Base station device, terminal device, control method, and program that distinguish between measured value and estimated value

Abstract

In a base station device that communicates with a terminal device on the basis of a cellular communication standard of the third generation partnership project (3GPP): the terminal device is capable of inputting a measured value of the wireless quality of a received signal measured in the terminal device to a trained model generated by machine learning, thereby executing inference for estimating wireless quality that has not been measured in the terminal device; and the base station receives, from the terminal device, a report including at least one value from among the measured value and an estimated value generated through inference, said report furthermore including specification information indicating whether the value is measured or generated through inference, and either executes a prescribed process using the measured value without using the estimated value or executes the prescribed process using the estimated value without using the measured value, the value used being specified on the basis of the specification information.

Description

Base station apparatus, terminal apparatus, control method and program for distinguishing between measured values and estimated values

[0001] The present invention relates to a technology in which a base station performs communication control based on radio quality in a mobile communication system.

[0002] In the 3rd Generation Partnership Project (3GPP (registered trademark)), utilization of artificial intelligence (AI) and machine learning (ML) for control in mobile communication systems is under consideration. An AI / ML model may be used for inference using AI or ML. One example of an AI / ML model is a trained model generated by machine learning using training data. The AI / ML model may also be referred to as an AI model. For example, examples of use cases where an AI model is used in a terminal of a mobile communication system include beam management, positioning, radio channel quality prediction, and the like. Furthermore, an example of a use case where an AI model is used in both a base station and a terminal of a mobile communication system is compression and restoration of channel state information. Non-Patent Document 1 describes that AI models are used for radio quality estimation, handover determination, and the like, and also describes a method for evaluating AI models.

[0003] 3rd Generation Partnership Project, "TR 38.744 Study on Artificial Intelligence (AI) / Machine Learning (ML) for mobility in NR; (Release 19)"

[0004] When inference using an AI model is used in radio quality measurement performed by a terminal, a problem may arise based on the fact that a value included in a measurement report received by a base station from the terminal may be a value indicating a measured radio quality or a value indicating an estimated radio quality. The present invention provides a technology for a base station to execute appropriate processing based on a value indicating radio quality included in a measurement report even when an estimated value obtained using an AI model is included in the radio quality measurement report from the terminal in a mobile communication system.

[0005] A base station device according to one aspect of the present invention is a base station device that communicates with a terminal device based on the cellular communication standard of the Third Generation Partnership Project (3GPP), wherein the terminal device is capable of performing inference to estimate radio quality not measured by the terminal device by inputting measured values ​​of radio quality of a received signal measured by the terminal device into a trained model generated by machine learning, and the base station device includes receiving means for receiving a report from the terminal device which includes at least one of the measured value and the estimated value generated by the inference, and further includes specific information indicating whether the value was generated by measurement or estimation, and processing means which perform a predetermined process using the measured value without using the estimated value, or performs a predetermined process using the estimated value without using the measured value, as determined based on the specific information.

[0006] A terminal device according to one aspect of the present invention is a terminal device that communicates with a base station device based on the cellular communication standard of the Third Generation Partnership Project (3GPP), and comprises: an estimation means that performs inference to estimate radio quality not measured by the terminal device by inputting measured values ​​of radio quality of a received signal measured by the terminal device into a trained model generated by machine learning; and a transmission means that transmits to the base station device a report that includes at least one of the measured value and the estimated value generated by the inference, and further includes specific information indicating whether the value was generated by measurement or estimation.

[0007] According to the present invention, in a mobile communication system, even if the measurement report of wireless quality by a terminal includes an estimated value using an AI model, the base station will be able to perform appropriate processing based on the value indicating wireless quality included in the measurement report.

[0008] Other features and advantages of this disclosure will become apparent from the following description with reference to the accompanying drawings. In the accompanying drawings, the same or similar components are given the same reference numeral.

[0009] The attached drawings are included in the specification and constitute part thereof, illustrating embodiments of the present disclosure and used together with the description to explain the principles of the present disclosure. Figure 1 is a diagram showing an example configuration of a mobile communication system. Figure 2 is a diagram illustrating an example of operation for measuring radio quality by a terminal. Figure 3 is a diagram showing an example of the hardware configuration of a base station. Figure 4 is a diagram showing an example of the functional configuration of a base station. Figure 5 is a diagram showing an example of the functional configuration of a terminal. Figure 6 is a diagram showing an example of a message sequence exchanged between a base station and a terminal. Figure 7 is a diagram illustrating an example of the operation of a base station. Figure 8 is a diagram showing an example configuration of a mobile communication system. Figure 9 is a diagram showing an example of a message sequence exchanged between a base station and a terminal. Figure 10 is a diagram illustrating an example of the operation of a base station.

[0010] The embodiments will be described in detail below with reference to the attached drawings. Note that the following embodiments do not limit the invention as defined in the claims, and not all combinations of features described in the embodiments are essential to the invention. Two or more of the features described in the embodiments may be combined in any way. Furthermore, identical or similar configurations will be given the same reference numeral, and redundant descriptions will be omitted.

[0011] (System Configuration) Figure 1 shows an example of the configuration of a mobile communication system according to this embodiment. The mobile communication system of this embodiment is, for example, a cellular communication system compliant with the cellular communication standard of the Third Generation Partnership Project (3GPP®). The cellular communication standard may be Long Term Evolution (LTE), Fifth Generation Mobile Communication System (5G, NR), Beyond 5G, 6G, etc. However, it is not limited to these, and the following discussion can be applied to a mobile communication system compliant with any wireless communication standard. This mobile communication system is configured, for example, to include a base station 101 and a terminal 111. The base station 101 exchanges radio signals with the terminal 111 via a wireless medium. The base station 101 includes, for example, gNB (next Generation Node B), eNB (evolved Node B), etc. The base station 101 is connected to a core network (not shown). The core network may be, for example, an Evolved Packet System (EPS) or a 5G Core Network (5GC). Terminal 111 is a terminal used by a user and exchanges radio signals with base station 101 via a wireless medium. Terminal 111 may be called User Equipment (UE). Terminal 111 includes, for example, smartphones, mobile phones, personal computers, tablet terminals, wearable devices, IoT (Internet of Things) terminals, etc. For example, base station 101 and terminal 111 can communicate in a cell 121 provided by base station 101 using the connection established between base station 101 and terminal 111. Figure 1 shows an example where there is one base station 101 and one terminal 111, but in a mobile communication system, there may be two or more base stations 101 and two or more terminals 111. In this case, multiple terminals 111 may be connected to one base station 101, or one terminal 111 may be connected to multiple base stations 101.

[0012] In a mobile communication system, communication is controlled based on the state of the channel (radio quality) between the base station 101 and the terminal 111. For example, the base station 101 can control communication parameters such as selecting the beam to be used for communication, allocating radio resources, controlling the transmit power, and selecting the MCS, based on the radio quality between the terminal 111 and its own device. The base station 101 can also perform communication control such as executing a handover procedure, based on the radio quality between the terminal 111 and its own device or other nearby base stations. MCS is an abbreviation for Modulation and Coding Scheme. The base station 101 can also periodically or aperiodicly transmit a reference signal used to measure radio quality, causing the terminal 111 to perform radio quality measurement. The terminal 111 can perform radio quality measurement according to the instructions of the base station 101 and report the measurement results. The terminal 111 can also communicate with the base station 101 based on communication parameters and instructions notified by the base station 101.

[0013] Figure 2 shows an overview of the radio quality measurement performed at terminal 111. Base station 101 periodically or aperiodically transmits reference signals 201 to 205 used for measuring radio quality. Each of reference signals 201 to 205 may contain one or more reference signals. For example, base station 101 may transmit an SSB as a reference signal. SSB is an abbreviation for SS / PBCH Block, where SS is an abbreviation for Synchronization Signal and PBCH is an abbreviation for Physical Broadcast Channel. Base station 101 may transmit an SS burst set containing multiple SSBs. For example, if base station 101 can form multiple beams with different directivity, it may transmit each SSB included in the SS burst set using its respective beam. Terminal 111 can measure radio quality using signals received on radio resources during the period notified by base station 101. Values ​​indicating radio quality include RSRP, RSRQ, RSSI, SINR, etc. RSRP is an abbreviation for Reference Signal Received Power. RSRQ is an abbreviation for Reference Signal Received Quality. RSSI is an abbreviation for Received Signal Strength Indicator. SINR is an abbreviation for Signal to Interference plus Noise Ratio. For example, terminal 111 may be instructed by base station 101 to measure radio quality using an RRC Connection Reconfiguration message that includes measurement configuration. The measurement configuration may include information that identifies the period 211 to 215 during which terminal 111 should perform measurements and the radio resources to be used. Information regarding radio quality measurements not included in the measurement configuration may be individually notified by signals or messages indicating SSB or radio resource allocation. Furthermore, the measurement configuration may include information that identifies the information that terminal 111 should include in the measurement report.For example, terminal 111 may measure RSRP in each of the periods 211 to 215 according to the measurement settings and report information to base station 101 that identifies the beam with the highest RSRP. As an example, terminal 111 may use the SSB identification information contained in each received SSB to report to base station 101 the identification information of the SSB with the highest RSRP, or the identification information of a predetermined number of SSBs in descending order of RSRP. The reference signal transmitted by base station 101 may be CSI-RS. CSI-RS is an abbreviation for Channel State Information Reference Signal. For example, base station 101 may periodically transmit SSBs and, if necessary, transmit CSI-RS aperiodically. Base station 101 may also periodically transmit CSI-RS. Terminal 111 can report the radio quality measured using the received SSB, CSI-RS, or other reference signals to base station 101.

[0014] Furthermore, the control performed by the base station 101 based on the radio quality reported to the terminal 111 is not limited to beam selection. For example, if the base station 101 acquires information indicating the radio quality for each radio resource, it may allocate radio resources to the terminal 111 based on this information. The base station 101 may also perform controls such as transmit power control and MCS determination for the terminal 111 based on the acquired radio quality. In addition, the base station 101 may cause the terminal 111 to perform a handover based on the radio quality between the serving cell provided by its device and the surrounding cells reported by the terminal 111. For example, when the terminal 111 measures the radio quality of the surrounding cells, the reference signals 201 to 205 shown in Figure 2 may be SSB transmitted from the surrounding cells. The base station 101 may cause the terminal 111 to perform a handover to the surrounding cells when the pre-set handover conditions are met based on the measured radio quality. For example, base station 101 may cause terminal 111 to perform a handover to a surrounding cell if the RSRP of a surrounding cell exceeds a predetermined threshold above the RSRP of the cell provided by its own device. The handover conditions may be changed based on the communication status. For example, the handover conditions may be changed to reduce the number of handover failures based on the handover conditions and information indicating the success or failure of handovers performed using those conditions. Changes to the handover conditions may be performed by base station 101 or network functions connected to base station 101.

[0015] Artificial intelligence (AI) and machine learning (ML) inference can be applied to control systems in mobile communication systems. AI is an abbreviation for Artificial Intelligence. ML is an abbreviation for Machine Learning. When performing inference using AI or ML, AI / ML models can be used. An example of an AI / ML model is a trained model generated by machine learning using training data. AI / ML models can also be called AI models. A model used when inference is performed at either the base station 101 or the terminal 111 in a mobile communication system is called a one-sided model. For example, use cases in which a one-sided model is used at the terminal 111 of a mobile communication system include beam management, positioning, and radio channel quality prediction. On the other hand, a model used when inference is performed at both the base station 101 and the terminal 111 in a mobile communication system is called a two-sided model. One example of a use case where a two-sided model is used is the compression and restoration of channel state information. However, the use cases in mobile communication systems where an AI model is used are not limited to these.

[0016] By using inference with an AI model in the measurement of wireless quality performed by terminal 111, terminal 111 can efficiently perform wireless quality measurements. For example, if terminal 111 can estimate the wireless quality for periods 213 to 215 using the wireless quality measured in periods 211 and 212, it can report the estimated value to base station 101 instead of the measured value without performing measurements in periods 213 to 215. This can reduce the processing load on terminal 111 for performing wireless quality measurements. Terminal 111 may perform power-saving operations during periods when measurements are not performed, and may communicate data with base station 101 during those periods. Furthermore, terminal 111 may provide base station 101 with a value indicating the wireless quality for periods 213 to 215, estimated using the wireless quality measured in periods 211 and 212, as a future predicted value. This makes it possible to perform a handover earlier, taking into account the movement of terminal 111, in situations where terminal 111 is moving at high speed. Furthermore, the base station 101 may reduce the number of reference signals transmitted during beam selection, assuming that the terminal 111 performs wireless quality estimation. For example, the base station 101 may use a portion of the beam formed by its own device to transmit the reference signals included in each of the reference signals 201 to 205, and may not transmit reference signals for the remaining beams. In this case, it becomes possible to reduce the number of SSBs that make up each of the reference signals 201 to 205, and to shorten the period for transmitting SSBs using the same beam. The terminal 111 can estimate the wireless quality of the beams not used for transmitting reference signals by inference using an AI model. This makes it possible to shorten the time required for beam selection.

[0017] As described above, the use of inference using an AI model in the measurement of wireless quality performed by terminal 111 can make the measurement of wireless quality performed by terminal 111 more efficient. However, if the accuracy of the AI ​​model's inference is low, the communication quality may deteriorate if base station 101 controls communication using the estimated values. Communication quality can be indicated by, for example, throughput, latency, packet error rate, etc. For example, if there is a large error between the value indicating the actual wireless quality and the estimated value, inappropriate communication parameters may be selected. For example, inappropriate beam, frequency resources, transmit power, MCS, etc. may be selected. In this case, the likelihood of communication failure between base station 101 and terminal 111 increases, thus degrading the communication quality provided to terminal 111. For example, a decrease in throughput, an increase in latency, etc., may occur. Thus, if the wireless quality report by terminal 111 includes not only measured values ​​but also estimated values, the communication quality provided to terminal 111 may deteriorate due to the selection of inappropriate communication parameters. In contrast, base station 101 can select appropriate communication parameters by using only measured values ​​and highly accurate estimated values ​​from the wireless quality information included in the report by terminal 111. However, the base station 101 cannot distinguish which of the wireless quality information included in the report from the terminal 111 is a measured value and which is an estimated value.

[0018] Furthermore, if the AI ​​model's inference accuracy is low, a handover may be performed when it is determined that the handover conditions have been met even though they are not actually met. In this case, the likelihood of the handover failing increases. Then, the handover conditions may be changed based on the failure of the handover. When the handover conditions are changed in such a situation, the change may be a mixture of a change made because the previously used handover conditions were inappropriate and a change made because the AI ​​model's inference accuracy was low, which may result in inappropriate handover conditions not being set. In contrast, appropriate handover conditions can be set by determining whether or not to change the handover conditions based on the success or failure of the handover when the handover conditions are met based on measured values ​​or highly accurate estimated values ​​of wireless quality. However, the base station 101 and the network function that sets the handover conditions cannot distinguish which of the wireless quality information included in the report from the terminal 111 are measured values ​​and which are estimated values.

[0019] In light of these circumstances, the terminal 111 in this embodiment measures the wireless quality and, by inputting the measured wireless quality of the received signal to the AI ​​model, if it estimates the wireless quality that has not been measured, it transmits a report to the base station 101 that includes at least one of the measured or estimated value indicating the wireless quality, and specific information indicating whether the value indicating the wireless quality was generated by measurement or estimation. When the base station 101 receives a report from the terminal 111 that includes specific information indicating whether the value indicating the wireless quality was generated by measurement or estimation, it identifies whether each value was generated by measurement or estimation based on that specific information. The base station 101 then performs a predetermined process using the identified measured value without using the estimated value. Alternatively, the base station 101 performs a predetermined process using the identified estimated value without using the measured value.

[0020] For example, when base station 101 determines communication parameters to be used for communication with terminal 111 using measured values ​​or estimated values ​​included in the report from terminal 111, it uses information indicating the success or failure of communication performed using communication parameters determined based on estimated values, and does not use information indicating the success or failure of communication performed using communication parameters determined based on measured values ​​to determine the accuracy of the AI ​​model's inference. With this configuration, base station 101 can identify that the radio quality included in the report from terminal 111 is an estimated value, and can determine the accuracy of the AI ​​model's inference using the success or failure of communication based on the estimated value. As a result, if the accuracy of the AI ​​model's inference is low, base station 101 can perform processing to improve the accuracy of that inference. For example, base station 101 can perform processing such as shortening the measurement period at terminal 111, increasing the ratio of measurement to estimation, or shortening the period in which base station 101 transmits a reference signal. Also, base station 101 may instruct terminal 111 not to perform inference using the AI ​​model.

[0021] Furthermore, when the base station 101 decides whether or not to have terminal 111 perform a handover based on whether the handover conditions are met by the measured values ​​or estimated values ​​included in the report from terminal 111, it uses information indicating the success or failure of a handover when the handover conditions are met by measured values, and does not use information indicating the success or failure of a handover when the handover conditions are met by estimated values, to determine whether or not to change the handover conditions. With this configuration, the base station 101 can determine whether or not to change the handover conditions based on the success or failure of a handover when the handover conditions are met by measured values. As a result, appropriate changes to the handover conditions based on inappropriate conditions can be made without being affected by handovers that failed due to low accuracy of inference using the AI ​​model. An example of the configuration and operation of the base station 101 and terminal 111 operating in this manner is described below.

[0022] (Circuit Configuration) An example of the configuration of the base station 101 and terminal 111 will be described. Figure 3 is a diagram showing the hardware configuration of the base station 101. In one example, the base station 101 and terminal 111 are configured to include a processor 301, ROM 302, RAM 303, storage device 304, and communication circuit 305. The processor 301 is a computer configured to include one or more processing circuits, such as a general-purpose CPU (Central Processing Unit) or ASIC (Application-Specific Integrated Circuit). The processor 301 performs the overall processing of the device and the above-mentioned processing by reading and executing programs stored in the ROM 302 and storage device 304. The ROM 302 is a read-only memory in which information such as programs and various parameters related to the processing performed by the base station 101 and terminal 111 is recorded. The RAM 303 functions as a workspace when the processor 301 executes programs and is a random access memory in which temporary information is recorded. The storage device 304 is configured, for example, by a removable external storage device. The communication circuit 305 is configured to include, for example, circuits for wired or wireless communication between the base station 101 and the terminal 111. For example, the base station 101 and the terminal 111 can communicate with the other device using the communication circuit 305 for LTE or 5G and an antenna (not shown).

[0023] (Functional Configuration) Figure 4 shows an example of the functional configuration of the base station 101. The base station 101 is configured to include, for example, a radio quality information receiving unit 401, a communication parameter control unit 402, an inference accuracy determination unit 403, a measurement setting control unit 404, a handover execution unit 405, and a handover condition setting unit 406. Figure 4 shows the functional configuration of the base station 101 in this embodiment, and the general configuration of the base station 101 is omitted, for example. These functional units can be realized, for example, by the processor 301 executing a program stored in the ROM 302 or storage device 304 and controlling the communication circuit 305 as needed. However, it is not limited to this, and for example, dedicated hardware for realizing each function may be provided.

[0024] The wireless quality information receiving unit 401 receives reports from the terminal 111 that include values ​​indicating wireless quality. For example, the wireless quality information receiving unit 401 may receive measurement reports or CSI feedback from the terminal 111. CSI feedback may be information notifying wireless quality measured using CSI-RS. CSI feedback may be reported at a shorter interval than measurement reports. The values ​​indicating wireless quality may be measured values ​​of wireless quality measured by the terminal 111, or estimated values ​​of wireless quality estimated using those measured values. The wireless quality information receiving unit 401 may also receive specific information indicating whether each value indicating wireless quality is a measured value or an estimated value. The communication parameter control unit 402 controls the communication parameters used for communication with the terminal 111. For example, the communication parameter control unit 402 may control the communication parameters based on the values ​​indicating wireless quality received from the terminal 111.

[0025] The inference accuracy determination unit 403 determines the accuracy of the inference performed using the AI ​​model executed at terminal 111. For example, the inference accuracy determination unit 403 can determine the accuracy of the inference using the AI ​​model based on the success or failure of communication performed using communication parameters selected based on the estimated value of wireless quality from among the values ​​indicating wireless quality received from terminal 111. As an example, the inference accuracy determination unit 403 can determine that the accuracy of the inference using the AI ​​model is low based on the success rate of the communication falling below a predetermined threshold.

[0026] The measurement setting control unit 404 controls the settings for the wireless quality measurement to be performed by the terminal 111. For example, the measurement setting control unit 404 can control the settings for wireless quality measurement and estimation to be performed by the terminal 111 by notifying the terminal 111 of the measurement settings (Measurement Configuration). In addition, if the measurement setting control unit 404 determines that the accuracy of the inference using the AI ​​model performed by the terminal 111 is low, it can notify the terminal 111 of new measurement settings. For example, the new measurement settings may include settings to improve the accuracy of the inference using the AI ​​model performed by the terminal 111, or instructions indicating that the terminal 111 should not perform inference using the AI ​​model.

[0027] The handover execution unit 405 executes a process to cause terminal 111 to perform a handover. For example, the handover execution unit 405 determines whether the handover conditions are met based on a value indicating wireless quality received from terminal 111. If the handover conditions are met, the handover execution unit 405 may cause terminal 111 to perform a handover. The handover condition setting unit 406 sets handover conditions for determining whether or not to cause terminal 111 to perform a handover. For example, the handover condition setting unit 406 may determine whether or not to change the handover conditions based on the success or failure of a handover that was performed based on the measurement value of wireless quality among the values ​​indicating wireless quality received from terminal 111, which indicates that the handover conditions were met. Based on the determination that the handover conditions should be changed, the handover condition setting unit 406 may change the handover conditions.

[0028] Figure 5 shows an example of the functional configuration of terminal 111. Terminal 111 includes, for example, a wireless quality measurement unit 501, a wireless quality estimation unit 502, and a wireless quality reporting unit 503 as its functions. Figure 5 shows the functional configuration of base station 101 in this embodiment, and omits, for example, the general configuration of base station 101. These functional units can be realized, for example, by a processor 301 executing a program stored in ROM 302 or storage device 304 and controlling the communication circuit 305 as needed. However, it is not limited to this, and for example, dedicated hardware for realizing each function may be provided.

[0029] The wireless quality measurement unit 501 measures the wireless quality using the received signal. For example, the wireless quality measurement unit 501 measures the wireless quality based on the measurement settings notified by the base station 101. The wireless quality estimation unit 502 estimates the unmeasured wireless quality using the measured wireless quality values ​​measured by the wireless quality measurement unit 501. For example, the wireless quality estimation unit 502 estimates the wireless quality based on the measurement settings notified by the base station 101. The wireless quality reporting unit 503 reports the measured or estimated wireless quality to the base station 101. For example, the wireless quality reporting unit 503 may report using a Measurement Report or CSI feedback. As an example, it reports the wireless quality based on the measurement settings notified by the base station 101. The wireless quality reporting unit 503 may also include specific information in the measurement settings indicating whether each value representing the wireless quality is a measured value or an estimated value.

[0030] (Processing Flow) (First Embodiment) In the mobile communication system shown in Figure 1, the operation of the base station 101 and the terminal 111 when a terminal 111 connected to the base station 101 performs a measurement of radio quality will be described. Figure 6 shows an example of the processing performed by the base station 101 and the terminal 111 when a terminal 111 connected to the base station 101 performs a measurement of radio quality and reports the measurement results to the base station 101. The base station 101 is assumed to periodically transmit reference signals, as shown in the transmission of reference signals 201 to 205 in Figure 2. The reference signals may be SSB, CSI-RS, etc. The reference signals may also be other signals that can be used to measure radio quality. The terminal 111 is assumed to take the measured radio quality values ​​as input and perform inference using an AI model to estimate the radio quality of the unmeasured signals. For example, the terminal 111 is assumed to measure the radio quality of the received signal during a portion of the period 211 to 215 in Figure 2, and to estimate the radio quality during the other period 211 to 215. As an example, terminal 111 may perform wireless quality measurements during periods 211 and 215, and use the measured values ​​obtained in those measurements to estimate the wireless quality during periods 212 to 214. Alternatively, terminal 111 may use the measured values ​​obtained in the wireless quality measurement during period 211 to estimate the wireless quality during periods 212 to 214, or it may use the measured values ​​obtained in the wireless quality measurements during periods 211 and 215 to estimate the wireless quality during periods 212 to 214.

[0031] First, the base station 101 instructs the terminal 111 to perform a measurement of the radio quality (S601). For example, the base station 101 may instruct the terminal 111 to perform a measurement of the radio quality by notifying it of an RRC Connection Reconfiguration message that includes the measurement configuration. For example, the measurement configuration may include the timing and period in which the reference signal is transmitted, information identifying the radio resource to which the reference signal is transmitted, the period in which the measurement or estimation should be performed, information to be included in the measurement report, and information identifying the timing and period in which the measurement report should be performed. The base station 101 may also individually notify the terminal 11 of any information that is not included in the measurement configuration using other signals or messages. The information to be included in the measurement report may include a value indicating the radio quality and specific information that identifies whether the value was generated by measurement or estimation. The information to be included in the measurement report may also include information identifying the frequency resource that was measured or estimated, and information identifying the SSB or beam that was measured or estimated. Furthermore, the measurement settings may include instructions regarding inference using an AI model performed by terminal 111. For example, the measurement settings may include information that identifies the AI ​​model that terminal 111 should use for inference. The measurement settings may also include information indicating the period during which terminal 111 should perform measurements and the period during which it should perform inferences, or information indicating the proportion of time that terminal 111 should perform measurements and the proportion of time that it should perform inferences. Note that the measurement settings may include multiple settings. For example, if terminal 111 is to perform periodic measurements using SSB and non-periodic measurements using CSI-RS, the settings corresponding to each measurement may be included in one measurement setting. Note that base station 101 may notify terminal 111 of the settings corresponding to each measurement as separate measurement settings.

[0032] Base station 101 transmits reference signals (S602, S604, S610). The reference signals may be the same, or they may be a combination of different reference signals. For example, some reference signals may be SSB and others may be CSI-RS. The reference signals may be transmitted periodically or aperiodicly. Terminal 111 measures and estimates radio quality based on the measurement settings (S603, S605, S611). For example, terminal 111 may measure radio quality at a timing identified as appropriate based on the measurement settings, and estimate radio quality at a timing identified as appropriate based on the measurement settings.

[0033] Terminal 111 measures or estimates the radio quality and then reports the radio quality based on the measurement settings (S607, S613). For example, after measuring or estimating the radio quality, terminal 111 may request the base station 101 to allocate radio resources for reporting. For example, when terminal 111 receives the allocation of radio resources for reporting from base station 101 (S606), it reports a value indicating the radio quality generated by measurement or estimation using the allocated radio resources (S607). For example, terminal 111 may provide a report that includes the measured value of the radio quality measured in S603. Note that terminal 111 may not be able to report the radio quality immediately after measuring or estimating the radio quality. For example, after measuring or estimating the radio quality, terminal 111 may request the allocation of radio resources for reporting. In this case, it may take some time for base station 101 to allocate radio resources to terminal 111. For this reason, base station 101 may not be able to determine whether the value indicating the radio quality included in the received report is a measured value or an estimated value based on the timing of receiving the report. Furthermore, if the measurement settings include information indicating the wireless resources to be used for reporting, terminal 111 may use those wireless resources to submit a measurement report. In this case as well, if base station 101 does not know when the value indicating wireless quality received using those wireless resources was measured or estimated, it cannot determine whether that value is a measured value or an estimated value. For this reason, terminal 111 reports the value indicating wireless quality included in the report, associating it with specific information indicating whether that value is a measured value or an estimated value. For example, if the value indicating wireless quality included in the report is a measured value, terminal 111 may add the specific information "Measured" to the value, and if the value indicating wireless quality included in the report is an estimated value, it may add the specific information "Estimated" to the value. Furthermore, if the value indicating wireless quality included in the report is a measured value, terminal 111 may add specific information corresponding to the value indicating wireless quality being a measured value, such as "0" or "1", "True" or "False".In this case, if the value indicating wireless quality included in the report is an estimated value, terminal 111 may add specific information to the value corresponding to the estimated value, such as "1" or "0", "False" or "True". For example, if one report includes one value indicating wireless quality, terminal 111 may add one piece of specific information indicating whether that value is a measured value or an estimated value. On the other hand, if one report includes multiple values ​​indicating wireless quality, terminal 111 may add one piece of specific information to each value indicating whether that value is a measured value or an estimated value. In addition, if one report includes multiple values ​​indicating wireless quality, terminal 111 may add one piece of specific information to multiple values. For example, terminal 111 may add one piece of specific information to multiple measured values ​​of wireless quality indicating that these values ​​are measured values. Also, terminal 111 may add one piece of specific information to multiple estimated values ​​of wireless quality indicating that these values ​​are estimated values. In addition, terminal 111 may add specific information to either measured values ​​or estimated values, or both. For example, if the default value indicating wireless quality is a measured value, terminal 111 may add information to the estimated value indicating that the value is an estimated value. This reduces the amount of information communicated for specific information. Similarly, for reports where wireless quality estimation is not performed, terminal 111 may add a new area to include the wireless quality value generated by estimation, thereby indicating whether each wireless quality value is a measured value or an estimated value. If the value indicating wireless quality is composed of multiple elements, terminal 111 may add specific information to each element, or add one specific piece of information to multiple elements. For example, if CSI is composed of CQI, PMI, and RI, one specific piece of information may be added to CSI, or one specific piece of information may be added to each of CQI, PMI, and RI. CQI is an abbreviation for Channel Quality Indicator. PMI is an abbreviation for Precoding Matrix Indicator. RI is an abbreviation for Rank Indicator.Furthermore, terminal 111 may add specific information to the value indicating wireless quality in the format notified in the measurement settings notified by base station 101. For example, base station 101 may specify the format of the specific information to be added to the value indicating wireless quality in the measurement settings.

[0034] The base station 101 controls communication with the terminal 111 based on a value indicating radio quality included in the received report (S608). For example, the base station 101 may select the beam to be used for communication with the terminal 111, select the radio resources to be allocated to communication with the terminal 111, select parameters related to transmit power control to be used by the terminal 111, and select the MCS to be used for communication with the terminal 111. The base station 101 communicates data with the terminal 111 using the selected communication parameters (S609). The base station 101 may terminate data communication if successful, or retransmit the failed data if it fails. For example, when the base station 101 transmits data, it may determine the success or failure of the communication based on the acknowledgment received from the terminal 111. Also, when the terminal 111 transmits data, the base station 101 may determine the success or failure of the communication based on whether or not the data was successfully received by its own device. The base station 101 can store success / failure information indicating the success or failure of data communication, and identification information indicating whether the communication parameters used for that data communication were selected based on measured values ​​of radio quality or estimated values ​​of radio quality, in association with each other. Based on specific information contained in the report received from the terminal 111, the base station 101 can determine whether the values ​​indicating radio quality used to select the communication parameters are measured values ​​or estimated values. For example, the base station 101 stores success / failure information indicating the success or failure of the communication performed in S609, in association with identification information indicating that this communication was performed using communication parameters selected based on measured values ​​of radio quality.

[0035] Next, the base station 101 receives a measurement report from the terminal 111 (S613). For example, the measurement report may include a value indicating the radio quality estimated by the terminal 111 in S605. Based on the value indicating the radio quality included in the received report, the base station 101 controls communication with the terminal 111, similar to S608 (S614). The base station 101 communicates data with the terminal 111 using the selected communication parameters, similar to S609 (S615). When the base station 101 determines whether the data communication was successful or not, it stores success / failure information indicating the success or failure of the data communication and identification information indicating that this communication was performed using communication parameters selected based on the estimated value of the radio quality, in association with each other.

[0036] Then, the base station 101 determines the accuracy of the inference performed by the terminal 111 using the AI ​​model (S616). For example, the base station 101 may perform this determination periodically. Alternatively, the base station 101 may perform this determination based on whether a value indicating the communication quality in communication with the terminal 111 falls below a predetermined threshold. The value indicating the communication quality may be, for example, throughput, delay, packet error rate, etc. Figure 7 is a flowchart showing an example of the process for determining the accuracy of inference using the AI ​​model performed by the base station 101. First, the base station 101 acquires information indicating the success or failure of communication performed using communication parameters selected based on the estimated value of the radio quality (S701). For example, the base station 101 acquires information indicating the success or failure of communication performed using communication parameters selected based on the estimated value of the radio quality, based on success or failure information stored in its own device and identification information associated with each success or failure information that identifies whether the communication was performed using communication parameters selected based on the measured value or the estimated value of the radio quality. The base station 101 then determines whether the success rate of the communication is below a predetermined threshold (S702). For example, the success rate of the communication may be the packet error rate (PER) or the block error rate (BLER).

[0037] If the success rate of the communication falls below a predetermined threshold (YES in S702), the base station 101 performs a predetermined process. For example, as a predetermined process, the base station 101 may perform a process to improve the accuracy of inference using the AI ​​model in the terminal 111. As an example, the base station 101 may shorten the period during which the terminal 111 measures the wireless quality. For example, the base station 101 may instruct the terminal 111 to measure the wireless quality in period 213 in addition to periods 211 and 215 in Figure 2. In this case, wireless quality estimation may be performed in periods 212 and 214. By shortening the period during which wireless quality is measured, the amount of data input to the AI ​​model may increase. Also, by shortening the period during which wireless quality is measured, measured values ​​taken closer to the timing of wireless quality estimation may be input to the AI ​​model. In this way, by shortening the period during which the terminal 111 measures the wireless quality, the accuracy of inference using the AI ​​model may be improved. Furthermore, as a predetermined process, the base station 101 may shorten the period of the reference signals it transmits. For example, the base station 101 may transmit reference signal 203 when it has transmitted reference signals 201 and 205, and the terminal 111 is using these reference signals to measure the radio quality for periods 211 and 215, and using the acquired measurements to estimate the radio quality for periods 212 to 214. The base station 101 may then instruct the terminal 111 to measure the radio quality for periods 211, 213, and 215, respectively, and to estimate the radio quality for periods 212 and 214. In this way, when the terminal 111 is measuring the radio quality for each of the reference signals transmitted by the base station 101, the period during which the terminal 111 measures the radio quality can be further shortened by adding a reference signal transmitted by the base station 101. This can improve the accuracy of inference using the AI ​​model. Furthermore, as a predetermined process, the base station 101 may instruct the terminal 111 to measure the wireless quality during each period and report the measured wireless quality values, without performing inference using the AI ​​model. This can resolve the situation in which appropriate communication parameters are not selected due to the low accuracy of inference using the AI ​​model.On the other hand, if the communication success rate does not fall below a predetermined threshold (NO in S702), the base station 101 terminates the process of determining the accuracy of the inference using the AI ​​model.

[0038] Terminal 111 may be notified of new measurement settings in order to perform the predetermined processing described above (S617). The new measurement settings may include information indicating the period during which terminal 111 should measure wireless quality or estimate wireless quality. The new measurement settings may also include information indicating the period during which a reference signal is transmitted. Furthermore, the new measurement settings may include information indicating that terminal 111 should measure and report wireless quality without performing inference using an AI model. When terminal 111 receives new measurement settings from base station 101, it measures or estimates wireless quality based on those measurement settings (S617). For example, terminal 111 measures wireless quality based on the period during which wireless quality should be measured as indicated in the measurement settings. Also, terminal 111 estimates wireless quality based on the period during which wireless quality should be estimated as indicated in the measurement settings. If the measurement settings include information indicating that wireless quality should be measured without performing inference using an AI model, terminal 111 measures wireless quality without performing inference using an AI model.

[0039] (Second Embodiment) In the first embodiment, an example was described in which the base station 101 determines the accuracy of inference using an AI model based on the success or failure of communication performed using communication parameters selected based on estimated values ​​among the radio quality values ​​included in the report received from the terminal 111 connected to the device. In this example, an example is described in which the necessity of changing the handover conditions is determined based on the success or failure of a handover performed based on measured values ​​among the radio quality values ​​reported by the terminal 111. The determination of whether or not to change the handover conditions may be performed by the base station 101 or by other network functions. For example, the determination of whether or not to change the handover conditions may be performed by a SON server that provides SON functionality. SON is an abbreviation for Self-Organizing Network. For example, the SON server may provide an MRO function that optimizes the handover conditions to reduce handover failures based on the report from the terminal 111, including the success or failure of the handover. MRO is an abbreviation for Mobility Robustness Optimization.

[0040] Figure 8 shows an example of a mobile communication system in a second embodiment. The mobile communication system shown in Figure 8 consists of base stations 101 and 102, and terminals 111 and 112. The configuration of base stations 101 and 102 is the same as that of base station 101 in Figure 1. Similarly, the configuration of terminals 111 and 112 is the same as that of terminal 111 in Figure 1. In this example, base stations 101 and 102 may be referred to as base station 100 without distinction. Similarly, terminals 111 and 112 may be referred to as terminal 110 without distinction. Each of terminals 111 and 112 is assumed to have established a connection with base station 101. Furthermore, terminals 111 and 112 are located in an area where cell 121 provided by base station 101 and cell 122 provided by base station 102 overlap. For example, terminals 111 and 112 can receive signals from cell 121 and cell 122. Base station 101 can communicate with base station 102 via network 131. For example, base station 101 can communicate with base station 102 using the X2 interface or the Xn interface. For example, when causing terminals 111 and 112 to perform a handover from cell 121 to cell 122, base station 101 can send a handover request message to base station 102. The mobile communication system may include other network functions not shown. For example, it may include the SON server described above. The SON server is connected to each of the base stations 100 and can obtain information indicating the success or failure of the handover of terminal 110 via base station 100.

[0041] Figure 9 shows an example of the sequence of operations performed between base station 101, base station 102, terminal 111, and terminal 112 when base station 101 causes terminals 111 and 112 to perform a handover from cell 121 to cell 122. Base station 101, which provides cell 121 as the source cell for the handover when terminals 111 and 112 perform the handover, may be called the source base station. Similarly, base station 102, which provides cell 122 as the destination cell for the handover when terminals 111 and 112 perform the handover, may be called the target base station. First, the source base station 101 causes terminals 111 and 112 to perform a radio quality measurement (S901). For example, base station 101 may cause terminal 110 to perform a radio quality measurement by notifying terminal 110 of an RRC Connection Reconfiguration message that includes measurement settings (Measurement Configuration). For example, the measurement settings may include the timing and period during which a reference signal is transmitted in cell 122, information identifying the radio resource from which the reference signal is transmitted, the period during which measurement or estimation should be performed, information to be included in the measurement report, and information identifying the timing and period during which the measurement report should be performed. While the following explanation focuses on the measurement and estimation of the radio quality of cell 122, the same explanation can be applied to the measurement and estimation of the radio quality of cell 121. Furthermore, the same explanation can be applied if there are other surrounding cells besides cell 122. For example, if the handover condition is that the received power of target cell 122 exceeds the received power of source cell 121 by a threshold, then the radio quality of both source cell 121 and target cell 122 may be measured or estimated. The information to be included in the measurement report may include a value indicating the radio quality and specific information identifying whether that value was generated by measurement or estimation. The information to be included in the measurement report may also include information indicating that a predetermined condition has been met and information indicating whether the information used to determine that the predetermined condition has been met was a measured value or an estimated value. The specified conditions may be handover conditions. Furthermore, the information to be included in the measurement report may include information identifying the cell and frequency on which the measurement or estimation was performed.Furthermore, the measurement settings may include instructions regarding the inference performed by terminal 111 using an AI model. For example, the measurement settings may include information that identifies the AI ​​model that terminal 111 should use for inference. The measurement settings may also include information indicating the cycles in which terminal 111 should perform measurements and inferences, or information indicating the proportion of times terminal 111 should perform measurements and inferences.

[0042] The target base station 102 transmits a reference signal (S902, S906, S915). For example, the reference signal may be SSB. The terminal 110 measures or estimates the radio quality of cell 122 based on the measurement settings (S903, S907, S916). For example, the terminal 110 may measure the radio quality at a time when it is determined that it should measure the radio quality based on the measurement settings, and may estimate the radio quality at a time when it is determined that it should estimate the radio quality based on the measurement settings. After measuring or estimating the radio quality of cell 122, the terminal 110 reports the radio quality based on the measurement settings (S904, S908, S913, S917). For example, terminals 111 and 112 may report including the measured value of the radio quality of cell 122 measured in S903 (S904). Also, terminal 111 may report including the estimated value of the radio quality of cell 122 estimated in S907 (S908). Similarly, terminal 112 may provide a report that includes an estimated value of the wireless quality of cell 122, which was estimated in S907 (S913). Furthermore, terminal 112 may provide a report that includes a measured value of the wireless quality of cell 122, which was measured in S916 (S917). Thus, the value indicating the wireless quality of cell 122 reported by terminal 110 may contain a mixture of measured and estimated values. For this reason, terminal 111, as in the first embodiment, reports the value indicating the wireless quality included in the report by associating it with specific information indicating whether the value is a measured or estimated value.

[0043] Based on the received report, the source base station 101 determines whether or not to have the terminal 110 perform a handover to cell 122 (S905, S909, S914, S918). For example, the source base station 101 may determine whether the handover conditions are met based on a value indicating the radio quality of cell 122, and then determine whether or not to have the terminal 110 perform a handover to cell 122 based on the result of that determination. Also, if the received report indicates that the handover conditions have been met, the source base station 101 may have the terminal 110 perform a handover based on the receipt of the report. In this case, the source base station 101 may identify the cell 122 to be used for the handover in the received report. The handover conditions may be generated by the source base station 101 or provided by the SON server described above. For example, in determining the handover conditions in S905, source base station 101 determines whether the handover conditions are met based on the reports received in S904 from terminals 111 and 112 (S905). Suppose source base station 101 determines that the handover conditions are not met. On the other hand, in determining the handover conditions in S909, source base station 101 determines whether the handover conditions are met based on the estimated value of the radio quality of cell 122 included in the report received in S908 from terminal 111, and determines that the handover conditions are met (S909). In this case, source base station 101 initiates a procedure to have terminal 111 perform a handover to cell 122. For example, source base station 101 sends a Handover Request message to target base station 102. The target base station 102 responds to the source base station 101 using a Handover Request Acknowledgment message. The source base station 101 then notifies the terminal 111 of a Handover Command message (S910). Based on the Handover Command message received from the source base station 101, the terminal 111 performs a handover to the target base station 102 (S911). For example, the terminal 111 may perform a random access procedure.The source base station 101 determines whether the handover performed by terminal 111 was successful or not (S912). For example, the source base station 101 may determine that the handover was successful by receiving a Handover Complete message from terminal 111. The source base station 101 stores success / failure information indicating the success or failure of the handover and identification information indicating whether the value indicating the radio quality used to determine the execution of the handover was a measured value or an estimated value, in association with each other. For example, if the handover by terminal 111 fails, the source base station 101 stores success / failure information indicating that the handover failed and identification information indicating that the execution of this handover was determined based on an estimated value of the radio quality, in association with each other.

[0044] Subsequently, assume that the source base station 101 determines whether a handover condition is satisfied based on the estimated value of the radio quality of the cell 122 included in the report received from the terminal 112 in S913 in the determination of the handover condition in S914, and determines that the handover condition is not satisfied. On the other hand, assume that the source base station 101 determines whether the handover condition is satisfied based on the measured value of the radio quality of the cell 122 included in the report received from the terminal 112 in S917 in the determination of the handover condition in S918, and determines that the handover condition is satisfied. In this case, the source base station 101 performs a procedure for causing the terminal 112 to execute handover to the cell 122 in the same manner as in S909. For example, when the terminal 112 receives a Handover Command message from the source base station 101 (S919), the terminal 112 executes handover to the target base station 102 (S920). The source base station 101 determines whether the handover executed by the terminal 112 has succeeded (S921). For example, the source base station 101 can determine that the handover has succeeded by receiving a Handover Complete message from the terminal 112. The source base station 101 stores success / failure information indicating the success or failure of the handover in association with identification information that identifies whether the value indicating the radio quality used for determining the execution of the handover was a measured value or an estimated value. For example, when the handover of the terminal 112 fails, the source base station 101 stores success / failure information indicating that the handover failed in association with identification information indicating that the execution of the handover was determined based on the measured value of the radio quality.

[0045] The source base station 101 then determines whether or not to change the handover conditions (S922). Based on the determination result, the source base station 101 changes the handover conditions (S923). For example, the source base station 101 may make this determination periodically. Alternatively, the source base station 101 may make this determination based on the handover success rate falling below a predetermined threshold. Figure 10 is a flowchart showing an example of the handover condition determination process performed by the source base station 101. First, the source base station 101 obtains success / failure information indicating the success or failure of a handover performed based on the measurement value of radio quality satisfying the handover conditions (S1001). For example, the base station 101 may obtain success / failure information indicating the success or failure of a handover performed based on the measurement value of radio quality satisfying the handover conditions, based on the handover success / failure information stored in its own device and identification information associated with each success / failure information that identifies whether the execution of that handover was determined based on the measurement value of radio quality or an estimated value. The base station 101 then determines whether the success rate of the handover is below a predetermined threshold (S1002). If the success rate of the handover is below the predetermined threshold (YES in S1002), the base station 101 changes the handover conditions (S1003). For example, the base station 101 may change the handover conditions to ones that have a lower probability of failing than the handover conditions currently in use. On the other hand, if the success rate of communication is not below the predetermined threshold (NO in S1002), the base station 101 terminates the process of determining whether to change the handover conditions.

[0046] In this way, at the source base station 101, the decision to change the handover conditions is made using success / failure information indicating the success or failure of a handover performed based on the measurement value of the radio quality satisfying the handover conditions. As a result, the handover conditions are changed based on the inappropriateness of the handover conditions without being affected by handovers that failed due to the low accuracy of inference using the AI ​​model. Although the above explanation used an example where the base station 101 generates and changes the handover conditions, as mentioned above, the network functions of the base station 101 may also generate and change the handover conditions. For example, when the SON server generates or changes the handover conditions, the base station 101 associates the success / failure information indicating the success or failure of the handover with identification information that identifies whether the value indicating the radio quality used to determine whether the handover was performed was a measurement value or an estimated value, and notifies the SON server. The SON server performs the processing shown in Figure 10 based on the notified success / failure information and identification information to determine whether it is necessary to change the handover conditions. When the SON server changes the handover conditions, it notifies the base station 100 of the changed handover conditions.

[0047] As described above, according to the present embodiment, the base station 101 receives a report from the terminal 111 including specific information indicating whether a value indicating radio quality is a value generated by measurement or estimation, and identifies whether each value is generated by measurement or estimation based on the specific information. Then, the base station 101 determines the inference accuracy using an AI model based on the success rate of communication performed using communication parameters selected based on the estimated value of radio quality. With this configuration, when the inference accuracy of the AI model is low, the base station 101 can perform processing for improving the inference accuracy. Furthermore, the base station 101 determines whether the handover condition should be changed using the success rate of handover executed based on a handover condition being satisfied by a measured value of radio quality. With this configuration, appropriate handover condition modification can be performed based on inappropriate handover conditions without being affected by handover failures caused by low inference accuracy using an AI model. As described above, the base station 101 can appropriately control communication based on whether the value indicating radio quality notified by the terminal 111 is a measured value or an estimated value. This makes it possible to stabilize the radio quality provided to the radio terminal 111. Therefore, it becomes possible to contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs): "Build resilient infrastructure, promote inclusive and sustainable industrialization and foster innovation."

[0048] The invention is not limited to the above embodiments, and various modifications and changes can be made within the scope of the gist of the invention.

[0049] The present application claims priority based on Japanese Patent Application No. 2025-038519 filed on March 11, 2025, and the entire content of the description thereof is incorporated herein by reference.

Claims

1. A base station device that communicates with a terminal device based on the cellular communication standard of the Third Generation Partnership Project (3GPP), wherein the terminal device is capable of performing inference to estimate radio quality not measured by the terminal device by inputting measured values ​​of radio quality of a received signal measured by the terminal device into a trained model generated by machine learning, the base station device includes receiving means for receiving a report from the terminal device which includes at least one of the measured value and the estimated value generated by the inference, and further includes specific information indicating whether the value was generated by measurement or estimation, and processing means which perform a predetermined process using the measured value without using the estimated value, or performs a predetermined process using the estimated value without using the measured value, as determined based on the specific information.

2. The base station device according to claim 1, further comprising determination means for determining communication parameters to be used for the communication with the terminal device using the measured values ​​or estimated values ​​included in the report, wherein the processing means performs a process to determine the accuracy of the inference as a predetermined process, using information indicating the success or failure of the communication performed using the communication parameters determined based on the estimated values, without using information indicating the success or failure of the communication performed using the communication parameters determined based on the measured values.

3. The base station device according to claim 2, wherein the processing means further performs a process to set the period for measuring the radio quality performed by the terminal device based on the accuracy of the inference, as the predetermined process.

4. The base station device according to claim 2, further comprising a transmitting means for transmitting a reference signal used for measuring the radio quality performed by the terminal device, wherein the processing means further performs a process as a predetermined process for controlling the period for transmitting the reference signal based on the accuracy of the inference.

5. The base station device according to claim 2, wherein the processing means further performs a process of instructing the terminal device whether or not to perform the inference based on the accuracy of the inference, as the predetermined process.

6. The base station device according to claim 1, further comprising a determination means for determining whether or not to cause the terminal device to perform a handover based on whether the handover conditions are met by the measured values ​​or estimated values ​​included in the report, wherein the processing means, as a predetermined process, determines whether or not to change the handover conditions using information indicating the success or failure of the handovers performed by the terminal device in which the handover conditions were met by the measured values, without using information indicating the success or failure of the handovers in which the handover conditions were met by the estimated values.

7. The base station device according to claim 6, wherein the processing means further performs control to change the handover condition if it is determined to change the handover condition as a predetermined process, and to maintain the handover condition if it is determined not to change the handover condition.

8. The base station device according to any one of claims 1 to 7, wherein the report includes one piece of specific information associated with each of the measured values ​​or each of the estimated values ​​included in the report.

9. The base station device according to any one of claims 1 to 7, wherein the report includes at least one specific piece of information associated with one or more sets of the measured values ​​included in the report, or one specific piece of information associated with one or more sets of the estimated values ​​included in the report.

10. A terminal device that communicates with a base station device based on the cellular communication standard of the Third Generation Partnership Project (3GPP), comprising: an estimation means that performs inference to estimate radio quality not measured by the terminal device by inputting measured values ​​of radio quality of a received signal measured by the terminal device into a trained model generated by machine learning; and a transmission means that transmits to the base station device a report that includes at least one of the measured value and the estimated value generated by the inference, further including specific information indicating whether the value was generated by measurement or estimation.

11. The terminal device according to claim 10, wherein the report includes one specific piece of information associated with each of the measured values ​​or each of the estimated values ​​included in the report.

12. The terminal device according to claim 10, wherein the report includes at least one specific piece of information associated with one or more sets of the measured values ​​included in the report, or one specific piece of information associated with one or more sets of the estimated values ​​included in the report.

13. A control method performed by a base station device that communicates with a terminal device based on the cellular communication standard of the Third Generation Partnership Project (3GPP), wherein the terminal device is capable of performing inference to estimate radio quality not measured by the terminal device by inputting measured values ​​of radio quality of a received signal measured by the terminal device into a trained model generated by machine learning, and the control method comprises: receiving from the terminal device a report that includes at least one of the measured value and an estimated value generated by the inference, and further including specific information indicating whether the value was generated by measurement or estimation; and performing a predetermined process using the measured value without using the estimated value, or performing a predetermined process using the estimated value without using the measured value, as determined based on the specific information.

14. A control method performed by a terminal device that communicates with a base station device based on the cellular communication standard of the Third Generation Partnership Project (3GPP), comprising: performing inference to estimate radio quality not measured by the terminal device by inputting measured values ​​of radio quality of a received signal measured by the terminal device into a trained model generated by machine learning; and transmitting to the base station device a report that includes at least one of the measured value and the estimated value generated by the inference, further including specific information indicating whether the value was generated by measurement or estimation.

15. A program for causing a computer to operate as each of the means of the base station device described in any one of claims 1 to 9.

Images