JP7742868B2 - Control device, control method, and program for controlling network functions based on disaster information - Google Patents

Description

The present invention relates to technology for controlling network function resources.

As mobile communication systems evolve from a communications infrastructure to a life infrastructure, there is a need to ensure the reliability of mobile communication systems. In particular, in Japan, large-scale disasters such as earthquakes occur once every few years, and various systems have been developed to mitigate the impact of damage caused by earthquakes and other disasters on mobile communication systems or services. For example, Patent Document 1 describes technology that switches communication paths based on predicted seismic intensity values in earthquake information.

JP 2013-135295 A

On the other hand, when a disaster actually occurs, there is a possibility that a large amount of communication traffic will temporarily occur in the area affected by the disaster. The technology in Patent Document 1 cannot deal with such a situation. The present invention provides a technology in a mobile communication system that allocates the resources required to process communication traffic according to communication demand.

A control device according to one aspect of the present invention is a control device that controls resources of a processing device in which a User Plane Function (UPF) in a cellular communication standard of the Third Generation Partnership Project (3GPP) is implemented, and includes an acquisition means for acquiring disaster information that identifies the occurrence of a disaster, an identification means for identifying an area affected by the disaster based on the disaster information, and a setting means for setting resources of the processing device to be allocated for the UPF, and when the occurrence of a disaster is detected by the disaster information, the control device identifies the UPF associated with the area identified by the identification means or associated with a terminal located in the area, from among a plurality of UPFs whose associated areas are at least partially different. and the setting means performs the setting to increase the resources allocated to the identified UPF, wherein the setting means increases the resources allocated to the UPF based on the fact that the sum of a potential traffic volume, which is the volume of communication traffic predicted to be transmitted by the terminal located in the identified area, and an inflow traffic volume predicted to flow into the UPF from another UPF, exceeds a first threshold based on the processing capacity of the UPF, and the inflow traffic volume is determined based on the probability that the other UPF will cease to operate normally, which is determined based on the magnitude of the impact of the disaster in the area where the other UPF is operating, and the potential traffic volume of the other UPF .

According to the present invention, in a mobile communication system, the resources required to process communication traffic can be allocated according to communication demand.

FIG. 1 is a diagram illustrating an example of the configuration of a mobile communication system. FIG. 2 illustrates an example of a hardware configuration of a management server. FIG. 2 illustrates an example of a functional configuration of a management server. FIG. 10 is a diagram illustrating an operation flow of the management server.

The following embodiments are described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the scope of the claimed invention, and not all combinations of features described in the embodiments are necessarily essential to the invention. Two or more of the multiple features described in the embodiments may be combined in any desired manner. Furthermore, the same reference numbers are used for identical or similar components, and duplicate descriptions will be omitted.

(System configuration)
1 shows an example of the configuration of a mobile communication system according to this embodiment. The mobile communication system is, for example, a cellular communication system conforming to the cellular communication standard of the Third Generation Partnership Project (3GPP (registered trademark)). However, this is not limited to this, and the following discussion can be applied to a mobile communication system conforming to any wireless communication standard. The mobile communication system is, for example, configured to include terminals 101, 102, 103, 104, 105, 106, and 107, base stations 111, 112, 113, 114, 115, and 116, UPFs 121, 122, and 123, a data network 130, TAs 141, 142, and 143, AMF 150, SMF 160, a management server 170, and a disaster information server 180. UPF, TA, AMF, and SMF are abbreviations for User Plane Function, Tracking Area, Access and Mobility Management Function, and Session Management Function, respectively. Terminals 101 to 107 may be collectively referred to as terminals 100. Base stations 111 to 116 may be collectively referred to as base stations 110. UPFs 121 to 123 may be collectively referred to as UPF 120. TAs 141 to 143 may be collectively referred to as TA 140. While FIG. 1 illustrates an example in which three UPFs 121-123 are connected to one data network 130 via a wired network, the number of UPFs 120 may be one, two, or four or more. Furthermore, the UPFs 120 may be connected to two or more data networks 130, and each UPF 120 may be connected to a different data network 130. Similarly, two base stations 110 are illustrated as representatives of each UPF 120, but the number may be one, three, or more. Furthermore, two base stations 110 constituting each TA 140 are illustrated as representatives of each TA 140, but the number may be one, three, or more. Furthermore, although one UPF 120 is illustrated as corresponding to each TA 140, two or more UPFs 120 may be present. The number of terminals 100 connected to each base station 110 may be zero or more, and one terminal 100 may be connected to multiple base stations 110 simultaneously. Although one of each of the AMF 150, SMF 160, management server 170, and disaster information server 180 is illustrated as a representative example, there may be two or more of each. The devices may be interconnected via a wired network or a wireless network. In a mobile communication system, each function that performs processing for communication may be referred to as a network function. In this embodiment, the base station 110, UPF 120, AMF 150, SMF 160, etc. are examples of network functions.

The terminal 100 is a terminal used by a user and exchanges radio signals with the base station 110 via a wireless medium. The terminal 100 may be referred to as User Equipment (UE). The terminal 100 includes, for example, a smartphone, a mobile phone, a personal computer, a tablet terminal, a wearable terminal, an IoT (Internet of Things) terminal, etc. The base station 110 exchanges radio signals with the terminal 100 via a wireless medium. The base station 110 includes, for example, a gNB (next generation Node B), an eNB (evolved Node B), etc. The UPF 120 provides connection point functions such as routing for forwarding user plane data transmitted from the terminal 100 to the data network 130 and for forwarding user plane data received from the data network 130 to the destination terminal 100. User plane data is information exchanged between the terminal 110 and the data network 130, and may be referred to as data in this embodiment. When transferring data between the base station 110 and the data network 130, multiple UPFs 120 may be interposed. The data network 130 is a network that provides data services to the terminal 100, and includes, for example, an IP (Internet Protocol) service network of a public land mobile network (PLMN), the Internet, a corporate network, etc. A TA 140 is a unit of area used for location registration of the terminal 100. A TA 140 may be composed of one or more cells. A cell is a communicable range covered by each base station 110. The terminal 100 may be managed as being present within one of the TAs 140, with its location associated with the TA 140. In this embodiment, the presence of the terminal 100 within the TA 140 may also be expressed as the terminal 100 being present within the TA 140. For example, when communication traffic addressed to a certain terminal 100 occurs, all base stations 110 that constitute the TA 140 associated with and registered with the terminal 100 call the terminal 100. The AMF 150 handles procedures and management for the terminal 100, such as registration, connection, and movement to the network. The SMF 160 provides session management functions when the terminal 100 transmits data to the data network. For example, the SMF 160 assigns an IP address to the terminal 100, selects the UPF 120 to connect to, and manages PDU (Packet Data Unit) sessions. The management server 170 manages resources in the mobile communication system. The disaster information server 180 provides disaster information to each device constituting the mobile communication system. The disaster information server 180 is, for example, a server that provides emergency earthquake alerts provided by the Japan Meteorological Agency. The disaster information server 180 may also be a server that provides alerts such as tsunami alerts, fire alerts, and nationwide instantaneous alerts.

An overview of the procedure for the terminal 100 to access the mobile communication system and receive services is described below. When the terminal 100 is turned on, it performs an initial registration procedure. For example, the terminal 100 transmits a registration request message to the base station 110. When transmitting the registration request message, the terminal 100 may use a Subscription Concealed Identifier (SUCI) or a 5G Globally Unique Temporary Identifier (5G-GUTI) as information to identify the terminal 100. When the base station 110 receives the registration request message from the terminal 100, it selects an AMF 150 to connect to based on the identifier of the terminal 100, etc. The base station 110 may, for example, select the AMF 150 that is geographically closest to the terminal 100 and transmit a registration request message to this AMF 150. Each base station 110 may define one AMF 150 to be selected by the base station 110 by default. When the AMF 150 receives a registration request message, it authenticates the terminal 100, which is the sender, and performs registration processing. The AMF 150 can authenticate the terminal 100, for example, by querying an Authentication Server Function (AUSF) not shown. Furthermore, in the registration processing, the AMF 150 can register in a Unified Data Management (UDM) not shown that the AMF itself is the serving AMF for this terminal 100. The UDM is a database function of a mobile communication system that stores subscriber contract information and the like. Furthermore, the AMF 150 can notify the terminal 100 of a registration acceptance message including a 5G-GUTI, a registration area, a periodic registration update timer, and the like. The registration area included in the registration acceptance message may be, for example, the TA 140 in which the terminal 100 resides, or may include multiple surrounding TAs 140. A Tracking Area Identifier (TAI), which is an identifier that uniquely identifies each TA 140, may be assigned to the TA 140. Furthermore, the location of the terminal 100 may be managed by a Registration Area consisting of multiple TAs 140. In this embodiment, the location of the terminal 100 is managed by association with the TA 140, and each TA 140 is identified by a TAI. The terminal 100 notifies the AMF 150 of its location based on a periodic registration update timer. As an example, the terminal 100 periodically notifies the AMF 150 of the Cell Global Identifier (CGI) and TAI based on a periodic registration update timer. The CGI is an identifier that uniquely identifies a cell and can be composed of a PLMN identifier and a cell identifier. In this way, by the terminal 100 performing periodic location registration or location update, the AMF 150 can manage the location of each terminal 100 by linking it with the TA 140 in which the terminal 100 is located.

When data destined for the data network 130 is generated, the terminal 100 transmits a PDU session request message to the AMF 150 via the base station 110. For example, the PDU session request message may include information such as a PDU session identifier generated by the terminal 100 and the specified data network 150. The AMF 150 selects an appropriate SMF 160 based on the information included in the PDU session request message and forwards a PDU session establishment request message. The AMF 150 may select the SMF 160 based on, for example, the subscriber contract information of the terminal 100. The SMF 160 selects a UPF 120 required to establish the requested PDU session and allocates the PDU session. Data transfer between the terminal 100 and the data network 130 may go through multiple UPFs 120. In this case, tunneling is performed on a transfer route formed by the multiple UPFs 120, and user plane data may be transferred using the tunnel. The SMF 160 transmits information including address information of the UPF 120 that is the termination point of the tunnel and an accept message for setting up the PDU session to the AMF 150. The AMF 150 notifies the base station 110 of the address information of the UPF 120 that is the termination point of the tunnel, and notifies the terminal 100 of an accept message. When the AMF 150 is notified by the base station 110 that preparation for tunneling is complete, it notifies the SMF 160 of the address of the base station 110 as the other termination point of the tunneling. This opens a PDU session from the terminal 100 to the data network 130 via the base station 110. Note that different PDU sessions can be established between one terminal 100 and multiple UPFs 120. For example, if multiple applications used by the terminal 100 connect to different data networks 130, PDU sessions can be established individually between one terminal 100 and multiple different UPFs 120 corresponding to each data network 130. That is, the UPF 120 can be selected according to its physical location and the data network 130 to which it is connected. In addition, the AMF 150 or the SMF 160 can individually manage the UPF 120 to which each terminal 100 connects for each PDU session.

In this embodiment, an example has been shown in which AMF150 works in cooperation with SMF160 to manage the location of terminal 100 and establish a PDU session, but the procedures for managing the location of terminal 100 and establishing a PDU session, and the device that executes those procedures, are not limited to these. For example, AMF150 and SMF160 may execute the procedures as a single device, or another device may execute the procedures. For example, a Mobility Management Entity (MME) in the Evolved Packet Core (EPC) of Long Term Evolution (LTE) may execute the location management and session management of terminal 100.

As described above, when data to be transmitted to the data network 130 is generated at the terminal 100, the terminal 100 requests a PDU session for connection to the data network 130 and begins exchanging data with the data network 130 via the established PDU session. Furthermore, if no data exchange occurs for a certain period of time, the terminal 100 transitions to an inactive state. Therefore, when designing a mobile communication system, it is possible to assume that only a portion of the terminals 100 connected to the mobile communication system will be active at all times, rather than assuming that all terminals 100 will be active at all times. In this case, for example, if a disaster such as an earthquake occurs in a certain area, there may be a temporary surge in demand for communications to obtain disaster-related information, confirm the safety of users, etc. If a load exceeds the processing capacity of a specific network function, delays or discards of communication traffic may occur. For example, if a large number of users each begin collecting information using their terminals 100, a large number of PDU sessions will be required, which may result in a concentrated load on the UPF 120 installed in that area. In particular, immediately after a disaster occurs, there is a high demand for urgent and continuous information gathering, making communication traffic congestion difficult to resolve. Furthermore, as a result of congestion, communication traffic delays and discards may cause users to repeatedly resend information requests, further increasing the volume of communication traffic and placing a greater load on network functions. Therefore, mobile communication systems must quickly resolve the temporarily increased emergency load. In this embodiment, virtualization technology is applied to the network functions that make up the mobile communication system, and resources allocated to the network functions are controlled based on disaster information, thereby providing network functions with processing capabilities that meet demand. Resource control of network functions in this embodiment is described below.

(Network function virtualization and resource control by management server)
In this embodiment, a description will be given using the UPF 120 as an example of a network function to which the present technology is applied. The network function to which the present technology is applied is not limited to the UPF 120, and the following discussion can be applied to any network function constituting a mobile communication system. The UPF 120 in this embodiment is implemented as software that executes a network function on a virtualization layer installed on a general-purpose server. Such a virtualized network function may also be called a Virtualized Network Function (VNF). In other words, a network function realized by virtualization is a processing device implemented as software. The general-purpose server may be configured, for example, with a processing device such as an accelerator including a Central Processing Unit (CPU) and a Graphics Processing Unit (GPU), a storage medium such as a memory or a hard disk, and a communication device such as a Network Interface Card. The virtualization layer, also known as a hypervisor, is a control program that allows VNFs to run on general-purpose servers. Specifically, the UPF 120 is implemented as a VNF running on a general-purpose server and can process communication traffic, such as data transfer, using the resources of the general-purpose server through the hypervisor. Therefore, the processing capacity of the UPF 120 depends on the resources allocated to the UPF 120. Generally, a VNF can run on a single general-purpose server or across multiple general-purpose servers. Each network function may be executed by a single VNF or may be executed by multiple VNFs working together. Therefore, to increase the processing capacity of a network function (e.g., the UPF 120), additional general-purpose servers (or parts of the hardware that constitute them) or VNFs may be allocated. Releasing the general-purpose servers (or parts of the hardware that constitute them) or VNFs allocated to the UPF 120 can reduce the resources of the UPF 120. In this embodiment, regardless of the specific method for increasing or decreasing the processing capacity of the UPF 120, increasing or decreasing the processing capacity of the UPF 120 may be expressed as controlling resources. Adding resources for a network function may be called scale-out. Releasing resources for a network function may be called scale-in. In this embodiment, for example, the management server 170 controls the resources allocated to the UPF 120. The management server 170 may perform, for example, management of hardware resources constituting a general-purpose server, management of each VNF, orchestration of the entire virtualized network function, etc. The management server 170 may be implemented by individual servers that perform each function operating in cooperation with each other. For example, a function that manages hardware resources constituting a general-purpose server may be called a Virtualized Infrastructure Manager (VIM). A function that manages each VNF may be called a VNF Manager (VNFM). The function that orchestrates the entire virtualized network function may be called an NFV Orchestrator (NFVO). As an example, when scaling out the UPF 120, in the management server 170, the NFVO first checks the availability of resources with the VIM and reserves the resources. The NFVO instructs the VNFM to scale out the UPF 120. The NFVO may notify the VNFM of the identifiers of the reserved resources. The VNFM requests resources from the VIM. The VIM allocates resources such as general-purpose servers or VNFs. The VNFM uses the allocated resources to configure the UPF 120 for operation. This completes the scale-out of the UPF 120. The method by which the management server 170 controls the resources of the UPF 120 is not limited to this, and other functions may be included, or some functions may be omitted.

(Resource control based on disaster information)
In this embodiment, the management server 170 controls network function resources based on disaster information. First, the management server 170 identifies areas affected by the occurrence of a disaster. As an example, the management server 170 acquires disaster information from the disaster information server 180. For example, if the disaster information is an earthquake early warning, the disaster information may include the time of the earthquake, the estimated location of the earthquake (epicenter), the estimated magnitude of the earthquake, the name of the affected area, the predicted seismic intensity, and the predicted time of arrival of the main tremor in the area. Furthermore, earthquake information may be provided multiple times after the earthquake occurs. The affected areas and seismic intensity may differ for each piece of earthquake information. The management server 170 may select which information to use from multiple pieces of disaster information. For example, if the disaster information is used for processing requiring rapid response, the management server 170 uses the disaster information provided earlier. Furthermore, if the disaster information is used for processing requiring reliability, the management server 170 may use the disaster information provided later. For example, the management server 170 may request that the device itself be registered with an information distribution service provided by the disaster information server 180 in order to receive disaster information from the disaster information server 180. The management server 170 identifies areas that will be heavily affected by the disaster based on the acquired disaster information. For example, if the disaster information is an earthquake early warning, a heavily affected area may mean that the predicted or observed seismic intensity is high. For example, if the disaster information is a weather warning, a heavily affected area may mean that the alert level is high. Furthermore, if the disaster information is a major tsunami warning, a heavily affected area may mean that the tsunami height is high. For example, the management server 170 may identify areas where the magnitude of the impact of the disaster exceeds a predetermined threshold as areas affected by the disaster. Note that affected areas may also be identified based on the occurrence of the disaster itself, regardless of the magnitude of the impact of the disaster.

Then, the management server 170 identifies the UPF 120 that handles communication traffic occurring in the area affected by the disaster. As an example, the management server 170 first identifies the TA 140 that corresponds to the area (specific area) identified as the area affected by the disaster. For example, the management server 170 may have a list of areas covered by each of the TAs 140 that make up the mobile communication system. In this case, the management server 170 may use that list to identify the TA 140 that corresponds to the specific area. Furthermore, if the AMF 150 has a similar list, the management server 170 may request information about the TA 140 from the AMF 150. For example, the management server 170 notifies the AMF 150 of an inquiry including information about the specific area, and the AMF 150 provides a list of the TAs 140 that cover the specific area. The list of TAs 140 may include the TAI of each TA 140. If a device other than the AMF 150 has such a list, the management server 170 may make an inquiry to that device.

The management server 170 then identifies network functions that can handle communication traffic occurring in the identified TA 140. For example, if each UPF 120 is directly associated with the TA 140, the management server 170 may have a list of UPFs 120 associated with each TA 140. In this case, the management server 170 may identify the associated UPF 120 based on the identified TA 140. The management server 170 may also query network functions such as the AMF 150 and the SMF 160 to obtain information about the UPF 120. By querying the network functions that manage the location information of each terminal 100 and the PDU session between each terminal 100 and the UPF 120, the management server 170 may obtain information about the UPF 120 corresponding to the identified TA 140. On the other hand, when each UPF 120 is associated with each terminal 100, the management server 170 may, for example, identify the terminal 100 present in each TA 140 and identify the UPF 120 associated with this terminal 100. In this case, the management server 170 may inquire of the AMF 150 about information (such as an identifier) of the terminal 100 present in the identified TA 140. Furthermore, when the AMF 150 or the SMF 160 knows the UPF 120 associated with each terminal 100, the management server 170 may acquire information about the UPF 120 associated with each terminal 100 from the AMF 150 or the SMF 160. Note that, for example, when different UPFs 120 are associated with each type of data, such as a slice, a QoS, an application, or a network that provides data, the management server 170 may identify the associated UPF 120 based on the type of data. In this case, the management server 170 may identify applications that users are expected to use depending on the disaster situation, and may make inquiries to the AMF 150, SMF 160, etc. based on the identified applications. For example, if it is expected that usage of the safety confirmation service will increase, the management server 170 may inquire of the SMF 160, etc. about the UPF 120 that corresponds to the data network 130 that provides the safety confirmation service. In this embodiment, as an example, the description will be given assuming that the UPF 120 is directly associated with each TA 140, but the method by which the management server 170 identifies the UPF 120 associated with the identified TA 140 is not limited to this.

The management server 170 predicts the amount of communication traffic (potential traffic volume) that may occur in the identified network function. As an example, the management server 170 predicts the potential traffic volume of the UPF 120 based on the number of terminals 100 present in each TA 140. The management server 170 may, for example, specify the identified TA 140 and inquire of the AMF 150 about the number of terminals 100 present in that TA 140. The AMF 150 extracts the terminals 100 present in the TA 140 queried by the management server 170. For example, each terminal 100 notifies the AMF 150 to periodically update its own location registration information. Therefore, the AMF 150 may identify each terminal 100 present in each TA 140 using the location registration information of each terminal 100 for which the AMF 150 is registered as a serving AMF. The AMF 150 may then report the number of terminals 100 counted for each TA 140 to the management server 170. For example, the management server predicts the communication traffic volume by multiplying the number of terminals 100 by a predetermined communication traffic volume and a weighting factor based on the magnitude of the impact of the disaster. The predetermined communication traffic volume may be, for example, the maximum communication traffic volume per terminal or a communication traffic volume that ensures minimum communication. The predetermined communication traffic volume may be determined based on the results of past similar disasters, or may be determined based on the average communication traffic volume when a specific application is used. It may also be determined based on other methods. Furthermore, the weighting factor may be determined based on, for example, the magnitude of the impact of the disaster. For example, when an earthquake occurs, different weighting factors may be used for each seismic intensity. For example, if the seismic intensity ranges from highest to lowest, from highest to lowest, be 7, 6+, 6-low, 5+, and 5-low or less, the weighting factors may be set to decrease in stages from highest to lowest, such as 1.2, 0.8, 0.5, 0.1, and 0. For example, in one case (Case 1), when an earthquake occurs, seismic intensities of 7, 6-weak, and 5-up are observed at TAs 141 to 143 in FIG. 1 , respectively. Furthermore, suppose that terminals 100 Nu1, Nu2, and Nu3 are present at TAs 141 to 143, respectively. In this case, if the communication traffic volume per terminal is defined as Tru, the potential traffic volumes Tr1 to Tr3 at UPFs 121 to 123 are Tr1 = 1.2 × Nu1 × Tru, Tr2 = 0.5 × Nu2 × Tru, and Tr3 = 0.1 × Nu3 × Tru, respectively. In this way, the management server 170 can predict the potential traffic volume of UPF 120 based on the number of terminals 100 present at each TA 140. Furthermore, when a UPF 120 handles communication traffic from multiple TAs 140, the potential traffic volume of each UPF 120 can be expressed as the sum of the potential traffic volumes of those multiple TAs 140.

The management server 170 determines whether to add resources to a network function based on the resources allocated to the network function and the potential traffic volume. For example, the management server 170 calculates the processing capacity of the UPF 120 being evaluated based on the resources currently allocated to the UPF 120, and determines whether the potential traffic volume of the UPF 120 exceeds that processing capacity. The processing capacity of the UPF 120 can be expressed, for example, as the amount of communication traffic that can be processed within a unit time. For example, the processing capacity of the UPF 120 depends on the performance of resources such as general-purpose hardware and VNFs allocated to the UPF 120. Therefore, the management server 170 can determine the amount of communication traffic that the UPF 120 can process per unit time based on the performance of the resources allocated to the UPF 120. If the potential traffic volume exceeds the processing capacity of the UPF 120, the management server 170 can determine to allocate additional resources. Furthermore, if the potential traffic volume is lower than the processing capacity of the UPF 120, the management server 170 may determine not to allocate additional resources. Note that the ratio of the processing capacity of a given UPF 120 to the potential traffic volume of that UPF 120 may be referred to as the congestion risk level. The management server 170 may determine to allocate additional resources if the congestion risk level exceeds a predetermined threshold. For example, the management server 170 may determine to allocate additional resources if the congestion risk level exceeds 1. Furthermore, the management server 170 may determine not to allocate additional resources if the congestion risk level does not exceed 1.

The management server 170 may control resources based on the number of terminals present in an area affected by a disaster. For example, the management server 170 identifies an area affected by a disaster based on disaster information. If the identified area is directly associated with a network function (such as UPF 120), the management server 170 may add resources to the network function based on the number of terminals present in that area. As an example, the management server 170 may associate the amount of resources to be added based on the number of terminals 100 associated with UPF 120. For example, if the number of terminals 100 T1, T2, ..., Tn is associated with the amount of resources to be added R1, R2, ..., Rn, respectively, if the number of terminals present in the area associated with the target network function is less than T1, R1 resources are allocated. Similarly, if the number of present terminals is greater than or equal to T1 and less than T2, R2 resources are allocated, and if the number of present terminals is greater than or equal to Tn-1 and less than Tn, Rn-1 resources are allocated. In this way, the management server 170 determines the resources to add to the network functions based on the number of terminals 100, regardless of the extent of the impact of the disaster or the amount of communication traffic generated by the terminals 100, thereby reducing the amount of calculation required to allocate resources and allowing resources to be allocated to the processing performed by the network functions.

When a disaster occurs, physical damage to equipment installed in the affected area or disruptions such as power outages in the building where the equipment is installed may occur. In this case, equipment such as general-purpose servers on which network functions are running may cease to function. In this case, the processing that should be performed by the network function on the failed general-purpose server may be controlled so that a network function installed on a general-purpose server in another area performs the processing. In this embodiment, traffic flowing into a network function from another network function may be referred to as "inflow traffic." In this embodiment, the management server 170 may calculate the congestion risk taking into account the possibility of inflow traffic occurring due to the failure of surrounding UPFs 120 caused by the disaster. For example, the management server 170 may determine the likelihood that each UPF 120 will cease to function based on the magnitude of the disaster's impact on the area where the general-purpose servers on which the UPFs 120 are running are installed. As an example, if the disaster is an earthquake, the management server 170 may determine different failure risks for each seismic intensity. The failure risk is the probability that equipment or devices will cease to function properly due to the disaster. For example, if the seismic intensity is 7, 6, 5, or less, the failure risk may be determined to decrease in stages from high to low, such as 0.1, 0.05, and 0. As an example, when the failure risk is included, the potential traffic volume of a UPF 120 can be expressed as the sum of the potential traffic volume of that UPF 120 and the product of the potential traffic volumes of the surrounding UPFs 120 and the failure risk of each UPF 120. For example, in the above case 1, when the failure risks of UPFs 121 and 122 are included, the potential traffic volume Tr3' of UPF 123 can be calculated as Tr3' = 0.1 × Tr1 + 0.05 × Tr2 + Tr3. The congestion risk of UPF 123 in this case is Tr3' / UP123's processing capacity.

When management server 170 determines that resources should be added to a specific network function, it executes a procedure for allocating additional resources to that network function. For example, management server 170 adds resources by performing resource control of the network function as described above. The additional resources may be determined, for example, so that the processing capacity of the network function after the resources are added exceeds the potential traffic volume. As an example, management server 170 may allocate resources equivalent to the difference between the potential traffic volume and the processing capacity of UPF 120. If the processing capacities of UPFs 121 and 123 in Case 1 above are P1 and P3, respectively, the additional resources allocated to UPF 121 may be the resources necessary to process the communication traffic volume of Tr1-P1. Furthermore, the additional resources allocated to UPF 123 when considering the risk of failure may be the resources necessary to process the communication traffic volume of Tr3'-P3.

When the management server 170 allocates additional resources to a specific network function, it may release the allocated resources based on the satisfaction of certain conditions. If a disaster causes a temporary increase in demand for communications to disappear or ease, continuing to allocate resources to that network function may result in a decrease in the utilization rate of the general-purpose server on which that network function operates. For example, the management server 170 may monitor a parameter indicating the resource usage status of the network function to which additional resources have been allocated, and release the allocated resources if the measured value of that parameter falls below a certain threshold for a certain period of time. The measured value indicating the resource usage status may be, for example, the processing volume or utilization rate of the VNF. The measured value indicating the resource usage status may also be other measured values, such as the CPU utilization rate of the general-purpose server on which the network function operates or the amount of communication traffic processed by the network function within a unit time. The predetermined threshold may be a predetermined threshold, for example, a threshold based on the processing capacity of the network function before the additional resources were allocated. For example, if N times the amount of resources allocated before the additional resources were allocated, 1/N may be set as the resource usage threshold for releasing the additional resources.

When management server 170 performs resource control for surrounding network functions using a congestion risk level, including a failure risk level, taking into account the possibility that network functions in areas severely affected by a disaster will fail, it may release resources allocated to the surrounding network functions upon detecting that a network function that was at risk of failure is operating normally. For example, management server 170 calculates the failure risk level for each network function based on disaster information. If there is a network function with a non-zero failure risk, management server 170 may allocate additional resources to the surrounding network functions to process incoming traffic, taking into account the possibility that the network function will fail. Management server 170 then monitors the operation of network functions that are at risk of failure (with a non-zero failure risk). Based on the results of monitoring over a predetermined period of time, management server 170 determines whether the network functions are operating normally. If it determines that the network functions are operating normally, management server 170 may release resources allocated to the surrounding network functions to process incoming traffic. Releasing resources that were redundantly allocated based on the failure risk level enables more effective resource utilization.

In this way, the management server 170 can use disaster information to identify areas that will be severely affected by the disaster and add resources for network functions associated with those areas, thereby making it possible to handle the temporary increase in communication traffic.

(Circuit configuration)
Next, an example configuration of the management server 170 as described above will be described. FIG. 2 is a diagram showing the hardware configuration of the management server. In one example, the management server 170 includes a processor 201, a ROM 202, a RAM 203, a storage device 204, and a communication circuit 205. The processor 201 is a computer including one or more processing circuits, such as a general-purpose CPU (Central Processing Unit) or an ASIC (Application Specific Integrated Circuit). The processor 201 reads and executes programs stored in the ROM 202 or the storage device 204 to execute the overall processing of the device and each of the above-mentioned processes. The ROM 202 is a read-only memory that stores information such as programs and various parameters related to the processing executed by the management server 170. The RAM 203 functions as a workspace when the processor 201 executes a program and is a random access memory that stores temporary information. The storage device 204 is, for example, a removable external storage device. The communication circuit 205 is, for example, configured to include a circuit for communicating with other devices.

(Functional configuration)
FIG. 3 is a diagram showing an example of the functional configuration of the management server 170. The management server 170 is configured to include, as its functions, a disaster information acquisition unit 301, a disaster area identification unit 302, and a resource setting unit 303, for example. FIG. 3 shows the functional configuration of the management server 170 of this embodiment, and, for example, omits the general configuration of a management server. Note that these functional units can be realized, for example, by the processor 301 executing programs stored in the ROM 302 or the storage device 304 and controlling the communication circuit 305 as necessary. However, this is not limited to this, and, for example, dedicated hardware for realizing each function may be provided.

The disaster information acquisition unit 301 acquires disaster information from the disaster information server 180. The disaster area identification unit 302 identifies areas that will be severely affected by the disaster based on the acquired disaster information. The resource setting unit 303 allocates resources to network functions. In order to identify the network function (such as UPF 120) to which resources are to be allocated, the resource setting unit 303 identifies the network function associated with the area identified by the disaster area identification unit 302. For example, if UPF 120 is directly associated with an area such as TA 140, the resource setting unit 303 may identify the TA 140 that covers the area identified by the disaster area identification unit 302, and thereby identify the UPF 120 associated with that TA 140. The resource setting unit 303 may determine whether or not additional resources are required for UPF 120 based on the number of terminals 100 present in the identified TA 140 and the amount of communication traffic that may occur in the terminals 100. For example, if the number of terminals 100 present in the TA 140 associated with the UPF 120 exceeds a predetermined threshold, the resource setting unit 303 allocates additional resources to the UPF 120. On the other hand, if the number of present terminals 100 is less than a predetermined threshold, the resource setting unit 303 does not allocate additional resources. Note that, if multiple thresholds associated with the number of terminals 100 are set, the resource setting unit 303 may allocate resources according to the number of present terminals 100. The resource setting unit 303 may also predict potential traffic volume based on the terminals 100 and the magnitude of the disaster. The resource setting unit 303 may also allocate resources based on the assumption that a specific network function will be affected by a disaster, taking into account the amount of communication traffic that may flow into other network functions due to the failure. The resource setting unit 303 may determine whether to add resources by comparing the potential traffic volume and inflow traffic volume in the UPF 120 with the processing capacity of the UPF 120. Furthermore, after adding resources, the resource setting unit 303 may release the additionally allocated resources based on the resource utilization rate of the network function falling below a predetermined threshold. Furthermore, taking into consideration the possibility of failures in surrounding network functions, the resource setting unit 303 may release resources allocated for processing incoming traffic based on the detection that surrounding network functions are operating normally.

(Processing flow)
FIG. 4 shows an example of a processing flow when the management server 170 in this embodiment allocates resources to the UPF 120 based on disaster information.

First, when a disaster occurs, the management server 170 acquires disaster information from the disaster information server 180 (S401). The management server 170 identifies areas that will be severely affected by the disaster based on the acquired disaster information (S402). For example, if the disaster information is an earthquake early warning, areas with high seismic intensity may be identified as areas that will be severely affected by the disaster. The management server 170 identifies a UPF 120 associated with the identified area (S403). For example, if the UPF 120 is directly associated with a TA 140, the TA 140 that covers the identified area may be identified, thereby identifying the UPF 120 associated with this TA 140. The method for identifying the UPF 120 is not limited to this method. For example, the associated UPF 120 may be identified by querying the AMF 150 or SMF 160 that manages the location information of the terminal 100 and PDU session information between the UPF 120 and the terminal 100. The management server 170 predicts the potential traffic volume of the identified UPF 120 (S404). For example, the potential traffic volume of each UPF 120 can be predicted by tallying the number of terminals 100 present in the area identified as an area severely affected by the disaster for each UPF 120. Instead of predicting the communication traffic volume, the management server 170 may determine the resources to be added to each UPF 120 based on the number of terminals 100 that may be associated with each UPF 120. The management server 170 compares the potential traffic volume with the processing capacity of the UPF 120. If the predicted communication traffic volume exceeds the processing capacity of the UPF 120 (YES in S405), the management server 170 allocates resources to the UPF 120 and terminates the process (S406). If the predicted communication traffic volume is lower than the processing capacity of the UPF 120 (NO in S405), the management server 170 terminates the process without allocating resources to the UPF 120.

As described above, according to this embodiment, the management server 170 allocates additional resources to expand processing capacity to network functions associated with areas where communication traffic may temporarily increase, based on disaster information acquired from the disaster information server 180. This makes it possible to proactively increase resources in response to the concentration of load on specific network functions and communication traffic congestion that may occur when a disaster occurs, and to avoid interruptions or outages of communication services. This makes it possible to contribute to Goal 9 of the United Nations-led Sustainable Development Goals (SDGs), which is to "Develop resilient infrastructure, promote sustainable industrialization and foster innovation."

The invention is not limited to the above-described embodiments, and various modifications and variations are possible within the scope of the invention.

101: Terminal, 111: Base Station, 121: UPF, 130: Data Network, 140: TA, 150: AMF, 160: SMF, 170: Management Server, 180: Disaster Information Server

Claims (7)

A control device that controls resources of a processing device in which a user plane function (UPF) in a cellular communication standard of a Third Generation Partnership Project (3GPP) is implemented,
an acquisition means for acquiring disaster information for identifying the occurrence of a disaster;
an identification means for identifying an area affected by the disaster based on the disaster information;
a setting means for setting resources of the processing device to be allocated for the UPF, the setting means specifying, when the occurrence of a disaster is detected by the disaster information, the UPF associated with an area specified by the specifying means or associated with a terminal located in the area, from among a plurality of UPFs whose associated areas are at least partially different, and performing the setting so as to increase the resources to be allocated to the specified UPF ;
the setting means increases resources allocated to the UPF based on the fact that a sum of a potential traffic volume, which is a communication traffic volume predicted to be transmitted by the terminal located in the specified area, and an inflow traffic volume predicted to flow into the UPF from another UPF, exceeds a first threshold based on a processing capacity of the UPF;
The inflow traffic volume is determined based on a probability that the other UPF will not operate normally, which is determined based on the magnitude of the impact of the disaster in the area where the other UPF is operating, and the potential traffic volume of the other UPF.
A control device characterized by: 2. The control device according to claim 1, wherein the setting means estimates the potential traffic volume using the number of terminals located in the area identified by the identifying means and a coefficient associated with the magnitude of the disaster . 2. The control device according to claim 1, wherein the setting means, when it is detected that the other UPF is operating normally, releases the amount of resources allocated to process the incoming traffic volume from the resources of the UPF . 2. The control device according to claim 1, wherein the setting means increases the resources allocated to the UPF based on the occurrence of a disaster being detected by the disaster information, and then releases the increased amount of the resources of the UPF when the utilization rate of the resources allocated to the UPF becomes equal to or less than a second threshold. The control device according to claim 1, characterized in that the setting means identifies a tracking area corresponding to the area identified by the identification means, and identifies the UPF that handles communication traffic occurring in the identified tracking area, thereby identifying the UPF associated with the area identified by the identification means or associated with a terminal located in the area. 1. A control method executed by a control device that controls resources of a processing device that executes a User Plane Function (UPF) in a cellular communication standard of a Third Generation Partnership Project (3GPP), comprising:
an acquisition step of acquiring disaster information that identifies the occurrence of a disaster;
an identifying step of identifying an area affected by the disaster based on the disaster information;
a setting step of setting resources of the processing device to be allocated for the UPF, wherein, when an occurrence of a disaster is detected by the disaster information, the setting step identifies the UPF associated with the area identified by the identifying step or associated with a terminal located in the area from among a plurality of UPFs whose associated areas are different in at least some respects, and performs the setting so as to increase the resources to be allocated to the identified UPF ;
The setting step increases resources allocated to the UPF based on the fact that a sum of a potential traffic volume, which is a communication traffic volume predicted to be transmitted by the terminal located in the specified area, and an inflow traffic volume predicted to flow into the UPF from another UPF, exceeds a first threshold based on a processing capacity of the UPF;
The inflow traffic volume is determined based on a probability that the other UPF will not operate normally, which is determined based on the magnitude of the impact of the disaster in the area where the other UPF is operating, and the potential traffic volume of the other UPF.
A control method comprising: A computer provided in a control device that controls resources of a processing device that executes a user plane function (UPF) in a cellular communication standard of the Third Generation Partnership Project (3GPP),
Acquire disaster information that identifies the occurrence of a disaster;
Identifying an area affected by the disaster based on the disaster information;
A setting of resources of the processing device to be allocated for the UPF, in which, when an occurrence of a disaster is detected by the disaster information, the UPF associated with the identified area or associated with a terminal located in the identified area is identified from among a plurality of UPFs having at least a part of different associated areas, and the setting is made to increase the resources to be allocated to the identified UPF ;
In the setting, resources allocated to the UPF are increased based on the fact that a sum of a potential traffic volume, which is a communication traffic volume predicted to be transmitted by the terminal located in the specified area, and an inflow traffic volume predicted to flow into the UPF from another UPF, exceeds a first threshold based on a processing capacity of the UPF;
The inflow traffic volume is determined based on a probability that the other UPF will not operate normally, which is determined based on the magnitude of the impact of the disaster in the area where the other UPF is operating, and the potential traffic volume of the other UPF.
Program for.