Control device, optimization system, control method, and program

Abstract

In this optimization system for optimizing a radio environment in a radio communication network, to facilitate optimization of a radio quality of a radio communication network of a plurality of tenants, this control device uses radio environment data relating to the tenants providing the radio communication network to provide each of the tenants with feedback of tenant-specific setting information.

Description

Control device, optimization system, control method, and program

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

[0002] BACKGROUND ART There are known optimization techniques for wireless communication networks that use techniques such as radio wave propagation estimation, station layout design, installation, radio state recognition, and quality prediction in wireless communication networks to track or adapt the wireless communication network to the situation.

[0003] Ryuichi Takechi, Koji Ogawa, Masato Okuda, "Wireless Network Optimization Technology: SON", FUJITSU. 62, 4, pp.449-454 (July 2011). Multi-wireless proactive control technology Cradio(r) (Cradio 1.0 system),<https: / / www.rd.ntt / as / history / wireless / wi0519.html> ,Internet,[Retrieved November 1, 1993].

[0004] In conventional wireless communication network optimization techniques, data on the wireless communication network of a single tenant is collected and the wireless quality of the wireless communication network of that tenant is optimized. However, this method builds an optimization model for the wireless communication network that depends on the wireless environment of that tenant, which poses a problem in that when optimizing the wireless quality of the wireless communication network of another tenant, a separate optimization model must be built.

[0005] An embodiment of the present invention has been made in consideration of the above-mentioned problems, and enables an optimization system that optimizes the wireless environment of a wireless communication network to easily optimize the wireless quality of a wireless communication network for multiple tenants.

[0006] In order to solve the above problem, a control device according to an embodiment of the present invention provides a wireless communication network to a plurality of tenants, and based on wireless environment data of the plurality of tenants, feeds back configuration information of each tenant to the plurality of tenants.

[0007] According to an embodiment of the present invention, in an optimization system that optimizes the wireless environment of a wireless communication network, it becomes possible to easily optimize the wireless quality of a wireless communication network of multiple tenants.

[0008] FIG. 1 is a diagram illustrating an example of a configuration of an optimization system according to the present embodiment. FIG. 2 is a diagram for explaining an overview of optimization processing of a wireless communication network according to the present embodiment. FIG. 3 is a diagram for explaining an overview of processing of the optimization system according to the present embodiment. FIG. 4 is a diagram illustrating an overview of optimization processing of an estimation model according to Example 1. FIG. 5 is a sequence diagram illustrating an example of optimization processing of an estimation model according to Example 1. FIG. 6 is a diagram illustrating an overview of optimization processing of a control model according to Example 1. FIG. 7 is a sequence diagram illustrating an example of optimization processing of a control model according to Example 1. FIG. 8 is a diagram for explaining an example of optimization of a control model according to Example 1. FIG. 9 is a sequence diagram illustrating an example of optimization of a wireless communication network according to Example 1. FIG. 10 is a diagram illustrating an example of automatic control information according to Example 1. FIG. 11 is a diagram for explaining optimization of a model according to Example 2. FIG. 12 is a diagram illustrating an example of categorization of a wireless environment according to Example 2. FIG. 13 is a diagram illustrating an example of optimization of an estimation model by machine learning. FIG. 14 is a diagram illustrating an overall view of an optimization system according to Example 2. FIG. 15 is a diagram illustrating an example of a hardware configuration of a computer according to the present embodiment.

[0009] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

[0010] <Configuration of Optimization System> Fig. 1 is a diagram showing an example of the configuration of an optimization system according to this embodiment. The optimization system 100 is a system that optimizes the wireless environment of each tenant based on wireless environment data acquired from multiple tenants 10a, 10b, 10c, ... that provide a wireless communication network. In the following description, "tenant 10" will be used to refer to any tenant among the multiple tenants 10a, 10b, 10c, ... The tenant 10 is, for example, a business operator that provides a wireless communication network.

[0011] The wireless communication network may include various wireless communication networks such as a wireless local area network (LAN), a local 5G (5th Generation), and 5G, etc. The wireless communication network also includes various wireless devices such as a base station, a relay station, an access point, and a terminal.

[0012] In the example of FIG. 1 , the optimization system 100 includes a control device 101, an acquisition unit 102, an estimation unit 103, a determination unit 104, an analysis unit 105, an estimation model optimization unit 106, a control model optimization unit 107, a setting control unit 108, a classification unit 109, and a storage unit 110.

[0013] The control device 101 is an information processing device having a computer configuration or a system including multiple computers. The control device 101 is communicably connected to multiple tenants 10 and controls each functional block of the optimization system 100 to feed back configuration information for each tenant 10 to the multiple tenants 10 based on wireless environment data of the multiple tenants 10.

[0014] The acquisition unit 102 executes an acquisition process to acquire wireless environment data of a plurality of tenants 10 for which a wireless communication network is provided. For example, the acquisition unit 102 acquires wireless environment data of each tenant 10 from a plurality of tenants 10a, 10b, 10c, ... via the control device 101. This wireless environment data includes, for example, operation information of wireless devices constituting the wireless communication network of each tenant 10, quality information indicating communication quality, and the like.

[0015] The estimation unit 103 executes estimation processing to estimate the wireless quality of the wireless communication network of each tenant 10 based on the wireless environment data of each tenant 10. The estimation unit 103 according to this embodiment estimates the wireless quality of each tenant 10 by inputting the wireless environment data of each tenant 10 and the like into an estimation model optimized based on the wireless environment data of the multiple tenants 10.

[0016] The determination unit 104 executes a determination process to determine setting information that optimizes the wireless environment of each tenant 10, based on the wireless environment data of each tenant 10. The determination unit 104 according to this embodiment determines setting information that optimizes the wireless environment of each tenant 10 by inputting the wireless environment data of each tenant, etc., into a control model that has been optimized based on the wireless environment data of a plurality of tenants 10.

[0017] The analysis unit 105 analyzes the wireless quality of the wireless communication network of each tenant and executes an analysis process to detect a predetermined condition to be controlled. For example, the analysis unit 105 analyzes whether the wireless quality of each tenant 10 estimated by the estimation unit 103 satisfies the predetermined condition to be controlled.

[0018] The estimation model optimization unit 106 executes an estimation model optimization process to optimize an estimation model that estimates the wireless quality of each tenant 10 based on the wireless environment data of the plurality of tenants 10 .

[0019] The control model optimization unit 107 executes a control model optimization process that optimizes a control model that outputs setting information that optimizes the wireless environment of each tenant 10 based on the wireless environment data of the multiple tenants 10 .

[0020] The setting control unit 108 executes a setting control process to set the setting information determined by the determination unit 104 in each tenant 10 or the wireless device of each tenant 10 under control of the control device 101 .

[0021] The classification unit 109 categorizes the wireless environment of each tenant 10 and executes a classification process to classify a plurality of tenants 10 according to the patterns of their wireless environments.

[0022] The memory unit 110 stores various data, information, programs, etc., including, for example, wireless environment data of multiple tenants 10 acquired by the acquisition unit 102, estimation results by the estimation unit 103, and setting information determined by the determination unit 104.

[0023] Each of the functional blocks, such as the acquisition unit 102, the estimation unit 103, the determination unit 104, the analysis unit 105, the estimation model optimization unit 106, the control model optimization unit 107, the setting control unit 108, and the classification unit 109, is realized, for example, by a program executed by a computer included in the optimization system 100. The storage unit 110 is realized, for example, by a storage device or the like of the computer included in the optimization system 100. Note that each of the above functional blocks is not limited to being implemented by a physical machine, and may be realized by a program executed by a virtual machine or the like on a cloud.

[0024] With the above configuration, the optimization system 100 grasps the wireless environment of each tenant 10, estimates the wireless quality of each tenant 10, and when a specified condition to be controlled is detected, performs closed-loop control by feeding back setting information that optimizes the wireless quality of each tenant 10.

[0025] 2 is a diagram for explaining an overview of the optimization process of a wireless communication network according to this embodiment. The optimization system 100 grasps the wireless environment by acquiring wireless environment data indicating the wireless environment of the wireless communication network 201 from, for example, wireless devices 202 constituting the wireless communication network 201 and terminals 203 connected to the wireless devices 202 (step S1).

[0026] Next, the optimization system 100 estimates the wireless quality of the wireless communication network 201 (step S2). For example, the optimization system 100 estimates the wireless quality of the wireless communication network 201 by inputting the wireless environment data of the wireless communication network 201 into an estimation model 211 that estimates the wireless quality of the wireless communication network.

[0027] Furthermore, when the wireless quality of the wireless communication network 201 satisfies a predetermined control target condition, the optimization system 100 determines setting information that optimizes the wireless quality of the wireless communication network 201. For example, the optimization system 100 determines the setting information by inputting wireless environment data of the wireless communication network 201 to a control model 212 that outputs setting information that optimizes the wireless quality of the wireless communication network 201. Furthermore, the optimization system 100 dynamically controls the wireless quality of the wireless communication network 201 by feeding back the determined setting information to the wireless devices 202 and the like of the wireless communication network 201 (step S3).

[0028] The optimization system 100 can optimize the wireless communication network quality based on the wireless environment data of the wireless communication network 201 in operation by repeatedly performing the processes of steps S1 to S3.

[0029] 3 is a diagram for explaining an overview of the processing of the optimization system according to this embodiment. The optimization system 100 according to this embodiment performs the wireless communication network optimization 301 described in FIG. 2 for each of the multiple tenants 10a, 10b, 10c, 10d, .... The optimization system 100 also performs model optimization 302, which optimizes the estimation model 211 and control model 212 common to the multiple tenants 10, based on the wireless environment data of the multiple tenants 10.

[0030] For example, when estimation conditions and radio environment data are input to the estimation model 211 (step S11), the estimation model 211 outputs an estimated environment (step S12). The optimization system 100 optimizes the estimation model 211 so as to minimize the difference between the estimated environment output by the estimation model 211 and newly acquired radio environment data (actual measured environment) (step S13).

[0031] Furthermore, when the control model 212 receives control conditions and wireless environment data (step S11), it outputs an optimal control environment together with setting information for optimizing the communication quality of the wireless communication network (step S12). The optimization system 100 optimizes the control model 212 so as to minimize the difference between the optimal control environment output by the control model 212 and newly acquired wireless environment data (actual measured environment) (step S13).

[0032] At this time, the optimization system 100 according to this embodiment uses the radio environment data of the plurality of tenants 10 to optimize the estimation model 211 and the control model 212 common to each tenant 10 .

[0033] Therefore, according to this embodiment, it is not necessary to optimize the estimation model 211 and the control model 212 for each tenant 10, and therefore the wireless quality of the wireless communication networks of multiple tenants 10 can be easily optimized.

[0034] Furthermore, according to this embodiment, compared to optimization in a single tenant, the estimation model 211 and the control model 212 can be optimized using a large amount of diverse data under the same propagation environment, thereby improving the estimation accuracy of the estimation model 211 and the control accuracy of the control model 212.

[0035] [Example 1] (Optimization process of estimation model) Fig. 4 is a diagram illustrating an outline of optimization process of an estimation model according to Example 1. As illustrated in Fig. 4, the estimation model 211 is configured to output an estimated environment 420 when an estimation condition 411 and an estimation target environment 401 are input. Here, the estimation target environment 401 is wireless environment data acquired from an actual environment 400 (e.g., a wireless communication network of a tenant 10). Furthermore, the estimated environment 420 is, for example, an estimated value of wireless quality of the actual environment 400.

[0036] Furthermore, the optimization system 100 can hold information according to the environment of each tenant as tenant preset values ​​412 and set information dependent on the environment. The tenant preset values ​​412 can include, for example, three-dimensional model data including materials at the installation location of the wireless device 202 (materials and heights of structures, points, floors, etc.). The tenant preset values ​​412 can also include the model number, gain, frequency, output, calibration data, and antenna specifications of the antenna of the existing wireless device 202 or terminal 203.

[0037] The estimation model optimization unit 106 optimizes the estimation model 211 so as to minimize the difference between the estimated environment 420 output by the estimation model 211 and the actual measurement environment 402 measured in the actual environment 400. Here, the actual measurement environment 402 is the actual measurement value of the wireless quality in the actual environment 400.

[0038] By repeatedly performing this optimization process for the estimation model, the optimization system 100 can constantly improve the estimation model 211 .

[0039] <Processing Flow> Fig. 5 is a sequence diagram illustrating an example of the optimization process of the estimation model according to Example 1. This process illustrates a specific example of the optimization process of the estimation model described with reference to Fig. 4 .

[0040] In step S500, the administrator 523 who manages the wireless area sets the estimation conditions, the estimation target environment, etc. in the control device 101 of the optimization system 100, and then the optimization system 100 repeatedly executes the processing of steps S501 to S510 (optimization loop).

[0041] For example, in steps S501 and S502 of “Loop N,” which is the Nth optimization loop, the wireless device 521 and the reflector / movable base station 522 in the actual environment 400 transmit measurement data of the wireless environment to the optimization system 100.

[0042] In step S503, the control device 101 transmits the received measured data to the acquisition unit 102. In response to this, in step S504, the acquisition unit 102 stores the acquired measured data in the storage unit 110 or the like as measured environment information 531b.

[0043] In step S505, the control device 101 transmits the received actual measurement data to the estimation unit 103. In response to this, in step S506, the estimation unit 103 estimates the wireless quality of the actual environment 400 using the received actual measurement data and the estimation model 211, and stores the estimated wireless quality in the storage unit 110 or the like as estimated environment information 532b.

[0044] In step S507, the control device 101 transmits the received actual measurement data to the determination unit 104. In response to this, in step S508, the determination unit 104 determines setting information that optimizes the wireless quality of the actual environment 400 using the received actual measurement data and the control model 212, and stores the determined setting information in the storage unit 110 or the like as optimal control parameters 533b.

[0045] In step S509, the estimation model optimization unit 106 acquires data such as the estimated environment information 532a stored in the memory unit 110 in the N-1th optimization loop, "Loop N-1," and the measured environment information 531b stored in the memory unit 110 in "Loop N."

[0046] In step S510, the estimation model optimization unit 106 optimizes the estimation model 211 used by the estimation unit 103 to estimate wireless quality so as to minimize the difference 534 between the acquired estimated environment information 532a and the actual measured environment information 531b.

[0047] The optimization process of the estimation model shown in FIG. 5 is executed across the tenants 10 independently of each other each time an optimization loop for optimizing the wireless quality of the wireless communication network is executed.

[0048] (Optimization Process of Control Model) FIG. 6 is a diagram illustrating an overview of the optimization process of the control model according to the first embodiment. As illustrated in FIG. 6 , the control model 212 is configured to output an optimal control environment 620 and optimal control parameters 621 when a control condition 611 and a controlled environment 601 are input. Here, the controlled environment 601 is wireless environment data acquired from a real environment 600 (e.g., a wireless communication network of a tenant 10). The optimal control environment 620 is, for example, an estimated value of the wireless quality of the real environment 600 after the optimal control parameters 621 are set in the real environment 600. The optimal control parameters 621 are setting information to be set in a wireless device 603 or the like in the real environment in order to optimize the wireless quality of the real environment 600.

[0049] Furthermore, the optimization system 100 can set information depending on the environment by holding information according to the environment of each tenant as tenant preset values ​​412. The tenant preset values ​​412 are similar to the tenant preset values ​​412 described in FIG. 4, and therefore will not be described here.

[0050] The control model optimization unit 107 optimizes the control model 212 so as to minimize the difference between the optimal control environment 620 output by the estimation model 211 and a measured environment 602 after control that is measured in the actual environment 600. Here, the measured environment 602 after control is an actual measurement value of wireless quality measured in the actual environment 600 after reflecting the optimal control parameters 621 in wireless devices 603 and the like in the actual environment 600.

[0051] By repeatedly performing this control model optimization process, the optimization system 100 can constantly improve the control model 212 .

[0052] <Processing Flow> Fig. 7 is a sequence diagram illustrating an example of the optimization process of the control model according to Example 1. This process illustrates a specific example of the optimization process of the control model described with reference to Fig. 6 .

[0053] In step S700, the administrator 523 who manages the wireless area sets the control conditions, etc. in the control device 101 of the optimization system 100, and then the optimization system 100 repeatedly executes the processing of steps S701 to S710 (optimization loop).

[0054] For example, in steps S701 and S702 of “Loop N,” which is the Nth optimization loop, the wireless device 521 and the reflector / movable base station 522 in the actual environment 400 transmit measurement data of the wireless environment to the optimization system 100.

[0055] In step S703, the control device 101 transmits the received measured data to the acquisition unit 102. In response to this, in step S704, the acquisition unit 102 stores the acquired measured data in the storage unit 110 or the like as measured environment information 721b.

[0056] In step S705, the control device 101 transmits the received actual measurement data to the determination unit 104. In response to this, in step S706, the determination unit 104 stores the wireless quality of the actual environment 600 after control, which is estimated using the received actual measurement data and the control model 212, as optimal controlled environment information 722a in the storage unit 110, etc. Furthermore, in step S707, the determination unit 104 determines setting information that optimizes the wireless quality of the actual environment 400 using the received actual measurement data and the control model 212, and stores the determined setting information as optimal control parameters 713b in the storage unit 110, etc.

[0057] In step S708, the estimation model optimization unit 106 acquires data such as the optimal control environment information 722a stored in the memory unit 110 in the N-1th optimization loop, "Loop N-1," and the actual measurement environment information 721b stored in the memory unit 110 in "Loop N."

[0058] In step S709, the estimation model optimization unit 106 optimizes the control model 212 used by the determination unit 104 to determine the optimal control parameters 621 so that the difference 724 between the acquired optimal control environment information 722a and the measured environment information 721b is minimized.

[0059] The optimization process of the control model shown in FIG. 7 is executed across the tenants 10 independently of each other each time an optimization loop for optimizing the wireless quality of the wireless communication network is executed.

[0060] (Example of Optimization of Control Model) FIG. 8 is a diagram for explaining an example of optimization of the control model according to the first embodiment.

[0061] In step S21, it is assumed that the optimization system 100 detects interference between access points (APs) (a decrease in throughput) in a terminal 802 connected to an AP 801a.

[0062] In step S22, the optimization system 100 detects interference and performs dynamic control to reduce the transmission power level of the AP 802b adjacent to the AP 801a by a predetermined value. By reducing the transmission power of the AP 801b in this way, interference between the APs can be reduced, and the throughput of the terminal 802 can be expected to improve.

[0063] In step S23, if interference still occurs after the control even when the transmission power of AP 801b is reduced by a predetermined value (if the throughput of terminal 802 does not improve), the control model optimization unit 107 of the optimization system 100 optimizes the control model 212. For example, the optimization system 100 changes the predetermined value to a larger value. This enables the optimization system 100 to perform the dynamic control of step S22 with higher accuracy from the next time.

[0064] (Wireless Environment Optimization Process) Fig. 9 is a sequence diagram illustrating an example of the optimization process of the wireless communication network according to the first embodiment. This process illustrates, for example, a specific example of the optimization process of the wireless communication network described in Fig. 2. Note that the following description will be given on the assumption that the wireless device information indicating the wireless quality of the wireless device 911 of the wireless communication network 910 and the terminal information indicating the wireless quality of the terminal collected by the terminal information collection device 912 are sequentially stored in the storage unit 110 by the acquisition unit 102.

[0065] In step S901, the optimization system 100 receives an analysis setting instructing an analysis of wireless quality from an information terminal (hereinafter referred to as the administrator terminal 901) used by an administrator who manages the wireless communication network 910, and then executes processing from step S902 onwards.

[0066] In step S902 , the control device 101 transmits the accepted analysis settings to the analysis unit 105 .

[0067] In step S903, the analysis unit 105 stores the analysis settings received from the control device 101 in the storage unit 110, etc. In Fig. 9, dashed arrows indicate ACK (ACKnowledgement) or the like in response to the immediately preceding process.

[0068] The optimization system 100 also repeatedly executes steps S904 to S907 (wireless quality analysis loop) 332. For example, in steps S904 and S905, the analysis unit 105 periodically references the wireless device information and terminal information stored in the storage unit 110.

[0069] Furthermore, in step S906, the control device 101 periodically polls the analysis unit 105 for analysis execution asynchronously with the processing of steps S904 and S905. In response to this, in step S907, the analysis unit 105 analyzes the acquired wireless device information and terminal information and transmits the analysis result to the control device 101. For example, the analysis unit 105 analyzes the acquired wireless device information and terminal information, and if a predetermined control target condition and a predetermined design target condition are not detected, transmits the analysis result "no abnormality" to the control device 101.

[0070] On the other hand, for example, if a predetermined control target condition occurs in the wireless communication network 910 in step S908, information indicating the occurrence of the predetermined control target condition is included in the wireless device information or terminal information referenced by the analysis unit 105 in steps S909 and S910. In this case, the analysis unit 105 transmits a "fault detected" to the control device 101 in response to the analysis execution polling in steps S911 and S912, indicating that the predetermined control target condition has been detected.

[0071] For example, the optimization system 100 stores automatic control information 1000 as shown in FIG. 10 in advance in the storage unit 110 or the like.

[0072] FIG. 10 is a diagram illustrating an example of automatic control information according to the first embodiment. In the example of FIG. 10, the automatic control information 1000 includes information such as a "control target condition," a "control content," a "recovery condition," and a "case where recovery is not performed" as items. The "control target condition" stores the control target condition detected in the analysis process 931 by the analysis unit 105. The "control content" stores the control content to be performed by the optimization system 100 when the "control target condition" is detected. The "recovery condition" stores the recovery condition for determining that recovery from the control target condition has occurred. The "case where recovery is not performed" stores the processing content to be performed when the "control target condition" occurs and the "recovery condition" is not satisfied even if the "control content" is performed.

[0073] The analysis unit 105 detects, for example, based on such automatic control information 1000, from the wireless device information and the terminal information, that a control target condition has occurred.

[0074] In step S913, when the control device 101 receives a "fault detection" from the analysis unit 105 indicating that a specified control target condition has been detected, it sends a determination request to the determination unit 104 requesting the determination of wireless control parameters.

[0075] In step S914, the determination unit 104 determines wireless control parameters that optimize the wireless communication network 910, for example, using the control model 212 optimized by the processing of Figure 7, and stores the determined wireless control parameters in the memory unit 110, etc.

[0076] In addition, in step S915 , the determination unit 104 transmits the determined radio control parameters to the control device 101 .

[0077] In step S916, the analysis unit 105 notifies the administrator terminal 901 of the occurrence of a failure by email, a GUI, or the like. This process may be performed by the control device 101 after step S912, for example.

[0078] In step S917, in response to the failure notification, the administrator who manages the wireless communication network 910 uses the administrator terminal 901 to send a recovery instruction to the optimization system 100. However, this process is optional. The optimization system 100 may automatically execute the processes from step S918 onwards without relying on a recovery instruction from the administrator terminal 901.

[0079] In steps S918 and S919, the control device 101 acquires the wireless control parameters stored by the determination unit 104 in the storage unit 110 or the like.

[0080] In steps S920 and S921, the control device 101 requests the setting control unit 108 to control the wireless device based on the acquired wireless control parameters.

[0081] In step S921 , the setting control unit 108 controls the wireless device 911 in accordance with the request for control of the wireless device received from the control device 101 .

[0082] In this way, when the optimization system 100 detects a control target condition, it determines wireless control parameters and, based on the determined wireless control parameters, for example, automatically controls the wireless device 911. Furthermore, by repeatedly executing the same process after control, the optimization system 100 can continuously optimize the configuration of the wireless communication network 910 so as to always satisfy the wireless communication quality required by the user or terminal.

[0083] 9 is an example. For example, the control device 101 may feed back wireless control parameters that optimize the wireless quality of the wireless communication network 910 to the tenant 10 that provides the wireless communication network 910. In this case, the process of controlling the wireless device 911 based on the wireless control parameters may be executed by the system of the tenant 10.

[0084] 11 is a diagram illustrating an overview of processing of an optimization system according to Example 2. Similar to Example 1, the optimization system 100 according to Example 2 performs optimization 301 of a wireless communication network for each of a plurality of tenants 10a, 10b, 10c, 10d, and so on.

[0085] 11, the classification unit 109 of the optimization system 100 categorizes the wireless environment of each tenant 10 and classifies multiple tenants 10 by environmental pattern (step S31). If heterogeneous environments are treated as the same group and modeled, this may result in noise, making it difficult to optimize. However, by dividing them into environmental patterns, appropriate modeling is possible for each environment. The following categorization methods for dividing them into environmental patterns are possible.

[0086] Fig. 12 is a diagram illustrating an example of categorization of a wireless environment according to Example 2. As illustrated in Fig. 12, possible categorization methods include categorization according to communication patterns, categorization according to terminal vendors, and categorization according to propagation environments.

[0087] In categorizing according to communication patterns, wireless environments may be classified according to the ratio of uplink and downlink communications, the amount of data, etc. Examples of categorization in this case include "Internet of Things (IoT) terminal environment (mainly small amounts of uplink data)" and "video distribution environment (mainly large amounts of downlink data)."

[0088] In categorizing according to terminal vendor, the vendor may be identified by the MAC address of the collected data source, and the wireless environment may be classified by vendor. In this case, examples of categorization may include "vendor A environment," "vendor B environment," or "vendor A and B mixed environment."

[0089] In categorization according to propagation environment, wireless environments may be classified according to propagation characteristics, such as "outdoor urban / outdoor natural environment" or "indoor office / indoor factory environment."

[0090] In the example of FIG. 11, the classification unit 109 classifies the tenants 10a and 10b into "environment pattern A" and the tenants 10c and 10d into "environment pattern B."

[0091] Furthermore, the estimation model optimization unit 106 and the control model optimization unit 107 of the optimization system 100 optimize models such as the estimation model 211 and the control model 212 for each categorized wireless environment (step S32).

[0092] In this case, a machine learning model to be used can be selected for each classified wireless environment pattern. In the example of Fig. 11, a deep learning model, a convolutional neural network (CNN), is applied to "environment pattern A," and a recurrent neural network (RNN) is applied to "environment pattern B."

[0093] As an example of a CNN, as shown in FIG. 13, a technique is known in which various map data images are input to a CNN 1300 to extract optimal parameters for radio wave propagation and use them to predict radio wave propagation (Reference Technical Document 1: T. Imai, K. Kitao, M. Inomata, "Radio Propagation Prediction Model Using Convolutional Neural Networks by Deep Learning," 2019 13th European Conference on Antennas and Propagation).

[0094] As an example of an RNN, a technology is known that uses past received power as input to predict future received power (Reference Technical Document 2: Motoharu Sasaki et al., "RNN Based Prediction of Path Loss Fading Distribution by Interval Estimation," 2022 16th European Conference on Antennas and Propagation.).

[0095] Fig. 14 is a diagram illustrating an overall view of an optimization system according to Example 2. In the example of Fig. 14, the optimization system 100 classifies tenant (1) into environment pattern A and tenants (2) and (3) into environment pattern B.

[0096] In this case, the optimization system 100 uses collected data 1411 of tenant (1) to optimize an estimation model 1421 of environment pattern A and a control model 1431 of environment pattern A in the same manner as in Example 1 described with reference to Figures 4 to 7. The optimization system 100 also uses collected data 1412 of tenant (2) and collected data 1413 of tenant (3) to similarly optimize an estimation model 1422 of environment pattern B and a control model 1432 of environment pattern B.

[0097] Furthermore, the optimization system 100 optimizes the wireless quality of the wireless communication network 1440 of tenant (1) using the estimation model 1421 of environment pattern A and the control model 1431 of environment pattern A in the same manner as in Example 1 described with reference to Fig. 9. At this time, the optimization system 100 can apply the preset value 1441 of tenant (1) as described above.

[0098] Similarly, the optimization system 100 optimizes the wireless quality of the wireless communication network 1450 of tenant (2) using the estimation model 1422 of environment pattern B and the control model 1432 of environment pattern B in the same manner as in Example 1 described with reference to Fig. 9. At this time, the optimization system 100 can apply the preset value 1451 of tenant (2) as described above.

[0099] Similarly, the optimization system 100 optimizes the wireless quality of the wireless communication network 1460 of tenant (3) using the estimation model 1422 of environment pattern B and the control model 1432 of environment pattern B in the same manner as in Example 1 described with reference to Fig. 9. At this time, the optimization system 100 can apply the preset value 1461 of tenant (3) as described above.

[0100] In this way, according to the optimization system 100 of Example 2, it is possible to optimize (improve, enhance) the model for each categorized environment, thereby further improving the accuracy of optimizing the wireless quality of the wireless communication network.

[0101] Furthermore, in the second embodiment, each tenant has ownership of the collected data itself. Also, for design parameters according to the environment, individual preset values ​​can be applied to each tenant.

[0102] <Hardware Configuration Example> The control device 101 has, for example, the hardware configuration of a computer 1500 as shown in Fig. 15 . Furthermore, the optimization system 100 is realized, for example, by a plurality of computers 1500. Fig. 15 is a diagram showing an example of the hardware configuration of a computer according to this embodiment. In the example of Fig. 15 , the computer 1500 has a processor 1501, a memory 1502, a storage device 1503, a communication device 1504, an input device 1505, an output device 1506, a bus B, etc.

[0103] The processor 1501 is, for example, an arithmetic unit such as a CPU (Central Processing Unit) that executes predetermined programs to realize various functions. The memory 1502 is a storage medium readable by the computer 1500, and includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The storage device 1503 is a computer-readable storage medium, and includes, for example, a HDD (Hard Disk Drive), an SSD (Solid State Drive), various optical disks, and a magneto-optical disk.

[0104] The communication device 1504 includes one or more pieces of hardware (communication devices) for communicating with other devices via a wireless or wired network. The input device 1505 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that receives input from the outside. The output device 1506 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside. Note that the input device 1505 and the output device 1506 may be integrated into one device (e.g., an input / output device such as a touch panel display).

[0105] The bus B is commonly connected to the above components and transmits, for example, address signals, data signals, and various control signals. The processor 1501 is not limited to a CPU, and may be, for example, a DSP (Digital Signal Processor), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0106] (Supplementary Note) The control device 101 and optimization system 100 in this embodiment may be realized not only by a dedicated device but also by a general-purpose computer. In this case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed. Note that the term "computer system" here includes hardware such as an OS and peripheral devices.

[0107] Furthermore, the term "computer-readable recording medium" includes various storage devices such as portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as storage devices 1503 built into computer systems. Furthermore, the term "computer-readable recording medium" may also include devices that dynamically store a program for a short period of time, such as a communication line when transmitting a program via a network such as the Internet or a communication line such as a telephone line, and devices that store a program for a certain period of time, such as volatile memory within a computer system that serves as a server or client in such cases.

[0108] Furthermore, the above program may be one that realizes part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in a computer system, or may be one that is realized using hardware such as a PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array).

[0109] <Effects of the embodiment> According to the present embodiment, the optimization system 100 that optimizes the wireless environment of a wireless communication network can easily optimize the wireless quality of the wireless communication networks of multiple tenants.

[0110] Furthermore, in this embodiment, the estimation model 211 and the control model 212 common to the multiple tenants 10 are optimized using radio environment data acquired from the multiple tenants 10. Therefore, compared to optimization for a single tenant, a large amount of diverse data under the same propagation environment is optimized, and therefore the accuracy of the estimation model 211 and the control model 212 becomes higher.

[0111] The estimation model 211 and the control model 212 optimized by the optimization system 100 can also be used by an external system via, for example, an API (Application Programming Interface) of the control device 101 or the like.

[0112] Furthermore, according to Example 2, multiple tenants 10 are categorized by their wireless environment, and the estimation model 211 and the control model 212 are optimized for each wireless environment, thereby further improving the accuracy of the estimation model 211 and the control model 212.

[0113] Summary of Embodiments This specification discloses at least the control device, optimization system, control method, and program of the following paragraphs: (1) A control device that feeds back configuration information of multiple tenants to which it provides a wireless communication network, based on wireless environment data of the multiple tenants. (2) The control device described in paragraph 1, wherein the control device controls an optimization system having: an estimation model optimization unit that optimizes an estimation model that estimates wireless quality of each tenant based on the wireless environment data of the multiple tenants; and an estimation unit that estimates wireless quality of each tenant using the estimation model. (3) The control device described in paragraph 2, wherein the optimization system has a classification unit that categorizes wireless communication networks provided by the multiple tenants by wireless environment, the estimation model optimization unit optimizes the estimation model that differs for each wireless environment, and the estimation unit estimates wireless quality of each tenant using the estimation model according to the wireless environment of each tenant. (4) The control device according to paragraph 1 or 2, wherein the optimization system comprises: a control model optimization unit that optimizes a control model that outputs configuration information that optimizes the wireless quality of each tenant based on wireless environment data of the multiple tenants; and a determination unit that determines configuration information for each tenant using the control model. (5) The optimization system comprises: a classification unit that categorizes wireless communication networks provided by the multiple tenants by wireless environment, the control model optimization unit optimizes the control model that differs for each wireless environment, and the determination unit determines configuration information for each tenant using the control model according to the wireless environment of each tenant. (6) An optimization system comprising: an acquisition unit that acquires wireless environment data for multiple tenants that provide a wireless communication network; and a control device that feeds back configuration information for each tenant to the multiple tenants based on the wireless environment data of the multiple tenants. (7) A control method in which a control device feeds back configuration information for each tenant to the multiple tenants based on the wireless environment data of the multiple tenants that provide a wireless communication network.(Clause 8) A program, or a storage medium storing the program, that causes a control device to feed back setting information of each tenant to a plurality of tenants based on wireless environment data of the tenants to which a wireless communication network is provided.

[0114] Although the present embodiment has been described above, the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims.

[0115] 100 Optimization system 10, 10a, 10b, 10c, 10d Tenant 101 Control device 102 Acquisition unit 103 Estimation unit 104 Determination unit 105 Analysis unit 106 Estimation model optimization unit 107 Control model optimization unit 108 Setting control unit 109 Classification unit 211 Estimation model 212 Control model 1500 Computer

Claims

1. A control device that feeds back configuration information of a plurality of tenants to which a wireless communication network is provided, based on wireless environment data of the plurality of tenants.

2. The control device according to claim 1, wherein the control device controls an optimization system having: an estimation model optimization unit that optimizes an estimation model that estimates wireless quality of each tenant based on wireless environment data of the multiple tenants; and an estimation unit that estimates wireless quality of each tenant using the estimation model.

3. The control device according to claim 2, wherein the optimization system has a classification unit that categorizes the wireless communication networks provided by the multiple tenants by wireless environment, the estimation model optimization unit optimizes the estimation model that differs for each wireless environment, and the estimation unit estimates the wireless quality of each tenant using the estimation model corresponding to the wireless environment of each tenant.

4. The control device according to claim 2 or 3, wherein the optimization system comprises: a control model optimization unit that optimizes a control model that outputs setting information that optimizes the wireless quality of each tenant based on wireless environment data of the multiple tenants; and a determination unit that determines the setting information for each tenant using the control model.

5. The control device according to claim 4, wherein the optimization system has a classification unit that categorizes the wireless communication networks provided by the multiple tenants by wireless environment, the control model optimization unit optimizes the control model that differs for each wireless environment, and the determination unit determines the configuration information for each tenant using the control model that corresponds to the wireless environment of each tenant.

6. An optimization system comprising: an acquisition unit that acquires wireless environment data of a plurality of tenants for which a wireless communication network is provided; and a control device that feeds back configuration information of each tenant to the plurality of tenants based on the wireless environment data of the plurality of tenants.

7. A control method in which a control device feeds back configuration information of each of a plurality of tenants to which a wireless communication network is provided, based on wireless environment data of the plurality of tenants.

8. A program that causes a control device to feed back configuration information of each of a plurality of tenants to which a wireless communication network is provided, based on wireless environment data of the plurality of tenants.

Images