Control apparatus, control method, and program

Abstract

Provided is a control apparatus comprising: a trained model storage unit that stores a trained model generated by using a plurality of data sets each including an uplink signal received from a terminal by a base station capable of executing communication using a plurality of antennas, at least one of a transmission output and a transmission timing of the uplink signal, reception quality of the uplink signal, and a propagation path status between the base station and the terminal; an estimation unit that acquires, as an estimated propagation path status, a propagation path status at a future time point outputted by inputting, to the trained model, the uplink signal received by the base station, said at least one of the transmission output and the transmission timing of the uplink signal, and the reception quality of the uplink signal; and a determination unit that determines a weight pattern of beams of the plurality of antennas on the basis of the estimated propagation path status. The control apparatus may execute RAN control and AI processing (including RAN control AI processing such as a RAN intelligent controller (RIC), and non-RAN control AI processing).

Description

Control Device, Control Method, and Program

[0001] The present invention relates to a control device, a control method, and a program.

[0002] Patent Document 1 describes that "the control device (1) predicts the probability distribution of the radio quality when downlink transmission from the base station (2) to the wireless terminal (3) or uplink transmission from the wireless terminal (3) to the base station (2) is performed. The control device (1) calculates the block error rate or the expected value of reliability of downlink or uplink transmission using the obtained probability distribution. The control device (1) determines the value of each of one or more parameters related to downlink or uplink transmission in consideration of the obtained expected value." [Prior Art Document] [Patent Document] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2023-175178

[0003] When communicating with a terminal using a plurality of antennas provided in a base station, the beam weights of the plurality of antennas are determined based on the uplink signal from the terminal. The base station determines the beam weights and the like based on the timing when the uplink signal is received. However, when the terminal is moving, the position and timing when the terminal transmits the uplink signal are shifted from the position and timing when the terminal receives the downlink signal with the adjusted beam weights from the base station. As a result, at the position of the destination of the terminal, the downlink signal may be transmitted from the base station to the terminal with a non-optimal beam weight, and the communication performance may deteriorate.

[0004] In particular, when performing cooperative communication between a plurality of base stations or a plurality of antenna devices such as D-MIMO (Distributed Multi-Input Multi-Output), the timing of cooperation and the like may be shifted due to the movement of the terminal, and the communication performance may deteriorate.

[0005] The control device 300 according to this embodiment has a configuration that contributes to solving such problems. For example, the control device 300 learns the movement tendency of the terminal in advance, estimates the transmission path conditions at a future point in time after the terminal has moved, and determines the beam weights of multiple antennas based on the estimation results. This makes it possible to communicate with beam weights optimized for the terminal's position after it has moved, for example.

[0006] For example, one of several antennas may be the closest antenna to the terminal at the time the terminal transmits an uplink signal. However, due to the terminal's subsequent movement, the relative positions of the terminal and the multiple antennas may change, and by the time the terminal receives a downlink signal from the base station, that antenna may no longer be the closest antenna to the terminal. In such cases, the control device 300 according to this embodiment can estimate the transmission path conditions based on the relative positions of the terminal and the multiple antennas at a future point in time after the terminal has moved, and determine the beam weights of the multiple antennas. Therefore, even after the terminal has moved, communication services can be provided to the terminal by transmitting beams with optimal beam weights.

[0007] For example, the control device 300 performs the above estimation using a learning model generated by AI (Artificial Intelligence) or the like. For example, the learning model takes at least the uplink signal received by the base station from the terminal as input and outputs the transmission path status of the terminal at a future point in time.

[0008] The control device 300 according to this embodiment may be located on an information processing infrastructure located in a region. On such information processing infrastructure, for example, control of a general RAN (Radio Access Network) may be performed in conjunction with the AI ​​processing described above. By running the RAN control function on a high-performance GPU (Graphics Processing Unit) server instead of a general-purpose server, the surplus computing resources can be utilized for AI processing. Examples of AI processing include AI processing related to RAN control (sometimes referred to as RAN control AI processing) and AI processing not related to RAN control (sometimes referred to as non-RAN control AI processing).

[0009] An example of AI-based RAN control processing is RIC (RAN Intelligent Controller). RIC is a technology that uses AI to optimize RAN wireless resources and automate RAN operations. RIC includes Non-RT RIC (Non-Real Time RIC) and Near-RT RIC (Near-Real Time RIC). Non-RT RIC is sometimes called Centralized RIC. Non-RT RIC is located within an SMO (Service Management and Orchestration) that manages and orchestrates the RAN.

[0010] Non-RT RICs generate and notify policies related to RAN control and send information to Near-RT RICs. For example, Non-RT RICs generate a learning model for RAN control by performing machine learning using data collected from the RAN and send it to Near-RT RICs.

[0011] Near-RT RICs are sometimes called Distributed RICs. Compared to Non-RT RICs, Near-RT RICs are located closer to RAN nodes (RU (Radio Unit), DU (Distributed Unit), CU (Central Unit)) and perform tasks such as controlling RAN nodes and resources. Near-RT RICs perform processing that is more real-time than Non-RT RICs. For example, Near-RT RICs perform inference processing related to RAN control using learning models acquired from Non-RT RICs. RAN control AI processing is not limited to RICs.

[0012] According to one embodiment of the present invention, a control device is provided. The control device may include a learning model storage unit that stores a learning model generated using a plurality of datasets, each dataset including an uplink signal received by a base station from a target terminal and at least one of the transmission output and transmission timing of the uplink signal transmitted by the terminal, the reception quality of the uplink signal from each of the plurality of antennas, and the transmission path status between the base station and the terminal estimated based on the respective uplink signals. The learning model takes at least an uplink signal received by the base station from a target terminal as input and outputs the transmission path status between the base station and the target terminal at a future point in time. The control device may also include an estimation unit that inputs the uplink signal received by the base station from the target terminal, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal, and the reception quality of the uplink signal from each of the plurality of antennas to the learning model and obtains the transmission path status at a future point in time output from the learning model as the estimated transmission path status between the base station and the target terminal at a future point in time. The control device may include a determination unit that determines the beam weight patterns of the plurality of antennas at a future point in time based on the estimated transmission path conditions.

[0013] Any of the above-mentioned control devices may include a learning model storage unit that stores a learning model generated using a plurality of datasets, which include an uplink signal received by the second base station from a target terminal as input and the transmission path status between the first base station, the second base station, and the target terminal at a future point in time as output. The dataset is generated using a plurality of datasets, which include an uplink signal received by the second base station from a target terminal as input and the transmission path status between the first base station, the second base station, and the target terminal at a future point in time as output. The dataset is generated using a plurality of datasets, which include an uplink signal received by the second base station from a target terminal as input and the transmission path status between the first base station, the second base station, and the terminal. The dataset is generated using a plurality of datasets, which include an uplink signal received by the second base station from a target terminal as input and the transmission path status between the first base station, the second base station, and the terminal at a future point in time as output. Any of the control devices described above may include an estimation unit that inputs the uplink signal received by the first base station and the second base station from the target terminal, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal, the reception quality of the uplink signal from each of the plurality of antennas of the first base station, and the reception quality of the uplink signal from each of the plurality of antennas of the second base station to the learning model, and obtains the transmission path status at the future time output from the learning model as the estimated transmission path status at the future time between the first base station, the second base station, and the target terminal. Any of the control devices described above may include a determination unit that determines the beam weight patterns of the plurality of antennas of the first base station and the plurality of antennas of the second base station at the future time based on the estimated transmission path status.

[0014] Any of the above control devices may include a generation unit that generates the learning model using a plurality of the above datasets. Any of the above control devices may include a quality acquisition unit that acquires the service quality of communication with the target terminal using the weight pattern determined by the determination unit. Any of the above control devices may include a model update unit that updates the learning model based on the service quality acquired by the quality acquisition unit so that the service quality is improved when the transmission path condition at a future point in time output by the learning model is used as the estimated transmission path condition.

[0015] In any of the above-mentioned control devices, the dataset may further include at least one of the terminal's speed and direction of movement. In any of the above-mentioned control devices, the dataset may further include map information of a region corresponding to the base station's coverage area.

[0016] In any of the above-mentioned control devices, the dataset may further include location information related to the location from which the terminal transmitted the uplink signal. In any of the above-mentioned control devices, the estimation unit may further input information representing the current location of the target terminal to the learning model and obtain the transmission path status output from the learning model as the estimated transmission path status at a future point in time between the base station and the target terminal.

[0017] According to one embodiment of the present invention, a control method is provided. The control method may include a learning model storage step, which stores a learning model generated using a plurality of datasets, each dataset including an uplink signal received by a base station from a target terminal and an estimated transmission path status between the base station and the target terminal based on the uplink signal received by the base station from a target terminal, at least one of the uplink signal received by the base station from a terminal, at least one of the uplink signal transmission output and transmission timing of the uplink signal transmitted by the terminal, the reception quality of the uplink signal from each of the plurality of antennas, and the estimated transmission path status between the base station and the terminal based on the respective uplink signals. The control method may also include an estimation step, in which the uplink signal received by the base station from the target terminal, at least one of the uplink signal received by the target terminal and the uplink signal reception quality of each of the plurality of antennas are input to the learning model, and the transmission path status at a future point in time output from the learning model is obtained as the estimated transmission path status between the base station and the target terminal at a future point in time. The control method may include a decision step of determining the beam weight patterns of the plurality of antennas at a future point in time based on the estimated transmission path conditions.

[0018] Any of the above control methods may include a learning model storage step in which a first base station capable of performing cooperative communication using multiple antennas and a second base station capable of performing cooperative communication using multiple antennas and coordinating with the first base station to provide communication services to a terminal store a learning model that takes at least an uplink signal received by the second base station from a target terminal as input and stores the transmission path status at a future point in time between the first base station and the target terminal as output, which is generated using a plurality of datasets that include an uplink signal received from a terminal located at the first base station, at least one of the transmission output and transmission timing of the uplink signal from the terminal, the reception quality of the uplink signal from each of the plurality of antennas of the first base station, the reception quality of the uplink signal from each of the plurality of antennas of the second base station, the transmission path status between the first base station and the terminal estimated based on each of the uplink signals, and the transmission path status between the second base station and the terminal estimated based on the uplink signal received by the second base station. Any of the above control methods may include an estimation step in which the uplink signal received by the first base station and the second base station from the target terminal, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal, the reception quality of the uplink signal for each of the plurality of antennas of the first base station, and the reception quality of the uplink signal for each of the plurality of antennas of the second base station are input to the learning model, and the transmission path status for the future time output from the learning model is obtained as the estimated transmission path status for the future time between the first base station, the second base station, and the target terminal. Any of the above control methods may include a determination step in which the beam weight patterns of the plurality of antennas of the first base station and the plurality of antennas of the second base station at the future time are determined based on the estimated transmission path status.

[0019] According to one embodiment of the present invention, a program is provided for causing a computer to execute any of the control methods described above.

[0020] It should be noted that the above summary of the invention does not enumerate all the necessary features of the present invention. Furthermore, subcombinations of these features may also constitute an invention.

[0021] An example of the prior art is shown in general terms. An example of the prior art is shown in general terms. An example of system 30 is shown in general terms. An example of system 30 is shown in general terms. An example of system 30 is shown in general terms. An example of the functional configuration of the control device 300 is shown in general terms. An example of the processing flow by the control device 300 is shown in general terms. An example of system 30 is shown in general terms. An example of the hardware configuration of a computer 1200 that functions as a control device 300, an information processing platform 400, or a management platform 500 is shown in general terms.

[0022] The present invention will be described below through embodiments, but these embodiments are not intended to limit the scope of the claims. Furthermore, not all combinations of features described in the embodiments are necessarily essential to the solution of the invention.

[0023] Figures 1 and 2 schematically show an example of the prior art. In the example shown in Figures 1 and 2, a base station 100 capable of performing communication using multiple antennas forms a cell 10, and a terminal 80 is located at position A within the cell 10. In the example shown in Figures 1 and 2, the base station 100 has multiple antennas, including antenna 110 and antenna 120. The description of antennas other than antenna 110 and antenna 120 will be omitted hereafter, but those skilled in the art will understand that the other antennas may be controlled in the same way as antennas 110 and antenna 120.

[0024] In the example shown in Figure 1, terminal 80 transmits an uplink signal. Here, the uplink signal transmitted by terminal 80 that reaches antenna 110 is distinguished as uplink signal 82, and the uplink signal that reaches antenna 120 is distinguished as uplink signal 84.

[0025] The base station 100 may instruct the terminal 80 on the strength and phase of the uplink signal, and the terminal 80 may transmit the uplink signal in accordance with the instruction from the base station 100. The base station 100 may estimate the conditions of the transmission path between itself and the terminal 80 by comparing the strength and phase of the uplink signal instructed to the terminal 80 with the uplink signals 82 and 84 actually received. For example, the conditions of the transmission path refer to the degree of attenuation of radio wave strength and the degree of phase shift of radio waves in the transmission path.

[0026] The base station 100 may estimate the status of the transmission path between each of the multiple antennas and the terminal 80. For example, the base station 100 estimates the status of the transmission path between antenna 110 and terminal 80 based on the uplink signal 82 received by antenna 110. For example, the base station 100 estimates the status of the transmission path between antenna 120 and terminal 80 based on the uplink signal 84 received by antenna 120.

[0027] The base station 100 determines the beamweights of antennas 110 and 120 based on the estimated transmission path conditions. The determination of beamweights may include determining the transmission direction of the beam and determining the phase of the beam's radio waves.

[0028] Figure 2 shows the state after the base station 100 has determined the beamweight in Figure 1, and is transmitting beams 11 and 12 from antennas 110 and 120 with the determined beamweight. At this time, terminal 80 has moved from position A, where it transmitted the uplink signal in Figure 1, to position B.

[0029] In the example shown in Figure 2, beams 11 and 12 are beams formed with beam weights determined based on the transmission path conditions at the time terminal 80 is located at position A. Therefore, when terminal 80 moves to position B, the beam weights are no longer optimal.

[0030] Thus, if terminal 80 moves during the period between the time it transmits the uplink signal and the time it receives beams from multiple antennas, it may not be possible to transmit the beam to terminal 80 with the optimal beamweight. This can be particularly problematic when the terminal moves at a high speed and covers a large distance in a short time, or when the time interval for adjusting the beamweight is large.

[0031] Figures 3 and 4 schematically show an example of system 30. System 30 comprises a control device 300 and a base station 100 capable of performing communication using multiple antennas. In the example shown in Figure 3, the control device 300 is located on an information processing infrastructure 400 located in region 40. The control device 300 may be connected to the information processing infrastructure 400. The control device 300 may be located on the base station 100.

[0032] In the example shown in Figure 3, the base station 100 is connected to the information processing infrastructure 400. In the example shown in Figure 3, the base station 100 has multiple antennas, including antenna 110 and antenna 120. The description of the other antennas other than antenna 110 and antenna 120 will be omitted hereafter, but those skilled in the art will understand that the other antennas may be controlled in the same way as antennas 110 and antenna 120.

[0033] In the example shown in Figure 3, terminal 80 is located at position A within the cell 10 formed by base station 100. Base station 100 receives uplink signals from terminal 80. Here, the uplink signals transmitted by terminal 80 that reach antenna 110 are distinguished as uplink signal 82, and the uplink signals that reach antenna 120 are distinguished as uplink signal 84.

[0034] The control device 300 obtains the transmission path status of the target terminal 80 at a future point in time as an estimated transmission path status. In the example shown in Figure 3, the control device 300 obtains the transmission path status of the target terminal 80 at a future point in time based on the uplink signal transmitted by the target terminal 80 at position A. For example, the control device 300 obtains the status of the transmission path 86 between the antenna 110 and the target terminal 80 at a future point in time. For example, the control device 300 obtains the status of the transmission path 88 between the antenna 120 and the target terminal 80 at a future point in time.

[0035] The control device 300 acquires the estimated transmission path status using a learning model. The method by which the control device 300 acquires the estimated transmission path status will be described below.

[0036] The control device 300 stores a learning model generated using multiple datasets that include uplink signals received by the base station 100 from the terminal 80, at least one of the transmission output and transmission timing of uplink signals transmitted by the terminal 80, the reception quality of the uplink signals from each of the multiple antennas, and the transmission path conditions between the base station 100 and the terminal 80 estimated based on each uplink signal. The learning model takes at least the uplink signals that the base station 100 receives from the target terminal 80 as input and outputs the transmission path conditions between the base station 100 and the target terminal 80 at a future point in time.

[0037] The control device 300 may generate the learning model. Alternatively, another device different from the control device 300 may generate the learning model, and the control device 300 may store the model. Hereafter, the case in which the control device 300 generates the learning model will be explained as an example.

[0038] In the example shown in Figure 3, for example, by comparing the transmission output of the uplink signal transmitted by terminal 80 from position A with the uplink signal 82 received by antenna 110 and considering propagation loss, the approximate distance between antenna 110 and terminal 80 can be determined. Similarly, by comparing the transmission output of the uplink signal transmitted by terminal 80 from position A with the uplink signal 84 received by antenna 120 and considering propagation loss, the approximate distance between antenna 120 and terminal 80 can be determined. In the same way, the approximate distance between terminal 80 and the other antennas of base station 100 can also be determined. From the relationship of these distances, it can be determined that terminal 80 is located at position A.

[0039] In the example shown in Figure 3, for example, by comparing the transmission timing of the uplink signal sent by terminal 80 from position A with the uplink signal 82 received by antenna 110, and taking into account phase differences, the approximate distance between antenna 110 and terminal 80 can be determined. Similarly, by comparing the transmission timing of the uplink signal sent by terminal 80 from position A with the uplink signal 84 received by antenna 120, the approximate distance between antenna 120 and terminal 80 can be determined. In the same way, the approximate distance between terminal 80 and the other antennas of base station 100 can also be determined. From the relationship of these distances, it can be determined that terminal 80 is located at position A.

[0040] In other words, the base station 100 can acquire data estimating the position of terminal 80 based on the uplink signal received from terminal 80 and at least one of the transmission output and transmission timing of the uplink signal transmitted by terminal 80. For example, the control device 300 can learn the movement trends of terminal 80 by acquiring and learning similar data in a time series. This allows the control device 300 to estimate the position of terminal 80 at a future point in time.

[0041] For example, the control device 300 stores a data set in which the reception quality of the uplink signal 82 by the antenna 110, the transmission path status between the antenna 110 and the terminal 80 estimated using the same, the reception quality of the uplink signal 84 by the antenna 120, and the transmission path status between the antenna 120 and the terminal 80 estimated using the same are recorded together with the data. For example, the control device 300 generates a learning model that uses at least the uplink signal received by the base station 100 from the target terminal 80 as an input and outputs the future transmission path status between the base station 100 and the target terminal 80 by learning using the data set.

[0042] The data set may further include position information related to the position where the terminal 80 transmitted the uplink signal. The control device 300 may obtain position information related to the position where the terminal 80 transmitted the uplink signal based on the GPS position information of the terminal 80 and include it in the data set. The control device 300 may estimate the position information of the terminal 80 from the data of the position of the terminal 80 derived by the above method and include it in the data set.

[0043] The data set may further include at least one of the moving speed and the moving direction of the terminal 80. As a result, for example, the position of the terminal 80 at a future time point can be estimated based on the moving speed and the moving direction of the terminal 80, so that the accuracy of the position estimation of the terminal 80 is improved. Subsequently, the accuracy of the estimated transmission path status is improved, and a more appropriate beam weight pattern can be determined.

[0044] The control device 300 may obtain the moving speed and the moving direction of the terminal 80 based on the time-series change of the GPS position information of the terminal 80 and include them in the data set. The control device 300 may estimate the moving speed and the moving direction of the terminal 80 from the time-series change of the data of the position of the terminal 80 derived by the above method and include them in the data set.

[0045] The dataset may further include map information for the area corresponding to the coverage area of ​​the base station 100. In the example shown in Figure 3, the dataset may include map information for the area 40 corresponding to cell 10, which is the coverage area of ​​the base station 100. This allows for a more specific estimation of possible routes that the terminal 80 may take, based on the map information. For example, if the terminal 80 is continuously moving along a railway line on the map, it can be estimated that there is a high probability that the terminal 80's future location will be on that railway line.

[0046] The control device 300 inputs the uplink signal received by the base station 100 from the target terminal 80, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal 80, and the reception quality of the uplink signal from each of the multiple antennas into the learning model, and acquires the output transmission path status at a future point in time as the estimated transmission path status at a future point in time between the base station 100 and the target terminal 80. In the example shown in Figure 3, the control device 300 inputs the uplink signal 82 received by antenna 110, the reception quality of uplink signal 82, the uplink signal 84 received by antenna 120, the reception quality of uplink signal 84, and at least one of the transmission output and transmission timing of the uplink signal into the learning model. As a result, the learning model outputs the transmission path status at a future point in time. For example, the learning model outputs the status of the transmission path 86 at a future point in time between antenna 110 and the target terminal 80. For example, the learning model outputs the status of the transmission path 88 at a future point in time between antenna 120 and the target terminal 80.

[0047] The control device 300 may further input information representing the current location of the target terminal 80 into the learning model and acquire the transmission path status output from the learning model as an estimated transmission path status at a future point in time between the base station 100 and the target terminal 80. The method by which the control device 300 acquires information representing the current location of the target terminal 80 may be the aforementioned method based on GPS location information, the method based on the transmission output of the uplink signal, and the method based on the transmission timing of the uplink signal.

[0048] The control device 300 acquires the future transmission path status output from the learning model as the estimated transmission path status at a future time between the base station 100 and the target terminal 80. Based on the estimated transmission path status, the control device 300 determines the weight patterns of the beams of a plurality of antennas at a future time. For example, the control device 300 determines the beam weight of the antenna 110 based on the status of the transmission path 86 at a future time. For example, the control device 300 determines the beam weight of the antenna 120 based on the status of the transmission path 88 at a future time.

[0049] FIG. 4 shows a state where, after the control device 300 determines the weight pattern of the beam in FIG. 3, the beams 11 and 12 are transmitted to the antennas 110 and 120 with the determined beam weights. At this time, the terminal 80 is moving from the position A when the uplink signal was transmitted in FIG. 3 to the position B.

[0050] In this way, since the control device 300 causes the base station 100 to transmit the beams 11 and 12 with the weight pattern of the beam determined based on the estimated transmission path of the terminal 80 at a future time, even when the target terminal 80 moves, the beam can be transmitted to the terminal 80 with the optimal beam weight pattern. As a result, for example, compared with the prior art shown in FIGS. 1 and 2, the communication efficiency between the base station 100 and the target terminal 80 is improved.

[0051] The control device 300 may update the learning model. For example, the control device 300 actually communicates with the target terminal 80 using the weight pattern determined by the above method and acquires the service quality of the communication. For example, the control device 300 updates the learning model so that the service quality improves when the future transmission path status output by the learning model is regarded as the estimated transmission path status based on the acquired service quality.

[0052] For example, the control device 300 updates its learning model so that the degree of phase shift in the future transmission path between the antenna 110 and the terminal 80 is adjusted based on the actual communication service quality. This adjusts the beamweight of the antenna 110 determined by the control device 300, and adjusts the phase of the radio waves of beam 11. As a result, if the service quality deteriorates due to a phase difference between the radio waves of beam 11 and beam 12 reaching the target terminal 80, the phase of beam 11 can be adjusted to approach the phase of beam 12, thereby improving the service quality.

[0053] Figures 5 and 6 schematically show an example of system 30. The examples shown in Figures 5 and 6 will primarily be explained in terms of differences from the examples shown in Figures 3 and 4.

[0054] In the example shown in Figure 5, the system 30 comprises a control device 300, a base station 100, and a base station 200. In the example shown in Figure 5, the control device 300 is located on the information processing infrastructure 400. The control device 300 may be connected to the information processing infrastructure 400. The control device 300 may be located on the base station 100 or the base station 200.

[0055] Base stations 100 and 200 are capable of performing coordinated communication using multiple antennas. In the example shown in Figure 5, base station 200 provides communication services to terminal 80 in cooperation with base station 100. Coordinated communication may include D-MIMO. Coordinated communication may include CoMP (Coordinated Multi-Point Transmission). Base station 100 may be an example of a first base station, and base station 200 may be an example of a second base station.

[0056] In the example shown in Figure 5, the base station 200 has multiple antennas, including antenna 210 and antenna 220. The description of the other antennas besides antenna 210 and antenna 220 will be omitted hereafter, but those skilled in the art will understand that the other antennas may be controlled in the same way as antennas 210 and antenna 220.

[0057] In the example shown in Figure 5, the multiple antennas of base station 100 may be located in the same place or in different places. In the example shown in Figure 5, the multiple antennas of base station 200 may be located in the same place or in different places.

[0058] In the example shown in Figure 5, the multiple antennas may perform coordinated operation as a whole by each independently transmitting and receiving signals. For example, some or all of antennas 110, 120, 210, and 220 may perform coordinated operation as a whole by independently transmitting and receiving signals. For example, the coordinated control performed by the multiple antennas may include D-MIMO. For example, the coordinated control performed by the multiple antennas may include CoMP.

[0059] In the example shown in Figure 5, terminal 80 is located at position A, which is within cell 10 formed by base station 100 and within cell 20 formed by base station 200. In the example shown in Figure 5, terminal 80 is located at base station 100. Base stations 100 and 200 receive uplink signals from terminal 80. Here, the uplink signals transmitted by terminal 80 are distinguished as follows: uplink signal 82 reaches antenna 110, uplink signal 84 reaches antenna 120, uplink signal 92 reaches antenna 210, and uplink signal 94 reaches antenna 220.

[0060] The control device 300 obtains the transmission path status of the target terminal 80 at a future point in time as an estimated transmission path status. In the example shown in Figure 5, the control device 300 obtains the transmission path status of the target terminal 80 at a future point in time based on the uplink signal transmitted by the target terminal 80 at position A.

[0061] For example, the control device 300 acquires the status of the transmission line 86 at a future point in time between the antenna 110 and the target terminal 80. For example, the control device 300 acquires the status of the transmission line 88 at a future point in time between the antenna 120 and the target terminal 80. For example, the control device 300 acquires the status of the transmission line 96 at a future point in time between the antenna 210 and the target terminal 80. For example, the control device 300 acquires the status of the transmission line 98 at a future point in time between the antenna 220 and the target terminal 80.

[0062] The control device 300 acquires the estimated transmission path status using a learning model. The method by which the control device 300 acquires the estimated transmission path status will be described below.

[0063] The control device 300 stores a learning model generated using multiple datasets, each of which includes the uplink signal received from the terminal 80, at least one of the transmission output and transmission timing of the uplink signal from the terminal 80, the reception quality of the uplink signal from each of the multiple antennas of the base station 100, the reception quality of the uplink signal from each of the multiple antennas of the base station 200, and the transmission path conditions between the base station 100 and the base station 200 and the terminal 80 estimated based on each uplink signal. The uplink signal received from the terminal 80 may include the uplink signal received by the base station 100 and the uplink signal received by the base station 200.

[0064] The learning model takes at least the uplink signal received by the base station 200 from the target terminal 80 as input and outputs the transmission path status at a future point in time between the base station 100 and the base station 200 and the target terminal 80. In the example shown in Figure 5, the learning model takes at least the uplink signal 92 and the uplink signal 94 as input and outputs the status of the transmission path 86 at a future point in time, the status of the transmission path 88 at a future point in time, the status of the transmission path 96 at a future point in time, and the status of the transmission path 98 at a future point in time.

[0065] The control device 300 may generate the learning model, or another device different from the control device 300 may generate the learning model and the control device 300 may store the model, as in the example shown in Figure 3.

[0066] The control device 300 inputs the uplink signals received by base stations 100 and 200 from the target terminal 80, at least one of the transmission output and transmission timing of the uplink signals transmitted by the target terminal 80, the reception quality of the uplink signals from each of the multiple antennas of base station 100, and the reception quality of the uplink signals from each of the multiple antennas of base station 200 into a learning model, and acquires the output transmission path status at a future point in time as the estimated transmission path status at a future point in time between base stations 100 and 200 and the end of the target terminal 80. In the example shown in Figure 5, the control device 300 inputs the following to the learning model: the uplink signal 82 received by antenna 110, the reception quality of the uplink signal 82, the uplink signal 84 received by antenna 120, the reception quality of the uplink signal 84, the uplink signal 92 received by antenna 210, the reception quality of the uplink signal 92, the uplink signal 94 received by antenna 220, the reception quality of the uplink signal 94, and at least one of the transmission output and transmission timing of the uplink signal.

[0067] This allows the learning model to output the transmission path status at a future point in time. For example, the learning model outputs the status of transmission path 86 at a future point in time between antenna 110 and target terminal 80. For example, the learning model outputs the status of transmission path 88 at a future point in time between antenna 120 and target terminal 80. For example, the learning model outputs the status of transmission path 96 at a future point in time between antenna 210 and target terminal 80. For example, the learning model outputs the status of transmission path 98 at a future point in time between antenna 220 and target terminal 80.

[0068] The control device 300 acquires the transmission path conditions at a future point in time output from the learning model as the estimated transmission path conditions at a future point in time between the base stations 100 and 200 and the target terminal 80. Based on the estimated transmission path conditions, the control device 300 determines the beam weight patterns of multiple antennas at a future point in time.

[0069] For example, the control device 300 determines the beamweight of antenna 110 based on the condition of transmission line 86 at a future point in time. For example, the control device 300 determines the beamweight of antenna 120 based on the condition of transmission line 88 at a future point in time. For example, the control device 300 determines the beamweight of antenna 210 based on the condition of transmission line 96 at a future point in time. For example, the control device 300 determines the beamweight of antenna 120 based on the condition of transmission line 98 at a future point in time.

[0070] Figure 6 shows the state after the control device 300 has determined the beam weight pattern in Figure 3, instructing antennas 110 and 120 to transmit beams 11 and 12, and antennas 210 and 220 to transmit beams 21 and 22. At this time, terminal 80 has moved from position A, where it transmitted the uplink signal in Figure 5, to position B.

[0071] In this way, the control device 300 causes the base station 100 to transmit beams 11 and 12 and the base station 200 to transmit beams 21 and 22 using a beam weight pattern determined based on the estimated transmission path of the terminal 80 at a future point in time. Therefore, even if the target terminal 80 moves, coordinated beams can be transmitted to the terminal 80 with an optimal beam weight pattern. As a result, the efficiency of coordinated communication between the base stations 100 and 200 and the target terminal 80 is improved compared to, for example, conventional coordinated communication that does not consider the movement of the terminal between uplink signal transmission and beam weight determination.

[0072] The control device 300 may actually communicate with the target terminal 80 using the weight pattern determined by the method described above, and update the learning model based on the quality of service of that communication, as is common to the examples shown in Figures 3 and 4.

[0073] Figure 7 schematically shows an example of the functional configuration of the control device 300. In the example shown in Figure 7, the control device 300 includes a learning model generation unit 310, a learning model storage unit 320, an estimation unit 330, a determination unit 340, a quality acquisition unit 350, and a model update unit 360. It is not necessarily required that the control device 300 include all of these.

[0074] The learning model generation unit 310 generates various learning models. The learning model generation unit 310 may store various datasets. The learning model generation unit 310 may store multiple datasets. The learning model generation unit 310 may generate a learning model using multiple datasets.

[0075] The learning model generation unit 310 stores a dataset that includes, for example, the uplink signal received by the base station 100 from the terminal 80, at least one of the transmission output and transmission timing of the uplink signal transmitted by the terminal 80, the reception quality of the uplink signal from each of the multiple antennas, and the transmission path conditions between the base station 100 and the terminal 80 estimated based on each uplink signal. The learning model generation unit 310 may generate a learning model using multiple such datasets. The learning model may be one that takes the uplink signal received by the base station 100 from the target terminal 80 as input and outputs the transmission path conditions between the base station 100 and the target terminal 80 at a future point in time.

[0076] The learning model generation unit 310 stores a dataset that includes, for example, the uplink signal received from the terminal 80, at least one of the transmission output and transmission timing of the uplink signal from the terminal 80, the reception quality of the uplink signal from each of the multiple antennas of the base station 100, the reception quality of the uplink signal from each of the multiple antennas of the base station 200, and the transmission path conditions between the base station 100 and the base station 200 and the terminal 80 estimated based on each uplink signal. The learning model generation unit 310 may generate a learning model using multiple such datasets. The learning model may be one that takes the uplink signal received by the base station 200 from the target terminal 80 as input and outputs the transmission path conditions between the base station 100 and the base station 200 and the target terminal 80 at a future point in time.

[0077] The learning model storage unit 320 stores various learning models. For example, the learning model storage unit 320 stores a learning model that takes the uplink signal that the base station 100 receives from the terminal 80 as input and outputs the transmission path status between the base station 100 and the target terminal 80 at a future point in time. This model is generated using multiple datasets that include the uplink signal that the base station 100 receives from the terminal 80, at least one of the transmission output and transmission timing of the uplink signal transmitted by the terminal 80, the reception quality of the uplink signal from each of the multiple antennas, and the transmission path status between the base station 100 and the terminal 80 estimated based on each uplink signal.

[0078] For example, the learning model storage unit 320 stores a learning model that takes as input the uplink signal that base station 200 receives from the target terminal 80, and outputs the transmission path status at a future point in time between base station 100 and base station 200 and the target terminal 80. This model is generated using a plurality of datasets that include the uplink signal received from terminal 80, at least one of the transmission output and transmission timing of the uplink signal from terminal 80, the reception quality of the uplink signal from each of the multiple antennas of base station 100, the reception quality of the uplink signal from each of the multiple antennas of base station 200, and the transmission path status between base station 100 and base station 200 and terminal 80 estimated based on each uplink signal.

[0079] The estimation unit 330 acquires the estimated transmission path status at a future point in time between the base station and the terminal 80. For example, the estimation unit 330 inputs the uplink signal received by the base station 100 from the target terminal 80, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal 80, and the reception quality of the uplink signal from each of the multiple antennas into a learning model, and acquires the output transmission path status at a future point in time as the estimated transmission path status at a future point in time between the base station 100 and the target terminal 80.

[0080] The estimation unit 330 may further input information representing the current location of the target terminal 80 into the learning model, and obtain the transmission path status output from the learning model as the estimated transmission path status at a future point in time between the base station 100 and the target terminal 80.

[0081] For example, the estimation unit 330 inputs the uplink signals received by base stations 100 and 200 from the target terminal 80, at least one of the transmission output and transmission timing of the uplink signals transmitted by the target terminal 80, the reception quality of the uplink signals from each of the multiple antennas of base station 100, and the reception quality of the uplink signals from each of the multiple antennas of base station 200 into a learning model, and obtains the output transmission path status at a future point in time as the estimated transmission path status at a future point in time between base stations 100 and 200 and the end of the target terminal 80.

[0082] The determination unit 340 determines the beam weight patterns of multiple antennas. For example, the determination unit 340 determines the beam weight patterns of multiple antennas at a future point in time based on the estimated transmission path conditions. The determination unit 340 may use the determined beam weight patterns to control the multiple antennas of the base station and provide communication services to the target terminal 80.

[0083] The quality acquisition unit 350 acquires the service quality of communication between the base station and the terminal 80. For example, the quality acquisition unit 350 acquires the service quality of communication with the target terminal 80 using the weight pattern determined by the determination unit 340.

[0084] The model update unit 360 updates various learning models. For example, based on the service quality acquired by the quality acquisition unit 350, the model update unit 360 updates the learning model so that the service quality improves when the transmission path conditions at a future point in time output by the learning model are used as the estimated transmission path conditions.

[0085] Figure 8 schematically shows an example of the processing flow by the control device 300. In step 102 (steps may be abbreviated as S), the learning model generation unit 310 takes the uplink signal from the terminal 80 as input and generates a learning model that outputs the transmission path conditions at a future point in time.

[0086] In S104, the learning model storage unit 320 stores the learning model generated by the learning model generation unit 310 in S102. The learning model storage unit 320 may also acquire and store a learning model generated by another device different from the learning model generation unit 310.

[0087] In S106, the estimation unit 330 inputs the uplink signal etc. from the target terminal 80 into the learning model stored in the learning model storage unit 320 in S104 to obtain the estimated transmission path status.

[0088] In S108, the determination unit 340 determines the beam weight patterns of the multiple antennas of the base station 100 based on the estimated transmission path conditions acquired by the estimation unit 330 in S106. Here, the determination unit 340 uses the determined beam weight patterns to cause the multiple antennas of the base station to transmit beams.

[0089] In S110, the quality acquisition unit 350 acquires the service quality of communication with the target terminal 80 using the weight pattern determined by the determination unit 340 in S108.

[0090] In S112, the model update unit 360 updates the learning model based on the service quality acquired by the quality acquisition unit 350 in S110.

[0091] Figure 9 schematically shows an example of system 30. In the example shown in Figure 9, system 30 comprises a management infrastructure 500 and a plurality of information processing infrastructures 400. The management infrastructure 500 may manage the plurality of information processing infrastructures 400. The management infrastructure 500 may be a data center that manages the plurality of information processing infrastructures 400. The management infrastructure 500 may be configured as a plurality of devices. The management infrastructure 500 may be implemented on a virtualization infrastructure consisting of a plurality of devices. The management infrastructure 500 may be implemented as a single device. That is, the management infrastructure 500 may be a management device.

[0092] In the example shown in Figure 9, the configuration of each of the multiple information processing infrastructures 400 is the same as in the examples shown in Figures 3 to 6. Note that in Figure 9, the control device 300, terminal 80, base station antennas, uplink signals, transmission lines, and beams are omitted from the description.

[0093] In the example shown in Figure 9, the control device 300 may be located on the management infrastructure 500. In this case, the control device 300 may be further located on at least one of the multiple information processing infrastructures 400. Alternatively, the control device 300 may not be located on the management infrastructure 500, but on each of the multiple information processing infrastructures 400.

[0094] The management infrastructure 500 may be called the Core Brain, and the information processing infrastructure 400 may be called the Regional Brain. Note that Figure 9 illustrates a case where a single-layer information processing infrastructure 400 is located below the management infrastructure 500, but this is not the only example. The information processing infrastructure 400 may have multiple layers. For example, if two layers of information processing infrastructure 400 are located below the management infrastructure 500, the management infrastructure 500 may be called the Core Brain, the lower layer of information processing infrastructure 400 may be called the Regional Brain, and the further lower layer of information processing infrastructure 400 may be called the Sub-Regional Brain. In this case, the control device 300 may be located in the Regional Brain. In any of the above cases, the control device 300 may be connected to the information processing infrastructure 400 or management infrastructure 500 instead of being located on them.

[0095] Figure 10 schematically shows an example of the hardware configuration of a computer 1200 that functions as a control device 300, an information processing platform 400, or a management platform 500. A program installed on the computer 1200 can cause the computer 1200 to function as one or more "parts" of the apparatus according to this embodiment, or to cause the computer 1200 to execute operations associated with the apparatus according to this embodiment or such one or more "parts", and / or to cause the computer 1200 to execute a process or a stage of such process according to this embodiment. Such a program may be executed by the CPU 1212 to cause the computer 1200 to execute specific operations associated with some or all of the blocks in the flowcharts and block diagrams described herein.

[0096] The computer 1200 according to this embodiment includes a CPU 1212, a GPU 1213, a RAM 1214, and a graphics controller 1216, which are interconnected by a host controller 1210. The computer 1200 also includes input / output units such as a communication interface 1222, a storage device 1224, a DVD drive, and an IC card drive, which are connected to the host controller 1210 via an input / output controller 1220. The DVD drive may be a DVD-ROM drive and a DVD-RAM drive, etc. The storage device 1224 may be a hard disk drive and a solid-state drive, etc. The computer 1200 also includes legacy input / output units such as a ROM 1230 and a keyboard, which are connected to the input / output controller 1220 via an input / output chip 1240.

[0097] The CPU 1212 operates according to the programs stored in the ROM 1230 and RAM 1214, thereby controlling each unit. The graphics controller 1216 acquires the image data generated by the CPU 1212 and stores it in the frame buffer provided in the RAM 1214 or within itself, so that the image data is displayed on the display device 1218.

[0098] The communication interface 1222 communicates with other electronic devices via a network. The storage device 1224 stores programs and data used by the CPU 1212 in the computer 1200. The DVD drive reads programs or data from a DVD-ROM or the like and provides them to the storage device 1224. The IC card drive reads programs and data from an IC card and / or writes programs and data to an IC card.

[0099] The ROM 1230 stores boot programs and / or hardware-dependent programs of the computer 1200, which are executed by the computer 1200 when activated. The input / output chip 1240 may also connect various input / output units to the input / output controller 1220 via USB ports, parallel ports, serial ports, keyboard ports, mouse ports, etc.

[0100] The program is provided on a computer-readable storage medium such as a DVD-ROM or IC card. The program is read from the computer-readable storage medium and installed on a storage device 1224, RAM 1214, or ROM 1230, which are examples of computer-readable storage media, and executed by the CPU 1212. The information processing described within these programs is read by the computer 1200, resulting in coordination between the program and the various types of hardware resources described above. The apparatus or method may be configured to realize the operation or processing of information in accordance with the use of the computer 1200.

[0101] For example, when communication is performed between a computer 1200 and an external device, the CPU 1212 may execute a communication program loaded into the RAM 1214 and, based on the processing described in the communication program, instruct the communication interface 1222 to perform communication processing. Under the control of the CPU 1212, the communication interface 1222 reads transmission data stored in a transmission buffer area provided in a recording medium such as the RAM 1214, storage device 1224, DVD-ROM, or IC card, transmits the read transmission data to the network, or writes received data received from the network to a reception buffer area or the like provided on the recording medium.

[0102] Furthermore, the CPU 1212 may read all or necessary parts of a file or database stored on an external recording medium such as a storage device 1224, a DVD drive (DVD-ROM), or an IC card into the RAM 1214, and perform various types of processing on the data in the RAM 1214. The CPU 1212 may then write the processed data back to the external recording medium.

[0103] Various types of information, such as various types of programs, data, tables, and databases, may be stored on the recording medium and subjected to information processing. The CPU 1212 may perform various types of processing on the data read from the RAM 1214, including various types of operations, information processing, conditional judgments, conditional branching, unconditional branching, information retrieval / replacement, etc., as described throughout this disclosure and specified by the program instruction sequence, and write the results back to the RAM 1214. The CPU 1212 may also retrieve information in files, databases, etc., within the recording medium. For example, if a plurality of entries are stored in the recording medium, each having an attribute value of a first attribute associated with an attribute value of a second attribute, the CPU 1212 may search among the plurality of entries for an entry that matches the specified condition for the attribute value of the first attribute, read the attribute value of the second attribute stored in that entry, and thereby obtain the attribute value of the second attribute associated with the first attribute that satisfies a predetermined condition.

[0104] The program or software module described above may be stored on or near the computer 1200 in a computer-readable storage medium. Alternatively, a recording medium such as a hard disk or RAM provided within a server system connected to a dedicated communication network or the Internet can be used as a computer-readable storage medium, thereby providing the program to the computer 1200 via the network.

[0105] In this embodiment, blocks in the flowchart and block diagram may represent a stage in a process in which an operation is performed or a "part" of a device that has the role of performing an operation. A particular stage and "part" may be implemented by a dedicated circuit, a programmable circuit supplied with computer-readable instructions stored on a computer-readable storage medium, and / or a processor supplied with computer-readable instructions stored on a computer-readable storage medium. The dedicated circuit may include digital and / or analog hardware circuits, and may include integrated circuits (ICs) and / or discrete circuits. The programmable circuit may include reconfigurable hardware circuits, such as field-programmable gate arrays (FPGAs) and programmable logic arrays (PLAs), which include logical AND, logical OR, exclusive OR, negated AND, negated OR, and other logical operations, flip-flops, registers, and memory elements.

[0106] A computer-readable storage medium may include any tangible device capable of storing instructions to be executed by a suitable device, and as a result, a computer-readable storage medium having instructions stored therein will comprise a product that includes instructions that can be executed to create means for performing operations specified in a flowchart or block diagram. Examples of computer-readable storage media may include electronic storage media, magnetic storage media, optical storage media, electromagnetic storage media, semiconductor storage media, etc. More specific examples of computer-readable storage media may include floppy disks, diskettes, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), electrically erasable programmable read-only memory (EEPROM), static random access memory (SRAM), compact disk read-only memory (CD-ROM), digital multipurpose disk (DVD), Blu-ray® disk, memory stick, integrated circuit card, etc.

[0107] Computer-readable instructions may include assembler instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk®, Java®, C++, and conventional procedural programming languages ​​such as the C programming language or similar programming languages.

[0108] Computer-readable instructions may be provided locally or via a wide area network (WAN) such as a local area network (LAN) or the internet to a processor or programmable circuit of a general-purpose computer, a special-purpose computer, or another programmable data processing device, so that the processor or programmable circuit of the programmable data processing device, such as a computer, may execute the computer-readable instructions to generate means for performing operations specified in a flowchart or block diagram. Here, the computer may be a PC (personal computer), a tablet computer, a smartphone, a workstation, a server computer, a general-purpose computer, or a special-purpose computer, and may also be a computer system in which multiple computers are connected. Such a computer system in which multiple computers are connected is also called a distributed computing system and is a computer in a broad sense. In a distributed computing system, multiple computers execute a program collectively by each computer executing a part of the program and passing data during program execution between computers as needed.

[0109] Examples of processors include computer processors, central processing units (CPUs), processing units, microprocessors, digital signal processors, controllers, and microcontrollers. A computer may have one or more processors. In a multiprocessor system with multiple processors, each processor executes a portion of the program, and the processors collectively execute the program by passing program execution data between them as needed. For example, in the execution of multitasks, each of the multiple processors may execute a portion of each task in small chunks by switching tasks at each time slice. In this case, which part of a program each processor executes changes dynamically. Which part of a program each of the multiple processors executes may also be statically determined by multiprocessor-aware programming.

[0110] The invention according to this embodiment allows for the appropriate adjustment of the beam weight patterns of multiple antennas at a base station in response to the movement of a terminal, thereby enabling the provision of communication services to the terminal. This provides a more efficient and comfortable communication infrastructure. Consequently, it can contribute to achieving at least one of the Sustainable Development Goals (SDGs): Goal 7 "Affordable and Clean Energy," Goal 9 "Industry, Innovation and Infrastructure," and Goal 11 "Sustainable Cities and Communities."

[0111] Although the present invention has been described above using embodiments, the technical scope of the present invention is not limited to the scope described in the above embodiments. It will be apparent to those skilled in the art that various modifications or improvements can be made to the above embodiments. It will be clear from the claims that such modified or improved forms may also be included in the technical scope of the present invention.

[0112] It should be noted that the execution order of operations, procedures, steps, and stages in the devices, systems, programs, and methods shown in the claims, specifications, and drawings is not explicitly stated as "before" or "prior to," and that these can be performed in any order unless the output of a previous operation is used in a later operation. Even if the operation flow in the claims, specifications, and drawings is described using phrases such as "first," and "next," for convenience, this does not mean that it is mandatory to perform the operations in that order.

[0113] 10 Cell, 11 Beam, 12 Beam, 20 Cell, 21 Beam, 22 Beam, 30 System, 40 Region, 80 Terminal, 82 Uplink signal, 84 Uplink signal, 86 Transmission path, 88 Transmission path, 92 Uplink signal, 94 Uplink signal, 96 Transmission path, 98 Transmission path, 100 Base station, 110 Antenna, 120 Antenna, 200 Base station, 210 Antenna, 220 Antenna, 300 Control device, 310 Learning model generation unit, 320 Learning model storage unit, 330 Estimation unit, 340 Decision unit, 350 Quality acquisition unit, 360 Model update unit, 400 Information processing infrastructure, 500 Management infrastructure, 1200 Computer, 1210 Host controller, 1212 CPU, 1213 GPU, 1214 RAM, 1216 Graphics controller, 1218 Display device, 1220 Input / Output controller, 1222 Communication interface, 1224 Storage device, 1230 ROM, 1240 Input / Output chip

Claims

1. A control device comprising: a learning model storage unit that stores a learning model generated using a plurality of datasets including an uplink signal received by a base station from a target terminal and an output of the transmission path status at a future point in time between the base station and the target terminal, the uplink signal received by the base station from a target terminal and an output of the transmission path status at a future point in time between the base station and the target terminal, the uplink signal received by the base station from the target terminal and an output of an uplink signal received by the base station from a target terminal and an output of an uplink signal at a future point in time between the base station and the target terminal; an estimation unit that inputs the uplink signal received by the base station from the target terminal and an output of the uplink signal received by the target terminal and an output of the uplink signal received by the target terminal and an output of the uplink signal received by the plurality of antennas to the learning model and acquires the transmission path status at a future point in time output from the learning model as an estimated transmission path status at a future point in time between the base station and the target terminal; and a determination unit that determines the beam weight pattern of the plurality of antennas at the future point in time based on the estimated transmission path status.

2. A learning model storage unit stores a learning model that stores a learning model generated using a plurality of datasets, which include an uplink signal received by the second base station from a target terminal as input and the transmission path status between the first base station, the second base station, and the target terminal at a future point in time as output, and which takes an uplink signal received by the second base station from a target terminal as input and outputs the transmission path status between the first base station, the second base station, and the target terminal as output, the learning model stored using a plurality of datasets which include an uplink signal received by the second base station from a target terminal, at least one of the transmission output and transmission timing of the uplink signal from the terminal, the reception quality of the uplink signal from each of the plurality of antennas of the first base station, the reception quality of the uplink signal from each of the plurality of antennas of the second base station, and the transmission path status between the first base station, the second base station, and the terminal estimated based on the respective uplink signals. A control device comprising: an estimation unit that inputs the uplink signal received by the first base station and the second base station from the target terminal, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal, the reception quality of the uplink signal of each of the plurality of antennas of the first base station and the reception quality of the uplink signal of each of the plurality of antennas of the second base station to the learning model and acquires the transmission path status of the future time output from the learning model as the estimated transmission path status of the future time between the first base station, the second base station and the target terminal; and a determination unit that determines the beam weight patterns of the plurality of antennas of the first base station and the plurality of antennas of the second base station at the future time based on the estimated transmission path status.

3. The control device according to claim 1 or 2, comprising a generation unit that generates the learning model using a plurality of the aforementioned datasets.

4. The control device according to any one of claims 1 to 3, further comprising: a quality acquisition unit that acquires the service quality of communication with the target terminal using the weight pattern determined by the determination unit; and a model update unit that updates the learning model based on the service quality acquired by the quality acquisition unit so that the service quality is improved when the transmission path condition at a future point in time output by the learning model is used as the estimated transmission path condition.

5. The control device according to any one of claims 1 to 4, wherein the dataset further includes at least one of the moving speed and moving direction of the terminal.

6. The control device according to claim 1, wherein the dataset further includes map information of a region corresponding to the coverage area of ​​the base station.

7. The control device according to claim 1 or 6, wherein the dataset further includes location information related to the location from which the terminal transmitted the uplink signal, and the estimation unit further inputs information representing the current location of the target terminal to the learning model, and obtains the transmission path status output from the learning model as an estimated transmission path status at a future point in time between the base station and the target terminal.

8. A control method comprising: a learning model storage step of storing a learning model that takes at least the uplink signal received by the base station from a target terminal as input and outputs the transmission path status at a future point in time between the base station and the target terminal, generated using a plurality of datasets including an uplink signal received by the base station from a terminal, at least one of the transmission output and transmission timing of the uplink signal transmitted by the terminal, the reception quality of the uplink signal of each of the plurality of antennas, and the transmission path status between the base station and the terminal estimated based on the respective uplink signals; an estimation step of inputting the uplink signal received by the base station from the target terminal, at least one of the transmission output and transmission timing of the uplink signal received by the target terminal, and the reception quality of the uplink signal of each of the plurality of antennas to the learning model and obtaining the transmission path status at a future point in time output from the learning model as the estimated transmission path status at a future point in time between the base station and the target terminal; and a determination step of determining the beam weight pattern of the plurality of antennas at the future point in time based on the estimated transmission path status.

9. A learning model storage stage stores a learning model that stores, which is generated using a plurality of datasets, each dataset including an uplink signal received by the second base station from a target terminal as input and the transmission path status between the first base station and the target terminal at a future point in time as output, and which is generated using a plurality of datasets including an uplink signal received by the first base station from a terminal located at the first base station, at least one of the transmission output and transmission timing of the uplink signal from the terminal, the reception quality of the uplink signal from each of the plurality of antennas of the first base station, the reception quality of the uplink signal from each of the plurality of antennas of the second base station, the transmission path status between the first base station and the terminal estimated based on each of the uplink signals, and the transmission path status between the second base station and the terminal estimated based on the uplink signal received by the second base station, the first base station capable of performing cooperative communication using a plurality of antennas and providing communication services to a terminal in cooperation with the first base station. A control method comprising: an estimation step of inputting the uplink signal received by the first base station and the second base station from the target terminal, at least one of the transmission output and transmission timing of the uplink signal transmitted by the target terminal, the reception quality of the uplink signal of each of the plurality of antennas of the first base station and the reception quality of the uplink signal of each of the plurality of antennas of the second base station into the learning model, and obtaining the transmission path status of the future time output from the learning model as the estimated transmission path status of the future time between the first base station, the second base station and the target terminal; and a determination step of determining the beam weight patterns of the plurality of antennas of the first base station and the plurality of antennas of the second base station at the future time based on the estimated transmission path status.

10. A program for causing a computer to perform the control method described in claim 8 or 9.

Images