COMMUNICATION METHOD, DATA TRANSMISSION ENTITY AND USER EQUIPMENT - Patent application

Abstract

A communication method according to one aspect of the present invention is for a mobile communication system. The communication method has a step in which a data transmission entity derives a trained model with respect to a prescribed block on the basis of a signal received from a data reception entity. In addition, the communication method has a step in which the data transmission entity calculates output data with respect to input data, by using the prescribed block or the trained model. Further, the communication method has a step in which the data transmission entity performs a prescribed process when the likelihood of the input data and / or the likelihood of the output data is not more than a threshold value.

Description

Communication Method

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

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

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

[0004] A communication method according to one aspect is a communication method in a mobile communication system. The communication method includes a step in which a data transmitting entity derives a trained model for a predetermined block based on a signal received from a data receiving entity. The communication method also includes a step in which the data transmitting entity calculates output data for input data using either the predetermined block or the trained model. The communication method further includes a step in which the data transmitting entity performs a predetermined process when the likelihood of the input data and / or the likelihood of the output data is equal to or less than a threshold.

[0005] Also, a communication method according to one aspect is a communication method in a mobile communication system. The communication method includes a step in which a data transmitting entity derives a trained model for a predetermined block based on a signal received from a data receiving entity. The communication method also includes a step in which the data transmitting entity calculates output data for input data using either the predetermined block or the trained model. The communication method further includes a step in which the data transmitting entity transmits the input data and the output data to the data receiving entity. The communication method also includes a step in which the data receiving entity calculates the likelihood of the input data and / or the likelihood of the output data.

[0006] Furthermore, a communication method according to one aspect is a communication method in a mobile communication system. The communication method includes a step in which a first user device derives a trained model for a predetermined block based on a signal received from a base station. The communication method also includes a step in which the first user device calculates output data for input data using the predetermined block and / or the trained model. The communication method also includes a step in which the first user device calculates a likelihood of the input data and / or a likelihood of the output data. The communication method also includes a step in which the first user device associates the input data and / or the output data with the likelihood and transmits the input data and / or the output data and the likelihood to the base station. The communication method also includes a step in which the base station transmits the input data and / or the output data and the likelihood to a second user device. The communication method also includes a step in which the second user device determines whether to use the input data and / or the output data as training data based on the likelihood.

[0007] Furthermore, a communication method according to one aspect is a communication method in a user equipment (UE) that generates a measurement report using a measurement report model, the communication method including a step of calculating, by the UE, a likelihood of input data to be input to a predetermined block included in the measurement model and a likelihood of first output data to be output from the predetermined block, and a step of discarding, by the UE, the input data and / or the first output data when the likelihood of the input data and / or the first output data is equal to or less than a first threshold.

[0008] 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. FIGS. 7A to 7C are diagrams showing an example of the configuration of a mobile communication system according to the first embodiment. FIG. 8 is a diagram showing an example of the configuration of a data transmitting entity and a data receiving entity according to the first embodiment. FIG. 9 is a diagram showing an example of the configuration of a first operation scenario according to the first embodiment. FIG. 10 is a diagram showing an example of the operation of the first operation scenario according to the first embodiment. FIG. 11 is a diagram showing an example of the configuration of a second operation scenario according to the first embodiment. FIG. 12 is a diagram showing an example of the configuration of a third operation scenario according to the first embodiment. FIG. 13 is a diagram showing an example of the operation of the third operation scenario according to the first embodiment. FIG. 14 is a diagram showing an example of the configuration of the third operation scenario according to the first embodiment. Fig. 15 is a diagram showing a configuration example of a fourth operation scenario according to the first embodiment. Fig. 16 is a diagram showing a configuration example of a measurement reporting model according to the first embodiment. Fig. 17 is a diagram showing an operation example of the fourth operation scenario according to the first embodiment. Fig. 18 is a diagram showing an operation example according to the second embodiment. Fig. 19 is a diagram showing a configuration example according to the third embodiment. Fig. 20 is a diagram showing an operation example according to the third embodiment. Fig. 21 is a diagram showing an operation example according to the fourth embodiment. Fig. 22 is a diagram showing an operation example according to the fifth embodiment. Fig. 23 is a diagram showing a configuration example of a mobile communication system including a dedicated center.

[0009] In machine learning technology, when machine learning is performed using idiosyncratic data, the learning accuracy decreases compared to when machine learning is performed using appropriate data. Furthermore, in mobile communication systems, when processing is performed using idiosyncratic data, the processing may not be performed appropriately.

[0010] Therefore, an object of the present disclosure is to provide a communication method in a mobile communication system that can improve learning accuracy in machine learning, and also to provide a communication method in a mobile communication system that can perform appropriate processing.

[0011] [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.

[0012] (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.

[0013] 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.

[0014] The UE 100 is a mobile wireless communication device. The UE 100 may be any device that is used by a user. For example, the UE 100 may be a mobile phone terminal (including a smartphone) and / or a tablet terminal, a notebook PC, a communication module (including a communication card or a chipset), a sensor or a device provided in a sensor, a vehicle or a device provided in a vehicle (Vehicle UE), or an aircraft or a device provided in an aircraft (Aerial UE).

[0015] 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").

[0016] 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.

[0017] 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.

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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 and encoding / decoding of baseband signals. The CPU executes programs stored in the memory to perform various processes.

[0022] 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.

[0023] 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.

[0024] 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.

[0025] The control unit 230 performs various controls and processes in the gNB 200. Such processes include processes for each layer, which will be 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 in the processes by the processor. The processor may include a baseband processor and a CPU. The baseband processor performs modulation / demodulation and encoding / decoding of baseband signals. The CPU executes programs stored in the memory to perform various processes.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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 the successfully decoded DCI as DCI addressed to the UE. The DCI transmitted from gNB200 has a CRC parity bit scrambled by the RNTI added.

[0030] In NR, UE100 can use a bandwidth narrower than the system bandwidth (i.e., the cell bandwidth). The gNB200 configures a bandwidth portion (BWP) consisting of contiguous PRBs (Physical Resource Blocks) for UE100. UE100 transmits and receives data and control signals in the active BWP. For example, up to four BWPs may be configurable for UE100. Each BWP may have a different subcarrier spacing. The frequencies of the BWPs may overlap with each other. When multiple BWPs are configured for 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 UE100, etc., and reduce UE power consumption.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] The PDCP layer performs header compression / decompression, encryption / decryption, and the like.

[0035] 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.

[0036] 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).

[0037] 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.

[0038] 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.

[0039] 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).

[0040] (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. The functional blocks of the AI / ML technology may be referred to as AI / ML blocks hereinafter. Fig. 6 shows an example of the configuration of an AI / ML block AB.

[0041] The AI / ML block AB 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.

[0042] The data collection unit A1 collects input data, specifically, learning data and inference data. The data collection unit A1 outputs the learning data to the model learning unit A2. The data collection unit A1 also outputs the inference data to the model inference unit A3. The data collection unit A1 may acquire data in the device on which the data collection unit A1 is provided as input data. The data collection unit A1 may also acquire data in another device as input data.

[0043] The model learning unit A2 performs model learning. Specifically, the model learning unit A2 optimizes the 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, in the case of y = ax + b, a (slope) and b (intercept) are parameters, and optimizing these corresponds to machine learning.

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

[0045] The model inference unit A3 performs model inference. Specifically, the model inference unit A3 infers an output from inference data using a trained model and outputs the inference result data to the data processing unit A4. For example, in the case of y = ax + b, x corresponds to the inference data and y corresponds to the inference result data. Note that "y = ax + b" is a model. A model in which the slope and intercept are optimized, for example, "y = 5x + 3", is a trained model. Here, there are various model approaches, such as linear regression analysis, neural networks, and decision tree analysis. The above "y = ax + b" can be considered a type of linear regression analysis. The model inference unit A3 may provide model performance feedback to the model learning unit A2.

[0046] The data processing unit A4 receives the inference result data and performs processing that utilizes the inference result data.

[0047] Hereinafter, a block including the data collection unit A1, the model learning unit A2, and the model inference unit A3 may be referred to as an AI / ML processing block AP. The AI / ML processing block AP derives a trained model from training data and outputs inference result data from inference data using the derived trained model.

[0048] (Application Example of First Embodiment) Next, an example in which machine learning technology is applied to the mobile communication system 1 will be described.

[0049] FIG. 7A is a diagram illustrating an example of the configuration of a mobile communication system 1 according to the first embodiment.

[0050] 7A, the mobile communication system 1 includes a data transmitting entity TE and a data receiving entity RE. The entities are, for example, a device, a functional block included in the device, or a hardware block included in the device.

[0051] The data transmitting entity TE is, for example, an entity in which machine learning is performed. The data transmitting entity TE performs machine learning based on the received signal received from the data receiving entity RE to derive a trained model. The data transmitting entity TE then uses the trained model to generate inference result data as an inference result. The data transmitting entity TE transmits a transmission signal including the inference result data to the data receiving entity RE.

[0052] The data receiving entity RE is, for example, an entity in which machine learning is not performed. The data receiving entity RE receives inference result data from the data transmitting entity and performs various processes using the received inference result data.

[0053] Figures 7(B) and 7(C) are diagrams showing an example of the configuration of the mobile communication system 1 according to the first embodiment. As shown in Figure 7(B), the data transmitting entity TE may be the UE 100, and the data receiving entity RE may be the gNB 200. Also, as shown in Figure 7(C), the data transmitting entity TE may be the gNB 200, and the data receiving entity RE may be the UE 100.

[0054] In this way, when machine learning technology is applied to the mobile communication system 1, there are cases where machine learning is performed in the UE 100 and cases where machine learning is performed in the gNB 200. In the following, in order to distinguish between the entities where machine learning is performed, without distinguishing between the UE 100 and the gNB 200, they may be described as a data transmitting entity TE and a data receiving entity RE.

[0055] (Configuration Examples of Data Transmitting Entity and Data Receiving Entity) Next, configuration examples of each entity TE and RE will be described.

[0056] FIG. 8 is a diagram illustrating an example of the configuration of a data transmitting entity TE and a data receiving entity RE according to the first embodiment.

[0057] As shown in FIG. 8, the data transmitting entity TE includes a receiving part TE-1, a transmitting part TE-2 and a controlling part TE-3.

[0058] The receiving unit TE-1 receives a signal (received signal) transmitted from the data receiving entity RE. The receiving unit TE-1 extracts received data from the received signal and outputs the extracted received data to the control unit TE-3. The receiving unit TE-1 corresponds to, for example, the receiving unit 110 of the UE 100. The receiving unit TE-1 also corresponds to, for example, the receiving unit 220 of the gNB 200.

[0059] The transmitter TE-2 receives input data and output data from the controller TE-3 and generates a transmission signal including the input data and output data. The transmitter TE-2 transmits the generated transmission signal to the data receiving entity RE. The transmitter TE-2 may transmit the input data and the output data in separate transmission signals instead of in a single transmission signal.

[0060] The control unit TE-3 controls the data transmission entity TE. The control unit TE-3 corresponds to, for example, the control unit 130 of the UE 100 or the control unit 230 of the gNB 200.

[0061] The control unit TE-3 includes an AI / ML processing block AP, which performs machine learning.

[0062] The control unit TE-3 also includes a legacy processing block LP. The legacy processing block LP is, for example, a block that is the target of a trained model derived in the AI / ML processing block AP. The AI / ML processing block AP derives a trained model targeted at the legacy processing block LP. The legacy processing block LP is a different block depending on the operation scenario in which machine learning is applied in the mobile communication system 1. For example, in an operation scenario using positioning accuracy enhancement, the legacy processing block LP becomes a position information generation unit. Furthermore, for example, in an operation scenario using beam management, the legacy processing block LP becomes an optimal beam determination unit. Furthermore, for example, in the case of an operation scenario using channel state information (CSI) feedback (CSI feedback enhancement), the legacy processing block LP becomes a CSI generating unit. Note that, although the example shown in FIG. 8 shows an example in which the legacy processing block LP is included in the control unit TE-3, the legacy processing block LP may be provided outside the control unit TE-3.

[0063] Furthermore, the control unit TE-3 includes a likelihood filter LF-1 for input data and a likelihood filter LF-2 for output data.

[0064] The input data likelihood filter LF-1 is provided before the legacy processing block LP and the AI / ML processing block AP, i.e., between the legacy processing block LP and the AI / ML processing block AP and the receiving unit TE-1. The input data likelihood filter LF-1 calculates a likelihood for the input data. When the likelihood is equal to or less than a threshold value (e.g., a first threshold value or a second threshold value), the input data likelihood filter LF-1 performs a predetermined process. The predetermined process may include discarding the input data, linking the input data to information indicating that the likelihood is equal to or less than the threshold value, or linking the input data to the likelihood. The input data likelihood filter LF-1 outputs input data whose likelihood is greater than the threshold value to the legacy processing block LP and the AI / ML processing block AP.

[0065] The output data likelihood filter LF-2 is provided downstream of the legacy processing block LP and the AI / ML processing block AP, i.e., between the legacy processing block LP and the AI / ML processing block AP and the transmitter TE-2. The output data likelihood filter LF-2 calculates a likelihood for the output data of the legacy processing block LP. The output data likelihood filter LF-2 performs a predetermined process when the likelihood of the output data from the legacy processing block LP is equal to or less than a threshold (e.g., a first threshold). The predetermined process may include discarding the output data, linking the output data to information indicating that the likelihood is equal to or less than the threshold, or linking the output data to the likelihood. The output data likelihood filter LF-2 calculates a likelihood for the output data from the AI / ML processing block AP (i.e., inference result data). The output data likelihood filter LF-2 performs a predetermined process when the likelihood of the output data from the AI / ML processing block AP is equal to or less than a threshold (e.g., a second threshold). The predetermined processing may include discarding the output data or associating the output data with information indicating that the likelihood is equal to or less than a threshold value.

[0066] In FIG. 8, the likelihood filter LF-2 for output data is shown as two blocks in the drawing, but it may be one block.

[0067] The input data is data input to the legacy processing block LP and the AL / ML processing block AP. The input data is also inference data input to the model inference unit A3 of the AI / ML processing block AP. The input data may be learning data input to the model learning unit A2 of the AI / ML processing block AP. Specific examples of input data vary depending on the operation scenario. Specific examples of input data will be explained as appropriate when explaining each operation scenario.

[0068] Furthermore, output data is data output from the legacy processing block LP and the AL / ML processing block AP. The output data is also inference result data output from the model inference unit A3 of the AI / ML processing block AP. Specific examples of output data also differ depending on the operation scenario. Specific examples of output data will also be explained appropriately when explaining each operation scenario.

[0069] Furthermore, likelihood is, for example, an index representing plausibility. The higher the likelihood, the more likely the data is, and the lower the likelihood, the less likely the data is.

[0070] The data receiving entity RE includes a transmitting part RE-1, a receiving part RE-2 and a controlling part RE-3.

[0071] The transmitter RE-1 transmits a signal (received signal) to the data transmitting entity TE. The transmitter RE-1 corresponds to, for example, the transmitter 120 of the UE 100 or the transmitter 210 of the gNB 200.

[0072] The receiver RE-2 receives a signal (transmission signal) transmitted from the data transmitting entity TE. The receiver RE-2 extracts data from the signal and outputs the extracted data to the control unit RE-3. The receiver RE-2 corresponds to, for example, the receiver 110 of the UE 100 or the receiver 220 of the gNB 200.

[0073] The control unit RE-3 controls the data receiving entity RE. The control unit RE-3 corresponds to, for example, the control unit 130 of the UE 100 or the control unit 230 of the gNB 200. The control unit RE-3 includes a data processing unit A4 of the AI / ML block AB.

[0074] Thus, in the first embodiment, the likelihood filters LF-1 and LF-2 in the data transmitting entity TE make it possible to discard input data and / or output data whose likelihood is equal to or less than a threshold value.

[0075] Specifically, first, a data transmitting entity (e.g., a data transmitting entity TE) derives a trained model for a predetermined block (e.g., a legacy processing block) based on a signal (e.g., a received signal) received from a data receiving entity (e.g., a data receiving entity RE). Second, the data transmitting entity calculates output data for input data using either the predetermined block or the trained model. Third, the data transmitting entity performs a predetermined process when the likelihood of the input data and / or the output data is equal to or less than a threshold. The predetermined process is one of a process in which the data transmitting entity discards the input data and / or the output data, a process in which the data transmitting entity associates information indicating that the likelihood of the input data and / or the output data is equal to or less than a threshold with the input data and / or the output data, and a process in which the data transmitting entity associates the likelihood with the input data and / or the output data.

[0076] As a result, for example, when output data for input data is calculated using a trained block, input data and / or output data with likelihoods below a threshold are discarded. Therefore, the AI / ML processing block AP can perform machine learning using appropriate data without using idiosyncratic data. Therefore, in the first embodiment, it is possible to improve learning accuracy compared to when machine learning is performed using idiosyncratic data.

[0077] Furthermore, for example, the data transmitting entity TE can discard input data and / or output data whose likelihood is equal to or less than a threshold. Therefore, the data transmitting entity TE will not transmit anomalous data whose likelihood is equal to or less than a threshold to the data receiving entity RE. Therefore, the data receiving entity RE will not perform processing using the anomalous data, and therefore can perform processing more appropriately than when processing is performed using the anomalous data.

[0078] Furthermore, when information indicating that the likelihood of input data and / or the likelihood of output data is equal to or less than a threshold is associated with the input data and / or the output data, even if the data receiving entity RE receives the input data and / or the output data from the data transmitting entity TE, the data receiving entity RE can discard the input data and / or the output data because the information is associated with the input data and / or the output data. Furthermore, when the likelihood is associated with the input data and / or the output data, the data receiving entity RE can discard the input data and / or the output data based on the likelihood. Therefore, the data receiving entity RE can perform processing without using peculiar data whose likelihood is equal to or less than a threshold, and therefore can perform more appropriate processing than when processing is performed using peculiar data.

[0079] 8, an example in which both the likelihood filter LF-1 for input data and the likelihood filter LF-2 for output data are used has been described, but the present invention is not limited to this. For example, the likelihood filter may include either the likelihood filter LF-1 for input data or the likelihood filter LF-2 for output data in the data transmitting entity TE. This is because, by discarding input data or output data whose likelihood is equal to or less than a threshold value using either of the likelihood filters, it is possible to improve learning accuracy and perform appropriate processing in the data receiving entity RE compared to a case in which anomalous data is included in both the input data and the output data.

[0080] (Filtering Method of Likelihood Filter) Here, the filtering method of the likelihood filters LF-1 and LF-2 will be described.

[0081] The filtering method using likelihood used in likelihood filters LF-1 and LF-2 may be determined using a known method. For example, the determination may be made as follows. In either case, likelihood filters LF-1 and LF-2 perform predetermined processing when the input data is equal to or less than a threshold, and output the input data as is when the input data is greater than the threshold.

[0082] That is, it may be determined whether the difference between the data input to the likelihood filters LF-1 and LF-2 and the data input immediately before that data is equal to or less than a threshold value.

[0083] Alternatively, the likelihood filters LF-1 and LF-2 may calculate the average value and deviation of the input data and determine whether the deviation of the input data is equal to or less than a threshold value.

[0084] Furthermore, the likelihood filters LF-1 and LF-2 may calculate the Euclidean distance for each piece of input data and determine whether the Euclidean distance is equal to or smaller than a threshold value.

[0085] Furthermore, the likelihood filters LF-1 and LF-2 may calculate a vector for the input data and determine whether the direction and magnitude of the vector are equal to or smaller than a threshold value.

[0086] Furthermore, the likelihood filters LF-1 and / or LF-2 may obtain filtered data for input data by machine learning as unsupervised learning.

[0087] Furthermore, the likelihood filters LF-1 and LF-2 may calculate the likelihood for the input data using a known calculation formula, such as a probability density function, and determine whether the calculated likelihood is below a threshold value.

[0088] In the following, a method using a calculation formula will be mainly described, but as mentioned above, the present invention is not limited to this.

[0089] The thresholds used in filtering by the likelihood filters LF-1 and LF-2 may be the same for the likelihood filter LF-1 for input data and the likelihood filter LF-2 for output data, or may be different. In the following description, it is assumed that the thresholds used in the likelihood filters LF-1 and LF-2 are the same. The thresholds may be changed depending on the environmental conditions of the UE 100 (for example, the movement speed of the UE 100).

[0090] In addition, the type of filter used in likelihood filters LF-1 and LF-2, the calculation method (or calculation formula), and / or the threshold value may be instructed from outside the data transmitting entity TE (which may be the data receiving entity RE).

[0091] (Operation Scenarios) Four operation scenarios will now be described as specific examples in which the AI / ML block AB is applied to the mobile communication system 1. The following four operation scenarios will be described in order.

[0092] (1.1) First Operation Scenario: Operation Scenario Using Positioning Accuracy Enhancement

[0093] (1.2) Second Operation Scenario: Operation Scenario Using Beam Management

[0094] (1.3) Third Operation Scenario: Operation Scenario with Channel State Information Feedback (CSI Feedback Enhancement)

[0095] (1.4) Fourth Operation Scenario: Operation Scenario Using Measurement Report Model

[0096] (1.1) First Operation Scenario First, the first operation scenario according to the first embodiment will be described.

[0097] FIG. 9 is a diagram illustrating an example of the configuration of a first operation scenario according to the first embodiment.

[0098] In the first operation scenario, UE 100 generates location information based on the angle of arrival (AoA) of a received signal. In the first operation scenario, UE 100 derives a trained model for location information generation unit 131 in AI / ML processing block AP based on the angle of arrival and location data, and obtains location data (i.e., inference result data) from the angle of arrival (i.e., inference data) using the trained model. In the example shown in FIG. 9 , the location data can be calculated by either calculating the location data (output data) using the location information generation unit 131 or calculating the location data (output data) using the AI / ML processing block AP.

[0099] In the first operation scenario shown in FIG. 9, the data transmitting entity TE is the UE100, and the data receiving entity RE is the gNB200. Also, in the first operation scenario shown in FIG. 9, the legacy processing block LP is the location information generation unit 131. The location information generation unit 131 generates location information of the UE100 from the angle of arrival (input data). The location information generation unit 131 outputs the generated location information as location data (output data). Furthermore, in the first operation scenario shown in FIG. 9, the input data is the angle of arrival of the received signal, and the output data is location data. Note that the location data includes the location data output from the location information generation unit 131 and the location data (i.e., inference result data) output from the AI / ML processing block AP.

[0100] As shown in FIG. 9 , UE 100 includes likelihood filters LF-1 and LF-2. The likelihood filter LF-1 for input data is provided between receiver 110 and location information generator 131. The likelihood filter LF-1 for input data is also provided between receiver 110 and AI / ML processing block AP. The likelihood filter LF-1 for input data calculates the likelihood of the angle of arrival (input data). When the likelihood is equal to or less than a threshold, the likelihood filter LF-1 performs predetermined processing, such as discarding the angle of arrival. The likelihood filter LF-1 for input data outputs data of an angle of arrival whose likelihood is greater than the threshold to transmitter 120, AI / ML processing block AP, and location information generator 131.

[0101] The output data likelihood filter LF-2 is provided between the position information generation unit 131 and the AI / ML processing block AP. The output data likelihood filter LF-2 is also provided between the AI / ML processing block AP and the transmission unit 120. The output data likelihood filter LF-2 calculates the likelihood of the position data (output data) output from the position information generation unit 131 and the position data (output data) output from the AI / ML processing block AP. When the likelihood is equal to or less than a threshold, the likelihood filter LF-2 performs predetermined processing, such as discarding the position data. On the other hand, the likelihood filter LF-2 outputs position data whose likelihood is greater than the threshold to the AI / ML processing block and the transmission unit 120.

[0102] The transmitter 120 includes input data (arrival angle data) whose likelihood is greater than a threshold and output data (location data) whose likelihood is greater than a threshold in a transmission signal and transmits it to the gNB 200.

[0103] 9 illustrates an example in which the likelihood filter LF-1 for input data and the likelihood filter LF-2 for output data are included in the UE 100. However, this is not limiting. As described above, either the likelihood filter LF1 or the likelihood filter LF-2 may be included in the UE 100.

[0104] (1.1.1) Example of Operation in First Operation Scenario Next, an example of operation in the first operation scenario according to the first embodiment will be described.

[0105] FIG. 10 is a diagram illustrating an example of an operation of the first operation scenario according to the first embodiment.

[0106] As shown in Figure 10, in step S10, gNB200 transmits a signal (received signal) to UE100.

[0107] In step S11, UE100 decides to transmit data to gNB200. The data to be transmitted is input data and / or output data. UE100 may receive an instruction to transmit data from gNB200. For example, gNB200 may transmit a message (e.g., an RRC message) including instruction information indicating an instruction to transmit data to UE100. UE100 may decide to transmit data in response to receiving the instruction information.

[0108] In step S12, the gNB 200 may configure a likelihood for the UE 100. The configuration may include the type of filter used in the likelihood filters LF-1 and / or LF-2. Alternatively, the configuration may specify an input value for input data input to the likelihood filters LF-1 and / or LF-2. The input value may indicate a range of values ​​that the input data can take. Alternatively, the configuration may specify an output value for output data output from the likelihood filters LF-1 and / or LF-2. The output value may indicate a range of values ​​that the output data can take. Alternatively, if machine learning is used in the likelihood filters LF-1 and / or LF-2, the configuration may include a specific name of the learning model, a type of architecture for the learning model (such as linear regression analysis or DNN (Deep Neural Network)), and / or an application type of the learning model (such as for stationary or mobile). The configuration may include a likelihood threshold. The UE 100 performs a predetermined process on the input data and / or output data based on the threshold. The gNB 200 may perform the setting by transmitting a message (e.g., an RRC message) including the setting. Alternatively, the setting may be hard-coded in the UE 100 in advance.

[0109] In step S13, the UE 100 calculates location data (output data) from the arrival angle (input data) using the location information generation unit 131 or a trained model. When a trained model is used, it is assumed that the AI / ML processing block AP has derived at least a trained model for the location information generation unit 131.

[0110] In step S14, UE 100 calculates the likelihood for the input data and / or output data and checks the likelihood. If the likelihood is equal to or less than a threshold, the input data and / or output data are subjected to a predetermined process (e.g., discarded). If the likelihood is greater than the threshold, the input data and / or output data are output to transmission unit 120. That is, input data likelihood filter LF-1 discards data of arrival angles whose likelihood is equal to or less than the threshold, and outputs data of arrival angles whose likelihood is greater than the threshold to location information generation unit 131, AI / ML processing block AP, and transmission unit 120. For example, when data of arrival angles greater than the arrival angle estimated from the received signal received by UE 100 is input as input data to likelihood filter LF-1, the likelihood becomes equal to or less than the threshold. Furthermore, output data likelihood filter LF-2 discards location data whose likelihood is equal to or less than the threshold, and outputs location data whose likelihood is greater than the threshold to AI / ML processing block AP and transmission unit 120. For example, if the location data indicates an impossible location (for example, if UE 100 is in Japan and the location data indicates New York), the likelihood may be below the threshold.

[0111] In step S15, UE 100 transmits a transmission signal to gNB 200. For example, transmitter 120 transmits a transmission signal including arrival angle data (input data) and / or location data (output data, which is location data output from location information generator 131 or location data output from AI / ML processing block AP) whose likelihood is greater than a threshold.

[0112] (1.1.2) Other Examples in the First Operation Scenario In the first operation scenario, the arrival angle of the received signal at the UE 100 has been described as an example of input data, but this is not limiting. Examples of input data used in the first operation scenario include the following.

[0113] - Reception level for each antenna in UE100, reception phase for each antenna in UE100, or reception time difference for each antenna in UE100 (OTDOA: Observed Time Difference Of Arrival) - Positioning information (positioning information (longitude, latitude, and altitude) obtained from a GNSS (Global Navigation Satellite System) reception signal, positioning information obtained using DL-TDOA (Downlink time difference of arrival), or positioning information obtained using Multi-RTT (Round Trip Time)) - Indicators representing reception signal quality (RSRP (Reference Signal Received Power), RSRQ (Reference Signal Received Quality), SINR (Signal to Interference plus Noise Ratio), output waveform of an AD converter for a reception signal in the receiving unit 110, etc.) - Line-of-sight information for UE100 (line-of-sight information (LOS (Line Of Sight)) and non-line-of-sight information (NLOS (Non Line Of Sight))) - Measurement timing of received signal at UE100 - RF fingerprint (cell ID and reception quality in the cell of the cell ID) - Reception information of beacons used in wireless LAN (Local Area Network) such as Wi-Fi (registered trademark) or short-range wireless communication such as Bluetooth (registered trademark) - Movement speed of UE100 At least two of these input data may be used in appropriate combination.

[0114] In the above-described input data, the positioning information acquired from the GNSS reception signal may be calculated by the GNSS receiver 150 included in the UE 100 based on the GNSS reception signal, as shown in FIG. 9 . In this case, the likelihood filter LF-1 for input data is provided between the GNSS receiver 150 and the position information generation unit 131, and calculates a likelihood for the positioning information, performs predetermined processing according to a threshold, and outputs the positioning information to the position information generation unit 131. Furthermore, in the above-described input data, the movement speed of the UE 100 may also be calculated by the GNSS receiver 150. The movement speed of the UE 100 may be acquired by a speed sensor in the UE 100.

[0115] (1.2) Second Operation Scenario Next, the second operation scenario according to the first embodiment will be described.

[0116] The second operating scenario is an operating scenario using beam management. In the beam management operating scenario, for example, the UE 100 determines an optimal beam based on a CSI reference signal (CSI-RS) transmitted from the gNB 200.

[0117] FIG. 11 is a diagram illustrating an example of the configuration of the second operation scenario according to the first embodiment.

[0118] 11, in the second operation scenario, the legacy processing block LP is an optimal beam determination unit 132. The optimal beam determination unit 132 determines the optimal beam from among multiple beams transmitted from the gNB 200 based on the CSI-RS. The AI / ML processing block AP derives a trained model with the optimal beam determination unit 132 as the target block.

[0119] In the example shown in FIG. 11 , the input data is a CSI reference signal (CSI-RS), and the output data is data representing an optimal beam. Therefore, as shown in FIG. 11 , the likelihood filter LF-1 for input data calculates the likelihood for the CSI-RS. If the likelihood is equal to or less than a threshold, the likelihood filter LF-1 performs predetermined processing, such as discarding the CSI-RS, and if the likelihood is greater than the threshold, the likelihood filter LF-1 outputs the CSI-RS to the optimal beam determination unit 132, the AI / ML processing block AP, and the transmission unit 120. Furthermore, the likelihood filter LF-2 for output data calculates the likelihood for data representing the optimal beam. If the likelihood is equal to or less than a threshold, the likelihood filter LF-2 performs predetermined processing, such as discarding the data representing the optimal beam, and if the likelihood is greater than the threshold, the likelihood filter LF-2 outputs the data representing the optimal beam to the AI / ML processing block AP and the transmission unit 120.

[0120] The transmitter 120 transmits a transmission signal to the gNB 200, the transmission signal including input data (CSI-RS) and output data (optimal beam) having a likelihood greater than a threshold.

[0121] As for the likelihood filters LF-1 and LF-2 shown in FIG. 11, either one of the likelihood filters LF-1 or LF-2 may be included in the UE 100, as in the first operation scenario.

[0122] The operation example in the second operation scenario can also operate in the same manner as the first operation example (FIG. 10). However, in FIG. 10, instead of calculating the location information (step S13), the optimal beam can be determined. In this case, the UE 100 may determine the optimal beam using the optimal beam determination unit 132. The UE 100 may determine the optimal beam using a trained model derived in the AI / ML processing block AP.

[0123] (1.2.1) Other Examples of the Second Operation Scenario In the example shown in Fig. 11, the CSI-RS is used as input data, but this is not limiting. For example, the following may be used as input data.

[0124] - Synchronization signal block (SSB) received from gNB200 - Indicator representing received signal quality (RSRP, RSRQ, SINR, or output waveform of an AD converter, etc.) - Bit error rate (BER) of the received signal or block error rate (BLER) of the received signal - Number of beams (number of beams) or beam pattern transmitted from gNB200 - Measurement values ​​for the beams themselves - Movement speed of UE100 At least two of these input data may be used in appropriate combination.

[0125] (1.3) Third Operation Scenario Next, a third operation scenario according to the first embodiment will be described.

[0126] The third operating scenario is an operating scenario using CSI feedback. The operating scenario using CSI feedback is, for example, an operating scenario in which UE 100 feeds back channel state information (CSI) to gNB 200 based on CSI-RS transmitted from gNB 200.

[0127] FIG. 12 is a diagram illustrating an example of the configuration of the third operation scenario according to the first embodiment.

[0128] As shown in FIG. 12, in the third operation scenario, the legacy processing block LP is the CSI generation unit 133. The CSI generation unit 133 generates CSI indicating the downlink channel state between the UE 100 and the gNB 200 based on the CSI-RS. The CSI includes at least one of a channel quality indicator (CQI), a precoding matrix indicator (PMI), and a rank indicator (RI). The gNB 200 performs downlink scheduling and the like based on the CSI. The AI / ML processing block AP derives a trained model with the CSI generation unit 133 as the target block.

[0129] 12, the input data is a CSI-RS, and the output data is CSI. Therefore, as shown in FIG. 12, the likelihood filter LF-1 for input data calculates the likelihood for the CSI-RS. If the likelihood is equal to or less than a threshold, the likelihood filter LF-1 performs predetermined processing, such as discarding the CSI-RS, and if the likelihood is greater than the threshold, the likelihood filter LF-1 outputs the CSI-RS to the optimal beam determination unit 132, the AI / ML processing block AP, and the transmission unit 120. Furthermore, the likelihood filter LF-2 for output data calculates the likelihood for the CSI. If the likelihood is equal to or less than a threshold, the likelihood filter LF-2 performs predetermined processing, such as discarding the CSI, and if the likelihood is greater than the threshold, the likelihood filter LF-2 outputs the CSI to the AI / ML processing block AP and the transmission unit 120.

[0130] The transmitter 120 transmits a transmission signal including input data (CSI-RS) and output data (CSI) having a likelihood greater than a threshold to the gNB 200. The transmitter 120 may include the input data having a likelihood smaller than the threshold and the output data having a likelihood smaller than the threshold, and information indicating that the likelihood is smaller than the threshold, in the transmission signal and transmit it to the gNB 200. Alternatively, the transmitter 120 may include the associated input data and / or output data and the likelihood in the transmission signal and transmit it to the gNB 200.

[0131] As for the likelihood filters LF-1 and LF-2 shown in FIG. 12, either one of the likelihood filters LF-1 or LF-2 may be included in the UE 100, as in the first operation scenario.

[0132] (1.3.1) Example of Operation in the Third Operation Scenario Next, an example of operation in the third operation scenario according to the first embodiment will be described.

[0133] 13 is a diagram showing an example of operation of the third operation scenario according to the first embodiment. Steps S20 to S22 are the same as steps S10 to S12 of the example of operation of the first operation scenario (FIG. 10), respectively.

[0134] In step S23, the UE 100 calculates the CSI using either the CSI generating unit 133 or the trained model. The trained model is a trained model for the CSI generating unit 133 derived by the AI / ML processing block AP.

[0135] In step S24, the UE 100 checks the likelihood of the CSI-RS and / or the likelihood of the CSI. The UE 100 may discard the CSI-RS whose likelihood is equal to or less than a threshold. The UE 100 may discard the CSI whose likelihood is equal to or less than a threshold. In this case, instead of discarding the CSI-RS, the likelihood filter LF-1 may associate information indicating that the likelihood is equal to or less than a threshold with the CSI-RS. Alternatively, the likelihood filter LF-1 may associate the CSI-RS with the likelihood. The likelihood filter LF-1 may output the associated CSI-RS and the information to the transmitting unit 120. The likelihood filter LF-1 may store the associated CSI-RS and the information in an internal memory of the UE 100 without outputting them. Furthermore, instead of discarding the CSI, the likelihood filter LF-2 may also associate information indicating that the likelihood is equal to or less than a threshold with the CSI. The likelihood filter LF-2 may output the associated CSI and the information to the transmitter 120. Alternatively, the likelihood filter LF-2 may associate the CSI and the likelihood. The likelihood filter LF-2 may output the associated CSI and the likelihood to the transmitter 120. Therefore, the UE 100 may include the associated CSI-RS and the information in a transmission signal and transmit it to the gNB 200. The UE 100 may include the associated CSI and the information in a transmission signal and transmit it to the gNB 200 (step S25).

[0136] In step S25, the UE 100 transmits a transmission signal including CSI-RS and / or CSI having a likelihood greater than the threshold to the gNB 200. As described above, the UE 100 may transmit a transmission signal including CSI-RS and / or CSI having a likelihood smaller than the threshold and information indicating that the likelihood is smaller than the threshold to the gNB 200. Alternatively, as described above, the UE 100 may transmit the linked CSI-RS and / or CSI and the likelihood to the gNB 200 in a transmission signal.

[0137] (1.3.1) Other examples of the third operation scenario In the example shown in Figure 12, an example was described in which CSI-RS is used as input data, but this is not limited to this, and for example, the following may be used as input data.

[0138] - An index representing the received signal quality (RSRP, RSRQ, SINR, or output waveform of an AD converter, etc.) - The bit error rate (BER) of the received signal or the block error rate (BLER) of the received signal - The movement speed of UE100 At least two of these input data may also be used in appropriate combination.

[0139] 12, an example has been described in which the data transmitting entity TE is the UE 100 and the data receiving entity RE is the gNB 200, but the third operation scenario is not limited to this. The data transmitting entity TE may be the gNB 200 and the data receiving entity RE may be the UE 100.

[0140] FIG. 14 is a diagram illustrating another example of the configuration of the third operation scenario according to the first embodiment.

[0141] As shown in Figure 14, the input data is a Sounding Reference Signal (SRS) transmitted from the UE 100. The output data is CSI, and the legacy processing block LP is the CSI generation unit 231, as in the case of Figure 12. The likelihood filters LF-1 and LF-2 are included in the control unit 230 of the gNB 200. Either one of the likelihood filters LF-1 and LF-2 may be included in the control unit 230.

[0142] The operation example shown in Figure 14 can be implemented by replacing UE 100 with gNB 200 in the operation example shown in Figure 13. In this case, the likelihood filter setting in step S22 may be performed by the gNB 200 itself.

[0143] (1.4) Fourth Operation Scenario Next, a fourth operation scenario according to the first embodiment will be described.

[0144] The fourth operating scenario is an operating scenario using a measurement report model.

[0145] The operating scenario according to the measurement report model is an operating scenario in which UE100 transmits a measurement report to gNB200 based on a beam formed by a signal (received signal) transmitted from gNB200.

[0146] (1.4.1) Example of the Configuration of the Measurement Report Model FIG. 15 is a diagram showing an example of the configuration of the fourth operation scenario according to the first embodiment.

[0147] 15 is included in, for example, the control unit 130 of the UE 100. However, a part of the measurement report model ML may be included in the receiving unit 220.

[0148] As shown in FIG. 15 , the measurement reporting model ML includes a Layer 1 (L1) filter 161, a Beam Consolidation / Selection unit 162, Layer 3 (L3) filters (Layer 3 filtering for cell quality, L3 Beam filtering) 163 and 165, an evaluation unit (Evaluation of reporting criteria) 164, and a Beam Selection unit (Beam Selection for reporting) 166.

[0149] The L1 filter 161 receives the beam (measured value) output from the antenna element, smooths the beam, and outputs the smoothed beam to the beam combining / selecting unit 162 and the L3 filter 165. The L1 filter 161 is implementation-dependent.

[0150] The beam combiner / selector 162 combines the smoothed beams (measurements) and outputs the cell quality.

[0151] The L3 filter 163 adds a hysteresis value to the cell quality and outputs the result. The hysteresis value can prevent a phenomenon in which handover is performed multiple times in a short period of time in the UE 100 (a so-called ping-pong phenomenon).

[0152] The evaluation unit 164 evaluates the cell quality output from the L3 filter 163 and determines whether or not the cell quality may be reported as a measurement report. The evaluation unit 164 outputs the measurement report. The measurement report is transmitted to the gNB 200 via the transmission unit 120.

[0153] The L3 filter 165 adds a hysteresis value to the smoothed beam and outputs it.

[0154] The beam selection unit 166 selects a beam (measurement value) to be reported in the measurement report from the beams output from the L3 filter. The beam selection unit 166 includes the selected beam in the measurement report and transmits it to the gNB 200 via the transmission unit 120.

[0155] The settings of the beam combining / selecting unit 162, the L3 filters 163 and 165, the evaluating unit 164, and the beam selecting unit 166 are set by RRC setting parameters.

[0156] In the first embodiment, for each block included in the measurement report model ML configured in this manner, a likelihood is calculated for the input data input to each block and the output data output from that block, and when the likelihood is below a threshold, the input data and / or output data is discarded.

[0157] Specifically, first, the user equipment (e.g., UE 100) calculates the likelihood of input data input to a predetermined block included in a measurement report model (e.g., measurement report model ML) and the likelihood of first output data output from the predetermined block. Second, when the likelihood of the input data and / or the likelihood of the first output data is equal to or less than a first threshold, the user equipment discards the input data and / or the first output data.

[0158] As a result, for example, input data and / or output data whose likelihood is equal to or less than the first threshold is discarded, so that the UE 100 or the gNB 200 can perform processing without using specific data, etc. Therefore, in the mobile communication system 1, it is possible to perform appropriate processing without using specific data.

[0159] The example shown in Fig. 15 represents a case where the L3 filter 163 is the target block. As shown in Fig. 15, a likelihood filter LF-1 for input data is provided before the L3 filter 163, and a likelihood filter LF-2 for output data is provided after the L3 filter. In the example shown in Fig. 15, the input data is cell quality. The output data is cell quality to which a hysteresis value is added.

[0160] First, in the UE 100, a trained model may be derived by machine learning for the likelihood calculated by the likelihood filters LF-1 and LF-2, and the likelihood may be calculated using the derived trained model. In this case, for example, the likelihood filter LF-1 may be used as a legacy processing block LP, and a trained model for the likelihood filter LF-1 may be derived by the AI / ML processing block AP. Then, the trained model may be used to obtain the likelihood as output data. Similarly, a trained model for the likelihood filter LF-2 may be derived by the AI / ML processing block AP, and the likelihood may be obtained from the trained model. Alternatively, the likelihood may be calculated using the known filtering method for likelihood described above, without using machine learning.

[0161] Second, the UE 100 may derive a learned model using each block (i.e., legacy processing block LP) included in the measurement report model ML as a target block. Then, the UE 100 may use either the block or the learned block to obtain output data for input data, and calculate likelihood for the input data and / or output data.

[0162] Specifically, first, the user equipment (e.g., UE 100) derives a learned model for a predetermined block based on input data (e.g., data input to L3 filter 163) and first output data (e.g., data output from L3 filter 163). Second, the user equipment either calculates first output data for the input data using the predetermined block, or calculates second output data for the input data (e.g., data output from the learned model of the L3 filter) using the learned model. Third, the user equipment discards the input data and / or the second output data when the likelihood of the input data and / or the likelihood of the second output data is equal to or less than a second threshold.

[0163] This means that, for example, a trained model will no longer be derived using peculiar data whose likelihood is below the second threshold, and compared to deriving a trained model using peculiar data, it becomes possible to perform machine learning using an appropriate trained model, thereby improving learning accuracy.

[0164] Fig. 16 is a diagram illustrating an example of the configuration of a measurement report model ML when deriving a trained model for the L3 filter 163. As shown in Fig. 16, the AI / ML processing block AP derives a trained model with the L3 filter 163 as a target block from the cell quality (input data) input to the L3 filter 163 and the smoothed cell quality (output data) output from the L3 filter 163.

[0165] In the measurement reporting model ML, the legacy processing block LP is each block of the measurement reporting model ML. In the example of Fig. 16, the L3 filter 163 is the legacy processing block LP. Furthermore, the input data is data input to each block of the measurement reporting model (or the trained model of each block). In the example of Fig. 16, the input data is cell quality. Furthermore, the output data is data output from each block of the measurement reporting model (and the trained model of each block). In the example of Fig. 16, the output data is smoothed cell quality.

[0166] 16, the likelihood filter LF-1 for input data calculates the likelihood of the cell quality (input data), discards the cell quality when the likelihood is equal to or less than a threshold (e.g., a first threshold), and outputs the cell quality to the L3 filter 163 when the likelihood is greater than the threshold. The likelihood filter LF-2 for output data calculates the likelihood of the smoothed cell quality (output data), discards the smoothed cell quality when the likelihood is equal to or less than a threshold, and outputs the cell quality to the evaluation unit 164 or the transmission unit 120 when the likelihood is greater than the threshold.

[0167] In this way, the likelihood filters LF-1 and LF-2 discard data whose likelihood is equal to or less than a threshold, and therefore the AI / ML processing block AP can derive a trained model with higher training accuracy than when idiosyncratic data is included. This makes it possible to improve the training accuracy in the mobile communication system 1.

[0168] Note that one of the likelihood filters LF-1 and LF-2 may be included in the measurement report model ML, rather than both.

[0169] (1.4.2) Example of Operation in Fourth Operation Scenario Next, an example of operation in the fourth operation scenario according to the first embodiment will be described.

[0170] FIG. 17 is a diagram illustrating an example of an operation of the fourth operation scenario according to the first embodiment.

[0171] Steps S30 to S32 are the same as steps S10 to S12 of the operational example in the first operating scenario. However, in step S30, the gNB200 may transmit a transmission signal (beam) by forming a beam using SSB or CSI-RS. Also, in step S32, the gNB200 may specify input values ​​for each block of the measurement reporting model ML (L1 filter 161, beam combining / selecting unit 162, L3 filters 163 and 165, evaluation unit 164, and / or beam selecting unit 166). In step S32, the gNB200 may specify output values ​​for each block of the measurement reporting model ML.

[0172] In step S33, the UE 100 calculates a measurement report. The UE 100 may calculate the measurement report using each block of the measurement report model ML. The UE 100 may calculate the measurement report using a learned model targeted at each block of the measurement report model ML.

[0173] In step S34, the UE 100 checks the likelihood. For example, the UE 100 calculates likelihoods in a likelihood filter LF-1 for input data provided in the front stage of each block of the measurement report model ML and a likelihood filter LF-2 for output data provided in the rear stage of each block of the measurement report model ML, and checks the likelihoods. The likelihood filters LF-1 and LF-2 discard input data and output data whose likelihoods are equal to or less than a threshold, and output input data and output data whose likelihoods are greater than the threshold.

[0174] In step S35, the UE 100 transmits a transmission signal. In this case, the transmission signal includes input data and / or output data whose likelihood is greater than the threshold, similarly to the first operation scenario.

[0175] (1.4.3) Other Examples of the Fourth Operation Scenario When deriving a learned model for each block of the measurement report model ML, the AI / ML processing block AP may be configured to input data that is unrelated to the input data input to each block. Examples of the unrelated data include location data of the UE 100 or movement speed data of the UE 100. The AI / ML processing block AP may acquire the location data or movement speed data from the GNSS receiver 150.

[0176] Second Embodiment Next, a second embodiment will be described, focusing mainly on the differences from the first embodiment.

[0177] In the second embodiment, in the data transmitting entity TE, when the likelihood of input data and / or output data is below a threshold, the calculation of output data is switched from a legacy processing block LP to a trained model, or from the trained model to a legacy processing block LP.

[0178] Specifically, first, when a data transmitting entity (e.g., a data transmitting entity TE) is calculating output data using a specified block (e.g., a legacy processing block LP), if the likelihood of the input data and / or the likelihood of the output data is less than or equal to a first threshold, it switches to calculating the output data using a trained model, and when the data transmitting entity is calculating output data using a trained model, if the likelihood of the input data and / or the likelihood of the output data is less than or equal to a second threshold, it switches to calculating the output data using a specified block.

[0179] In this way, the data transmitting entity TE switches the entity that calculates the output data when the likelihood of the input data and / or output data is equal to or less than the threshold, so that it is possible to prevent output data with a likelihood equal to or less than the threshold from being transmitted to the data receiving entity RE, and therefore the data receiving entity RE can perform appropriate processing without using peculiar data with a likelihood equal to or less than the threshold.

[0180] Note that whether the calculation entity before switching is the legacy processing block LP or the learned model can be set, for example, as follows.

[0181] First, the data receiving entity RE may instruct or configure the data transmitting entity TE to be the calculation subject. For example, the data receiving entity RE may instruct or configure the data transmitting entity TE by transmitting a message (e.g., an RRC message) including information (instruction information or configuration information) indicating whether the output data is to be calculated by the legacy processing block LP or the trained model.

[0182] Second, the data transmitting entity TE may select whether to use the legacy processing block LP or the trained model based on the likelihood of the input and / or output data.

[0183] Third, the data transmitting entity TE may select whether to use the legacy processing block LP or the trained model based on the power consumption, for example, the data transmitting entity TE may select the one with the lower power consumption for the first calculation.

[0184] Fourth, the calculation subject before switching may be hard-coded in advance in the data transmitting entity TE.

[0185] The second embodiment can be applied to each of the above-mentioned operation scenarios. For example, the operation scenario for improving position accuracy (first operation scenario) can be implemented using the configuration example of Fig. 9. Furthermore, for example, the operation scenario for beam management (second operation scenario) can be implemented using the configuration example of Fig. 11. Furthermore, for example, the operation scenario for CSI feedback (third operation scenario) can be implemented using Figs. 12 and 14. Furthermore, for example, the operation scenario for measurement and reporting model ML (fourth operation scenario) can be implemented using the configuration examples of Figs. 15 and 16.

[0186] (Example of Operation According to Second Embodiment) Next, an example of operation according to the second embodiment will be described.

[0187] 18 is a diagram illustrating an example of operation according to the second embodiment. FIG. 18 illustrates an example of operation in the case where the data transmitting entity TE is the UE 100 and the data receiving entity RE is the gNB 200.

[0188] Steps S40 to S42 shown in FIG. 18 are the same as steps S10 to S12 in the operation example (FIG. 10) of the first operation scenario according to the first embodiment.

[0189] In step S43, the UE 100 calculates output data. The UE 100 calculates the output data using either the legacy processing block LP or the learned model. In the case of an operation scenario for improving location accuracy (first operation scenario), the legacy processing block LP is the location information generation unit 131, and the output data is location data. In the case of an operation scenario for beam management (second operation scenario), the legacy processing block LP is the optimal beam determination unit 132, and the output data is data representing the optimal beam. In the case of an operation scenario for CSI feedback (third operation scenario), the legacy processing block LP is the CSI generation unit 133 or 231, and the output data is CSI. In the case of an operation scenario for the measurement reporting model ML (fourth operation scenario), the legacy processing block LP is each block of the measurement reporting model ML, and the output data is the output data of each block.

[0190] In step S44, the UE 100 checks the likelihood of the input data and / or output data. The likelihood filter LF-1 for input data and / or the likelihood filter LF-2 for output data calculates and checks the likelihood. Then, when the likelihood is equal to or less than a threshold, the UE 100 switches the calculation entity of the output data. For example, when the UE 100 is calculating output data for input data using the legacy processing block LP, if the likelihood of the input data and / or output data is equal to or less than a threshold (e.g., a first threshold), the UE 100 switches to calculating output data using a trained model. In this case, the UE 100 can be considered to switch to calculating output data using a trained model, assuming that the learning accuracy of the trained model is higher than in a constant case. Also, for example, when the UE 100 is calculating output data for input data using the trained model, if the likelihood of the input data and / or output data is equal to or less than a threshold (e.g., a second threshold), the UE 100 switches to calculating output data using the legacy processing block LP. In this case, the UE 100 can be considered to switch to the legacy processing block LP, assuming that the learning accuracy of the learned model is lower than in a constant case.

[0191] In addition, when UE100 calculates output data using either the legacy processing block LP or the learned model, if the likelihood of the input data and / or output data is greater than a threshold (first threshold or second threshold), it continues as is without switching the entity that calculates the output data.

[0192] In step S45, UE100 transmits a transmission signal including input data and output data to gNB200.

[0193] (Another Example of the Second Embodiment) In the second embodiment, an example has been described in which the UE 100 obtains output data by switching the calculation subject of the output data (step S45). For example, the UE 100 may calculate the output data using both the legacy processing block LP and the learned model. In this case, the UE 100 may compare the likelihoods of the output data from the legacy processing block LP and the output data from the learned model, and transmit the output data having the higher likelihood to the gNB 200 (step S45). Alternatively, the UE 100 may calculate the output data using both the legacy processing block LP and the learned model, and transmit two output data to the gNB 200. In this case, the UE 100 may calculate the likelihood of each of the two output data, associate the calculated likelihood with each output data, and transmit each output data and each likelihood to the gNB 200 (step S45).

[0194] In the second embodiment, the data transmitting entity TE is the UE 100, and the data receiving entity RE is the gNB 200. For example, as described above, the second embodiment can also be applied to the CSI feedback operation scenario (third operation scenario), and can be implemented even if the gNB 200 is the data transmitting entity TE and the UE 100 is the data receiving entity RE (FIG. 14). In the operation example shown in FIG. 18, by replacing the UE 100 with the gNB 200 and the gNB 200 with the UE 100, an operation example in which the gNB 200 is the data transmitting entity TE and the UE 100 is the data receiving entity RE can also be implemented.

[0195] Third Embodiment Next, a third embodiment will be described, focusing mainly on the differences from the first embodiment.

[0196] In the third embodiment, the data receiving entity RE calculates the likelihood. Specifically, first, a data transmitting entity (e.g., a data transmitting entity TE) derives a learned model for a predetermined block (e.g., a legacy processing block LP) based on a signal received from the data receiving entity (e.g., the data receiving entity RE). Second, the data transmitting entity calculates output data for input data using either the predetermined block or the learned model. Third, the data transmitting entity transmits the input data and output data to the data receiving entity. Fourth, the data receiving entity calculates the likelihood of the input data and / or the likelihood of the output data.

[0197] In this way, since the likelihood is calculated in the data receiving entity RE, the data receiving entity RE can discard the input data and output data received from the data transmitting entity TE based on the calculated likelihood. Therefore, the data receiving entity RE can process the input data and output data without using peculiar data whose likelihood is equal to or less than a threshold. Therefore, it is possible to perform appropriate processing in the mobile communication system 1.

[0198] (Configuration Example According to Third Embodiment) Next, a configuration example according to the third embodiment will be described.

[0199] FIG. 19 is a diagram illustrating an example of a configuration according to the third embodiment.

[0200] As shown in FIG. 19, the data receiving entity RE further comprises a likelihood filter LF-3.

[0201] First, the likelihood filter LF-3 may calculate the likelihood for the input data and output data output from the receiving unit RE-2, and discard the input data and output data when the likelihood is below a threshold.

[0202] Second, when the likelihood is equal to or less than a threshold, the likelihood filter LF-3 may generate instruction information instructing the transmitter RE-1 to switch between calculating output data using the legacy processing block LP and calculating output data using the trained model. The instruction information is output to the transmitter RE-1 via the control unit RE-3. The transmitter RE-1 transmits a message including the instruction information to the data transmitting entity TE. The message may be transmitted as an RRC message or the like. Note that when the likelihood filter LF-3 receives both output data from the legacy processing block LP and output data from the trained model as output data from the data transmitting entity TE, the likelihood filter LF-3 may select either the output data from the legacy processing block LP or the output data from the trained model according to the likelihood. The likelihood filter LF-3 may output information indicating which output data has been selected to the transmitter RE-1. The transmitter RE-1 may transmit a message (e.g., an RRC message) including the information to the data transmitting entity TE.

[0203] Third, the likelihood filter LF-3 may output the calculated likelihood as likelihood information to the transmitter RE-1 via the controller RE-3. The transmitter RE-1 may transmit a message (e.g., an RRC message) including the likelihood information to the data transmitting entity TE. The data transmitting entity TE may switch the entity that calculates the output data (the legacy processing block LP or the trained model) based on the likelihood information.

[0204] The likelihood filter LF-3 may check the likelihood of either the input data or the output data.

[0205] The first to fourth operation scenarios described in the first embodiment can also be applied to the third embodiment (FIGS. 9, 11, 12, 14, 15, and 16).

[0206] (Example of Operation According to Third Embodiment) Next, an example of operation according to the third embodiment will be described.

[0207] Fig. 20 is a diagram illustrating an example of operation according to the third embodiment. In Fig. 20, an example in which the data transmitting entity TE is the UE 100 and the data receiving entity RE is the gNB 200 will be described.

[0208] In FIG. 20, steps S50 and S51 are the same as steps S10 and S11, respectively, in the operation example according to the first embodiment (FIG. 10).

[0209] In step S53, the UE 100 calculates output data for the input data using either the legacy processing block LP or the learned model. As in the first embodiment, the legacy processing block LP, the input data, and the output data differ depending on each applied operation scenario.

[0210] In step S54, UE100 transmits a transmission signal including input data and output data to gNB200.

[0211] In step S55, the gNB200 checks the likelihood of the input data and / or output data. The gNB200 calculates the likelihood of the input data and / or output data, and when the likelihood is greater than a threshold, performs processing using the input data and / or output data. On the other hand, when the likelihood of the input data and / or output data is equal to or less than a threshold, the gNB200 may discard the input data and / or output data. Alternatively, when the likelihood of the input data and / or output data is equal to or less than a threshold, the gNB200 may instruct the data transmission entity TE to switch the calculation entity of the output data. Regarding the switching, if the UE100 has calculated the output data using the legacy processing block LP, the gNB200 instructs the UE100 to switch to the learned model. On the other hand, if the UE100 has calculated the output data using the learned model, the gNB200 instructs the UE100 to switch to the legacy processing block LP. The gNB 200 may transmit a message (e.g., an RRC message) including instruction information instructing the UE 100 to switch. When the gNB 200 (data receiving entity RE) receives both output data from the legacy processing block LP and output data from the learned model as output data, the gNB 200 may select either the output data from the legacy processing block LP or the output data from the learned model according to the likelihood.

[0212] In step S56, when gNB200 calculates the likelihood of the input data and / or output data, it may transmit the likelihood as likelihood information to UE100. gNB200 may transmit a message (e.g., an RRC message) including the likelihood information to UE100.

[0213] In step S57, the UE 100 may switch the entity that calculates the output data to either the legacy processing block LP or the trained model based on the likelihood information. Regarding the switching, if the UE 100 has calculated the output data using the legacy processing block LP, the UE 100 switches to the trained model, and if the UE 100 has calculated the output data using the trained model, the UE 100 switches to the legacy processing block LP.

[0214] (Another example of the third embodiment) In the third embodiment, as in the second embodiment, in the operation scenario (third operation scenario) of CSI feedback, the data transmitting entity TE may be the gNB 200 and the data receiving entity RE may be the UE 100. In this case, in the operation example shown in FIG. 20 , it is possible to implement by replacing UE 100 with gNB 200 and gNB 200 with UE 100, respectively.

[0215] Fourth Embodiment Next, a fourth embodiment will be described. The fourth embodiment will also be described mainly focusing on the differences from the first embodiment.

[0216] In the fourth embodiment, a data receiving entity RE transmits configuration information related to the likelihood to a data transmitting entity TE. Specifically, the data receiving entity (e.g., the data receiving entity RE) transmits a message including a method for calculating the likelihood to a data transmitting entity (e.g., the data transmitting entity TE).

[0217] As a result, for example, the data transmitting entity TE can calculate the likelihood according to the likelihood calculation method instructed by the data receiving entity RE, and therefore can also prevent data with a likelihood equal to or less than a threshold from being transmitted to the data receiving entity RE. Therefore, the data receiving entity RE does not perform processing using anomalous data, and can perform appropriate processing.

[0218] In the fourth embodiment, first, a data receiving entity (e.g., a data receiving entity RE) transmits a message including a minimum likelihood value to a data transmitting entity (e.g., a data transmitting entity TE), and second, the data transmitting entity transmits input data and / or output data whose likelihood of the input data and / or the output data is equal to or greater than the minimum value to the data receiving entity.

[0219] As a result, for example, the data receiving entity RE can receive data whose likelihood is equal to or greater than the minimum value, and therefore does not receive anomalous data whose likelihood is less than the minimum value, and therefore does not perform processing using anomalous data. Thus, the data receiving entity RE can perform appropriate processing.

[0220] In addition, the first to fourth operation scenarios described in the first embodiment can also be applied to the fourth embodiment (FIGS. 9, 11, 12, 14, 15, and 16).

[0221] (Example of Operation According to Fourth Embodiment) Next, an example of operation according to the fourth embodiment will be described.

[0222] Fig. 21 is a diagram illustrating an example of operation according to the fourth embodiment. In the example illustrated in Fig. 21, the UE 100 is a data transmitting entity TE, and the gNB 200 is a data receiving entity RE.

[0223] In FIG. 21, steps S60 and S61 are the same as steps S10 and S11, respectively, in the operation example according to the first embodiment (FIG. 10).

[0224] In step S62, the gNB 200 may transmit a likelihood calculation method to the UE 100. The likelihood calculation method may be the filtering method of the likelihood filters LF-1 and LF-2 described in the first embodiment. The likelihood calculation method may indicate what calculation formula is used to calculate the likelihood. The gNB 200 may transmit information indicating the calculation method to each UE 100 by individual signaling of an RRC message. The gNB 200 may transmit information indicating the calculation method to multiple UEs 100 by broadcast signaling of an RRC message (for example, a system information block (SIB)).

[0225] In step S63, the gNB 200 transmits the lowest likelihood value to the UE 100. The gNB 200 may transmit the lowest value to each UE 100 using individual signaling of an RRC message. The gNB 200 may also transmit the lowest value to multiple UEs 100 using broadcast signaling of an RRC message (e.g., SIB).

[0226] In step S64, the UE 100 calculates output data for the input data using either the legacy processing block LP or the learned model. As in the first embodiment, the legacy processing block LP, the input data, and the output data differ depending on each applied operation scenario.

[0227] In step S65, the UE 100 calculates a likelihood for the input data and / or the output data, and confirms the likelihood.

[0228] In step S66, if the likelihood is equal to or greater than the minimum value, the UE 100 transmits the input data and / or output data having the likelihood in a transmission signal. On the other hand, if the likelihood is less than the minimum value, the UE 100 does not transmit the input data and / or output data having the likelihood to the gNB 200.

[0229] (Another example of the fourth embodiment) In the fourth embodiment, as in the second embodiment, in the operation scenario (third operation scenario) of CSI feedback, the data transmitting entity TE may be the gNB 200 and the data receiving entity RE may be the UE 100. In this case, in the operation example shown in FIG. 21 , it is possible to implement by replacing UE 100 with gNB 200 and gNB 200 with UE 100, respectively.

[0230] Fifth Embodiment Next, a fifth embodiment will be described.

[0231] The fifth embodiment is an embodiment in which UE 100-1 links input data and output data to likelihood, and UE 100-2 determines whether or not to use the input data and / or output data as learning data based on the linked likelihood.

[0232] Specifically, a first user apparatus (e.g., UE100-1) derives a learned model for a predetermined block (e.g., legacy processing block LP) based on a signal received from a base station (e.g., gNB200). Second, the first user apparatus calculates output data for input data using the predetermined block and / or the learned model. Third, the first user apparatus calculates the likelihood of the input data and / or the likelihood of the output data. Fourth, the first user apparatus associates the input data and / or the output data with the likelihood and transmits the input data and / or the output data and the likelihood to the base station. Fifth, the base station transmits the input data and / or the output data and the likelihood to a second user apparatus. Sixth, the second user apparatus determines whether to use the input data and / or the output data as training data based on the likelihood.

[0233] As a result, for example, even in the UE 100-2 that cannot calculate the likelihood, it is possible to determine whether or not to use the input data and / or output data as learning data based on the likelihood. Therefore, in the UE 100-2, by appropriately using the input data and / or output data used in the UE 100-1 as learning data, it is possible to improve the learning accuracy. In addition, in the UE 100-2, by not using input data and / or output data whose likelihood is below a threshold, it is also possible to perform appropriate processing.

[0234] In the fifth embodiment, the UE 100-1 may associate information indicating that the likelihood is equal to or greater than a threshold with input data and / or output data whose likelihood is equal to or greater than a threshold. By such an association process, the UE 100-2 that has received the input data and / or the output data can determine, based on the information, to use the input data and / or the output data as learning data.

[0235] In the fifth embodiment, among the first to fourth operating scenarios described in the first embodiment, the operating scenarios in which the data transmitting entity TE is UE100 and the data receiving entity RE is gNB200 are applicable (Figures 9, 11, 12, 15, and 16).

[0236] (Example of Operation According to Fifth Embodiment) Next, an example of operation according to the fifth embodiment will be described.

[0237] FIG. 22 is a diagram illustrating an example of operation according to the fifth embodiment.

[0238] In FIG. 22, steps S70 and S71 are the same as steps S10 and S11, respectively, in the operation example according to the first embodiment (FIG. 10).

[0239] In step S72, the gNB 200 may transmit to the UE 100 linking instruction information that instructs linking of input data and output data with likelihood. The linking instruction information may be information that instructs linking of information indicating that the likelihood is equal to or greater than a threshold, for input data and output data whose likelihood is equal to or greater than a threshold. The gNB 200 may transmit the linking instruction information to each UE 100 by individual signaling of an RRC message. The gNB 200 may transmit the linking instruction information to multiple UEs 100 by broadcast signaling (e.g., SIB) of an RRC message.

[0240] In step S73, the UE 100 calculates output data for the input data using either the legacy processing block LP or the learned model. As in the first embodiment, the legacy processing block LP, the input data, and the output data differ depending on each applied operation scenario.

[0241] In step S74, the UE 100 calculates the likelihood for the input data and / or the output data.

[0242] In step S75, the UE 100 associates the input data and / or output data with the likelihood. The UE 100 may perform the association in accordance with the association instruction information (step S72). Alternatively, the UE 100 may associate information indicating that the likelihood is equal to or greater than a threshold with the input data and output data whose likelihood is equal to or greater than a threshold. The association may be performed in accordance with the association instruction information (step S72).

[0243] In step S76, the UE 100 transmits the input data and / or output data to be linked and the likelihood in a transmission signal to the gNB 200. As in the fourth embodiment, the UE 100 may transmit input data and / or output data having a likelihood equal to or greater than the minimum value.

[0244] In step S77, gNB200 includes the input data and / or output data and likelihood received from UE100-1 in a transmission signal and transmits it to UE100-2.

[0245] In step S78, the UE 100-2 checks the associated likelihood. When the likelihood is greater than a threshold, the UE 100-2 uses the input data and / or output data having the likelihood as learning data. On the other hand, when the likelihood is equal to or less than the threshold, the UE 100 does not use the input data and / or output data having the likelihood as learning data.

[0246] [Other Embodiments] In the first embodiment described above, an example has been described in which the mobile communication system 1 includes a data transmitting entity TE and a data receiving entity RE. For example, the mobile communication system 1 may include a dedicated center 500. FIG. 23 is a diagram illustrating an example configuration of the mobile communication system 1 including the dedicated center 500. The dedicated center 500 is, for example, a device that exclusively calculates likelihoods when calculating likelihoods in likelihood filters LF-1 and LF-2. As shown in FIG. 23, the UE 100 transmits a likelihood calculation instruction to the dedicated center 500 via the gNB 200. The likelihood calculation instruction may include input data and output data to be used for likelihood calculation. The input data and output data may be transmitted separately from the likelihood calculation instruction. The dedicated center 500 calculates the likelihood in accordance with the likelihood calculation instruction. The dedicated center 500 transmits the calculated likelihood as likelihood data to the UE 100 via the gNB 200. 23 shows an example in which likelihood is calculated in likelihood filters LF-1 and LF-2 of UE 100, but it is also applicable, for example, to a case in which likelihood is calculated by likelihood filter LF-3 of gNB 200. In this case, gNB 200 transmits a likelihood calculation instruction to dedicated center 500 and receives likelihood data from dedicated center 500.

[0247] In the above-described embodiments, supervised learning has been mainly described, but the present invention is not limited to this. For example, the first to fifth embodiments may be applied to unsupervised learning or reinforcement learning.

[0248] Furthermore, a program (information processing program) that causes a computer to execute each process or function according to each of the above-described embodiments may be provided. Alternatively, a program (e.g., a mobile communication program) that causes the mobile communication system 1 to execute each process or function according to the above-described embodiments may be provided. The program may be recorded on a computer-readable medium. Using a computer-readable medium, 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-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. Such a recording medium may be memory included in the UE 100 and the gNB 200.

[0249] As used in this disclosure, the terms "based on" and "depending on" 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." Furthermore, 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. Furthermore, as used in this disclosure, the term "or" 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.

[0250] Although the embodiments have been described in detail above with reference to the drawings, the specific configuration is not limited to the above, and various design changes can be made without departing from the spirit of the invention. Furthermore, it is also possible to combine the embodiments, operation examples, or processes within a consistent range.

[0251] This application claims priority from Japanese Patent Application No. 2022-139680 (filed September 2, 2022), the entire contents of which are incorporated herein by reference.

[0252] (Supplementary Note) (Supplementary Note 1) A communication method in a mobile communication system, comprising: a step in which a data transmitting entity derives a learned model for a predetermined block based on a signal received from a data receiving entity; a step in which the data transmitting entity calculates output data for input data using either the predetermined block or the learned model; and a step in which the data transmitting entity performs predetermined processing when the likelihood of the input data and / or the likelihood of the output data is below a threshold.

[0253] (Supplementary Note 2) The communication method according to Supplementary Note 1, wherein the predetermined processing is one of: a processing by the data transmission entity of discarding the input data and / or the output data; a processing by the data transmission entity of associating the input data and / or the output data with information indicating that the likelihood of the input data and / or the likelihood of the output data is equal to or less than the threshold; and a processing of associating likelihood with the input data and / or the output data.

[0254] (Supplementary Note 3) The communication method according to Supplementary Note 1 or Supplementary Note 2, wherein the data receiving entity is a base station and the data transmitting entity is a user equipment, and the step of performing the predetermined processing includes a step in which the base station sets the likelihood for the user equipment.

[0255] (Supplementary Note 4) The threshold is a first threshold or a second threshold, and the predetermined processing is a processing of: when the data transmitting entity is calculating the output data using the predetermined block, switching to calculating the output data using the trained model when the likelihood of the input data and / or the likelihood of the output data is equal to or less than the first threshold; and when the data transmitting entity is calculating the output data using the trained model, switching to calculating the output data using the predetermined block when the likelihood of the input data and / or the likelihood of the output data is equal to or less than the second threshold. A communication method described in any of Supplementary Note 1 to Supplementary Note 3.

[0256] (Supplementary Note 5) The communication method according to any one of Supplementary Notes 1 to 4, further comprising the step of the data receiving entity transmitting a message including a method for calculating the likelihood to the data transmitting entity.

[0257] (Supplementary Note 6) A communication method as described in any of Supplementary Notes 1 to 5, further comprising a step in which the data receiving entity transmits a message including a minimum likelihood value to the data transmitting entity, and the predetermined processing is a processing in which the data transmitting entity transmits to the data receiving entity the input data and / or the output data whose likelihood of the input data and / or the output data is equal to or greater than the minimum value.

[0258] (Supplementary Note 7) A communication method in a mobile communication system, comprising: a step in which a data transmitting entity derives a learned model for a predetermined block based on a signal received from a data receiving entity; a step in which the data transmitting entity calculates output data for input data using either the predetermined block or the learned model; a step in which the data transmitting entity transmits the input data and the output data to the data receiving entity; and a step in which the data receiving entity calculates the likelihood of the input data and / or the likelihood of the output data.

[0259] (Supplementary Note 8) The communication method according to Supplementary Note 7, further comprising the step of the data receiving entity discarding the input data and / or the output data when the likelihood of the input data and / or the likelihood of the output data is equal to or less than a threshold.

[0260] (Supplementary Note 9) The communication method described in Supplementary Note 7 or Supplementary Note 8 further comprises a step in which the data receiving entity transmits to the data transmitting entity a message including instruction information instructing the data receiving entity to switch to either calculating the output data using the specified block or calculating the output data using the trained model when the likelihood of the input data and / or the likelihood of the output data is equal to or lower than a threshold.

[0261] (Supplementary Note 10) The communication method according to any one of Supplementary Notes 7 to 9, further comprising the step of the data receiving entity transmitting a message including likelihood information representing the likelihood to the data transmitting entity.

[0262] (Supplementary Note 11) A communication method in a mobile communication system, comprising: a step by a first user device deriving a trained model for a predetermined block based on a signal received from a base station; a step by the first user device calculating output data for input data using the predetermined block and / or the trained model; a step by the first user device calculating a likelihood of the input data and / or the likelihood of the output data; a step by the first user device linking the input data and / or the output data with the likelihood and transmitting the input data and / or the output data and the likelihood to the base station; a step by the base station transmitting the input data and / or the output data and the likelihood to a second user device; and a step by the second user device deciding whether to use the input data and / or the output data as training data based on the likelihood.

[0263] (Supplementary Note 12) The communication method according to Supplementary Note 11, further comprising a step in which the base station transmits to the first user device a message including linking instruction information instructing the first user device to link the input data and / or the output data with the likelihood, wherein the step of transmitting to the base station includes a step in which the first user device links the input data and / or the output data with the likelihood in accordance with the instruction information.

[0264] (Supplementary Note 13) A communication method in a user equipment that generates a measurement report using a measurement report model, comprising: a step in which the user equipment calculates a likelihood of input data input to a predetermined block included in the measurement report model and a likelihood of first output data output from the predetermined block; and a step in which the user equipment discards the input data and / or the first output data when the likelihood of the input data and / or the likelihood of the first output data is equal to or less than a first threshold.

[0265] (Supplementary Note 14) The communication method according to Supplementary Note 13, further comprising a step in which the user device derives a trained model for the specified block based on the input data and the first output data, wherein the calculating step includes either a step in which the user device calculates the first output data for the input data using the specified block, or a step in which the user device calculates second output data for the input data using the trained model, and wherein the discarding step includes a step in which the user device discards the input data and / or the second output data when the likelihood of the input data and / or the likelihood of the second output data is equal to or less than a second threshold.

[0266] 1: Mobile communication system 100 (100-1, 100-1): UE 110: Receiving unit 120: Transmitting unit 130: Control unit 131: Position information generating unit 132: Optimal beam determining unit 133: CSI generating unit 150: GNSS receiver 200: gNB 210: Receiving unit 220: Transmitting unit 230: Control unit 231: CSI generating unit A1: Data collecting unit A2: Model learning unit A3: Model inference unit A4: Data processing unit TE: Data transmitting entity RE: Data receiving entity

Claims

1. A communication method in a mobile communication system, comprising: A data transmitting entity deriving a trained model for a given block based on a signal received from the data receiving entity; The data transmitting entity calculates output data for input data using either the predetermined block or the trained model; and performing a predetermined process when the likelihood of the input data and / or the likelihood of the output data is equal to or less than a threshold value. Communication methods.

2. The predetermined process is the data transmitting entity discarding the input data and / or the output data; A process in which the data transmitting entity associates, with the input data and / or the output data, information indicating that the likelihood of the input data and / or the likelihood of the output data is equal to or less than the threshold; and A process of associating a likelihood with the input data and / or the output data. The communication method according to claim 1.

3. The data receiving entity is a base station and the data transmitting entity is a user equipment, The step of performing the predetermined process includes the base station setting the likelihood for the user equipment. The communication method according to claim 1.

4. The threshold value is a first threshold value or a second threshold value, The predetermined process is When the data transmitting entity is calculating the output data using the predetermined block, if the likelihood of the input data and / or the likelihood of the output data is equal to or less than the first threshold, switching to calculation of the output data using the trained model; When the data transmitting entity calculates the output data using the trained model, if the likelihood of the input data and / or the likelihood of the output data is equal to or less than the second threshold, the data transmitting entity switches to calculation of the output data using the predetermined block. The communication method according to claim 1.

5. The method further comprises the data receiving entity transmitting a message including the method for calculating the likelihood to the data transmitting entity. The communication method according to claim 1 .

6. The method further includes the data receiving entity transmitting a message to the data transmitting entity, the message including the lowest likelihood value; The predetermined process is a process in which the data transmitting entity transmits, to the data receiving entity, the input data and / or the output data, the likelihood of which is equal to or greater than the minimum value. The communication method according to claim 1.

7. A communication method in a mobile communication system, comprising: A data transmitting entity deriving a trained model for a given block based on a signal received from the data receiving entity; The data transmitting entity calculates output data for input data using either the predetermined block or the trained model; the data transmitting entity transmitting the input data and the output data to the data receiving entity; the data receiving entity calculating the likelihood of the input data and / or the likelihood of the output data. Communication methods.

8. The data receiving entity may further comprise discarding the input data and / or the output data when the likelihood of the input data and / or the likelihood of the output data is below a threshold. The communication method according to claim 7.

9. The data receiving entity may further transmit, to the data transmitting entity, a message including instruction information instructing the data receiving entity to switch to either calculation of the output data using the predetermined block or calculation of the output data using the trained model when the likelihood of the input data and / or the likelihood of the output data is equal to or lower than a threshold. The communication method according to claim 7.

10. The data receiving entity may further include transmitting a message including likelihood information representing the likelihood to the data transmitting entity. The communication method according to claim 7.

11. A communication method in a mobile communication system, comprising: A first user device derives a trained model for a predetermined block based on a signal received from a base station; The first user device calculates output data for input data using the predetermined block and / or the trained model; the first user device calculating a likelihood of the input data and / or a likelihood of the output data; The first user device associates the input data and / or the output data with the likelihood, and transmits the input data and / or the output data and the likelihood to the base station; the base station transmitting the input data and / or the output data and the likelihood to a second user equipment; and determining whether or not the second user device uses the input data and / or the output data as training data based on the likelihood. Communication methods.

12. The base station further transmits, to the first user device, a message including binding instruction information instructing binding of the input data and / or the output data to the likelihood; The transmitting to the base station includes associating the input data and / or the output data with the likelihood in accordance with the instruction information by the first user device. The communication method according to claim 11.

13. A communication method in a user equipment for generating a measurement report using a measurement report model, comprising: The user equipment calculates a likelihood of input data to be input to a predetermined block included in the measurement report model and a likelihood of first output data to be output from the predetermined block; and discarding the input data and / or the first output data when the likelihood of the input data and / or the likelihood of the first output data is equal to or less than a first threshold. Communication methods.

14. The user device further comprises deriving a trained model for the predetermined block based on the input data and the first output data; The calculating includes any one of the user device calculating the first output data for the input data using the predetermined block, and the user device calculating the second output data for the input data using the trained model; The discarding includes the user device discarding the input data and / or the second output data when a likelihood of the input data and / or a likelihood of the second output data is equal to or less than a second threshold. The communication method according to claim 13.

15. A data transmission entity comprising: A control unit that derives a learned model for a predetermined block based on a signal received from a data receiving entity, The control unit calculates output data for input data using either the predetermined block or the trained model, The control unit performs a predetermined process when the likelihood of the input data and / or the likelihood of the output data is equal to or less than a threshold. Data sending entity.

16. A data transmission entity comprising: Deriving a trained model for the given block based on a signal received from the data receiving entity; A control unit that calculates output data for input data using either the predetermined block or the trained model; a transmitter for transmitting the input data and the output data to the data receiving entity; The control unit calculates the likelihood of the input data and / or the likelihood of the output data. Data sending entity.

17. A user device, comprising: A control unit that derives a trained model for a predetermined block based on a signal received from a base station; A transmitter unit, The control unit calculates output data for input data using the predetermined block and / or the trained model, The control unit calculates a likelihood of the input data and / or a likelihood of the output data, The transmission unit links the input data and / or the output data with the likelihood, and transmits the input data and / or the output data and the likelihood to the base station. User equipment.

18. A user equipment for generating a measurement report using a measurement report model, comprising: a control unit that calculates a likelihood of input data input to a predetermined block included in the measurement report model and a likelihood of first output data output from the predetermined block; The control unit discards the input data and / or the first output data when the likelihood of the input data and / or the likelihood of the first output data is equal to or less than a first threshold. User equipment.