Traffic prediction device, traffic prediction program, and traffic prediction method

Abstract

To provide a communication system in which a user apparatus is connected to a network, capable of improving a prediction accuracy of a traffic expected to occur.SOLUTION: A traffic prediction device for predicting a traffic occurring in a communication system in which a user apparatus is connected to a network, includes: information collection means for collecting information including attribute-specific population information, attribute-specific service utilization ratio information, and service bandwidth information in an interested area; information holding means for holding information including population transition information at a time of interest in the interested area; and calculation processing means for calculating a total bandwidth of the traffic occurring in the interested area based on the information collected by the information collection means and the information held by the information holding means.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a traffic prediction device, a traffic prediction program, and a traffic prediction method, and can be applied to predicting traffic in a communication carrier network (for example, a mobile communication network), for example. [Background technology]

[0002] Conventionally, in mobile communication networks (communication carrier networks), wireless base stations are generally divided into a slave station communication device that communicates with user devices and a master station communication device that controls the slave station communication device. The master station communication device and the slave station communication device are connected by a transmission path such as optical fiber, and multiple wireless devices are installed within the wireless base station. The number of wireless devices per wireless base station is determined by the number of mobile terminals that the wireless base station must handle and the amount of traffic set based on predicted usage frequency. Furthermore, mobile terminals must be able to be used even during periods of high traffic, resulting in a significant number of wireless devices.

[0003] On the other hand, traffic fluctuates depending on the time of day, with low traffic late at night and early in the morning and high traffic from noon to night. Furthermore, with the launch of 5G services in recent years, high-definition image transmission services and services aimed at advancing cities using digital twin technology have become widespread. In recent years, services have become more diversified and large-capacity, and the differences in traffic depending on the time of day, etc., are becoming increasingly large.

[0004] Conventional wireless devices in mobile communication networks are installed assuming high traffic times, and operate even during low traffic times, resulting in unnecessary power consumption. Therefore, wireless devices in mobile communication networks must be operated efficiently, satisfying QoS (Quality of Service) while minimizing power consumption.

[0005] To achieve the above-mentioned goal of efficient operation, there are conventional technologies disclosed in Patent Documents 1 and 2 that observe traffic in a communication network, calculate future traffic volume based on the observed data, and determine the communication bandwidth of the network.

[0006] Patent Document 1 describes a method and device for calculating the required bandwidth for mobile communication services. The technology described in Patent Document 1 calculates the normalized observed traffic volume for each hour, normalized by the 24-hour average value of the total traffic volume on weekdays or holidays in a certain area, calculates the monthly average, and calculates the maximum traffic volume for a future month, thereby improving the accuracy of calculating the bandwidth facility volume required to achieve QoS.

[0007] Patent Document 2 describes how the accuracy of future traffic predictions can be improved by taking into account not only traffic information obtained within a communication network but also external factors (weather information and calendar information) as input data. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] Japanese Patent Application Publication No. 2018-037960 [Patent Document 2] Japanese Patent Application Laid-Open No. 2012-253445 Summary of the Invention [Problem to be solved by the invention]

[0009] However, in general, in a mobile communication network, the amount of traffic generated by a user device may differ even when compared on a daily basis. Furthermore, when traffic prediction is performed using the average value of past traffic, as in the conventional technology, there is a problem that accurate traffic prediction becomes difficult when, for example, a sudden event or disaster occurs. Furthermore, in general, in a mobile communication network, there is a problem that the amount of traffic generated in each area varies depending on the age composition of the area, but the conventional technology has not been able to solve this problem.

[0010] In view of the above problems, there is a need for a traffic prediction device, a traffic prediction program, and a traffic prediction method that can improve the accuracy of predicting traffic that is expected to occur for each area in which user devices are located in a communication system that connects user devices to a network. [Means for solving the problem]

[0011] The first invention of the present invention is a traffic prediction device that predicts traffic generated in a communication system that connects user devices used by users to a network, characterized by having: information collection means that collects information including attribute-specific population information on the population for each user attribute in a target area; attribute-specific service usage rate information that indicates the network service usage rate for each user attribute in the target area; and service bandwidth information on the bandwidth used by the user for each network service; information storage means that stores information including population trend information on population trends in the target area at a future target time; and calculation processing means that calculates the total bandwidth of traffic generated in the user devices that access the network from the target area at the target time based on the information collected by the information collection means and the information stored by the information storage means.

[0012] A second traffic prediction program of the present invention is characterized in that it causes a computer installed in a traffic prediction device that predicts traffic generated in a communication system that connects user devices used by users to a network to function as: information collection means that collects information including attribute-specific population information on the population for each user attribute in a target area, attribute-specific service usage rate information that indicates the network service usage rate for each user attribute in the target area, and service bandwidth information on the bandwidth used by the user for each network service; information storage means that stores information including population trend information on population trends in the target area at a future target time; and calculation processing means that calculates the total bandwidth of traffic generated in the user devices accessing the network from the target area at the target time based on the information collected by the information collection means and the information stored by the information storage means.

[0013] A third aspect of the present invention is a traffic prediction method performed by a traffic prediction device that predicts traffic generated in a communication system that connects user devices used by users to a network, the traffic prediction device having information collection means, information holding means, and calculation processing means, wherein the information collection means collects information including attribute-specific population information on the population for each user attribute in a target area, attribute-specific service usage rate information indicating the network service usage rate for each user attribute in the target area, and service bandwidth information on the bandwidth used by the user for each network service, the information holding means holds information including population trend information on population trends in the target area at a future target time, and the calculation processing means calculates, based on the information collected by the information collection means and the information held by the information holding means, a total bandwidth of traffic generated in the user devices accessing the network from the target area at the target time. [Effects of the Invention]

[0014] According to the present invention, in a communication system that connects user devices to a network, it is possible to improve the accuracy of predicting traffic that is expected to occur for each area in which user devices exist. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 2 is a block diagram showing a functional configuration of a traffic prediction unit (traffic prediction device) according to the embodiment. [Figure 2] 1 is a diagram illustrating a configuration of a communication system according to an embodiment. [Figure 3] 10 is a diagram showing an example of information acquired about an area of ​​interest by an area / service information collection unit according to the embodiment. FIG. [Figure 4] 10 is a diagram (part 1) illustrating an example of information acquired about an area of ​​interest by a population transition information collection unit according to the embodiment. FIG. [Figure 5] 10 is a flowchart illustrating an operation of a traffic prediction unit according to the embodiment when performing a traffic prediction process. [Figure 6] FIG. 10 is a diagram (part 1) illustrating a process when a traffic prediction unit according to an embodiment performs a traffic prediction process. [Figure 7] FIG. 10 is a diagram (part 2) illustrating an example of information acquired about an area of ​​interest by a population transition information collection unit according to the embodiment. [Figure 8] FIG. 10 is a diagram (part 2) illustrating a process when the traffic prediction unit according to the embodiment performs traffic prediction processing. [Figure 9] FIG. 10 is a diagram showing the configuration of a communication system according to a modified embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0016] (A) Main embodiment Hereinafter, an embodiment of a traffic prediction device, a traffic prediction program, and a traffic prediction method of the present invention will be described in detail with reference to the drawings. In this embodiment, an example will be described in which the traffic prediction device, the traffic prediction program, and the traffic prediction method of the present invention are applied to a traffic prediction unit.

[0017] (A-1) Configuration of the embodiment FIG. 2 is a block diagram showing the overall configuration of a communication system 1 according to this embodiment.

[0018] The communication system 1 includes N (N is an integer equal to or greater than 1) slave station communication devices 20 (20-1 to 20-N) as antenna devices constituting a wireless access network in a mobile communication network (e.g., a so-called 5G mobile communication network), and a master station communication device 10 as a signal processing device that controls each slave station communication device 20 to form the wireless access network. The slave station communication devices 20 transmit and receive wireless signals to and from mobile terminals TE, which are user devices used by users who subscribe to the communication system 1 (mobile communication network). Furthermore, the master station communication device 10 is connected to a core network CN (e.g., a core network of a communication carrier). That is, the communication system 1 is configured to connect the mobile terminals TE to the core network CN via the slave station communication devices 20 and the master station communication device 10. Each of the slave station communication devices 20-1 to 20-N is assigned an area (hereinafter simply referred to as an "area" or "mesh area") corresponding to a predetermined mesh code, and the communication system 1 performs roaming processing or the like to connect to (accommodate) a mobile terminal TE residing within its assigned area. In this embodiment, areas A-1 to AN are assigned to the slave station communication devices 20-1 to 20-N, respectively. Note that in this embodiment, each area corresponds to a 500m mesh code, but the mesh code to which each area corresponds is not limited. Note that in this embodiment, one area (mesh area) is assigned to one slave station communication device 20, but multiple areas (mesh areas) may be assigned to one slave station communication device 20.

[0019] Hereinafter, communication from the master station communication device 10 side (core network CN side) to the slave station communication device 20 side (mobile terminal TE side) will be referred to as "downstream communication," and communication from the slave station communication device 20 side (mobile terminal TE side) to the master station communication device 10 side (core network CN side) will be referred to as "upstream communication."

[0020] As shown in FIG. 1, the master station communication device 10 includes a communication control unit 11 that controls each slave station communication device 20 to perform communication control for forming a wireless access network.

[0021] The communication control unit 11 includes a traffic prediction unit 12 that performs a process (hereinafter referred to as "traffic prediction process") to predict traffic occurring in each area (each slave station communication device 20). In this embodiment, the communication control unit 11 performs communication control (e.g., allocation of bandwidth for upstream communication / downstream communication) for each area (each slave station communication device 20) based on the result of the traffic prediction process by the traffic prediction unit 12. The traffic prediction process performed by the communication control unit 11 will be described later. Furthermore, the communication process based on the result of the traffic prediction process performed by the communication control unit 11 is not limited, and various types of traffic control (e.g., bandwidth control) may be performed.

[0022] Next, the detailed configuration of the traffic prediction unit 12 will be described.

[0023] The traffic prediction unit 12 performs a process (traffic prediction process) to predict traffic that will occur in an arbitrary area (hereinafter referred to as an "area of ​​interest") at an arbitrary future time (hereinafter referred to as an "interesting time") based on information that can be collected at the current time t0. Note that in this embodiment, the traffic prediction unit 12 will be described as predicting downstream communication traffic, but it may also be configured to predict upstream communication traffic. In the following, the interest times will be represented as t1, t2, t3, etc., in order from the previous time in the time series.

[0024] Incidentally, the amount of traffic generated in each area varies depending on the population staying there, and the services (network services) that are frequently used also vary depending on the age group of the population staying there. For example, people in their 20s tend to use video viewing and social networking services a lot, while people in their 50s and 60s tend to use email and web browsing services a lot. Therefore, external factors such as weather information and calendar information alone are insufficient, making it difficult to accurately predict traffic. In other words, to improve the accuracy of traffic forecasts for a certain area, it is desirable to make traffic forecasts that take into account the population by attribute, such as age, staying in that area.

[0025] Therefore, the traffic prediction unit 12 performs traffic prediction processing taking into consideration the visiting population for each user attribute (e.g., user age and generation) in the area of ​​interest. The "visiting population" is statistical data expressed by various values, such as the monthly average value (by weekday / holiday) of the number of people sampled at each specified time (e.g., 4:00, 10:00, 14:00, 20:00) and daily values ​​(e.g., the number of people sampled on a specified day corresponding to the date and time when the traffic prediction processing is performed).

[0026] FIG. 1 is a block diagram showing the functional configuration of the traffic prediction unit 12. As shown in FIG.

[0027] The traffic prediction unit 12 has an area and service information collection unit 121 as information collection means, a population transition information collection unit 122 as information storage means, a population calculation unit 123, a service bandwidth calculation unit 124, a traffic calculation unit 125, and a traffic prediction output unit 126. Hereinafter, the population calculation unit 123, the service bandwidth calculation unit 124, and the traffic calculation unit 125 will be collectively referred to as the "calculation processing means."

[0028] The traffic prediction unit 12 may be configured, for example, by installing a program (including, for example, the traffic prediction program according to the embodiment) in a computer having a memory and a processor. The traffic prediction unit 12 may be configured using a dedicated computer, or may be configured using a computer on which other functions (programs) are also installed.

[0029] The area and service information collection unit 121 has the function of collecting information including information on the visiting population by age group (by user attribute) and by time of day for each area (hereinafter referred to as "visiting population information by age group / time"), information on the service usage rate by age group (hereinafter referred to as "service usage rate information by age group"), and information on the bandwidth usage for each service (hereinafter referred to as "service bandwidth information"). Details of the information collected by the area and service information collection unit 121 will be described later. The means by which the area and service information collection unit 121 collects the information is not limited. For example, the information may be obtained from open data (e.g., data downloadable by accessing a specific URL on the Internet) or from data from a specific provider (e.g., data transmitted from a server of a specific provider that handles data such as the visiting population for each area). The area and service information collection unit 121 may also collect information registered in advance by a system administrator or the like.

[0030] The population trend information collecting unit 122 performs a process (hereinafter referred to as a "population trend information storing process") to store information on the transition of the visiting population (e.g., rate of increase or decrease) at a future time of interest (hereinafter referred to as "population trend information"). The population trend information collecting unit 122 performs a process of estimating and storing the transition of the total population (number of people ratio) for each area (hereinafter referred to as "total population trend information") and the transition of the population ratio by age group (hereinafter referred to as "age-group population ratio transition information") as population trend information. The population trend information collecting unit 122 may store the total population trend information and the age-group population ratio transition information based on, for example, the current age-group / time-group visiting population information for each area. The population trend information storing process performed by the population trend information collecting unit 122 will be described in detail below.

[0031] The population calculation unit 123 performs a process of calculating the visiting population by age group at a future time of interest for each area (hereinafter referred to as "population calculation process"). The population calculation unit 123 performs the population calculation process based on the visiting population information by age group / time group collected by the area / service information collection unit 121, the results of the population transition information storage process by the population transition information collection unit 122, etc.

[0032] The service bandwidth calculation unit 124 performs a process of calculating the service usage bandwidth by age group at a time of interest for each area (hereinafter referred to as "service bandwidth calculation process"). The service bandwidth calculation unit 124 performs the service bandwidth calculation process based on the information acquired by the area / service information collection unit 121 (service usage ratio information by age group and service bandwidth information) and the results of the population calculation process by the population calculation unit 123.

[0033] The traffic calculation unit 125 performs a process (hereinafter referred to as a "traffic calculation process") to calculate a predicted value of the traffic volume (total bandwidth used) at a time of interest for each area. The traffic calculation unit 125 calculates the predicted value of the traffic volume (total bandwidth) from the result of the service bandwidth calculation unit 124.

[0034] The traffic prediction output unit 126 has a function of outputting the traffic calculation processing result performed by the traffic calculation unit 125. The format and destination in which the traffic prediction output unit 126 outputs the traffic calculation processing result are not limited. For example, when the traffic prediction unit 12 performs traffic calculation processing for an arbitrary area based on control from the communication control unit 11, the traffic prediction unit 12 may obtain the current traffic calculation processing result for that area and return it to the traffic prediction unit 12. Furthermore, the traffic prediction output unit 126 may transmit the traffic calculation processing result to an external server or the like.

[0035] Next, specific examples of information collected by the area and service information collecting unit 121 (information on visiting population by age group and time, information on service usage ratio by age group, and service bandwidth information) will be described.

[0036] FIG. 3 is a diagram showing an example of information (visitor population information by age group / time at a specific time, service usage rate information by age group, and service bandwidth information) acquired by the area / service information collection unit 121 for the area of ​​interest.

[0037] FIG. 3(a) shows information on the visiting population by age group and time in area A-1 at an arbitrary time t0.

[0038] Here, the visiting population information by age group and time is assumed to be information indicating the population composition by time in each area (population composition by attribute; in this embodiment, population composition by age group). The visiting population information by age group and time shown in FIG. 3(a) indicates the visiting population by age group. In this embodiment, as shown in FIG. 3, the user attribute "age" is expressed as one of five values: 20s (29 years old or younger), 30s (30 to 39 years old), 40s (40 to 49 years old), 50s (50 to 59 years old), and 60s (60 years old or older). In the following description, the user attribute "age" is expressed as one of the above five values, but the present invention is not limited to the value used to express the user attribute "age". For example, values ​​such as "teens" (19 years old or younger) and "70s" (70 years old or older) may be set for the age group.

[0039] FIG. 3(b) shows information on service usage ratios by age group at an arbitrary time t0 in area A-1.

[0040] Here, the service usage ratio information by age group is assumed to be information showing the usage ratio of each service (network service) by age group, as shown in FIG. 3(b).

[0041] In this embodiment, the services are described as being classified into any one of video (for example, video streaming service), email (for example, sending and receiving email data to and from a server), and web browsing (for example, browsing web pages on the Internet), as shown in Fig. 3. The number and types of services classified are not limited to the above three types, and various combinations can be applied.

[0042] The service usage rate information by age group shown in Figure 3(b) shows the usage rate (unit: [%]) of each service by age group. For example, in Figure 3(b), for people in their 20s, it is [video 50%, email 25%, web browsing 25%] (total 100%). Note that for the service usage rate information by age group, different data may be applied for each area and each time (for example, each specified time), or uniform data may be applied for each area and each time.

[0043] FIG. 3(c) shows service band information at an arbitrary time t0 in area A-1.

[0044] Here, as shown in FIG. 3(c), the service bandwidth information indicates the bandwidth (unit: [Mbps]) required for each service. For example, in FIG. 3(c), it is [video: 30 Mbps, email: 5 Mbps, web browsing: 10 Mbps]. The service bandwidth information may use a general value that satisfies QoS. Also, different data may be applied to each area and each time (for example, each specified time), or uniform data may be applied to each area and each time.

[0045] FIG. 4 is a diagram showing an example of population transition information (total population transition information and age-specific population proportion transition information) acquired by the population transition information collecting unit 122 for the area of ​​interest.

[0046] FIG. 4(a) is a diagram showing an example of total population transition information for an area of ​​interest when the time of interest is t1.

[0047] The total population trend information shown in Figure 4(a) displays the total population at the current time t0 and the total population at time t1, assuming time t1. The total population trend information shown in Figure 4(a) displays the total population at each time in percentage. For example, if the population information by age group and time in the area of ​​interest at time t0 is as shown in Figure 3(a), the total population in the area of ​​interest at time t0 will be 2,000. If the total population trend information at time t0 is as shown in Figure 4(a) (t0: 100%, t1: 100%), this indicates that the proportion of the total population is the same at t0 and t1, and therefore the total population at time t1 will also be 2,000, just like t0.

[0048] FIG. 4(b) is a diagram showing an example of information on transition of population ratio by age group for an area of ​​interest at a time of interest t1.

[0049] The population percentage transition information by age group shown in Figure 4(a) shows the population percentage by age group at time t1 in [%]. The population percentage transition information by age group shown in Figure 4(a) is [20s: 80%, 30s: 10%, 40s: 5%, 50s: 5%, 60s: 0%]. As described above, if the total population of the area of ​​interest at time t1 is 2,000 based on the total population transition information, etc., multiplying the percentage of each age group in the population percentage transition information by age group by the total population (2,000) gives the population by age group: [20s: 1,600, 30s: 200, 40s: 100, 50s: 100, 60s: 0].

[0050] As described above, the population calculation unit 123 of this embodiment can ascertain the visiting population by age group at the time of interest based on the population transition information acquired from the population transition information collection unit 122. Note that the specific configuration of the population transition information held by the population transition information collection unit 122 and the method of holding the population transition information are not limited, as long as the population calculation unit 123 can hold information that enables it to ascertain the visiting population by age group at the time of interest.

[0051] Next, a specific example of the population transition information storage process performed by the population transition information collection unit 122 will be described.

[0052] For example, the population trend information collecting unit 122 may generate and store population trend information for each designated time of the visiting population sample (for each designated time of the visiting population sample that is the source of the visiting population information by age group / time; for example, 4:00 AM on weekdays, 10:00 AM on weekdays, 2:00 PM on weekdays, 8:00 PM on weekdays, 4:00 AM on holidays, 10:00 AM on holidays, 2:00 PM on holidays, and 8:00 PM on holidays) based on the visiting population information by age group / time of the area of ​​interest collected by the population trend information collecting unit 122 (information on the visiting population by age group for each time series) and using the population at the current time t0 as the base (the population at the current time t0 is 100%), and may select population trend information for a designated time corresponding to the date and time of the visiting population information by age group / time (for example, the designated time closest to the time of interest) and supply this information to the population calculation unit 123. Furthermore, for example, the population trend information collecting unit 122 may select a designated time corresponding to the time of interest (for example, select a designated time closest to the time of interest), and generate and store population trend information corresponding to the designated time selected based on the visiting population information by age group / time, etc. Furthermore, for example, the population trend information collecting unit 122 may store population trend information for each area and for each specified time in advance (for example, by operation of a system administrator, etc.), and may store population trend information for the specified time corresponding to the date and time of the time of interest. Furthermore, the population trend information collecting unit 122 may adjust (increase or decrease) each value of the population trend information according to the weather forecast (for example, temperature and weather) for the time of interest in the area of ​​interest, the details of events being held, etc. Also, for example, the population trend information collecting unit 122 may predict future population increases or decreases based on population increases or decreases in areas that have experienced past torrential rain (localized heavy rain that falls in a short period of time), or may obtain information on the location and date and time of future events (hereinafter referred to as "event information") from information on social networking services (SNS) (for example, information posted on the timeline of an SNS), and predict future population increases or decreases for each area based on the obtained event information.

[0053] (A-2) Operation of the embodiment Next, a traffic prediction process (traffic prediction method according to the embodiment) performed by the traffic prediction unit 12 according to the embodiment will be described.

[0054] FIG. 5 is a flowchart showing the operation of the traffic prediction unit 12 when performing traffic prediction processing (processing for predicting traffic occurring in an area of ​​interest at a time of interest).

[0055] First, the population calculation unit 123 calculates the total visiting population in the area of ​​interest at the time of interest based on information supplied from the area / service information collection unit 121 and the population transition information collection unit 122 (S101), and then performs a process of calculating the population by age group (population calculation process) (S102). Specifically, the population calculation unit 123 acquires visiting population information by age group / time at the time of interest in the area of ​​interest from the area / service information collection unit 121. The population calculation unit 123 also acquires total population transition information and age group population proportion transition information for the area of ​​interest at the time of interest from the population transition information collection unit 122. Then, the population calculation unit 123 calculates the total population in the area of ​​interest at the time of interest based on the acquired visiting population information by age group / time and total population transition information. Then, the population calculation unit 123 calculates the visiting population by age group in the area of ​​interest at the time of interest by multiplying each value of the age group population proportion transition information (population proportion by age group) by the total population.

[0056] Next, the service bandwidth calculation unit 124 performs a process (service bandwidth calculation process) to calculate the service usage bandwidth by age group at the time of interest (hereinafter referred to as the "service usage bandwidth by age group") based on the information (service usage ratio information by age group and service bandwidth information) acquired from the area / service information collection unit 121 and the result of the population calculation process by the population calculation unit 123 (visiting population by age group) (S103). Specifically, the service bandwidth calculation unit 124 calculates the number of users of each service by age group (hereinafter referred to as the "number of service users by age group") by multiplying each value of the service usage ratio information by age group by the visiting population by age group. Then, the service bandwidth calculation unit 124 calculates each value of the service usage bandwidth by age group by multiplying each value of the number of service users by age group by each bandwidth in the service bandwidth information.

[0057] Next, the traffic calculation unit 125 calculates the total bandwidth of the traffic occurring in the area of ​​interest at the time of interest by summing up the service usage bandwidths by age group calculated by the service bandwidth calculation unit 124, and obtains it as the result of the traffic prediction process.Then, the traffic prediction output unit 126 outputs the result of the traffic prediction process (total bandwidth) in a predetermined format (S104).

[0058] In the above-described manner, the traffic prediction unit 12 performs traffic prediction processing.

[0059] Next, an example of the specific process and results of traffic prediction processing when the traffic prediction unit 12 operates according to the flowchart in Fig. 5 will be described. Here, an example of traffic prediction processing at future target times t1 and t2 will be described. Note that the following procedure for traffic prediction (calculation processing) is merely an example, and the present invention is not limited to the following procedure.

[0060] [Traffic prediction processing for time t1] First, a case will be described in which the traffic prediction unit 12, at time t0, performs a process of predicting traffic that will occur in area A-1 at a future time t1 of interest. That is, the area of ​​interest here is area A-1.

[0061] Here, it is assumed that the visiting population information by age group / time, service usage rate information by age group, and service bandwidth information for the area of ​​interest A-1 at time t0 are the contents shown in Fig. 3. It is also assumed that the total population transition information and population rate transition information by age group for the area of ​​interest A-1 at time t1 of interest are the contents shown in Fig. 4.

[0062] FIG. 6 is a diagram showing a process in which the traffic predicting unit 12, at time t0, performs a process of predicting traffic that will occur in the area A-1 at a future time t1 of interest.

[0063] In step S101, the population calculation unit 123 calculates the total number of people and the visiting population by age group at the target time t1 from the visiting population information by age group / time (FIG. 3(a)), total population transition information (FIG. 4(a)), and age group population proportion transition information (FIG. 4(b)). Here, as shown in FIG. 6(a), the total population at time t1 is 2000 x 100 (%), which is 2000 people. Furthermore, the population for each age group at time t1 is multiplied by the population proportion for that age group, resulting in [20s: 1600 people, 30s: 200 people, 40s: 100 people, 50s: 100 people, 60s: 0 people], as shown in FIG. 6(a).

[0064] Next, in step S102, the population calculation unit 123 calculates the number of service users by age group at the target time t1 (FIG. 6(b)) by multiplying the service usage ratio information by age group (FIG. 3(b)) by the result (visiting population by age group) calculated in step S101. Note that here, if the calculation result is a decimal point, the number of users is rounded up to avoid a bandwidth shortage.

[0065] Next, in step S103, the service bandwidth calculation unit 124 can obtain the service usage bandwidth by age group (Figure 6(c)) by multiplying the number of service users by age group (Figure 6(b)) by the corresponding value of the service bandwidth information at time t0 (Figure 3(c)).

[0066] Next, in step S104, the traffic calculation unit 125 calculates the total bandwidth (traffic) that will occur in the area of ​​interest A-1 at the future time t1 by adding up each bandwidth (each bandwidth by age group / service) shown in the service usage bandwidth by age group (FIG. 6(c)). In this case, as shown in FIG. 6(d), the predicted value of the total bandwidth of the traffic that will occur in the area of ​​interest A-1 at the future time t1 is 35725 Mbps.

[0067] [Traffic prediction processing for time t2] Next, a case will be described in which the traffic predicting unit 12, at time t0, performs a process of predicting traffic that will occur in the area A-1 at a future time t2 of interest.

[0068] FIG. 7 is a diagram showing total population transition information and age-specific population ratio transition information at the target time t2 of the target area A-1.

[0069] FIG. 8 is a diagram showing a process in which the traffic predicting unit 12 predicts traffic that will occur in the area A-1 at a future time t2 of interest.

[0070] Then, in step S101, the population calculation unit 123 calculates the total number of people and the visiting population by age group at the target time t2 from the visiting population information by age group / time (FIG. 3(a)), the total population transition information (FIG. 7(a)), and the population proportion transition information by age group (FIG. 7(b)). Here, as shown in FIG. 8(a), the total population at time t2 is 2000 x 100 (%), which is 2000 people. Furthermore, the population for each age group at time t2 is multiplied by the population proportion for that age group, resulting in [20s: 0 people, 30s: 100 people, 40s: 100 people, 50s: 200 people, 60s: 1600 people], as shown in FIG. 8(a).

[0071] Next, in step S102, the population calculation unit 123 multiplies the service usage ratio information by age group (Figure 3(b)) by the result (visiting population by age group) obtained in step S101 to obtain the number of service users by age group at the target time t2 (Figure 8(b)).

[0072] Next, in step S103, the service bandwidth calculation unit 124 can obtain the service usage bandwidth by age group (Figure 8(c)) by multiplying the number of service users by age group (Figure 8(b)) by the corresponding value of the service bandwidth information at time t0 (Figure 3(c)).

[0073] Next, in step S104, the traffic calculation unit 125 calculates the total bandwidth (traffic) that will occur in the area of ​​interest A-1 at the future time t2 of interest by adding up each bandwidth (each bandwidth by age / service) shown in the service usage bandwidth by age group (FIG. 8(c)). In this case, as shown in FIG. 8(d), the predicted value of the total bandwidth of the traffic that will occur in the area of ​​interest A-1 at the future time t2 of interest is 24760 Mbps.

[0074] [Comparison with conventional traffic prediction processing] Here, we will explain the case where the total bandwidth of traffic occurring in the area of ​​interest A-1 at times t1 and t2 is calculated simply by using the total population of the area of ​​interest A-1, without using the service usage bandwidth by age group (Figures 6(c) and 8(c)). Here, as with the conventional technology, the average bandwidth for each service is used to calculate the predicted total bandwidth. In this case, the average bandwidth for each service is (30 + 5 + 10) / 3 = 15 Mbps, and multiplying this by the visiting population of 2,000 people results in 30,000 Mbps. In other words, if the traffic prediction process for times t1 and t2 were performed using the conventional technology, the predicted total bandwidth would both be 30,000 Mbps. However, because the service usage bandwidth by age group for the area of ​​interest A-1 (Figures 6(c) and 8(c)) is not taken into consideration, there is a high possibility that the accuracy will be lower than that of the traffic prediction unit 12 of this embodiment.

[0075] (A-3) Effects of the embodiment According to this embodiment, the following effects can be achieved.

[0076] In this embodiment, the traffic prediction unit 12 calculates the service usage bandwidth by age group from the service usage ratio information by age group, and performs traffic prediction based on the service usage bandwidth by age group, thereby improving prediction accuracy. This has the effect of satisfying the QoS in the communication system 1 and suppressing excess power consumption (for example, suppressing an increase in power consumption of each device due to excessive bandwidth allocation).

[0077] (B) Other embodiments The present invention is not limited to the above-described embodiment, and may include modified embodiments such as those exemplified below.

[0078] (B-1) In the above embodiment, the transmission path between the master station communication device 10 and each slave station communication device 20 is not limited, and various transmission paths can be applied. The type of the transmission path between the master station communication device 10 and each slave station communication device 20 is not limited, and may be, for example, various PtP (Point-to-Point) type transmission paths (e.g., various Ethernet (registered trademark) lines, digital leased lines, etc.) or various PtMP (Point-to-Multipoint) type transmission paths (e.g., PON (Passive Optical Network) type connection configuration, branchable digital leased lines, etc.).

[0079] FIG. 9 shows an example of the configuration of a communication system 1A in which a PON type transmission line is applied to the transmission line between the master station communication device 10 and each slave station communication device 20 (20-1 to 20-N).

[0080] 9, a master station communication device 10 and each slave station communication device 20 (20-1 to 20-N) are connected via an optical communication network system 30, which is a PON-type transmission path. The optical communication network system 30 has an OLT (Optical Line Terminal) 31 connected to the master station communication device 10, and ONUs (Optical Network Units) 32 (32-1 to 32-N) connected to each slave station communication device 20 (20-1 to 20-N). In the optical communication network system 30, the OLT 31 and each ONU 32 (32-1 to 32-N) are connected via optical fibers 33, which serve as optical transmission paths branched into N by a splitter 34.

[0081] (B-2) In the above embodiment, an example has been described in which the traffic prediction unit 12 serving as the traffic prediction device of the present invention is mounted in the master station communication device 10, but there are no limitations on the device in which the traffic prediction unit 12 is located. For example, the traffic prediction unit 12 may be configured as an independent traffic prediction device.

[0082] (B-3) In the traffic prediction unit 12 of each of the above embodiments, the concept of age is used as a user attribute, and traffic prediction processing is performed based on the population by age group and the service usage rate thereof, but the applied user attribute is not limited to age, and other concepts may be used. For example, in the traffic prediction unit 12 of each of the above embodiments, the user attributes may be classified based on other attributes such as gender, occupation, etc. (or classified using a combination of multiple user attributes) and traffic prediction processing may be performed. [Explanation of symbols]

[0083] 1...communication system, 10...master station communication device, 11...communication control unit, 12...traffic prediction unit, 20, 20-1 to 20-N...substation communication devices, 20-1...substation communication device, 30...optical communication network system, OLT...31, ONU...32-1 to 32-N, 33...optical fiber, 34...splitter, 121...service information collection unit, 122...population trend information collection unit, 123...population calculation unit, 124...service bandwidth calculation unit, 125...traffic calculation unit, 126...traffic prediction output unit, A, A-1 to AN...area, CN...core network, TE...mobile terminal

Claims

1. A traffic prediction device for predicting traffic generated in a communication system that connects user devices used by users to a network, comprising: an information collecting means for collecting information including attribute-specific population information relating to the population for each user attribute in a target area, attribute-specific service usage rate information indicating the network service usage rate for each user attribute in the target area, and service bandwidth information relating to the bandwidth used by the user for each network service; an information storage means for storing information including population transition information regarding population transition in the area of ​​interest at a future time of interest; a calculation processing means for calculating a total bandwidth of traffic generated in the user devices accessing the network from the area of ​​interest at the time of interest based on the information collected by the information collecting means and the information held by the information holding means; A traffic prediction device comprising:

2. The calculation processing means Calculating the population for each user attribute in the area of ​​interest at the time of interest based on the attribute-specific population information and the population transition information; determining a service usage bandwidth by user attribute, which is a bandwidth used by the user for each user attribute and each network service in the area of ​​interest at the time of interest, based on the population by user attribute in the area of ​​interest at the time of interest and the attribute-by-attribute service usage ratio information; Calculating the total bandwidth in the area of ​​interest at the time of interest based on the service bandwidth for each user attribute and the service bandwidth information 2. The traffic prediction device according to claim 1, wherein:

3. 3. The traffic prediction device according to claim 2, wherein the user attribute is the age of the user.

4. A computer installed in a traffic prediction device that predicts traffic generated in a communication system that connects a user device used by a user to a network, an information collecting means for collecting information including attribute-specific population information relating to the population for each user attribute in a target area, attribute-specific service usage rate information indicating the network service usage rate for each user attribute in the target area, and service bandwidth information relating to the bandwidth used by the user for each network service; an information storage means for storing information including population transition information regarding population transition in the area of ​​interest at a future time of interest; a calculation processing means for calculating a total bandwidth of traffic generated in the user devices accessing the network from the area of ​​interest at the time of interest based on the information collected by the information collecting means and the information held by the information holding means; A traffic prediction program characterized by functioning as follows.

5. A traffic prediction method performed by a traffic prediction device that predicts traffic generated in a communication system that connects user equipment used by users to a network, comprising: the traffic prediction device includes an information collection means, an information storage means, and a calculation processing means; the information collecting means collects information including attribute-specific population information relating to the population for each user attribute in the area of ​​interest, attribute-specific service usage rate information indicating the network service usage rate for each user attribute in the area of ​​interest, and service bandwidth information relating to the bandwidth used by the user for each network service; the information storage means stores information including population transition information regarding population transition in the target area at a future target time; The calculation processing means calculates the total bandwidth of traffic generated in the user devices accessing the network from the area of ​​interest at the time of interest based on the information collected by the information collecting means and the information held by the information holding means. A traffic prediction method comprising:

Images