Control device and base station control method

Abstract

To allow traffic prediction to be performed with high accuracy in a wireless communication system that performs a transmission termination process.SOLUTION: A control device includes a prediction unit that predicts group traffic of a group including a plurality of base station devices using individual traffic for each base station device, and a control unit that determines whether to stop or operate the base station devices in the group based on the predicted group traffic. The group is composed of base station devices that form a macro cell and base station devices that form small cells, and the small cells have areas that overlap with the macro cell.SELECTED DRAWING: Figure 8

Description

Technical Field

[0001] The present invention relates to a control device and a base station control method.

Background Art

[0002] In a wireless communication system, power saving of the communication system is achieved by optimally deploying base station devices according to the communication traffic volume. Further, power saving may be further achieved by performing a wave stop process of stopping some base station devices for a certain period according to the communication traffic volume.

[0003] In a wireless communication system, in order to stably continue the communication service even when performing the wave stop process, for example, the base station device to be stopped is determined based on the traffic demand prediction for each base station device.

[0004] Techniques related to traffic prediction are described in, for example, the following documents.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] However, regarding the base station device that is stopped during the wave stop process, since the traffic of the cell during the wave stop is 0 and some of the traffic data that is an input for prediction is missing, it is difficult to perform accurate traffic demand prediction.

[0007] In addition, for base station apparatuses that do not stop during the outage process, traffic from terminal apparatuses in other cells concentrates, and the data serving as input for prediction becomes a larger value than during normal operation, making it difficult to perform high-precision traffic demand prediction.

[0008] Therefore, one disclosure can provide a control apparatus and a base station control method capable of performing traffic prediction with high precision in a wireless communication system that performs an outage process.

Means for Solving the Problem

[0009] A prediction unit that predicts group traffic of a group including a plurality of base station apparatuses using individual traffic for each base station apparatus, and a control unit that determines stop or operation of the base station apparatuses in the group according to the predicted group traffic, wherein the group is composed of a base station apparatus that constitutes a macro cell and a base station apparatus that constitutes a small cell, and the small cell has a region overlapping with the macro cell.

Effect of the Invention

[0010] One disclosure can perform traffic prediction with high precision in a wireless communication system that performs an outage process.

Brief Description of the Drawings

[0011]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Embodiments for Carrying Out the Invention

[0012] [First Embodiment] The first embodiment will be described.

[0013] [Regarding the wireless communication system 10] FIG. 1 is a diagram showing a configuration example of the wireless communication system 10. The wireless communication system 10 includes terminal devices 100-1 to m (m is an integer), base station devices 200-1 to n (n is an integer), a server machine 300, and a network 400. The wireless communication system 10 is a communication system that supports outage processing. The outage processing is, for example, a process of predicting the traffic (hereinafter may be referred to as traffic) between the base station devices 200-1 to n (hereinafter may be referred to as the base station device 200) and the terminal devices 100-1 to m (hereinafter may be referred to as the terminal device 100), and stopping some of the base station devices for a predetermined period according to the prediction result.

[0014] The terminal device 100 is a communication device that wirelessly connects to the base station device 200 and performs wireless communication, and is, for example, a tablet terminal or a smartphone. The terminal device 100 wirelessly connects to the base station device 200 that constitutes the cell (communication area constituted by the base station device 200) of the destination and performs wireless communication.

[0015] The base station device 200 is a communication device that wirelessly connects to the terminal device 100 and performs wireless communication, and is, for example, an eNodeB or a gNodeB. The base station device 200 performs wireless communication with the terminal device 100 or relays communication between the terminal device 100 and other communication devices. In relaying, the base station device 200 transmits the signal received from the terminal device 100 to the destination communication device via the connected network 400.

[0016] The server machine 300 is a device that performs (or gives an instruction for) a wave stop process, and is, for example, a computer or a server machine. The server machine 300 collects traffic from the base station device 200. The server machine 300 performs traffic prediction using the collected traffic. The server machine 300 determines a base station device 200 to be stopped according to the traffic prediction, and gives an instruction to the base station device 200 to stop.

[0017] The network 400 is a network that mediates communication between the server machine 300, the base station device 200, and other communication devices, and is, for example, the Internet or an intranet.

[0018] In the wireless communication system 10, in traffic prediction, in addition to the traffic of the base station device 200, traffic in units of groups including a plurality of base station devices 200 is predicted. Also, in determining the base station device 200 to be a stop target in the wave stop process, in addition to the traffic of a single base station device 200, traffic prediction for each group is also used.

[0019] <Configuration example of server machine 300> FIG. 2 is a diagram showing a configuration example of the server machine 300. The server machine 300 includes a CPU (Central Processing Unit) 310, a storage 320, a memory 330, and a communication circuit 350.

[0020] The storage 320 is an auxiliary storage device such as a flash memory, an HDD (Hard Disk Drive), or an SSD (Solid State Drive) that stores programs and data. The storage 320 stores a wave stop process program 321, a traffic prediction model 322, a model learning program 323, and a traffic collection program 324.

[0021] The traffic prediction model 322 is an AI (Artificial Intelligence) learning model that predicts traffic in units of base station apparatuses 200. The traffic prediction model 322 is, for example, a trained model that has learned, as teacher data, the traffic volume for each time period (predetermined time unit) of each base station apparatus 200, feature quantities related to the traffic volume, and the like. The server machine 300 can predict the traffic of a certain base station apparatus 200 at a certain time period using the traffic prediction model 322.

[0022] The memory 330 is an area for loading programs stored in the storage 320. Also, the memory 330 may be used as an area where the program stores data.

[0023] The communication circuit 350 is a device that transmits and receives packets to and from the base station apparatus 200 and other communication devices to perform communication, and is, for example, a network interface circuit. The communication circuit 350 communicates with the base station apparatus 200 and acquires packets transmitted and received on the network 400.

[0024] The CPU 310 is a processor that loads the programs stored in the storage 320 into the memory 330, executes the loaded programs, constructs each part, and realizes each process.

[0025] By executing the outage processing program 321, the CPU 310 constructs a prediction unit and a control unit and performs outage processing. The outage processing is a process of predicting the traffic for each base station apparatus 200, accordingly predicting the traffic for each group, determining the base station apparatus 200 to be stopped from the traffic prediction of the group, and stopping the base station apparatus 200. By performing the outage processing, the power of the base station apparatus 200 can be suppressed, and power saving is realized.

[0026] The CPU 310 constructs a prediction unit by executing the base station traffic prediction module 3211 included in the outage processing program 321, and performs base station traffic prediction processing. The base station traffic prediction processing is processing for predicting the traffic for each base station apparatus 200. The server machine 300 predicts the traffic of each base station apparatus 200, for example, using the traffic prediction model 322. Also, the outage processing may include, for example, processing for classifying the base station apparatuses 200 into groups. The base station apparatuses 200 belonging to each group may be determined in advance and stored in a memory, or may be determined as part of the outage processing.

[0027] The CPU 310 constructs a prediction unit by executing the group traffic prediction module 3212 included in the outage processing program 321, and performs group traffic prediction processing. The group traffic prediction processing is processing for predicting the traffic for each group. The server machine 300 predicts the traffic for each group, for example, by adding up the traffic of a plurality of base station apparatuses 200 belonging to the group. Also, when the traffic prediction model 322 is capable of predicting the traffic for each group, the server machine 300 may predict the traffic for each group using the traffic prediction model.

[0028] The CPU 310 constructs a prediction unit and a control unit by executing the base station to be stopped determination module 3213 included in the outage processing program 321, and performs base station to be stopped determination processing. The base station to be stopped determination processing is processing for determining the base station apparatus 200 for which outage processing is to be performed according to the predicted traffic for each group or the predicted traffic for each base station apparatus. The server machine 300 selects, for example, a group for which outage is possible from the predicted traffic for each group, and determines one or a plurality of the base station apparatuses 200 within the group as the base stations to be stopped.

[0029] The CPU 310 performs traffic prediction processing using the traffic prediction model 322. The traffic prediction model 322 is, for example, a learned model that has learned the traffic of each base station device 200 as teacher data. The traffic prediction processing is, for example, processing for predicting the traffic for each base station device 200. Note that when the traffic prediction model 322 learns using the traffic for each group as teacher data, the traffic prediction processing becomes processing for predicting the traffic for each group. Hereinafter, unless otherwise specified, the traffic prediction processing is assumed to be processing for predicting the traffic for each base station device 200.

[0030] The CPU 310 constructs a learning unit and performs model learning processing by executing a model learning program 323. The model learning processing is processing for learning the traffic prediction model 322. The server machine 300 performs model learning processing using the traffic of each base station device 200 as teacher data.

[0031] The CPU 310 constructs a collection unit and performs traffic collection processing by executing a traffic collection program 324. The traffic collection processing is processing for collecting traffic from each base station device 200 and is performed, for example, periodically. The server machine 300 separates the collected traffic for each base station device 200 and stores it in an internal memory. Then, the server machine 300 uses the stored traffic to learn the traffic prediction model.

[0032] <Traffic Prediction> Traffic prediction will be described. FIG. 3 is a diagram showing an example of the relationship between actual measured traffic and predicted traffic. Note that the measured traffic is assumed to be teacher data in the learning of the traffic prediction model 322.

[0033] FIG. 3(A) is a diagram showing an example of the relationship between measured traffic and predicted traffic during normal operation (operation without performing outage processing). The cell configuration is such that small cells B and C exist inside macro cell A. According to FIG. 3(A), the measured traffic at a certain time for each of cells A, B, and C is such that A is slightly higher, and B and C have close values. The predicted traffic output by the traffic prediction model learned using this measured traffic as teacher data is, as shown in FIG. 3(A), such that A is slightly higher, and B and C have close values.

[0034] FIG. 3(B) is a diagram showing an example of the relationship between measured traffic and predicted traffic during outage processing operation in which the base station apparatuses 200 constituting cells B and C are stopped. The cell configuration is the same as in FIG. 3(A), but since cells B and C are not transmitting radio waves, they are in a state where they do not substantially exist. According to FIG. 3(B), the traffic of cell A becomes a high value. This is because, since cells B and C do not substantially exist, traffic concentrates on cell A. On the other hand, the traffic of cells B and C is 0. The predicted traffic output by the traffic prediction model learned using this measured traffic as teacher data is, as shown in FIG. 3(B), such that cell A has a high value, and each of B and C has a value close to 0.

[0035] Thus, in a communication system that performs outage processing, the traffic prediction model is affected by the traffic in a state where some of the base station apparatuses 200 are stopped. Therefore, in determining the base station apparatuses 200 to be stopped, for example, if the predicted traffic during outage processing is used, since the traffic of cells B and C is close to 0, cells B and C will be determined as the base stations to be stopped. In this way, when the measured traffic during outage processing operation is mixed, it becomes difficult to predict the demand traffic of each cell, and it may be impossible to make an appropriate determination of the base stations to be stopped.

[0036] Therefore, in the first embodiment, the server machine 300 determines the base stations to be stopped using the predicted traffic for each group.

[0037] <Wave stop processing> The wave stop processing will be described. The wave stop processing is performed, for example, periodically or irregularly. Also, the period for stopping the base station to be stopped is, for example, the period until the next wave stop processing is performed, or a predetermined time (a preset time, a time according to the traffic volume, etc.).

[0038] FIG. 4 is a diagram showing an example of a processing flowchart of the wave stop processing S100. The server machine 300 acquires traffic from each base station device 200 (S100-1). The server machine 300 uses the acquired traffic to train the traffic prediction model 322.

[0039] The server machine 300 predicts the traffic (individual traffic) of each base station device 200 (S100-2). The server machine 300 predicts the traffic after a predetermined time (predetermined time), for example, from the latest traffic, using the traffic prediction model 322.

[0040] The server machine 300 checks whether the base stations to be stopped for all groups (including groups with one base station device 200) have been determined (S100-3).

[0041] If there is a group for which the determination of the base stations to be stopped has not been made (No in S100-3), the server machine 300 predicts the traffic of a certain group (group traffic) (S100-4). Then, the server machine 300 determines the base stations to be stopped for a certain group (S100-5). The server machine 300 repeats processes S100-4 and S100-5 until the base stations to be stopped for all groups are determined.

[0042] Note that, for example, the determination of the base station to be stopped may be such that the base station apparatuses in the group are not determined to stop. For example, a group including a cell with extremely high traffic may be determined to have no base station to be stopped because if some of the base station apparatuses 200 are stopped, the traffic will concentrate on the non-stopped base station apparatuses 200, which may exceed the limit of the terminal apparatuses 100 that can be accommodated.

[0043] If there is no group for which the determination of the base station to be stopped has not been made (Yes in S100-3), the server machine 300 transmits a stop instruction to the base station to be stopped (S100-6) and ends the process.

[0044] FIG. 5 is a diagram showing an example of a sequence of the outage process. Assume that the base station apparatuses 200-1 to 3 belong to the same group 1.

[0045] The server machine 300 performs polling on the base station apparatuses 200-1 to 3 in group 1 in order to collect traffic (S10 to S12).

[0046] When the base station apparatuses 200-1 to 3 receive the polling, they collect traffic (S13 to S15) and transmit the collected results (traffic data) to the server machine 300 (S16 to S18).

[0047] The server machine 300 predicts the traffic of each of the base station apparatuses 200-1 to 3 (S19, S100-2 in FIG. 4). Then, the server machine 300 performs traffic prediction for group 1 (S20, S100-4 in FIG. 4).

[0048] Based on the traffic prediction for group 1, the server machine 300 determines that the base station apparatus to be outaged in group 1 is the base station apparatus 200-3 (S21, S100-5 in FIG. 4).

[0049] The server machine 300 transmits a wave stop instruction to the base station to be stopped (for example, the base station device 200-3) (S21, S100-6 in FIG. 4). When the base station device 200-3 receives the wave stop instruction, it stops transmitting and receiving radio waves.

[0050] <Group classification> The group classification (way of grouping) will be described. FIG. 6 is a diagram showing an example of a group.

[0051] In the group classification in FIG. 6, for example, one cell (base station device 200) belongs to one group and does not belong to a plurality of groups.

[0052] Also, in the group classification in FIG. 6, a macro cell and a small cell whose entire cell area is included in the area of the macro cell are regarded as one group.

[0053] Group 1 is a group composed of one macro cell and one small cell. All areas of the small cell B are completely included in the area of the macro cell A.

[0054] Group 2 is a group composed of one macro cell and two (a plurality of) small cells. All areas of the small cells D and E are completely included in the area of the macro cell A.

[0055] Group 3 is a group composed of one macro cell. Also, Group 4 is a group composed of one small cell. In this way, when there are no cells with overlapping areas, one cell forms one group.

[0056] Groups 5 and 6 are groups in the case where a part of the area of the macro cells overlaps. The macro cell H and the macro cell I have a part of their areas overlapping, but since the area of one cell does not completely include the area of the other cell, the macro cell H and the macro cell I are each in a separate group (the macro cell H is Group 5 and the macro cell I is Group 6).

[0057] Groups 7 and 8 are groups in the case where a partial area of a small cell overlaps with the area of a macro cell. The macro cell J and the small cell K have a partial area that overlaps, but since the area of the macro cell does not completely include the area of the small cell, the macro cell J and the small cell K are each in separate groups (macro cell J is in group 7 and small cell K is in group 8).

[0058] In FIG. 6, a macro cell and a small cell whose entire cell area is included in the area of the macro cell are regarded as one group. However, even if not all cell areas overlap, if a predetermined ratio or more of the areas overlap, they may be regarded as the same group.

[0059] FIG. 7 is a diagram showing an example of groups. The group classification in FIG. 7 is such that a macro cell and a small cell whose area is a predetermined ratio or more of the area of the macro cell are regarded as one group. Note that the ratio of the area of the small cell that overlaps with the area of the macro cell may be referred to as the overlap ratio. Also, the threshold value is, for example, 50%.

[0060] In FIG. 7(A), group 1 is composed of one macro cell and one small cell. Since 50% or more of the small cell B overlaps with the macro cell A, the small cell B is classified into the same group 1 as the macro cell A.

[0061] On the other hand, in FIG. 7(B), group 1 is composed of one macro cell and group 2 is composed of one small cell. Since the small cell B overlaps with the macro cell A by about 10% and thus the overlap ratio is less than the threshold value, the small cell B and the macro cell A are classified into separate groups.

[0062] <Predicted traffic of groups> The predicted traffic of the group will be described. The predicted traffic of the group is, for example, the total value (sum value) of the traffic of the base station apparatuses 200 included in the group. The server machine 300 distributes the predicted traffic of the group to the traffic for each individual base station apparatus (the individual traffic after correction) in order to determine the base station to be stopped.

[0063] One method of distribution is a method in which an area ratio is calculated from the area of the cell of each base station apparatus 200 in the group, and the distribution is made to each base station apparatus 200 according to the area ratio. The higher the area ratio, the larger the amount of traffic.

[0064] FIG. 8 is a diagram showing an example of the distribution of the predicted traffic of the group. When the server machine 300 distributes the predicted traffic of the group, which is the sum of the predicted traffic for each individual base station apparatus 200, according to, for example, the area ratio, the traffic after correction in FIG. 8 can be obtained. As a result, either or both of cell B and cell C will be the base station to be stopped.

[0065] In addition, since the traffic of each base station apparatus 200 changes depending on the base station to be stopped, it is preferable to determine the base station to be stopped and predict the traffic of the base station apparatus 200 simultaneously or in parallel. For example, the server machine 300 calculates the traffic of each base station apparatus 200 from the predicted traffic of the group according to the area ratio of the cells, determines the base station to be stopped, and performs a radio simulation in which the traffic of the stopped base station apparatus 200 is allocated (corrected) as an increase in the traffic of the operating base station apparatus. A base station apparatus with little (low degree of decrease or no) impact on the service even when stopped may be used as the base station to be stopped.

[0066] Note that since it is necessary to verify the stop of all the base station apparatuses 200 and determine the optimal base station to be stopped, the calculation time can also be shortened by utilizing AI to learn the optimal outage pattern. [Other Embodiments] The cell and the overlap ratio may be assumed and calculated as follows. In the following, it is assumed that radio waves are radiated evenly in all directions without considering the directivity of the radio waves. · Determine the radius of the radio wave of the base station apparatus 200. For example, it can be calculated from the Friis propagation formula using the antenna height, radio wave frequency, radio wave output, etc. · Plot the cell of the base station apparatus 200 on a plane from the position coordinates of the base station apparatus 200 and the radius of the radio wave, and obtain the ratio at which the areas of the cells of other base station apparatuses 200 overlap (overlap ratio). Note that it is also possible to more accurately obtain the coverage relationship using an advanced radio wave simulation considering buildings and terrain.

[0067] Also, the outage processing may be executed by a device other than the server machine 300. For example, it may be executed by the base station apparatus 200 that constitutes the macro cell.

[0068] Summarizing below, it becomes as follows in the appendix.

[0069] (Appendix 1) A prediction unit that predicts the group traffic of a group including a plurality of base station apparatuses using the individual traffic of each base station apparatus, A control unit that determines the stop or operation of the base station apparatuses in the group according to the predicted group traffic, and The group is composed of a base station apparatus that constitutes a macro cell and a base station apparatus that constitutes a small cell, The small cell has an area that overlaps with the macro cell Control device.

[0070] (Appendix 2) All areas of the small cell overlap with the area of the macro cell The control device according to Appendix 1.

[0071] (Appendix 3) The ratio at which the area of the small cell overlaps with the area of the macro cell is equal to or greater than a first threshold value The control device according to Appendix 1.

[0072] (Appendix 4) The control unit sums up the individual traffic of each of the plurality of base station devices and calculates the group traffic. The control device according to Appendix 1.

[0073] (Appendix 5) The control unit allocates the group traffic according to the area of the cells constituted by the plurality of base station devices, and calculates the corrected individual traffic of each of the plurality of base station devices. The control device according to Appendix 1.

[0074] (Appendix 6) The control unit calculates the individual traffic of each of the plurality of base station devices by using the group traffic, adds the first individual traffic of the first base station device to be stopped as an increment to the second individual traffic of the second base station device not to be stopped, and determines the stop or operation of the base station device in consideration of the increment. The control device according to Appendix 1.

[0075] (Appendix 7) The group traffic is output from a learning model that predicts the group traffic learned from the traffic measured by each of the plurality of base station devices. The control device according to Appendix 1.

[0076] (Appendix 8) The control unit controls to stop the base station device to be stopped. The control device according to Appendix 1.

[0077] (Appendix 9) A prediction step of predicting the group traffic of a group including a plurality of base station devices by using the individual traffic of each base station device, and a control step of determining the stop or operation of the base station device among the plurality of base station devices in the group according to the group traffic. The group is composed of a base station device constituting a macro cell and a base station device constituting a small cell. The small cell has a base station control method for an area overlapping with the macro cell.

Explanation of Signs

[0078] 10: Wireless communication system 100: Terminal device 200: Base station device 300: Server machine 310: CPU 320: Storage 321: Idle wave processing program 3211: Base station traffic prediction module 3212: Group traffic prediction module 3213: Determination module for base station to be stopped 322: Traffic prediction model 323: Model learning program 324: Traffic collection program 330: Memory 350: Communication circuit 400: Network

Claims

1. A prediction unit that predicts group traffic of a group including a plurality of base station apparatuses using individual traffic for each base station apparatus; A control unit that determines stop or operation of the base station apparatuses in the group according to the predicted group traffic, and the group is composed of a base station apparatus that constitutes a macro cell and a base station apparatus that constitutes a small cell, the small cell has an area overlapping with the macro cell Control device.

2. All areas of the small cell overlap with the area of the macro cell The control device according to claim 1.

3. The ratio of the area of the small cell overlapping with the area of the macro cell is equal to or greater than a first threshold value The control device according to claim 1.

4. The control unit allocates the group traffic according to the area of the cells constituted by the plurality of base station apparatuses, and calculates corrected individual traffic for each of the plurality of base station apparatuses The control device according to claim 1.

5. The control unit calculates individual traffic for each of the plurality of base station apparatuses using the group traffic, adds the first individual traffic of a first base station apparatus to be stopped as an increment to the second individual traffic of a second base station apparatus not to be stopped, and determines stop or operation of the base station apparatus in consideration of the increment The control device according to claim 1.

6. The group traffic is output from a learning model that predicts the group traffic learned from the measured traffic of each of the plurality of base station apparatuses The control device according to claim 1.

7. A prediction step of predicting group traffic of a group including a plurality of base station apparatuses using individual traffic for each base station apparatus; A control step of determining stop or operation of the base station apparatuses in the group according to the group traffic, and the group is composed of a base station apparatus that constitutes a macro cell and a base station apparatus that constitutes a small cell, the small cell has an area overlapping with the macro cell Base station control method.

Images