Communication control device and communication control method

Abstract

This communication control device comprises: an acquisition unit that acquires, from each of a plurality of lower-level devices that are communication devices wirelessly connected to a terminal, wireless control information used for controlling path switching between the plurality of lower-level devices and at least one upper-level device that is a communication device connected to any of the lower-level devices; a traffic prediction unit that predicts the future traffic volume in each of the plurality of lower-level devices on the basis of the wireless control information, and predicts the future traffic volume in the upper-level device on the basis of the prediction result obtained thereby; and a switching determination unit that predicts whether future congestion will occur on the basis of the predicted future traffic volume in the upper-level device and determines the execution of the path switching.

Description

Communication control device and communication control method

[0001] The present invention relates to a communication control device and a communication control method. This application claims priority based on PCT / JP2024 / 38645 filed in Japan on October 30, 2024, and PCT / JP2025 / 814 filed in Japan on January 14, 2025, and incorporates the contents thereof herein by reference.

[0002] When multiple upper-level devices and multiple lower-level devices communicate with each other, or when communication is simultaneously performed by multiple communication devices, congestion may occur and communication delays may occur. Congestion can sometimes be resolved by switching the communication path (hereinafter referred to as "path switching") by a switching device installed in the communication path and changing the connection relationship between the upper-level device and the lower-level device. If the communication traffic (hereinafter simply referred to as "traffic") concentrated on a specific communication device (for example, an upper-level device) can be dispersed to multiple communication devices by path switching, congestion can be resolved and communication delays can be reduced.

[0003] Conventionally, in a communication system having a switching device capable of changing the connection relationship between an upper-level device and a lower-level device, information indicating the amount of traffic (hereinafter simply referred to as "traffic volume") measured by the upper-level device or the lower-level device is collected, and based on the collected traffic volume, it is determined whether congestion has occurred, and a path switching is performed. A conventional communication system, for example, calculates a congestion value indicating the degree of congestion based on the traffic volume of each communication path, and performs path switching when the congestion value of any communication path exceeds a threshold.

[0004] However, generally, path switching takes a certain amount of time. Therefore, in a conventional communication system, it may be in a congested state from the occurrence of congestion until the path switching is completed. Thus, the time required for path switching is also one of the factors causing communication delays. Since it takes time to collect the actual traffic volume and then perform path switching, it is difficult to determine in real time (for example, within a few milliseconds) whether congestion has occurred and execute path switching within a predetermined period (for example, within several tens of milliseconds).

[0005] In contrast, there is a technology that avoids congestion by predicting future traffic volume and switching routes in advance based on the predicted value, rather than switching routes based on actual traffic volume (see, for example, Patent Document 1). Such technology can predict and prevent congestion from occurring, thereby further reducing communication delays.

[0006] Generally, traffic often exhibits periodicity due to burst traffic, such as video traffic and traffic in time-division duplex (TDD) communications (see, for example, Non-Patent Document 1). Therefore, especially in situations where terminals do not move much, it is possible to predict future traffic volume with greater accuracy by identifying the periodicity of traffic from past changes in traffic volume and making predictions based on the identified periodicity.

[0007] International Publication No. 2023 / 181139

[0008] "4G / LTE-TDD", ShareTechnote, [Retrieved October 18, 2024], Internet (URL: https: / / www.sharetechnote.com / html / LTE_TDD_Overview.html)

[0009] On the other hand, in situations where terminals are frequently in motion, the number of terminals within the coverage area of ​​the communication station connecting to that terminal fluctuates randomly, making it difficult to identify traffic periodicity from past traffic volumes. Consequently, in the past, there was a challenge in accurately predicting future traffic volumes and achieving low-latency communication in situations where terminals were frequently in motion.

[0010] To address these challenges, one possible approach is to predict changes in the number of terminals by forecasting terminal movement, and then predict future traffic volume based on the predicted changes in the number of terminals. However, this method requires predicting the movement of each individual terminal. Therefore, especially in situations where there are many terminals within a communication station's coverage area, the computational load required for predicting the movement of each terminal becomes enormous.

[0011] In view of the above circumstances, the present invention aims to provide a technology that can predict traffic volume with greater accuracy while suppressing an increase in computational load, even in situations where terminals are frequently moved.

[0012] One aspect of the present invention is a communication control device comprising: an acquisition unit that acquires wireless control information used for controlling route switching between a plurality of lower-level devices, which are communication devices wirelessly connected to a terminal, and at least one upper-level device, which is a communication device connected to any of the lower-level devices; a traffic prediction unit that predicts the future traffic volume of each of the plurality of lower-level devices based on the wireless control information and predicts the future traffic volume of the upper-level device based on the prediction results; and a switching determination unit that predicts whether or not future congestion will occur based on the predicted future traffic volume of the upper-level device and makes a decision on whether to execute the route switching.

[0013] Furthermore, one aspect of the present invention is a communication control method executed by a computer, comprising: an acquisition step of acquiring wireless control information used for controlling route switching between a plurality of lower-level devices, which are communication devices wirelessly connected to a terminal, and at least one upper-level device, which is a communication device connected to any of the lower-level devices; a prediction step of predicting the future traffic volume at the plurality of lower-level devices based on the wireless control information and predicting the future traffic volume at the upper-level device based on the prediction results; and a switching determination step of predicting whether or not future congestion will occur based on the predicted future traffic volume at the upper-level device and making a decision on whether to execute the route switching.

[0014] This invention makes it possible to predict traffic volume with greater accuracy while suppressing the increase in computational load, even in situations where terminals are frequently in motion.

[0015] This is a diagram showing the overall configuration of the communication system 1 in an embodiment of the present invention. This is a diagram showing the overall configuration of the mobile communication system 10 in a first embodiment of the present invention. This is a block diagram showing the functional configuration of the switching instruction device 17 in a first embodiment of the present invention. This is a diagram for explaining the future traffic volume prediction processing by the switching instruction device 17 in a first embodiment of the present invention. This is a flowchart showing the operation of the switching instruction device 17 in a first embodiment of the present invention. This is a block diagram showing the functional configuration of the switching instruction device 17a in a second embodiment of the present invention. This is a diagram for explaining the future traffic volume prediction processing by the switching instruction device 17a in a second embodiment of the present invention. This is a diagram showing an example of the coverage area of ​​the distributed station 13 in a second embodiment of the present invention. This is a flowchart showing the operation of the switching instruction device 17a in a second embodiment of the present invention. This is a block diagram showing the functional configuration of the switching instruction device 17b in a third embodiment of the present invention. This is a flowchart showing the operation of the switching instruction device 17b in a third embodiment of the present invention. This is a block diagram showing the functional configuration of the switching instruction device 17c in a fourth embodiment of the present invention. This is a flowchart showing the operation of the switching instruction device 17c in a fourth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching instruction device 17d in a fifth embodiment of the present invention. This is a flowchart showing the operation of the switching instruction device 17d in a fifth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching instruction device 17e in a sixth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17e in the sixth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17f in the seventh embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17f in the seventh embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17g in a modified example of the seventh embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17g in a modified example of the seventh embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17h in the eighth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17h in the eighth embodiment of the present invention.This is a block diagram showing the functional configuration of the switching indicator device 17i in the ninth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17i in the ninth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17j in the first modified example of the ninth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17j in the first modified example of the ninth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17k in the second modified example of the ninth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17k in the second modified example of the ninth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17l in the tenth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17l in the tenth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17m in the tenth modified example of the present invention. This is a flowchart showing the operation of the switching indicator device 17m in the tenth modified example of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17n in the eleventh embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17n in the eleventh embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17o in the twelfth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17o in the twelfth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17p in the thirteenth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17p in the thirteenth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17q in the fourteenth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17q in the fourteenth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17r in the fifteenth embodiment of the present invention. This is a flowchart showing the operation of the switching indicator device 17r in the fifteenth embodiment of the present invention. This is a block diagram showing the functional configuration of the switching indicator device 17s in the sixteenth embodiment of the present invention. This is a diagram showing the hardware configuration of the switching indicator device in the first to sixteenth embodiments of the present invention and each modified example.

[0016] Embodiments of the present invention will be described in detail below with reference to the drawings.

[0017] [Overview] First, before describing the individual embodiments of the present invention in detail, we will explain the basic system configuration that is common to each embodiment described below.

[0018] Figure 1 shows the overall configuration of a communication system 1 in an embodiment of the present invention. The communication system 1 includes a plurality of lower-level devices 3, a plurality of upper-level devices 4, and a switching instruction device 7. The lower-level devices 3 and the upper-level devices 4 are connected via a network 5. The network 5 is composed of one or more transfer devices (not shown; devices corresponding to the transfer device 15 shown in Figure 2, which will be described later). The transfer device also functions as a switching device that can change the connection relationship between the lower-level devices 3 and the upper-level devices 4.

[0019] In Figure 1, the four lower-level devices 3 are labeled "lower-level device 3-1" to "lower-level device 3-4," and the two upper-level devices 4 are labeled "upper-level device 4-1" and "upper-level device 4-2." However, the number of lower-level devices 3 and upper-level devices 4 is not limited to these numbers and can be any number. Hereafter, the direction from lower-level device 3 to upper-level device 4 will be referred to as "up," and the direction from upper-level device 4 to lower-level device 3 will be referred to as "down."

[0020] The lower-level device 3 transmits an uplink signal to the connected upper-level device 4. Network 5 forwards the uplink signal to the upper-level device 4 according to the communication path between the lower-level device 3 and the upper-level device 4. The upper-level device 4 also transmits a downlink signal to the connected lower-level device 3. Network 5 forwards the downlink signal to the lower-level device 3 according to the communication path between the lower-level device 3 and the upper-level device 4.

[0021] The switching instruction device 7 instructs each device in the communication system 1 to switch routes. The switching instruction device 7 uses information such as traffic volume and the number of terminals obtained from the lower-level device 3 to predict in advance the traffic volume of some or all communication links (hereinafter referred to as "links") in the communication path between the lower-level device 3 and the upper-level device 4. For each link, the switching instruction device 7 uses the predicted traffic volume information and the switching threshold to determine whether congestion will occur.

[0022] A switching threshold is a threshold used to determine whether or not congestion requiring route switching will occur. The switching threshold is predetermined for each link, for example, based on the link's transmission capacity. An indicator used to represent the link's transmission capacity is, for example, the link rate. The link rate is the maximum communication speed of the link.

[0023] The switching instruction device 7 determines that route switching for load balancing should be performed if congestion is predicted to occur on any of the links. The switching instruction device 7 instructs each device (transmitter, downstream device 3, and upstream device 4, etc.) to switch routes so that at least a portion of the traffic transmitted on the link where congestion is predicted to occur is transmitted on the link where congestion is not predicted to occur. This prevents congestion delays on the links and makes it possible to improve the efficiency of bandwidth utilization.

[0024] As shown in Figure 1, the switching instruction device 7 has a prediction unit 72 and a switching decision unit 75. The prediction unit 72 acquires wireless control information from each device (subordinate device 3, etc.) that is used to predict the traffic volume of each link. Wireless control information includes, for example, the traffic volume observed by the device, the traffic allocation to the device, the number of connected terminals to the device, the location information of the device, the location information and radio wave environment of the terminals, and the required communication quality (hereinafter referred to as "required quality"). Note that the wireless control information may be one or more of these pieces of information. The prediction unit 72 uses the wireless control information to predict the future traffic volume for each link (hereinafter referred to as "future traffic volume").

[0025] The switching determination unit 75 determines in advance whether congestion will occur for each link, using the predicted future traffic volume and the switching threshold. If congestion is predicted, the switching determination unit 75 decides to perform a route switch for load balancing. For example, the switching determination unit 75 decides to perform a route switch from the link where congestion is predicted to a link with lower link utilization. The switching determination unit 75 instructs each device (transfer device, lower-level device 3, and upper-level device 4, etc.) to execute the process for route switching.

[0026] Several embodiments of applying the above-described communication system 1 to a mobile communication system will be described in detail below.

[0027] <First Embodiment> The first embodiment of the present invention will be described below.

[0028] [Overall Configuration of the Communication System] Figure 2 shows the overall configuration of the mobile communication system 10 in the first embodiment of the present invention. The mobile communication system 10 is an example of the communication system 1 described above.

[0029] The mobile communication system 10 is, for example, a fifth-generation mobile communication system (5G). The mobile communication system 10 includes a terminal station 11 (terminal), an antenna station 12, a distributed station 13, an aggregation station 14, a transfer device 15, a switching instruction device 17, and a resource allocation device 16.

[0030] The terminal station 11, antenna station 12, distributed station 13, and aggregation station 14 are the UE (User Equipment), RU (Radio Unit), DU (Distributed Unit), and CU (Central Unit) of the fifth-generation mobile communication system, respectively. The distributed station 13 is an example of a lower-level device 3, and the aggregation station 14 is an example of a higher-level device 4. However, the combination of the device corresponding to the lower-level device 3 and the device corresponding to the higher-level device 4 is not limited to the distributed station 13 and the aggregation station 14, but is arbitrary. The forwarding device 15 is an example of a forwarding device that constitutes the network 5.

[0031] The transfer device 15 also functions as a switching device that can change the connection relationship between the distributed station 13 and the aggregation station 14. The transfer function and switching function of the transfer device 15 may be implemented by separate devices. Furthermore, the switching instruction device 17, the transfer device 15, and the resource allocation device 16 do not necessarily have to be separate devices; they may be configured as an integrated device.

[0032] The aggregation station 14 of the mobile communication system 10 is connected to the upper-level network 20.

[0033] Hereafter, the distributed stations 13 of M units (where M is an integer greater than or equal to 1) will be referred to as "distributed station 13-1" to "distributed station 13-M," respectively. Also, below, the K under distributed station 13-m (where m is an integer greater than or equal to 1 and less than or equal to M) ｍ Stand (K ｍ An antenna station 12 with (where N is an integer of 1 or more) will be referred to as "antenna station 12-m". Similarly, N aggregation stations 14 (where N is an integer of 2 or more) will be referred to as "aggregation station 14-1" through "aggregation station 14-N", respectively. Therefore, the mobile communication system 10 shown in Figure 2 has M=4, K １ = 2, K ２ = 2, K ３ = 2, K ４ This is an example of a mobile communication system where N = 2 and N = 2.

[0034] Terminal station 11 uses radio resources allocated from distributed station 13 to send and receive radio signals with antenna station 12. The radio resources allocated from distributed station 13 include, for example, information indicating the start and end timings of time intervals in which the transmission and reception of radio signals are permitted. The start and end timings are represented, for example, by slots. A slot is a scheduling unit for data transmission and reception in a radio frame. The allocated radio resources may also include information indicating the coding rate and modulation scheme, etc.

[0035] Antenna station 12 receives uplink data from terminal station 11 via a wireless signal. Antenna station 12-m sets the received uplink data as an uplink signal. Then, antenna station 12-m transmits the uplink signal to distributed station 13-m via a wired interface.

[0036] Furthermore, antenna station 12-m receives downlink signals from distributed station 13-m via a wired interface. Antenna station 12 transmits the downlink data destined for terminal station 11, which is set in the received downlink signal, to terminal station 11 via a wireless signal.

[0037] Distributed station 13-m is K ｍ Uplink signals are received from each of the antenna stations 12-m. The uplink signals received by the distributed station 13-m include the uplink data received by antenna station 12-m from its subordinate terminal stations 11. The distributed station 13 generates an uplink signal by aggregating the uplink data from each terminal station 11 and transmits the generated uplink signal to the aggregation station 14 to which it is connected.

[0038] Furthermore, the distributed station 13 receives a downlink signal from the aggregation station 14 to which it is connected, which contains downlink data destined for its subordinate terminal stations 11. The distributed station 13-m converts the received downlink signal into a downlink signal corresponding to the radio signal transmitted from each antenna station 12-m. Then, the distributed station 13-m transmits the converted downlink signal to the antenna station 12-m corresponding to that downlink signal.

[0039] The aggregation station 14 aggregates the uplink signals received from its subordinate distributed stations 13 and forwards them to the higher-level network 20. The aggregation station 14 also receives downlink signals from the higher-level network 20 that contain downlink data destined for terminal stations 11, and forwards the received downlink signals to the distributed stations 13 connected to the destination terminal station 11.

[0040] The forwarding device 15 is connected to the distributed station 13, the aggregation station 14, and the switching instruction device 17. The forwarding device 15 also functions as a device that performs route switching. The forwarding device 15 forwards signals according to the communication path between the distributed station 13 and the aggregation station 14. That is, the forwarding device 15 forwards the uplink signal received from the distributed station 13 to the destination aggregation station 14 according to the communication path.

[0041] Further, the transfer device 15 transfers the downstream signal received from the concentration station 14 to the destination distribution station 13 along the communication path. The transfer of the signal along the communication path is executed under the instruction from the switching instruction device 17. The resource allocation device 16 manages the resources of the concentration station 14.

[0042] The switching instruction device 17 is connected to the distribution station 13, the concentration station 14, and the transfer device 15. Further, the switching instruction device 17 may be connected to the resource allocation device 16. The switching instruction device 17 instructs the transfer device 15 to switch the path between the distribution station 13 and the concentration station 14.

[0043] In the mobile communication system 10 of the present embodiment, the communication path between the distribution stations 13 and the concentration stations 14 is determined by the connection relationship between the distribution stations 13 and the concentration stations 14. Therefore, in the mobile communication system 10, the transfer device 15 is controlled to change the concentration station 14 to which the distribution station 13 is connected, thereby performing path switching between the distribution station 13 and the concentration station 14.

[0044] Note that the switching instruction device 17 of the present embodiment is configured to determine the presence or absence of future congestion based on the prediction result of the upstream traffic volume. Therefore, hereinafter, "traffic" shall mean the upstream signal traffic unless otherwise specified.

[0045] [Functional Configuration of Switching Instruction Device] Hereinafter, the functional configuration of the switching instruction device 17 will be described. The switching instruction device 17 is a device that predicts the future traffic volume in each concentration station 14 to predict the presence or absence of congestion, and performs control to execute path switching as necessary. The switching instruction device 17 in the first embodiment predicts the future traffic volume in each distribution station 13 based on the recent trend (variation amount) of the traffic volume per terminal station 11 in each distribution station 13. Then, the switching instruction device 17 predicts the future traffic volume in each concentration station 14 based on the future traffic volume in each distribution station 13.

[0046] Figure 3 is a block diagram showing the functional configuration of the switching instruction device 17 in the first embodiment of the present invention. As shown in Figure 3, the switching instruction device 17 is configured to include a prediction unit 172 and a switching determination unit 175. The prediction unit 172 is configured to include a traffic prediction unit 174.

[0047] The prediction unit 172 receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. The radio control information in the first embodiment includes data that associates the traffic volume at each distributed station 13 with the number of terminal stations 11 that are located within the coverage area of ​​each distributed station 13 and communicate with it (hereinafter referred to as "number of connected terminals") with time. The traffic prediction unit 174 of the prediction unit 172 uses the radio control information to predict the future traffic volume at each distributed station 13.

[0048] Furthermore, the prediction unit 172 calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172 outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0049] The switching decision unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172 and a predetermined switching threshold. For example, the switching decision unit 175 calculates a predicted link utilization rate based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172 and a predetermined switching threshold. The switching decision unit 175 then predicts whether or not future congestion will occur based on the predicted link utilization rate.

[0050] The predetermined switching threshold is a value set in advance based on, for example, the link rate and bandwidth of the upstream link. The switching determination unit 175 determines, for example, if there is an aggregation station 14 where it is determined that congestion will occur because the amount of traffic will exceed the bandwidth of the upstream link in the future, it will perform a route switch to distribute the load to a link with available bandwidth. When the switching determination unit 175 determines that a route switch should be performed, it issues instructions to each device (distributed station 13, aggregation station 14 and forwarding device 15) to change the aggregation station 14 to which the distributed station 13 is connected for route switching.

[0051] Specifically, the switching determination unit 175 instructs the aggregation station 14 of the destination before the switch (hereinafter referred to as the "source") to release the connection with the distributed station 13, and instructs the aggregation station 14 of the destination after the switch (hereinafter referred to as the "destination") to connect with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch the route so that signal transfer is carried out via the communication path after the destination switch.

[0052] [Predicting Future Traffic Volume] The prediction process for future traffic volume at each distributed station 13, performed by the prediction unit 172, will be explained in more detail below.

[0053] The prediction unit 172 acquires wireless control information from each distributed station 13, including the traffic volume and the number of connected terminals at that distributed station 13, for example, on a unit period basis. The traffic prediction unit 174 of the prediction unit 172 calculates the traffic volume per terminal station 11 for several past periods (for example, the most recent five unit periods) based on the traffic volume and the number of connected terminals for each unit period. Then, the traffic prediction unit 174 predicts the future traffic volume per terminal station 11 at each distributed station 13 (for example, the traffic volume for the next unit period) based on the trend of the traffic volume per terminal station 11 over several past periods.

[0054] At this time, the traffic prediction unit 174 predicts the future traffic volume per terminal station 11 in each distributed station 13 by weighting the prediction values ​​so that more recent fluctuations in traffic volume are given greater weight. The following will explain in more detail the future traffic volume prediction process with such weighting using numerical examples.

[0055] Figure 4 is a diagram illustrating the future traffic volume prediction process by the switching instruction device 17 in the first embodiment of the present invention. Figure 4 shows data generated based on radio control information acquired from a single distributed station 13. As shown in Figure 4, the data is in tabular format, with the time, traffic volume, number of connected terminals, traffic volume per terminal station, and traffic volume fluctuation amount associated with each other.

[0056] The prediction unit 172 acquires wireless control information, including time, traffic volume, and number of connected terminals, from each distributed station 13, as shown in Figure 4. As shown in Figure 4, the prediction unit 172 sequentially acquires wireless control information, including traffic volume and number of connected terminals, at times t(-4), t(-3), t(-2), t(-1), and t(0), for example.

[0057] Note that times t(-4), t(-3), t(-2), t(-1), and t(0) are times within a unit period, with time t(0) being the most recent time. For example, as shown in Figure 4, the traffic volume at time t(-4) in a certain distributed station 13 is 2.0 [Gbps], and the number of connected terminals is 10 [units]. Also, for example, as shown in Figure 4, the traffic volume at time t(-3) in the same distributed station 13 is 2.0 [Gbps], and the number of connected terminals is 12 [units].

[0058] Here, as an example, we will explain a case where future traffic volume is predicted based on data from the five most recent time periods. Therefore, Figure 4 shows the data for the five most recent time periods. However, the number of time periods used for the data used in the prediction is not limited to five, but can be any number.

[0059] The traffic prediction unit 174 of the prediction unit 172 calculates the traffic volume per terminal station 11 at each time by dividing the traffic volume value by the number of connected terminals at the same time.

[0060] Therefore, the calculated traffic volume per terminal station 11 at each time point is as shown in Figure 4: 0.2 [Gbps] at time t(-4), 0.17 [Gbps] at time t(-3), 0.175 [Gbps] at time t(-2), 0.136 [Gbps] at time t(-1), and 0.24 [Gbps] at time t(0).

[0061] Next, the traffic prediction unit 174 of the prediction unit 172 calculates the amount of change in the amount of traffic per terminal station 11 for each unit period based on the calculated amount of traffic per terminal station 11 at each time point. The amount of change in the amount of traffic here refers to the difference (one unit period) between the preceding and succeeding time points.

[0062] For example, the traffic volume per terminal station 11 is 0.2 [Gbps] at time t(-4) and 0.17 [Gbps] at time t(-3). Therefore, as shown in Figure 4, the change in traffic volume per terminal station 11 between time t(-4) and t(-3) is -0.03 [Gbps].

[0063] In this way, the traffic prediction unit 174 determines that the change in the amount of traffic per terminal station 11 between time t(-4) and t(-3) is -0.03 [Gbps], the change in the amount of traffic per terminal station 11 between time t(-3) and t(-2) is +0.005 [Gbps], the change in the amount of traffic per terminal station 11 between time t(-2) and t(-1) is -0.039 [Gbps], and the change in the amount of traffic per terminal station 11 between time t(-1) and t(0) is +0.104 [Gbps].

[0064] The traffic prediction unit 174 predicts the future traffic volume per terminal station 11 at each distributed station 13 based on the trend of the fluctuation amount of traffic volume per terminal station 11 calculated in this way. At this time, as described above, the traffic prediction unit 174 predicts the future traffic volume per terminal station 11 by weighting the prediction so that the more recent fluctuation amount of traffic volume is given more weight in the predicted value.

[0065] For example, the traffic prediction unit 174 assigns progressively smaller weights to the following fluctuations in traffic volume per terminal station 11: the fluctuation in traffic volume per terminal station 11 between the most recent time interval t(-1) to t(0), the fluctuation in traffic volume per terminal station 11 between the time interval t(-2) to t(-1), the fluctuation in traffic volume per terminal station 11 between the time interval t(-3) to t(-2), and the fluctuation in traffic volume per terminal station 11 between the time interval t(-4) to t(-3). For example, the traffic prediction unit 174 assigns weights to each of the five fluctuations in the ratio of 8:4:2:1, starting from the most recent unit period. Note that the weight values ​​are not limited to these and can be arbitrary.

[0066] As a result, the formula used by the traffic prediction unit 174 to predict the future traffic volume per terminal station 11 in a given distributed station 13 is, for example, as shown in equation (1) below.

[0067] Future traffic volume per terminal station 11 = 0.24 + 0.104 × (8 / 15) - 0.039 × (4 / 15) + 0.005 × (2 / 15) - 0.03 × (1 / 15) ... (1)

[0068] Based on the wireless control information shown in Figure 4, the value of the future traffic volume per terminal station 11 in a given distributed station 13 is 0.284 [Gbps] by calculating the above equation (1). The traffic prediction unit 174 can predict the future traffic volume in the distributed station 13 by multiplying this future traffic volume per terminal station 11 by the future number of connected terminals in the distributed station 13.

[0069] Furthermore, the future number of connected terminals may be a value predicted based on, for example, the fluctuation in the number of connected terminals over the past few periods, or it may be the value of the most recent number of connected terminals (for example, in the case of Figure 4, 10 [units] at the most recent time, time t(0)). Therefore, for example, if the future number of connected terminals is 10 [units], the future traffic volume at distributed station 13 is predicted to be 2.84 [Gbps] (= (future traffic volume per terminal station 11) 0.284 [Gbps] × (future number of connected terminals) 10 [units]).

[0070] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17 will be described below. Figure 5 is a flowchart showing the operation of the switching instruction device 17 in the first embodiment of the present invention. The operation of the switching instruction device 17 shown in the flowchart of Figure 5 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0071] First, the prediction unit 172 obtains wireless control information from each distributed station 13, which associates the time, traffic volume, and number of connected terminals (step S001). Next, the traffic prediction unit 174 of the prediction unit 172 calculates the traffic volume per terminal station 11 at each time for each distributed station 13 by dividing the traffic volume value by the number of connected terminals (step S002).

[0072] Next, the traffic forecasting unit 174 calculates the amount of change in the amount of traffic per terminal station 11 for each unit period based on the calculated amount of traffic per terminal station 11 at each time (step S003). Next, the traffic forecasting unit 174 predicts the future amount of traffic per terminal station 11 based on the calculated change in the amount of traffic per terminal station 11 for each unit period (step S004).

[0073] At this time, the traffic prediction unit 174 predicts the future traffic volume per terminal station 11 by weighting the prediction values ​​so that more recent fluctuations in traffic volume are given greater consideration, as described above.

[0074] Next, the traffic prediction unit 174 calculates the future traffic volume at each distributed station 13 by multiplying the future traffic volume per terminal station 11 by the future number of connected terminals at that distributed station 13 (step S005). Next, the traffic prediction unit 174 calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14 and the future traffic volume at each distributed station 13 (step S006).

[0075] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172 and a predetermined switching threshold (step S007). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S008, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S009).

[0076] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregation station 14 to which the distributed station 13 is connected (step S010). Specifically, the switching determination unit 175 instructs the aggregation station 14 from which the switching is initiated to release the connection with the distributed station 13, and instructs the aggregation station 14 to which the switching is initiated to establish a connection with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is performed via the communication path after the connection destination has been switched.

[0077] This completes the operation of the switching instruction device 17 as shown in the flowchart of Figure 5.

[0078] As described above, the switching instruction device 17 in the first embodiment of the present invention predicts the future traffic volume at each distributed station 13 based on the recent trend (fluctuation) of the traffic volume per terminal station 11 at each distributed station 13. Then, the switching instruction device 17 predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion is determined to occur, the switching instruction device 17 performs route switching to distribute the load to links with available bandwidth.

[0079] With this configuration, the switching instruction device 17 in the first embodiment of the present invention can predict future traffic volume without having to perform movement predictions for each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17 can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0080] <Second Embodiment> A second embodiment of the present invention will be described below.

[0081] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the second embodiment (hereinafter referred to as "mobile communication system 10a") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10a is an example of the communication system 1 described above. The mobile communication system 10a in the second embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device. Hereinafter, the switching instruction device in the second embodiment will be referred to as "switching instruction device 17a".

[0082] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17a will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17a is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17a in this embodiment and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0083] In the second embodiment, the switching instruction device 17a predicts the future traffic volume at each distributed station 13 based on the traffic volume and number of connected terminals of the distributed station 13 that is the target of future traffic volume prediction (hereinafter referred to as "the distributed station 13 to be predicted") and the traffic volume and number of connected terminals of other distributed stations 13 adjacent to the distributed station 13 in coverage area (hereinafter referred to as "the adjacent distributed station 13").

[0084] More specifically, in the second embodiment, the switching instruction device 17a predicts the future traffic volume at each distributed station 13, taking into account the inflow of terminal stations 11 from adjacent distributed stations 13 to the distributed station 13 to be predicted. Then, the switching instruction device 17 predicts the future traffic volume at each aggregate station 14 based on the future traffic volume at each distributed station 13.

[0085] Figure 6 is a block diagram showing the functional configuration of the switching instruction device 17a in a second embodiment of the present invention. As shown in Figure 6, the switching instruction device 17a is configured to include a prediction unit 172a and a switching determination unit 175. The prediction unit 172a is configured to include a traffic prediction unit 174a.

[0086] The prediction unit 172a receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. In the second embodiment, the radio control information includes data that associates the most recent traffic volume value at each distributed station 13 with the most recent number of connected terminals at each distributed station 13. The traffic prediction unit 174a of the prediction unit 172a predicts the future traffic volume at each distributed station 13 using the radio control information. At this time, the traffic prediction unit 174a predicts the future traffic volume at each distributed station 13, taking into account the influx of terminal stations 11 into the target distributed station 13 due to the movement of terminal stations 11 from the coverage area of ​​an adjacent distributed station 13 to the coverage area of ​​the target distributed station 13.

[0087] Furthermore, the prediction unit 172a calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172a outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0088] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172a and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0089] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0090] [Predicting Future Traffic Volume] The prediction process for future traffic volume at each distributed station 13, performed by the prediction unit 172a, will be explained in more detail below.

[0091] The prediction unit 172a acquires radio control information from each distributed station 13, including the traffic volume and the number of connected terminals at that distributed station 13. The traffic prediction unit 174a of the prediction unit 172a predicts the future traffic volume at the distributed station 13, taking into account the inflow of terminal stations 11 from adjacent distributed stations 13 to the distributed station 13, based on the traffic volume and number of connected terminals of the distributed station 13 to be predicted and the traffic volume and number of connected terminals of the adjacent distributed station 13.

[0092] At this time, the traffic prediction unit 174a predicts the future traffic volume per terminal station 11 in the target distribution station 13 by weighting the ease with which terminal stations 11 can flow in from each adjacent distribution station 13 to the target distribution station 13. The ease with which terminal stations 11 can flow in from each adjacent distribution station 13 to the target distribution station 13 may be determined in advance based on, for example, past movement records of terminal stations 11, or it may be determined based on the distance (positional relationship) between the coverage area of ​​the target distribution station 13 and the terminal station 11. A more specific explanation will be given below with numerical examples.

[0093] Figure 7 is a diagram illustrating the future traffic volume prediction processing by the switching instruction device 17a in the second embodiment of the present invention. Figure 7 shows tabular data where the distributed station ID, traffic volume, number of connected terminals, and traffic volume per terminal station are associated at a certain point in time. The distributed station ID is identification information that identifies the distributed station 13. In other words, Figure 7 shows data for each distributed station 13.

[0094] Furthermore, as an example, we will assume that the target distributed station 13 is the distributed station 13 with distributed station ID "3", and that the coverage area of ​​each distributed station 13 is arranged as shown in Figure 8.

[0095] Figure 8 shows an example of the coverage area of ​​a distributed station 13 in a second embodiment of the present invention. As shown in Figure 8, the coverage area of ​​distributed station 13 with distributed station ID "3" is adjacent to the coverage areas of distributed stations 13 with distributed station IDs "1", "2", and "4", respectively. Therefore, if the distributed station 13 to be predicted is distributed station 13 with distributed station ID "3", then the distributed stations 13 adjacent to it are distributed stations 13 with distributed station IDs "1", "2", and "4".

[0096] The prediction unit 172a acquires wireless control information from each distributed station 13, including the distributed station ID, traffic volume, and number of connected terminals, from the data shown in Figure 7. For example, as shown in Figure 7, the traffic volume at distributed station 13 with distributed station ID "1" is 2.0 [Gbps] and the number of connected terminals is 10 [units]. Also, as shown in Figure 7, the traffic volume at the predicted distributed station 13 with distributed station ID "3" is 2.1 [Gbps] and the number of connected terminals is 12 [units].

[0097] The traffic prediction unit 174a of the prediction unit 172a calculates the traffic volume per terminal station 11 for each distributed station 13 by dividing the traffic volume value by the number of connected terminals.

[0098] Therefore, as shown in Figure 7, the calculated traffic volume per terminal station 11 is 0.2 [Gbps] for distributed station 13 with distributed station ID "1", 0.075 [Gbps] for distributed station 13 with distributed station ID "2", 0.175 [Gbps] for distributed station 13 with distributed station ID "3", 0.12 [Gbps] for distributed station 13 with distributed station ID "4", and 0.14 [Gbps] for distributed station 13 with distributed station ID "5".

[0099] The traffic prediction unit 174 predicts the future traffic volume per terminal station 11 in the target distribution station 13 based on the traffic volume per terminal station 11 in each distribution station 13 calculated in this manner. At this time, the traffic prediction unit 174 predicts the future traffic volume per terminal station 11 in the target distribution station 13 by considering the inflow ratio of terminal stations 11 from adjacent distribution stations 13 to the target distribution station 13.

[0100] Here, as an example, the ratio of terminal stations 11 flowing from an adjacent distributed station 13 to the distributed station 13 being predicted is assumed to be as follows. Note that, as shown below, the number of terminal stations 11 remaining within the coverage area of ​​the distributed station 13 being predicted is not the number of terminals flowing in, but the number of terminal stations 11 remaining within the coverage area of ​​that distributed station 13.

[0101] The ratio of terminals flowing from the adjacent distributed station ID "1" to the predicted distributed station 13 is 2:1:8:1. Note that the ratio of terminal stations 11 flowing from the adjacent distributed station ID "2" to the predicted distributed station 13 is not limited to this and is arbitrary.

[0102] Note that the distributed station 13 with distributed station ID "5" does not fall under the category of adjacent distributed station 13 because its coverage area is not adjacent to that of the distributed station 13 with distribution station ID "3" that is the target of prediction. In this embodiment, the calculation is performed assuming that there is no inflow of terminal stations 11 from distributed stations 13 other than adjacent distributed stations 13 to the distributed station 13 that is the target of prediction.

[0103] The traffic prediction unit 174a predicts the future traffic volume per terminal station 11 in the target distributed station 13 by weighting the numbers of terminals entering the network (and the number of terminals remaining at the target distributed station 13) according to the ratio. As a result, the calculation formula used by the traffic prediction unit 174a to predict the future traffic volume per terminal station 11 in the target distributed station 13 is, for example, equation (2) below.

[0104] Future traffic volume per terminal station 11 = 0.2 × 2 + 0.075 × 1 + 0.175 × 8 + 0.12 × 1 ... (2)

[0105] Based on the wireless control information shown in Figure 7, the value of the future traffic volume per terminal station 11 at the distributed station 13 targeted for prediction of distributed station ID "3" is 0.166 [Gbps] by calculating the above equation (2). The traffic prediction unit 174a can predict the future traffic volume at the distributed station 13 by multiplying this future traffic volume per terminal station 11 by the future number of connected terminals at the distributed station 13.

[0106] Furthermore, the future number of connected terminals may be a value predicted by considering, for example, the inflow ratio of terminal stations 11 to the predicted distributed station 13, or the value of the most recent number of connected terminals at the predicted distributed station 13 (for example, 12 [units] in the case of Figure 7) may be used. Therefore, for example, if the future number of connected terminals is 12 [units], the future traffic volume at the distributed station 13 is predicted to be 1.99 [Gbps] (= 0.166 [Gbps] × 12 [units]).

[0107] In the above, a specific calculation example was described when the distributed station 13 with distributed station ID "3" is the distributed station 13 to be predicted. However, the traffic prediction unit 174a similarly predicts the future traffic volume for distributed stations 13 with distributed station IDs "1", "2", "4", and "5". Then, based on the predicted future traffic volume at each distributed station 13, the traffic prediction unit 174a predicts the future traffic volume at the aggregation station 14.

[0108] In this embodiment, the prediction unit 172a predicts future traffic volume using the traffic volume and number of connected terminals at the distributed station 13 at the most recent time. However, it may also be configured in combination with the configuration described in the first embodiment above. That is, the prediction unit 172a may calculate the amount of change (trend) in traffic volume per terminal station 11 using the traffic volume and number of connected terminals at the distributed station 13 over the past few periods, and further predict future traffic volume considering the inflow ratio of terminal stations 11 to the distributed station 13 to be predicted.

[0109] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17a will be described below. Figure 9 is a flowchart showing the operation of the switching instruction device 17a in the second embodiment of the present invention. The operation of the switching instruction device 17a shown in the flowchart of Figure 9 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0110] First, the prediction unit 172a obtains wireless control information from each distributed station 13, which associates the distributed station ID, traffic volume, and number of connected terminals (step S101). Next, the traffic prediction unit 174a of the prediction unit 172a calculates the traffic volume per terminal station 11 for each distributed station 13 by dividing the traffic volume value by the number of connected terminals (step S102).

[0111] Next, the traffic prediction unit 174a predicts the future traffic volume per terminal station 11 in the target distribution station 13, taking into account the inflow ratio of terminal stations 11 from each adjacent distribution station 13 to the target distribution station 13 (step S103).

[0112] Next, the traffic prediction unit 174a calculates the future traffic volume at each distributed station 13 by multiplying the future traffic volume per terminal station 11 by the future number of connected terminals at that distributed station 13 (step S104). Next, the traffic prediction unit 174a calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14 and the future traffic volume at each distributed station 13 (step S105).

[0113] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172a and a predetermined switching threshold (step S106). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S107, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S108).

[0114] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregate station 14 to which the distributed station 13 is connected (step S109). Specifically, the switching determination unit 175 instructs the source aggregate station 14 to release the connection with the distributed station 13 and instructs the destination aggregate station 14 to connect with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is carried out via the communication path after the connection destination switch.

[0115] This completes the operation of the switching instruction device 17a as shown in the flowchart of Figure 9.

[0116] As described above, the switching instruction device 17a in the second embodiment of the present invention predicts the traffic volume per terminal station 11 at the distribution station 13 to be predicted, taking into account the inflow ratio of terminal stations 11 from each adjacent distribution station 13, and predicts the future traffic volume at each distribution station 13. Then, the switching instruction device 17a predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distribution station 13. If there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion is determined to occur, the switching instruction device 17a performs route switching to distribute the load to links with available bandwidth.

[0117] With this configuration, the switching instruction device 17a in the second embodiment of the present invention can predict future traffic volume without performing movement predictions for each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17a can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0118] <Third Embodiment> A third embodiment of the present invention will be described below.

[0119] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the third embodiment (hereinafter referred to as "mobile communication system 10b") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10b is an example of the communication system 1 described above. The mobile communication system 10b in the third embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device. Hereinafter, the switching instruction device in the third embodiment will be referred to as "switching instruction device 17b".

[0120] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17b will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17b is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17b and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0121] In the third embodiment, the switching instruction device 17b predicts the movement of terminal stations 11 based on the amount of change in the position of each terminal station 11 located within the coverage area of ​​the distribution station 13 to be predicted, and each terminal station 11 located around the said coverage area. Based on the movement prediction, the switching instruction device 17b predicts the future number of connected terminals within the coverage area of ​​the distribution station 13 to be predicted, and predicts the future traffic volume at each distribution station 13 based on the traffic volume at each distribution station 13 and the predicted number of connected terminals. Based on the future traffic volume at each distribution station 13, the switching instruction device 17b predicts the future traffic volume at each aggregation station 14.

[0122] Figure 10 is a block diagram showing the functional configuration of the switching instruction device 17b in a third embodiment of the present invention. As shown in Figure 10, the switching instruction device 17b is configured to include a prediction unit 172b and a switching determination unit 175. The prediction unit 172b is configured to include a future position estimation unit 173b and a traffic prediction unit 174b.

[0123] The prediction unit 172b receives radio control information from each distributed station 13, which is used to predict future traffic volume at each distributed station 13. In the third embodiment, the radio control information includes the most recent traffic volume value at each distributed station 13 and the location information of each terminal station 11 within the coverage area of ​​the distributed station 13.

[0124] The future position estimation unit 173b of the prediction unit 172b uses the position information for each terminal station 11 included in the radio control information to predict the movement of each terminal station 11 and estimate the number of connected terminals for each distributed station 13 in the future. Specifically, the future position estimation unit 173b identifies the position of each terminal station 11 located within the coverage area of ​​the distributed station 13 and in the surrounding area, based on the position information for each terminal station 11 included in the radio control information.

[0125] The location of each terminal station 11 located within the coverage area of ​​a distributed station 13 is determined based on radio control information obtained from that distributed station 13. Furthermore, the location of each terminal station 11 located in the area surrounding the coverage area of ​​a distributed station 13 is determined based on radio control information obtained from an adjacent distributed station 13.

[0126] The surrounding area referred to here is the range in which the terminal station 11 can enter the coverage area of ​​the distributed station 13 from the outside during the period to be predicted. For example, if the coverage area is a circular area centered on the location of the distributed station 13, then the area within the coverage area and its surroundings referred to here would be, for example, the area within a predetermined distance from the location of the distributed station 13 (for example, within a radius of several kilometers), or a circular area centered on the location of the distributed station 13 with a radius 1.2 times the radius of the coverage area.

[0127] The future location estimation unit 173b performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area, for each distributed station 13. The future location estimation unit 173b also performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area at predetermined unit intervals.

[0128] The future position estimation unit 173b then calculates the amount of change in the position of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11). Based on the calculated change amounts, the future position estimation unit 173b estimates the future position of each terminal station 11. The future position estimation unit 173b outputs position information indicating the estimated future position of each terminal station 11 to the traffic prediction unit 174b.

[0129] The future position of each terminal station 11 may be estimated, for example, by using the fluctuation amount in the most recent unit period as the predicted value of the future fluctuation amount, or by using the average value of the fluctuation amounts over multiple past unit periods as the predicted value of the future fluctuation amount.

[0130] The traffic prediction unit 174b of the prediction unit 172b obtains location information indicating the future location of each terminal station 11 from the future location estimation unit 173b. Based on the future location of each terminal station 11 estimated by the future location estimation unit 173b, the traffic prediction unit 174b calculates the future number of connected terminals for each distributed station 13.

[0131] The traffic prediction unit 174b predicts the future traffic volume at each distributed station 13 based on the calculated future number of connected terminals at each distributed station 13. For example, the traffic prediction unit 174b predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0132] Furthermore, the prediction unit 172b calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172b outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0133] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172b and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0134] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0135] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17b will be described below. Figure 11 is a flowchart showing the operation of the switching instruction device 17b in the third embodiment of the present invention. The operation of the switching instruction device 17b shown in the flowchart of Figure 11 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0136] First, the prediction unit 172b obtains radio control information from each distributed station 13, which includes the most recent traffic volume value at the distributed station 13 and the location information of each terminal station 11 within the coverage area of ​​the distributed station 13 (step S201). Next, the future location estimation unit 173b of the prediction unit 172b identifies the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding areas, based on the location information of each terminal station 11 included in the radio control information (step S202).

[0137] Next, the future position estimation unit 173b calculates the amount of change in the position of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11) (step S203). Next, the future position estimation unit 173b estimates the future position of each terminal station 11 based on the calculated amount of change (step S204).

[0138] Next, the traffic prediction unit 174b of the prediction unit 172b calculates the future number of connected terminals for each distributed station 13 based on the future location of each terminal station 11 estimated by the future location estimation unit 173b (step S205).

[0139] Next, the traffic prediction unit 174b predicts the future traffic volume at each distributed station 13 based on the calculated future number of connected terminals at each distributed station 13 (step S206). For example, the traffic prediction unit 174b predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0140] Next, the traffic prediction unit 174b calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13 (step S207).

[0141] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172b and a predetermined switching threshold (step S208). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S209, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S210).

[0142] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregation station 14 to which the distributed station 13 is connected (step S211). Specifically, the switching determination unit 175 instructs the aggregation station 14 from which the switching is initiated to release the connection with the distributed station 13, and instructs the aggregation station 14 to which the switching is initiated to establish a connection with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is performed through the communication path after the connection destination has been switched.

[0143] This completes the operation of the switching instruction device 17b as shown in the flowchart of Figure 11.

[0144] As described above, in the third embodiment of the present invention, the switching instruction device 17b estimates the future location of each terminal station 11 based on the location information of each terminal station 11 located within the coverage area of ​​each distributed station 13 and in the surrounding area. Based on the estimated future location of each terminal station 11, the switching instruction device 17b estimates the future number of connected terminals at each distributed station 13. Based on the recent traffic volume at each distributed station 13, the recent number of connected terminals, and the estimated future number of connected terminals, the switching instruction device 17b predicts the future traffic volume at each distributed station 13.

[0145] The switching instruction device 17b then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If the switching instruction device 17b determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0146] Furthermore, in this third embodiment, the switching instruction device 17b performs movement prediction only for terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas. Therefore, compared to the case where movement prediction is performed for all terminal stations 11, including terminal stations 11 located in other areas, the increase in computational load can be suppressed.

[0147] With this configuration, the switching instruction device 17b in the third embodiment of the present invention can predict future traffic volume without needing to predict the movement of all terminal stations 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to, for example, frequent movement of terminal stations 11. As a result, the switching instruction device 17b can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0148] <Fourth Embodiment> A fourth embodiment of the present invention will be described below.

[0149] The switching instruction device 17b in the third embodiment described above performed movement prediction only for terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas, thereby limiting movement prediction to only a portion of terminal stations 11. In contrast, the switching instruction device in the fourth embodiment described below (hereinafter referred to as "switching instruction device 17c") has a configuration that further limits the number of terminal stations 11 for which movement prediction is performed, thereby reducing the computational load. Specifically, the switching instruction device 17c in the fourth embodiment further narrows the target to only terminal stations 11 with high required quality (high priority) among the terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas, and performs movement prediction.

[0150] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the fourth embodiment (hereinafter referred to as "mobile communication system 10c") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10c is an example of the communication system 1 described above. The mobile communication system 10c in the fourth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0151] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17c will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17c is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17c and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0152] In the fourth embodiment, the switching instruction device 17c identifies the required quality (priority) for each terminal station 11 located within the coverage area of ​​the distributed station 13 to be predicted, and in the vicinity of said coverage area.

[0153] The required quality specified here includes, for example, the delay requirement level and the priority level of QoS (Quality of Service). The priority level of QoS, as used here, refers to, for example, the 5QI (5G network Quality of service class Identifier) ​​value. However, the required quality (priority) used is not limited to this information; any information can be used as long as a priority level can be set.

[0154] The switching instruction device 17c targets terminal stations 11 whose specified required quality (priority) is higher than a predetermined required quality for movement prediction. On the other hand, the switching instruction device 17c does not perform movement prediction for terminal stations 11 whose specified required quality is lower than or equal to a predetermined required quality, and considers their current position to be their future position (i.e., terminal stations 11 with low required quality are considered not to move).

[0155] The switching instruction device 17c predicts the movement of each terminal station 11 that is the target of the movement prediction, based on the amount of change in their respective positions. Based on the movement prediction, the switching instruction device 17c predicts the future number of connected terminals within the coverage area of ​​the target distributed station 13, and predicts the future traffic volume at each distributed station 13 based on the traffic volume at each distributed station 13 and the predicted number of connected terminals. Then, based on the future traffic volume at each distributed station 13, the switching instruction device 17c predicts the future traffic volume at each aggregation station 14.

[0156] Figure 12 is a block diagram showing the functional configuration of the switching instruction device 17c in a fourth embodiment of the present invention. As shown in Figure 12, the switching instruction device 17c is configured to include a prediction unit 172c and a switching determination unit 175. The prediction unit 172c is configured to include a future position estimation unit 173c and a traffic prediction unit 174c.

[0157] The prediction unit 172c receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. The radio control information in the fourth embodiment includes the most recent traffic volume value at each distributed station 13, location information of each terminal station 11 within the coverage area of ​​the distributed station 13, and information indicating the required quality (priority) for each terminal station 11 (hereinafter referred to as "required quality information").

[0158] The future position estimation unit 173c of the prediction unit 172c uses the position information for each terminal station 11 included in the radio control information to predict the movement of each terminal station 11 and estimate the number of connected terminals for each distributed station 13 in the future. Specifically, the future position estimation unit 173c identifies the position of each terminal station 11 located within the coverage area of ​​the distributed station 13 and in the surrounding area, based on the position information for each terminal station 11 included in the radio control information.

[0159] The future location estimation unit 173c performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area, for each distributed station 13. The future location estimation unit 173c also performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area at predetermined unit intervals.

[0160] Furthermore, the future position estimation unit 173c identifies the required quality (priority) for each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding areas, based on the required quality information included in the wireless control information. The future position estimation unit 173c targets terminal stations 11 whose identified required quality is higher than a predetermined required quality for movement prediction. On the other hand, the switching instruction device 17c does not perform movement prediction for terminal stations 11 whose identified required quality is below a predetermined required quality, and considers their current position to be their future position.

[0161] The future location estimation unit 173c then calculates the amount of change in the location of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11) for only the terminal stations 11 that have been narrowed down according to the required quality (priority). Based on the calculated change amounts, the future location estimation unit 173c estimates the future location of each terminal station 11. For terminal stations 11 with high required quality, the future location estimation unit 173c outputs location information indicating the estimated future location of each terminal station 11, and for terminal stations 11 with low required quality, it outputs location information indicating the current location (as location information indicating the future location) to the traffic prediction unit 174c.

[0162] The future position of each terminal station 11 may be estimated, for example, by using the fluctuation amount in the most recent unit period as the predicted value of the future fluctuation amount, or by using the average value of the fluctuation amounts over multiple past unit periods as the predicted value of the future fluctuation amount.

[0163] The traffic prediction unit 174c of the prediction unit 172c obtains location information indicating the future location of each terminal station 11 from the future location estimation unit 173c. Based on the obtained future locations of each terminal station 11, the traffic prediction unit 174c calculates the future number of connected terminals for each distributed station 13.

[0164] The traffic prediction unit 174c predicts the future traffic volume at each distributed station 13 based on the calculated future number of connected terminals at each distributed station 13. For example, the traffic prediction unit 174c predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0165] Furthermore, the prediction unit 172c calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172c outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0166] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172c and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0167] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0168] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17c will be described below. Figure 13 is a flowchart showing the operation of the switching instruction device 17c in the fourth embodiment of the present invention. The operation of the switching instruction device 17c shown in the flowchart of Figure 13 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0169] First, the prediction unit 172c obtains radio control information from each distributed station 13, which includes the most recent traffic volume value at the distributed station 13, the location information of each terminal station 11 within the coverage area of ​​the distributed station 13, and the required quality information for each terminal station 11 (step S301). Next, the future location estimation unit 173c of the prediction unit 172c identifies the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding areas, based on the location information for each terminal station 11 included in the radio control information (step S302).

[0170] Next, the future position estimation unit 173c identifies the required quality (priority) for each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding areas, based on the required quality information included in the radio control information (step S303).

[0171] Next, the future position estimation unit 173c determines whether it is necessary to estimate the future position of each terminal station 11 based on the specified required quality (step S304). Specifically, the future position estimation unit 173c determines that terminal stations 11 whose specified required quality is higher than a predetermined required quality are targets for movement prediction. On the other hand, the switching instruction device 17c does not perform movement prediction for terminal stations 11 whose specified required quality is below a predetermined required quality, and considers the current position to be the future position.

[0172] Next, the future location estimation unit 173c calculates the amount of change in the location of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11) for only the terminal stations 11 that have been narrowed down according to the required quality (priority) (step S305). Next, the future location estimation unit 173c estimates the future location of each terminal station 11 based on the calculated amount of change (step S306). For terminal stations 11 with high required quality, the future location estimation unit 173c outputs location information indicating the estimated future location of each terminal station 11, and for terminal stations 11 with low required quality, it outputs location information indicating the current location (as location information indicating the future location) to the traffic prediction unit 174c.

[0173] Next, the traffic prediction unit 174c of the prediction unit 172c calculates the future number of connected terminals at each distributed station 13 based on the future location of each terminal station 11 obtained (step S307).

[0174] Next, the traffic prediction unit 174c predicts the future traffic volume at each distributed station 13 based on the calculated future number of connected terminals at each distributed station 13 (step S308). For example, the traffic prediction unit 174c predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the future number of connected terminals to the most recent number of connected terminals.

[0175] Next, the traffic prediction unit 174c calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13 (step S309).

[0176] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172c and a predetermined switching threshold (step S310). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S311, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S312).

[0177] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregation station 14 to which the distributed station 13 is connected (step S313). Specifically, the switching determination unit 175 instructs the aggregation station 14 from which the switching is initiated to release the connection with the distributed station 13, and instructs the aggregation station 14 to which the switching is initiated to establish a connection with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is performed via the communication path after the connection destination has been switched.

[0178] This completes the operation of the switching instruction device 17c as shown in the flowchart of Figure 13.

[0179] As described above, in the fourth embodiment of the present invention, the switching instruction device 17c estimates the future location of each terminal station 11 based on the location information of each terminal station 11 located within the coverage area of ​​each distributed station 13 and in the surrounding area. At this time, the switching instruction device 17c estimates the future location only for terminal stations 11 with high required quality (priority), and omits the future location estimation process for terminal stations 11 with low required quality, considering their current location as their future location. Based on the estimated future location of each terminal station 11, the switching instruction device 17c estimates the future number of connected terminals at each distributed station 13. Based on the recent traffic volume at each distributed station 13, the recent number of connected terminals, and the estimated future number of connected terminals, the switching instruction device 17c predicts the future traffic volume at each distributed station 13.

[0180] The switching instruction device 17c then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If the switching instruction device 17c determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0181] Furthermore, in this fourth embodiment, the switching instruction device 17c focuses on predicting movement only for terminal stations 11 with high required quality (priority), thus reducing the increase in computational load compared to the case where movement prediction is performed for all terminal stations 11.

[0182] With this configuration, the switching instruction device 17c in the fourth embodiment of the present invention can predict future traffic volume without needing to predict the movement of all terminal stations 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to, for example, frequent movement of terminal stations 11. As a result, the switching instruction device 17c can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0183] <Fifth Embodiment> A fifth embodiment of the present invention will be described below.

[0184] The switching instruction device 17b in the third embodiment described above performed movement prediction only for terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas, thereby limiting movement prediction to only a portion of terminal stations 11. In contrast, the switching instruction device in the fifth embodiment described below (hereinafter referred to as "switching instruction device 17d") has a configuration that further limits the number of terminal stations 11 for which movement prediction is performed, thereby reducing the computational load. Specifically, the switching instruction device 17d in the fifth embodiment further narrows the target to only terminal stations 11 with a large recent change in position (amount of movement) among the terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas, and performs movement prediction.

[0185] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the fifth embodiment (hereinafter referred to as "mobile communication system 10d") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10d is an example of the communication system 1 described above. The mobile communication system 10d in the fifth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0186] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17d will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17d is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and controls the execution of route switching as necessary. However, the switching instruction device 17d and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0187] In the fifth embodiment, the switching instruction device 17d calculates the amount of positional variation in a direct return for each terminal station 11 located within the coverage area of ​​the distributed station 13 to be predicted, and in the vicinity of said coverage area. The switching instruction device 17d targets terminal stations 11 for which the calculated variation amount is higher than a predetermined variation amount for movement prediction. On the other hand, the switching instruction device 17d does not perform movement prediction for terminal stations 11 for which the identified variation amount is less than or equal to the predetermined variation amount, and considers their current position to be their future position (i.e., terminal stations 11 with little movement are considered not to move).

[0188] The switching instruction device 17d predicts the movement of each terminal station 11 that is the target of the movement prediction, based on the amount of change in their respective positions. Based on the movement prediction, the switching instruction device 17d predicts the future number of connected terminals within the coverage area of ​​the distribution station 13 that is the target of the prediction, and predicts the future traffic volume at each distribution station 13 based on the traffic volume at each distribution station 13 and the predicted number of connected terminals. Then, based on the future traffic volume at each distribution station 13, the switching instruction device 17d predicts the future traffic volume at each aggregation station 14.

[0189] Figure 14 is a block diagram showing the functional configuration of the switching instruction device 17d in a fifth embodiment of the present invention. As shown in Figure 14, the switching instruction device 17d is configured to include a prediction unit 172d and a switching determination unit 175. The prediction unit 172d is configured to include a future position estimation unit 173d and a traffic prediction unit 174d.

[0190] The prediction unit 172d receives radio control information from each distributed station 13, which is used to predict future traffic volume at each distributed station 13. In the fifth embodiment, the radio control information includes the most recent traffic volume value at each distributed station 13 and the location information of each terminal station 11 within the coverage area of ​​the distributed station 13.

[0191] The future position estimation unit 173d of the prediction unit 172d uses the position information for each terminal station 11 included in the radio control information to predict the movement of each terminal station 11 and estimate the number of connected terminals for each distributed station 13 in the future. Specifically, the future position estimation unit 173d identifies the position of each terminal station 11 located within the coverage area of ​​the distributed station 13 and in the surrounding area, based on the position information for each terminal station 11 included in the radio control information.

[0192] The future location estimation unit 173d performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area, for each distributed station 13. The future location estimation unit 173d also performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area at predetermined unit intervals.

[0193] The future position estimation unit 173d calculates the amount of change in the position of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11) based on the position information of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding areas. The future position estimation unit 173d then determines that terminal stations 11 whose calculated change amount is greater than a predetermined change amount are subject to movement prediction. On the other hand, the future position estimation unit 173d does not perform movement prediction for terminal stations 11 whose calculated change amount is less than or equal to the predetermined change amount, and considers their current position to be their future position.

[0194] The future position estimation unit 173d estimates the future position of each terminal station 11 that has been targeted for movement prediction, based on the recent change in the position of the terminal station 11 calculated above. For terminal stations 11 for which movement prediction has been performed, the future position estimation unit 173d outputs position information indicating the estimated future position of each terminal station 11, and for terminal stations 11 for which movement prediction has not been performed, it outputs position information indicating the current position (as position information indicating the future position) to the traffic prediction unit 174d.

[0195] The future position of each terminal station 11 may be estimated, for example, by using the fluctuation amount in the most recent unit period as the predicted value of the future fluctuation amount, or by using the average value of the fluctuation amounts over multiple past unit periods as the predicted value of the future fluctuation amount.

[0196] The traffic prediction unit 174d of the prediction unit 172d obtains location information indicating the future location of each terminal station 11 from the future location estimation unit 173d. Based on the obtained future locations of each terminal station 11, the traffic prediction unit 174d calculates the future number of connected terminals for each distributed station 13.

[0197] The traffic prediction unit 174d predicts the future traffic volume at each distributed station 13 based on the calculated future number of connected terminals at each distributed station 13. For example, the traffic prediction unit 174d predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0198] Furthermore, the prediction unit 172d calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172d outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0199] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172d and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0200] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0201] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17d will be described below. Figure 15 is a flowchart showing the operation of the switching instruction device 17d in the fifth embodiment of the present invention. The operation of the switching instruction device 17d shown in the flowchart of Figure 15 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0202] First, the prediction unit 172d obtains radio control information from each distributed station 13, which includes the most recent traffic volume value at the distributed station 13 and the location information of each terminal station 11 within the coverage area of ​​the distributed station 13 (step S401). Next, the future location estimation unit 173d of the prediction unit 172d identifies the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding areas, based on the location information of each terminal station 11 included in the radio control information (step S402).

[0203] Next, the future position estimation unit 173d calculates the amount of change in the position of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11) based on the position information of each terminal station 11 located within the coverage area of ​​the distributed station 13 and the surrounding area. (Step S403).

[0204] Next, the future position estimation unit 173d determines whether it is necessary to estimate the future position of each terminal station 11 based on the calculated amount of change in the position of the terminal station 11 (step S404). Specifically, the future position estimation unit 173d determines that terminal stations 11 whose calculated change is greater than a predetermined change amount are subject to movement prediction. On the other hand, the future position estimation unit 173d does not perform movement prediction for terminal stations 11 whose calculated change is less than or equal to a predetermined change amount, and considers their current position to be their future position.

[0205] Next, the future location estimation unit 173d estimates the future location of each terminal station 11 based on the calculated fluctuation amount, focusing only on the terminal stations 11 that have been narrowed down according to the fluctuation amount of their most recent location (step S405). The future location estimation unit 173d outputs location information indicating the estimated future location of each terminal station 11 for terminal stations 11 with a large fluctuation amount of their most recent location, and location information indicating the current location for terminal stations 11 with a small fluctuation amount of their most recent location (as location information indicating the future location) to the traffic prediction unit 174d.

[0206] Next, the traffic prediction unit 174d of the prediction unit 172d calculates the future number of connected terminals at each distributed station 13 based on the future location of each terminal station 11 obtained (step S406).

[0207] Next, the traffic prediction unit 174d predicts the future traffic volume at each distributed station 13 based on the calculated future number of connected terminals at each distributed station 13 (step S407). For example, the traffic prediction unit 174d predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0208] Next, the traffic prediction unit 174d calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13 (step S408).

[0209] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172d and a predetermined switching threshold (step S409). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S410, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S411).

[0210] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregate station 14 to which the distributed station 13 is connected (step S412). Specifically, the switching determination unit 175 instructs the source aggregate station 14 to release the connection with the distributed station 13, and instructs the destination aggregate station 14 to connect with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is carried out via the communication path after the connection destination switch.

[0211] This completes the operation of the switching instruction device 17d as shown in the flowchart of Figure 15.

[0212] As described above, in the fifth embodiment of the present invention, the switching instruction device 17d estimates the future location of each terminal station 11 based on the location information of each terminal station 11 located within the coverage area of ​​each distributed station 13 and in the surrounding area. At this time, the switching instruction device 17d estimates the future location only for terminal stations 11 with a large recent change in location (recent movement), and omits the future location estimation process for terminal stations 11 with a small recent change in location (recent movement), considering their current location as their future location. Based on the estimated future location of each terminal station 11, the switching instruction device 17d estimates the future number of connected terminals at each distributed station 13. Based on the recent traffic volume at each distributed station 13, the recent number of connected terminals, and the estimated future number of connected terminals, the switching instruction device 17d predicts the future traffic volume at each distributed station 13.

[0213] The switching instruction device 17d then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If the switching instruction device 17d determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform route switching to distribute the load to links with available bandwidth.

[0214] Furthermore, in this fifth embodiment, the switching instruction device 17d focuses on predicting movement only for terminal stations 11 with a large amount of recent positional change (recent movement), thus reducing the increase in computational load compared to the case where movement prediction is performed for all terminal stations 11.

[0215] With this configuration, the switching instruction device 17d in the fifth embodiment of the present invention can predict future traffic volume without needing to predict the movement of all terminal stations 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to, for example, frequent movement of terminal stations 11. As a result, the switching instruction device 17d can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0216] Furthermore, a configuration combining the configuration of the fifth embodiment described above and the configuration of the fourth embodiment described above may also be used. That is, in the fifth embodiment, the configuration narrows down the terminal stations 11 that are the target of movement prediction to only those terminal stations 11 with a large amount of recent movement, and in the fourth embodiment described above, the configuration narrows down the terminal stations 11 that are the target of movement prediction to only those terminal stations 11 with a high required quality (priority), but a configuration that performs both of these narrowings may also be used.

[0217] <Sixth Embodiment> A sixth embodiment of the present invention will be described below.

[0218] In the third embodiment described above, the switching instruction device 17b was configured to perform movement predictions only for terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas. In contrast, the switching instruction device in the sixth embodiment described below (hereinafter referred to as "switching instruction device 17e") has a configuration that further limits the number of terminal stations 11 for which movement predictions are performed to reduce the computational load. Specifically, the switching instruction device 17e in the sixth embodiment narrows down the predictions to only distributed stations 13 with a large number of recently connected terminals within the coverage area, or distributed stations 13 with a large amount of recently accessed traffic, to perform predictions for the number of connected terminals in the future.

[0219] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the sixth embodiment (hereinafter referred to as "mobile communication system 10e") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10e is an example of the communication system 1 described above. The mobile communication system 10e in the sixth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0220] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17e will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17e is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17e and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0221] In the sixth embodiment, the switching instruction device 17e determines for each distributed station 13 whether or not to perform movement prediction for each terminal station 11 based on the number of connected terminals at each distributed station 13. Specifically, the switching instruction device 17e determines to perform movement prediction for each terminal station 11 if the number of most recent connected terminals at each distributed station 13 exceeds a predetermined number of connected terminals. The predetermined number of connected terminals here is, for example, 80% of the number of connected terminals where congestion may occur.

[0222] On the other hand, if the number of connected terminals at each distributed station 13 is less than or equal to a predetermined number of connected terminals, the switching instruction device 17e determines for each distributed station 13 whether or not to perform a movement prediction for each terminal station 11 based on the amount of traffic at each distributed station 13. Specifically, if the amount of traffic at each distributed station 13 exceeds a predetermined amount, the switching instruction device 17e decides to perform a movement prediction for each terminal station 11. The predetermined amount of traffic here is, for example, the amount of traffic that may cause congestion.

[0223] Specifically, the switching instruction device 17e determines that it will not perform movement prediction for each terminal station 11 if the number of most recent connected terminals at each distributed station 13 is less than or equal to a predetermined number of connected terminals, and the amount of most recent traffic at each distributed station 13 is less than or equal to a predetermined amount of traffic. In other words, if the number of most recent connected terminals and the amount of traffic are less than or equal to predetermined values, the switching instruction device 17e predicts that even if there is an influx of traffic from a terminal station 11, it will not be enough to cause congestion, and therefore omits the process of predicting the movement of the terminal station 11.

[0224] When the switching instruction device 17e determines that it is necessary to predict the movement of each terminal station 11, it makes a movement prediction based on the amount of change in the position of each terminal station 11 located within and around the coverage area of ​​the distribution station 13 to be predicted. Then, the switching instruction device 17b predicts the future number of connected terminals within the coverage area of ​​the distribution station 13 to be predicted based on the movement prediction, and predicts the future traffic volume at each distribution station 13 based on the traffic volume of the distribution station 13 and the predicted number of connected terminals.

[0225] On the other hand, if the switching instruction device 17e determines that it does not need to perform movement predictions for each terminal station 11, it considers the number of most recent connected terminals at the distributed station 13 as the number of future connected terminals. Then, the switching instruction device 17e predicts the future traffic volume at each distributed station 13 based on the traffic volume and the number of most recent connected terminals at each distributed station 13. Then, the switching instruction device 17e predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13.

[0226] Figure 16 is a block diagram showing the functional configuration of the switching instruction device 17e in the sixth embodiment of the present invention. As shown in Figure 16, the switching instruction device 17e is configured to include a prediction execution decision unit 171, a prediction unit 172e, and a switching decision unit 175. The prediction unit 172e is configured to include a future position estimation unit 173e and a traffic prediction unit 174e.

[0227] The prediction unit 172e receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. The radio control information in the sixth embodiment includes the value of the most recent traffic volume at the distributed station 13, the value of the most recent number of connected terminals at the distributed station 13, and the location information of each terminal station 11 within the coverage area of ​​the distributed station 13.

[0228] The prediction execution decision unit 171 of the prediction unit 172b determines for each distributed station 13 whether or not to perform movement prediction for each terminal station 11, based on the number of most recent connected terminals at each distributed station 13. The prediction execution decision unit 171 determines to perform movement prediction for each terminal station 11 for distributed stations 13 where the number of most recent connected terminals exceeds a predetermined number of connected terminals.

[0229] On the other hand, for distributed stations 13 where the number of connected terminals is less than or equal to a predetermined number of connected terminals, the prediction execution decision unit 171 decides for each distributed station 13 whether or not to perform movement prediction for each terminal station 11 based on the most recent traffic volume at the distributed station 13. For distributed stations 13 where the most recent traffic volume exceeds a predetermined traffic volume, the prediction execution decision unit 171 decides to perform movement prediction for each terminal station 11.

[0230] The prediction execution decision unit 171 determines that it does not perform movement prediction for each terminal station 11 for distributed stations 13 where the number of connected terminals is less than or equal to a predetermined number of connected terminals and the most recent traffic volume is less than or equal to a predetermined traffic volume. The prediction execution decision unit 171 then outputs information to the prediction unit 172e indicating whether or not movement prediction for each terminal station 11 is necessary for each distributed station 13.

[0231] The future position estimation unit 173e of the prediction unit 172e determines that it is necessary to predict the movement of each terminal station 11 in the distributed station 13, and uses the location information of each terminal station 11 included in the radio control information to predict the movement of each terminal station 11 and estimate the future number of connected terminals for each distributed station 13. Specifically, the future position estimation unit 173e identifies the location of each terminal station 11 that exists within the coverage area of ​​the distributed station 13 and in the surrounding area, based on the location information of each terminal station 11 included in the radio control information.

[0232] The future position estimation unit 173e performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area, for each distributed station 13 that it has determined to perform movement prediction for each terminal station 11. The future position estimation unit 173b also performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area at predetermined unit intervals.

[0233] The future position estimation unit 173e then calculates the amount of change in the position of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11). Based on the calculated change amounts, the future position estimation unit 173e estimates the future position of each terminal station 11. The future position estimation unit 173e outputs position information indicating the estimated future position of each terminal station 11 to the traffic prediction unit 174e.

[0234] The future position of each terminal station 11 may be estimated, for example, by using the fluctuation amount in the most recent unit period as the predicted value of the future fluctuation amount, or by using the average value of the fluctuation amounts over multiple past unit periods as the predicted value of the future fluctuation amount.

[0235] The traffic prediction unit 174e of the prediction unit 172e obtains location information indicating the future location of each terminal station 11 from the future location estimation unit 173e. Based on the future location of each terminal station 11 estimated by the future location estimation unit 173e, the traffic prediction unit 174e calculates the future number of connected terminals for each distributed station 13 that it has decided to predict the movement of each terminal station 11.

[0236] On the other hand, for each distributed station 13 where the traffic prediction unit 174e has determined that it does not perform movement prediction for each terminal station 11, it considers the most recent number of connected terminals as the future number of connected terminals.

[0237] The traffic prediction unit 174e predicts the future traffic volume at each distributed station 13 based on the future number of connected terminals at each distributed station 13. For example, the traffic prediction unit 174e predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0238] Furthermore, the prediction unit 172e calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172e outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0239] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172e and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0240] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0241] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17e will be described below. Figure 17 is a flowchart showing the operation of the switching instruction device 17e in the sixth embodiment of the present invention. The operation of the switching instruction device 17e shown in the flowchart of Figure 17 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0242] First, the prediction unit 172e obtains radio control information from each distributed station 13, including the most recent traffic volume value at the distributed station 13, the most recent number of connected terminals at the distributed station 13, and the location information of each terminal station 11 within the coverage area of ​​the distributed station 13 (step S501). Next, the prediction execution decision unit 171 of the prediction unit 172e determines for each distributed station 13 whether or not to perform movement prediction for each terminal station 11, based on the most recent number of connected terminals at each distributed station 13 and a predetermined threshold (step S502).

[0243] Next, the prediction execution decision unit 171 determines whether to perform movement prediction for each terminal station 11 if the number of recently connected terminals exceeds a predetermined threshold (step S502, YES) (proceed to step S504). On the other hand, if the number of recently connected terminals is below a predetermined threshold (step S502, NO), the prediction execution decision unit 171 determines for each distributed station 13 whether to perform movement prediction for each terminal station 11 based on the recent traffic volume at the distributed station 13 and the predetermined threshold (step S503).

[0244] Next, the prediction execution decision unit 171 determines that for distributed stations 13 whose recent traffic volume exceeds a predetermined threshold (step S503, YES), it will perform movement prediction for each terminal station 11 (proceed to step S504). On the other hand, the prediction execution decision unit 171 determines that for distributed stations 13 whose recent traffic volume is below a predetermined threshold (step S503, NO), it will not perform movement prediction for each terminal station 11 (proceed to step S508). The prediction execution decision unit 171 outputs information to the prediction unit 172e indicating whether or not movement prediction for each terminal station 11 is necessary for each distributed station 13.

[0245] Next, the future position estimation unit 173e of the prediction unit 172e determines that it is necessary to predict the movement of each terminal station 11, and identifies the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area based on the location information of each terminal station 11 included in the radio control information (step S504).

[0246] The future position estimation unit 173e performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area, for each distributed station 13 that it has determined to perform movement prediction for each terminal station 11. The future position estimation unit 173b also performs the above process to identify the location of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area at predetermined unit intervals.

[0247] Next, the future position estimation unit 173e calculates the amount of change in the position of each terminal station 11 (for example, the amount of change in the distance traveled and the direction of travel for each terminal station 11) based on the identified position (step S505). Next, the future position estimation unit 173e estimates the future position of each terminal station 11 based on the calculated change (step S506). The future position estimation unit 173e outputs position information indicating the estimated future position of each terminal station 11 to the traffic prediction unit 174e.

[0248] Next, based on the future location of each terminal station 11 estimated by the future location estimation unit 173e, the number of future connected terminals for each distributed station 13 that has decided to perform movement prediction for each terminal station 11 is calculated (step S507). On the other hand, for each distributed station 13 that has decided not to perform movement prediction for each terminal station 11, the traffic prediction unit 174e considers the number of most recent connected terminals based on radio control information as the number of future connected terminals.

[0249] Next, the traffic prediction unit 174e calculates the future traffic volume at each distributed station 13 based on the future number of connected terminals at each distributed station 13 (step S508). For example, the traffic prediction unit 174e predicts the future traffic volume by multiplying the value of the most recent traffic volume at each distributed station 13 by the ratio of the number of future connected terminals to the number of recent connected terminals.

[0250] Next, the traffic prediction unit 174e calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13 (step S509).

[0251] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172e and a predetermined switching threshold (step S510). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S511, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S512).

[0252] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregation station 14 to which the distributed station 13 is connected (step S513). Specifically, the switching determination unit 175 instructs the aggregation station 14 from which the switching is initiated to release the connection with the distributed station 13, and instructs the aggregation station 14 to which the switching is initiated to establish a connection with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is performed via the communication path after the connection destination has been switched.

[0253] This completes the operation of the switching instruction device 17e as shown in the flowchart of Figure 17.

[0254] As described above, in the sixth embodiment of the present invention, the switching instruction device 17e estimates the future location of each terminal station 11 based on the location information of each terminal station 11 located within the coverage area of ​​the distributed station 13 and its surrounding area, for distributed stations 13 that satisfy at least one of the following conditions: the number of most recent connected terminals exceeds a predetermined number of connected terminals, and the amount of recent traffic exceeds a predetermined amount of traffic. Based on the estimated future location of each terminal station 11, the switching instruction device 17e estimates the future number of connected terminals for each distributed station 13. On the other hand, for distributed stations 13 where the number of most recent connected terminals is less than or equal to a predetermined number of connected terminals, and the amount of recent traffic is less than or equal to a predetermined amount of traffic, the switching instruction device 17e decides not to perform the above-mentioned movement prediction for each terminal station 11 (predicting that congestion will not occur even if there is an influx of terminal stations 11), and considers the number of most recent connected terminals in the distributed station 13 to be the number of future connected terminals.

[0255] The switching instruction device 17e then predicts the future traffic volume for each distributed station 13 based on the most recent traffic volume, the most recent number of connected terminals, and the future number of connected terminals at each distributed station 13. Based on the future traffic volume at each distributed station 13, the switching instruction device 17e then predicts the future traffic volume at each aggregation station 14. If the switching instruction device 17e determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it performs route switching to distribute the load to links with available bandwidth.

[0256] In this sixth embodiment, the switching instruction device 17e performs movement prediction only for terminal stations 11 located within the coverage area of ​​each distributed station 13 and its surrounding areas, provided that at least one of the following conditions is met: the number of most recent connected terminals exceeds a predetermined number of connected terminals, and the amount of most recent traffic exceeds a predetermined amount of traffic.

[0257] In other words, the switching instruction device 17e in the sixth embodiment estimates the future number of connected terminals at a distributed station 13 by performing movement predictions for each terminal station 11 for distributed stations 13 where congestion is expected to occur, but omits such movement predictions for each terminal station 11 for distributed stations 13 where congestion is expected to occur less likely. Therefore, the switching instruction device 17e in the sixth embodiment can suppress the increase in computational load compared to the case where movement predictions are performed for all terminal stations 11.

[0258] With this configuration, the switching instruction device 17e in the sixth embodiment of the present invention can predict future traffic volume without needing to predict the movement of all terminal stations 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to, for example, frequent movement of terminal stations 11. As a result, the switching instruction device 17e can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0259] Furthermore, the configuration of the sixth embodiment described above may be combined with some of the configurations of the first to fifth embodiments mentioned above. For example, the configuration of the sixth embodiment may be combined with at least one of the configurations of the fourth and fifth embodiments mentioned above.

[0260] In other words, in the sixth embodiment, the configuration narrowed down the target terminal stations 11 for movement prediction to only those terminal stations 11 that communicate with distributed stations 13 that satisfy predetermined conditions (i.e., terminal stations 11 that communicate with distributed stations 13 that satisfy at least one of the following conditions: the number of recently connected terminals exceeds a predetermined number of connected terminals, and the amount of recent traffic exceeds a predetermined amount of traffic). In the fourth embodiment described above, the configuration narrowed down the target terminal stations 11 for movement prediction to only those terminal stations 11 with high required quality (priority). In the fifth embodiment described above, the configuration narrowed down the target terminal stations 11 for movement prediction to only those terminal stations 11 with a large amount of recent movement. For example, all of these configurations may be combined to perform three types of narrowing, or any of these configurations may be combined to perform two types of narrowing.

[0261] The switching instruction devices 17, 17a to 17e in the first to sixth embodiments described above were configured to predict changes in traffic volume caused by fluctuations in the number of connected terminals of the distributed station 13, mainly due to the movement of terminal stations 11, and to predict the occurrence of congestion. However, the factors that change the traffic volume are not necessarily limited to fluctuations in the number of connected terminals of the distributed station 13.

[0262] For example, even within a single communication area covered by a distributed station 13, there may be terminal stations 11 with high signal strength and terminal stations 11 with low signal strength due to differences in the wireless environment. Generally, terminal stations 11 with low signal strength will have relatively less traffic. Also, for example, terminal stations 11 where handover is performed will temporarily experience less traffic during the handover. Furthermore, for example, if a failure occurs in any of the devices along the communication path, such as terminal station 11, base stations such as distributed stations 13 or aggregation stations 14, or transmission devices 15, the amount of traffic will decrease even if the wireless environment is otherwise good.

[0263] The switching instruction devices 17f to 17k in the seventh to ninth embodiments and the modified versions of each embodiment described below predict changes in traffic volume caused by factors other than such fluctuations in the number of connected terminals. The switching instruction devices 17f to 17k then determine whether congestion is occurring based on the prediction results and have a configuration to execute route switching if congestion is determined to be occurring.

[0264] <Seventh Embodiment> The seventh embodiment of the present invention will be described below.

[0265] The switching instruction device (hereinafter referred to as "switching instruction device 17f") in the seventh embodiment described below acquires information indicating the radio wave strength for each terminal station 11 (hereinafter referred to as "radio wave strength information") from the distributed station 13, and for each terminal station 11, it determines whether the terminal station 11 has a relatively high radio wave strength or a relatively low radio wave strength. The switching instruction device 17f then has a configuration that predicts the future traffic volume, taking into account that the throughput of the terminal station 11 with a relatively low radio wave strength will be relatively low.

[0266] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the seventh embodiment (hereinafter referred to as "mobile communication system 10f") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10f is an example of the communication system 1 described above. The mobile communication system 10f in the seventh embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0267] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17f will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17f is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17f and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0268] In the seventh embodiment, the switching instruction device 17f acquires radio control information from the distributed station 13, including radio wave strength information for each terminal station 11 located within the coverage area of ​​the distributed station 13 to be predicted, and determines for each terminal station 11 whether it has a relatively high radio wave strength or a relatively low radio wave strength. More specifically, the switching instruction device 17f distinguishes each terminal station 11 based on whether or not the radio wave strength is sufficient to meet the throughput required in advance for the communication service realized by the mobile communication system 10f (hereinafter referred to as "required throughput").

[0269] The radio wave strength referred to here may be the received signal strength when the signal transmitted from the terminal station 11 is received at the distributed station 13, or it may be the received signal strength when the signal transmitted from the distributed station 13 is received at the terminal station 11.

[0270] The switching instruction device 17f then predicts the future traffic volume at each distributed station 13 by predicting that terminal stations 11 with relatively high signal strength will have correspondingly high traffic volumes, and terminal stations 11 with relatively low signal strength will have correspondingly low traffic volumes. Based on the future traffic volume at each distributed station 13, the switching instruction device 17f then predicts the future traffic volume at each aggregation station 14.

[0271] Figure 18 is a block diagram showing the functional configuration of the switching instruction device 17f in the seventh embodiment of the present invention. As shown in Figure 18, the switching instruction device 17f is configured to include a prediction unit 172f and a switching determination unit 175. The prediction unit 172f is configured to include a deterioration condition detection unit 176f and a traffic prediction unit 174f.

[0272] The prediction unit 172f receives radio control information from each distributed station 13 that is used to predict the future traffic volume at each distributed station 13. The radio control information in the seventh embodiment includes the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13, radio wave strength information for each terminal station 11 within the coverage area of ​​the distributed station 13, and a value of the requested throughput that has been requested in advance from the distributed station 13.

[0273] The radio wave strength information referred to here is, for example, the value of RSRP (Reference Signal Received Power). Alternatively, information indicating communication quality, such as the values ​​of RSRQ (Reference Signal Received Quality) and SINR (Signal-to-Interference-plus-Noise Ratio), may be used instead of radio wave strength information.

[0274] The degradation condition detection unit 176f of the prediction unit 172f sorts each terminal station 11 based on the radio wave strength information for each terminal station 11 included in the wireless control information and the value of the requested throughput. More specifically, the degradation condition detection unit 176f uses the value of the requested throughput as a threshold and counts the number of terminal stations 11 whose radio wave strength is lower than the threshold. The degradation condition detection unit 176f identifies the number of terminal stations 11 that satisfy the requested throughput (i.e., terminal stations 11 whose radio wave strength is equal to or greater than the threshold) by subtracting the number of terminal stations 11 whose radio wave strength is lower than the threshold from the number of connected terminals of terminal stations 11 within the coverage area of ​​the distributed station 13.

[0275] In this embodiment, the degradation condition detection unit 176f is configured to identify the number of terminal stations 11 that satisfy the required throughput (i.e., terminal stations 11 whose signal strength is above the threshold) by counting the number of terminal stations 11 whose signal strength is below the threshold, and then subtracting the counted value from the total number of connected terminals (determination) of the distributed station 13. However, the reverse calculation method may also be used. That is, the degradation condition detection unit 176f may be configured to identify the number of terminal stations 11 whose signal strength is below the threshold by counting the number of terminal stations 11 that satisfy the required throughput (i.e., terminal stations 11 whose signal strength is above the threshold), and then subtracting the counted value from the total number of connected terminals (determination) of the distributed station 13.

[0276] The degradation condition detection unit 176f then outputs to the traffic prediction unit 174f information indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is lower than the threshold. The degradation condition detection unit 176f performs the above process to identify the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is lower than the threshold for each distributed station 13.

[0277] The traffic prediction unit 174f of the prediction unit 172f obtains information from the degradation condition detection unit 176f indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is below a threshold. The traffic prediction unit 174f predicts the future traffic volume at each distributed station 13 by multiplying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is below a threshold by a predetermined throughput value and summing them up.

[0278] For example, the traffic prediction unit 174f multiplies the number of terminal stations 11 that meet the required throughput for each distributed station 13 by the value of the required throughput, and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by one-hundredth of the value of the required throughput. Then, the traffic prediction unit 174f predicts the future traffic volume at each distributed station 13 by summing the two calculated values.

[0279] Furthermore, the prediction unit 172f calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172f outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0280] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172f and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0281] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0282] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17f will be described below. Figure 19 is a flowchart showing the operation of the switching instruction device 17f in the seventh embodiment of the present invention. The operation of the switching instruction device 17f shown in the flowchart of Figure 19 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0283] First, the prediction unit 172f receives radio control information from each distributed station 13 that is used to predict the future traffic volume at each distributed station 13 (step S601). As described above, the radio control information in this embodiment includes, for example, the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13, radio wave strength information for each terminal station 11 within the coverage area of ​​the distributed station 13, and the value of the requested throughput that has been requested in advance from the distributed station 13.

[0284] Next, the degradation condition detection unit 176f of the prediction unit 172 counts the number of terminal stations 11 whose signal strength is lower than the threshold value, using the requested throughput value as the threshold, based on the signal strength information for each terminal station 11 included in the wireless control information and the value of the requested throughput (step S602). Next, the degradation condition detection unit 176f identifies the number of terminal stations 11 whose signal strength is lower than the threshold value by subtracting the number of terminal stations 11 whose signal strength is lower than the threshold value from the number of connected terminals of terminal stations 11 in the coverage area of ​​the distributed station 13 (step S603).

[0285] Next, the traffic prediction unit 174f of the prediction unit 172f obtains information from the degradation condition detection unit 176f indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is lower than the threshold. The traffic prediction unit 174f multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is lower than the threshold by a predetermined throughput value (step S604). As described above, for example, the traffic prediction unit 174f multiplies the number of terminal stations 11 that meet the required throughput by the required throughput value and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by one-hundredth of the required throughput value.

[0286] Next, the traffic prediction unit 174f predicts the future traffic volume at each distributed station 13 by summing the two values ​​calculated above (step S605). Next, the prediction unit 172f calculates the future traffic volume at each aggregated station 14 based on the connection relationship between the distributed stations 13 and the aggregated station 14 and the future traffic volume at each distributed station 13 (step S606).

[0287] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172f and a predetermined switching threshold (step S607). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S608, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S609).

[0288] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregation station 14 to which the distributed station 13 is connected (step S610). Specifically, the switching determination unit 175 instructs the aggregation station 14 from which the switching is initiated to release the connection with the distributed station 13, and instructs the aggregation station 14 to which the switching is initiated to establish a connection with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is performed via the communication path after the connection destination has been switched.

[0289] This completes the operation of the switching instruction device 17f as shown in the flowchart of Figure 19.

[0290] As described above, the switching instruction device 17f in the seventh embodiment of the present invention acquires radio wave strength information for each terminal station 11 from the distributed station 13 and determines for each terminal station 11 whether the terminal station 11 has a relatively high radio wave strength or a relatively low radio wave strength. The switching instruction device 17f then predicts the future traffic volume at each distributed station 13, taking into account that the throughput of the terminal station 11 with a relatively low radio wave strength will be relatively low.

[0291] The switching instruction device 17f then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If the switching instruction device 17f determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0292] Furthermore, in this manner, the switching instruction device 17f in the seventh embodiment predicts future traffic based on the strength of radio waves at terminal stations 11 within the coverage area of ​​each distributed station 13. Therefore, compared to conventional technology that predicts future traffic by predicting the movement of all terminal stations 11, it can suppress the increase in computational load.

[0293] With this configuration, the switching instruction device 17f in the seventh embodiment of the present invention can predict future traffic volume without needing to predict the movement of each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17f can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0294] (Variations of the seventh embodiment) Hereinafter, variations of the seventh embodiment of the present invention will be described.

[0295] In the seventh embodiment described above, the wireless control information received by the prediction unit 172f from each distributed station 13 included the number of connected terminals of the terminal station 11 within the coverage area of ​​the distributed station 13, the radio wave strength information for each terminal station 11 within the coverage area of ​​the distributed station 13, and the requested throughput value that had been requested in advance from the distributed station 13. In contrast, in the switching instruction device (hereinafter referred to as "switching instruction device 17g") in the modified version of the seventh embodiment described below, the wireless control information received from each distributed station 13 includes the radio wave strength information for each terminal station 11 within the coverage area of ​​the distributed station 13, and the requested throughput value that had been requested in advance from the distributed station 13.

[0296] In other words, in the modified version of the seventh embodiment, the wireless control information does not include information indicating the total number of connected terminals (the base number) of terminal stations 11 within the coverage area of ​​the distributed station 13. The modified version of the seventh embodiment assumes a case where information indicating the total number of connected terminals (the base number) cannot be received from each distributed station 13.

[0297] The switching instruction device g in the modified seventh embodiment described below, similar to the switching instruction device f in the seventh embodiment described above, acquires radio wave strength information for each terminal station 11 from the distributed station 13 and determines for each terminal station 11 whether the terminal station 11 has a relatively high radio wave strength or a relatively low radio wave strength. The switching instruction device 17g then has a configuration that predicts future traffic volume, taking into account that the throughput of terminal stations 11 with relatively low radio wave strength will be relatively low.

[0298] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the modified version of the seventh embodiment (hereinafter referred to as "mobile communication system 10g") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation is omitted. The mobile communication system 10g is an example of the communication system 1 described above. The mobile communication system 10g in the modified version of the seventh embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0299] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17g will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17g is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17g and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0300] In the modified version of the seventh embodiment, the switching instruction device 17g acquires radio wave strength information for each terminal station 11 located within the coverage area of ​​the distribution station 13 to be predicted, and for each terminal station 11, it distinguishes between terminal stations 11 with relatively high radio wave strength and terminal stations 11 with relatively low radio wave strength. More specifically, the switching instruction device 17g distinguishes each terminal station 11 based on whether or not the radio wave strength is sufficient to meet the required throughput pre-required for the communication service realized by the mobile communication system 10g.

[0301] The switching instruction device 17g then predicts the future traffic volume at each distributed station 13 by predicting that terminal stations 11 with relatively high signal strength will have a correspondingly high traffic volume, and terminal stations 11 with relatively low signal strength will have a correspondingly low traffic volume. Based on the future traffic volume at each distributed station 13, the switching instruction device 17g then predicts the future traffic volume at each aggregation station 14.

[0302] Figure 20 is a block diagram showing the functional configuration of a switching instruction device 17g in a modified example of the seventh embodiment of the present invention. As shown in Figure 20, the switching instruction device 17g is configured to include a prediction unit 172g and a switching determination unit 175. The prediction unit 172g is configured to include a deterioration condition detection unit 176g and a traffic prediction unit 174g.

[0303] The prediction unit 172g receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. In the modified version of the seventh embodiment, the radio control information includes radio wave strength information for each terminal station 11 within the coverage area of ​​the distributed station 13, and a value of the requested throughput that has been requested in advance from the distributed station 13.

[0304] The degradation condition detection unit 176g of the prediction unit 172g sorts each terminal station 11 based on the radio wave strength information for each terminal station 11 and the value of the required throughput included in the wireless control information. More specifically, the degradation condition detection unit 176g uses the value of the required throughput as a threshold and counts the number of terminal stations 11 whose radio wave strength is lower than the threshold and the number of terminal stations 11 that meet the required throughput (i.e., terminal stations 11 whose radio wave strength is equal to or greater than the threshold).

[0305] The future location estimation unit 173g then outputs information to the traffic prediction unit 174g indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is below the threshold. The future location estimation unit 173g performs the above process of identifying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is below the threshold for each distributed station 13.

[0306] The traffic prediction unit 174g of the prediction unit 172g obtains information from the degradation condition detection unit 176g indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is below a threshold. The traffic prediction unit 174g multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is below a threshold by a predetermined throughput value, and then sums the two calculated values ​​to predict the future traffic volume at each distributed station 13.

[0307] For example, the traffic prediction unit 174g multiplies the number of terminal stations 11 that meet the required throughput for each distributed station 13 by the value of the required throughput, and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by one-hundredth of the value of the required throughput. Then, the traffic prediction unit 174g predicts the future traffic volume at each distributed station 13 by summing the two calculated values.

[0308] Furthermore, the prediction unit 172g calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172g outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0309] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172g and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0310] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0311] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17g will be described below. Figure 21 is a flowchart showing the operation of the switching instruction device 17g in a modified example of the seventh embodiment of the present invention. The operation of the switching instruction device 17g shown in the flowchart of Figure 21 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0312] First, the prediction unit 172g receives radio control information from each distributed station 13 that is used to predict the future traffic volume at each distributed station 13 (step S701). As described above, the radio control information in this embodiment includes radio wave strength information for each terminal station 11 within the coverage area of ​​the distributed station 13 and the value of the requested throughput that has been requested in advance from the distributed station 13. As described above, the radio control information in this modified example does not include information indicating the total number of connected terminals (denominator) of terminal stations 11 within the coverage area of ​​the distributed station 13.

[0313] Next, the degradation condition detection unit 176g of the prediction unit 172 counts and identifies the number of terminal stations 11 whose signal strength is lower than the threshold and the number of terminal stations 11 that satisfy the required throughput (i.e., terminal stations 11 whose signal strength is equal to or greater than the threshold) based on the signal strength information for each terminal station 11 included in the wireless control information and the value of the required throughput, using the value of the required throughput as a threshold (step S702).

[0314] Next, the traffic prediction unit 174g of the prediction unit 172g obtains information from the degradation condition detection unit 176g indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is lower than the threshold. The traffic prediction unit 174g multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose signal strength is lower than the threshold by a predetermined throughput value (step S703). As described above, for example, the traffic prediction unit 174g multiplies the number of terminal stations 11 that meet the required throughput by the required throughput value and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by one-hundredth of the required throughput value.

[0315] Next, the traffic prediction unit 174g predicts the future traffic volume at each distributed station 13 by summing the calculated values ​​(step S704). Next, the prediction unit 172g calculates the future traffic volume at each aggregated station 14 based on the connection relationship between the distributed stations 13 and the aggregated station 14 and the future traffic volume at each distributed station 13 (step S705).

[0316] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172g and a predetermined switching threshold (step S706). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S707, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S708).

[0317] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregate station 14 to which the distributed station 13 is connected (step S709). Specifically, the switching determination unit 175 instructs the source aggregate station 14 to release the connection with the distributed station 13 and instructs the destination aggregate station 14 to connect with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is carried out via the communication path after the connection destination switch.

[0318] This completes the operation of the switching instruction device 17g as shown in the flowchart of Figure 21.

[0319] As described above, the switching instruction device 17g in the modified seventh embodiment of the present invention acquires radio wave strength information for each terminal station 11 from the distributed station 13 and determines for each terminal station 11 whether the terminal station 11 has a relatively high radio wave strength or a relatively low radio wave strength. The switching instruction device 17g then predicts the future traffic volume at each distributed station 13, taking into account that the throughput of the terminal station 11 with a relatively low radio wave strength will be relatively low.

[0320] The switching instruction device 17g then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If the switching instruction device 17g determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0321] Furthermore, in this modified version of the seventh embodiment, the switching instruction device 17g predicts future traffic based on the radio wave intensity at the terminal station 11 within the coverage area of ​​each distributed station 13. Therefore, compared to conventional technology that predicts future traffic by predicting the movement of all terminal stations 11, it can suppress the increase in computational load.

[0322] With this configuration, the switching instruction device 17g in the modified seventh embodiment of the present invention can predict future traffic volume without needing to predict the movement of each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17g can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0323] <Eighth Embodiment> The eighth embodiment of the present invention will be described below.

[0324] The switching instruction device in the eighth embodiment described below (hereinafter referred to as "switching instruction device 17h") acquires information indicating the number of terminal stations 11 that will be handed over in the most recent time (hereinafter referred to as "handover information"). Based on the difference between the number of terminal stations 11 that will be handed over in the most recent time and the number of terminal stations 11 based on past handover information acquired one period ago (i.e., the number of terminal stations 11 that were handed over one period ago), the switching instruction device 17h determines the increase or decrease in the number of terminal stations 11 that will be handed over.

[0325] Terminal stations 11 undergoing handover will experience a temporary decrease in throughput during the handover process. Therefore, if the number of terminal stations 11 undergoing handover increases, the switching instruction device 17h estimates the number of terminal stations 11 whose throughput will temporarily decrease based on that increase. The switching instruction device 17h then has a configuration that predicts future traffic volume, taking into account the temporary decrease in throughput due to the handover process.

[0326] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the eighth embodiment (hereinafter referred to as "mobile communication system 10h") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation is omitted. The mobile communication system 10h is an example of the communication system 1 described above. The mobile communication system 10h in the eighth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0327] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17h will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17h is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17h and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0328] In the eighth embodiment, the switching instruction device 17h acquires handover information from the predicted distributed station 13, compares the number of terminal stations 11 undergoing handover with the number from one period prior, and identifies the increase or decrease. Based on the increase or decrease in the number of terminal stations 11 undergoing handover, and the total number of terminal stations 11 within the coverage area of ​​the predicted distributed station 13 and their increase or decrease (i.e., the base number of terminal stations 11 and the increase or decrease in that base number), the switching instruction device 17h identifies the number of terminal stations 11 undergoing handover and the number of terminal stations 11 not undergoing handover.

[0329] The switching instruction device 17h then predicts the future traffic volume at each distributed station 13 by predicting that terminal stations 11 where handover does not occur will have a correspondingly high traffic volume, and terminal stations 11 where handover does occur will have a correspondingly low traffic volume. Based on the future traffic volume at each distributed station 13, the switching instruction device 17h then predicts the future traffic volume at each aggregate station 14.

[0330] Figure 22 is a block diagram showing the functional configuration of the switching instruction device 17h in the eighth embodiment of the present invention. As shown in Figure 22, the switching instruction device 17h is configured to include a prediction unit 172h and a switching determination unit 175. The prediction unit 172h is configured to include a deterioration condition detection unit 176h and a traffic prediction unit 174h.

[0331] The prediction unit 172h receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. In the eighth embodiment, the radio control information includes the number of connected terminals at terminal stations 11 within the coverage area of ​​the distributed station 13 and the handover information of the distributed station 13.

[0332] The deterioration condition detection unit 176h of the prediction unit 172h determines the increase or decrease in the number of terminal stations 11 to be handed over based on the difference between the handover information contained in the most recently acquired wireless control information (i.e., the number of terminal stations 11 to be handed over most recently) and the handover information contained in the wireless control information acquired one period ago (i.e., the number of terminal stations 11 to be handed over one period ago).

[0333] Furthermore, the degradation condition detection unit 176h identifies the increase or decrease in the total number of connected terminals in the terminal station 11 within the coverage area of ​​the distributed station 13 (i.e., the increase or decrease in the base number) based on the difference between the total number of connected terminals in the terminal station 11 within the coverage area of ​​the distributed station 13 (base number) included in the most recently acquired radio control information and the total number of connected terminals in the terminal station 11 within the coverage area of ​​the distributed station 13 (base number) included in the radio control information acquired one period ago.

[0334] The degradation condition detection unit 176h identifies the number of terminal stations 11 that will be handed over and the number of terminal stations 11 that will not be handed over, based on the increase or decrease in the number of terminal stations 11 that will be handed over and the total number of connected terminals of terminal stations 11 within the coverage area of ​​the distributed station 13 and the increase or decrease in the number of connected terminals.

[0335] Furthermore, the degradation condition detection unit 176h considers terminal stations 11 in which handover does not occur to be terminal stations 11 that can meet the required throughput pre-required for the communication service realized by the mobile communication system 10h, and terminal stations 11 in which handover occurs to be terminal stations 11 that cannot meet the above-mentioned required throughput.

[0336] The degradation condition detection unit 176h performs the above process for each distributed station 13 to identify the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 that cannot meet the required throughput due to handover. The degradation condition detection unit 176h then outputs information indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 that cannot meet the required throughput due to handover to the traffic prediction unit 174h.

[0337] The traffic prediction unit 174h of the prediction unit 172h obtains information from the degradation condition detection unit 176h indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 that cannot meet the required throughput due to handover. The traffic prediction unit 174h predicts the future traffic volume at each distributed station 13 by multiplying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 that cannot meet the required throughput due to handover by a predetermined throughput value and summing them up.

[0338] For example, the traffic prediction unit 174h multiplies the number of terminal stations 11 that meet the required throughput for each distributed station 13 by the value of the required throughput, and multiplies the number of terminal stations 11 that cannot meet the required throughput due to handover by a value of 1 / 100 of the value of the required throughput. Then, the traffic prediction unit 174h predicts the future traffic volume at each distributed station 13 by summing the two calculated values.

[0339] Furthermore, the prediction unit 172h calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172h outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0340] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172h and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0341] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0342] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17h will be described below. Figure 23 is a flowchart showing the operation of the switching instruction device 17h in the eighth embodiment of the present invention. The operation of the switching instruction device 17h shown in the flowchart of Figure 23 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction results are performed at predetermined intervals.

[0343] The prediction unit 172h receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. As described above, the radio control information in this embodiment includes the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13 and the handover information of the distributed station 13.

[0344] Next, the deterioration condition detection unit 176h of the prediction unit 172h determines the increase or decrease in the number of terminal stations 11 to be handed over based on the difference between the handover information contained in the most recently acquired wireless control information (i.e., the number of terminal stations 11 to be handed over most recently) and the handover information contained in the wireless control information acquired one period ago (i.e., the number of terminal stations 11 to be handed over one period ago) (step S802).

[0345] Next, the degradation condition detection unit 176h identifies the increase or decrease in the total number of connected terminals in the terminal station 11 within the coverage area of ​​the distributed station 13 (i.e., the increase or decrease in the number of terminal stations 11) based on the difference between the total number of connected terminals in the terminal station 11 within the coverage area of ​​the distributed station 13 (descriptor number) included in the most recently acquired radio control information and the total number of connected terminals in the terminal station 11 within the coverage area of ​​the distributed station 13 (descriptor number) included in the radio control information acquired one period ago (step S803).

[0346] Next, the degradation condition detection unit 176h identifies the number of terminal stations 11 that will be handed over and the number of terminal stations 11 that will not be handed over, based on the increase or decrease in the number of terminal stations 11 that will be handed over and the total number of connected terminals of terminal stations 11 within the coverage area of ​​the distributed station 13 and the increase or decrease in the number of connected terminals (step S804). The degradation condition detection unit 176h performs the above process to identify the number of terminal stations 11 that will satisfy the required throughput and the number of terminal stations 11 that will not be able to satisfy the required throughput due to handover, for each distributed station 13.

[0347] Next, the traffic prediction unit 174h of the prediction unit 172h obtains information from the degradation condition detection unit 176h indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 that cannot meet the required throughput due to handover. The traffic prediction unit 174h multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 that cannot meet the required throughput due to handover by a predetermined throughput value (step S805). As described above, for example, the traffic prediction unit 174h multiplies the number of terminal stations 11 that meet the required throughput by the required throughput value and multiplies the number of terminal stations 11 that cannot meet the required throughput due to handover by one-hundredth of the required throughput value.

[0348] Next, the traffic prediction unit 174h predicts the future traffic volume at each distributed station 13 by summing the calculated values ​​(step S806). Next, the prediction unit 172h calculates the future traffic volume at each aggregated station 14 based on the connection relationship between the distributed stations 13 and the aggregated station 14 and the future traffic volume at each distributed station 13 (step S807).

[0349] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172h and a predetermined switching threshold (step S808). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S809, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S810).

[0350] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregate station 14 to which the distributed station 13 is connected (step S811). Specifically, the switching determination unit 175 instructs the source aggregate station 14 to release the connection with the distributed station 13 and instructs the destination aggregate station 14 to connect with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is carried out via the communication path after the connection destination switch.

[0351] This completes the operation of the switching instruction device 17g as shown in the flowchart of Figure 23.

[0352] As described above, the switching instruction device 17h in the eighth embodiment of the present invention acquires handover information from the distributed station 13 indicating the number of terminal stations 11 to be handed over, and identifies the increase or decrease in the number of terminal stations 11 to be handed over based on the difference between the number of terminal stations 11 to be handed over most recently and the number of terminal stations 11 to be handed over one period ago. If the number of terminal stations 11 to be handed over increases, the switching instruction device 17h estimates the number of terminal stations 11 to which throughput will temporarily decrease based on that increase. Then, the switching instruction device 17h predicts the future traffic volume at each distributed station 13, taking into account the temporary decrease in throughput due to the execution of the handover.

[0353] The switching instruction device 17h then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distribution station 13. If the switching instruction device 17h determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0354] Furthermore, in this eighth embodiment, the switching instruction device 17h predicts future traffic based on the handover information of each distributed station 13. Therefore, compared to conventional technology that predicts future traffic by predicting movement for all terminal stations 11, it can suppress the increase in computational load.

[0355] With this configuration, the switching instruction device 17h in the eighth embodiment of the present invention can predict future traffic volume without needing to predict the movement of terminal stations 11, even in situations where the number of connected terminals in the coverage area of ​​distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17h can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0356] <Ninth Embodiment> The ninth embodiment of the present invention will be described below.

[0357] The switching instruction device in the ninth embodiment described below (hereinafter referred to as "switching instruction device 17i") acquires information indicating the occurrence of a fault (hereinafter referred to as "fault information") from the distributed station 13. A fault, in this context, is a fault that can cause a decrease in communication quality in, for example, base station equipment such as the terminal station 11, distributed station 13, and aggregation station 14, and equipment installed on the communication path such as the forwarding device 15. In the following description, a decrease in communication quality is, as an example, a fault that causes a decrease in throughput.

[0358] The switching instruction device 17i identifies the number of terminal stations 11 whose throughput will decrease due to the effects of a failure, based on the acquired failure information. The switching instruction device 17i is configured to predict future traffic volume, taking into account the temporary decrease in throughput caused by the failure.

[0359] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the ninth embodiment (hereinafter referred to as "mobile communication system 10i") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation is omitted. The mobile communication system 10i is an example of the communication system 1 described above. The mobile communication system 10i in the ninth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0360] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17i will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17i is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17i and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0361] In the ninth embodiment, the switching instruction device 17i acquires fault information and radio information from the distributed stations 13 to be predicted, and identifies the number of terminal stations 11 whose throughput will decrease due to the effects of the fault and the number of terminal stations 11 that will not be affected by the fault and will satisfy the required throughput.

[0362] For example, the switching instruction device 17i acquires RAN (Radio Access Network) failure information (UE.anomalies) and radio information (RRC.ConnectionUEMean, TBS) from the RIC (RAN Intelligent Controller)'s rAPP (Non-RT RIC Application) as failure information, and identifies the number of terminal stations 11 affected by the failure (i.e., terminal stations 11 whose throughput will decrease) and the total number of terminal stations 11 within the coverage area of ​​the distributed station 13 to be predicted (the population). Then, based on the number of terminal stations 11 affected by the failure and the total number of terminal stations 11 (the population), the switching instruction device 17i identifies the number of terminal stations 11 that are not affected by the failure (i.e., terminal stations 11 that meet the required throughput).

[0363] The switching instruction device 17i then predicts the future traffic volume at each distributed station 13 by predicting that terminal stations 11 unaffected by the failure will have a correspondingly high traffic volume, and terminal stations 11 affected by the failure will have a correspondingly low traffic volume. Based on the future traffic volume at each distributed station 13, the switching instruction device 17i then predicts the future traffic volume at each aggregation station 14.

[0364] Figure 24 is a block diagram showing the functional configuration of the switching instruction device 17i in the ninth embodiment of the present invention. As shown in Figure 24, the switching instruction device 17i is configured to include a prediction unit 172i and a switching determination unit 175. The prediction unit 172i is configured to include a deterioration condition detection unit 176i and a traffic prediction unit 174i.

[0365] The prediction unit 172i receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. In the ninth embodiment, the radio control information includes the number of connected terminals at the terminal station 11 within the coverage area of ​​the distributed station 13, and fault information. As mentioned above, fault information is information indicating the occurrence of a fault that may cause a decrease in communication quality (a decrease in throughput in this embodiment) in devices installed on the communication path, such as the terminal station 11, base station equipment such as the distributed station 13 and aggregation station 14, and the forwarding device 15.

[0366] The degradation condition detection unit 176i of the prediction unit 172i identifies, based on the fault information included in the wireless control information and the number of connected terminals of the terminal station 11 within the coverage area of ​​the distributed station 13, the number of terminal stations 11 whose throughput may decrease due to the effects of a fault, and the number of terminal stations 11 whose throughput will not decrease because they are not affected by the fault.

[0367] More specifically, the degradation condition detection unit 176i counts the number of terminal stations 11 whose throughput may decrease due to the effects of a failure, based on the failure information included in the wireless control information. Then, the degradation condition detection unit 176i subtracts the number of terminal stations 11 whose throughput may decrease from the number of connected terminals (determinant) of terminal stations 11 within the coverage area of ​​the distributed station 13 included in the wireless control information, thereby identifying the number of terminal stations 11 that will not experience a failure and will not experience a decrease in throughput (i.e., terminal stations 11 that meet the required throughput).

[0368] Furthermore, if the failure information includes information indicating a failure at a specific distributed station 13, the degradation condition detection unit 176i may estimate that all terminal stations 11 connected to the distributed station 13 will be affected by the failure, and may consider the number of connected terminals to the distributed station 13 as the number of terminal stations 11 whose throughput may decrease (details will be explained in the first modified example described later). Similarly, if the failure information includes information indicating a failure at a communication device such as an aggregation station 14 or a transfer device 15, the degradation condition detection unit 176i may consider the number of terminal stations 11 under the communication device as the number of terminal stations 11 whose throughput may decrease.

[0369] The degradation condition detection unit 176i then outputs information to the traffic prediction unit 174i indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput decreases due to the effects of the failure. The degradation condition detection unit 176i performs the above process to identify the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput decreases due to the effects of the failure for each distributed station 13.

[0370] The traffic prediction unit 174i of the prediction unit 172i obtains information from the degradation condition detection unit 176i indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput will decrease due to the effects of a failure. The traffic prediction unit 174i predicts the future traffic volume at each distributed station 13 by multiplying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput will decrease due to the effects of a failure by a predetermined throughput value and summing them up.

[0371] For example, the traffic prediction unit 174i multiplies the number of terminal stations 11 that meet the required throughput for each distributed station 13 by the value of the required throughput, and multiplies the number of terminal stations 11 whose throughput will decrease due to the impact of a failure by one-hundredth of the value of the required throughput. Then, the traffic prediction unit 174i predicts the future traffic volume at each distributed station 13 by summing the two calculated values.

[0372] Furthermore, the prediction unit 172i calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172i outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0373] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172i and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0374] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0375] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17i will be described below. Figure 25 is a flowchart showing the operation of the switching instruction device 17i in the ninth embodiment of the present invention. The operation of the switching instruction device 17i shown in the flowchart of Figure 25 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0376] First, the prediction unit 172i receives radio control information from each distributed station 13 that is used to predict the future traffic volume at each distributed station 13 (step S901). As described above, the radio control information in this embodiment includes the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13, and fault information indicating the occurrence of a fault that may cause a decrease in throughput.

[0377] Next, the degradation condition detection unit 176i of the prediction unit 172 counts the number of terminal stations 11 whose throughput may decrease due to the effects of a failure, based on the failure information included in the wireless control information (step S902). Next, the degradation condition detection unit 176i identifies the number of terminal stations 11 whose throughput may decrease by subtracting the number of terminal stations 11 whose throughput may decrease from the number of connected terminals (determinant) of terminal stations 11 within the coverage area of ​​the distributed station 13 (step S903).

[0378] Next, the traffic prediction unit 174i of the prediction unit 172i obtains information from the degradation condition detection unit 176i indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput may decrease due to the effects of a failure. The traffic prediction unit 174i multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput may decrease due to the effects of a failure by a predetermined throughput value (step S904). As described above, for example, the traffic prediction unit 174i multiplies the number of terminal stations 11 that meet the required throughput by the value of the required throughput, and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by a value of 1 / 100 of the value of the required throughput.

[0379] Next, the traffic prediction unit 174i predicts the future traffic volume at each distributed station 13 by summing the two calculated values ​​(step S905). Next, the prediction unit 172i calculates the future traffic volume at each aggregated station 14 based on the connection relationship between the distributed stations 13 and the aggregated station 14 and the future traffic volume at each distributed station 13 (step S906).

[0380] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172i and a predetermined switching threshold (step S907). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S908, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S909).

[0381] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregate station 14 to which the distributed station 13 is connected (step S910). Specifically, the switching determination unit 175 instructs the source aggregate station 14 to release the connection with the distributed station 13 and instructs the destination aggregate station 14 to connect with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is carried out via the communication path after the connection destination switch.

[0382] This completes the operation of the switching instruction device 17i as shown in the flowchart of Figure 25.

[0383] As described above, the switching instruction device 17i in the ninth embodiment of the present invention acquires fault information from the distributed station 13 indicating the occurrence of a fault that may cause a decrease in communication quality, and identifies the number of terminal stations 11 affected by the fault and the number of terminal stations 11 that are not affected by the fault. The switching instruction device 17i then predicts the future traffic volume at each distributed station 13, taking into account that the throughput of the terminal stations 11 affected by the fault will be relatively low.

[0384] The switching instruction device 17i then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distribution station 13. If the switching instruction device 17i determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0385] Furthermore, in this manner, the switching instruction device 17i in the ninth embodiment predicts future traffic based on the number of terminal stations 11 whose communication quality may deteriorate due to the effects of a failure. Therefore, compared to conventional technology that predicts future traffic by predicting the movement of all terminal stations 11, it can suppress the increase in computational load.

[0386] With this configuration, the switching instruction device 17i in the ninth embodiment of the present invention can predict future traffic volume without needing to predict the movement of each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17i can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0387] (First Modification of the Ninth Embodiment) The first modification of the ninth embodiment of the present invention will be described below.

[0388] In the ninth embodiment described above, the switching instruction device 17i was configured to count terminal stations 11 whose communication quality may be degraded due to a failure, based on the failure information contained in the acquired wireless information. In contrast, the switching instruction device in the first modified example of the ninth embodiment described below (hereinafter referred to as "switching instruction device 17j") considers all terminal stations 11 connected to a distributed station 13 to be terminal stations 11 whose communication quality may be degraded due to a failure, if at least one of those terminal stations 11 is under its control.

[0389] In other words, the switching instruction device 17j in the first modified example of the ninth embodiment described below is configured to predict future traffic by considering the number of connected terminals of a distributed station 13 as the number of terminal stations 11 whose communication quality may be degraded due to a failure, for distributed stations 13 that have at least one terminal station 11 under their control. By having such a configuration, the switching instruction device 17j in the first modified example of the ninth embodiment described below can reduce the amount of computation required for predicting future traffic compared to the switching instruction device 17i in the ninth embodiment described above.

[0390] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the first modified example of the ninth embodiment (hereinafter referred to as "mobile communication system 10j") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation is omitted. The mobile communication system 10j is an example of the communication system 1 described above. The mobile communication system 10j in the first modified example of the ninth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0391] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17j will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17j is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17j and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0392] In the first modified example of the ninth embodiment, the switching instruction device 17j acquires fault information and radio information from the distributed stations 13 to be predicted, and identifies the number of terminal stations 11 whose throughput will decrease due to the effects of the fault and the number of terminal stations 11 that will not be affected by the fault and will satisfy the required throughput.

[0393] The switching instruction device 17j then predicts the future traffic volume at each distributed station 13 by predicting that terminal stations 11 unaffected by the failure will have a correspondingly high traffic volume, and terminal stations 11 affected by the failure will have a correspondingly low traffic volume. Based on the future traffic volume at each distributed station 13, the switching instruction device 17j then predicts the future traffic volume at each aggregation station 14.

[0394] Figure 26 is a block diagram showing the functional configuration of the switching instruction device 17j in a first modified example of the ninth embodiment of the present invention. As shown in Figure 26, the switching instruction device 17j is configured to include a prediction unit 172j and a switching determination unit 175. The prediction unit 172j is configured to include a deterioration condition detection unit 176j and a traffic prediction unit 174j.

[0395] The prediction unit 172j receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. In the first modified example of the ninth embodiment, the radio control information includes the number of connected terminals at the terminal station 11 within the coverage area of ​​the distributed station 13, and fault information. As mentioned above, fault information is information indicating the occurrence of a fault that may cause a decrease in communication quality (a decrease in throughput in this embodiment) in devices installed on the communication path, such as the terminal station 11, base station equipment such as the distributed station 13 and aggregation station 14, and the forwarding device 15.

[0396] The degradation condition detection unit 176j of the prediction unit 172j identifies, based on the failure information contained in the wireless control information, whether or not there is at least one terminal station 11 connected to the distributed station 13 to be predicted that could be affected by a failure and have its throughput reduced.

[0397] If even one terminal station 11 connected to the distributed station 13 being predicted is capable of experiencing a decrease in throughput, the degradation condition detection unit 176j considers the number of connected terminals within the coverage area of ​​the distributed station 13 to be the number of terminal stations 11 capable of experiencing a decrease in throughput. In other words, if even one terminal station 11 connected to the distributed station 13 being predicted is capable of experiencing a decrease in throughput, the degradation condition detection unit 176j considers all terminal stations 11 connected to the said distributed station 13 to be terminal stations 11 capable of experiencing a decrease in throughput due to the effects of a failure.

[0398] The degradation condition detection unit 176j then estimates the number of terminal stations 11 whose throughput may decrease due to the effects of a failure for each distributed station 13 by the above process. The degradation condition detection unit 176j also identifies the number of terminal stations 11 whose throughput may decrease by subtracting the number of terminal stations 11 identified above from the total number of connected terminals (determinant) of terminal stations 11 within the coverage area of ​​each distributed station 13 included in the wireless control information, thereby determining the number of terminal stations 11 that will not experience a failure and will not experience a decrease in throughput (i.e., terminal stations 11 that meet the required throughput).

[0399] The degradation condition detection unit 176j then outputs information to the traffic prediction unit 174j indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput decreases due to the effects of the failure. The degradation condition detection unit 176j performs the above process of identifying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput decreases due to the effects of the failure for each distributed station 13.

[0400] The traffic prediction unit 174j of the prediction unit 172j obtains information from the degradation condition detection unit 176j indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput will decrease due to the effects of a failure. The traffic prediction unit 174j predicts the future traffic volume at each distributed station 13 by multiplying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput will decrease due to the effects of a failure by a predetermined throughput value and summing them up.

[0401] For example, the traffic prediction unit 174j multiplies the number of terminal stations 11 that meet the required throughput for each distributed station 13 by the value of the required throughput, and multiplies the number of terminal stations 11 whose throughput will decrease due to the impact of a failure by one-hundredth of the value of the required throughput. Then, the traffic prediction unit 174j predicts the future traffic volume at each distributed station 13 by summing the two calculated values.

[0402] Furthermore, the prediction unit 172j calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172j outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0403] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172j and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0404] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0405] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17j will be described below. Figure 27 is a flowchart showing the operation of the switching instruction device 17j in the first modified example of the ninth embodiment of the present invention. The operation of the switching instruction device 17j shown in the flowchart of Figure 27 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0406] First, the prediction unit 172j receives radio control information from each distributed station 13 that is used to predict the future traffic volume at each distributed station 13 (step S1001). As described above, the radio control information in this embodiment includes the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13, and fault information indicating the occurrence of a fault that may cause a decrease in throughput.

[0407] Next, the degradation condition detection unit 176j of the prediction unit 172 identifies whether or not there is at least one terminal station 11 connected to the target distributed station 13 that could be affected by a failure and have its throughput reduced, based on the failure information contained in the radio control information. Then, the degradation condition detection unit 176j sums up the number of connected terminals for each of the distributed stations 13 that have been identified in this way (step S1002). Next, the degradation condition detection unit 176j subtracts the number of connected terminals for each of the distributed stations 13 identified above (i.e., the number of terminal stations 11 that could have their throughput reduced due to a failure and the number of terminal stations 11 that are considered to have their throughput reduced) from the total number of connected terminals for each of the terminal stations 11 in the coverage area of ​​each distributed station 13 to identify the number of terminal stations 11 that can satisfy the required throughput (i.e., terminal stations 11 that do not experience failures and have their throughput not reduced) (step S1003).

[0408] Next, the traffic prediction unit 174j of the prediction unit 172j obtains information from the degradation condition detection unit 176j indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput may decrease due to the effects of a failure. The traffic prediction unit 174j multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput may decrease due to the effects of a failure by a predetermined throughput value (step S1004). As described above, for example, the traffic prediction unit 174j multiplies the number of terminal stations 11 that meet the required throughput by the value of the required throughput, and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by a value of 1 / 100 of the value of the required throughput.

[0409] Next, the traffic prediction unit 174j predicts the future traffic volume at each distributed station 13 by summing the calculated values ​​(step S1005). Next, the prediction unit 172j calculates the future traffic volume at each aggregated station 14 based on the connection relationship between the distributed stations 13 and the aggregated station 14 and the future traffic volume at each distributed station 13 (step S1006).

[0410] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172j and a predetermined switching threshold (step S1007). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S1008, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S1009).

[0411] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregate station 14 to which the distributed station 13 is connected (step S1010). Specifically, the switching determination unit 175 instructs the source aggregate station 14 to release the connection with the distributed station 13, and instructs the destination aggregate station 14 to connect with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer takes place through the communication path after the connection destination has been switched.

[0412] This completes the operation of the switching instruction device 17j as shown in the flowchart of Figure 27.

[0413] As described above, in the first modified example of the ninth embodiment of the present invention, the switching instruction device 17j acquires fault information from the distributed station 13 indicating the occurrence of a fault that may cause a decrease in communication quality, and identifies the distributed station 13 that has at least one terminal station 11 under its control that is affected by the fault. The switching instruction device 17j considers all terminal stations 11 under the identified distributed station 13 to be terminal stations 11 that are affected by the fault. The switching instruction device 17j identifies the number of terminal stations 11 that are affected by the fault and the number of terminal stations 11 that are not affected by the fault. Then, the switching instruction device 17j predicts the future traffic volume at each distributed station 13, taking into account that the throughput of the terminal stations 11 affected by the fault will be relatively low.

[0414] The switching instruction device 17j then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distribution station 13. If the switching instruction device 17j determines that there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion will occur, it will perform a route switch to distribute the load to a link with available bandwidth.

[0415] Furthermore, in the first modified example of the ninth embodiment, the switching instruction device 17j predicts future traffic based on the number of terminal stations 11 whose communication quality may deteriorate due to the effects of a failure. Therefore, compared to conventional technology that predicts future traffic by predicting the movement of all terminal stations 11, it can suppress the increase in computational load.

[0416] With this configuration, the switching instruction device 17j in the first modified example of the ninth embodiment of the present invention can predict future traffic volume without needing to predict the movement of each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to frequent movement of terminal stations 11. As a result, the switching instruction device 17j can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0417] (Second Modification of the Ninth Embodiment) A second modification of the ninth embodiment of the present invention will be described below.

[0418] In the first modified example of the ninth embodiment described above, the switching instruction device 17j was configured to identify the number of terminal stations 11 that would satisfy the required throughput by subtracting from the total number of connected terminals of each distributed station 13 (the base number) that there is at least one terminal station 11 under its control whose throughput may decrease due to the effects of a failure, based on the acquired failure information.

[0419] In contrast, the switching instruction device in the second modified example of the ninth embodiment described below (hereinafter referred to as "switching instruction device 17k") counts, based on the acquired fault information, the number of connected terminals in a distributed station 13 that has at least one terminal station 11 under its control that could reduce throughput due to the effects of a fault, and the number of connected terminals in a distributed station 13 that does not have any terminal stations 11 under its control that could reduce throughput due to the effects of a fault.

[0420] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the second modified example of the ninth embodiment (hereinafter referred to as "mobile communication system 10k") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation is omitted. The mobile communication system 10k is an example of the communication system 1 described above. The mobile communication system 10k in the second modified example of the ninth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device.

[0421] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17k will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17k is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17k and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0422] In the second modified example of the ninth embodiment, the switching instruction device 17k acquires fault information and radio information from the distributed stations 13 to be predicted, and identifies the number of terminal stations 11 whose throughput will decrease due to the effects of the fault and the number of terminal stations 11 that will not be affected by the fault and will satisfy the required throughput.

[0423] The switching instruction device 17k then predicts the future traffic volume at each distributed station 13 by predicting that terminal stations 11 unaffected by the failure will have a correspondingly high traffic volume, and terminal stations 11 affected by the failure will have a correspondingly low traffic volume. Based on the future traffic volume at each distributed station 13, the switching instruction device 17k then predicts the future traffic volume at each aggregation station 14.

[0424] Figure 28 is a block diagram showing the functional configuration of the switching instruction device 17k in a second modified example of the ninth embodiment of the present invention. As shown in Figure 28, the switching instruction device 17k is configured to include a prediction unit 172k and a switching determination unit 175. The prediction unit 172k is configured to include a deterioration condition detection unit 176k and a traffic prediction unit 174k.

[0425] The prediction unit 172k receives radio control information from each distributed station 13, which is used to predict the future traffic volume at each distributed station 13. In the second modified example of the ninth embodiment, the radio control information includes the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13, and fault information. Fault information is, for example, information indicating the occurrence of a fault at the terminal station 11 that may cause a decrease in communication quality (a decrease in throughput in this embodiment).

[0426] The degradation condition detection unit 176k of the prediction unit 172k identifies, based on the failure information contained in the wireless control information, whether or not there is at least one terminal station 11 connected to the distributed station 13 to be predicted that could be affected by a failure and result in a decrease in throughput.

[0427] If even one terminal station 11 connected to the target distributed station 13 has a throughput that may decrease, the degradation condition detection unit 176k counts the number of connected terminals within the coverage area of ​​the distributed station 13 as the number of terminal stations 11 whose throughput may decrease. In other words, if even one terminal station 11 connected to the target distributed station 13 has a throughput that may decrease, the degradation condition detection unit 176k considers all terminal stations 11 connected to the said distributed station 13 to be terminal stations 11 whose throughput may decrease due to the effects of a failure.

[0428] Then, the degradation condition detection unit 176k estimates the number of terminal stations 11 whose throughput may decrease due to the effects of a failure for each distributed station 13, as described above, and sums them up.

[0429] Furthermore, if there are no terminal stations 11 among the terminal stations 11 connected to the distributed station 13 that are subject to prediction that could result in a decrease in throughput, the degradation condition detection unit 176k counts the number of connected terminals within the coverage area of ​​the distributed station 13 as the number of terminal stations 11 that are not affected by the occurrence of a failure and can meet the required throughput. In other words, if there are no terminal stations 11 among the terminal stations 11 connected to the distributed station 13 that are subject to prediction that could result in a decrease in throughput, the degradation condition detection unit 176k considers all terminal stations 11 connected to the said distributed station 13 to be terminal stations 11 that are not affected by the occurrence of a failure and can meet the required throughput.

[0430] Then, the degradation condition detection unit 176k estimates the number of terminal stations 11 that are not affected by the occurrence of a failure and can meet the required throughput for each distributed station 13, as described above, and sums them up.

[0431] The degradation condition detection unit 176k then outputs information to the traffic prediction unit 174k indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput decreases due to the effects of a failure. The degradation condition detection unit 176k performs the above process of identifying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput decreases due to the effects of a failure for each distributed station 13.

[0432] The traffic prediction unit 174k of the prediction unit 172k obtains information from the degradation condition detection unit 176k indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput will decrease due to the effects of a failure. The traffic prediction unit 174k predicts the future traffic volume at each distributed station 13 by multiplying the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput will decrease due to the effects of a failure by a predetermined throughput value and summing them up.

[0433] For example, the traffic prediction unit 174k multiplies the number of terminal stations 11 that meet the required throughput for each distributed station 13 by the value of the required throughput, and multiplies the number of terminal stations 11 whose throughput will decrease due to the effects of a failure by one-hundredth of the value of the required throughput. Then, the traffic prediction unit 174k predicts the future traffic volume at each distributed station 13 by summing the two calculated values.

[0434] Furthermore, the prediction unit 172k calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172k outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0435] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172k and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0436] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0437] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17k will be described below. Figure 29 is a flowchart showing the operation of the switching instruction device 17k in a second modified example of the ninth embodiment of the present invention. The operation of the switching instruction device 17k shown in the flowchart of Figure 29 is repeatedly started at predetermined intervals, for example. That is, the prediction of future traffic volume and the decision to execute route switching based on the prediction result are performed at predetermined intervals.

[0438] First, the prediction unit 172k receives radio control information from each distributed station 13 that is used to predict the future traffic volume at each distributed station 13 (step S1101). As described above, the radio control information in this embodiment includes the number of connected terminals at each terminal station 11 within the coverage area of ​​the distributed station 13, and fault information indicating the occurrence of a fault that may cause a decrease in throughput.

[0439] Next, the degradation condition detection unit 176k of the prediction unit 172 identifies, based on the failure information contained in the wireless control information, whether or not there is at least one terminal station 11 connected to the target distributed station 13 that could be affected by a failure and experience a decrease in throughput. Then, the degradation condition detection unit 176k sums up the number of connected terminals for each distributed station 13 that has been identified as having at least one terminal station 11 connected to the target distributed station 13 that could be affected by a failure and experience a decrease in throughput (step S1102).

[0440] Next, the degradation condition detection unit 176k sums up the number of connected terminals in each distributed station 13 that has been identified as having no terminal stations 11 connected to the distributed station 13 that could experience a decrease in throughput due to a failure, and uses this sum as the number of terminal stations 11 that satisfy the required throughput (step S1103).

[0441] Next, the traffic prediction unit 174k of the prediction unit 172k obtains information from the degradation condition detection unit 176k indicating the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput may decrease due to the effects of a failure. The traffic prediction unit 174k multiplies the number of terminal stations 11 that meet the required throughput and the number of terminal stations 11 whose throughput may decrease due to the effects of a failure by a predetermined throughput value (step S1104). As described above, for example, the traffic prediction unit 174k multiplies the number of terminal stations 11 that meet the required throughput by the value of the required throughput, and multiplies the number of terminal stations 11 whose signal strength is lower than the threshold by a value of 1 / 100 of the value of the required throughput.

[0442] Next, the traffic prediction unit 174k predicts the future traffic volume at each distributed station 13 by summing the two values ​​(step S1105). Next, the prediction unit 172k calculates the future traffic volume at each aggregated station 14 based on the connection relationship between the distributed stations 13 and the aggregated station 14 and the future traffic volume at each distributed station 13 (step S1106).

[0443] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172k and a predetermined switching threshold (step S1107). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S1108, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S1109).

[0444] Next, the switching determination unit 175 gives an instruction for path switching to change the aggregation station 14 to which the distributed station 13 is connected for each device (step S1110). Specifically, the switching determination unit 175 instructs the aggregation station 14 that is the source of switching to release the connection with the distributed station 13, and instructs the aggregation station 14 that is the destination of switching to connect to the distributed station 13. Further, the switching instruction device 17 instructs the transfer device 15 to perform path switching so that signal transfer is performed via the communication path after the connection destination is switched.

[0445] This completes the operation of the switching instruction device 17k shown in the flowchart of FIG. 29.

[0446] As described above, the switching instruction device 17k in the second modification of the ninth embodiment of the present invention acquires failure information indicating the occurrence of a failure that may cause a deterioration in communication quality from the distributed station 13, and determines the distributed station 13 that includes at least one terminal station 11 affected by the occurrence of the failure and the distributed station 13 that does not include any terminal station 11 affected by the occurrence of the failure. The switching instruction device 17k regards all the terminal stations 11 under the distributed station 13 that includes at least one terminal station 11 affected by the occurrence of the failure as terminal stations 11 affected by the occurrence of the failure. The switching instruction device 17k identifies the number of terminal stations 11 affected by the occurrence of the failure and the number of terminal stations 11 not affected by the occurrence of the failure. Then, the switching instruction device 17k predicts the future traffic volume in each distributed station 13 in consideration of the relatively low throughput of the terminal stations 11 affected by the occurrence of the failure.

[0447] Then, the switching instruction device 17k predicts the future traffic volume in each aggregation station 14 based on the future traffic volume in each distributed station 13. If there is an aggregation station 14 determined to have a traffic jam because the future traffic volume exceeds a predetermined switching threshold value, the switching instruction device 17k executes path switching for load distribution to a link with available bandwidth.

[0448] Note that, as described above, the switching instruction device 17k in the second modification of the ninth embodiment predicts future traffic based on the number of terminal stations 11 whose communication quality may deteriorate due to the occurrence of a failure. Therefore, compared with the conventional technology that predicts future traffic by performing movement prediction for all the terminal stations 11, an increase in the amount of calculation can be suppressed.

[0449] By having such a configuration, the switching instruction device 17k in the second modification of the ninth embodiment of the present invention can predict the future traffic volume without the need to perform movement prediction for each of the terminal stations 11 even in a situation where the number of connected terminals in the coverage area of the distributed station 13 fluctuates randomly due to, for example, a large amount of movement of the terminal stations 11. Thereby, even in a situation where the number of connected terminals of the distributed station 13 fluctuates randomly, the switching instruction device 17k can predict the presence or absence of occurrence of congestion in advance while suppressing an increase in the amount of calculation, and can realize low-latency communication.

[0450] Note that, among the configurations of the seventh embodiment (or its modification), the configuration of the eighth embodiment, and the configuration of the ninth embodiment (or its modification) described above, a configuration in which two or more of the configurations are combined may be used. That is, a configuration that takes into account that the throughput of the terminal station 11 with relatively low radio wave intensity becomes relatively low, which is the configuration of the seventh embodiment described above, a configuration that takes into account a temporary decrease in throughput due to the execution of a handover, which is the configuration of the eighth embodiment described above, and a configuration that takes into account a decrease in throughput due to the occurrence of a failure, which is the configuration of the ninth embodiment described above, may be combined to predict the future traffic volume, or a configuration that combines any two of these configurations to predict the future traffic volume may be used.

[0451] The switching instruction devices 17, 17a to 17k in the first to ninth embodiments and the modified versions thereof described above were configured to predict the future traffic volume at all distributed stations 13 based on wireless control information such as the recent trend (fluctuation) of the traffic volume per terminal station 11 acquired from the distributed station 13, the recent traffic volume and number of connected terminals at the distributed station 13, the recent fluctuation in the position of each terminal station 11, the radio wave strength for each terminal station 11, the increase or decrease in the number of terminal stations 11 that will undergo handover, and the number of terminal stations 11 whose communication quality will deteriorate due to the effects of failure.

[0452] However, if the number of terminal stations 11 connected to the distributed station 13 increases, the amount of wireless control information collected from each terminal station 11 will increase. If the amount of information increases too much, it will become difficult for the distributed station 13 to transmit wireless control information to the switching instruction devices 17, 17a to 17k at a higher frequency. Furthermore, if the amount of information increases too much, the load on the future traffic volume prediction processing will also increase, making it difficult for the switching instruction devices 17, 17a to 17k to predict future traffic volume at a higher frequency. As a result, the frequency of route switching will decrease, making it impossible for the switching instruction devices 17, 17a to 17k to perform congestion control in real time, and potentially making it impossible to achieve low-latency communication.

[0453] In contrast, the switching instruction devices 17l to 17s in the 10th to 16th embodiments and the modified versions of each embodiment described below further include a prediction necessity determination unit, which will be described later. The prediction necessity determination unit determines for each distributed station 13 whether it is necessary to perform future traffic volume prediction processing at the distributed station 13, based on a relatively small number of types of radio control information (for example, a single type of radio control information) obtained from the distributed station 13. The switching instruction devices 17l to 17s then perform future traffic volume prediction processing using a relatively large number of types of radio control information only for the distributed stations 13 that are deemed necessary. In other words, the switching instruction devices 17l to 17s in this embodiment omit the execution of future traffic volume prediction processing for some of the distributed stations 13.

[0454] By having such a configuration, the switching instruction devices 17l to 17s in the 10th to 16th embodiments and modified versions of each embodiment described below can suppress the increase in the amount of wireless control information to be collected and reduce the load on future traffic volume prediction processing, even when the number of terminal stations 11 housed in the distributed station 13 increases. As a result, the switching instruction devices 17l to 17s in the 10th to 16th embodiments and modified versions of each embodiment described below can perform congestion control more efficiently and realize low-latency communication, even when the number of terminal stations 11 housed in the distributed station 13 increases.

[0455] The following describes the tenth to sixteenth embodiments of the present invention and variations of each embodiment.

[0456] <Tenth Embodiment> A tenth embodiment of the present invention will be described below.

[0457] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the tenth embodiment (hereinafter referred to as "mobile communication system 10l") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation will be omitted. The mobile communication system 10l is an example of the communication system 1 described above. The mobile communication system 10l in the tenth embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device. Hereinafter, the switching instruction device in the tenth embodiment will be referred to as "switching instruction device 17l".

[0458] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17l will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17l is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17l in this embodiment and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0459] In the tenth embodiment, the switching instruction device 17l does not predict future traffic volume for all distributed stations 13, as in the switching instruction device 17 in the first embodiment described above. Instead, it determines whether or not to perform prediction processing for each distributed station 13, and predicts future traffic volume only for some of the distributed stations 13 that are deemed necessary.

[0460] On the other hand, the switching instruction device 17l omits the execution of future traffic volume prediction processing for distributed stations 13 that it has determined do not need to be performed. In this case, the switching instruction device 17l performs future traffic volume prediction processing at the aggregation station 14, which is a higher-level device, for distributed stations 13 that it has determined do not need to be performed, for example by substituting the most recent traffic volume value at the distributed station 13 or by substituting the future traffic volume value at the distributed station 13 that has been calculated in the past.

[0461] The switching instruction device 17l then predicts the future traffic volume at each distributed station 13 based on the recent trend (fluctuation) of the traffic volume per terminal station 11 at each distributed station 13. Then, the switching instruction device 17l predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13.

[0462] Figure 30 is a block diagram showing the functional configuration of the switching instruction device 17l in the tenth embodiment of the present invention. As shown in Figure 30, the switching instruction device 17l is configured to include a prediction necessity determination unit 170l, a prediction unit 172l, and a switching determination unit 175. The prediction unit 172l is configured to include a traffic prediction unit 174l.

[0463] The prediction necessity determination unit 170l performs a process (hereinafter referred to as "prediction necessity determination process") to determine for each distributed station 13 whether or not it is necessary to perform future traffic volume prediction processing at each distributed station 13. The future traffic volume prediction processing at each distributed station 13 is a process performed by the prediction unit (in this embodiment, the prediction unit 172l), as in the embodiments described above.

[0464] The prediction necessity determination unit 170l collects relatively few types of radio control information (hereinafter referred to as "prediction necessity determination information") used for prediction necessity determination processing from each distributed station 13. In this embodiment, the prediction necessity determination information is a single type of radio control information, such as the value of the most recent traffic volume at the distributed station 13, such as TBS (Transport Block Size). The prediction necessity determination unit 170l compares the value of the most recent traffic volume at the distributed station 13 with a predetermined threshold and performs prediction necessity determination processing for each distributed station 13.

[0465] The predetermined threshold is stored in advance in a storage medium (not shown) provided in, for example, the switching instruction device 17l. For example, the threshold S, which is obtained by the following equation (3), is set in advance as the predetermined threshold.

[0466] S = (Link rate of aggregation station 14) / (Number of connected terminals at distributed station 13) × 0.8 ... (3)

[0467] For example, the prediction necessity determination unit 170l compares the acquired traffic volume (TBS) value with a threshold S. If the traffic volume (TBS) value at a distributed station 13 is smaller than the threshold S, the prediction necessity determination unit 170l determines that there is a high possibility of congestion occurring, and therefore, it is necessary to perform prediction processing of the future traffic volume at the distributed station 13 (using the relatively large amount of radio control information from the traffic prediction unit 174l). In this way, the prediction necessity determination unit 170l performs prediction necessity determination processing for each distributed station 13.

[0468] The prediction necessity determination unit 170l may predict the amount of future traffic volume fluctuations at the distributed station 13 based on time-series traffic volume values, including values ​​acquired in the past, and compare the predicted amount of future traffic volume fluctuations with a predetermined threshold. The prediction necessity determination unit 170l may also be configured to perform a combination of a threshold comparison for future traffic volume fluctuations and a comparison for future traffic volume fluctuations.

[0469] Alternatively, the prediction necessity determination unit 170l may calculate the probability that the future traffic volume value at the distributed station 13 will exceed a threshold S based on time-series traffic volume values, including values ​​acquired in the past, and determine that the distributed station 13 is one that requires future traffic volume prediction processing (using relatively large amounts of radio control information by the traffic prediction unit 174l) if the calculated probability is higher than a predetermined threshold.

[0470] The prediction necessity determination unit 170l notifies the prediction unit 172l of information indicating the distributed stations 13 for which it has been determined that future traffic volume prediction processing is necessary. The prediction necessity determination unit 170l may also notify the prediction unit 172l of both information indicating the distributed stations 13 for which it has been determined that future traffic volume prediction processing is necessary, and information indicating the distributed stations 13 for which it has been determined that future traffic volume prediction processing is not necessary.

[0471] The prediction unit 172l acquires information from the prediction necessity determination unit 170l indicating the distributed station 13 for which it has been determined that future traffic volume prediction processing is necessary. The prediction unit 172l also receives relatively common types of radio control information from each distributed station 13 that are used to predict future traffic volume at the distributed station 13. The radio control information in the tenth embodiment includes data that associates the most recent traffic volume value at the distributed station 13 with the most recent number of connected terminals at the distributed station 13.

[0472] Furthermore, as mentioned above, the prediction unit 172l may obtain the most recent traffic volume values ​​from the prediction unit 170l instead of obtaining them again from the prediction unit 170l, since this information has already been acquired by the prediction unit 170l from each distributed station 13.

[0473] Furthermore, the prediction necessity determination unit 170l may be configured to pre-collect the above-mentioned relatively common types of radio control information used to predict future traffic volume at the distributed station 13, along with prediction necessity determination information (i.e., the value of the most recent traffic volume at the distributed station 13). In this case, the prediction unit 172l acquires the relatively common types of radio control information used to predict future traffic volume at the distributed station 13 from the prediction necessity determination unit 170l.

[0474] The traffic prediction unit 174l of the prediction unit 172l predicts the future traffic volume at each distributed station 13 using radio control information, but only for the distributed stations 13 for which it has been determined that future traffic volume prediction processing is necessary.

[0475] Furthermore, for distributed stations 13 where it is determined that future traffic volume prediction processing is unnecessary, the traffic prediction unit 174l omits the future traffic volume prediction processing and instead considers the most recent traffic volume value at the distributed station 13 as the future traffic volume. Alternatively, for distributed stations 13 where it is determined that future traffic volume prediction processing is unnecessary, the traffic prediction unit 174l may omit the future traffic volume prediction processing and instead use a previously calculated value of future traffic volume at the distributed station 13.

[0476] Furthermore, the prediction unit 172l calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172l outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0477] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction unit 172l and a predetermined switching threshold. The predetermined switching threshold is a value predetermined based on, for example, the bandwidth of the upper link. If the switching determination unit 175 determines that congestion will occur at any aggregation station 14 because the future traffic volume will exceed the bandwidth of the upper link, it decides to perform a route switch to distribute the load to links with available bandwidth. If the switching determination unit 175 decides to perform a route switch, it issues a route switch instruction to each device to change the aggregation station 14 to which the distributed station 13 is connected.

[0478] Specifically, the switching determination unit 175 instructs the source aggregation station 14 to release the connection with the distributed station 13, and instructs the destination aggregation station 14 to establish a connection with the said distributed station 13. Furthermore, the switching determination unit 175 instructs the transfer device 15 to switch routes so that signal transfer takes place via the communication path after the connection destination has been switched.

[0479] [Operation of the Switching Instruction Device] An example of the operation of the switching instruction device 17l will be described below. Figure 31 is a flowchart showing the operation of the switching instruction device 17l in the tenth embodiment of the present invention. The operation of the switching instruction device 17l shown in the flowchart of Figure 31 is repeatedly started at predetermined intervals, for example. That is, the determination of whether or not prediction is necessary, the prediction of future traffic volume, and the determination to execute route switching based on the prediction results are performed at predetermined intervals.

[0480] First, the prediction necessity determination unit 170l acquires prediction necessity determination information, which is a relatively small amount of radio control information, from each distributed station 13 (step S1201). As described above, the prediction necessity determination information in this embodiment is a single type of radio control information, which is the value of the most recent traffic volume. Next, the prediction necessity determination unit 170l compares the value of the most recent traffic volume at each distributed station 13 with a predetermined threshold and performs a prediction necessity determination process for each distributed station 13 (step S1202). Next, the prediction necessity determination unit 170l notifies the prediction unit 172l of information indicating the distributed station 13 for which it has been determined that future traffic volume prediction processing is necessary (step S1203).

[0481] Next, the prediction unit 172l acquires, from each distributed station 13, a plurality of types of radio control information that is relatively large and used for predicting the future traffic volume in the distributed station 13 (step S1204). For example, the prediction unit 172l acquires radio control information in which time, traffic volume, and the number of connected terminals are associated from each distributed station 13.

[0482] Next, the traffic prediction unit 174l of the prediction unit 172l executes, for only the distributed stations 13 for which it is determined that execution of the future traffic volume prediction process is necessary, the processes for predicting the future traffic volume in each distributed station 13 from step S1205 to step S1208 below.

[0483] The traffic prediction unit 174l calculates the traffic volume per terminal station 11 at each time for each distributed station 13 that is the target of the prediction process by dividing the value of the traffic volume by the number of connected terminals (step S1205). Next, the traffic prediction unit 174l calculates the amount of change in the traffic volume per terminal station 11 for each unit period based on the calculated traffic volume per terminal station 11 at each time (step S1206).

[0484] Next, the traffic prediction unit 174l predicts the future traffic volume per terminal station 11 based on the transition of the calculated amount of change in the traffic volume per terminal station 11 for each unit period (step S1207). At this time, as described above, the traffic prediction unit 174 may perform weighting so that the more recent the amount of change in the traffic volume is, the more it is considered in the prediction value, and predict the future traffic volume per terminal station 11.

[0485] Next, the traffic prediction unit 174l calculates the future traffic volume at each distributed station 13 by multiplying the future traffic volume per terminal station 11 by the future number of connected terminals at that distributed station 13 (step S1208). Next, the traffic prediction unit 174l calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14 and the future traffic volume at each distributed station 13 (step S1209).

[0486] As mentioned above, the traffic prediction unit 174l omits the prediction process for distributed stations 13 where it is determined that future traffic volume prediction processing is unnecessary. Instead, it calculates the future traffic volume at each aggregation station 14 by, for example, substituting the most recent traffic volume value at the distributed station 13 or by substituting a previously calculated future traffic volume value at the distributed station 13.

[0487] Next, the switching determination unit 175 predicts whether or not future congestion will occur based on the relationship between the future traffic volume at each aggregation station 14 calculated by the prediction unit 172l and a predetermined switching threshold (step S1210). Next, if there is an aggregation station 14 where the future traffic volume exceeds the predetermined switching threshold and congestion is determined to occur (step S1211, YES), the switching determination unit 175 decides to perform a route switch to distribute the load to links with available bandwidth and determines the communication path after the switch (step S1212).

[0488] Next, the switching determination unit 175 issues instructions to each device for route switching to change the aggregation station 14 to which the distributed station 13 is connected (step S1213). Specifically, the switching determination unit 175 instructs the aggregation station 14 from which the switching is initiated to release the connection with the distributed station 13, and instructs the aggregation station 14 to which the switching is initiated to establish a connection with the said distributed station 13. Furthermore, the switching instruction device 17 instructs the transfer device 15 to switch routes so that signal transfer is performed via the communication path after the connection destination has been switched.

[0489] This completes the operation of the switching instruction device 17l as shown in the flowchart of Figure 31.

[0490] As described above, in the tenth embodiment of the present invention, the switching instruction device 17l predicts the future traffic volume at each distributed station 13 based on the recent trend (fluctuation) of the traffic volume per terminal station 11 at each distributed station 13. Then, the switching instruction device 17l predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13. If there is an aggregation station 14 where the future traffic volume exceeds a predetermined switching threshold and congestion is determined to occur, the switching instruction device 17l performs route switching to distribute the load to links with available bandwidth.

[0491] With this configuration, the switching instruction device 17l in the tenth embodiment of the present invention can predict future traffic volume without having to perform movement predictions for each terminal station 11, even in situations where the number of connected terminals in the coverage area of ​​the distributed station 13 fluctuates randomly due to, for example, frequent movement of terminal stations 11. As a result, the switching instruction device 17l can predict in advance whether or not congestion will occur while suppressing an increase in computational load, even in situations where the number of connected terminals in the distributed station 13 fluctuates randomly, thereby realizing low-latency communication.

[0492] Furthermore, as described above, the switching instruction device 17l in the tenth embodiment of the present invention performs a prediction necessity determination process for each distributed station 13, using prediction necessity determination information, which is a relatively small number of types (single type) of wireless control information, to determine whether or not it is necessary to perform prediction processing for future traffic volume. Then, the switching instruction device 17l performs prediction processing for future traffic volume using a relatively large number of types of wireless control information, limited to only the distributed stations 13 that it has determined are necessary.

[0493] By having such a configuration, the switching instruction device 17l in the tenth embodiment of the present invention can omit the execution of future traffic volume prediction processing for some distributed stations 13. Therefore, even if the number of terminal stations 11 housed in the distributed stations 13 increases, it is possible to suppress the increase in the amount of wireless control information to be collected and reduce the load on future traffic volume prediction processing. As a result, the switching instruction device 17l can perform congestion control more efficiently even if the number of terminal stations 11 housed in the distributed stations 13 increases, and low-latency communication can be achieved.

[0494] (Modification of the 10th Embodiment) A modification of the 10th embodiment of the present invention will be described below.

[0495] [Overall Configuration of the Communication System] The overall configuration of the mobile communication system in the modified version of the 10th embodiment (hereinafter referred to as "mobile communication system 10m") is the same as the overall configuration of the mobile communication system 10 in the first embodiment shown in Figure 2 above, so the explanation is omitted. The mobile communication system 10m is an example of the communication system 1 described above. The mobile communication system 10m in the modified version of the 10th embodiment and the mobile communication system 10 in the first embodiment differ in the configuration of the switching instruction device. Hereinafter, the switching instruction device in the modified version of the 10th embodiment will be referred to as "switching instruction device 17m".

[0496] [Functional Configuration of the Switching Instruction Device] The functional configuration of the switching instruction device 17m will be described below. Similar to the switching instruction device 17 in the first embodiment described above, the switching instruction device 17m is a device that predicts the future traffic volume at each aggregation station 14, predicts whether or not congestion will occur, and performs control to execute route switching as necessary. However, the switching instruction device 17m in the modified version of the tenth embodiment and the switching instruction device 17 in the first embodiment described above differ in the processing that predicts the future traffic volume at each distributed station 13, which is a preliminary process to predicting the future traffic volume at each aggregation station 14.

[0497] In the modified version of the tenth embodiment, the switching instruction device 17m, similar to the switching instruction device 17l in the tenth embodiment described above, does not predict future traffic volume for all distributed stations 13, but rather determines whether or not to perform prediction processing for each distributed station 13, and predicts future traffic volume only for some of the distributed stations 13 that are deemed necessary.

[0498] On the other hand, the switching instruction device 17m omits the execution of future traffic volume prediction processing for distributed stations 13 that it has determined do not require it. In this case, the switching instruction device 17m performs future traffic volume prediction processing at the aggregation station 14, which is a higher-level device, for distributed stations 13 that it has determined do not require it, for example by substituting the most recent traffic volume value at the distributed station 13 or by substituting the future traffic volume value at the distributed station 13 that has been calculated in the past.

[0499] The switching instruction device 17m then predicts the future traffic volume at each distributed station 13 based on the number of connected terminals at each distributed station 13 and the signal strength values ​​of the radio waves from the terminal stations 11 connected to the distributed station 13. The switching instruction device 17m then predicts the future traffic volume at each aggregation station 14 based on the future traffic volume at each distributed station 13.

[0500] Figure 32 is a block diagram showing the functional configuration of a switching instruction device 17m in a modified example of the tenth embodiment of the present invention. As shown in Figure 32, the switching instruction device 17m is configured to include a prediction necessity determination unit 170m, a prediction unit 172m, and a switching determination unit 175. The prediction unit 172m is configured to include a traffic prediction unit 174m.

[0501] The prediction necessity determination unit 170m performs prediction necessity determination processing. As described above, prediction necessity determination processing is the process of determining for each distributed station 13 whether or not it is necessary to perform prediction processing of future traffic volume at the distributed station 13. The prediction processing of future traffic volume at the distributed station 13 is a process performed by the prediction unit (in this modified example, the prediction unit 172m), as in the embodiments described above.

[0502] The prediction necessity determination unit 170m collects prediction necessity determination information from each distributed station 13. As mentioned above, prediction necessity determination information is a relatively small amount of wireless control information used in the prediction necessity determination process. In this embodiment, the prediction necessity determination information consists of two types of wireless control information, for example, the number of connected terminals for each distributed station 13 and the signal strength value of the radio waves from the terminal station 11 connected to the distributed station 13.

[0503] Thus, the switching instruction device 17m in the modified version of the tenth embodiment described above differs in configuration from the switching instruction device 17l in the tenth embodiment described above, which performs prediction necessity determination processing using a single type of wireless control information (traffic volume (TBS)), in that it performs prediction necessity determination processing using multiple (two types) of wireless control information.

[0504] The prediction necessity determination unit 170m compares the number of connected terminals at the distributed station 13 with a predetermined threshold, and also compares it with the signal strength value of the radio waves from the terminal station 11, and performs a prediction necessity determination process for each distributed station 13.

[0505] More specifically, the prediction necessity determination unit 170m determines, for example, that if the number of connected terminals at the distributed station 13 is greater than a predetermined threshold, and the average signal strength of the radio waves from the terminal station 11 is higher than a predetermined threshold, then there is a high possibility of congestion occurring, and therefore the distributed station 13 is one that needs to perform future traffic volume prediction processing.

[0506] The predetermined threshold is pre-stored, for example, in a storage medium (not shown) provided by the switching instruction device 17m. In this manner, the prediction necessity determination unit 170m performs the prediction necessity determination process for each distributed station 13.

[0507] The prediction necessity determination unit 170m may predict the future fluctuations in the number of connected terminals and the future fluctuations in the signal strength at the distributed station 13 based on time-series values ​​of the number of connected terminals and signal strength, including values ​​acquired in the past, and compare the predicted future fluctuations in the number of connected terminals and the future fluctuations in the signal strength with predetermined thresholds. The prediction necessity determination unit 170m may also be configured to perform a combination of threshold comparisons for the future number of connected terminals and the future signal strength, and threshold comparisons for the future fluctuations in the number of connected terminals and the future fluctuations in the signal strength.

[0508] Alternatively, the prediction necessity determination unit 170m may calculate the probability that the future number of connected terminals and the future signal strength values ​​at the distributed station 13 will exceed their respective thresholds S, based on the time-series number of connected terminals and time-series signal strength values, including values ​​acquired in the past. If the calculated probability is higher than the respective predetermined threshold, the unit may determine that the distributed station 13 requires the execution of a future traffic volume prediction process (using relatively large amounts of wireless control information by the traffic prediction unit 174m).

[0509] The prediction necessity determination unit 170m notifies the prediction unit 172m of information indicating the distributed stations 13 for which it has been determined that future traffic volume prediction processing is necessary. The prediction necessity determination unit 170m may also notify the prediction unit 172m of both information indicating the distributed stations 13 for which it has been determined that future traffic volume prediction processing is necessary, and information indicating the distributed stations 13 for which it has been determined that future traffic volume prediction processing is not necessary.

[0510] The prediction unit 172m acquires information from the prediction necessity determination unit 170m indicating the distributed station 13 for which it has been determined that future traffic volume prediction processing is necessary. The prediction unit 172m also receives relatively common types of radio control information from each distributed station 13 that are used to predict future traffic volume at the distributed station 13. In the modified version of the tenth embodiment, the radio control information includes data that associates the most recent traffic volume value at the distributed station 13 with the most recent number of connected terminals at the distributed station 13.

[0511] Furthermore, the prediction necessity determination unit 170m may be configured to pre-collect the above-mentioned relatively common types of radio control information used to predict future traffic volume at the distributed station 13, along with prediction necessity determination information (i.e., the number of connected terminals for each distributed station 13 and the signal strength values ​​of radio waves from terminal stations 11 connected to the distributed station 13). In this case, the prediction unit 172m acquires the relatively common types of radio control information used to predict future traffic volume at the distributed station 13 from the prediction necessity determination unit 170m.

[0512] The traffic prediction unit 174m of the prediction unit 172m predicts the future traffic volume at each distributed station 13 using radio control information, but only for the distributed stations 13 for which it has been determined that future traffic volume prediction processing is necessary.

[0513] Furthermore, for distributed stations 13 where it is determined that future traffic volume prediction processing is unnecessary, the traffic prediction unit 174m omits the future traffic volume prediction processing and instead considers the most recent traffic volume value at the distributed station 13 as the future traffic volume. Alternatively, the traffic prediction unit 174m may omit the future traffic volume prediction processing for distributed stations 13 where it is determined that future traffic volume prediction processing is unnecessary and instead use a previously calculated value of future traffic volume at the distributed station 13.

[0514] Furthermore, the prediction unit 172m calculates the future traffic volume at each aggregate station 14 based on the connection relationship between the distributed stations 13 and the aggregate station 14, and the future traffic volume at each distributed station 13. The uplink traffic volume at the aggregate station 14 corresponds to the sum of the uplink traffic volumes received by the aggregate station 14 from each of the distributed stations 13 under its control. The prediction unit 172m outputs information indicating the future traffic volume at each aggregate station 14 to the switching determination unit 175.

[0515] The switching determination unit 175 predicts whether future congestion will occur based on the relationship between the future traffic volume for each aggregation station 14 obtained from the prediction u...

Claims

1. A communication control device comprising: an acquisition unit that acquires wireless control information used for controlling route switching between a plurality of lower-level devices, which are communication devices that wirelessly connect to a terminal, and at least one higher-level device, which is a communication device that connects to any of the lower-level devices; a traffic prediction unit that predicts a plurality of traffic volumes based on the wireless control information and predicts the future traffic volume at the higher-level device based on the prediction results; and a switching determination unit that predicts whether or not future congestion will occur based on the predicted future traffic volume at the higher-level device and makes a decision on whether to perform the route switching.

2. The communication control device according to claim 1, wherein the wireless control information includes information indicating the amount of traffic and the number of connected terminals in the lower-level device, and the traffic prediction unit calculates the amount of traffic per terminal in each of the multiple lower-level devices based on the wireless control information, and predicts the future amount of traffic in each of the multiple lower-level devices based on the calculated amount of traffic per terminal.

3. The communication control device according to claim 2, wherein the traffic prediction unit predicts future traffic volume in a plurality of lower-level devices by performing a weighting calculation based on the time-series fluctuation amount of traffic volume per terminal in the lower-level device, or by performing a weighting calculation based on the traffic volume per terminal in other adjacent lower-level devices.

4. The communication control device according to claim 1 or 2, wherein the wireless control information further includes location information of the terminal, and further comprises a future location estimation unit that estimates the movement of the terminals present in the coverage area and surrounding area of ​​the subordinate device based on the time series change of the location information, and estimates the future number of connected terminals in the subordinate device based on the result of the movement estimation, and the traffic prediction unit predicts the future amount of traffic in a plurality of subordinate devices based on the estimated number of connected terminals.

5. The communication control device according to claim 4, wherein the wireless control information further includes information indicating the priority of each communication between the terminal and the subordinate device, and the future position estimation unit performs the movement estimation only for the terminal performing the communication which has a predetermined priority.

6. The communication control device according to claim 4, wherein the future position estimation unit performs the movement estimation only for terminals whose movement amount estimated by the movement estimation is greater than a predetermined value.

7. The communication control device according to claim 4, further comprising a predictive execution decision unit that identifies a subordinate device in which the value of the traffic amount included in the wireless control information exceeds a predetermined value, or the number of connected terminals included in the wireless control information exceeds a predetermined value, wherein the future position estimation unit performs the movement estimation only for the terminals located in and around the coverage area of ​​the subordinate device identified by the predictive execution decision unit.

8. The communication control device according to claim 1, wherein the wireless control information includes information indicating the radio wave strength for each terminal, and the traffic prediction unit calculates the amount of traffic per terminal in a plurality of lower-level devices based on predetermined wireless communication quality values ​​that differ according to the radio wave strength.

9. The communication control device according to claim 8, wherein the traffic prediction unit divides a plurality of terminals into strong signal terminals, which are terminals with relatively high signal strength, and weak signal terminals, which are terminals with relatively low signal strength, based on the signal strength and a predetermined threshold, and calculates the amount of traffic per terminal in a plurality of lower-level devices based on the sum of a value obtained by multiplying the number of strong signal terminals by a first throughput value and a value obtained by multiplying the number of weak signal terminals by a second throughput value smaller than the first throughput value.

10. The communication control device according to claim 9, wherein the predetermined threshold is the throughput value required in the service or system used by the terminal.

11. The communication control device according to claim 1, wherein the wireless control information includes handover information indicating the terminal on which a handover is performed in the lower-level device, and the traffic prediction unit calculates the amount of traffic per terminal in a plurality of lower-level devices based on predetermined wireless communication quality values ​​that differ depending on whether or not the terminal on which the handover is performed.

12. The communication control device according to claim 11, wherein the traffic prediction unit, based on the handover information, divides the plurality of terminals into non-executing terminals, which are terminals on which the handover is not performed, and executing terminals, which are terminals on which the handover is performed, and calculates the traffic volume per terminal in the plurality of lower-level devices based on the sum of the number of non-executing terminals multiplied by a first throughput value and the number of executing terminals multiplied by a second throughput value smaller than the first throughput value.

13. The communication control device according to claim 1, wherein the wireless control information includes fault information which is information that can identify an affected terminal which is a terminal whose wireless communication quality may deteriorate due to the occurrence of a fault, and the traffic prediction unit calculates the amount of traffic per terminal in a plurality of lower-level devices based on predetermined wireless communication quality values ​​which differ depending on whether or not the terminal is affected.

14. The communication control device according to claim 13, wherein the traffic prediction unit, based on the failure information, divides the plurality of terminals into unaffected terminals, which are terminals that are not affected by the occurrence of the failure, and affected terminals, and calculates the traffic volume per terminal in the plurality of lower-level devices based on the sum of the number of unaffected terminals multiplied by a first throughput value and the number of affected terminals multiplied by a second throughput value that is smaller than the first throughput value.

15. The communication control device according to claim 13 or 14, wherein the wireless control information includes the fault information output from the control controller of the wireless access network.

16. The communication control device according to claim 1, further comprising a prediction necessity determination unit that determines for each of the plurality of subordinate devices whether or not to perform a prediction of the future traffic volume in the subordinate device based on first information included in the wireless control information, wherein the traffic prediction unit performs the prediction of the future traffic volume in the subordinate device based on second information included in the wireless control information, which has a larger amount of information than the first information, only for the subordinate device for which the prediction necessity determination unit has determined to perform the prediction, and does not perform the prediction for the subordinate device for which the prediction necessity determination unit has determined not to perform the prediction.

17. The communication control device according to claim 16, wherein the first information includes at least one of the following: information indicating the amount of traffic for each terminal connected to the lower-level device, and information indicating the radio wave strength in the lower-level device for each terminal connected to the lower-level device.

18. The communication control device according to claim 16, wherein the acquisition unit acquires the second information only from the lower-level devices that the prediction necessity determination unit has determined to perform the prediction.

19. The communication control device according to claim 16, further comprising a collection frequency determination unit that identifies a fluctuation amount of the traffic amount in the lower-level device based on the information included in the wireless control information, which indicates the traffic amount in the lower-level device, and determines the collection frequency of the wireless control information by the acquisition unit based on the identified fluctuation amount.

20. The communication control device according to claim 19, wherein the collection frequency determination unit determines that the collection frequency should be increased as the specified amount of variation becomes relatively larger.

21. The communication control device according to claim 16, wherein the traffic prediction unit detects that there is insufficient wireless control information necessary to make the prediction of the future traffic volume in the lower-level device, requests the lower-level device to provide additional wireless control information, and makes the prediction using the additional wireless control information.

22. The communication control device according to claim 16, wherein, when the switching determination unit detects that there is insufficient information necessary to predict whether or not congestion will occur in the future, it requests the lower-level device to provide additional wireless control information and performs the prediction using the additional wireless control information.

23. The communication control device according to claim 16, further comprising a collection frequency determination unit that determines the frequency of collection of the wireless control information by the acquisition unit, wherein the traffic prediction unit measures the processing load when predicting the future amount of traffic, and the collection frequency determination unit determines the collection frequency based on the processing load.

24. A communication control method executed by a computer, comprising: an acquisition step of acquiring wireless control information used for controlling route switching between a plurality of lower-level devices, which are communication devices wirelessly connected to a terminal, and at least one upper-level device, which is a communication device connected to any of the lower-level devices; a traffic prediction step of predicting the future traffic volume at the plurality of lower-level devices based on the wireless control information and predicting the future traffic volume at the upper-level device based on the prediction results; and a switching determination step of predicting whether or not future congestion will occur based on the predicted future traffic volume at the upper-level device and making a decision on whether to execute the route switching.

Images