Information processing device, terminal device, information processing method, and program

Abstract

We propose an information processing device, a terminal device, an information processing method, and a program that can further improve the estimation accuracy of the estimation model used to determine communication parameters. [Solution] The information processing device of the present disclosure includes a control unit. The control unit generates an estimation model for estimating data on communication characteristics in a second area using data on communication characteristics in a first area. The control unit generates an estimation model using data on communication characteristics in one or more first areas similar to the second area.

Description

[Technical Field]

[0001] This disclosure relates to an information processing device, a terminal device, an information processing method, and a program. [Background technology]

[0002] In wireless communication, techniques are known that achieve more optimal communication by appropriately controlling wireless resources and communication parameters. For example, optimal communication can be achieved by adaptively controlling communication parameters according to the conditions of the propagation path between the base station and the terminal device.

[0003] For example, when a base station transmits a known signal and a terminal device receives that signal, the terminal device can estimate the conditions of the propagation path. Furthermore, by having the terminal device feed back the estimated propagation path conditions to the base station, the base station can set optimal communication parameters for the terminal device.

[0004] Furthermore, if there is no feedback from the terminal device regarding the propagation path conditions, the base station can recognize the average propagation path conditions by using a statistical propagation model (e.g., a path loss model, an interference model, etc.) that is appropriate to the distance from the terminal device. [Prior art documents] [Patent Documents]

[0005] [Patent Document 1] International Publication No. 2024 / 053559 [Overview of the project] [Problems that the invention aims to solve]

[0006] However, radio wave propagation in wireless communication varies greatly depending on factors such as the presence or absence of obstacles between the base station (transmitting point) and the terminal device (receiving point). Furthermore, considering interference with adjacent cells and surrounding base stations, base stations determine communication parameters based on statistical information such as propagation models to minimize such interference.

[0007] Therefore, when communication is performed using communication parameters determined by the base station, there is a risk that suitable wireless communication may not be possible depending on the obstacles present in the space where the base station and the terminal device actually perform wireless communication (hereinafter also referred to as the "real space"). For example, if the base station determines the minimum transmission power to minimize interference, and an obstacle exists between the base station and the terminal device, there is a risk that the signal transmitted by the base station will not reach the terminal device due to this obstacle.

[0008] To avoid this, for example, there is a known technique of determining communication parameters using an estimation model generated in advance using machine learning or other techniques. When determining communication parameters using an estimation model, improving the estimation accuracy of the estimation model contributes to more suitable wireless communication control, i.e., determination of communication parameters.

[0009] Therefore, this disclosure proposes an information processing device, a terminal device, an information processing method, and a program that can further improve the estimation accuracy of the estimation model used to determine communication parameters.

[0010] It should be noted that the above-mentioned problems or objectives are merely one of several problems or objectives that can be solved or achieved by the multiple embodiments disclosed herein. [Means for solving the problem]

[0011] The information processing apparatus of the present disclosure includes a control unit. The control unit generates an estimation model for estimating data regarding the communication characteristics in a second area using data regarding the communication characteristics in a first area. The control unit generates the estimation model using the data regarding the communication characteristics in one or a plurality of the first areas similar to the second area.

Brief Description of the Drawings

[0012] [Figure 1] It is a diagram showing an example of radio wave propagation according to the proposed technology of the present disclosure. [Figure 2] It is a diagram showing an example of a learning area and a prediction area. [Figure 3] It is a diagram showing another example of a learning area and a prediction area. [Figure 4] It is a diagram showing an example of a learning area according to an embodiment of the present disclosure. [Figure 5] It is a diagram showing an example of a prediction area according to an embodiment of the present disclosure. [Figure 6] It is a diagram showing an example of communication processing according to an embodiment of the present disclosure. [Figure 7] It is a diagram showing an example of statistical processing according to an embodiment of the present disclosure. [Figure 8] It is a diagram showing a configuration example of a wireless communication system according to an embodiment of the present disclosure. [Figure 9] It is a block diagram showing a configuration example of a base station according to an embodiment of the present disclosure. [Figure 10] It is a block diagram showing a configuration example of a terminal device according to an embodiment of the present disclosure. [Figure 11] It is a diagram showing a configuration example of a control station according to an embodiment of the present disclosure. [Figure 12] It is a diagram showing an example of a candidate area and a prediction area in a first selection method according to an embodiment of the present disclosure. [Figure 13] It is a diagram showing an example of a candidate area and a prediction area in a second selection method according to an embodiment of the present disclosure. [Figure 14]This figure shows an example of candidate areas and predicted areas in a third selection method according to an embodiment of this disclosure. [Figure 15] This figure shows an example of candidate areas and predicted areas in a third selection method according to an embodiment of this disclosure. [Figure 16] This figure shows an example of a first candidate area in a fifth selection method according to an embodiment of the present disclosure. [Figure 17] This figure shows an example of a second candidate area in a fifth selection method according to an embodiment of the present disclosure. [Figure 18] This figure shows an example of a prediction area in a fifth selection method according to an embodiment of this disclosure. [Figure 19] This figure shows an example of a first comparative method according to an embodiment of the present disclosure. [Figure 20] This figure shows an example of a second comparative method according to an embodiment of the present disclosure. [Figure 21] This figure shows another example of a second comparative method according to the embodiments of this disclosure. [Figure 22] This figure shows an example of extracting a divided area from a learning area according to the embodiment of this disclosure. [Figure 23] This figure shows another example of a second comparative method according to the embodiments of this disclosure. [Figure 24] This flowchart shows an example of the information generation process flow according to the embodiment of this disclosure. [Figure 25] This figure shows an example of the hardware configuration of a device, etc. [Modes for carrying out the invention]

[0013] Embodiments of this disclosure will be described in detail below with reference to the attached drawings. In this specification and the drawings, components having substantially the same functional configuration are denoted by the same reference numerals, and redundant descriptions will be omitted.

[0014] Furthermore, in this specification and drawings, similar components of embodiments may be distinguished by adding at least one different alphabet and number after the same reference numeral. However, if there is no need to particularly distinguish each of the similar components, only the same reference numeral will be used. For example, multiple components having substantially the same functional configuration may be distinguished as terminal device 400_1 and terminal device 400_2 as needed. For example, if there is no need to particularly distinguish between terminal device 400_1 and terminal device 400_2, they will simply be referred to as terminal device 400.

[0015] The one or more embodiments (including examples, modifications, and applications) described below can each be implemented independently. On the other hand, at least some of the embodiments described below may be implemented in appropriate combination with at least some of the other embodiments. These embodiments may contain novel features that differ from each other. Therefore, these embodiments may contribute to solving different objectives or problems and may produce different effects.

[0016] <<1. Introduction>> <1-1. Background> In wireless communication, it is known that communication characteristics such as received power, throughput, interference power, SIR (Signal-to-Interference power ratio), and SINR (Signal-to-Interference and Noise power ratio) are strongly dependent on the radio wave propagation characteristics between the base station and the terminal equipment.

[0017] To improve communication characteristics, base stations are required to adaptively design communication parameters such as modulation schemes and coding rates according to the radio wave propagation characteristics around the communication area.

[0018] For example, when the received power at a terminal device is high, the base station can improve transmission efficiency by using a higher-order modulation scheme. On the other hand, when the received power is low, an increase in bit error rate and symbol error rate can be suppressed by using a lower-order modulation scheme.

[0019] Conventional radio wave propagation characteristic predictions have relied on empirical models, such as the Okumura model. These empirical models are constructed by statistically processing measured data in representative environments, such as large cities. They are used to estimate the global characteristics of radio wave propagation (such as path loss) based on information such as the distance between the transmitter and receiver.

[0020] On the other hand, in real-world environments, radio wave propagation characteristics fluctuate probabilistically due to factors such as shadowing and multipath fading. Since empirical models have difficulty predicting propagation characteristics with such probabilistic fluctuations, a new propagation prediction method is needed to replace empirical models.

[0021] <1-2. Challenges> However, radio wave propagation characteristics (communication characteristics) in wireless communication can vary significantly depending on factors such as the presence or absence of obstacles between the base station (transmitter) and the terminal device (receiver). This point will be explained using Figure 1.

[0022] Figure 1 shows an example of radio wave propagation according to the proposed technology of this disclosure. Figure 1(a) shows an example of radio wave propagation when there is no obstacle 600. Figure 1(b) shows an example of radio wave propagation when there is an obstacle 600.

[0023] For example, suppose base station 300 transmits a signal with transmission power Ptx. Without obstacle 600 (see Figure 1(a)), the radio waves will travel further compared to when obstacle 600 is present (see Figure 1(b)).

[0024] When determining communication parameters using the statistical propagation model described above, considering interference with neighboring cells and surrounding base stations, the base station will determine the communication parameters in a way that minimizes this interference. This is because the statistical propagation model differs from the actual propagation path conditions.

[0025] Here, statistical propagation models are generated by limiting the communication environment to typical propagation environments such as free space or urban areas. Therefore, using statistical propagation models to estimate radio wave intensity in a specific environment may degrade the estimation accuracy.

[0026] Therefore, in conventional interference design, base stations avoid interference with adjacent cells and surrounding base stations by taking into account fluctuations in radio wave intensity due to obstacles and adding a large margin to the interference power.

[0027] Thus, when communication parameters are determined with a large margin to minimize interference, the efficiency of wireless resource utilization is limited, which can lead to a significant degradation in communication performance.

[0028] Furthermore, if the terminal device estimates the propagation path conditions and provides feedback, the base station 300 can determine more accurate communication parameters according to the actual transmission path conditions. However, the base station 300 can only grasp the propagation path conditions at the location of the terminal device that provided the feedback.

[0029] Therefore, when a terminal device moves, it is not possible to accurately determine the propagation path conditions at the destination. Furthermore, it is not possible to accurately determine the propagation path conditions in locations where no terminal device is present.

[0030] Furthermore, when numerous terminal devices each provide feedback on the propagation path conditions, the communication resources required for this feedback become an overhead, reducing the overall efficiency of wireless resource utilization in the communication system.

[0031] Furthermore, depending on the use case, the base station may not receive sufficient feedback on the propagation path conditions from the terminal device. For example, the base station may receive no feedback at all, or very little, from the terminal device.

[0032] In such circumstances (environment, location, area), communication systems are required to accurately estimate the propagation path conditions of terminal devices and / or base stations. These propagation path conditions may include, for example, communication characteristics and the wireless communication environment.

[0033] Therefore, conventional techniques have been known that utilize estimation models (also called communication characteristic estimation models) that estimate (or predict) communication characteristics using actual measurement data in a predetermined communication area. For example, a communication system generates an estimation model using actual measurement data in this predetermined communication area and uses the generated estimation model to predict communication characteristics in the predetermined communication area.

[0034] In this case, the communication area from which the measured data for generating the estimation model is acquired (hereinafter also referred to as the learning area) and the communication area from which the communication characteristics are predicted (hereinafter also referred to as the prediction area) are the same predetermined area. Therefore, a certain degree of correlation is expected between the wireless environment in the learning area and the prediction area.

[0035] This correlation increases the probability that the communication characteristics between the learning area and the prediction area are similar. Therefore, by using this estimation model, the communication system can predict the communication characteristics (e.g., communication parameters) in the prediction area with greater accuracy.

[0036] One example of an estimation model is the Radio Environment Map (REM). A REM is a map that visualizes the spatial distribution of communication characteristics within a given communication area. In recent years, REMs have been increasingly used in a wide range of applications, including UAVs (Unmanned Aerial Vehicles) and DSA (Dynamic Spectrum Access).

[0037] REM is generated, for example, based on measured data regarding communication characteristics at one or more receiving points within a learning area. In other words, in order to create REM, it is necessary to acquire measured data regarding communication characteristics at one or more receiving points within a learning area.

[0038] In this case, the communication system described in Patent Document 1 generates a communication characteristics estimation model (an example of an estimation model) using measured data in a certain learning area, and then uses the generated communication characteristics model to predict the communication characteristics in the learning area. In other words, Patent Document 1 assumes that the learning area and the prediction area are the same.

[0039] Figure 2 shows an example of a learning area and a prediction area. In Figure 2, an example is shown where the learning area and the prediction area are the same.

[0040] In Figure 2, the hatched areas represent locations where measured data regarding communication characteristics were acquired. The communication system generates an estimation model using the measured data acquired at these locations. Subsequently, the communication system uses this estimation model to predict the communication characteristics at the hatched areas.

[0041] In the example in Figure 2, the learning area and the prediction area perfectly coincide geographically.

[0042] Figure 3 shows another example of the learning area and prediction area. In Figure 3, an example is shown where the prediction area is part of the learning area. Thus, even if the learning area and prediction area do not perfectly coincide geographically, if the prediction area is part of the learning area, the learning area and prediction area are assumed to be the same.

[0043] In this case, the communication system generates an estimation model using measured data acquired at hatched locations within the learning area. The communication system then uses this estimation model to predict the communication characteristics at hatched locations within the prediction area.

[0044] As shown in Figure 3, predictions made when the prediction area is contained within the learning area are generally called spatial interpolation. Conventional studies on REM (Reverse Engineering Method) have largely assumed this spatial interpolation.

[0045] In the examples shown in Figures 2 and 3, the prediction area is contained within the learning area. Therefore, the prediction area and the learning area are expected to be similar to some extent; in other words, the wireless environments of the prediction area and the learning area are expected to have a certain degree of correlation.

[0046] Communication characteristics are known to be highly dependent on the wireless environment within the communication area. Therefore, if there is a certain level of correlation between the wireless environments of the learning area and the prediction area, the probability that the communication characteristics between the two areas will also be similar increases.

[0047] Therefore, if the prediction area is contained within the learning area, that is, if the prediction area and the learning area are the same, it is considered that the estimation model generated using the measured data of the learning area can accurately predict the communication characteristics of the prediction area.

[0048] On the other hand, there are cases where the prediction area and the learning area are not the same. Figure 4 shows an example of a learning area according to an embodiment of this disclosure. Figure 5 shows an example of a prediction area according to an embodiment of this disclosure. Figures 4 and 5 show examples where the learning area and the prediction area are completely different areas.

[0049] In this embodiment, the statement that the learning area and the prediction area are "separate areas" means that the prediction area is not contained within the learning area. In other words, in this embodiment, when the learning area and the prediction area are "separate areas," it means that the learning area and the prediction area are not completely the same area, and the prediction area is not contained within the learning area.

[0050] For example, cases other than those shown in Figures 2 and 3 are cases where the learning area and prediction area are "different areas". For example, as shown in Figure 4, if the prediction area and learning area are geographically separated and not contiguous, then the learning area and prediction area are considered to be different areas.

[0051] Predicting wireless environments when the learning area and prediction area are in different locations is generally classified as a spatial extrapolation task.

[0052] Unlike spatial interpolation, spatial extrapolation is likely to involve differences in the wireless environment between the learning area and the prediction area. This can lead to a decrease in the correlation between communication characteristics in both areas. Consequently, spatial extrapolation generally results in lower prediction accuracy of communication characteristics compared to spatial interpolation.

[0053] Therefore, even when the learning area and the prediction area are different, that is, separate areas, it is necessary to predict communication characteristics with higher accuracy.

[0054] <1-3. Overview of the proposed technology> Figure 6 shows an example of communication processing according to an embodiment of this disclosure. The communication processing shown in Figure 6 is performed in a communication system. The communication system comprises a control station 100, a base station 300, and terminal devices 400_1 to 400_3. The base station 300 may be a base station of a cellular wireless communication system, as will be described later, or it may be a wireless LAN router such as a Wi-Fi® access point.

[0055] Here, terminal device 400_1 is assumed to be a device located within the prediction area, and terminal devices 400_2 and 400_3 are assumed to be devices located within the learning area. Note that the number of control stations 100 in the communication system is not limited to one, but may be two or more. The number of base stations 300 in the communication system is not limited to one, but may be two or more. Furthermore, the number of terminal devices 400 located within the prediction area is not limited to one, but may be two or more. The number of terminal devices 400 located within the learning area is not limited to two, but may be one or four or more.

[0056] First, the base station 300 transmits signals to terminal devices 400_2 and 400_3 at regular or non-periodic intervals (step S1). These signals are, for example, reference signals for measuring communication characteristics.

[0057] Terminal devices 400_2 and 400_3 measure signals to generate measured data and transmit the generated measured data to the control station 100 (step S2). The measured data includes, for example, information about communication characteristics. Information about communication characteristics includes, for example, information about received power (e.g., RSRP (Reference Signal Received Power), RSSI (Received Signal Strength Indicator), RSRQ (Reference Signal Received Quality)), interference power, SNR (Signal-to-noise ratio), SIR (Signal-to-Interference power ratio), SINR (Signal and interference-to-noise ratio), throughput, delay, Ping, and information about the location of terminal device 400.

[0058] In this case, the terminal device 400 may transmit the measured data to the control station 100 via the base station 300, or it may transmit the measured data to the control station 100 without going through the base station 300. That is, the terminal device 400 can transmit the measured data to the control station 100 via the cellular network.

[0059] Alternatively, the terminal device 400 may transmit the measured data to the control station 100 via Wi-Fi, the internet, or the like. Alternatively, the terminal device 400 may transmit the measured data to the control station 100 by being directly connected to the control station 100 via a cable.

[0060] The control station 100 is, for example, a cloud server or a database server with computing capabilities. The control station 100 may also be a frequency management system, such as the SAS (Spectrum Access System) used by the US CBRS (Citizens Broadband Radio Service).

[0061] The control station 100 performs statistical processing using the acquired measurement data (step S3). The statistical processing will be described later with reference to Figure 7.

[0062] The control station 100 notifies the base station 300 and / or terminal device 400_1 of statistical information regarding the results of statistical processing (step S4).

[0063] Furthermore, the statistical processing here is at least processing using measured data. For example, the statistical processing according to this embodiment includes the process of generating an estimation model for estimating the radio wave propagation characteristics between the base station 300 and the terminal device 400. The statistical processing may also include the process of estimating the radio wave propagation characteristics using the estimation model.

[0064] The control station 100 may notify the base station 300 and / or terminal equipment 400 of the estimated radio wave propagation characteristics as statistical information. Alternatively, the control station 100 may notify the base station 300 and / or terminal equipment 400 of communication parameters calculated based on the estimated radio wave propagation characteristics (for example, control information for communication between the base station 300 and terminal equipment 400) as statistical information.

[0065] The control station 100 may calculate statistical information in accordance with requests from the base station 300 and / or terminal equipment 400. In other words, the base station 300 and / or terminal equipment 400 may send a request to the control station 100 regarding the calculation of statistical information. The base station 300 and / or terminal equipment 400 may, for example, use its own communication unit, as described below, to request the control station 100 to transmit the statistical information. The control station 100 will then notify the base station 300 and / or terminal equipment 400 of the statistical information in response to this request.

[0066] In this example, the terminal devices 400_2 and 400_3 that transmit the measured data and the terminal device 400_1 that acquires the statistical information are assumed to be different devices, but these devices may be the same. That is, terminal device 400 may transmit the measured data and acquire the statistical information generated using this measured data.

[0067] Figure 7 shows an example of statistical processing according to the embodiment of this disclosure. The statistical processing shown in Figure 7 is performed, for example, in the control station 100.

[0068] The control station 100 compares, for example, at least one candidate area that is a candidate for the learning area with the prediction area (step S11). In the example in Figure 7, the control station 100 compares the first candidate area with the prediction area, and then compares the second candidate area with the prediction area.

[0069] The control station 100 selects a learning area according to the comparison results (step S12). For example, the control station 100 selects one or more candidate areas similar to the predicted area as the learning area.

[0070] The control station 100 generates an estimation model using the measured data acquired in the selected learning area (step S13).

[0071] The control station 100 estimates the estimation model using, for example, at least one of the following methods. - Statistical analysis methods such as linear regression analysis and multiple regression analysis - AI / ML (Artificial Intelligence, Machine Learning, Deep Learning) - Deep Learning - Multilayer perceptron (MLP) - Convolutional Neural Network (CNN) - Stochastic Gradient Descent - Decision Tree - Random Forest - Support Vector Machine -k-Nearest Neighbors - Bayesian Classifier (Naive Bayes Classifier)

[0072] In addition to the methods described above, numerous other methods exist for generating estimation models. It is desirable for the estimater to consider the principles of each method and select the appropriate one. Furthermore, when using AI / ML, the communication characteristics of the target area (i.e., the prediction area) can be considered the objective variable. Explanatory variables include, for example, various parameters, data, and information (e.g., communication environment information) described later.

[0073] Furthermore, while this explanation describes the case where the control station 100 generates an estimation model as part of statistical processing, the control station 100 may also perform predictions of communication characteristics using this estimation model, in addition to generating the estimation model, as part of statistical processing.

[0074] Furthermore, the control station 100 may perform a determination process to determine whether or not to perform the statistical processing related to the proposed technology before performing the statistical processing.

[0075] For example, before performing statistical processing, the control station 100 determines whether the learning area and the prediction area are the same, and decides whether or not to perform statistical processing based on the determination result. For example, if the control station 100 determines that the learning area and the prediction area are not the same, that is, that the learning area and the prediction area are different, it decides to perform statistical processing.

[0076] For example, the control station 100 plots the location information (e.g., latitude and longitude information) of the learning area and the prediction area on a map and determines whether the learning area and the prediction area are the same.

[0077] As described above, the control station 100 relating to the proposed technology generates an estimation model for estimating data on communication characteristics in a prediction area (an example of a second area), such as communication parameters, using data on communication characteristics in a learning area (an example of a first area), such as measured data.

[0078] The control station 100 generates an estimation model using data on communication characteristics in one or more learning areas similar to the prediction area. For example, the control station 100 compares candidate areas for the learning area with the prediction area and selects a learning area similar to the prediction area according to the comparison result.

[0079] When the prediction area and the learning area are similar, the correlation between the wireless environments of both areas is considered to be high. Therefore, the control station 100 can generate an estimation model using data on communication characteristics in one or more learning areas similar to the prediction area, thereby generating an estimation model using data from learning areas that have a high correlation with the wireless environment of the prediction area.

[0080] As a result, the control station 100 can generate an estimation model that can predict (estimate) data related to communication characteristics (e.g., communication parameters, etc.) with higher accuracy.

[0081] <<2. Example of a communication system configuration>> <2-1. Example of overall communication system configuration> Figure 8 shows an example configuration of a wireless communication system according to an embodiment of the present disclosure. The wireless communication system shown in Figure 8 comprises a control station 100, core networks 200A and 200B, base stations 300A1, 300A2, 300B1 and 300B2, and terminal devices 400A1, 400A2, 400B1 and 400B2.

[0082] The control station 100 connects to the core network 200A within the local network N1_A via network N1_P. The control station 100 also connects to the core network 200B within the local network N1_B via network N1_P.

[0083] Network N1_P is a communication network such as a LAN (Local Area Network), WAN (Wide Area Network), cellular network, fixed telephone network, regional IP (Internet Protocol) network, or the Internet. Network N1_P may include wired networks or wireless networks. Network N1_P may also be a data network connected to the core network 200. The data network may be a service network of a telecommunications carrier, such as an IMS (IP Multimedia Subsystem) network. Alternatively, the data network may be a private network, such as an internal corporate network. Although only one network N1_P is shown in the example in Figure 8, there is no limit to the number of networks N1_P.

[0084] Furthermore, while Figure 8 shows two local networks, N1_A and N1_B, the number of local networks is not limited to two. There can be one local network, three or more, or more.

[0085] In local network N1_A, core network 200A connects to base stations 300A1 and 300A2. The number of base stations 300A connected to core network 200A is not limited to two. There may be one base station 300A or three or more.

[0086] Base station 300A1 connects to terminal device 400A1 via wireless communication. Base station 300A2 connects to terminal device 400A2 via wireless communication. The number of terminal devices 400A connected to base station 300A is not limited to one; there may be two or more. Also, the number of terminal devices 400A1 connected to base station 300A1 may be different from the number of terminal devices 400A2 connected to base station 300A2.

[0087] Note that the configuration of local network N1_B is the same as that of local network N1_A, so the explanation will be omitted.

[0088] For example, control station 100 is an information processing device that controls a Dynamic Spectrum Access (DSA) system. Control station 100 can control radio resources and communication parameters for at least one of the local networks N1_A, N1_B, core networks 200A, 200B, and base stations 300A, 300B connected to the DSA. Here, radio resources mean resources in at least one of the time, frequency, MIMO layer, and spatial domains used for radio communication.

[0089] Note that the core networks 200A and 200B do not need to be installed. In this case, the control station 100 will be directly connected to the base stations 300A and 300B.

[0090] The core network 200 may be located within either the control station 100 or the base stations 300. The core network 200 may also be distributed and located within both the control station 100 and the base stations 300.

[0091] Local networks N1_A and N1_B are also called private networks and are networks whose communication coverage (an example of a communication area) is limited to a predetermined area or site. In local networks N1_A and N1_B, only pre-registered terminal devices 400 can connect to at least one of the base station 300, the control station 100, and the core network 200.

[0092] Furthermore, examples of radio access technologies (RATs) used for wireless communication between the base station 300 and the terminal device 400 include cellular communication systems such as 4G, 5G, 6G, LTE (Long Term Evolution), and NR (New Radio). Moreover, this radio access technology is not limited to cellular communication systems. For example, various wireless communication systems such as Wi-Fi, Bluetooth®, and LPWA (Low Power Wide Area) systems can also be used as this radio access technology.

[0093] Furthermore, in the example shown in Figure 8, the control station 100 is assumed to be connected to local networks N1_A and N1_B, but the network to which the control station 100 is connected may also be a public network to which subscribers can connect.

[0094] As described above, the core network 200 may be omitted or located within the control station 100 and / or base station 300.

[0095] Therefore, in order to simplify the explanation below, we will assume that the wireless communication system is a system that omits the core network 200. That is, the wireless communication system of this embodiment will include a control station 100, a base station 300, and a terminal device 400.

[0096] <2-2. Example of base station configuration> Next, the base station 300 will be described. The base station 300 is a communication device that operates a cell and provides wireless communication services to one or more terminal devices 400 located within the cell's coverage. The cell operates according to any wireless communication method, such as LTE or NR. The base station 300 is connected to the core network 200. The core network 200 is connected to the packet data network (not shown) via a gateway device (not shown). The base station 300 also operates beams identifiable by SSB (Synchronization Signal / PBCH Block) and can send and receive data to and from one or more terminal devices 400 via one or more beams.

[0097] The base station 300 may be composed of a collection of multiple physical or logical devices. For example, in this embodiment, the base station 300 may be distinguished into multiple devices of BBU (Baseband Unit) and RU, and may be interpreted as a collection of these multiple devices. Alternatively, in this embodiment, the base station 300 may be either or both of the BBU and RU. The BBU and RU may be connected by a predetermined interface (e.g., eCPRI). Alternatively, the RU may be referred to as a Remote Radio Unit (RRU) or Radio DoT (RD). Alternatively, the RU may correspond to a gNB-DU described later. Alternatively, the BBU may correspond to a gNB-CU described later. Alternatively, the RU may be connected to a gNB-DU described later. Alternatively, the BBU may correspond to a combination of gNB-CU and gNB-DU described later. Alternatively, the RU may be a device formed integrally with an antenna. The antenna of the base station 300 (for example, an antenna integrated with the RU) may employ an Advanced Antenna System and support MIMO (for example, FD-MIMO) and beamforming. The Advanced Antenna System may include, for example, 64 transmitting antenna ports and 64 receiving antenna ports.

[0098] Furthermore, multiple base stations 300 may be connected to one another. One or more base stations 300 may be included in a Radio Access Network (RAN). That is, base stations 300 may simply be referred to as RAN, RAN node, AN (Access Network), or AN node. In LTE, the RAN is called EUTRAN (Enhanced Universal Terrestrial RAN). In NR, the RAN is called NGRAN. In W-CDMA (UMTS), the RAN is called UTRAN. In LTE, base stations 300 are referred to as eNodeB (Evolved Node B) or eNB. That is, EUTRAN includes one or more eNodeB (eNB). Also, in NR, base stations 300 are referred to as gNodeB or gNB. That is, NGRAN includes one or more gNB. Furthermore, EUTRAN may include gNB (en-gNB) connected to the core network (EPC) in the LTE communication system (EPS). Similarly, NGRAN may include ng-eNB connected to the core network 5GC in a 5G communication system (5GS). Furthermore, or alternatively, if base station 300 is an eNB, gNB, etc., it may be referred to as 3GPP Access. Furthermore, or alternatively, if base station 300 is a radio access point (e.g., a Wi-Fi® access point), it may be referred to as Non-3GPP Access. Furthermore, or alternatively, base station 300 may be an optical extension device called an RRH (Remote Radio Head). Furthermore, or alternatively, if base station 300 is a gNB, base station 300 may be referred to as a combination of the aforementioned gNB CU (Central Unit) and gNB DU (Distributed Unit), or either of these. The gNB CU hosts multiple upper layers (e.g., RRC, SDAP, PDCP) of the Access Stratum for communication with the UE. On the other hand, the gNB-DU hosts multiple lower layers (e.g., RLC, MAC, PHY) of the Access Stratum.In other words, among the messages and information described later, RRC signalling (e.g., MIB, various SIBs including SIB1, RRCSetup message, RRCReconfiguration message) may be generated by the gNB CU, while DCI and various Physical Channels (e.g., PDCCH, PBCH) described later may be generated by the gNB-DU. Alternatively, among the RRC signalling, some configuration information, such as IE:cellGroupConfig, may be generated by the gNB-DU, and the remaining configuration information may be generated by the gNB-CU. These configuration information may be transmitted and received via the F1 interface described later. Base station 300 may be configured to communicate with other base stations 300. For example, if multiple base stations 300 are eNBs or a combination of eNB and en-gNB, the base stations 300 may be connected via the X2 interface. Furthermore, or alternatively, if multiple base stations 300 are gNBs or a combination of gn-eNB and gNB, the devices may be connected via the Xn interface. Furthermore, or alternatively, if multiple base stations 300 are a combination of gNB CUs and gNB DUs, the devices may be connected to each other via the F1 interface described above. The messages and information described later (RRC signalling or DCI information, Physical Channel) may be communicated between the multiple base stations 300 (for example, via the X2, Xn, and F1 interfaces).

[0099] Furthermore, as mentioned above, the base station 300 may be configured to manage multiple cells. The cells provided by the base station 300 are called Serving cells. A Serving cell includes PCells (Primary Cells) and SCells (Secondary Cells). When Dual Connectivity (e.g., EUTRA-EUTRA Dual Connectivity, EUTRA-NR Dual Connectivity (ENDC), EUTRA-NR Dual Connectivity with 5GC, NR-EUTRA Dual Connectivity (NEDC), NR-NR Dual Connectivity) is provided to a UE (e.g., terminal device 400), the PCells and zero or more SCell(s) provided by the MN (Master Node) are called a Master Cell Group. Furthermore, a Serving cell may also include PSCells (Primary Secondary Cells or Primary SCG Cells). That is, when Dual Connectivity is provided to a UE, the PSCells and zero or more SCell(s) provided by the SN (Secondary Node) are called a Secondary Cell Group (SCG). Unless otherwise specified (e.g., PUCCH on SCell), the Physical Uplink Control Channel (PUCCH) is transmitted by PCell and PSCell, but not by SCell. Similarly, Radio Link Failure is detected by PCell and PSCell, but not by SCell (and does not need to be detected). Because PCell and PSCell have special roles within Serving Cell(s), they are also called Special Cells (SpCell). A single cell may be associated with one Downlink Component Carrier and one Uplink Component Carrier. Furthermore, the system bandwidth corresponding to a single cell may be divided into multiple Bandwidth Parts.In this case, one or more Bandwidth Parts (BWPs) may be configured for the UE, and one Bandwidth Part may be used by the UE as the Active BWP. Furthermore, the radio resources (e.g., frequency band, numerology (subcarrier spacing), slot format) available to the terminal device 400 may differ for each cell, component carrier, or BWP.

[0100] Figure 9 is a block diagram showing an example configuration of a base station 300 according to the present disclosure. The base station 300 is a wireless communication device that communicates wirelessly with a terminal device 400. The base station 300 is a type of communication device. The base station 300 is also a type of information processing device.

[0101] The base station 300 shown in Figure 9 comprises a communication unit 310, a storage unit 320, a network communication unit 330, and a control unit 340. Note that the configuration shown in Figure 9 is a functional configuration, and the hardware configuration may differ. Furthermore, the functions of the base station 300 may be distributed and implemented across multiple physically separated configurations. For example, as mentioned above, the functions of the base station 300 may be distributed across a CU and a DU, or a CU, a DU, and an RU.

[0102] The communication unit 310 is a signal processing unit for wireless communication with other wireless communication devices (e.g., terminal device 400 and other base stations 300). The communication unit 310 operates according to the control of the control unit 340. If the other wireless communication device is terminal device 400, the communication unit 310 may be a wireless transceiver that supports one or more wireless access schemes. For example, the communication unit 310 supports both NR and LTE. In addition to NR and LTE, the communication unit 310 may also support W-CDMA and cdma2000. Furthermore, the communication unit 310 may support communication using NOMA. If the other wireless communication device is another base station 300, the communication unit 310 may have an X2 interface, an Xn interface, or an F1 interface.

[0103] The communication unit 310 comprises a receiving processing unit 311, a transmitting processing unit 312, and an antenna 313. The communication unit 310 may have multiple receiving processing units 311, transmitting processing units 312, and antennas 313. If the communication unit 310 supports multiple wireless access methods, each part of the communication unit 310 may be configured separately for each wireless access method. For example, the receiving processing unit 311 and the transmitting processing unit 312 may be configured separately for LTE and NR.

[0104] The receiving processing unit 311 processes the uplink signal received via the antenna 313. The receiving processing unit 311 operates as a receiving unit that receives the received signal. The receiving processing unit 311 comprises a wireless receiving unit 311a, a multiplexing / decoupling unit 311b, a demodulation unit 311c, and a decoding unit 311d.

[0105] The wireless receiver 311a performs functions on the uplink signal such as down-conversion, removal of unwanted frequency components, amplification level control, quadrature demodulation, conversion to a digital signal, removal of guard intervals (cyclic prefixes), and extraction of frequency domain signals by fast Fourier transform. The multiplexing / decoupling unit 311b separates the uplink channels and uplink reference signals, such as PUSCH (Physical Uplink Shared Channel) and PUCCH (Physical Uplink Control Channel), from the signal output from the wireless receiver 311a.

[0106] The demodulator 311c demodulates the received signal using a modulation scheme such as BPSK (Binary Phase Shift Keying) or QPSK (Quadrature Phase Shift Keying) for the modulation symbols of the uplink channel. The modulation scheme used by the demodulator 311c may be 16QAM (Quadrature Amplitude Modulation), 64QAM, or 256QAM. In this case, the signal points on the constellation do not necessarily have to be equidistant. The constellation may be a non-uniform constellation (NUC).

[0107] The decoding unit 311d performs decoding on the encoded bits of the demodulated uplink channel. The decoded uplink data and uplink control information are output to the control unit 340.

[0108] The transmission processing unit 312 performs the transmission processing of downlink control information and downlink data. Thus, the transmission processing unit 312 is an acquisition unit that acquires bit sequences such as downlink control information and downlink data from the control unit 340. The transmission processing unit 312 comprises an encoding unit 312a, a modulation unit 312b, a multiplexing unit 312c, and a wireless transmission unit 312d.

[0109] The encoding unit 312a encodes the downlink control information and downlink data input from the control unit 340 using encoding methods such as block coding, convolutional coding, and turbo coding. The encoding unit 312a may also perform encoding using polar code or LDPC (Low Density Parity Check Code).

[0110] The modulation unit 312b modulates the encoded bits output from the encoding unit 312a using a predetermined modulation scheme such as BPSK, QPSK, 16QAM, 64QAM, or 256QAM. In this case, the signal points on the constellation do not necessarily have to be equidistant. The constellation may be a heterogeneous constellation.

[0111] The multiplexer 312c multiplexes the modulation symbols and downlink reference signals for each channel and places them in predetermined resource elements. The wireless transmitter 312d performs various signal processing on the signals from the multiplexer 312c. For example, the wireless transmitter 312d performs processing such as time-domain to frequency-domain conversion using the Fast Fourier Transform, addition of guard intervals (cyclic prefixes), generation of baseband digital signals, conversion to analog signals, quadrature modulation, upconversion, removal of extraneous frequency components, and power amplification. The signals generated by the transmission processing unit 312 are transmitted from the antenna 313.

[0112] The memory unit 320 is a data read / write storage device such as DRAM (Dynamic Random Access Memory), SRAM (Static Random Access Memory), flash memory, or hard disk. The memory unit 320 functions as a storage means for the base station 300.

[0113] The network communication unit 330 is a communication interface for communicating with a node located higher up on the network (for example, the core network 200). For example, the network communication unit 330 may be a LAN (Local Area Network) interface such as a NIC (Network Interface Card). Alternatively, the network communication unit 330 may be an S1 interface or an NG interface for connecting to the core network node. The network communication unit 330 may be a wired interface or a wireless interface. The network communication unit 330 functions as a network communication means for the base station 300.

[0114] The control unit 340 is a controller that controls various parts of the base station 300. The control unit 340 is implemented by a processor (hardware processor) such as a CPU (Central Processing Unit) or an MPU (Micro Processing Unit). For example, the control unit 340 is implemented by the processor executing various programs stored in the memory device inside the base station 300, using RAM (Random Access Memory) or the like as a working area. The control unit 340 may also be implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array). CPUs, MPUs, ASICs, and FPGAs can all be considered controllers.

[0115] <2-3. Example of terminal device configuration> Next, an example of the configuration of the terminal device 400 according to the embodiment of this disclosure will be described using Figure 10. Figure 10 is a block diagram showing an example of the configuration of the terminal device 400 according to the embodiment of this disclosure.

[0116] The terminal device 400 is a wireless communication device that communicates wirelessly with the base station 300. The terminal device 400 may be, for example, a mobile phone, a smart device (smartphone or tablet), a PDA (Personal Digital Assistant), or a personal computer. Alternatively, the terminal device 400 may be a professional camera equipped with communication functions, or an M2M (Machine to Machine) device or an IoT (Internet of Things) device.

[0117] Furthermore, terminal device 400 may be capable of sidelink communication with other terminal devices 400. When performing sidelink communication, terminal device 400 may use automatic retransmission technology such as HARQ (Hybrid Automatic Repeat reQuest). Terminal device 400 may be capable of NOMA (Non Orthogonal Multiple Access) communication with base station 300. In addition, terminal device 400 may also be capable of NOMA communication in communication (sidelink) with other terminal devices 400. Furthermore, terminal device 400 may be capable of LPWA (Low Power Wide Area) communication with other communication devices (for example, base station 300 and other terminal devices 400). In addition, the wireless communication used by terminal device 400 may be wireless communication using millimeter waves. Furthermore, the wireless communication used by terminal device 400 (including sidelink communication) may be wireless communication using radio waves, or wireless communication using infrared or visible light (optical wireless).

[0118] The terminal device 400 may simultaneously connect to and communicate with multiple base stations 300 or multiple cells. For example, if one base station 300 can provide multiple cells, the terminal device 400 can perform carrier aggregation by using one cell as a pCell and other cells as sCells. Also, if multiple base stations 300 can each provide one or more cells, the terminal device 400 can achieve DC (Dual Connectivity) by using one or more cells managed by one base station 300 (MN (e.g., MeNB or MgNB)) as a pCell, or a pCell and sCell(s), and one or more cells managed by the other base station 300 (SN (e.g., SeNB or SgNB)) as a pCell(PSCell), or a pCell(PSCell) and sCell(s). DC may also be referred to as MC (Multi Connectivity).

[0119] Furthermore, when the communication area is supported via cells from different base stations 300 (multiple cells with different cell identifiers or the same cell identifier), it is possible to combine these multiple cells and communicate between the base station 300 and the terminal device 400 using carrier aggregation (CA), dual connectivity (DC), or multi-connectivity (MC) technologies. Alternatively, it is also possible for the terminal device 400 to communicate with these multiple base stations 300 via cells from different base stations 300 using coordinated multi-point transmission and reception (CoMP) technology.

[0120] The terminal device 400 comprises a communication unit 410, a storage unit 420, a network communication unit 430, an input / output unit 440, and a control unit 450. Note that the configuration shown in Figure 10 is a functional configuration, and the hardware configuration may differ. Furthermore, the functions of the terminal device 400 may be implemented in a distributed manner across multiple physically separated configurations.

[0121] The communication unit 410 is a signal processing unit for wireless communication with other wireless communication devices (e.g., base station 300 and other terminal devices 400). The communication unit 410 operates according to the control of the control unit 450. The communication unit 410 may be a wireless transceiver that supports one or more wireless access schemes. For example, the communication unit 410 supports both NR and LTE. In addition to NR and LTE, the communication unit 410 may also support W-CDMA and cdma2000. Furthermore, the communication unit 410 may also support communication using NOMA.

[0122] The communication unit 410 includes a receiving processing unit 411, a transmitting processing unit 412, and an antenna 413. The communication unit 410 may include multiple receiving processing units 411, transmitting processing units 412, and antennas 413.

[0123] The configuration of the communication unit 410, the receiving processing unit 411, the transmitting processing unit 412, and the antenna 413 is the same as that of the communication unit 310, the receiving processing unit 311, the transmitting processing unit 312, and the antenna 313 of the base station 300.

[0124] The memory unit 420 is a data read / write storage device such as DRAM, SRAM, flash memory, or hard disk. The memory unit 420 functions as a storage means for the terminal device 400.

[0125] The network communication unit 430 is a communication interface for communicating with other devices connected via the network. For example, the network communication unit 430 is a LAN interface such as a NIC. The network communication unit 430 may be a wired interface or a wireless interface. The network communication unit 430 functions as a network communication means for the terminal device 400. The network communication unit 430 communicates with other devices according to the control of the control unit 450.

[0126] The input / output unit 440 is a user interface for exchanging information with the user. For example, the input / output unit 440 is an operating device for the user to perform various operations, such as a keyboard, mouse, operation keys, or touch panel. Alternatively, the input / output unit 440 is a display device such as a liquid crystal display or an organic electroluminescence display. The input / output unit 440 may also be an audio device such as a speaker or buzzer. Furthermore, the input / output unit 440 may also be a lighting device such as an LED (light-emitting diode) lamp. The input / output unit 440 functions as an input / output means (input means, output means, operation means, or notification means) of the terminal device 400.

[0127] The control unit 450 is a controller that controls various parts of the terminal device 400. The control unit 450 is implemented by a processor such as a CPU, MPU, or GPU. For example, the control unit 450 is implemented by the processor executing various programs stored in the memory device inside the terminal device 400, using RAM or the like as a working area. The control unit 450 may also be implemented by an integrated circuit such as an ASIC or FPGA. A CPU, MPU, GPU, ASIC, and FPGA can all be considered controllers.

[0128] <2-4. Example of Control Station Configuration> Figure 11 shows an example configuration of a control station 100 according to an embodiment of the present disclosure. As described above, the control station 100 is an information processing device that controls, for example, a Dynamic Spectrum Access (DSA) system. As shown in Figure 11, the control station 100 comprises a communication unit 110, a storage unit 120, and a control unit 130.

[0129] (Communications Department 110) The communication unit 110 is a communication interface for communicating with other devices (e.g., base station 300). The communication unit 110 may be a network interface or an equipment connection interface. For example, the communication unit 110 may be a LAN interface such as a NIC, or a USB interface consisting of a USB (Universal Serial Bus) host controller, a USB port, etc. Furthermore, the communication unit 110 may be a wired interface or a wireless interface. The communication unit 110 functions as a communication means for the control station 100. The communication unit 110 communicates with the base station 300 according to the control of the control unit 130.

[0130] (Storage unit 120) The memory unit 120 is a data read / write storage device such as DRAM, SRAM, flash memory, or hard disk. The memory unit 120 functions as a storage means for the control station 100.

[0131] (Control unit 130) The control unit 130 is a controller that controls various parts of the control station 100. The control unit 130 is implemented by a processor such as a CPU, MPU, or GPU.

[0132] For example, the control unit 130 is implemented by the processor executing various programs stored in the memory device inside the control station 100 using RAM or the like as a working area. The control unit 130 may also be implemented by an integrated circuit such as an ASIC or FPGA. A CPU, MPU, GPU, ASIC, and FPGA can all be considered controllers.

[0133] The control unit 130 comprises a selection unit 131, an acquisition unit 132, a generation unit 133, a determination unit 134, and a notification unit 135. Each block constituting the control unit 130 (selection unit 131 to notification unit 135) is a functional block that indicates the function of the control unit 130.

[0134] These functional blocks may be software blocks or hardware blocks. For example, each of the functional blocks described above may be a single software module implemented in software (including microprograms), or a single circuit block on a semiconductor chip (die). Of course, each functional block may also be a single processor or a single integrated circuit. The configuration of the functional blocks is arbitrary.

[0135] The control unit 130 may be composed of functional units different from the functional blocks described above.

[0136] (Selection section 131) The selection unit 131 selects, for example, a learning area from which to acquire data to be used to generate the estimation model. The selection unit 131 selects, for example, a learning area to be used to generate the estimation model from candidate areas which are candidates for one or more learning areas.

[0137] For example, the selection unit 131 selects a learning area that is similar to the prediction area in which data regarding communication characteristics is predicted using the estimation model. For example, the selection unit 131 selects as a learning area an area from among the candidate areas that has a similar communication environment, or more specifically, an area with a high correlation of communication environments.

[0138] The selection unit 131 notifies the generation unit 133 of the selected learning area.

[0139] Details on how the selection unit 131 selects the learning area will be described later.

[0140] (Acquisition part 132) The acquisition unit 132 acquires information to be used by the generation unit 133 via the communication unit 110. For example, the acquisition unit 132 acquires actual measurement data for the predetermined area described above from the base station 300 and / or terminal device 400 via the communication unit 110. In this case, the acquisition unit 132 acquires actual measurement data for the estimated area from the base station 300 and / or terminal device 400.

[0141] The acquisition unit 132 receives information that may be transmitted from the local network, the core network 200, the base station 300, the terminal device 400, and at least one of the other communication nodes via the communication unit 110.

[0142] The acquisition unit 132 outputs the acquired information to the generation unit 133.

[0143] (Generation unit 133) The generation unit 133 executes a generation process to generate an estimation model using the actual measurement data acquired by the acquisition unit 132.

[0144] The generation unit 133 generates an estimation model based on measured data in at least one learning area. This estimation model is, for example, a communication characteristics estimation model used to estimate communication characteristics (radio wave propagation environment). The control station 100 generates the estimation model from measured data in the learning area, for example, by machine learning.

[0145] Here, the measured data used by the control station 100 to generate the estimation model is, for example, data relating to communication characteristics measured in the learning area selected by the selection unit 131.

[0146] The generation unit 133 outputs the generated estimation model to the determination unit 134.

[0147] (Decision Section 134) The determination unit 134 estimates estimated data regarding communication characteristics in the prediction area by performing estimation processing using an estimation model. For example, the determination unit 134 uses the estimated data to determine communication parameters between the base station 300 and the terminal device 400 in the prediction area.

[0148] The determination unit 134 estimates estimated data of communication characteristics in the prediction area using, for example, an estimation model generated based on actual data of communication characteristics measured in the learning area.

[0149] For example, the determination unit 134 inputs area data related to the prediction area into the estimation model. The determination unit 134 uses the output of the estimation model when area data is input as the estimated data. The area data may include, for example, communication environment information, which will be described later.

[0150] Furthermore, the area data may include actual measurement data within the prediction area. In this case, the number of actual measurement data points within the prediction area may be less than, for example, the total number of actual measurement data points within the prediction area.

[0151] Thus, the decision unit 134 may estimate the estimated data using the measured data in the prediction area. Alternatively, the decision unit 134 may correct the estimated data using the measured data in the prediction area. The decision unit 134 may also correct (retrain) the estimation model using the measured data in the prediction area.

[0152] The determination unit 134 determines, for example, the communication parameters between the base station 300 and the terminal device 400 in the prediction area using estimated data. The determination unit 134 outputs the determined communication parameters to the notification unit 135.

[0153] (Notification Department 135) The notification unit 135 notifies the base station 300 and / or terminal device 400 of the communication parameters determined by the determination unit 134.

[0154] <<3. How to select a learning area>> As described above, the control station 100 selects a learning area from one or more candidate areas. For example, the control station 100 selects one or more learning areas by comparing the predicted area with the candidate areas. For example, the control station 100 selects one or more candidate areas as learning areas whose communication environment is similar to that of the predicted area.

[0155] For example, the control station 100 calculates statistics related to the communication environment in one or more candidate areas and one prediction area, according to the unit space domain. The control station 100 determines the correlation (also referred to as similarity) between one or more candidate areas and one prediction area by comparing the statistics of one or more candidate areas with the statistics of the prediction area. The control station 100 selects a learning area according to the determination result.

[0156] The following describes some examples of selection methods performed by the control station 100.

[0157] For the sake of simplicity, in the following explanation, we will assume that the size of the prediction area and the size of the candidate area are the same.

[0158] <3-1. First Selection Method> For example, the control station 100 determines the correlation of the wireless environment (communication environment) using obstacle information regarding obstacles in the candidate area and the predicted area. For example, the control station 100 determines the correlation of the wireless environment according to the statistical amount of the height of obstacles in the candidate area and the predicted area.

[0159] In this embodiment, obstacles refer to buildings, trees, walls, signs, pillars, vehicles, and other objects that may affect the communication characteristics of the wireless environment.

[0160] First, the control station 100 divides the candidate area and the prediction area into one or more unit spatial regions. For example, the control station 100 divides the candidate area and the prediction area into grids. A grid is an example of a unit spatial region.

[0161] Here, the control station 100 divides the candidate area and the prediction area into grids of the same size. As mentioned above, since the candidate area and the prediction area are the same size, the number of grids included in the candidate area and the prediction area will be the same.

[0162] Figure 12 shows examples of candidate areas and predicted areas in a first selection method according to an embodiment of the present disclosure. Figure 12(a) shows an example of a first candidate area, Figure 12(b) shows an example of a second candidate area, and Figure 12(c) shows an example of a predicted area.

[0163] Here, we assume there are two candidate areas: a first candidate area and a second candidate area. However, the number of candidate areas is not limited to two; there may be one, three, or more. Furthermore, if the first and second candidate areas are not distinguished, they will simply be referred to as "candidate areas." Similarly, if the candidate areas and predicted areas are not distinguished, they will simply be referred to as "areas."

[0164] As shown in Figure 12, each area is divided into, for example, a square grid. Each grid represents, for example, any receiving point within each area.

[0165] The control station 100 calculates, for example, statistics regarding the height of obstacles in the grid. These statistics may be, for example, the average or maximum value of the obstacle heights, or other statistics. For example, if the grid contains multiple obstacles of different heights, the control station 100 may use the statistics of the heights of the multiple obstacles (e.g., average value, maximum value, etc.) as the height of the obstacles in that grid.

[0166] In Figure 12, the difference in obstacle heights in each grid is represented by the difference in hatching. For example, grids with the same hatching in each area have obstacles of the same or similar height. Here, similar obstacle heights may mean, for example, that the difference in obstacle heights between grids with the same hatching is less than a predetermined value.

[0167] Here, the receiving points in each area are represented by a square grid, but the receiving points may be represented in other ways. For example, each area may be divided into grids other than squares (in other words, unit spatial regions). Examples of grids other than squares include rectangular and circular grids. Also, the receiving points may be represented by points. Examples of points representing receiving points include grid intersections, points represented by latitude and longitude, etc.

[0168] Furthermore, while we have assumed here that each area is divided into 25 grids, the number of divisions is not limited to this. For example, the number of divisions in an area (in other words, the number of grids or receiving points) may be 24 or less, or 26 or more.

[0169] The control station 100, for example, compares the candidate area with the predicted area and selects the candidate area that is similar to the predicted area as the learning area. That is, when selecting the learning area, the control station 100 uses the similarity in the height of obstacles between the candidate area and the predicted area.

[0170] In the example shown in Figure 12, the first candidate area (see Figure 12(a)) is more similar to the predicted area (see Figure 12(c)) than the second candidate area (see Figure 12(b)). Therefore, the control station 100 selects the first candidate area as the learning area and generates an estimation model using this first candidate area.

[0171] For example, the control station 100 may calculate the similarity of obstacle heights in the candidate area and the prediction area using the difference in the average heights of the obstacles. For example, the control station 100 calculates the similarity using the following equation (1).

[0172]

number

[0173] Here, H X This is the average height of obstacles in candidate area X (in units of, for example, meters). For example, if candidate area X is divided into N grids and the height of obstacles is calculated for each grid, then H X This represents the average height of obstacles across all N grids.

[0174] Similarly, H is the average height of obstacles in the prediction area (in units of, for example, meters). For example, if the prediction area is divided into N grids and the height of obstacles is calculated for each grid, H will be the average height of obstacles in all N grids.

[0175] D X This D is the difference in the average height of obstacles between the candidate area X and the predicted area. X The smaller the value, that is, the smaller the difference in the average height of obstacles between candidate area X and the predicted area, the more the control station 100 determines that candidate area X and the predicted area are similar, that is, that they have a high degree of similarity.

[0176] Control station 100 is, for example, D XSelect the candidate area with the smallest, that is, the candidate area with the highest similarity as the learning area. Alternatively, the control station 100, for example, D X From the ones with smaller D, that is, from the ones with higher similarity, L candidate areas may be selected as the learning area. The control station 100, for example, D X May select the candidate area where D is less than a predetermined threshold as the learning area.

[0177] Here, H X is assumed to be the average value of the heights of obstacles in all grids, but H X is not limited to the average value and may be a statistic other than the average value, such as the maximum value.

[0178] In addition, the control station 100 may estimate the probability distribution of the heights of obstacles in each grid of the candidate area and the prediction area, and calculate the similarity of the distribution based on KL divergence or the like. The control station 100 selects the candidate area with the smallest calculated KL divergence, that is, the candidate area with the highest similarity, as the learning area. Alternatively, the control station 100 may select L candidate areas as the learning area from the ones with smaller KL divergence, that is, from the ones with higher similarity. For example, the control station 100 may select the candidate area where the KL divergence is less than a predetermined threshold, that is, the candidate area with a similarity higher than a predetermined value, as the learning area.

[0179] In addition, for example, when calculating the statistic of the height of obstacles in each area, the control station 100 may calculate the statistic taking into account the elevation of each area.

[0180] Furthermore, the control station 100 may, for example, calculate the similarity between the candidate area and the predicted area using an algorithm for calculating image similarity (hereinafter also referred to as the image similarity algorithm) for an image in which the grid of each area is considered as a pixel and the statistical quantity in the grid (in this case, the height of the obstacle) is considered as the pixel value. Examples of image similarity algorithms include MSE (Mean Square Error), MAE (Mean Absolute Error), NCC (Normalized Cross-Correlation), and PSNR (Peak Signal-to-Noise Ratio).

[0181] For example, the control station 100 can calculate similarity using an image similarity algorithm for candidate image information in which the grid of the candidate area is considered as pixels and the height of obstacles in the grid is considered as pixel values, and for candidate image information in which the grid of the predicted area is considered as pixels and the height of obstacles in the grid is considered as pixel values.

[0182] Alternatively, for example, the control station 100 may generate an estimation model using similarity as a weighting coefficient. For example, the control station 100 may use the above D X An estimation model is generated using these as weight coefficients. For example, control station 100 selects the first to third learning areas R01, R02, and R03, and generates an estimation model that estimates the average RSRP at a certain point in the prediction area using measured RSRP data in these first to third learning areas R01, R02, and R03.

[0183] In this case, the control station 100 is D as shown in equation (2) below. X You may also calculate a weighted average using these as weighting coefficients.

[0184]

number

[0185] Here, D1, D2, and D3 are the differences in the mean values ​​of the statistics regarding obstacle height between the first to third learning areas R01, R02, and R03 and the prediction area, respectively (hereinafter also simply referred to as the differences in mean values). P1, P2, and P3 (dBm) are the RSRPs of arbitrary points in the first to third learning areas R01, R02, and R03, respectively.

[0186] By using the reciprocal of the difference in mean values ​​as the weight coefficient, the control station 100 can multiply the RSRP of the learning area with a small difference in mean values ​​by a larger weight, thereby enabling it to predict the RSRP of the prediction area with higher accuracy.

[0187] In this example, it is assumed that the control station 100 predicts the RSRP of the prediction area, but the communication characteristics predicted by the control station 100 are not limited to RSRP. Other communication characteristics are similarly D X This can be calculated as a weighting coefficient.

[0188] <3-2. Second Selection Method> For example, the control station 100 uses the density of obstacles as obstacle information for the candidate area and the predicted area to determine the correlation of the wireless environment.

[0189] Furthermore, the control station 100 divides each area into a grid in the same manner as the first selection method.

[0190] Figure 13 shows examples of candidate areas and predicted areas in a second selection method according to an embodiment of the present disclosure. Figure 13(a) shows an example of a first candidate area, Figure 13(b) shows an example of a second candidate area, and Figure 13(c) shows an example of a predicted area.

[0191] Here, we assume there are two candidate areas, a first candidate area and a second candidate area. However, the number of candidate areas is not limited to two; there may be one, three or more, or any number of candidate areas.

[0192] As shown in Figure 13, each area is divided into, for example, a square grid. Each grid represents, for example, any receiving point within each area.

[0193] The control station 100 determines, for example, the presence or absence of obstacles in each grid. In Figure 13, the presence or absence of hatching in each grid indicates the presence or absence of obstacles. For example, grids with hatching in each area are grids where some kind of obstacle exists. White grids without hatching are grids where no obstacles exist.

[0194] The control station 100, for example, compares the candidate area with the predicted area and selects the candidate area that is similar to the predicted area as the learning area. That is, when selecting the learning area, the control station 100 uses the similarity between the candidate area and the predicted area in terms of the presence or absence of obstacles.

[0195] In the example shown in Figure 13, the first candidate area (see Figure 13(a)) is more similar to the predicted area (see Figure 13(c)) than the second candidate area (see Figure 13(b)). Therefore, the control station 100 selects the first candidate area as the learning area and generates an estimation model using this first candidate area.

[0196] For example, the control station 100 may calculate the similarity of obstacle densities in the candidate area and the prediction area using the difference in obstacle densities. For example, the control station 100 calculates the similarity using the following equation (3).

[0197]

number

[0198] Here, a X b is the density of obstacles in candidate area X. X ζ represents the density of obstacles in the predicted area. X This represents the difference in obstacle density between the candidate area X and the predicted area.

[0199] This ζ XThe smaller the value, that is, the smaller the difference in the density of obstacles between candidate area X and the predicted area, the more the control station 100 determines that candidate area X and the predicted area are similar, that is, that they have a high degree of similarity.

[0200] Control station 100 is, for example, ζ X The candidate area with the smallest size is selected as the learning area. Alternatively, the control station 100 selects, for example, ζ X The control station 100 may select L candidate areas as learning areas, starting with the smallest ones. X Candidate areas where the value is below a predetermined threshold may be selected as the learning area.

[0201] The density of obstacles may be defined, for example, by the number of obstacles per unit area in each area. In this case, it is desirable that the unit area values ​​for the candidate area and the prediction area be unified, i.e., the same.

[0202] Furthermore, similar to the first selection method, the control station 100 may estimate the probability distribution of obstacle density in each grid of the candidate area and the prediction area, and calculate the similarity of said distributions based on KL divergence or the like.

[0203] The control station 100 may, similar to the first selection method, for example, use pixels to represent the grid of each area and binary pixel values ​​to indicate the presence or absence of obstacles in the grid, and calculate the similarity between the candidate area and the predicted area using an algorithm that calculates image similarity.

[0204] The control station 100, similar to the first selection method, for example, the control station 100, ζ X You may also use this as a weighting coefficient to predict communication characteristics.

[0205] <3-3. Third Selection Method> For example, the control station 100 determines the correlation of the wireless environment based on the propagation characteristics of the candidate area and the predicted area. Examples of propagation characteristics include parameters that can affect communication characteristics, such as diffraction loss, clutter loss, reflectance, refractive index, distance attenuation and its attenuation coefficient, shadowing and its mean value and standard deviation, fading deviation and clearance coefficient.

[0206] The control station 100 estimates the propagation characteristics, for example, by simulation. Alternatively, the control station 100 may measure the propagation characteristics at some points in each area and use the measured propagation characteristics to determine the correlation between the wireless environment of the candidate area and the predicted area.

[0207] Alternatively, the control station 100 may use the measured propagation characteristics in the candidate area and the propagation characteristics estimated by simulation in the prediction area.

[0208] Alternatively, measured propagation characteristics may be used at some locations within each area, while propagation characteristics estimated by simulation may be used at the remaining locations, meaning that measured and simulated values ​​may be mixed within each area.

[0209] Furthermore, by using the propagation characteristics measured by the control station 100 to determine the correlation between the wireless environment of the candidate area and the predicted area, the influence of estimation errors from simulations can be reduced, and the accuracy of correlation determination can be further improved.

[0210] Figure 14 shows examples of candidate areas and predicted areas in a third selection method according to an embodiment of the present disclosure. Figure 14(a) shows an example of a first candidate area, Figure 14(b) shows an example of a second candidate area, and Figure 14(c) shows an example of a predicted area.

[0211] Here, we assume there are two candidate areas, a first candidate area and a second candidate area. However, the number of candidate areas is not limited to two; there may be one, three or more, or any number of candidate areas.

[0212] As shown in Figure 14, each area is divided into, for example, a square grid. Each grid represents, for example, any receiving point within each area.

[0213] The control station 100 acquires propagation characteristics for each grid, for example. In Figure 14, the differences in propagation characteristics of each grid are represented by differences in hatching. For example, grids with the same hatching in each area have the same or similar propagation characteristics. Here, similar propagation characteristics may mean, for example, that the difference in propagation characteristics of grids with the same hatching is less than a predetermined value.

[0214] The control station 100, for example, compares the candidate area with the predicted area and selects the candidate area that is similar to the predicted area as the learning area. In other words, when selecting the learning area, the control station 100 uses the similarity of the propagation characteristics between the candidate area and the predicted area.

[0215] In the example shown in Figure 14, the first candidate area (see Figure 14(a)) is more similar to the predicted area (see Figure 14(c)) than the second candidate area (see Figure 14(b)). Therefore, the control station 100 selects the first candidate area as the learning area and generates an estimation model using this first candidate area.

[0216] For example, the control station 100 may calculate the similarity of propagation characteristics in the candidate area and the prediction area using the difference in propagation characteristics. For example, the control station 100 calculates the similarity using the following equation (4).

[0217]

number

[0218] Here, c XThis is the average value of the propagation characteristics in candidate area X. X λ is the average value of the propagation characteristics in the prediction area. X This represents the difference in the average values ​​of propagation characteristics between the candidate area X and the predicted area.

[0219] This λ X The smaller the value, that is, the smaller the difference in the density of obstacles between candidate area X and the predicted area, the more the control station 100 determines that candidate area X and the predicted area are similar, that is, that they have a high degree of similarity.

[0220] Control station 100 is, for example, λ X The candidate area with the smallest value is selected as the learning area. Alternatively, the control station 100 selects, for example, λ X The control station 100 may select L candidate areas as learning areas, starting with the smallest ones. For example, λ X Candidate areas where the value is below a predetermined threshold may be selected as the learning area.

[0221] In addition, similar to the first and second selection methods, the control station 100 may estimate the probability distribution of propagation characteristics in each grid of the candidate area and the prediction area, and calculate the similarity of said distributions based on KL divergence or the like.

[0222] The control station 100 may, similar to the first and second selection methods, for example, use pixels to represent the grid of each area and pixel values ​​to represent the propagation characteristics in the grid, and calculate the similarity between the candidate area and the predicted area using an algorithm that calculates image similarity.

[0223] The control station 100, similar to the first and second selection methods, for example, the control station 100, λ X You may also use this as a weighting coefficient to predict communication characteristics.

[0224] Furthermore, the control station 100 may use ray tracing or existing propagation models such as those shown below when estimating propagation characteristics. The propagation model used is not limited to the examples listed below, and other propagation models may also be used. - Okumura-Hata style - Extended Qin style - Free-space propagation loss -ITM model -ITU-R P.452 -ITU-R P.2108 - Knife-edge diffraction model - Propagation models for macrocells or microcells as defined by 3GPP(registered trademark)

[0225] Furthermore, the control station 100 may estimate the propagation characteristics by taking into account information about the antenna. Examples of information about the antenna include, for example, the antenna pattern, and more specifically, at least one of the following: antenna height, antenna gain, tilt angle, and beam direction.

[0226] <3-4. Fourth Selection Method> For example, the control station 100 determines the correlation of the wireless environment based on LOS / NLOS information in the candidate area and the predicted area.

[0227] The control station 100 can determine LOS / NLOS in each grid of the area using information such as the height of surrounding structures, the distance between the transmitter and receiver, and the height of the transmitter and receiver.

[0228] The control station 100 may determine LOS / NLOS using, for example, at least one of the following methods. - Ray tracing simulation - Propagation models capable of LOS / NLOS determination, such as ITU-R P.452. - Calculation of shielding ratio in the first Fresnel zone - Utilization of topographic and building data

[0229] When determining LOS / NLOS using the shielding ratio of the first Fresnel zone, the control station 100 considers a receiving point to be LOS if, for example, the first Fresnel radius at that point is secured by r(%). For example, if the first Fresnel radius is 10m and 60% of that (r=60), or 4m, is secured, the control station 100 considers it to be LOS, and if it is shielded for a longer distance than 4m (in other words, 60% is not secured), it considers it to be NLOS.

[0230] Furthermore, when using terrain and building data to determine LOS / NLOS, the control station 100 considers, for example, a path connecting the transmitting and receiving points and determines whether or not this path is obstructed by terrain or buildings.

[0231] For example, if LOS / NLOS determinations are made at multiple determination points within the grid, the control station 100 may perform LOS / NLOS determinations on the grid according to the determination results at each determination point.

[0232] For example, if the number of locations determined to be LOS at each determination point is greater than the number of locations determined to be NLOS, the control station 100 determines that the grid is an LOS grid. Also, for example, if the number of locations determined to be LOS at each determination point is less than the number of locations determined to be NLOS, the control station 100 determines that the grid is an NLOS grid. For example, if the number of locations determined to be LOS at each determination point is the same as the number of locations determined to be NLOS, the control station 100 determines that the grid is a mixed LOS / NLOS grid (hereinafter also simply referred to as a mixed grid).

[0233] For example, if at least one of the determination points within the grid is determined to be LOS, the control station 100 determines that the grid is an LOS grid. Alternatively, for example, if at least one of the determination points within the grid is determined to be NLOS, the control station 100 determines that the grid is an NLOS grid.

[0234] For example, if the number of locations determined to be LOS within a grid is greater than or equal to a first threshold, the control station 100 determines that the grid is an LOS grid. If the number of locations determined to be LOS is less than the first threshold, and the number of locations determined to be NLOS is greater than or equal to a second threshold, the control station 100 determines that the grid is an NLOS grid. Furthermore, if the number of locations determined to be LOS is less than the first threshold, and the number of locations determined to be NLOS is less than the second threshold, the control station 100 determines that the grid is a mixed grid.

[0235] Note that the first threshold and the second threshold may be the same or different.

[0236] For example, if the number of locations determined to be NLOS within a grid is equal to or greater than a second threshold, the control station 100 may determine that the grid is an NLOS grid. In this case, if the number of locations determined to be NLOS is less than the second threshold, and the number of locations determined to be LOS is equal to or greater than the first threshold, the control station 100 determines that the grid is an LOS grid. Furthermore, if the number of locations determined to be LOS is less than the first threshold, and the number of locations determined to be NLOS is less than the second threshold, the control station 100 determines that the grid is a mixed grid.

[0237] Figure 15 shows examples of candidate areas and predicted areas in a third selection method according to an embodiment of the present disclosure. Figure 15(a) shows an example of a first candidate area, Figure 15(b) shows an example of a second candidate area, and Figure 15(c) shows an example of a predicted area.

[0238] Here, we assume there are two candidate areas, a first candidate area and a second candidate area. However, the number of candidate areas is not limited to two; there may be one, three or more, or any number of candidate areas.

[0239] As shown in Figure 15, each area is divided into, for example, a square grid. Each grid represents, for example, any receiving point within each area.

[0240] The control station 100 determines, for example, whether the grid in each area is LOS / NLOS. Here, the control station 100 also determines whether the grid is LOS / NLOS or mixed LOS / NLOS.

[0241] Figure 15 shows, for example, that a white grid is an NLOS grid, a grid with coarse hatching is an LOS grid, and a grid with fine hatching is a mixed grid.

[0242] Here, the receiving points in each area are represented by a square grid, but the receiving points may be represented in other ways. For example, each area may be divided into grids other than squares (in other words, unit spatial regions). Examples of grids other than squares include rectangular and circular grids. Also, the receiving points may be represented by points. Examples of points representing receiving points include grid intersections, points represented by latitude and longitude, etc.

[0243] Furthermore, while we have assumed here that each area is divided into 25 grids, the number of divisions is not limited to this. For example, the number of divisions in an area (in other words, the number of grids or receiving points) may be 24 or less, or 26 or more.

[0244] The control station 100, for example, compares the candidate area with the predicted area and selects the candidate area that is similar to the predicted area as the learning area. That is, when selecting the learning area, the control station 100 uses the similarity in the height of obstacles between the candidate area and the predicted area.

[0245] In the example shown in Figure 15, the first candidate area (see Figure 15(a)) is more similar to the predicted area (see Figure 15(c)) than the second candidate area (see Figure 15(b)). Therefore, the control station 100 selects the first candidate area as the learning area and generates an estimation model using this first candidate area.

[0246] For example, the control station 100 compares the difference in the number of NLOS grids between the candidate area and the prediction area, and selects the candidate area with a smaller difference as the learning area. For example, the control station 100 selects the candidate area with the smallest difference in the number of NLOS grids as the learning area.

[0247] Alternatively, the control station 100 may, for example, select L candidate areas as the learning area from the ones with smaller differences in the number of NLOS grids. For example, the control station 100 may select the candidate areas where the difference in the number of NLOS grids is less than a predetermined threshold as the learning area.

[0248] Here, it is assumed that the control station 100 selects the learning area according to the number of NLOS grids. However, instead of the NLOS grids, the learning area may be selected according to the number of LOS grids.

[0249] Also, the control station 100 may, in the same way as the first to third selection methods, estimate the probability distribution of the LOS / NLOS determination results in each grid of the candidate area and the prediction area, and calculate the similarity of the distribution based on the KL divergence or the like. The control station 100 selects the candidate area with the smallest calculated similarity as the learning area. Alternatively, the control station 100 may select L candidate areas as the learning area from the ones with smaller similarities, and for example, may select the candidate areas where the similarity is less than a predetermined threshold as the learning area.

[0250] Also, the control station 100 may, in the same way as the first to third selection methods, for example, regard the grids of each area as pixels, regard the numerical values representing the LOS / NLOS determination results as pixel values, and calculate the similarity between the candidate area and the prediction area using an algorithm for calculating the similarity of images.

[0251] The control station 100 may, in the same way as the first to third selection methods, for example, the control station 100 may predict the communication characteristics using the difference in the number of the above NLOS grids (or LOS grids) as a weight coefficient.

[0252] <3-5. The fifth selection method> In the first to fourth selection methods described above, the control station 100 divides the area into grids, for example, and selects a learning area according to the information for each grid. The selection of learning areas by the control station 100 is not limited to dividing the area into grids. For example, the control station 100 may select a learning area according to map information.

[0253] Figure 16 shows an example of a first candidate area in a fifth selection method according to an embodiment of this disclosure. Part of the map information shown in Figure 16 is the first candidate area R01.

[0254] As shown in Figure 16, the first candidate area R01 corresponds to a section of a major arterial road such as an expressway.

[0255] Figure 17 shows an example of a second candidate area in a fifth selection method according to an embodiment of this disclosure. Part of the map information shown in Figure 17 is the second candidate area R02.

[0256] As shown in Figure 17, the second candidate area R02 corresponds to a part of a densely built-up area such as a suburban residential area.

[0257] Figure 18 shows an example of a prediction area in a fifth selection method according to an embodiment of this disclosure. Part of the map information shown in Figure 18 is the prediction area R11.

[0258] As shown in Figure 18, the predicted area R11 corresponds to a section of a major arterial road such as a highway.

[0259] The control station 100, for example, compares map information of candidate areas and prediction areas, and selects candidate areas similar to the prediction areas as learning areas.

[0260] In the examples shown in Figures 16-18, the first candidate area R01 (see Figure 16) is more similar to the prediction area R11 (see Figure 18) than the second candidate area R02 (see Figure 17). Therefore, the control station 100 selects the first candidate area R01 as the learning area and generates an estimation model using this first candidate area R01.

[0261] The control station 100 may select a learning area based on, for example, the difference in road shape between the candidate area and the prediction area, specifically, at least one difference such as road width, road length, and the number of intersections. Alternatively, the control station 100 may select a learning area based on, for example, the difference in obstacles between the candidate area and the prediction area, specifically, at least one difference such as the shape of buildings and houses, and the abundance of trees.

[0262] The control station 100 may, for example, use image processing or other techniques to select a learning area similar to the prediction area, or it may select a learning area according to instructions from a user or the like.

[0263] <<4. Comparison Methods Based on Area Size>> In the first to fourth selection methods described above, it was assumed that the candidate area and the predicted area were the same size. However, the candidate area and the predicted area may be different in size. Here, we describe an example of how the control station 100 compares the candidate area and the predicted area when their sizes are different.

[0264] (First comparison method) Figure 19 is a diagram illustrating an example of a first comparison method according to an embodiment of the present disclosure. In Figure 19, candidate area R03 and prediction area R12 are shown. Although Figure 19 shows the case where prediction area R12 is larger than candidate area R03, candidate area R03 and prediction area R12 may be swapped. That is, candidate area R03 may be larger than prediction area R12.

[0265] In the first comparison method, the control station 100 divides areas of different sizes, for example, into the same number of grids. That is, the control station 100 divides each area so that the number of grids is the same.

[0266] When the control station 100 divides an area, it compares the candidate area and the predicted area according to the first to fourth selection methods described above, etc., and selects a learning area.

[0267] (Second comparison method) FIG. 20 is a diagram showing an example of the second comparison method according to an embodiment of the present disclosure. In FIG. 20, a candidate area R03 and a predicted area R12 are shown. In FIG. 20, a case where the predicted area R12 is larger than the candidate area R03 is shown, but the candidate area R03 and the predicted area R12 may be interchanged. That is, the candidate area R03 may be larger than the predicted area R12.

[0268] In the second comparison method, the control station 100 divides areas of different sizes, for example, into grids of the same size. When the control station 100 divides an area, it compares the candidate area and the predicted area according to the first to fourth selection methods described above, etc., and selects a learning area.

[0269] Here, when selecting a learning area according to the first to fourth selection methods described above, it may be desirable that the number of grids in the candidate area and the predicted area is the same. In this case, the control station 100 may cut off a part of the area with a larger size to generate an area (hereinafter, also referred to as a divided area) having the same size as the area with a smaller size. The control station 100 compares the divided area with the predicted area (or the candidate area) and selects a learning area.

[0270] FIG. 21 is a diagram showing another example of the second comparison method according to an embodiment of the present disclosure. In FIG. 21, it is assumed that the candidate area R03 is larger than the predicted area R12. In this case, the control station 100 divides the candidate area R03 and extracts a divided area having the same size as the predicted area R12 from the candidate area R03.

[0271] Figure 21 shows an example in which the control station 100 extracts a first divided area R21 and a second divided area R22 from candidate area R03. However, the number of divided areas extracted by the control station 100 is not limited to two; it may be one or three or more.

[0272] Furthermore, the control station 100 may divide the candidate area R03 into a grid and then extract the divided areas, or it may extract the divided areas and then divide these divided areas into a grid.

[0273] Furthermore, when the control station 100 extracts multiple divided areas, it may extract divided areas that partially overlap.

[0274] Furthermore, as described above, cases in which the prediction area is contained within the learning area (corresponding to the candidate area described above) are classified as cases in which the learning area and the prediction area are the same. The selection methods according to this embodiment (for example, the first to fifth selection methods) can also be applied to cases in which the prediction area is contained within the learning area.

[0275] In this case, the learning area (corresponding to the candidate area mentioned above) is larger than the prediction area. Therefore, the control station 100 applies either the first comparison method or the second comparison method to compare the learning area (corresponding to the candidate area mentioned above) with the prediction area and selects the learning area to be used to generate the estimation model.

[0276] For example, the control station 100 may extract a divided area from the learning area that is the same size as the prediction area, and generate an estimation model according to the divided area.

[0277] Figure 22 shows an example of extracting a divided area from a learning area according to the embodiment of this disclosure.

[0278] In the example shown in Figure 22, the control station 100 extracts a first divided area R23 and a second divided area R24 from within the learning area (corresponding to the candidate area described above) if the prediction area is contained within the learning area. The control station 100 then selects a learning area from the extracted divided areas to be used for training the estimation model.

[0279] Here, the control station 100 may, for example, extract a divided area of ​​the same size as the prediction area from a candidate area larger than the prediction area, and select a learning area by applying the first to fifth selection methods to this divided area as a candidate area. In this case, the control station 100 may, for example, extract a divided area randomly.

[0280] Alternatively, the control station 100 may, for example, extract divided areas by considering their similarity to the prediction area, or in other words, their correlation, in the same manner as the first to fifth selection methods. In this case, since the correlation has already been considered, the control station 100 can use the extracted divided areas as learning areas to generate the estimation model.

[0281] Alternatively, the control station 100 may use techniques such as pattern matching to extract divided areas from the candidate areas that are similar to the predicted area (for example, areas with high correlation in propagation characteristics, etc.).

[0282] Alternatively, the control station 100 may extract divided areas considering the geographical continuity of the candidate areas. For example, the control station 100 may extract geographically separated areas as divided areas. This allows the control station 100 to further reduce the bias in the wireless environment between the extracted divided areas.

[0283] The control station 100 generates an estimation model according to the extracted (or selected) divided area.

[0284] In this example, we have extracted the divided areas from the candidate area, but the control station 100 may also extract the divided areas from the prediction area. In this case, the control station 100 may generate the estimation model only for the divided areas within the prediction area.

[0285] Alternatively, the control station 100 may use a divided area instead of a predicted area when comparing the candidate area and the predicted area. In this case, the control station 100 selects a learning area from among the candidate areas according to the result of comparing the candidate area and the divided area. The control station 100 generates an estimation model in the predicted area according to the selected learning area.

[0286] Furthermore, in this example, the control station 100 cuts off a portion of the larger area to create an area of ​​the same size as the smaller area (hereinafter also referred to as the divided area). Alternatively, the control station 100 may integrate the smaller areas to create an area of ​​the same size as the larger area (hereinafter also referred to as the integrated area).

[0287] Figure 23 shows another example of a second comparison method according to an embodiment of the present disclosure. In Figure 23, the predicted area is larger than the candidate area. In this case, the control station 100 combines the multiple candidate areas to generate a combined area that is the same size as the predicted area.

[0288] Figure 23 shows an example in which control station 100 integrates candidate areas R04 to R08 to generate integrated area R30. However, the number of candidate areas that control station 100 integrates is not limited to five; it may be four or fewer, or six or more. Furthermore, the shapes and / or sizes of the candidate areas may all be the same, or at least some of them may differ.

[0289] Furthermore, the control station 100 may generate an integrated area by integrating divided areas, each containing at least a portion of the candidate area.

[0290] Furthermore, the control station 100 may divide the candidate area into a grid and then generate the integrated area, or it may generate the integrated area and then divide this integrated area into a grid.

[0291] <<5. Others>> This section provides definitions (explanations) of terms used in the embodiments described above.

[0292] <5-1. About Terminology> <5-1-1. Regarding communication characteristics> For example, the communication characteristics are one of the following, or a combination thereof. - Characteristics based on radio wave propagation from the transmitting point to the receiving point (downlink, uplink, and sidelink) - Communication parameters at the transmitting point, receiving point, base station 300, terminal device 400 and / or communication node

[0293] The characteristics based on radio wave propagation from the transmitting point to the receiving point include at least one of the following: characteristics related to received power, characteristics related to communication speed (throughput), and characteristics related to delay.

[0294] The characteristics related to received power include information on at least one of the following: received power, interference power, RSRP, RSRQ (Reference Signal Received Quality), RSSI (Received Signal Strength Indicator), SNR, and SINR.

[0295] The characteristics related to communication speed include information on at least one of the following: downlink throughput, uplink throughput, and sidelink throughput.

[0296] The latency-related characteristics include information about at least one of latency, jitter, and ping value.

[0297] Communication parameters at the transmitting point, receiving point, base station 300, terminal equipment 400 and / or communication node include dynamically determined parameters and / or semi-statically determined parameters.

[0298] The dynamically determined parameters include at least one of the following pieces of information: - Information regarding MCS (Modulation and Coding Scheme) - Information regarding transmission power - Information regarding beam control - Information regarding MIMO (Multi Input Multi Output) multiplexing.

[0299] Quasi-statically determined parameters include at least one of the range, maximum value, minimum value, mean value, and median value of parameters that the base station 300 or terminal equipment 400 can select (or is permitted to the base station 300 or terminal equipment 400). An example of a quasi-statically determined parameter is the maximum transmit power.

[0300] <5-1-2. About the Area> In this embodiment, the area (location, base) can be given by any of the following, or a combination thereof. This area corresponds to the predetermined area (learning area, candidate area, and / or prediction area) described above. - Coverage area of ​​a designated base station 300, terminal device 400, or communication node -Land or buildings owned or managed by a designated business operator - Areas designated in advance by specified businesses or government agencies. - Areas divided according to a predetermined method

[0301] An area includes coverage that can be connected to, for example, one communication node (such as a base station 300 or terminal device 400). Also, for example, if multiple base stations 300 are installed within a single site, those base stations 300 are connected to a single core network. In other words, a site can be defined as a coverage area covered by at least one base station 300 connected to a single core network.

[0302] For example, a base station 300 in a private network or an access point name (APN) in the core network may be set as a prediction area for each location. In other words, the same access point name is set for the same location (learning area), and if the access point names are different, they are recognized as different locations.

[0303] One possible method for dividing an area is one based on location information. In this method, the area is divided based on location information such as latitude and longitude, according to predetermined distances.

[0304] <5-1-3. Information on communication environment> Communication environment information is information that may be included in measured data. Communication environment information can be used in statistical processing (e.g., generation of estimation models). Communication environment information can be used to generate estimated data using estimation models. In other words, communication environment information can be an explanatory variable in the estimation model.

[0305] Communication environment information includes at least one of static or quasi-static information and dynamic information. Static or quasi-static information is fixed information or information that is not updated frequently. Static or quasi-static information may be information in a higher communication layer (e.g., application layer, RRC (Radio Resource Control) layer, etc.). Dynamic information is information that is updated frequently. Dynamic information may be information in a lower communication layer (e.g., physical layer, etc.).

[0306] Communication environment information includes, for example, at least one of the following pieces of information: - Map information -Structure information - Equipment information regarding base station 300 or terminal equipment 400 - Sensing information acquired through sensing devices - Wireless communication information related to wireless communication

[0307] (Map information) The map information here refers to information that allows for the recognition of the location and size of structures, base stations 300, and terminal devices 400, etc. The location may be absolute location information such as latitude and longitude, or it may be relative location information within the area.

[0308] Map information includes, for example, topographic information, office layout diagrams, and campus maps.

[0309] (Structure information) The structural information here includes elements that affect radio wave propagation. Examples of these effects include reflection, diffraction, and transmission.

[0310] Structures include, for example, buildings, walls, plantations, roads, signs, traffic lights, road signs, pillars, buildings, ground, glass, windows, desks, cabinets, etc.

[0311] Structural information includes, for example, the location, shape, size, and material of the structure, as well as parameters related to radio wave propagation within that structure (such as dielectric constant and conductivity).

[0312] Structural information is generated and constructed based on, for example, the map information mentioned above. In addition, structural information may also be generated and constructed based on information acquired from sensing devices, as described later.

[0313] (Device information relating to base station 300 or terminal device 400) The device information relating to the base station 300 or terminal device 400 here includes at least one of the following pieces of information: - Antenna information about antennas - Capability information regarding the functions and capabilities supported by wireless communication. - Shape information relating to the shape and weight of the base station 300 and / or terminal device 400 - Location information of the fixed base station 300 and / or fixed terminal device 400

[0314] The antenna information here includes, for example, at least one piece of information regarding the antenna configuration, beam pattern, number of antenna elements, and configuration of the antenna elements of the base station 300 and / or terminal equipment 400.

[0315] (Sensing information acquired through sensing devices) The sensing information acquired through the sensing device here includes object information relating to an object detected through the sensing device, and / or influence information relating to fluctuations and / or effects on wireless communication caused by the detected object.

[0316] Here, sensing devices include cameras and sensors. Sensors include photoelectric sensors, fiber sensors, laser sensors, color sensors, proximity sensors, overcurrent displacement sensors, contact displacement sensors, ultrasonic sensors, image discrimination sensors, pressure sensors, vibration sensors, and inertial measurement sensors.

[0317] Three-dimensional spatial information (for example, the structural information mentioned above) is generated from sensing information acquired through sensing devices. For example, if the sensing information is images or videos acquired in real time by a camera, the three-dimensional spatial information can be generated in real time using, for example, photogrammetry or volumetric capture technology.

[0318] The objects detected by the sensing device include various devices such as the sensing device itself, devices other than the sensing device, and terminal devices 400 that transmit information acquired by the sensing device. Furthermore, the objects detected by the sensing device include the structures mentioned above, as well as objects other than structures.

[0319] The terminal device 400 that transmits sensing information acquired by the sensing device may or may not be equipped with the sensing device. If the terminal device 400 and the sensing device are separate devices, it is preferable that the terminal device 400 acquires sensing information from the sensing device, for example, by wired or wireless communication.

[0320] Furthermore, the sensing information acquired through sensing (sensing information acquired through a sensing device) may include various types of sensing information in addition to object detection information. For example, the sensing information acquired through sensing may include beam information (e.g., information on beam pattern and beam angle) related to the beam transmitted from the base station 300 and / or terminal equipment 400.

[0321] The sensing device described above can detect not only stationary objects such as structures, but also moving objects such as people and robots. For example, the sensing device transmits information about the detected moving object as sensing information via the terminal device 400.

[0322] A sensing device may transmit sensing information when it detects a moving object and / or when it stops detecting a moving object. Alternatively, the sensing device may transmit sensing information at regular intervals.

[0323] (Radio communication information regarding wireless communication) The wireless communication information used here includes, for example, at least one of the following pieces of information: - Communication information regarding RAT (Radio Access Technology) and frequencies - Information regarding the transmission power of the base station 300 or terminal device 400 - Scenario information regarding communication environment scenarios - Restriction information regarding conditions and limitations on wireless communication available on the local network. - Quality information regarding communication quality in wireless communication

[0324] Communication information related to RAT includes, for example, information about LTE, NR, Wi-Fi, Bluetooth®, etc. Communication information related to frequency includes information about at least one of the frequency band, center frequency, and frequency bandwidth.

[0325] Information regarding the transmission power of the base station 300 or terminal device 400 includes, for example, information indicating the transmission power of the SS / PBCH (Synchronization Signal and Physical Broadcast Channel) block included in the SS / PBCH (System information block type 1) contained in the SIB1 (System information block type 1), which is control information broadcast from the base station 300 (ss-PBCH-BlockPower).

[0326] Scenario information regarding communication environment scenarios includes, for example, information on urban areas, suburban areas, rural areas, indoor offices, indoor factories, etc.

[0327] The scenario information may further include information about radio wave propagation models (e.g., path loss models) corresponding to the communication environment scenario. The radio wave propagation models may also correspond to LOS and NLOS environments.

[0328] The constraint information here includes information about the conditions and restrictions on wireless communications permitted on the local network.

[0329] These conditions and constraints may include, for example, information about available RATs, the area where wireless communication is possible (geographic information (such as two-dimensional planar information and / or three-dimensional spatial information including height)), the upper limit of interference power outside the area, the maximum transmittable power, transmittable frequency information, transmittable time information, and the location of the base station 300.

[0330] These conditions and constraints may, for example, be predetermined or defined. Furthermore, these conditions and constraints may be determined and / or modified based on information received from a designated server or storage device (e.g., a SAS (Spectrum Access System) server).

[0331] Quality information relating to communication quality in wireless communication includes, for example, at least one of the following pieces of information measured or estimated by the terminal device 400 in wireless communication: - Received power - Interference power -RSRP -RSRQ -RSSI -SNR - Downlink throughput - Uplink throughput - Latency -jitter -Ping value

[0332] <5-1-4. About Location Information> Location information refers to the location information of the base station 300 and / or terminal device 400. For example, location information may be included in the measured data taken in the area.

[0333] Location information includes absolute location information such as latitude, longitude, and / or altitude, obtained from systems such as GPS (Global Positioning System) and GNSS (Global Navigation Satellite System).

[0334] Alternatively, location information may include relative location information obtained through beacons or UWB (Ultra-WideBand).

[0335] Absolute or relative location information may be areas demarcated by predetermined distances or methods.

[0336] Furthermore, the information (such as measured data) according to this embodiment may be information linked to location information.

[0337] <5-2. About virtual space estimation information> The virtual space estimation information includes, for example, information regarding radio wave propagation at at least one of the base station 300 and the terminal device 400. The virtual space estimation information is, for example, data (information) obtained by simulation.

[0338] Virtual space estimation information may be used as needed when generating estimation models. For example, virtual space estimation information may be used in addition to, or in place of, measured data in generating estimation models.

[0339] Virtual space estimation information can be used as explanatory variables in the estimation model. Alternatively, virtual space estimation information may be used as the dependent variable in the estimation model. For example, virtual space estimation information can be used as student data and teacher data in the estimation model.

[0340] The virtual space estimation information includes, for example, at least one of the following pieces of information: -LOS / NLOS information - Simulation information -Calculated information based on LOS / NLOS information and simulation information.

[0341] (LOS / NLOS information) LOS / NLOS information indicates whether the environment between the base station 300 and the terminal device 400 is an LOS environment or an NLOS environment.

[0342] An LOS environment is also called a line-of-sight environment. An LOS environment is a situation in which there are no obstacles 600 such as structures or people in a straight line between the base station 300 and the terminal device 400, and the base station 300 and the terminal device 400 can directly transmit and receive waves between them. In this case, wireless communication between the base station 300 and the terminal device 400 is carried out not only by direct waves but also by reflected waves and diffracted waves.

[0343] An NLOS environment is also called a non-line-of-sight environment. An NLOS environment is a situation in which there are obstacles 600, such as structures or people, in the straight line between the base station 300 and the terminal device 400, and the base station 300 and the terminal device 400 cannot directly transmit or receive waves between them. In this case, wireless communication between the base station 300 and the terminal device 400 is performed via reflected waves, diffracted waves, or other means besides direct waves.

[0344] (Simulation information) The simulation information includes information regarding the simulation results of radio wave propagation in wireless communication between the base station 300 and the terminal device 400. The simulation information includes, for example, path information for one or more paths (transmitted wave, incoming wave, ray) obtained by ray tracing simulation.

[0345] This path includes direct waves, reflected waves, diffracted waves, transmitted waves, etc., between the base station 300 and the terminal device 400. Generally, there are various structures between the base station 300 and the terminal device 400. Therefore, the signal (radio wave) transmitted from the transmitting point (e.g., base station 300) travels through various paths, becoming multiple paths, before reaching the receiving point (e.g., terminal device 400).

[0346] Path information may include at least one of the following pieces of information: - Received power at the receiving point -Transmit power at the transmission point -path loss - Propagation distance - Number of reflections - Number of diffractions - Number of passes - Phase variation - Launch angle at the transmission point -Angle of arrival at the receiving point - Path arrival order (the chronological order of arrival among multiple paths) - Number of paths

[0347] (Calculated information) The calculated information is generated and calculated based on the LOS / NLOS information and simulation information mentioned above. The calculated information may include at least one of the following pieces of information: - Path loss at the receiving point - Received power - Interference power -RSRP -RSRQ -RSSI -SNR - Downlink throughput - Uplink throughput - Latency -jitter -Ping value

[0348] Here, an example of the virtual space estimation information generation process (information generation process) according to this embodiment will be explained using Figure 24. Figure 24 is a flowchart showing an example of the flow of the information generation process according to this disclosure.

[0349] The information generation process shown in Figure 24 may be performed, for example, by the control station 100 when using virtual space estimation information (simulation data) to generate estimation models and / or estimation data.

[0350] The control station 100 first constructs a virtual communication environment for the area (prediction area and / or learning area) (step S101).

[0351] For example, a virtual communication environment is a three-dimensional virtual space of that area. For example, a virtual communication environment is generated based on communication environment information in that area. For example, a virtual communication environment includes structures (buildings, ground surface, etc.) within that area.

[0352] Next, the control station 100 performs a radio wave propagation simulation between the base station 300 and the terminal device 400 in a virtual communication environment (step S102). Various methods can be used for the radio wave propagation simulation, such as LOS / NLOS environment determination in the virtual space, or ray tracing simulation.

[0353] The control station 100 generates virtual space estimation information based on the simulation results (step S103).

[0354] The information generation process may be performed by a device other than the control station 100. In this case, the control station 100 obtains virtual space estimation information (simulation data) from the device performing the information generation process. Furthermore, the timing of the information generation process is not limited to the generation of the estimation model and / or the estimation data. The control station 100 can perform the information generation process at any time.

[0355] <5-3. Regarding estimation accuracy> For example, if the control station 100 estimates the actual communication characteristics using an estimation model, the estimation accuracy may be given. Alternatively, the estimation accuracy may be given to the estimation model. This estimation accuracy can be further used when using (utilizing) the data estimated by the communication characteristics estimation model.

[0356] The accuracy of the communication characteristics and / or estimation model in the prediction area can be determined by any one or a combination of the following: - Number of learning areas - Accuracy of determining the correlation of wireless environments between each learning area and prediction area. - How to select a learning area - Method for selecting measurement locations - Accuracy of generating the estimation model - Correlation (similarity) of wireless environments in prediction and learning areas. - Accuracy of simulation data in the prediction area and / or learning area - Observation accuracy (measurement accuracy) of actual data in the prediction area and / or learning area.

[0357] Correlations (similarity) of communication environments include, for example, the average height of structures within each area, the density of structures, and the similarity of the height of base stations (including installation location (altitude), building height, antenna height, etc.).

[0358] For example, the accuracy of simulation data can be determined based on the accuracy (precision, accuracy) of the communication environment information used in the simulation to generate the simulation data.

[0359] The observational accuracy of the measured data may include at least one of the accuracy of the communication characteristics and the accuracy of the location information. For example, the error in the measured data may include the measurement error of RSRP and / or the error in the location information obtained by GPS.

[0360] Furthermore, if an observation error specific to the terminal device 400 occurs, information indicating that terminal device 400 may be included in the observation accuracy of the measured data.

[0361] <5-4. Use Cases> The statistical information according to the embodiments described above can be used in various processes, controls, and use cases.

[0362] For example, data on communication characteristics estimated in a prediction area (e.g., predicted values ​​as statistical information) can be used for cell design within that prediction area (such as the location of the transmission point and the setting of the maximum transmission power of the transmission point).

[0363] This cell design may further be performed based on the estimation accuracy of communication characteristics in the predicted area. Furthermore, this cell design may be performed at the control station 100, at the base station 300, or in the core network, etc.

[0364] For example, data on communication characteristics estimated in a prediction area (e.g., predicted values ​​as statistical information) can be used to control communication parameters (transmit power, MCS (Modulation and coding scheme), beam, etc.) at the transmitting and / or receiving points within that prediction area.

[0365] The control of communication parameters may be further performed based on the estimation accuracy of communication characteristics in the predicted area. This control of communication parameters may be performed at the control station 100, at the base station 300, or in the core network, etc.

[0366] The statistical information according to this embodiment can be used, for example, in the design of interference power and separation distance in frequency sharing. The control station 100 can utilize the statistical information of communication characteristics to estimate interference power, thereby effectively utilizing available frequencies in time and space while avoiding interference with the protected system.

[0367] Furthermore, the technology according to this embodiment may be applied to 6G joint communication and sensing.

[0368] <5-5. Normalization of measured data> In this embodiment, if there are multiple learning areas, the simulation data and / or measured data in the learning areas are normalized (an offset is given) by a predetermined method, for example.

[0369] The prescribed method may be carried out based on communication environment information in each learning area (for example, information on transmission power from base station 300).

[0370] Here, communication environment information refers to the information described above. Communication environment information includes, for example, information about the transmission power of base station 300, information about the frequencies used for communication (carrier frequency, frequency bandwidth, etc.), and information about the learning area (coverage, indoor / outdoor information, etc.).

[0371] For example, normalization is performed so that the communication environment information in each learning area is the same as that of the others.

[0372] Simulation data of communication characteristics in a designated area (one of several learning areas) is normalized so that the communication environment information in that designated area is the same as the communication environment information in each of the other learning areas (the remaining learning areas other than the designated area).

[0373] The communication characteristics data for a given area (one of several learning areas) is estimated taking into account the normalization described above.

[0374] The measured data of communication characteristics in a designated area (one of several learning areas) is normalized so that the communication environment information in that designated area is the same as the communication environment information in each of the other learning areas (the remaining learning areas other than the designated area).

[0375] The normalization described above is performed so that the communication environment information in each learning area becomes the communication environment information in the predetermined area (one of the multiple learning areas) described above.

[0376] Normalization is performed, for example, by a device that generates the estimation model (in this embodiment, the control station 100). Note that the device that performs normalization and the device that generates the estimation model may be different.

[0377] The following are specific examples of normalization.

[0378] For example, consider the case where the communication characteristics of the simulation data and / or measured data in learning area R01 are RSRP for the downlink.

[0379] Consider the case where the transmission power of base station 300 in learning area R01 is B1 (dBm), the transmission power of base station 300 in learning area R02 is B2 (dBm), and the transmission power of base station 300 in learning area R03 is B3 (dBm).

[0380] To normalize the transmitted power so that it equals S (dBm), the control station 100 applies an offset of S-B1 (dBm) to the RSRP in learning area R01. The control station 100 applies an offset of S-B2 (dBm) to the RSRP in learning area R02. The control station 100 applies an offset of S-B3 (dBm) to the RSRP in learning area R03. In this way, the control station 100 performs normalization by applying an offset to each RSRP (each data).

[0381] Here, S(dBm) could be the transmission power of base station 300 in the prediction area where the estimation is being performed.

[0382] <<6. Hardware Configuration Example>> Figure 25 shows an example of the hardware configuration of the device. The control station 100 described above is implemented, for example, by the computer 1000 shown in Figure 25.

[0383] Computer 1000 has a CPU 1100, RAM 1200, ROM 1300, HDD (Hard Disk Drive) 1400, communication interface 1500, and input / output interface 1600. The various parts of computer 1000 are connected by bus 1050.

[0384] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, and controls various parts. For example, the CPU 1100 loads the programs stored in the ROM 1300 or HDD 1400 into the RAM 1200 and executes processing corresponding to the various programs.

[0385] ROM1300 stores boot programs such as the BIOS (Basic Input Output System) executed by CPU1100 when computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0386] HDD1400 is a computer-readable storage medium that non-temporarily stores programs executed by CPU1100 and data used by such programs. Specifically, HDD1400 is a storage medium that stores a program for the information processing method according to this disclosure, which is an example of program data 1450.

[0387] The communication interface 1500 is an interface for the computer 1000 to connect to an external network 1550 (e.g., the Internet). For example, the CPU 1100 can receive data from other devices or transmit data it has generated to other devices via the communication interface 1500.

[0388] The input / output interface 1600 is an interface for connecting the input / output device 1650 and the computer 1000. For example, the CPU 1100 receives data from input devices such as a keyboard or mouse via the input / output interface 1600. The CPU 1100 also transmits data to output devices such as a display, speaker, or printer via the input / output interface 1600. The input / output interface 1600 may also function as a media interface for reading programs stored on a predetermined storage medium (media) that is readable by a computer. Examples of media include optical storage media such as DVDs (Digital Versatile Discs) and PDs (Phase Change Rewritable Disks), magneto-optical storage media such as MOs (Magneto-Optical Disks), tape media, magnetic storage media, or semiconductor memory.

[0389] When computer 1000 functions as the control station 100 described above, the CPU 1100 of computer 1000 implements the functions of the control unit 170 by executing a program loaded onto RAM 1200. The program may also be stored on HDD 1400. The CPU 1100 reads and executes program data 1450 from HDD 1400, but as an alternative, the program may be obtained from another device via an external network 1550.

[0390] Each of the above components may be made up of general-purpose materials, or it may be made up of hardware specialized for the function of each component. Such a configuration may be modified as appropriate depending on the technological level at the time of implementation.

[0391] <<7. Other Embodiments>> The processing according to the above-described embodiment may be carried out in various other forms besides those described above.

[0392] For example, in each of the embodiments described above, the control station 100 performs statistical processing such as selecting a learning area, generating an estimation model, and performing prediction processing using the estimation model. However, the device that performs these processes is not limited to the control station 100. For example, the base station 300 may perform at least one of these processes. In this case, the base station 300 obtains information from the control station 100 and / or the terminal device 400 to perform at least one of these processes.

[0393] Alternatively, the terminal device 400 may perform at least one of these processes. In this case, the terminal device 400 obtains information from the control station 100 and / or base station 300 to perform at least one of these processes.

[0394] Furthermore, in the embodiments described above, the same single device (for example, control station 100) performed all of the learning area selection process, the estimation model generation process, and the prediction process using the estimation model. However, the devices that perform these processes may be different.

[0395] For example, the control station 100 may perform the selection process and the generation process, and the base station 300 may perform the prediction process. In this case, the control station 100 obtains the information used for the selection process and the generation process from the base station 300 and / or the terminal device 400. The base station 300 obtains the information used for the prediction process from the control station 100 and / or the terminal device 400.

[0396] Alternatively, for example, the control station 100 may perform a selection process, the base station 300 may perform a generation process, and the terminal device 400 may perform a prediction process. In this case, the control station 100 obtains information to be used in the selection process from the base station 300 and / or the terminal device 400. The base station 300 obtains information to be used in the generation process from the control station 100 and / or the terminal device 400. The terminal device 400 obtains information to be used in the prediction process from the control station 100 and / or the base station 300.

[0397] Alternatively, for example, the base station 300 may perform the selection process and the generation process, and the terminal device 400 may perform the prediction process. In this case, the base station 300 obtains the information used for the selection process and the generation process from the control station 100 and / or the terminal device 400. The terminal device 400 obtains the information used for the prediction process from the control station 100 and / or the base station 300. Note that the terminal device 400 may perform the selection process and the generation process, and the base station 300 may perform the prediction process.

[0398] In other words, in this embodiment, the control station 100 may be replaced with a base station 300 or a terminal device 400 as appropriate. The base station 300 may be replaced with a control station 100 or a terminal device 400 as appropriate. The terminal device 400 may be replaced with a control station 100 or a base station 300 as appropriate.

[0399] For example, the control device that controls the control station 100, base station 300, and terminal device 400 of the above embodiment may be implemented by a dedicated computer system or by a general-purpose computer system.

[0400] For example, a communication program for performing the above-described operations is stored in a computer-readable recording medium such as an optical disc, semiconductor memory, magnetic tape, or flexible disk and distributed. Then, for example, the control device is configured by installing the program on a computer and executing the above-described process. In this case, the control device may be an external device (e.g., a personal computer) of the control station 100, base station 300, and terminal device 400. Alternatively, the control device may be an internal device (e.g., control units 130, 340, 450) of the control station 100, base station 300, and terminal device 400.

[0401] Alternatively, the above communication program may be stored on a disk device provided by a server on a network such as the Internet, and made available for download to a computer. Furthermore, the above functions may be realized through the cooperation of an OS (Operating System) and application software. In this case, the parts other than the OS may be stored on a medium and distributed, or the parts other than the OS may be stored on a server device and made available for download to a computer.

[0402] Furthermore, among the processes described in each of the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above documents and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0403] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0404] Furthermore, the embodiments and modifications described above can be combined as appropriate, provided that the processing content is not inconsistent.

[0405] Furthermore, the effects described herein are merely illustrative and not limiting; other effects may also occur.

[0406] Furthermore, for example, this embodiment can also be implemented as any configuration that constitutes a device or system, such as a processor as a system LSI (Large Scale Integration), a module using multiple processors, a unit using multiple modules, or a set with additional functions added to a unit (i.e., a configuration of a part of a device).

[0407] In this embodiment, a system refers to a collection of multiple components (devices, modules (parts), etc.), regardless of whether all components are located in the same enclosure. Therefore, multiple devices housed in separate enclosures and connected via a network, and a single device containing multiple modules within a single enclosure, are both considered systems.

[0408] Furthermore, for example, this embodiment can adopt a cloud computing configuration in which a single function is shared and processed collaboratively by multiple devices via a network.

[0409] In the embodiments described above, the case in which the control station 100 determines the control information of the base station 300 and / or terminal equipment 400 has been explained, but the embodiments are not limited to this. Each of the embodiments described above can be used for the purpose of determining and designing transmission parameters and / or reception parameters, the number of base stations 300 and / or terminal equipment 400 to be installed (maximum number to be installed), the installation location, and / or the installation direction (horizontal direction, tilt angle, etc.).

[0410] <<8. Conclusion>> The effects described in this disclosure are illustrative and not limited to those disclosed. Other effects may also occur.

[0411] While embodiments of this disclosure have been described above, the technical scope of this disclosure is not limited to the embodiments described above, and various modifications are possible without departing from the spirit of this disclosure. Furthermore, components from different embodiments and modifications may be combined as appropriate.

[0412] Furthermore, this technology can also be configured as follows. (1) The system includes a control unit that generates an estimation model for estimating the data relating to the communication characteristics in a second area using data relating to the communication characteristics in a first area, The control unit generates the estimation model using the data relating to the communication characteristics in one or more of the first areas that are similar to the second area. Information processing device. (2) The information processing apparatus according to (1), wherein the control unit generates the estimation model using the data relating to the communication characteristics in one or more of the first areas, the communication environment of which is similar to that of the second area. (3) The information processing apparatus according to (1) or (2), wherein the control unit generates the estimation model using the data relating to the communication characteristics in the first area, which has a high similarity between the statistics relating to the communication environment in the first area and the statistics relating to the communication environment in the second area. (4) The information processing apparatus according to (3), wherein the control unit generates the estimation model using the data relating to the communication characteristics of N (N is an integer between 1 and M) of the first areas that have a high similarity among M (M is an integer of 2 or more) of the first areas. (5) The information processing apparatus according to (3), wherein the control unit generates the estimation model using data relating to the communication characteristics in a first area, where the similarity between the statistical quantity relating to the communication environment in the first area and the statistical quantity relating to the communication environment in the second area is greater than or equal to a predetermined value. (6) The information processing apparatus according to any one of (3) to (5), wherein the control unit generates the estimation model using the data relating to the communication characteristics in the first area which has a high similarity between the statistical quantity corresponding to the unit spatial region of the first area and the statistical quantity corresponding to the unit spatial region of the second area. (7) The aforementioned similarity is calculated using an image similarity algorithm for first image information in which the unit space region of the first area is considered a pixel and the statistical quantity corresponding to the unit space region is considered a pixel value, and second image information in which the unit space region of the second area is considered a pixel and the statistical quantity corresponding to the unit space region is considered a pixel value. (6) The information processing device described above. (8) The information processing apparatus according to any one of (3) to (7), wherein the statistic is the statistic relating to obstacles in the first area and the second area. (9) The information processing apparatus according to (7), wherein the statistic is a statistic relating to at least one of the height and density of the obstacle. (10) The information processing apparatus according to any one of (3) to (7), wherein the statistic is the statistic relating to the propagation characteristics of the first area and the second area. (11) The information processing apparatus according to any one of (3) to (7), wherein the statistic is the statistic regarding LOS / NLOS of the first area and the second area. (12) The control unit generates the estimation model using the data relating to the communication characteristics in the first area, with the similarity between the statistics relating to the communication environment in the first area and the statistics relating to the communication environment in the second area as weights. An information processing device described in any one of (1) to (11). (13) The information processing apparatus according to (1) or (2), wherein the control unit generates the estimation model using the data relating to the communication characteristics in the first area, which has a high similarity to the map information relating to the second area. (14) The information processing apparatus according to any one of (1) to (13), wherein the first area is an area obtained by dividing a third area which is larger than the second area. (15) The information processing device according to any one of (1) to (13), wherein the first area is an area formed by integrating several fourth areas that are smaller than the second area. (16) The control unit, A request for the calculation of the data is obtained from a terminal device located in the second area. The data is calculated using the estimation model described above. The calculated data is transmitted to the terminal device. An information processing device described in any one of (1) to (15). (17) It includes an acquisition unit that acquires data on communication characteristics in a second area estimated by an estimation model, The estimation model is generated using the data relating to the communication characteristics in one or more first areas similar to the second area. Terminal device. (18) The terminal device according to (17), further comprising a communication unit that transmits a request for the calculation of said data to an information processing device that calculates said data relating to the communication characteristics using the estimation model. (19) This includes generating an estimation model for estimating the data relating to the communication characteristics in a second area using data relating to the communication characteristics in a first area, The estimation model is generated using the data relating to the communication characteristics in one or more of the first areas that are similar to the second area. Information processing methods. (20) Computer It functions as a control unit that generates an estimation model for estimating the data on communication characteristics in a second area using data on communication characteristics in a first area. The estimation model is generated using the data relating to the communication characteristics in one or more of the first areas that are similar to the second area. program. [Explanation of Symbols]

[0413] 100 control stations 110,310,410 Communications Department 120,320,420 storage section 130,340,450 Control Unit 300 base stations 330,430 Network Communications Department 400 terminal devices 440 Input / output section

Claims

1. The system includes a control unit that generates an estimation model for estimating the data relating to the communication characteristics in a second area using data relating to the communication characteristics in a first area, The control unit generates the estimation model using the data relating to the communication characteristics in one or more of the first areas that are similar to the second area. Information processing device.

2. The information processing apparatus according to claim 1, wherein the control unit generates the estimation model using the data relating to the communication characteristics in one or more first areas where the communication environment is similar to that of the second area.

3. The information processing apparatus according to claim 1, wherein the control unit generates the estimation model using the data relating to the communication characteristics in the first area, which has a high similarity between the statistics relating to the communication environment in the first area and the statistics relating to the communication environment in the second area.

4. The information processing apparatus according to claim 3, wherein the control unit generates the estimation model using the data relating to the communication characteristics of N (N is an integer between 1 and M) first areas with high similarity among M (M is an integer of 2 or more) first areas.

5. The information processing apparatus according to claim 3, wherein the control unit generates the estimation model using data relating to the communication characteristics in a first area, among a plurality of first areas, in which the similarity between the statistical quantity relating to the communication environment in the first area and the statistical quantity relating to the communication environment in the second area is higher than a predetermined value.

6. The information processing apparatus according to claim 3, wherein the control unit generates the estimation model using the data relating to the communication characteristics in the first area which has a high similarity between the statistical quantity corresponding to the unit spatial region of the first area and the statistical quantity corresponding to the unit spatial region of the second area.

7. The aforementioned similarity is calculated using an image similarity algorithm for first image information in which the unit space region of the first area is considered a pixel and the statistical quantity corresponding to the unit space region is considered a pixel value, and second image information in which the unit space region of the second area is considered a pixel and the statistical quantity corresponding to the unit space region is considered a pixel value. The information processing apparatus according to claim 6.

8. The information processing apparatus according to claim 3, wherein the statistic is the statistic relating to obstacles in the first area and the second area.

9. The information processing apparatus according to claim 7, wherein the statistic is a statistic relating to at least one of the height and density of the obstacle.

10. The information processing apparatus according to claim 3, wherein the statistic is the statistic relating to the propagation characteristics of the first area and the second area.

11. The information processing apparatus according to claim 3, wherein the statistic is the statistic regarding LOS / NLOS of the first area and the second area.

12. The control unit generates the estimation model using the data relating to the communication characteristics in the first area, with the similarity between the statistics relating to the communication environment in the first area and the statistics relating to the communication environment in the second area as weights. The information processing apparatus according to claim 1.

13. The information processing apparatus according to claim 1, wherein the control unit generates the estimation model using the data relating to the communication characteristics in the first area, which has a high similarity to the map information relating to the second area.

14. The information processing apparatus according to claim 1, wherein the first area is an area obtained by dividing a third area which is larger than the second area.

15. The information processing apparatus according to claim 1, wherein the first area is an area formed by integrating a plurality of fourth areas that are smaller than the second area.

16. The control unit, A request for the calculation of the data is obtained from a terminal device located in the second area. The data is calculated using the estimation model described above. The calculated data is transmitted to the terminal device. The information processing apparatus according to claim 1.

17. It includes an acquisition unit that acquires data on communication characteristics in a second area estimated by an estimation model, The estimation model is generated using the data relating to the communication characteristics in one or more first areas similar to the second area. Terminal device.

18. The terminal device according to claim 17, further comprising a communication unit that transmits a request for the calculation of said data to an information processing device that calculates said data regarding the communication characteristics using the estimation model.

19. This includes generating an estimation model for estimating the data relating to the communication characteristics in a second area using data relating to the communication characteristics in a first area, The estimation model is generated using the data relating to the communication characteristics in one or more of the first areas that are similar to the second area. Information processing methods.

20. Computer It functions as a control unit that generates an estimation model for estimating the data on communication characteristics in a second area using data on communication characteristics in a first area. The estimation model is generated using the data relating to the communication characteristics in one or more of the first areas that are similar to the second area. program.

Images