In the second embodiment, an example is described in which the UE100 sends log information to the gNB200 indicating whether the failure was due to the UE100's decision or a network instruction.

[0191] Specifically, firstly, the user device sends log information to the base station that includes the result of the conditional handover, and execution trigger information indicating either that the conditional handover was performed using an AI / ML model or that the conditional handover was performed using network settings without using an AI / ML model.

[0192] This type of information gathering enables the network to optimize its area of ​​operation and the control of the inference model in the UE100. Furthermore, this optimization allows the UE100 to appropriately perform wireless communication using AI / ML models, for example.

[0193] (2.1) MDT Here, we will describe the MDT used in the second embodiment.

[0194] In some cases, the radio conditions in a coverage area are measured through drive tests conducted by operators. Information collected through drive tests can be used to optimize base station settings and base station antenna tilt. However, conducting drive tests by operators can be time-consuming and costly. Therefore, 3GPP is considering having the UE100 measure and report the information collected during drive tests. This would reduce operating expenses (OPEX). The collective term for the techniques used to minimize the execution of drive tests is, for example, MDT (Minimization of Drive Test).

[0195] MDT specifies two methods for acquiring and reporting measurement information in the UE100: Immediate MDT and Logged MDT. Immediate MDT is a method in which the UE100 in an RRC connected state acquires and reports measurement information. In Immediate MDT, processing is carried out based on the RRC settings (MeasurementConfiguration) and reporting procedures related to the measurement. On the other hand, Logged MDT is a method in which the UE100 in an RRC idle or RRC inactive state acquires and reports measurement information. In Logged MDT, the UE100 records (logs) the measurement results; that is, it does not immediately report the measurement results to the gNB200, but rather reports the measurement results to the gNB200 in response to a request from the gNB200 after acquiring the measurement results. In the case of Logged MDT, the UE100 processes based on the RRC settings (LoggedMeasurementConfiguration).

[0196] (2.2) Example of operation according to the second embodiment Next, an example of operation according to the second embodiment will be described.

[0197] Figure 21 is a diagram illustrating an example of operation according to the second embodiment. It is assumed that UE100 has been configured for MDT from gNB200.

[0198] In step S601, UE100 performs a conditional handover and initiates access to the target cell.

[0199] In step S602, UE100 records log information upon completion of the conditional handover. The log information may be any of the following:

[0200] (Y1) Result of the conditional handover: Specifically, this is information indicating either that the conditional handover was successful or that the conditional handover failed.

[0201] (Y2) Conditional handover trigger: Specifically, this may be information indicating, for example, that inference by an AI / ML model was used as the trigger for the execution of a conditional handover, or that network settings were used as the trigger for the execution of a conditional handover. Alternatively, it may be information indicating, for example, that a conditional handover was performed using an AI / ML model, or that a conditional handover was performed using network settings without using an AI / ML model. Such information indicating the trigger for a conditional handover may be referred to as "execution trigger information".

[0202] (Y3) Identification information of the AI / ML model used in UE100: This identification information may be, for example, the model ID of the AI / ML model. Alternatively, this identification information may be represented by the model attributes of the AI / ML model (for example, either a proprietary model or an open format model).

[0203] (Y4) Current radio information: for example, the radio quality of the source cell and target cell.

[0204] (Y5) Timestamp, or location information representing the position of UE100, etc.

[0205] In step S603, UE100 may notify gNB200 that it is recording log information. UE100 may make this notification using an RRC message or MAC CE.

[0206] In step S604, gNB200 may request UE100 to send log information. gNB200 may make this request using an RRC message, MAC CE (Control Element), or DCI.

[0207] In step S605, UE100 sends log information to gNB200. UE100 may also send log information using a measurement report. UE100 may also send log information using other RRC messages.

[0208] In step S606, the gNB200 uses the received log information to perform area optimization and optimize the AI / ML model control in the UE100.

[0209] (2.3) Other examples according to the second embodiment The second embodiment describes an example in which a conditional handover is performed in UE100, but is not limited to this. For example, the second embodiment can also be implemented even when a legacy handover is performed in UE100. In this case, it can be implemented by replacing "execution of conditional handover" in step S601 with "execution of legacy handover". In addition, UE100 may use an AI / ML model to infer the timing of the legacy handover execution. The input (inference data) to the AI / ML model is the same as in the first embodiment. Furthermore, it can be implemented by replacing "(Y1) execution result of conditional handover" with "execution result of legacy handover".

[0210] In other words, the user device (e.g., UE100) sends log information to the base station (e.g., gNB200) that includes the result of the handover and execution trigger information indicating either that the handover was performed using an AI / ML model or that the handover was performed using network settings without using an AI / ML model.

[0211] This allows the network to optimize areas and inference model control in the UE100 based on log information, even in the case of legacy handover. Such optimizations also enable the UE100 to properly perform wireless communication using AI / ML models, for example.

[0212] [Other embodiments] The first embodiment described above primarily describes supervised learning, but is not limited thereto. For example, the first embodiment may be applied to unsupervised learning or reinforcement learning.

[0213] Each of the above-described operation flows can be performed not only independently, but also in combination of two or more operation flows. For example, some steps of one operation flow may be added to another operation flow, or some steps of one operation flow may be replaced with some steps of another operation flow. It is not necessary to execute all steps in each flow; only some steps may be executed.

[0214] In the embodiments and examples described above, an example in which the base station is an NR base station (gNB) was described, but the base station may also be an LTE base station (eNB) or a 6G base station. Furthermore, the base station may also be a relay node such as an IAB (Integrated Access and Backhaul) node. The base station may also be a DU of an IAB node. Furthermore, UE100 may be an MT (Mobile Termination) of an IAB node.

[0215] Furthermore, the term "network node" primarily refers to a base station, but may also refer to a core network device or a part of a base station (CU, DU, or RU). Additionally, a network node may consist of a combination of at least a part of the core network device and at least a part of a base station.

[0216] Furthermore, a program (e.g., an information processing program) that causes a computer to execute each of the processes or functions according to the above-described embodiment may be provided. Alternatively, a program (e.g., a mobile communication program) that causes the mobile communication system 1 to execute each of the processes or functions according to the above-described embodiment may be provided. The program may be recorded on a computer-readable medium. Using a computer-readable medium, it is possible to install the program on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transient recording medium. The non-transient recording medium is not particularly limited, but may be a recording medium such as a CD-ROM or DVD-ROM. Such a recording medium may be the memory included in the UE100 and gNB200. Furthermore, the circuits that execute each of the processes performed by the UE100 or gNB200 may be integrated, and at least a part of the UE100 or gNB200 may be configured as a semiconductor integrated circuit (chipset, SoC: System on a chip).

[0217] The functions realized by UE100 or gNB200 (network node) may be implemented in circuitry or processing circuitry, including general-purpose processors, application-specific processors, integrated circuits, ASICs (Application Specific Integrated Circuits), CPUs (a Central Processing Unit), conventional circuits, and / or combinations thereof, programmed to realize the described functions. A processor, including transistors and other circuits, is considered circuitry or processing circuitry. A processor may be a programmed processor that executes a program stored in memory. In this specification, circuitry, unit, and means are hardware programmed to realize or perform the described functions. Such hardware may be any hardware disclosed herein, or any hardware known to be programmed to realize or perform the described functions. If such hardware is a processor that is considered to be a type of circuitry, then such circuitry, means, or unit is a combination of hardware and software used to constitute such hardware and / or processor.

[0218] The phrases “based on” and “depending on / in response to” used in this disclosure do not mean “based solely on” or “depending solely on” unless otherwise specified. “Based on” means both “based solely on” and “at least partially on.” Similarly, “depending on” means both “at least partially on” and “at least partially on.” The terms “include,” “comprise,” and variations thereof do not mean that only the listed items are included; they mean that only the listed items may be included, or that additional items may be included in addition to the listed items. Furthermore, the term “or” used in this disclosure is not intended to mean exclusive OR. Additionally, any reference to elements using designations such as “first,” “second,” etc., used in this disclosure does not generally limit the quantity or order of those elements. These designations may be used herein as a convenient way to distinguish between two or more elements. Therefore, references to the first and second elements do not imply that only two elements may be adopted therein, or that the first element must precede the second element in any way. In this disclosure, where articles are added by translation, such as a, an, and the in English, these articles shall be plural unless it is clearly indicated by the context that they are not.

[0219] Although the embodiments have been described in detail above with reference to the drawings, the specific configuration is not limited to those described above, and various design changes can be made without departing from the gist of the work. Furthermore, it is possible to combine each embodiment, each operation example, or each process, as long as it is not contradictory.

[0220] This application claims priority to Japanese Patent Application No. 2023-017192 (filed on February 7, 2023), the entirety of which is incorporated into the specification of this application.

[0221] (Note) (Note 1) A communication control method in a mobile communication system, The network node has the step of setting a predetermined range of wireless quality to the user device, The predetermined range of wireless quality refers to the range of wireless quality in which the user device is permitted to determine the timing of conditional handover using an AI / ML model. Communication control method.

[0222] (Note 2) The setting step includes setting a wireless quality threshold on the user device that allows the network node to perform the conditional handover without using the AI / ML model. The communication control method described in Appendix 1.

[0223] (Note 3) The setting step includes setting a first condition for the conditional handover, which includes a predetermined range of the wireless quality, and a second condition for the conditional handover, which includes a threshold for the wireless quality, on the user device. The user device further includes the steps of: if the wireless quality satisfies the first condition, it performs the conditional handover at the execution timing inferred using the AI / ML model; and if the AI / ML model is unable to appropriately infer the execution timing, and the wireless quality satisfies the second condition, it performs the conditional handover without using the AI / ML model. The communication control method described in Appendix 1 or Appendix 2.

[0224] (Note 4) The user device further includes the step of sending log information to the network node, which includes the result of the conditional handover and execution trigger information indicating either that the conditional handover was performed using the AI / ML model or that the conditional handover was performed using network settings without using the AI / ML model. A communication control method as described in any of Appendix 1 to Appendix 3.

[0225] (Note 5) A communication control method in a mobile communication system, The user device sends log information to a network node, which includes the result of the handover and execution trigger information indicating either that the handover was performed using an AI / ML model or that the handover was performed using network settings without using the AI / ML model. Communication control method. [Explanation of symbols]

[0226] 1: Mobile communication systems 20:5GC(CN) 100:UE 110: Receiver 120: Transmitter 130: Control Unit 200:gNB 210: Transmitter 220: Receiver 230: Control Unit A1: Data Collection Department A2: Model Learning Department A3: Model inference section A4: Data Processing Section TE: Sending entity RE: Receiving Entity

Claims

1. A communication control method in a mobile communication system, The network node sets a predetermined range of wireless quality for the user device. The predetermined range of wireless quality refers to the range of wireless quality in which the user device is permitted to determine the timing of conditional handover using an AI (Artificial Intelligence) / ML (Machine Learning) model. Communication control method.

2. The above setting includes setting a wireless quality threshold on the user device that allows the network node to perform the conditional handover without using the AI / ML model. The communication control method according to claim 1.

3. The setting described above includes setting a first condition for the conditional handover, which includes a predetermined range of the wireless quality, and a second condition for the conditional handover, which includes a threshold value for the wireless quality, on the user device. The user device further includes the following: if the wireless quality satisfies the first condition, it performs the conditional handover at the execution timing inferred using the AI / ML model; and if the AI / ML model is unable to appropriately infer the execution timing, and the wireless quality satisfies the second condition, it performs the conditional handover without using the AI / ML model. The communication control method according to claim 2.

4. The user device further transmits log information to the network node, which includes the result of the conditional handover and execution trigger information indicating either that the conditional handover was performed using the AI / ML model or that the conditional handover was performed using network settings without using the AI / ML model. The communication control method according to claim 1.

5. A communication control method in a mobile communication system, The user device transmits log information to the network node, which includes the result of the handover and execution trigger information indicating either that the handover was performed using an AI / ML model or that the handover was performed using network settings without using the AI / ML model. Communication control method.

6. A network node in a mobile communication system, It has a control unit that sets a predetermined range of wireless quality for the user device, The predetermined range of wireless quality refers to the range of wireless quality in which the user device is permitted to determine the timing of conditional handover using an AI (Artificial Intelligence) / ML (Machine Learning) model. Network node.

7. A user device in a mobile communication system, It has a receiving unit that receives information including a predetermined range of wireless quality from a network node, The predetermined range of wireless quality refers to the range of wireless quality in which the user device is permitted to determine the timing of conditional handover using an AI (Artificial Intelligence) / ML (Machine Learning) model. User device.

Images